Re: ?Naro cel ei nau cepoa sia? ['naru,gil enQ,gibua'Za]

--- In conlang@yahoogroups.com, Shreyas Sampat <ssampat@j...> wrote:

 > On the other side of the sword, I disagree that they behave like
 > distinct phonemes, based on my loose understanding of the idea of
 > phoneme.  My grounds: Taot Toua and Tao Ttoua both result in |tt|
 > behavior, regardless of morpheme/syllable boundaries.  This suggests to
 > me that |tt| is a cluster, rather than a phoneme in its own right.  Were
 > it phonemic, I'd expect to find a distinction between the two
 > pronunciations (maybe [taut.toa] vs. [tau.toa]).  Compare English
 > "guano" /b&t.SIt/ vs. */b&tS)It/.
 > Forgive me for being non-rigorous, as usual.

That makes a whole lot of sense, and reinforces the Tao Ttoans'
view of their language.

However, the case is slightly different for the prenasalised
stops.  |Nt| is pronounced [nd] medially but [d] initially.  When
|n| + |t| come together through phrasal sandhi, no medialisation
takes place: It's pronounced [nt].  However, if |n| + |nt| come
together, the result is [nd] (basically, the final pronunciation
of |n|, [n], plus the initial pronunciation of |nt|, [d]).

So the prenasalised stops are closer to phoneme status than the
geminate stops are, as far as I understand it.  Then again |nt|
is always [nd] within a word, [nt] doesn't exist medially...

Thanks for your thought-provoking input.



-- Christian Thalmann