Message-Id: <199310100923.AA15835@eli.CS.YALE.EDU>
Date:         Sun, 10 Oct 1993 05:21:59 -0400
Reply-To: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: Logical Language Group <lojbab@ACCESS.DIGEX.NET>
Subject:      litru and klama
To: Erik Rauch <rauch>
Status: RO

What I think I'm saying is that an asymptotic route is better expressed by
litru than by klama, since it has no destination.  As to how to express it,
we have the motion tensors such as approaching, passing by, orbiting, etc.

My argument on metaphorical use of Lojban words is this:
1) it seems to me permissible to stretch the meaning of a gismu reasonably far
as long as all elements of the place structure remain intact under the
stretching.  There should be an relationships between each of the places and
the general concept that correspon with the "pure" form of the meaning, i.e.
the one intended when the word was chosen, and all of these relationships
should be CONSISTENT.
2) tanru and lujvo have to similarly abide by place structure rules.  If a
tanru is made, you cannot introduce new places that are not in the source
gismu, and the governing place strcuture is that of the final term - all
other places from other terms are added using be/bei.  This means that you
must have some heavy literalism, and added terms for some types of metaphorical
stretching.
   lujvo, whether you follow Nick's rules exactly, or not, should have places
at least suggested by the source gismu.  I have argued that there may be
some cases where additional places are warranted that are not in the source
gismu, but my current feeling is that what this should represent is that you
are implicitly leaving out a clarifying term of the compound that WOULD bring
in these places (like for example leaving out the mei/moi in interpreting
number rafsi).
3)tanru that are based on analogy are probably permissible, in that you
could always add a term like "simsa" that would clarify.  Thus, even if you
aren;t sure that mind-out-of-body travel is a litru, it is surely a litru-simsa.
4) the thing you must always be wary about in making analogies, leaving out
terms from your lujvo, etc. is the cultural factor.  You should invent new
concepts based on your audience, and if that audience is of a different
culture, metaphor that is not extremely obvious will be unintelligible.  At
this point, almost all Lojban is written either for the nonce and hence can
be (and is) quite sloppy, or it is being arhcived where it might be read by
anyone, and hence has an audience who is culturally indeterminate.  Culturally
dependent metaphor is thus extremely risky.
5) Related to 4, is the question of quanitity.  If you write a long text and
throw in one metaphorical usage throughout the text, with good contextual
support that clarifies the metaphor, this s probably fine.  If your sentence
has two or three metaphors, and all surrounding sentences have metapjorical
elements, you will have problems.  Michael Helsem's Lojban poetry has always
been almost unintelligibl to be because he relies almost totally on metaphor
to achieve his effects.  I cannot figure out what his metaphor has to do
with the context (which is especially sparse in poetry anyway), and hence it
comes across to me as randomly strung words (complicated by multitudinous
grammar errors ans style experimentations that force a literal translation
to English follwoed by concentrated interpretation in order to figure out
anything about what he is writing).

But no, I am NOT opposed to metaphor in general.  It just must be constrained
by the features that make Lojban a Loglan.

lojbab