Ambisyllabic (was: Romanized Orthography of My Conlang)

Eric Christopherson wrote:>
>Is the /p/ merely non-aspirated, or is there no release as well? I =
remember
>the discussion long ago about how English final unvoiced consonants (in =
some
>dialects at least) seem to have no release -- I initially thought my
>idiolect only did it with /t/, but it appears to hold true for my /p/ =
and
>/k/ also, now that I think about it. So I would pronounce <hap> [h{p_}] =
(for
>the X-SAMPA impaired, { is "ae ligature" and _} means "no audible =
release"),
>but I can't tell if I pronounce <happy> [h{pi] or [h{p_}i].
This is perhaps due to a phenomenon known as ambisyllabicity: The=20
simultaneous presence of a segment in two adjoining syllables. In=20
English words like 'happy', it is indeed difficult to say how exactly=20
one should divide the word up into syllables; [h{.pi] vs. [h{p.i].=20
Normally, /p/ in English is aspirated at the onset of stressed=20
syllables. But in 'happy', the /p/ is not aspirated and is explained=20
by assuming that /p/ is simultaneous in both syllables. Ambisyllabic=20
consonants are preceded by a stressed syllable, compare; 'happy', =
'city',=20
vs. 'upon', 'attack'. In words with an ambisyllabic consonant, it is=20
assumed that the stress attracts the voiceless stop into the preceding=20
syllable.

In another language, Danish, ambisyllabicity also exists. For instance,=20
/papir/ 'paper' is pronounced roughly as either [papiR?] or as [pApiR?]=20
(where [a] is a low front vowel and [A] a low back vowel). The=20
pronounciation of /a/ in the last case means that /p/ belongs to the=20
preceding syllable because /a/ is always pronounced as [A] before a=20
labial or velar consonant. But, on the other hand, /p/ must also belong=20
to the following syllable because it is aspirated. The consonant /p/ is=20
otherwise never aspirated in syllable final position. This paradox is=20
solved by saying that /p/ in /papir/ is ambisyllabic.

Ambisyllabicity is pretty neat, eh? Has anyone experimented with this=20
phenomenon? I've been fooling around with this idea of ambisyllibicity=20
for some consonants in Boreanesian. Notably after minor syllables. This=20
would make minor syllables seem even shorter and some sequences seem =
more=20
like consonant-clusters rather sequences of minor syllables. I have yet=20
to find the exact details though.

-kristian- 8)