Johanna,

I wonder if what's happening to "both" might be the laxing of a tense vowel in an unstressed environment.  "Both" normally carries very little stress, although it can.  Of course, that sort of tense/lax alternation isn't normally conditioned just by stress in American English, so this would be an odd case anyway.  

The Arabic voiced pharyngeal fricative is a different sort of sound because of its greater constriction, which will spread tension all the way down to the glottis, so that it also affects glottal state (voicing quality, for phonetically innocent readers).  Pharygealization of alveolars (/T/) in Arabic will pull the tongue back, at least in Saudi Arabic, so that /ta/ and /Ta/ sound to English speakers as if they have two different vowels, a low central and a low back.   However, in Iraqi Arabic that contrast isn't very strong.  But this affects the quality of the following vowel, whereas what I'm describing in English is not conditioned by a neighboring consonant.  I don't know if it occurs in English elsewhere than before a palato-alveolar.  I have heard some very back, almost r-colored vowels in "watch" and even in "water", so it may be more general in Upper South.

Herb


 
Herb,

It might interest you to know of a phenomenon that occurs out here in 
California. People from certain regions of Northern CA (it seems to be 
the central/easterly regions) pronounce the word 'both' with a very 
back, mid-low vowel very similar to the British vowel in a word like 
'bought'. It seems that this pronunciation is limited to this one word 
(then again, how many English words end in '-oth'? There's 'sloth', but 
many people pronounce that -o- with an 'ah' sound. It would be 
interesting to see how these Californians pronounce the last name 
'Toth'.  'Loath', as in 'I am loath to approach the subject' is pretty 
frozen, and I don't know how many current speakers are aware of it).

I've never heard the pharyngealization analysis for 'warsh' before. 
Very interesting. I speak some Arabic, and the 'warsh' I hear does not 
sound much like the voiced pharyngeal fricative. Maybe my native 
English is biasing me too much toward the retroflex /r/. I do believe 
the NoCal 'both' is a case of pharyngealization -- but the only 
motivation for it is the /o/. -th- being a front sound, it doesn't seem 
to offer any reason for backing the tongue. All the same, it doesn't 
happen in other words with -o-, such as 'loaf', etc.

You might want to post this to the list if it doesn't show up there. 
Still can't figure out what the !@!?! is going on with my subscription.

Johanna


On Jul 7, 2006, at 8:29 PM, Stahlke, Herbert F.W. wrote:

Marshall,

Your message appears at the end of an increasingly long posting, so 
I've deleted everything above it.

The perceived /r/ in "warsh" is in fact not an /r/, even though English 
speakers hear it as that.  In fact it's pharyngealization, that is, 
constriction of the pharynx.  The same thing also happens, I 
understand, in some Dutch dialects.  What's going on is this.  The 
vowel in "wash", at least outside of the low back vowel merger zone, 
which goes from Pittsburgh to the Mississippi and then covers most of 
the US west of the Mississippi, is as low and as far back as the tongue 
can be in the mouth without producing a fricative, in this case a 
pharyngeal fricative like those found in Arabic.  Now, a little known 
fact of articulatory phonetics is that when we retroflex for an [r], we 
also push the back of the tongue into the pharyngeal cavity (the 
throat).  It's this effect, actually, rather than the curling back of 
the tongue tip, that produces [r]-coloring.  So if you pronounce "wash" 
with a low back rounded vowel and than pull the tongue back just a 
little bit more, you'll get what to an Arabic speaker sounds like 
pharyngealization and to an English speaker like /r/.

As I commented in my response to Bruce, the first of the two vowels 
usually has to be a mid central or a low vowel, regardless of what the 
second is.  So "see about", where "see" has a high front vowel, doesn't 
become "seer about".  I'm not sure what's going on in Appalachian 
speech, although I've heard these forms.  What is possible is a sort of 
analogical leveling.  Since Appalachian tends to reduce final tense 
vowels, as in "swallow", to schwa, the schwa-final words merge with the 
er-final words, and they all become er-final.  But this is just a 
guess.

What you're calling /l/ insertion is actually the insertion of a schwa 
to break up a cluster.  /l/+consonant clusters have been unstable in 
English for a long time.  Look at all the post-vocalic /l/ we don't 
pronounce, as in talk, walk, yolk, etc.  Appalachian handles the 
problem by splitting the cluster instead.  Another example is 
"athelete".

The use of "taken" for "took" is usually described as analogical 
leveling as well, reducing two forms to one and using the past 
participle for both.  However, I can't help wondering, given the 
overall weakness of /v/ in English, whether it doesn't result from the 
deletion of "'ve", and the perfect and past merge.  /v/ has been 
disappearing for a long time.  "Lord" was "hlavord" in Early Middle 
English, and in Old English "lady" was "hlafdige", with the /f/ 
sounding like /v/.  OE didn't use the letter /v/.  And look at Modern 
English forms like "I'd a been there if I coulda."

Herb


Dr. Johanna Rubba, Associate Professor, Linguistics
Linguistics Minor Advisor
English Department
California Polytechnic State University, San Luis Obispo
E-mail: [log in to unmask]
Tel.: 805.756.2184
Dept. Ofc. Tel.: 805.756.2596
Dept. Fax: 805.756.6374
URL: http://www.cla.calpoly.edu/~jrubba

To join or leave this LISTSERV list, please visit the list's web interface at:
     http://listserv.muohio.edu/archives/ateg.html
and select "Join or leave the list"

Visit ATEG's web site at http://ateg.org/