"Lojban adjective"

["Mark E. Shoulson" <cbmvax!uunet!cuvmb.cc.columbia.edu!shoulson>]

>Date:         Fri, 14 Feb 1992 10:42:00 EST
>From: "61510::GILSON"
 <gilson%61510.decnet%CCF2.NRL.NAVY.MIL@cuvmb.cc.columbia.edu>

>>Bruce writes:
>>>You mean in Lojban you can't modify a proper name by a Lojban adjective? You
>>>can't talk about "the first Elizabeth" or "the most populous Springfield in
>>>the United States"? This seems to be a weakness of the language if so.
>>Watch it.  What is a Lojban "adjective"?  Lojban has no adjectives.  The
>>part of speech doesn't apply.

>Sorry. Let me explain what I mean. A sumti may modify another. If I say (sorry
>for reverting to Old Loglan, but I do not know the modern Lojban words) "le
>narmi glida grupa cefli," as JCB did in the very first passage that appeared
>in the Loglan article in Scientific American in 1960, we may agree that the
>role of "cefli" is different from the other three predicate words. It is a
>head in an AH construction, a noun in English terminology. The other three

A sumti may modify another using such contructs as "pe" and "ne".  You are
dealing with the case of a *brivla* modifying a *brivla* (or mre generally
a selbri on a selbri). For a simpler example, "le ctuca cukta" (meaning
irrelevant here).  It's *not* a case of "cukta" acting as a noun, and
"ctuca" as a modifier on that noun, i.e. an adjective.  Rather, "cukta" is
a brivla, which I find actually closer to a verb than a noun.  "ctuca"
modifies that brivla, and the whole construct is now still a selbri, which
is more verb than noun, to me.  Only *then* is the "le" tacked on to the
beginning, sumtifying the whole thing and making it a "noun".  Yes, this is
nit-picky, and it does wind up with basically the same effect, of a
modified noun which can be glossed, usually, as an adjective-noun pair.
The point is, though, that And's perception of cmene as classifiers
notwithstanding, a cmene is not a brivla and thus cannot participate in
this whole party.  The problem you dislike, as I said in my last letter, is
the fact that cmene cannot (without sneaky loopholes) be part of tanru, and
thus modified in this tanruic, "adjectival" fashion.  On the whole, it
probably would, as you say, be nice if they could, giving us such things as
"Elizabeth the First" as a unit, as opposed to "Elizabeth who is First".
Prescriptivally (oog, what creative spelling), those two uses, {*la pamoi
.elizabet.} and {la .elizabet. poi pamoi} should be equivalent, but I would
really look at them as being different, from a sort of stylistic point of
view (Nick, can you clarify?  I don't know stylistics.)  After all, tanru
just show a relationship which can always be clarified, and {poi} expansion
on a sumtified tanru is often very natural and reasonable.  And even in
English, if someone looks at you funny when you say "New York" you can
always explain "yeah, this is the "new" one.  There's an older York in
England."  So I guess I agree with you that I'd like to see that construct.

Trouble is, I can't think of a way to do it that doesn't mangle the
morphology beyond all recognition.  Only thing I can see that works is {la
pamoi mela .elizabet.}, which is, as I said, sorta ugly.  Hmmmm.  Maybe I
could get used to it.  {la cnino mela .iork.}?  I'm not so sure.

>are attributes in successively nested AH constructions, or adjectives in
 English
>terminology. I maintain that although a _word_ in Lojban is not a noun, a verb,
>or an adjective, we can still talk of its usages in those terms, defining a
>"verb" as the word used as the main selbri in a sentence/clause, a "noun"
>as any of the words filling the sumti-places of that word, and an "adjective"
>as any sumti that modifies another in an AH type construction.

Yes, I often think of sumti in terms of nouns and brivla as verbs and tanru
as adverb/verb or adjective/noun or noun/noun combinations, but when
dealing with Lojban syntax (as opposed to semantics), you have to keep
straight just how each word is functioning from a Lojbanic point of view.

~mark (shoulson@ctr.columbia.edu)