Why is 'aunt' becoming 'awnt'?
wor...@my-dejanews.com
All my life I've heard 'aunt', but more and more I'm hearing 'awnt'. Not to
mention 'Mawzda' (Mazda) and 'pahsta' (pasta). Anyone have an explanation?
It seems more common the the U.S., but since 'monkey see, monkey do' often
motivates us Canucks, the matter has relevance up here.
Interesting morphological 'aside': Why must I use different letters to render
'pahsta'? Or does 'pawsta' work? Does it matter?
Contact me at Nitpickers Anonymous for that last bit :>
-----== Posted via Deja News, The Leader in Internet Discussion ==-----
http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
Aaron J. Dinkin
> All my life I've heard 'aunt', but more and more I'm hearing 'awnt'. Not to
> mention 'Mawzda' (Mazda) and 'pahsta' (pasta). Anyone have an explanation?
> It seems more common the the U.S., but since 'monkey see, monkey do' often
> motivates us Canucks, the matter has relevance up here.
I really wish more people would learn and use the ASCII IPA. I really have
only a vague idea of what pronunciations you're asking about. At any rate,
here's what I have to say about the words you mention.
I pronounce "aunt" /ant/. It's the most common pronunciation in New England
and a number of other dialects. Hearing "aunt" as /&nt/ (like the word
"ant") irritates me but I've become used to it.
I've never heard any pronunciation for "Mazda" and "pasta" but /'mazd@/ and
/'past@/. In other words, I suspect that the weird new pronunciations for
you are the only ones I'm familiar with.
Please be warned that only in comparatively very few dialects (mostly
Canada and western U.S.) do "ah" and "aw" represent the same phoneme. When
you use them to represent the same phoneme you run the risk of confusing a
whole lot of people.
-Aaron J. Dinkin
Dr. Whom
Mark Odegard
**Please note Spam Trap** On Sat, 27 Jun 1998 23:32:09 -0500,
adi...@commschool.org (Aaron J. Dinkin) in
<adinkin-ya0231800...@news.nii.net> wrote
|I pronounce "aunt" /ant/. It's the most common pronunciation in New England
|and a number of other dialects. Hearing "aunt" as /&nt/ (like the word
|"ant") irritates me but I've become used to it.
The 'awnt' pronunciation, when I hear it, sounds closer to
/A~nt/ (nasalized). I say /&nt/ (the vowel in cat), just like
the word 'ant'. /ant/ looks like a nonrhotic pronunciation of
'aren't', but I have no doubt that people do pronounce 'aunt'
that way.
[...]
|Please be warned that only in comparatively very few dialects (mostly
|Canada and western U.S.) do "ah" and "aw" represent the same phoneme. When
|you use them to represent the same phoneme you run the risk of confusing a
|whole lot of people.
Comparatively "few" dialects, perhaps, but in my experience,
these "few" dialects actually represent the majority of US
speakers.
--
Mark Odegard. (descape to email)
Emailed copies of responses are very much appreciated.
wor...@my-dejanews.com
Can I repost this? I was advised to use ASCII IPA, so here goes:
In article <6n3r79$2i9$1...@nnrp1.dejanews.com,I wrote (paraphrased):
> Living on the west coast of Canada, I've always heard /&nt/, but more and more
> I'm hearing /ant/. >
> Not to mention /Mazdu/ instead of /M&zdu/ and /pastu/ for /p&stu/.
> It seems more common the the U.S., but it's spreading here now. Are we just
> linguistic copycats?>
murr...@compmore.net
In article <6n3r79$2i9$1...@nnrp1.dejanews.com>,
wor...@my-dejanews.com wrote:
>
> All my life I've heard 'aunt', but more and more I'm hearing 'awnt'. Not to
> mention 'Mawzda' (Mazda) and 'pahsta' (pasta). Anyone have an explanation?
> It seems more common the the U.S., but since 'monkey see, monkey do' often
> motivates us Canucks, the matter has relevance up here.
>
> 'pahsta'? Or does 'pawsta' work? Does it matter?
>
> Contact me at Nitpickers Anonymous for that last bit :>
>
> http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
>
Rawthuh.
John Goodwin
>I've never heard any pronunciation for "Mazda" and "pasta" but /'mazd@/ and
>/'past@/. In other words, I suspect that the weird new pronunciations for
>you are the only ones I'm familiar with.
There is a TV presenter called Lloyd Grossman, who I believe comes
from Boston, who uses the most strange vowel sounds I have ever heard.
In fact he's been diagnosed as having irritable vowel syndrome.
His pronuciations would be something like:
Aunt Awwwwwnt
Mazda Mawwwwwzder
Pasta Paaarster
JG
a1a5...@bc.sympatico.ca
On Sun, 28 Jun 1998 05:53:03 GMT, ode...@means.netscape (Mark
Odegard) wrote:
>adi...@commschool.org (Aaron J. Dinkin)
[ ]
>|Please be warned that only in comparatively very few dialects (mostly
>|Canada and western U.S.) do "ah" and "aw" represent the same phoneme. When
>|you use them to represent the same phoneme you run the risk of confusing a
>|whole lot of people.
>
>Comparatively "few" dialects, perhaps, but in my experience,
>these "few" dialects actually represent the majority of US
>speakers.
>--
>Mark Odegard.
I have, in Canada, only once experienced the confusion creation
mentioned by Aaron J. Dinkin and that was when I asked, in a
department store, where they kept the "saws'. My pronunciation
is pretty much standard for UK English and I do voice the "s" at
the end of "saws" so that the clerk (notice my flawless use of
the North American vocabulary) could not have thought I was being
snippy.
All the same, I was reduced to mime to find out where I might
buy the hacksaw I needed.
Gary Williams, Business Services Accounting
>All my life I've heard 'aunt', but more and more I'm hearing 'awnt'. Not to
>mention 'Mawzda' (Mazda) and 'pahsta' (pasta). Anyone have an explanation?
Interesting question, and one which cries out for the use of ASCII IPA, since
in the Northeastern United States, the word spelled "aunt" has long been
pronounced /ant/ (or would it be /Ant/?), not, as we midwesterners say, /&nt/,
just like the insect.
I think I have noted the phenomenon myself, although only in a few speakers.
If the feature were generalized beyond the single word "aunt", might it not be
part of the Northern Cities Vowel Shift, which was discussed here some time
ago.
>Interesting morphological 'aside': Why must I use different letters to render
>'pahsta'? Or does 'pawsta' work? Does it matter?
For me, 'ah' represents the vowel of 'father' or 'pot'. 'Aw' represents a more
back vowel, the one in paw or caught.
Gary Williams
wor...@my-dejanews.com
In article <6n4lsh$8ak$1...@nnrp1.dejanews.com>,
murr...@compmore.net wrote:
> Leapt the sea to jolly ol' England 'ave we?
>
> Rawthuh.
..............
Not at all! My pronunciation is strictly west coast Canadian (p&sta and
M&zda, if I'm getting the IPA right. It takes getting used to).
However, the alternates are becoming more common (pasta and Mazda), and I'm
wondering why. Someone want to explain the North Atlantic Vowel Shift in
simple terms? It would be most appreciated.
wor...@my-dejanews.com
In article <6n5lsa$her$1...@news.cudenver.edu>,
will...@ahecas.AHEC.EDU wrote:
> Interesting question, and one which cries out for the use of ASCII IPA, since
> in the Northeastern United States, the word spelled "aunt" has long been
> pronounced /ant/ (or would it be /Ant/?), not, as we midwesterners say, /&nt/,
> just like the insect.
Thanks for the tip about ASCII IPA. Another regular tipped me off as well,
which I very much appreciated.
We say / &nt / here, too. I'm just wondering _why_ they say / ant / in the
Northeastern U.S. I know it's not new, but it seems so British to me that I'm
always surprised to hear Americans saying it.
> I think I have noted the phenomenon myself, although only in a few speakers.
> If the feature were generalized beyond the single word "aunt", might it not be
> part of the Northern Cities Vowel Shift, which was discussed here some time
> ago.
Thanks again. In my last post, I stupidly referred to the North Atlantic
Vowel Shift, only because I didn't check your post carefully. I also got the
IPA wrong - should have explained that we also say / p&sta / and /M&zda /
here, but nobody says /p&stu/ and /M&zdu/ !
Sorry about that, everyone. Not used to all this high-level stuff. :>
Skitt
wor...@my-dejanews.com wrote in message
<6n689i$s26$1...@nnrp1.dejanews.com>...
>In article <6n4lsh$8ak$1...@nnrp1.dejanews.com>,
> murr...@compmore.net wrote:
>
>
>> Leapt the sea to jolly ol' England 'ave we?
>>
>> Rawthuh.
>..............
>
>Not at all! My pronunciation is strictly west coast Canadian (p&sta
and
>M&zda, if I'm getting the IPA right. It takes getting used to).
