lo selma'o cu mo

[CJ FINE <cbmvax!uunet!bradford.ac.uk!C.J.Fine>]

I had the idea a few weeks ago of trying to write a cmavo dictionary in
Lojban, which has led me to some thoughts about the meaning of "cmavo".

I was using the baselined gi'uste, wherein the x2 of "cmavo" is its
"type" or "class", so I got into difficulties thinking what (in Lojban)
would go in the x2.

".a" is a cmavo of selma'o A ...

zo .a cmavo ma

I first thought of
zo .a cmavo zo .a

but "zo .a" is a word, not a class. I settled on the rather
unsatisfactory device of making names for the selma'o, and noted that I
would post the question at some time. This allowed me to write the
following:

ni'o zo .a cmavo la .ac noi rebla ke sumti ke logji jonma'o ku'o lo
vlina bele lampru sumti beile lamba'i sumti

".a" is a cmavo-belonging-to Ac (which-are tail kind-of argument kind-of
logical join-words) [meaning] the alternation of the preceding sumti and
the following sumti.

[I wavered between "noi" and "no'u le'i"].

ni'o zo .a'a ci'orma'o cmavo la .uic noi nalte'i galma'o ku'o ledu'u le
cusku cu jundi

".a'a" is a emotion-word cmavo belonging-to UIc (which-are non-specific
modify-words) [meaning] the expresser is attentive

Then I looked at the summary of the recent place-structre changes (not
yet baselined). cmavo is now
	"x1 is a structure-word exemplified-by-word x2 with meaning-function
		x3 in language x4"

I can see why this change has been made, and incidentally, it solves my
first problem!
	zo .a cmavo zo .a
	.i zo .a'a cmavo zo .ui
Simple!

But now I have problems with my incidental clause

ni'o zo .a cmavo zo .a noi rebla ke sumti ke logji jonma'o ku'o lo vlina
bele lamprusumti beile lamba'i sumti

This is still true (ju'oje'i zo .a rebla ke sumti ke logji jonma'o) but
I don't think it is what I wanted to say. (It is certainly not what I
intended to say). My incidental clause was meant to qualify what I have
always thought of as the selma'o - the class of cmavo.
[If I used the other form I considered - "no'u le'i" - it would be
clearly incorrect. zo .a valsi gi'enoiseni'ibo klesi]

There are three things that concern me here.

One is that I am not sure I can easily say what I want to. I think I may
be stuck with
	zo .a cmavo zo .a noi selma'o lo rebla ke sumti ke logji jonma'o
		ku'o li'o
which is a bit long-winded.

Secondly, I do not know the extension of "lo selma'o be da"
Is it true that
	zo .a'a cmavo zo sa'e
as well as
	zo .a'a cmavo zo .ui

May I if I wish start distinguishing lo cmavo be zo .ui from lo cmavo be
zo mu'a, since I regard them as distinct semantic sets, even though they
are syntactically identical?

Third, I suspect we need to watch our use of "selma'o" in English
discussion. "A cmavo of selma'o PA" is OK, as you can think of selma'o
there as meaning "pattern"; but I have certaily thought of "selma'o" as
meaning "ma'oklesi", and I would be surprised if we haven't used it that
way in discussions. (I haven't looked)

The last problem is a little akin to one somebody posted recently about
letterals - is there a way of referring to "the symbol C" as opposed to
"the sumti I am referring to as cy"? When I say
	zo .ui selma'o zo .a'a
is "zo .ui" a word, a symbol of no intrinsic meaning, or something else?
It seems to me that it can only be a word (which is not to say that it
is necessarily being used for its meaning), in which case "selma'o"
means "is a word which is the pattern for cmavo x2 ...".

I hope this rather rambling disquisition is understandable.

		co'omi'e kolin