[YG Conlang Archives] > [ceqli group] > messages [Date Index] [Thread Index] >
Rex May - Baloo wrote:
>
> on 2/27/02 9:50 PM, Mike Wright at darwin@hidden.email wrote:
>
> > Rex May - Baloo wrote:
> >>
> >> I forgot to say, I'm willing to abandon bdomen for beli, but of course kofi
> >> doesn't fit the phonology.
> >
> > Took me a while to understand this point, and the reason for it. I'm
> > beginning to adapt, but it feels very strange--reasonable, but not intuitive.
>
> Yes. Self-segregating morphology was originally a Loglan idea, and was
> initially effectuated by allowing a limited number of word shapes, mainly
> CVCCV, CCVCV and CV(V).
> Ceqli might not exist if my recommendation to change those rules to the
> Ceqli system had been considered and adopted.
>
> Interestingly, Mandarin seems to have SSM itself, accidentally, as a
> by-product of the fact that virtually all morphemes are monosyllablic
> (morphemes, not words, tho I realize that gets a little fuzzy sometimes in
> Mandarin),
Well, "word" tends to be used in ways that don't reveal the
morphological structure of the language, but it's comfortable. On the
other hand, the native terms, <zi4> (monosyllabic character) and <ci2>
(polysyllabic lexical item) don't always work, either. Educated
Chinese tend to overemphasize the importance of the written language
in analyzing syntax. I'm sure you've seen me fighting about this on sci.lang.
Notice that Mandarin, Vietnamese, Japanese, Korean, Malay, and all the
Malayo-Polynesian languages (as far as I know) get by with very
limited syllable shapes. This is what makes all these languages
relatively easy to pronounce (if tone and register are ignored). As
I've said, if I were designing a conlang, I'd go with something like
Japanese, Malay, or Italian. However, that doesn't fit with the SSM
approach. (Chinese aren't necessarily analytical about morphemes in
their daily speaking. In fact, they don't have to be able to analyze
words like <jing1ji2> "economics", and most are not aware that words
like <bo1li4> "glass" and <hu4die4> "butterfly" are not monosyllabic
morphemes at all.)
Native Japanese, Korean, and Malay morphemes can be either
monosyllabic or polysyllabic. Korean has a bit broader selection of
syllables than Japanese, since it can have final consonants, as can Malay.
All these languages have a problem with matching the sounds of loan
words, Japanese being hilariously horrible at it. I guess that would
be a problem for Txeqli, if you want the loans to bear a fairly close
resemblance to the originals.
Even Mandarin morphemes are not unchanging atoms of sound. They are
subject to stress and even to tone changes. Hokkien even has lots of
contractions, such as /sim mi~? laN/ (literally, "what person?") ->
/siaN/ (who?). Mandarin has only a handful of recognizable
contractions, but they are there.
> tho I suppose it's possible to comfuse jin yi with ji ni.
Certainly not for a native speaker, since each syllable also has a
tone and a stress level. Even two high-level tones in a row may not be
at exactly the same pitch. Also, the /i/ in <jin> is closer to [I],
while the one in <ji> is clearly [i]. So, we can say that Mandarin /i/
has two allophones, [I] and [i], which occur in different
environments. In fact, there are two even more distinctive allophones
of /i/, [z-] and [z.-] (syllabic "z" and syllabic retroflex "z"), both
written as <i> in Pinyin, as in <si> and <shi>. These are always
conditioned by environment, and never form minimal pairs. That is,
you'll never see any two of {[z.-], [z-], [i]} following the same
consonant. The first follows only retroflex sibilants, the second
follows only alveolar sibilants, and the third follows all others,
except for velars {/k'/, /k/, /h/}. Natural languages are full of
"rules" like this that develop and shift over time. The vast majority
involve the natural flow of movements of the vocal aparatus, combined
with what can only be described as laziness. People just naturally
take the easy way out. That's why you have things like
schwa-buffering. Too bad language development is a chaotic system, so
we normally only know about these things in retrospect. Anyhow, that's
why Yankees talk funny.
--
Mike Wright
http://www.CoastalFog.net
_______________________________________________________
"When they wired us humans up, they really should have
labeled the wires--don't you think?" -- Ed