>
>However, the alternates are becoming more common (pasta and Mazda),
and I'm
>wondering why. Someone want to explain the North Atlantic Vowel Shift
in
>simple terms? It would be most appreciated.
Could it have something to do with not mangling the original language
pronunciations of those words?
--
Skitt http://www.geocities.com/TheTropics/5537/
K1912
a1a51640 wrote:
>I have, in Canada, only once experienced the confusion creation
>mentioned by Aaron J. Dinkin and that was when I asked, in a
>department store, where they kept the "saws'. My pronunciation
>is pretty much standard for UK English and I do voice the "s" at
>the end of "saws" so that the clerk (notice my flawless use of
>the North American vocabulary) could not have thought I was being
>snippy.
>
>All the same, I was reduced to mime to find out where I might
>buy the hacksaw I needed.
Nevermind your flawless use of the North American vocabulary--was the hacksaw
you needed small enough to fit inside a cake?
gk
K1912
Aaron J. Dinkin
> In article <6n3r79$2i9$1...@nnrp1.dejanews.com>, wor...@my-dejanews.com writes:
>
> >All my life I've heard 'aunt', but more and more I'm hearing 'awnt'. Not to
> >mention 'Mawzda' (Mazda) and 'pahsta' (pasta). Anyone have an explanation?
>
> Interesting question, and one which cries out for the use of ASCII IPA, since
> in the Northeastern United States, the word spelled "aunt" has long been
> pronounced /ant/ (or would it be /Ant/?), not, as we midwesterners say,
> /&nt/, just like the insect.
Well, in the Northeast, it's /ant/, not /Ant/, because Northeastern (or
rather, Boston) accents distinguish between /a/, the vowel in "father"
/'faD@r/, and /A/, the vowel in "bother" /'bAD@r/. "Aunt" shares the vowel
of "father", not "bother".
In those non-Boston dialects in which "father" and "bother" rhyme (usually
both with /A/) but "aunt" and "ant" don't, /Ant/ is probably more
appropriate.
> I think I have noted the phenomenon myself, although only in a few speakers.
> If the feature were generalized beyond the single word "aunt", might it not
> be part of the Northern Cities Vowel Shift, which was discussed here some
> time ago.
Oh, no, the Boston pronunciation of "aunt" is much older than that. Around
the eighteenth century in southern England, /&/ before nasals and unvoiced
fricatives sporadically became /a/ in "staff", "class", "path", "aunt",
"half", "dance", and so forth. This remains the case in RP and, although
it's on its way out, is still present in some words for many speakers in
Boston. (My father pronounces "ask" with /a/; my mother uses /a/ in "half"
and "can't"; my grandparents use it in many more, including "bath".) In
most other dialects this disappeared. In just about all dialects that
originally suffered this change, however, "aunt" with /a/ remains, even in
those which have lost all other sign of it - probably because it
distinguishes "aunt" from "ant".
> >Interesting morphological 'aside': Why must I use different letters to
> >render 'pahsta'? Or does 'pawsta' work? Does it matter?
>
> For me, 'ah' represents the vowel of 'father' or 'pot'. 'Aw' represents a
> more back vowel, the one in paw or caught.
For me, "ah" represents the vowel of "father". "Aw" represents a more back
vowel, the one in "pot", "paw", or "caught".
Aaron J. Dinkin
> On Sun, 28 Jun 1998 05:53:03 GMT, ode...@means.netscape (Mark
> Odegard) wrote:
>
>
> >adi...@commschool.org (Aaron J. Dinkin)
> [ ]
> >|Please be warned that only in comparatively very few dialects (mostly
> >|Canada and western U.S.) do "ah" and "aw" represent the same phoneme. When
> >|you use them to represent the same phoneme you run the risk of confusing a
> >|whole lot of people.
> >
> >Comparatively "few" dialects, perhaps, but in my experience,
> >these "few" dialects actually represent the majority of US
> >speakers.
I don't know about the _majority_ of US speakers. Virtually every dialect
east of the Mississippi, I think, distinguishes between "ah" and "aw" in
some way: /a/ versus /A/ in Boston, /A/ versus /O/ in New York, something
similar to New York in Chicago, and so forth. But I'll grant you that an
awful lot of US (and Canadian) speakers don't distinguish between "ah" and
"aw". Still, though, virtually every dialect of English in the Eastern
Hemisphere makes this distinction doubly so, and I'd bet that the majority
of distinct dialects are in that hemisphere. The Canadian and western US
speakers, numerous though they may be, have very few different dialects
between them.
> I have, in Canada, only once experienced the confusion creation
> mentioned by Aaron J. Dinkin and that was when I asked, in a
> department store, where they kept the "saws'. My pronunciation
> is pretty much standard for UK English and I do voice the "s" at
> the end of "saws" so that the clerk (notice my flawless use of
> the North American vocabulary) could not have thought I was being
> snippy.
>
> All the same, I was reduced to mime to find out where I might
> buy the hacksaw I needed.
The clerk undoubtedly used /A/ in "saws", and the first guess of anyone
with such a dialect on hearing the standard UK English pronunciation of
"saws" - namely, /sO:z/ - would probably guess "sores".
Joseph C Fineman
adi...@commschool.org (Aaron J. Dinkin) writes:
>In article <6n3r79$2i9$1...@nnrp1.dejanews.com>, wor...@my-dejanews.com wrote:
>> All my life I've heard 'aunt', but more and more I'm hearing
>> 'awnt'. Not to mention 'Mawzda' (Mazda) and 'pahsta' (pasta).
>> Anyone have an explanation? It seems more common the the U.S., but
>> since 'monkey see, monkey do' often motivates us Canucks, the
>> matter has relevance up here.
>I pronounce "aunt" /ant/. It's the most common pronunciation in New
>England and a number of other dialects. Hearing "aunt" as /&nt/ (like
>the word "ant") irritates me but I've become used to it.
My native childhood pronunciation of "aunt" was /&nt/, same as "ant".
However, in elementary school (1940s, Beverly Hills, CA), there was a
class called Speech, in which we were taught a refined orthoepy that
as far as I know was native to no class or region, but invented by the
writers of speech books. In it, "ant" was /&nt/, but "aunt" was
/ant/, with a vowel intermediate between the /A/ in "father" & the /&/
in "ant". (This same vowel was prescribed for "ask", "plant", etc.)
Being a good boy & a natural snob, as well as envying anyone who made
a distinction I didn't, I adopted this pronunciation & still use it
when I am being formal.
I have never heard "awnt", if by that is meant the vowel /O/ commonly
heard in "flaunt" & "taunt"; but I can believe it exists. It might be
a spelling pronunciation; or it might be a product of the same shift
that affects the /A/ in "hot", "God", etc. in some (I think it's)
middle-western dialects. Some years ago Ogden Nash wrote a hilarious
poem after hearing a song in which a singer of that persuasion
"apawlogized from the bawtom" of her heart. I quote from memory:
...I have been to Iowa, where they raise hawgs and worship strange
gawds, and now, here I am, in life's autumn.
Confronted by somebody's apawlogies and bautumn.
I myself say /hOg/, but I would only say /gOd/ by way of burlesque, as
in the limerick
There was once a young lady named Maud,
A pious society fraud.
At tea, I am told,
She was proper and cold,
But on the veranda -- my Gawd!
--- Joe Fineman j...@world.std.com
||: There's never time to do it right, but there's always time to :||
||: do it over. :||
Mark Odegard
**Please note Spam Trap** On Sun, 28 Jun 1998 20:00:03 -0500,
adi...@commschool.org (Aaron J. Dinkin) in
<adinkin-ya0231800...@news.nii.net> wrote
|will...@ahecas.AHEC.EDU wrote:
|
|> For me, 'ah' represents the vowel of 'father' or 'pot'. 'Aw' represents a
|> more back vowel, the one in paw or caught.
|
|For me, "ah" represents the vowel of "father". "Aw" represents a more back
|vowel, the one in "pot", "paw", or "caught".
And for me, and lots of us other US native-speakers, all these
words share *same* vowel. And it does make it rough for us
trying to understand what the rest of you attempt to convey (and
vice versa, no doubt).
Buddy
It seems that some people are uncomfortable with (ah) /A/ in at least
some contexts and tend to convert it, usually to (aw) /O/ or (ahr)
/AR/.
I think this may happen more with recent foreign imports. The Shah of
Iran became the Shaw, for some people. Same thing with 'spa' being
pronounced (spahr) /spAR/ by some.
This would account for (MAHZ duh) /'mAzd@/ becoming (MAWZ duh)
/'mOzd@/. It's possible that you are seeing 'pasta' from the other
side, having learned a converted form first and then hearing the
original form later.
I have personally never heard (PAS tuh) /'p&st@/ except to refer to the
guy in charge of the local church.
