TECH: experimental cmavo "xo'e"

[bob@GNU.AI.MIT.EDU]

Andrew Rosta <ucleaar@ucl.ac.uk> says

   In the case of, say, _judri_:

       judri  x1 is address of x2 in system x3

   putting xo'e in the x3 place means *not* that the address is not part
   of a system, but that it is unspecified as to whether the address is
   or isn't part of a system.

Good point!

   So, _xo'e_ really ought to be used frequently: each time one uses a
   brivla, for each sumti slot one should think "do I really want to
   claim that this sumti exists?"

   This is an annoying conclusion, ... All those 'by standard' places
   could be dumped, for example.

This is an _interesting_ conclusion.  Suppose a future community of
Lojban speakers start to use "xo'e" frequently.  This means that the
present notion of how gismu should be created is wrong.  Better to
create predicates with fewer places, and then to use `added' places as
needed.  This would bring Lojban much closer to natural languages that
use prepositions and case endings.

I think we should continue to construct gismu as we do now.  The
concept embedded in a gismu is supposed to be irreducible; all the
places are needed.  If I use "xo'e", you should consider the
possibility that I am speaking in a non-sensical manner.  The
presumption is that you have to have a maker if something is
made/assembled/built.  Likewise, the concept of an address is supposed
to include the notion that it is in some sort of system.  (Of course
the system may be trivial or obvious, but in that case, you can elide
the place or use "zo'e".)

Aristotle was wrong to argue that complex entities necessarily imply a
teleology; but he was not stupid.  The world is such that I can look
out my window and see a complex entity `made' without a maker, but it
took centuries of discoveried and thinking before anyone figured this
out.  Indeed, I don't think that you should use "zbasu" without the
`maker' place except occasionally; and you certainly should not speak
of a tree being made (except by microbiologists) except in the context
of teaching people that trees are instances of complex entities that
that appear without a teleology.  (In other words, the default
presumption that made things have a maker should be kept...and trees
do not fall into the category of made things.)

"Mark E. Shoulson" <shoulson@ctr.columbia.edu> says:

    ...what does "x1 makes x2 period" mean?

Maybe this utterance does *not* mean anything!  Perhaps everything
made must be made/assembled/built _out of_ something....or maybe not.
I certainly cannot think of anything made (in contrast to designed)
that is not made _out of_ something, whether it be waves of light or
atoms or gravitational fields or the emotions of a crowd....

Bear in mind that when we English speakers think of the places of a
gismu, we think of things that are easily added or deleted from the
concept.  Places are not primary to us English speakers.  But to a
fluent lojbanist, to speak an utterance without at least an implicit
place (as with "zo'e") ... as awkward as speaking a sentence without
tense.  It will take conscious effort.

When you use "xo'e" you are claiming that a fundamental understanding
is somehow wrong.

All this makes for yet another aspect of language to look for in the
communication habits of fluent Lojban speakers.

    Robert J. Chassell               bob@gnu.ai.mit.edu
    Rattlesnake Mountain Road        (413) 298-4725
    Stockbridge, MA 01262-0693 USA