From: Logical Language Group <lojbab>
Message-Id: <199502231711.AA02659@access2.digex.net>
Subject: Carterian formula (was: Gricean formula?)
Date: Thu, 23 Feb 1995 12:11:15 -0500 (EST)
Content-Transfer-Encoding: 7bit
X-From-Space-Date: Thu Feb 23 12:11:23 1995
X-From-Space-Address: lojbab

Jim Carter writes (quoted by Chris Bogart):

> In a dictionary words are defined in one or two sentences, but for
> guaspi these sentences are considered to be merely a learning aid.
> The effective definition is a set of lists of thus-related referents.
> For example,  the referent set of ``eats'' includes a
> member list with our example rat  in first case and our example cheese
> in second,  as well  as  numerous  other members containing other rats,
> foods, and so on ad (almost literally) infinitum.
> Other predicates  like cu,pair, have referent sets
> that  are  actually infinite.

This definition doesn't work for Lojban/Loglan, and in fact I have suggested
to Carter that it is buggy in general (see the file "cowan" in the guaspi
directory on www.math.ucla.edu).  "x1 has a heart" and "x1 has kidneys" have
the same referent sets (neglecting partly dissected animals, etc.).  But we
don't want to call them the same predicate.

> When you speak  an  argument  in  a nonsentence you call the
> listener's attention to its referents.  For example,
> 
> ^:i |va -jiw /vn -sper -jiol        {Hey, a crocodile!}
> 
> When you speak a sentence or a subordinate assertion you do the
> same thing: you call the listener's attention to the members of its referent
> set.  (Thanks to John Parks-Clifford, editor of {\it The Loglanist}, for this
> insight~\cite{TL43}.) Thus in:
> 
> ^:i |qnu !qo -jan /tara /jun !kseo |zey !ju
>             {John, the rat is after your cheese!}
> 
> your knowledge of the referent set of \trw-jun,hunt, includes a
> member which John will want to append to the ones he knows, before the cheese
> is stolen.  This is the ultimate meaning of the \guaspi\ sentence.

The second half of this works all right for Lojban/Loglan, but the first half
applies only to Loglan and -gua!spi, since the Lojban form for "A rat!" is not
"lo ratcu"/"pa ratcu" but simply "ratcu".  (In Loglan, that's an imperative,
and in -gua!spi I don't know what it is.)

> A guaspi sentence or argument expresses a relation between specific
> referents, and this specific referent set member is called an ``event''.
> (Frequently the sentence represents several similar events.)

I don't know whether Lo??an can accept this definition or not.

-- 
John Cowan		sharing account <lojbab@access.digex.net> for now
		e'osai ko sarji la lojban.