Buddy
=====================
wor...@my-dejanews.com wrote:
>
> In article <6n5lsa$her$1...@news.cudenver.edu>,
> will...@ahecas.AHEC.EDU wrote:
>
> > Interesting question, and one which cries out for the use of ASCII IPA, since
> > in the Northeastern United States, the word spelled "aunt" has long been
> > pronounced /ant/ (or would it be /Ant/?), not, as we midwesterners say, /&nt/,
> > just like the insect.
>
> Thanks for the tip about ASCII IPA. Another regular tipped me off as well,
> which I very much appreciated.
>
> We say / &nt / here, too. I'm just wondering _why_ they say / ant / in the
> Northeastern U.S. I know it's not new, but it seems so British to me that I'm
> always surprised to hear Americans saying it.
>
> > I think I have noted the phenomenon myself, although only in a few speakers.
> > If the feature were generalized beyond the single word "aunt", might it not be
> > part of the Northern Cities Vowel Shift, which was discussed here some time
> > ago.
>
> Thanks again. In my last post, I stupidly referred to the North Atlantic
> Vowel Shift, only because I didn't check your post carefully. I also got the
> IPA wrong - should have explained that we also say / p&sta / and /M&zda /
> here, but nobody says /p&stu/ and /M&zdu/ !
>
> Sorry about that, everyone. Not used to all this high-level stuff. :>
>
> -----== Posted via Deja News, The Leader in Internet Discussion ==-----
> http://www.dejanews.com/rg_mkgrp.xp Create Your Own Free Member Forum
--
***********************
For email, please use budd...@yahoo98.com, but remove the
ninety-eight.
=============================================
Rex Knepp
Aaron J. Dinkin wrote:
>
> I've never heard any pronunciation for "Mazda" and "pasta" but /'mazd@/ and
> /'past@/. In other words, I suspect that the weird new pronunciations for
> you are the only ones I'm familiar with.
>
>
I have heard < 'p&s t@ > for 'pasta' and < 'm&z d@ > for 'Mazda' on many
occasions; particularly while I lived in Colorado. Here in Texas, the
denizens tend to pronouce 'pasta' < sp@ 'gE ti >. They don't speak the
word 'Mazda,' preferring instead 'Suburban.'
Oddly, the people I've heard pronounce 'Mazda' < 'm&z d@ > also tend to
pronounce 'Toyota' < taI 'joU t@ >. Go figure.
Oh, and in my native Hoosier accent, 'aunt' is < &nt >, though many
of the black aquaintances of my youth pronounced it < Ant > or even
< Ont >.
-30-
rex
Mark Schaefer
> In article <6n4lsh$8ak$1...@nnrp1.dejanews.com>,
> murr...@compmore.net wrote:
>
>
> > Leapt the sea to jolly ol' England 'ave we?
> >
> > Rawthuh.
> ..............
>
> Not at all! My pronunciation is strictly west coast Canadian (p&sta and
> M&zda, if I'm getting the IPA right. It takes getting used to).
>
> However, the alternates are becoming more common (pasta and Mazda), and I'm
> wondering why. Someone want to explain the North Atlantic Vowel Shift in
> simple terms? It would be most appreciated.
I know of no North Atlantic Vowel Shift. As others have similarly stated,
I have always pronounced Mazda as [ma:zd@] and pasta as [past@]. (I think
my Italian grandmother would whoop me upside the head if I were to say
something like [p&st@]).
What I think is really the operative thing here is that British English
(and perhaps Canadian, too) does a much better job of Anglicizing foreign
words. That is, Brits tend to pronounce foreign words as if they had
always been English words. Americans, on the other hand, whether by
virtue of larger-sized immigrant communities, tend to retain the foreign
pronunciation:
Original American British
-------- -------- -------
pasta [pasta] [past@] [p&st@]
taco [tako] [tako] [t&ko]
garage [ga'RaZ] [g@'raZ][g@'radZ] ['g&rIdZ] ['g&raZ]
Nazi [natsi] [natsi] [n&tsi]
In addition to shortening the vowel from [a] -> [&], the British are more
likely to bring the stress forward (as would be expected in a Germanic
language).
While these rules are not always hard and fast (cf. o-REG-ano (Am) vs.
o-re-GA-no (Br.)) they are fairly descriptive of the differences. I had
this point driven home to me on a visit to England and Warwick Castle. I
read one of the placards alound relaying the history of the Castle and the
Earl of Warwick, Henri Beauchamp. As I read the name [anri boSAmp] my
friends turned to me in shock and said "Who?? That's Henry Beecham!"
Mark Schaefer--------------------------------------------------------
Foundry Democracy Project |"The strength of a political system
Foundry United Methodist | depends on the full and willing
1500 16th Street, N.W. | participation of its citizens."
Washington, D.C. 20036 | Social Principles,
(202) 265-8017 | Methodist Book of Discipline
---------------------------------------------------------------------
"Taxation without representation is tyranny."
* * *Support Voting Representation for the District of Columbia* * *
---------------------------------------------------------------------
http://home.dn.net/~schaefer/fdp.html
---------------------------------------------------------------------
Curtis Cameron
Aaron J. Dinkin wrote:
>Well, in the Northeast, it's /ant/, not /Ant/, because Northeastern (or
>rather, Boston) accents distinguish between /a/, the vowel in "father"
>/'faD@r/, and /A/, the vowel in "bother" /'bAD@r/. "Aunt" shares the vowel
>of "father", not "bother".
I grew up pronouncing "aunt" as a homophone of "ain't". If you have
ever heard Opie call for "Aint Bea", this is the same. Of course, I
grew up in the sticks.
-Curtis Cameron
WGS-84 33.033N, 96.724W
Gary Williams, Business Services Accounting
In article <35973F28...@earthlink98.net>,
Buddy <budd...@yahoo98.com (Buddy)> writes:
>It seems that some people are uncomfortable with (ah) /A/ in at least
>some contexts and tend to convert it, usually to (aw) /O/ or (ahr)
>/AR/.
There is another factor that may be involved; our ear may exaggerate the
difference between the vowel being produced and the vowel we expect.
I was once listening to a song in Spanish which contained the phrase "nada me
falta", and I thought it odd that the singer used a fairly flat /&/ vowel in
"falta", since Spanish is noted for the consistency of its vowels. On very
close listening, I decided that nothing had happened to the /a/ of "falta";
what had happened was that _my_ /a/ becomes an /A/ (hope I got that right; I
could also say my "ah" becomes an "aw") before -l-. Since the singer did _not_
move the vowel back, as I expected, I perceived that she had moved it forward.
Could the poster be hearing /pasta/ (pahsta), but the /a/ is so much farther
back than the /&/ he expects that he perceives /pAsta/ (pawsta)?
Gary Williams
Gary Williams, Business Services Accounting
In article <6n8ada$vlu$1...@news.cudenver.edu>, will...@ahecas.AHEC.EDU
(Gary Williams, Business Services Accounting) writes:
Based on advice from Buddy, not only in response to my specific question in
another thread, but also within the very message to which I was replying,
I should have put /O/ wherever I put /A/ below, and possibly should have put
/A/ wherever I put /a/.
Curtis Cameron
wor...@my-dejanews.com wrote:
>Not at all! My pronunciation is strictly west coast Canadian (p&sta and
>M&zda, if I'm getting the IPA right. It takes getting used to).
>
>However, the alternates are becoming more common (pasta and Mazda), and I'm
>wondering why. Someone want to explain the North Atlantic Vowel Shift in
>simple terms? It would be most appreciated.
Probably because p&sta is and always was a minority pronunciation, and
in our modern world, the dialects are gradually getting blended
together. In a hundred years, we'll probably all sound like Ted Koppel
(which is not a good thing IMO).
John Harper
In article <35973F28...@earthlink98.net>,
Buddy <budd...@yahoo98.com (Buddy)> wrote:
>It seems that some people are uncomfortable with (ah) /A/ in at least
>some contexts and tend to convert it, usually to (aw) /O/ or (ahr)
>/AR/.
>
>Iran became the Shaw, for some people. Same thing with 'spa' being
>pronounced (spahr) /spAR/ by some.
A little over 30 years ago I was in a train that had just arrived at
Bath Spa station, England, and was surprised to hear the loudspeaker
saying "Bawth Spaw". But British Rail's public address systems have
never been noted for their hi-fi quality :-) except at Leeds.
The last time I was there I was amazed to hear every word with crystal
clarity (in a Yorkshire accent, but that's to be expected in Leeds!)
John Harper, School of Mathematical and Computing Sciences,
Victoria University, Wellington, New Zealand
e-mail john....@vuw.ac.nz phone (+64)(4)471 5341 fax (+64)(4)495 5045
wor...@my-dejanews.com
In article <6n8aqq$u2m$1...@news.cudenver.edu>,
will...@ahecas.AHEC.EDU wrote:
>
> In article <6n8ada$vlu$1...@news.cudenver.edu>, will...@ahecas.AHEC.EDU
> (Gary Williams, Business Services Accounting) writes:
>
> Based on advice from Buddy, not only in response to my specific question in
> another thread, but also within the very message to which I was replying,
> I should have put /O/ wherever I put /A/ below, and possibly should have put
> /A/ wherever I put /a/.
