Re: Extra Syllabic Consonants

Chris wrote:
<<At various times I've read about extra syllabic consonants in
descriptions of languages...>>
<snip>

That's quite a Bella Coola word.  Like Ray said, though, I'd have to
hear it, because, quite frankly, I see potential nuclei all over the
place
(e.g., [K]).

Extra syllabicity comes in many forms.  Sometimes, though, it's a
way for phonologists to explain historical phenomena.  That is, one
of the credos of many modern phonologists (e.g., those who do
OT, but it's certainly not limited to OT) is that there *must* be a
synchronic explanation for everything.  Extra-metrical consonants
come into play whenever there's an apparent irregularity in the
stress system.  (I believe this is the story for Classical Latin...?
The
last mora is extra-metrical...?)

When it comes to conlangs, I'd argue that the historical explanation
is far more valuable than the synchronic analysis.  For example, in
Gweydr, trisyllabic forms are stressed initially:

(1)
(a) /'T&.li.n&/ "mother"
(b) /'pA.li.nu/ "animal"
(c) /'ko.ru.na/ "child"
(d) /'Si.li.n&/ "woman"
(e) /'bo.ze.nA/ "man"

There are a bunch of forms like these.  Then there's a word like
this one:

(2)
(a) /e.'zi.ni/ "sword"

There's a trick here, though.  The word for "sword" ends in an
orthographic "y" (i.e., a palatal approximant).  I suppose that
a synchronic explanation would have to do something like add
a couple ghosts consonants on to the end to push the stress over
(i.e., because a word that ends in more than one consonant is
stressed on the last syllable).  However, this explanation
is only important insofar as one wants a synchronic explanation.
Once one has it, then what?

Now, from the conlanging point of view, if there were a bunch
of words like those in (1) and only one or two like those in (2)
and there was *no* explanation whatsoever, I say that'd be rather
sloppy.  It would either look like the conlanger made a couple
mistakes, or that there are inexplicable irregularities simply for
the sake of irregularity.  It's fine if there are *synchronically*
inexplicable irregularities in a conlang, but ideally, the conlanger
should be able to explain why they're there.  In this case, it has
to do with the historical forms.

So, for a word like "mother", the old form looked the same, at
which point stress was assigned, and then it passed on to the
present unchanged:

(3)
(a) *T&lin& ['T&.li.n&] > T&lin& ['T&.li.n&]

The old form of "sword", however, looked like this:

(4)
(a) *esinj [e.'sinj]

Betwen that period and the present, several things happened.
First, unvoiced fricatives voiced intervocalically.  Second, word-final
approximants (i.e., /j/ and /w/) became syllabic when preceded
by a consonant.  The result: A trisyllabic form much like "mother":

(5)
(a) *esinj [e.'sinj] > *ezinj [e.'zinj] > *ezini [e.'zi.ni]

I imagine that a synchronic account would or could argue that the
last vowel is actually a consonant.  And, who knows, but there
might be some truth to this.  However, if you really buy into the
recent work on near-mergers, perhaps the two [i]'s in "sword" are
consistently phonetically different, so that the distinction is in fact
preserved in the synchronic state of the language.

(Note: For a quick intro to near-mergers: There have been studies
recently that argue that segments which have been argued to have
merged are actually phonetically distinct.  Two examples are word-
final obstruents in German and Russian.  Voiced stops in both
languages are said to devoice word-finally.  However, phoneticians
have found that German and Russian speaker maintain consistent
phonetic differences between word-final voiced stops and corresponding
word-final voiceless stops, even though they often can't conciously
tell the difference between the two.  The argument, then, is that
though the stops do devoice, they're still phonologically distinct.)

Oh, for languages with syllabic fricatives, check out IT Berber...
I'm failing to find a link.  Well, I do remember one word:

/nZm/ (no schwas) "to manage to escape"

Of course, that's a voiced fricative, but I believe it has voiceless
nuclei as well.

-David
*******************************************************************
"sunly eleSkarez ygralleryf ydZZixelje je ox2mejze."
"No eternal reward will forgive us now for wasting the dawn."

-Jim Morrison

http://dedalvs.free.fr/

Reply