|
krtisfranks wrote:
> I have four quibbles with the definition as-presented.
>
> (1) A small one: I dislike the "x_1" appearing here. I would just use
"x"
> or a description such as "the referent of the immediately following
> construct". Not a big deal though.
>
> (2) Is that construct a selbri or a sumti or what? I think that "sumti"
> makes the most sense as an answer (because it covers things as diverse
as
> "lo mlatu", "li ci", "lu mi klama li'u", and more), but I would then be
> unsure of the selma'o for the word.
I was not sure how to make the wording succint and simple. As the selma'o
(NAhE) and use of the expression "predicate modifier" implies, gu'y is a
prefix to a tanru-unit, like SE cmavo, although it can also be used in
NAhE-BO constructions governing a sumsni. I could have opted for a mere SE
selma'o, but I thought that allowing for the NAhE-BO construction would
allow for some more flexibility in usage.
Here is how I envision it being used:
• gu'y bakni = "x1 is a herd of cows (of species x2)"
• lo gu'y bakni = a certain herd of cows
• mi pu viska gu'y bo le do bakni = "I saw the herd of your cows".
>
> (3) When it says "composed of x1", it means "composed, at least in part,
> of x1" – in other words, the sum is involving two input, at least one of
> which is x1, and the other of which may not be specified (but may be x1)
–
> yes?
>
The definition wording is assuming the semantics of Xorlo plural
variables/constants, i.e. in "gu'y bo le do bakni", "le do bakni" would be
a plural constant (which can be understood as a covert set) among which
all your cows are. But the particle can be adapted to CLL lect by applying
gu'y bo to whatever type gunma's x2 is expecting, be it a mathematical
set (le'i bakni…) or whatever is the standard in the dialect at hand.
> (4) I dislike the fact that this word ends with "y". To me, that implies
> that it is the letteral for "gu" (which would be a relatively fine
> letteral to have). I actually dislike "y" occurring in almost all except
a
> special class of cmavo, but I would prefer all "y"s to be strictly
medial
> unless they are for letterals. Also, out of curiosity, what is the
> etymology for this choice?
The word is etymologically based on gunma; I, too, am not overly fond of
y-final cmavo, but I thought it would be of a greater phonological
compatibility with gunma (with an unstressed final -ma) than gy'a for
example (it is less easy to justify how a stressed -u- would transform
into a -y-). The Lojban language already mandates that speakers should be
able to discriminate unstressed -y- from unstressed -a-, compare for
example cakyREspa with ca ka REspa, or mantyGAPci with a
hypothetical mantaGAPci.
I could reassign the cmavo to gy'a, but this won't remove the fact
Lojban already requires the aformentioned -y/-a distinction.
|