>
> >There is another factor that may be involved; our ear may exaggerate the
> >difference between the vowel being produced and the vowel we expect.
> >Could the poster be hearing /pasta/ (pahsta), but the /a/ is so much farther
> >back than the /&/ he expects that he perceives /pAsta/ (pawsta)?
I really don't think so. This whole discussion is a real eye-opener for me,
though, because 'father', 'caught' and 'paw' are all /A / or /O/ in western
Canada. They are what used to be called 'short o'.
I don't think we have /a/ here, except in diphthongs like 'out'.
For example, I pronounce 'cat', and 'ask' exactly the same way, and according
to my IPA guidelines, /a/ exists only when the two are pronounced differently
from one another and from 'father' (which to me, rhymes with 'bother').
I also pronounce the first vowel in 'pasta' the same way as the vowels in
'cat' and 'ask'. Ditto for the first vowel in 'Mazda', and the second in
'Sinatra'. The last vowels are schwas. So I guess I say /p&:st@/ and
/M&:sd@/ and /C@n&:tr@/. Or, if you prefer, 'SihNAHtruh'. (NAHt to rhyme
with 'cat')
Oh, and by the the way, the poster is a 'she'. Not that there's anything
wrong with that. ;)
Chris Conner
In article <6n9ocm$a32$1...@nnrp1.dejanews.com>, <wor...@my-dejanews.com> wrote:
>In article <6n8aqq$u2m$1...@news.cudenver.edu>,
>I also pronounce the first vowel in 'pasta' the same way as the vowels in
>'cat' and 'ask'. Ditto for the first vowel in 'Mazda', and the second in
>'Sinatra'. The last vowels are schwas. So I guess I say /p&:st@/ and
>/M&:sd@/ and /C@n&:tr@/. Or, if you prefer, 'SihNAHtruh'. (NAHt to rhyme
>with 'cat')
Okay, let's see if Canada's distinct immigration experience has
had any effect on you. How do you pronounce "basmati"?
--
Chris Conner
Gary Williams, Business Services Accounting
>> >Could the poster be hearing /pasta/ (pahsta), but the /a/ is so much farther
>> >back than the /&/ he expects that he perceives /pAsta/ (pawsta)?
>Oh, and by the the way, the poster is a 'she'. Not that there's anything
>wrong with that. ;)
Oops....the perfect opportunity to demonstrate by use the excellence of the
Finnish pronoun, and I blew it!
Gary Williams
a1a5...@bc.sympatico.ca
On 30 Jun 1998 08:19:26 GMT, cco...@u.washington.edu (Chris
Conner) wrote:
>
> Okay, let's see if Canada's distinct immigration experience has
>had any effect on you. How do you pronounce "basmati"?
>
>--
>Chris Conner
>
From recent BBC listenining I suspect that "modern RP" renders
this as "bessmettee". They have had an immigration experience
too, compounded by cockney.
Aaron J. Dinkin
> I really don't think so. This whole discussion is a real eye-opener for me,
> though, because 'father', 'caught' and 'paw' are all /A / or /O/ in western
> Canada. They are what used to be called 'short o'.
They're all /A/. /O/ is the vowel you probably use in "sore", "port", and
the like (and one of the few ASCII IPA characters you can begin a sentence
with).
(Some older Boston accents use /A/ non-rhotically in "port", making it a
homophone, or nearly so, of "pot".)
> I don't think we have /a/ here, except in diphthongs like 'out'.
>
> For example, I pronounce 'cat', and 'ask' exactly the same way, and according
> to my IPA guidelines, /a/ exists only when the two are pronounced differently
> from one another and from 'father' (which to me, rhymes with 'bother').
That's not quite true. I think it's safe to say that everyone who
distinguishes /a/ from /A/ uses /a/ in "father", but "ask" is a stretch.
Lots of people who have /a/ still use /&/ in "ask". And there is no
three-way distinction: in any dialect, "ask" shares either the vowel of
"cat" or that of "father".
> I also pronounce the first vowel in 'pasta' the same way as the vowels in
> 'cat' and 'ask'. Ditto for the first vowel in 'Mazda', and the second in
> 'Sinatra'. The last vowels are schwas. So I guess I say /p&:st@/ and
> /M&:sd@/ and /C@n&:tr@/. Or, if you prefer, 'SihNAHtruh'. (NAHt to rhyme
> with 'cat')
You mean /m&zd@/ and /s@n&tr@/; /M/, /s/, and /C/ are different segments.
And you're going to confuse a lot of people if you use "NAH" to represent
/n&/. For most people "ah" represents either /a/ or /A/.
Aaron J. Dinkin
> Could the poster be hearing /pasta/ (pahsta), but the /a/ is so much farther
> back than the /&/ he expects that he perceives /pAsta/ (pawsta)?
That could be. It's also possible that the poster's dialect (which poster
was this again?) doesn't contain /a/, so when he hears [a] in an English
word he interprets it as /A/ (since dialects that don't have /a/ tend to
have /A/ where those that use /a/ have /a/).
Aaron J. Dinkin
> It seems that some people are uncomfortable with (ah) /A/ in at least
> some contexts and tend to convert it, usually to (aw) /O/ or (ahr)
> /AR/.
>
> This would account for (MAHZ duh) /'mAzd@/ becoming (MAWZ duh)
> /'mOzd@/. It's possible that you are seeing 'pasta' from the other
> side, having learned a converted form first and then hearing the
> original form later.
I told Wordwiz that she'd get in trouble for not using ASCII IPA. When she
wrote "mawzda", she meant /'mAzd@/ in the first place. She speaks with a
Canadian accent, in which <aw> represents /A/, not /O/ - so "law" is
pronounced /lA/, for example.
wor...@my-dejanews.com
In article <3598fc02...@news.bctel.ca>,
Given that I'm a fan of Uncle Ben, I seldom pronounce it at all. If I did,
however,it would be /b&zm&:ti/. I enjoyed the "bessmettee", though.
And thanks to those of you who are still correcting my IPA. I'm learning lots.
Chris Conner
In article <6ncgar$iof$1...@nnrp1.dejanews.com>, <wor...@my-dejanews.com> wrote:
>In article <3598fc02...@news.bctel.ca>,
> a1a5...@bc.sympatico.ca wrote:
>>
>> On 30 Jun 1998 08:19:26 GMT, cco...@u.washington.edu (Chris
>> Conner) wrote:
>>
>> >
>> > Okay, let's see if Canada's distinct immigration experience has
>> >had any effect on you. How do you pronounce "basmati"?
>> >
>> >--
>> >Chris Conner
>> >
>> From recent BBC listenining I suspect that "modern RP" renders
>> this as "bessmettee". They have had an immigration experience
>> too, compounded by cockney.
>
>
>Given that I'm a fan of Uncle Ben, I seldom pronounce it at all. If I did,
>however,it would be /b&zm&:ti/. I enjoyed the "bessmettee", though.
I expect that most western Americans would pronounce it
/bAz'mA:ti/, but my erstwhile roommate, a Canadian from a southern Indian
family, told me that's just the way white people say it. To him, it was
/'bAzmAti/, near as I can make out.
--
Chris Conner
Buddy
Aaron J. Dinkin wrote:
> I told Wordwiz that she'd get in trouble for not using ASCII IPA. When she
> wrote "mawzda", she meant /'mAzd@/ in the first place. She speaks with a
> Canadian accent, in which <aw> represents /A/, not /O/ - so "law" is
> pronounced /lA/, for example.
I understand now. Wordwiz was familiar with /'m&zd@/ and /'p&st@/ and
then heard them with /A/, which probably are merely the forms that are
closer to the original foreign pronunciation. A matter of Wordwiz
becoming linguistically more cosmopolitan.
Thanks.
Buddy
Buddy
wor...@my-dejanews.com wrote:
> I really don't think so. This whole discussion is a real eye-opener for me,
> though, because 'father', 'caught' and 'paw' are all /A / or /O/ in western
> Canada. They are what used to be called 'short o'.
>
>
> For example, I pronounce 'cat', and 'ask' exactly the same way, and according
> to my IPA guidelines, /a/ exists only when the two are pronounced differently
> from one another and from 'father' (which to me, rhymes with 'bother').
>
> 'cat' and 'ask'. Ditto for the first vowel in 'Mazda', and the second in
> 'Sinatra'. The last vowels are schwas. So I guess I say /p&:st@/ and
> /M&:sd@/ and /C@n&:tr@/. Or, if you prefer, 'SihNAHtruh'. (NAHt to rhyme
> with 'cat')
Now I see. You're accustomed to hearing 'aunt', 'Mazda', and 'pasta'
with /&/ and are now hearing it with /A/.
But, I think /A/ is closer to the original pronunciation of these
words. I also think /A/ is what you'd hear primarily in movies and TV.
