From: John Clifford <sjepark@u...>
Subject: da blanu de

From my occasional drops-in on Lojban list (MyGAWD are they on the second
place of nitcu again/still?!)I see that things are stuck in the same cycling
rut. But I see one historical question arising anew (i.e., I haven't seen
it for several years) namely, why Loglan blanu was two-placed. So, because
i actually liked that feature and the related one, I offer a recap:
The background is a linguistic/philosophic discussion in the late 60's about
semantic primes. Two features of this slid into Loglan at the time: that
all predicates were inherently potential, becoming actualized only in
context (borrowed from Quine eventually, I think — or at least blamed on
him) and that all "absolute" terms were actually relative. It was not clear
to what they were relative and trying to work this out was what led to
Lojban dropping this feature (that and the fact that you and John did not
like it) (The potential meaning was dropped even in Instiloglan in practice
at least because no one could figure out how to tell when the context had
actualized a term and when not.) But the cases underlying the original
comparative blanu remain to be dealt with, e.g, that a blue house is usually
much less blue than a blue sky or a blue sapphire but more so than a blue
baby and so on. The classics are things like tall dwarfs and short giants
(OK, small elephants and enormous ants). The skipped second place was
normally taken to be the typical of the named class of the first term: a
tall dwarf was (quite correctly) one taller than the typical dwarf, and so
on. But this came in conflict with the usual elision variable "something,"
leaving everything blanu apparently (well, bluer than SOMEthing).
The new "by standard..." does not solve that problem, nor does appeal to
general paradigm cases. But people seem to get by just by using the rule
they use in English (or whatever), which is what the original insight was an
attempt to make explicit.


To clarify what I said in my posting to cogling ...

Cross-cultural studies of colour terms have shown that while there is
considerable variation in the number of terms (anything from two upwards), some
things are pretty constant. One is that however many colour terms a language
has, they occur in a specific order (Berlin & Kay, 1969). Thus if a language
has only three colour terms, they will be black (dark), white (light) and red; if it has five, they will be those three plus yellow and green, and so on.

The second point is that there is substantial agreement on prototypical colours.
A speaker of a language which has a word for green but not for blue may describe
a blue object as "green", but their idea of a typical green will be the same as
for someone who also has the word "blue" (Heider (now Rosch), 1972, I think).

In other words, we don't need to worry too much about the colour gismu.

co'o mi'e robin.

    • These studies were mainly about hue. The issue with {blanu zdani} is largely about intensity (and whatever the other dimension is). In addition, I suspect some infection from the English difference beween "That house is blue" and "That is a blue house."

The ONLY thing we seem to agree about blanu is the relationship between the
x1 and the concept "blue", so a truthful place structure for blanu might be
"x1 pertains to some concept of the color blue".


Lojban is haunted by the ghost of a lost sumti place... In ''((lapoi pelxu
ku'o trajynobli))'' the following exchange occurs:

...''lu levi selsanga goi ko'i cu blanu za'e fe le dunra tsani be le
mexno dazyplu li'u .i ko'a bacru lu ko'i blanu fe la cart. poi ba'o
porpi li'e''

(where la cart. is Chartres cathedral).