Organic quality of language

[cbmvax!uunet!math.ucla.edu!jimc]

> Date: Mon, 29 Oct 90 11:19:12 -0800 (PST)
> From: Michael Urban <cbmvax!uunet!monty!rand.org!urban>
> Subject: Re: response to J. Prothero book review and comments of 12 Oct 90
> 
> ...
> Linguistic tinkerers like the Idists underestimated the organic quality of
> Esperanto, or of any living language.  Indeed, one of the valuable aspect of
> Lojban or Loglan, if either ever develops a substantial population of fluent
> speakers, will be to observe the extent to which the common usages of the
> language diverge from the prescriptive definitions.  Such effects will, I
> think, be easier to isolate and analyze in a language that was created `from
> whole cloth' than in an a-posteriori language like Esperanto.

In -gua!spi I feel the same force.  Looking forward to eventually using
a program to "understand" live (human-generated) -gua!spi text, I have
a policy that everything must be analysable into primitive predicates
(gismu) and their arguments, because such a program will have a hard
enough time "learning" 800 to 1200 gismu, much less 20,000 lujvo with
subtle distinctions.  When I translate "dunk" into "push-under-water"
as a compound, it is obvious that the compound, backmapped into
primitives suitably linked, represents well the meaning of the English
hyperspecialized word.  But take something like "rage".  The closest
compound I can make is "animal-anger" as a metaphor, not a lawful
compound, and the best rendering I can make of English "I rage about
<event>" is "My animal part is angry about <event>".  

Not that I can actually produce such a phrase assembly using standard
 -gua!spi compounding rules.  In some situations the required meaning is
too complicated for simple rules to support it.  What are we to do in
that case?  One possibility is to require the speaker to not use
compounds, to say the intended phrases in full.  -gua!spi words are so
short that such a requirement is not onerous.  Another response is to
imitate English and French, producing a gismu for "rage"; rage is a
very fundamental emotion that differs from anger and a gismu would be
quite reasonable, but where do you draw the line?  

And this translation does not have the "feel" which I am used to in
English.  Am I being Franco/Anglophilic, being used to a separate word
for everything? Or is something really lost, a la 1984, when you reduce
specialized words to equivalent gismu assemblies (rather than a Lojban
lujvo with a meaning possibly independent from the component rafsi)? 
As a metaphor from another area of art, compare the pieces for brass of
Gabrieli with the vocal motets of Lasso; brass is more rough and, well,
brazen than the wimpy voice, yet clearly it is not correct to say that
either one is unsuitable; each has merit in its own way.  Similarly,
English with its large vocabulary and -gua!spi with its simple encodings
each have their own merits, and it is not proper to reject the -gua!spi
style simply because it isn't English.  In fact, one might decide that
"animal-anger" helps the speaker better understand what rage really is.

But is -gua!spi (or the gismu assembly approach) adequate to represent all
required meanings?  How could one decide such a question?  

		-- jimc