I think what you are perceiving as a change in the pronunciation of
'aunt' may be merely a matter of your hearing more samples drawn from a
larger pool of speakers outside your regional accent. Is this a
possibility?
Buddy
Michael Hardy
Mark Schaefer (Mark_S...@notpartofmyaddress.csgi.com) wrote:
> Brits tend to pronounce foreign words as if they had
> always been English words.
I think that's right. During the `Gulf War' I heard
British broadcast journalists saying something like `kyoo-wait'
(I'm not ready to try ASCII IPA), while American broadcast
journalists said `koo-wait'.
> Americans, on the other hand, whether by virtue of larger-sized
> immigrant communities, tend to retain the foreign pronunciation:
Or rather: to change the foreign pronunciation no more than
is necessary to make it pronouneable by Americans. Whereas the
British make further changes beyond those.
> Original American British
> -------- -------- -------
> pasta [pasta] [past@] [p&st@]
OK so far.
> taco [tako] [tako] [t&ko]
I'm assuming by `original' you mean the way it is pronounced
in Spanish. English-speaking North Americans definitely do _not_
pronounce the final `o' in `taco' the way it is pronounced in Spanish.
Most of them _think_ they do. They cannot hear the difference, which
becomes conspicuous to me when I hear English-speaking North Americans
speaking Spanish and pronouncing the `o' in `taco' like the `o' in
English `spoke'. _Now_ I'll take a stab at ASCII IPA: I think you
should have written [takoU] for the way Americans pronounce this word.
Mike Hardy
--
Michael Hardy
ha...@math.unc.edu
Evan Kirshenbaum
ha...@math.unc.edu (Michael Hardy) writes:
> > taco [tako] [tako] [t&ko]
>
>
> I'm assuming by `original' you mean the way it is pronounced
> in Spanish. English-speaking North Americans definitely do _not_
> pronounce the final `o' in `taco' the way it is pronounced in Spanish.
> Most of them _think_ they do. They cannot hear the difference, which
> becomes conspicuous to me when I hear English-speaking North Americans
> speaking Spanish and pronouncing the `o' in `taco' like the `o' in
> English `spoke'. _Now_ I'll take a stab at ASCII IPA: I think you
> should have written [takoU] for the way Americans pronounce this word.
Right on the last vowel. I'd say that the first is /A/ for most
Americans, who don't distinguish between central and back lower
vowels. Merriam-Webster, who note the two pronunciations in words
that are distinguished (such as "father" (central, /a/) and "bother"
(back, /A/) use /A/ for "taco".
I'm sure I'm going to be sorry I asked (because I'm equally sure that
the answer is going to be "we don't"), but how do Brits pronounce the
Japanese word for "octopus", "tako", when ordering sushi? In the US,
it's /tAkoU/, just like the Mexican dish.
--
Evan Kirshenbaum +------------------------------------
HP Laboratories |In the beginning, there were no
1501 Page Mill Road, Building 1U |reasons, there were only causes.
Palo Alto, CA 94304 | Daniel Dennet
Aaron J. Dinkin
In article <v9hk95x...@garrett.hpl.hp.com>, Evan Kirshenbaum
<ev...@garrett.hpl.hp.com> wrote:
> Right on the last vowel. I'd say that the first is /A/ for most
> Americans, who don't distinguish between central and back lower
> vowels. Merriam-Webster, who note the two pronunciations in words
> that are distinguished (such as "father" (central, /a/) and "bother"
> (back, /A/) use /A/ for "taco".
That's wrong. Speakers who distinguish between /a/ and /A/ - that is,
between "father" and "bother" - use /a/ in "taco", not /A/. "Taco" doesn't
rhyme with "Rocko". At least not in Boston.
Ash Nallawalla
Chris Conner wrote in message
<6ncl76$1ncs$1...@nntp1.u.washington.edu>...
> I expect that most western Americans would pronounce it
>/bAz'mA:ti/, but my erstwhile roommate, a Canadian from a
southern Indian
>family, told me that's just the way white people say it. To him,
it was
>/'bAzmAti/, near as I can make out.
It is /'bA:sm@ti/ (means "fragrant"), a Hindi word (i.e. North
Indian). I can't find an ASCII IPA symbol for the "soft t" sound
as it is not present in English. IOW, it does not rhyme with the
word "tea".
Peter K W Tan
Mark Schaefer (Mark_S...@notpartofmyaddress.csgi.com) wrote:
: What I think is really the operative thing here is that British English
: (and perhaps Canadian, too) does a much better job of Anglicizing foreign
: words. That is, Brits tend to pronounce foreign words as if they had
: always been English words. Americans, on the other hand, whether by
: virtue of larger-sized immigrant communities, tend to retain the foreign
: pronunciation:
It's rather difficult to generalise, I think. Think of British
pronunciations of 'tomato', 'charade', or 'van Gogh'. Although _some_
British speakers might use a long i sound (IPA /aI/) for 'Italy', they'd
never do it for 'Iran'. Etc. etc.
Peter
Chris Conner
In article <6nelkl$aja$1...@quole.melbpc.org.au>,
Ash Nallawalla <a...@home.melbpc.org.au> wrote:
[about pronouncing "basmati"]
>It is /'bA:sm@ti/ (means "fragrant"), a Hindi word (i.e. North
>Indian).
Thank you. I'll try to use it some day.
>I can't find an ASCII IPA symbol for the "soft t" sound
>as it is not present in English. IOW, it does not rhyme with the
>word "tea".
Are you thinking of the tongue-flap sound represented in ASCII IPA
by /*/? Something similar?
--
Chris Conner
Aaron J. Dinkin
In article <6nfbqm$el4$1...@nntp1.u.washington.edu>, cco...@u.washington.edu
(Chris Conner) wrote:
Oh, wait a sec - is it retroflex t? The ASCII IPA for that is /t./.
Ash Nallawalla
Chris Conner wrote in message
>>I can't find an ASCII IPA symbol for the "soft t" sound
>>as it is not present in English. IOW, it does not rhyme with
the
>>word "tea".
>
> Are you thinking of the tongue-flap sound represented in ASCII
IPA
>by /*/? Something similar?
I think not, from the example given at the ASCII IPA page
(pedal/petal), but I doubt that one symbol could cover a similar
relationship between a soft D and a hard D sound. Indian
transliteration schemes seem to require knowledge of the sounds.
Go to the textarea box at the bottom of this page:
http://dworkin.wustl.edu/~hari/itrans.html and type "basmati
baa.Nsamatii", then Submit. The second ITRANS transliteration is
needed to generate the correct Devanagari but is barely
recognisable as the first word. (I am not suggesting that ASCII
IPA could or should accommodate every sound known to man.)
It may be easier to find a first generation Indian migrant who
knows Hindi and ask for a pronunciation of "Basmati".
M. Ranjit Mathews
Ash Nallawalla wrote:
> Chris Conner wrote in message
> <6ncl76$1ncs$1...@nntp1.u.washington.edu>...
>
> > I expect that most western Americans would pronounce it
> >/bAz'mA:ti/, but my erstwhile roommate, a Canadian from a
> southern Indian
> >family, told me that's just the way white people say it. To him,
> it was /'bAzmAti/, near as I can make out.
>
> It is /'bA:sm@ti/ (means "fragrant"), a Hindi word (i.e. North
> Indian). I can't find an ASCII IPA symbol for the "soft t" sound
> as it is not present in English. IOW, it does not rhyme with the
> word "tea".
There must be an IPA symbol because it's used a lot in European
languages (French pate', Spanish Quixote, Russian tut, and German
Strasse, for instance). It's a laminodental plosive.
To me, the most natural way to represent it might be to put a horizontal
rectangle below the t to turn it from an apicoalveolar plosive to a
laminal. Of course, that's not an ASCII representation. The character
set shown in http://www.cs.brown.edu/~dpb/ipa/desc.html , however, seems
to represent it with a horizontal bracket below the t. This indicates a
dental t. Unfortunately, there might be no way to specify that it's both
laminal and dental.
The corresponding voiced consonant is represented by a capital D; the
capital T, however, is not available as it's used to represent a
fricative dental (theta). Perhaps [t!] (turned t) represents this sound
in ASCII IPA. If this is the case, I pronounce it /bVsm@t!I/
I've crossposted this to sci.lang so that linguists who speak the
European languages mentioned above can tell us how they represent a
laminodental t in ASCII IPA.
M. Ranjit Mathews
M. Ranjit Mathews wrote:
Correction: It looks like the [D] (thorn ?) can be fricative. If this is
the case and [t!] is not correct, we need a character to place after T /D
in [T]/[D] to indicate that the consonant is plosive rather than fricative,
given that the capitalization of the letter already makes it dental.
Alternatively, we need to hear from European ASCII IPA cognescenti.
M. Ranjit Mathews
Ash Nallawalla wrote:
> Chris Conner wrote in message
> <6nfbqm$el4$1...@nntp1.u.washington.edu>...
