Pronunciation Questions

[Campbell, R Joe]

John,

   I have a few comments on some details that you described below -- I 
rearranged your text to relate two points that you made.

> 	The devoiced n I'm talking about is an aspiration (without vocal
> chord vibration) through the mouth and nose which is different from
> the final h, also breathed through the mouth and nose.
--------

> The devoiced n has more air going
> through the nose than the mouth, while the final h has practically
> all the air going through the mouth.

***********

  If the devoiced n has more air going through the nose than the mouth, 
it must be because the buccal passageway is nearly blocked (or totally 
blocked) by the tip of tongue at the alveolar ridge (or the teeth).  
Sounds that have at least some degree of obstruction in the air flow 
through the mouth are classified as consonants, so this sounds like a 
voiceless nasal consonant.  And, of course,
the velum is lowered, opening the passageway to the nasal cavity.

   If the final h has practically all the air going through mouth, then the
velum is "practically" raised, closing the passageway to the nasal cavity.
And h's are like vowels, having open, unobstructed air flow through the mouth.

   Is this an accurate characterization of the facts?


********************

> I am talking right now with one of the native speakers,
> and he can definitely feel
>  the difference between the two.


********************

   I don't want to wave a red flag at you about using the judgement of
native speakers about pronunciation, but I have a few comments on the danger
in it.

   1.  In one of my early linguistics classes, my professor, Henry R. Kahane,
a famous Romance philologist who had soaked up the principles of American
Structuralism, was trying to make the point that you *had* to take the 
data that the native speaker presented you with.  In his marked German 
accent (University of Berlin Ph.D), he glared at us and pointed his 
finger and said:

  "Zee native speaker iss always right"

And then he smiled slyly and said:

   "Zat iss rule number one.  Rule number two iss: 'Never turn your back on
    zee native speaker."

   Then he explained that linguistic analysis really was about real 
linguistic behavior, *not* about our possibly puristic preconceptions 
(the alliteration is free) about what the speaker meant to say.  But 
rule number two simply reminded us that the native speaker is not a 
trained phonetician, nor does he have access to the intricacies of 
syntactic analysis.  Further, he has prejudices that are not part of 
the data.  This is not to say that those very prejudices are not 
interesting to the analyst and worthy of explanation.  ...but they are 
not phonetic data.

2.  In the 1930's, Edward Sapir published his famous article, "The 
Psychological Reality of Phonemes", in which he described an incident 
with his "native interpreter" of Southern Paiute, Tony.  When Tony was 
asked to write a particular word, he demonstrated that he believed that 
he was pronouncing something other than what he actually pronounced.


3.  In 1970-71, a speaker of the Hueyapan, Morelos, dialect of Nahuatl spent
the academic year in Bloomington to help me as co-teacher and informant in
teaching Indiana University's first course in Nahuatl.  We worked in my 
office several hours a week, as well as presenting the material in 
class three times a week.  After a month of work, I said to her 
apologetically that I realized that I didn't write Nahuatl in the 
traditional way that it appeared in some many documents, but was she 
familiar enough with my k's and w's and ts's for her to write some 
words down?
   She said "Quemah" (or at the time, actually, "kemah").  So I said "lo oigo"
and she wrote "nikagi".  We did several more words (with Sapir smiling 
over my shoulder the whole time) and then I said "mujer" --  she wrote 
"sobatl"!!
She had been pronouncing each of the words as we proceeded and this 
time I expressed mild surprise, saying that I just didn't hear it that 
well.  (She had clearly said "soatl".)  She repeated it over and over 
and I kept expressing my inability to hear anything between the "o" and 
the "a".  Finally, she leaned
over the desk and said "ssssooooo aaatl".

   I soon realized that although there was no [b] in her pronunciation, 
there was certainly reason for her to believe that there *was*.

   Here are the relevant data from Hueyapan (in a 'phonetic' looking 
transcription):

    wetsi            he falls
  nivetsi            I fall
   oets              he fell

The facts are straightforward:

   1.  /w/ is the initial phoneme in the stem meaning "fall".
   2.  /w/ becomes [v] intervocally (i.e. between vowels);
       [v] is written as either "v" or "b" in the
         language of literacy of Hueyapan (i.e., Spanish).
   3.  /w/ deletes when immediately followed by /o/.

Thus, when a speaker of Nahuatl from Hueyapan either says or hears
a sequence of /o/ followed by another vowel, it is perfectly reasonable
for them to intuit and *strongly believe* that there is a /w/, which 
they believe is realized as [v] immediately after the /o/.

Iztayohmeh,

Joe




_______________________________________________
Nahuatl mailing list
Nahuatl at lists.famsi.org
http://www.famsi.org/mailman/listinfo/nahuatl