Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

[NGL] Translation: Ku Valaksiya à Weksaforda

0 views
Skip to first unread message

Jack Durst

unread,
Oct 11, 1999, 3:00:00 AM10/11/99
to N...@onelist.com
Responding to Jerry's comment on whether sylables can divide including the
next consonant, a little linguistics may well help.

A sylable is formally defined as being based around its core, the vouel,
in every language yet discovered, sylables are based on the most sonorous
element and every sylable must contain such an element. in some languages,
this can include R, L, Y, W, even Z, but in most, as with the case of NGL,
it's a vouel or a dypthong.

After the vouel, the next most important element for the sylable is the
onset. Native speakers of all languages will tend to favor valid onsets
as a point for deviding sylables over all other possible places of
division. Least important of all is the coda, which is generally a
consonant, an off-glide, or a consonant cluster.

As NGL has no initial or final consonant clusters, This natural hierarchy
of interpretation provides an easy rule for dividing NGL words into
sylables. Since it's more important that a sylable have an onset than a
coda, when there are only single consonants dividing sylables they are
assumed, in the abscence of better information, to begin the next
sylable. So, {bokvaibaco} naturally devides into BOK-vai-ba-co.

This is where "in the abscence of better information" becomes a key word,
as NGL morphemes are not consistantly composed of sylables with onsets.
Given that one recognizes a morpheme, the logical thing to do is pronounce
the morphemes in their normal sylabic form (but de-stressed), so the
sylables devide on the boundaries of known morphemes instead of the
standard sylabic boundaries. {bokvaibaco} becomes BOK-vaib-ac-o (correct)
or BOK-vaib-a-co (incorrect, but substantially less likely a derivation
anyway.)

So, the answer to your question is "maybe". In a strict a-priori analysis
of a stream of speach given only the phonology, TIR-i-be is impossible;
but given that one knows the morphemes involved to be {tibe} and {-ir-} it
does become at least plausible as an interpretation.

However, the point is moot for the interpretation of anything except
poetry in the language as the sylable is totally irrelevant as a marker
for morpheme boundaries, which must be found by searching for recognized
substreams in the words, and there are *no* phonological conditions
applicable to consonants which are conditioned by the sylable.

Sincerely,
Jack Durst
Sp...@sierra.net
[this posting written in Net English]

0 new messages