> >>I can't find an ASCII IPA symbol for the "soft t" sound
> >>as it is not present in English. IOW, it does not rhyme with
> the
> >>word "tea".
> >
> > Are you thinking of the tongue-flap sound represented in ASCII
> IPA
> >by /*/? Something similar?
Eureka! It is /t[/.
Look up http://www.cs.brown.edu/~dpb/ascii-ipa.html
Keith C. Ivey
"M. Ranjit Mathews" <ran...@austin.ibm.com> wrote:
>M. Ranjit Mathews wrote:
>> The character
>> set shown in http://www.cs.brown.edu/~dpb/ipa/desc.html , however, seems
>> to represent it with a horizontal bracket below the t. This indicates a
>> dental t. Unfortunately, there might be no way to specify that it's both
>> laminal and dental.
>>
>> The corresponding voiced consonant is represented by a capital D; the
>> capital T, however, is not available as it's used to represent a
>> fricative dental (theta). Perhaps [t!] (turned t) represents this sound
>> in ASCII IPA. If this is the case, I pronounce it /bVsm@t!I/
>
>Correction: It looks like the [D] (thorn ?) can be fricative. If this is
>the case and [t!] is not correct, we need a character to place after T /D
>in [T]/[D] to indicate that the consonant is plosive rather than fricative,
>given that the capitalization of the letter already makes it dental.
If you're using Kirshenbaum's ASCII IPA system (described at
http://www.hpl.hp.com/personal/Evan_Kirshenbaum/IPA/faq.html), the
symbols you're looking for are [t[] and [d[] (awkward because opening
square bracket is used to indicate dentality even though brackets are
used to surround phones). The [t!] (turned "t") is used for a dental
click, which I don't believe occurs in any of the languages you
mentioned.
Keith C. Ivey <kci...@cpcug.org>
http://cpcug.org/user/kcivey/
Washington, DC
M. Ranjit Mathews
Here are all the Malayalam, Hindi and Tamil consonants in ASCII IPA. I
leave out the aspirated consonants; you just add an h to get those. That
should cover all the consonant phonemes in Indic and Dravidian
languages.
/k/ /g/ /N/
/q/
/&c/ // /n^/, /c/, /J/ (AIPA has no alveolar J; /c/,/J/ are palatals;
rare in Malayalam)
/z/
/t./ /d./ /n./
/*./ (a retroflex flap like the 2nd 'r' in in "marwari")
/t[/ /d[/ /n[/
/t/ /d/ (Malayalam uses a double 'r' to represent both sounds)
/p/ /b/ /m/
/y/ /r<trl>/ /l[/ // (AIPA has no bilabial 'v', /v<lbd>/ is the closest)
/S/ /s./ /s/ /h/
/l./ /l^/ // /n/ (colloquial tamil often mispronounces /l^/ as /l./)
Devanagari has no /n/ - it uses the same letter as /n[/)
AIPA has no retroflex trill, I suggest using /r./
Devanagari has no /r./; it is represented as a loop on top of a
consonant or as a consonant that comes with a closed schwa that cannot
be replaced by another vowel - eg. rishi
AIPA has no nasal sound. I suggest a '~', which is the appropriate
diacritic in IPA - eg. /nahi:~/ (naheen = no)
Aaron J. Dinkin
In article <359D7DD0...@austin.ibm.com>, "M. Ranjit Mathews"
<ran...@austin.ibm.com> wrote:
> AIPA has no nasal sound. I suggest a '~', which is the appropriate
> diacritic in IPA - eg. /nahi:~/ (naheen = no)
Yes it has, and that's it.
Ash Nallawalla
M. Ranjit Mathews wrote in message
<359D4576...@austin.ibm.com>...
ASCII
>> IPA
>> >by /*/? Something similar?
>
>Eureka! It is /t[/.
>
>Look up http://www.cs.brown.edu/~dpb/ascii-ipa.html
Oh. I assumed that *the* ASCII-IPA home page was
http://homepages.tcp.co.uk/~laker/ipa/ .
--
User Groups on the Web: http://easyrsvp.com/ugotw
User Group Newsletters: http://easyrsvp.com/ugnotw
Design for a New Aussie Flag: http://easyrsvp.com/flag
Remove "home." from my address to write to me.
Larry Phillips
Joseph C Fineman wrote:
> My native childhood pronunciation of "aunt" was /&nt/, same as "ant".
Same here. We always considered those that said /ant/ to be pretentious.
> ... "ant" was /&nt/, but "aunt" was
> /ant/, with a vowel intermediate between the /A/ in "father" & the /&/
> in "ant". (This same vowel was prescribed for "ask", "plant", etc.)
For me, "ant", "plant", and "ask" all have the same vowel sound, as do
"flap", "chap", "tap", "slap", etc.
> I have never heard "awnt", if by that is meant the vowel /O/ commonly
> heard in "flaunt" & "taunt"; but I can believe it exists.
That's the one... it is common in parts of eastern Canada (Ontario and
the Maritime Provinces). It sounds excessively pretentious to my ear.
> It might be
> a spelling pronunciation; or it might be a product of the same shift
> that affects the /A/ in "hot", "God", etc. in some (I think it's)
> middle-western dialects.
In my speech, the cowels in "hot" and "God" rhyme with those in
"flaunt", "taunt", "bought", "clot", ant "cop".
--
------------------------------------------------------------
Sixty billion gigabits can do much. It even does windows.
-- Fred Pohl, Beyond the Blue Event Horizon, 1980
Larry Phillips
Mark Schaefer wrote:
> What I think is really the operative thing here is that British
> English (and perhaps Canadian, too) does a much better job of
> Anglicizing foreign words. That is, Brits tend to pronounce foreign
> words as if they had always been English words. Americans, on the
> other hand, whether by virtue of larger-sized immigrant communities,
> tend to retain the foreign pronunciation:
>
> Original American British
> -------- -------- -------
> pasta [pasta] [past@] [p&st@]
> taco [tako] [tako] [t&ko]
> garage [ga'RaZ] [g@'raZ][g@'radZ] ['g&rIdZ] ['g&raZ]
> Nazi [natsi] [natsi] [n&tsi]
I think you over-generalize. I often notice words in British English
that are left with their foreign pronunciation. 'Blanc Mange',
'restaurant', and 'Agincourt' come immediately to mind. In Canadian
and American English, these tend to be heavily anglicized.
M. Ranjit Mathews
Ash Nallawalla wrote:
> >Look up http://www.cs.brown.edu/~dpb/ascii-ipa.html
>
> Oh. I assumed that *the* ASCII-IPA home page was
> http://homepages.tcp.co.uk/~laker/ipa/ .
Laker's IPA list seems to be a subset of Kirshenbaum's list. I haven't,
however, compared them closely.
Mark Odegard
[Posted, e-mailed] **Please note Spam Trap** On Sat, 04 Jul
1998 00:35:56 -0500, "M. Ranjit Mathews" <ran...@austin.ibm.com>
in <359DBF3C...@austin.ibm.com> wrote
|Laker's IPA list seems to be a subset of Kirshenbaum's list. I haven't,
|however, compared them closely.
Evan K. has the full story. Evan devised the system as I
understand it.
The system we use here on a.u.e., as reflected on Markus' page,
is adapted to the English language, with additional symbols to
represent French and German as, for all practical purposes,
these are the *only* languages in *this* newsgroup whose
pronunciation is a really consistent topic.
The system, as I understand it, however, covers the entire IPA,
though certains letters are more cumbersome in their
representation. Evan posted some time ago that one the of
Spanish R's (the tapped one) is almost impossible to represent
and essentially has to be spelled out in <angle brackets>.
The whole thing was designed as a make-do until 'Unicode'
arrived (an all-inclusive character set that includes just about
everything, from IPA to Linear B to runes, down to the last
unique sqiggle found in Latvian), something that was expected
imminently, but, like the parousia, seems to be something that
will happen in an uncertain, unknown future.
As for representing Indian languages which use a syllabary -- in
their own particular syllabary -- I don't know, but I imagine
provision has been made.
God help us when we can read the Aaronic Benediction (or the
alt.balkan.wars.flamethrowers) in Kyrillic, Kanji, Hebrew,
Arabic, Greek, Armenian, etc, all on a single screen, in a
single news posting, without any fiddling of software or default
settings.
--
Mark Odegard. (descape to email)
Emailed copies of responses are very much appreciated.
David McMurray
Larry Phillips <lar...@home.com> wrote:
[snip]
> > I have never heard "awnt", if by that is meant the vowel /O/ commonly
> > heard in "flaunt" & "taunt"; but I can believe it exists.
>
> That's the one... it is common in parts of eastern Canada (Ontario and
> the Maritime Provinces).
If you are merely defining eastern Canada [1] for the geographically
challenged, you could be right - I have read that this pronunciation is
heard in parts of the Maritimes.
If you are suggesting that Ontario is one of the parts of eastern Canada
in which it is common, my experience suggests otherwise.
