[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
RE: [lojban] Why is there so much irregularity in cmavo/gismu?
At 05:43 PM 11/13/01 -0500, Craig wrote:
>> Oh, and I seem to rememeber you using 'xu' and '.ui' the last time this
>> came up. Is 'xu' being in UI the only thing that bothers you about se
>> cmavo? If you want to junk something, I'd hope there is at least more
than
>> one instance of it annoying you.
I missed this comment the first time, so I will respond now. The answer is
that xu and .ui are a particularly blatant example, but far from the only
one. I also don't think that lerfu should be a seperate cmavo when the most
common use is as pro-sumti - which should put them in KOhA.
Except that pro-sumti are not the ONLY use of lerfu, so limiting them to
KOhA would either be too restrictive or misleading.
>I mostly disagree with Craig's hatred of selma'o. He's railing against
>the whole system because of, apparently, a mistaken idea of their
>purpose.
I like the idea of putting cmavo in classes by function.
The problem with this is that we did not design cmavo or their classes
functionally. The language was designed to have an unambiguous formal
grammar. The original concept was that there were three parts of speech:
brivla, cmavo, and cmene. But an unambiguous formal grammar required that
cmavo be divided up, because some cmavo did not make any sense in some
positions.
I dislike how it
was implemented. If two things are not identical in their function within
the sentence (the ''se cmavo'') then they should not be put in the same
selma'o, e.g. xu, the attitudinal that has nothing to do with attitudes.
Easy - just call it "selma'o UI" and not "attitudinals". Coin a different
word for "attitudinal" as well as a different one for functional
class. But don't expect much agreement as to what the functional classes
of cmavo are, because we have in many cases devised new functions for cmavo
merely because they were grammatically permitted in some circumstances that
were not necessarily "designed" to make sense.
Furthermore, consider the other problem with UI. Is "(some UI) le broda cu
brode" asserting that 'le broda cu brode'? You definitely can't say yes and
include .e'o, .e'u, or many people's interpretation of .a'o - but you can't
say no and include .ui, zo'o, or .iu, to name a few examples. This is a
serious grammatical difference.
No. That is a a semantic difference. The structure of the sentence is
identical.
Maybe call one UI and one XU, or something,
because then it is possible to say, for instance, "all UI have the same
function, the expressing of emotion, while all XU have the function of
removing the assertive value of the bridi to express something about it
without having to claim it is true or nesting it inside of a ''nu'' phrase."
But there is no agreement as to which members of UI fill which "functions".
I would like to see the x2 of cmavo filled by the actual function of the
cmavo, rather than some arbitrary capital letters.
You want a different word, which associates words with their function as
opposed to their grammatical roles.
>However, I believe that from usage some selma'o will eventually combine.
>What is the grammatical distinction between ZEhA, ZAhO, FAhA, and PU?
The same as the distinction between UI and XU, in my hypothetical example.
If you look in the formal grammar, these clearly have a grammatical
distinction.
For example you can say
puza'o le broda cu klama
But if you try to say
za'opu le broda cu klama
the parser will insert a KU after za'o, and it will not be part of the
sumti tcita for le broda.
lojbab
--
lojbab lojbab@lojban.org
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA 703-385-0273
Artificial language Loglan/Lojban: http://www.lojban.org