RE^whatever:literalism

[pycyn@aol.com]

lojbab:
<<, if a
>lujvo works, then the fact that it is not literal should not count
>against it

I don't think it will, IF it is "required" per your above.  But given two 
tanru for a concept, one literal and the other non-literal, I think the 
literal will win every time.>>
I don't think it should count against it even if it is not required.  As
for your empirical claim, my sense is that non-literal ones, if
striking, have done rather well.  It is hard to tell now, of course,
because in the course of being successful they change the concepts
involved and become, in consequence, literal.  But "blade hammer,"
to overwork a good example, was non-literal when coined.

<<>^robber mammal^ is not the same as ^mammal^ or ^robber^, but it
>is already implicit in ^mammal^

What is?>>
^robber mammal^, as the English says.

<<"robber-appearance-mammal" makes "robber-mammal" 
explicit.  But why is the latter "new" if it applies to raccoons, but the 
former is not?>>
"robber-appearance mammal" is excluded on the ground that it is
culturally restricted -- I don't think that African or Asian (maybe
even European) robbers archetypically wear dominoes.  It is
"robber behaving mammal" if you will (see what I mean about the
confusion that comes even with literal tanru). That is admittedly
anthropecentric, but I think allowable nonetheless.  However, taken
literally, it applies to most mammals -- humans, surely, rabbits,
gophers, bears, wolves,  and so on.  To make apply to raccoons
specifically requires that something have a special meaning in this
context (I suspect it is "robber" but "mammal" and even "behaving"
may be involved).  

<<I am not the least comfortable with washer-bear, nor did I see sign that 
anyone else was.  Someone pointed out that it was another source language's 
tanru for racoon, not that it should be the Lojban tanru.>>
Sorry, maikl did not say explicitly that I noticed that "wash bear"
was from the German, so I took it as his suggestion in preference to
"wash cat," then cowan said the resulting Lojban lujvo might work. 
Relative to the flat rejection of "wash cat" these seemed
comfortable.  We now seem to have a vote for "dog" as the base ,
which I don't see, myself.  (Someone -- xod? --mentioned a while
back how much personal; aesthetics got wrapped into these and this
may be a case.)

<<>Of course, I like it because it clearly
>opens the way for otters and weasels (and skunks, regularly cats in
>English -- ahah! is that the prolem?).

No.>>
Not quite sure what is being denied: that the problem is that skunks
come into it somehow or that parallels to English usage come into
it somehow.  From other remarks it is clear that parallels to English
usage do come into it, as reasons for objecting to tanru and lujvo,
so I suppose you mean that skunks aren't relevant.

<<Borrowing for plants and animals would almost always use the Linnean 
name.  Lojban does have a cultural bias towards scientism %^)>>
The actual tendency, insofar as there is one, seems to be toward the
native word in the home area -- see Cowan's comment in the
present case and a mass of food words a while back.  That is also
recommended in The Book.

<<Because we are European language speakers and know our own language and 
culture.  We *assume* we are biased towards European metaphors, so we 
eschew them.>>
Guilty until proven innocent, with a guarantee that the latter proof
will not be allowed to come forward -- since the form has already
been rejected.  Not a great plan and a long way from "let usage
decide": it can't decide in favor of something we are forbidden to
use.

<<>But, as these
>threads bear witness, "perfectly literal" tanru and lujvo mislead
>people all the time.

No they don't.  The mean exactly what the literal tanru/lujvo seem to 
mean.  That meaning just doesn't seem to coincide with some people's 
primary meaning for the English term being translated.  the solution of 
course is to realize that there WON'T be a one-to-one mapping of English to 
lOjban words.>>
Hey, we just got boggled on an ENGLISH tanru (why are modifier-
modifed tanru not disallowed, BTW, since they follow a common
Englis pattern?), imagine what happens with tanru in languages we
don't know so well.  My experience is that I forget one possible
way of combining components -- often, embarassingly, the one
most like English (I expect a foreign language to be foreign) -- and
that turns out to be just what the creator had in mind.

xod:
<<Do we have a test for determining which category a tanru falls
into?>>

Roughly, does it pick out the right thing used literally; if not, then it
is non-literal.  Surprise!

<< le tsali lojban >>  this does not seem to parse, though it is clear
as English.

<<I suppose a non-literal tanru is a X type of Y which is not really a Y?

This is context-switching. Inasmuch as the non-literal context is assumed,
it really is a Y. If you must be a stickler, it's not a Y and the whole
tanru falls apart.>>
Or an X type of Y that is not really Xish or one that isn't really
either or....  What context?  In use, I suppose that the literal context
is assumed and the effectiveness of the non-literal case comes
precisely from the leap -- or fall -- to understanding.  How can a
tanru fall apart?  I suppose that the worst that can happen is that
they miss their intended target and so end with the hearer thinking
something false or falsely said.  As noted, this is not a peculiarity of
non-literal tanru.

<<na'e banzu xamgu mu'u le du'u na kakne casnu la lojban bau la lojban >>
Metalinguistic uses are always the last to come: look at Aristotle
struggling to do it for Greek or Ockham for Medieval Latin
(another artifical language).  

pier:
<<lo'u lumci ke pu'o gerku li'u .i e'u mi'o te cmene ra 
lu lumci gerku li'u
a zo lumge'u
a zo porkiono
a zo mabrlprokiono>>
Does "pro" really mean {pu'o} here?  I'm not sure what it means,
but this seems odd.  Is {mabrlprokiono} legal against
{mabryprokiono}?  I.e., is fuhivla rule different from lujvo at this
point?