> It sounds excessively pretentious to my ear.
It sounds American to both of mine - at least that part of America where
Oprah rules.
[1] You left out Québec - are you making a political statement?
--
David
David McMurray
Larry Phillips <lar...@home.com> wrote:
> Mark Schaefer wrote:
>
> > What I think is really the operative thing here is that British
> > English (and perhaps Canadian, too) does a much better job of
> > Anglicizing foreign words. That is, Brits tend to pronounce foreign
> > words as if they had always been English words. Americans, on the
> > other hand, whether by virtue of larger-sized immigrant communities,
> > tend to retain the foreign pronunciation:
[examples snipped]
> I think you over-generalize. I often notice words in British English
> that are left with their foreign pronunciation. 'Blanc Mange',
> 'restaurant', and 'Agincourt' come immediately to mind. In Canadian
> and American English, these tend to be heavily anglicized.
It's not clear to me what you mean by "heavily anglicized". I say and
hear "blancmange" with a schwa in the first vowel position; as far as I
know, the British do as well. Apart from that, the pronunciation seems
very close to the French.
I thought that the [pick a variety] English pronunciation of
"restaurant" shifted the stress to the first syllable and (usually)
sounded the final "t". I didn't know there was a transatlantic
difference.
As for "Agincourt", you could be right. I don't pronounce the Ontarian
"Agincourt" in the same way as I pronounce the French one, but perhaps
most North Americans do use the anglicized version for both whereas the
British do not.
--
David
Markus Laker
"M. Ranjit Mathews" <ran...@austin.ibm.com> wrote:
> Ash Nallawalla wrote:
>
> > >Look up http://www.cs.brown.edu/~dpb/ascii-ipa.html
> >
> > Oh. I assumed that *the* ASCII-IPA home page was
> > http://homepages.tcp.co.uk/~laker/ipa/ .
>
> Laker's IPA list seems to be a subset of Kirshenbaum's list. I haven't,
> however, compared them closely.
That's right. The ASCII IPA scheme that we use on a.u.e was devised by
a team led by Evan Kirshenbaum and is documented on his Web site. A
subset of that scheme -- enough to transcribe English, but not much more
-- appears in Mark Israel's FAQ. The FAQ material is suitable only as a
reference: there is no tutorial and, of course, there are no sound
samples.
My Web page is an expansion of the information given in the FAQ. I've
added a tutorial and some sound samples, and I've shown how to combine
pure vowels into a few diphtongs and triphthongs that aren't mentioned
in the FAQ, but no new symbols are introduced. The subset of ASCII IPA
documented on my Web page is the same as the one in the FAQ.
Incidentally, neither the FAQ nor my Web page claims to be a definite
reference. Both document have pointers to Evan's Web site. I'd also
like to put on record that I know much less about ASCII IPA than either
Mark or Evan, both of whom were kind enough to help while the page was
being built.
Markus
--
a.u.e FAQ and resources: http://homepages.tcp.co.uk/~laker/aue/
Remove the 'skip this bit' bit of my email address to reply.
Markus Laker
adi...@commschool.org (Aaron J. Dinkin) wrote:
> I think no dialect of English has a phoneme /i:/. I'm not speaking of
> allophones here, merely phonemes, and to say that there's a phoneme /i:/
> implies that there's a distinct phoneme /i/ that can be contrasted with
> /i:/. There isn't.
'Puttee' /'pVti:/; 'putty' /'pVti/.
I could quote more common words -- bargee, (argy) bargy; bailee, bailey;
bumpee, bumpy -- but in each case the stress is thrown on to the last
syllable by the final 'ee'. We could discuss until kingdom come whether
it's the stress or the length of the [i] that distinguishes these pairs
of words.
Markus
[Follow-ups trimmed to exclude s.c.i, where this doesn't seem relevant.]
Peter T. Daniels
M. Ranjit Mathews wrote:
>
> Peter T. Daniels wrote:
> > See K. P. Mohanan in *The World's Writing Systems* (edited by me and
> > William Bright; Oxford University Press, 1996) for a compact yet
> > intelligible description of Malayalam writing (and of the other
> > contemporary standardized Indian scripts by other specialists).
>
> OK, in due course. ISBN, please.
0-19-507993-0
>
> > M. Ranjit Mathews wrote:
> Interesting. I thought people from Laos knew Lao even if their native language was
> different.
Relations between minority ethnic groups and central authorities can't
be generalized; the Hmong people in the US are those recruited by the
CIA to fight against the communist government and probably had little
truck with the government anyway.
>
> > > > Oh, but wait a minute; Malayalam has three different flavours of "r",
> > > > hasn't it? Loosely speaking, /r[/, /r/ and /r./ to go with its /t[/,
> > > > /t/ and /t./. I'm not sure what they are more precisely.
> >
> > Malayalam is said to have SIX different nasals (with one too few letters
> > to notate them all distinctly) and thus to be of very great interest
> > typologically.
>
> /n[/ /n./ /n/ /N/ /n^/ are the only 5 I can think of. The nasal modifier applied to
> consonants introduces either an /n[/ or /N/. /^n/ and /n"/ are not used in Malayalam.
> That seems to cover all the ASCII IPA nasals. What is the sixth nasal ?.
m
(In order: m n[ n n. n^ N)
This from Bill Bright's article on the Dravidian scripts in Sanford
Steever (ed.)'s *The Dravidian Languages* (Routledge, 1997); it has no
chapter on Malayalam(!), and I typeset it: "A geminate alveolar nasal
[n_n_], derived historically and morphophonemically from certain
occurrences of n_r_, is in phonetic contrast with the geminate dental
nasal nn [nn]; however, both are written nn. Thus the nasals are written
the same in [panni] 'pig' and [kan_n_i] 'unmarried girl'."
(underscore indicates bar under the preceding letter)
> > The labial approximant is written in IPA with a symbol that looks rather
> > like an upsilon. (Pullum & Ladusaw call it "script v," which isn't a
> > great label.) I thought ASCII IPA provides an equivalent for each IPA
> > letter; do check your list again. (I've never printed out the various
> > schemes.)
>
> An upsilon is in the vowel quadrilateral; a /u/ can be pronounced /U/ in Malayalam.
> Of course, we can make it do double duty like the French do with their u, but that
> might not be in accordance with AIPA conventions. My chart shows a number of IPA
> symbols that don't have an AIPA representation.
Brian mentioned the two letters don't look alike; but the vowel-letter
doesn't look like an upsilon--it's symmetrical and can be called
"jug-eared u". (IPA used to use small cap U for this sound, and this
symbol is still sometimes seen in "speech" textbooks.)
--
Peter T. Daniels gram...@worldnet.att.net
M. Ranjit Mathews
Aaron J. Dinkin wrote:
> In article <359E797A...@austin.ibm.com>, "M. Ranjit Mathews"
> > Thanks. How would you represent a short 'i' in ASCII IPA, as in the
> > Italian pronounciation of "pizza" or the Indian pronounciation of
> > "Shiva" ?
>
> What's wrong with /i/?
>
> > Americans seem to pronounce both these as /i:/;
>
> I think no dialect of English has a phoneme /i:/. I'm not speaking of
> allophones here, merely phonemes, and to say that there's a phoneme /i:/
> implies that there's a distinct phoneme /i/ that can be contrasted with
> /i:/. There isn't.
>
> > I think /I/ might be closer to the Italian and Indian pronounciations, but
> > not close enough because it can get confused with Indian words that do use an
> > /I/.
>
> So again, why not just use /i/?
I was wondering whether the AIPA scheme allowed one to convey that the 'i' in a
given Indian word (Siva) is to be pronounced as a short and not as an /i:/ (sita)
or an /I/ (rishi). I thought that this might be possible because Europeans use
this vowel, eg. German "bitte", Italian "pizza", Russian "adin". If there is no
AIPA for a short 'i', how does one distinguish between the 'i's in German "bitte"
and "wiener" ?
M. Ranjit Mathews
Peter T. Daniels wrote:
> 0-19-507993-0
Thanks.
> > That seems to cover all the ASCII IPA nasals. What is the sixth nasal ?.
> m (In order: m n[ n n. n^ N)
BTW, in the Malayalam alphabet, the order is N n^ n. n[ m n
> Malayalam is said to have SIX different nasals (with one too few letters
> to notate them all distinctly) and thus to be of very great interest
> typologically.
> This from Bill Bright's article on the Dravidian scripts in Sanford
> Steever (ed.)'s *The Dravidian Languages* (Routledge, 1997); it has no
> chapter on Malayalam(!), and I typeset it: "A geminate alveolar nasal
> [n_n_], derived historically and morphophonemically from certain
> occurrences of n_r_, is in phonetic contrast with the geminate dental
> nasal nn [nn]; however, both are written nn. Thus the nasals are written
> the same in [panni] 'pig' and [kan_n_i] 'unmarried girl'."
