<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
Well, I decided to just share one observation about final Western
Muskogean /i/ as [e] (I don't recall noticing that Haas had said this
about Choctaw...).<br>
<br>
First, just to clarify re what Bob's talking about below, this
certainly does not happen in either Choctaw (Ct) or Chickasaw (Cs) with
long /i:/. as in borrowed words like Ct <i>tiih</i> / Cs <i>tii</i>
'tea' or other (not too common) words like the Cs interjection <i>kii</i>
'oh!' -- in these words, /i:/ is [i:].<br>
<br>
(Note that words I've written in italics (or, if you don't receive
those, not in //'s or []'s) are in orthography, but I don't think
there's anything unclear here. Note further that I am ignoring Ct final
-<i>h</i>. Most people don't hear this anyway!)<br>
<br>
However, there is a final [e] that's very important, and that I think
is a real problem for phonemics. This is the last sound in the Ct
(rarely also Cs) verb ending often written "hoke" [ho:ke:], which does
something like affirm the truth of the preceding (and is thought by
some to be the source for English <i>okay</i> -- this is really in
some dictionaries of English!). I honestly don't feel that we
understand the best phonemicization of this. I write this ending, when
it occurs in Cs (not often), as -<i>hookay</i>. The <i>ay</i># just
shows I am puzzled (but <i>ay</i> can be pronounced as [e:] elsewhere,
e.g. in <i>áyya'sha</i> 'they are there'). <br>
<br>
One more thing about final [e], possibly related to Haas's observation.
I frequently teach a class where ordinary UCLA students with no special
background in linguistics (mostly monolingual speakers of English or
bilinguals in Spanish or some Asian language) learn to pronounce and
transcribe Chickasaw. I am always amazed (though no longer surprised)
that about half of them reliably will pronounce a word like <i>malili</i>
'run', with final /i/, with final [e], rather than final [i]. Of course
I'm influenced by the phonemics, but I really hear this vowel only as
[i]. And so does about half (or even more) of a typical class. But the
others feel strongly that this final vowel is [e] (they pronounce it
this way, and we typically discuss the contrast in perception, which
they validate). So my "theory" is that there is some other factor
different from formant height or whatever that these students are
attending to, which is shared with English /e/ but not English /i/
(what? I don't know...). Because they note this feature, they interpret
the vowel as /e/ (and thus [e]). <br>
<br>
I'm certainly not saying that Haas was like my students, and in fact I
haven't done the same "experiment" with Choctaw. But this is all a
puzzle to me.<br>
<br>
Pam<br>
<br>
R. Rankin wrote:<br>
<blockquote type="cite" cite="mid00b901c4d730$40e666e0$2ab5ed81@Rankin">
<blockquote type="cite">. . .Haas states that the allophone [e] for
'i' occurs with length and in utterance final position before glottal
stop. I don't think this was mentioned by Dorsey in the dictionary
however.
<br>
</blockquote>
<br>
No, Dorsey always writes the length distinction as if it were a quality
distinction. No one has had the courage to undertake a complete
analysis of JOD's use of the breve and other vowel diacritics (it would
be a massive undertaking). But in Biloxi, it may make a tremendous
difference. All linguists since Dorsey have already screwed up Biloxi
by collapsing the two series of stops, and I'm afraid the vowels are no
different. It is especially important in the SE, where [e] is an
allophone of /i/ and [epsilon] is an allophone of /e/, to figure out
JOD's transcription. As a starting point, I'd look for his
<e-breve> to represent short /e/. Then his <e> with no
diacritics will be either [+long] or an allophone of /i/ (or both,
unfortunately). But there are other E's (e.g., with circumflex) to
deal with too.
<br>
<br>
Haas's comment about [e] representing long /i:/ I find especially
interesting, because I think the Ofo rule is the same, and I hadn't
reread Haas when doing my Ofo analysis. I'll double check. Thanx for
the tip.
<br>
<br>
Bob <br>
<br>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
Pamela Munro,
Professor, Linguistics, UCLA
UCLA Box 951543
Los Angeles, CA 90095-1543
<a class="moz-txt-link-freetext" href="http://www.linguistics.ucla.edu/people/munro/munro.htm">http://www.linguistics.ucla.edu/people/munro/munro.htm</a>
</pre>
</body>
</html>