Re: Romanized Orthography of My Conlang

Going back to the 22nd of October:
I wrote:
> Ed Heil wrote:
> > Oh, surely "reheat" is /ri$hit/ and "singing" is /siN$iN/ where $ is
> > syllable division, no?
>
> Perhaps, but it seems rather artificial to say draw syllable-boundaries
> that way.  The only way I know of to indicate objectively
> syllable-boundaries is that English vowels are short before
> syllable-final voiceless obstruents (/sit/ vs. /si:/ or /si:d/)
Well, I remembered something: English has a special rule for
syllabification, where intervocalic consonants go with the preceding
vowel IF that vowel is stressed, which explains why "happy" is [h&pi]
and not *[h&:p_hi], the allophonic shortening of the vowel and the
unaspirated form of /p/ are due to the fact that the word is syllabified
as /h&p.i/ rather than /h&.pi/ (also, /&/ normally can't occur in
syllable-final position)

So, /rihit/ is /ri'hit/ while /siNiN/ is /'siN.iN/, due to the stress.
Thus, it's probably true that /N/ can't occur before a stressed vowel
(which pretty much goes without saying since, as far as I can tell, /N/
only occurs in English at the end of roots and before velar consonants)

--
"It has been postulated that, given an infinite number of monkeys
bashing away at an infinite number of keyboards, we could eventually
reproduce the entire works of Shakespeare.  Thanks to the Internet, we
now know this to be incorrect." - Anonymous
http://members.tripod.com/~Nik_Taylor/X-Files
http://members.tripod.com/~Nik_Taylor/Books.html
ICQ: 18656696
AIM Screen-Name: NikTailor