Re: Length

Nik Taylor wrote:
>
>Fakaltin=E1su:
>> So, any thoughts?
>
>Forget it, I just made up my mind, I'll go with doubling.
>
>Still, any comments on this change?
Well, even though you already made up your mind, I still=20
wanna share something from Boreanesian which your options=20
#2 and #3 reminded me of. (These options opted to represent=20
length through a syllable final <h> or <k>).

In Boreanesian, all major syllables (with primary stress)=20
are either long vowels or rising diphthongs. But I have=20
analysed the long vowels as being derived from diphthongs,=20
and could therefore be represented by a syllable final=20
consonant. Let me explain:

There are three approximants in Boreanesian: /j/ palatal,=20
/w/ labial, and /3/ velar. Combined with the four vowels=20
of Boreanesian, we get the following grid of diphthongs=20
and long vowels:

     /aj/    /aw/    /a3/
     /@j/    /@w/    /@3/
     /ij/    /iw/    /i3/
     /uj/    /uw/    /u3/

The palatal approximant /j/ is essentially a nonsyllabic=20
version of /i/ and together creates a long vowel: /ij/ ->=20
[i:]. Similarly, the labial approximant /w/ is essentially=20
a nonsyllabic version of /u/ which together creates a long=20
vowel: /uw/ -> [u:]. The velar approximant is a nonsyllabic=20
version of centrality and together with the central vowels=20
/a/ and /@/ creates long vowels: [a:] [@:].

Other combinations are simply diphthongs.

So vowel length can be represented, depending on the vowel,=20
by different approximants. I'm not sure of the Romanized=20
orthography, but this is basically how I'm representing=20
Boreanesian in the tentatively practical IPA orthography.

-kristian- 8)