Re: [lojban] Re: mi'e

[Jorge "Llambías" <jjllambias2000@yahoo.com.ar>]

--- John Cowan <cowan@mercury.ccil.org> wrote: 
> It's far from easy to design a morphology that is self-segregating at
> both the word and the morpheme level while still allowing a modicum of
> naturalness in the morphemes, and I know of no conlang except Loglan/
> Lojban that even attempts it. 

What about something like this: every morpheme consists
of one or more consonants followed by one or more vowels.
A {y} can be optionally pronounced between consonants, and
a {'} can appear between vowels.

So for example {ba}, {ba'a}, {bra}, {bba} are all morphemes,
the last one can be pronounced /byba/. Morphemes that start
with a single consonant are cmavo and the rest are gismu.

For example, {barda} could be {bra} and {gerku} could 
be {grku}, with additional possible pronunciations
/gyrku/, /gryku/ or /gyryku/.

> (Gua\spi bypasses the problem by having
> no multi-morpheme words and no tanru/lujvo distinction.)

The difficulty for Lojban came mainly from the requirement
that gismu had to have CCVCV or CVCCV form, which is nice
enough for gismu, but everything else was patches upon patches 
over that. 

To distinguish tanru from lujvo all we would need is a separator
cmavo. With the above morphology, we could have {bragrku}
for {brage'u} and {brake grku} for {barda gerku}, for example.

And this morphology doesn't use stress at all to segregate
words.

mu'o mi'e xorxes




__________________________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo
http://search.yahoo.com