Re: A Self-segregating morphology (was: Guinea pigs invited)

On 12/16/05, Gary Shannon <fiziwig@...> wrote:

>
> Vowels are irrelevant to the identity of the word, so
> when words are compounded the first vowel is changed
> to match the new word length: "galo" + "haki" =
> "golohaki"; "ki" + "kedaso" = "kokedaso", OR
> "kakadaso".
...

> With perhaps a few dozen "standard" vowel sequence
> patterns, possibly choosen based on intial consonant,
> the language could have a very melodic and varied
> cadence. For example, suppose that two-syllable words
> beginning in "k" use the pattern "au", four-syllable
> words beginning with "d" always used the pattern
> "oaiu" while four-syllable words beginning with "n"
> always used the pattern "aaui".
So long as we have separate consonant and vowel "tiers", why not
assign meaning to the various vowel melodies, to give us a "binyamin"
root-and-pattern system?  That is, the choice of a particular vowel
melody would neither be fixed nor arbitrary, but determined by some
semantic feature of the word.  The first one or two vowels would
indicate the word length, the remaining would indicate some sort of
meaning.

For nouns, say, it could be an innate semantic feature such as gender
or diminuation/augmentation, in which case the root would still be
learnable as a whole, although derivational sets (same consonants,
different vowels) would exist.

Or it could be a grammatical feature such as plurality, case, etc., in
which case the consonant and vowel tiers would need to be assembled
"on the fly", in a sense, as the sentence is constructed.

Now, if ease of acquisition and use is the game here, the first would
be the simplest choice; it would most likely be *easier* to acquire
than meaningless patterns.  The vowel pattern for a given
"round"-gendered root would be reinforced by the same pattern in every
other round-gendered root.

And/Or:

It seems like compounding is the main morphological game here, rather
than affixation, so why not a system of binyamin indicating the
semantic relationship between the compounded roots?  After all,
compounds will have a lot of unused vowel space.  We can argue all we
want about the relative semantic transparency of compounding vs.
affixation, but certain there's room for further transparency...
consider the different semantic relationships between "vitamin pill",
"detox pill", "pain pill", "horse pill", etc.  The vocalic melody
could take care of the difference between "X made of Y", "X that does
Y", "X for Y", etc. etc.

A fun puzzle to work on: make it recursive, so that you can compound a
root with another compound and not lose the additional semantic
information in the vowel melody of the second.


--
Patrick Littell

University of Pittsburgh
Fall 05 Office Hours: Friday, 1:00-2:00 by appointment
G17, Cathedral of Learning

CCBC
Voice Mail: ext 744
Fall 05 Office Hours: W 5:00-6:00, by appointment
Building 9, room 102

Replies