> (underscore indicates bar under the preceding letter)
As far as I know, maiden is /kAnj@/ (kanya) I still don't understand which nasal would
pose a typological challenge. I don't know what geminate means. Do you mean that n_ is a
non apical /n/ that souds somewhat like /n[/ because the tongue is not pointed ? One
example of a historical "nr" is the Tamil /onr.u/ which becomes /on[n[u/ in Malayalam, but
this double n aounds EXACTLY the same as the one in /pan[n[i/.
There is a distinct Malayalam letter for each of N, n^, n. n[, m and n . There is a nasal
modifier to consonants but this is pronounced as m, n[ or N depending on context and can
therefore be rendered as such in an IPA transcription. I haven't heard a maiden called
kan_n_i and can't think of an nn that sounds only slightly different from the one in
/pan[n[i/. This double n IS orthographically distinct from the other 5 in /pinni/,
/pAn.n.i/, /b^hANNi/, /pAn^n^i/, and /pAmmi/.
M. Ranjit Mathews
Keith C. Ivey wrote:
> "M. Ranjit Mathews" <ran...@austin.ibm.com> wrote:
>
> >I was wondering whether the AIPA scheme allowed one to convey that the 'i' in a
> >given Indian word (Siva) is to be pronounced as a short and not as an /i:/ (sita)
> >or an /I/ (rishi). I thought that this might be possible because Europeans use
> >this vowel, eg. German "bitte", Italian "pizza", Russian "adin". If there is no
> >AIPA for a short 'i', how does one distinguish between the 'i's in German "bitte"
> >and "wiener" ?
>
> As Aaron J. Dinkin asked, what's wrong with /i/? If you want to
> distinguish between two phonemes that differ only in length, the
> logical solution seems to be to use /i/ for one and /i:/ for the
> other. Length is alwys relative. IPA doesn't define the colonlike
> "long" symbol as indicating a specific number of milliseconds, after
> all.
I just thought there might be some convention, perhaps used by Europeans, to shorten
an /i/ which English readers treat as long by default but Kirshenbaum doesn't provide
a half long symbol. Branner's ';' notation would allow /Si;v@/ to convey a half long
'i' and /sit[A/ a long 'i' but might introduce confusion in Kirshenbaum's scheme
where ';' is used to palatalize consonants.
Keith C. Ivey
"M. Ranjit Mathews" <ran...@austin.ibm.com> wrote:
>I was wondering whether the AIPA scheme allowed one to convey that the 'i' in a
>given Indian word (Siva) is to be pronounced as a short and not as an /i:/ (sita)
>or an /I/ (rishi). I thought that this might be possible because Europeans use
>this vowel, eg. German "bitte", Italian "pizza", Russian "adin". If there is no
>AIPA for a short 'i', how does one distinguish between the 'i's in German "bitte"
>and "wiener" ?
As Aaron J. Dinkin asked, what's wrong with /i/? If you want to
distinguish between two phonemes that differ only in length, the
logical solution seems to be to use /i/ for one and /i:/ for the
other. Length is alwys relative. IPA doesn't define the colonlike
"long" symbol as indicating a specific number of milliseconds, after
all.
Keith C. Ivey <kci...@cpcug.org>
http://cpcug.org/user/kcivey/
Washington, DC
Brian M. Scott
>I don't know what geminate means.
Double (Latin <geminus> 'twin; paired, double').
Brian M. Scott
Peter T. Daniels
M. Ranjit Mathews wrote:
>
> Aaron J. Dinkin wrote:
>
> > In article <359E797A...@austin.ibm.com>, "M. Ranjit Mathews"
> > > Thanks. How would you represent a short 'i' in ASCII IPA, as in the
> > > Italian pronounciation of "pizza" or the Indian pronounciation of
> > > "Shiva" ?
> >
> > What's wrong with /i/?
> >
> > > Americans seem to pronounce both these as /i:/;
> >
> > I think no dialect of English has a phoneme /i:/. I'm not speaking of
> > allophones here, merely phonemes, and to say that there's a phoneme /i:/
> > implies that there's a distinct phoneme /i/ that can be contrasted with
> > /i:/. There isn't.
The difference between "beet" and "bit" is traditionally called "length"
(and indeed the former is longer than the latter); in the Trager-Smith
phonemicization that no one seems to have heard of any more, they are
/biyt/ and /bit/ respectively.
> >
> > > I think /I/ might be closer to the Italian and Indian pronounciations, but
> > > not close enough because it can get confused with Indian words that do use an
> > > /I/.
> >
> > So again, why not just use /i/?
>
> I was wondering whether the AIPA scheme allowed one to convey that the 'i' in a
> given Indian word (Siva) is to be pronounced as a short and not as an /i:/ (sita)
> or an /I/ (rishi). I thought that this might be possible because Europeans use
> this vowel, eg. German "bitte", Italian "pizza", Russian "adin". If there is no
> AIPA for a short 'i', how does one distinguish between the 'i's in German "bitte"
> and "wiener" ?
This continued concern with AIPA (as opposed to IPA) is exactly
equivalent to asking whether Morse Code provides for this or that
spelling.
It doesn't matter; AIPA is entirely dependent on IPA, and if an IPA
symbol exists that isn't provided for in AIPA, then an AIPA symbol
should simply be invented.
And again, *please* distinguish between phonemic and phonetic
transcriptions!!
Peter T. Daniels
M. Ranjit Mathews wrote:
>
> Peter T. Daniels wrote:
>
> > 0-19-507993-0
>
> Thanks.
>
> > > That seems to cover all the ASCII IPA nasals. What is the sixth nasal ?.
> > m (In order: m n[ n n. n^ N)
>
> BTW, in the Malayalam alphabet, the order is N n^ n. n[ m n
I daresay there is NOT ONE Indic script in which the labials (e.g. m)
come before the dentals (e.g. n).
>
> > Malayalam is said to have SIX different nasals (with one too few letters
> > to notate them all distinctly) and thus to be of very great interest
> > typologically.
>
> > This from Bill Bright's article on the Dravidian scripts in Sanford
> > Steever (ed.)'s *The Dravidian Languages* (Routledge, 1997); it has no
> > chapter on Malayalam(!), and I typeset it: "A geminate alveolar nasal
> > [n_n_], derived historically and morphophonemically from certain
> > occurrences of n_r_, is in phonetic contrast with the geminate dental
> > nasal nn [nn]; however, both are written nn. Thus the nasals are written
> > the same in [panni] 'pig' and [kan_n_i] 'unmarried girl'."
> > (underscore indicates bar under the preceding letter)
>
> As far as I know, maiden is /kAnj@/ (kanya) I still don't understand which nasal would
> pose a typological challenge. I don't know what geminate means. Do you mean that n_ is a
> non apical /n/ that souds somewhat like /n[/ because the tongue is not pointed ? One
> example of a historical "nr" is the Tamil /onr.u/ which becomes /on[n[u/ in Malayalam, but
> this double n aounds EXACTLY the same as the one in /pan[n[i/.
I wouldn't be able to distinguish spoken Malayalam from Telugu, Tamil,
or for that matter Sinhalese if I heard all of them together, but I can
look at Bill's phonetic description and know exactly what the
distinction between [n] and [n_] is. Both of them are apical (involving
the tip of the tongue); [n] is dental, meaning the tonguetip goes
against the front teeth, as in most languages of Europe; [n_] is
alveolar, meaning the tonguetip goes against the gumridge, as in
English. Malayalam is unique, or nearly so, among the languages of the
world in having such a phonemic contrast. It's not that any one of the
sounds is unusual; having six different ones is.
You don't say whether you're a native speaker of Malayalam; I suspect
not, since [nj] is closer to the pronunciation of [n^] (palatal nasal)
than to [n_], but if you come from a language that has both (e.g. Hindi
or Spanish or French), you might hear [n_] as [n^].
(BTW careful about your formulation: Modern Tam. /onr.u/ doesn't
"become" Mal. /on[n[u/; both forms derive from a common ancestor:
"Between 800 and 1200 CE the western dialects of Tamil, geographically
separated from the others by the Western Ghats, developed into
Malayalam" (Steever p. 6))
>
> There is a distinct Malayalam letter for each of N, n^, n. n[, m and n .
Sorry, but there just isn't. In the usual way of laying out an Indic
script chart, there are 5 rows (k c t. t p), and the last column is the
nasals; the letter Bill is talking about is the 4th one down, a pair of
arches with in-curved outsides and vertical middle post.
There is a nasal
> modifier to consonants but this is pronounced as m, n[ or N depending on context and can
> therefore be rendered as such in an IPA transcription. I haven't heard a maiden called
> kan_n_i and can't think of an nn that sounds only slightly different from the one in
> /pan[n[i/. This double n IS orthographically distinct from the other 5 in /pinni/,
> /pAn.n.i/, /b^hANNi/, /pAn^n^i/, and /pAmmi/.
What does the 6th letter look like, and why isn't it in even the most
recent materials on the language? Has the same authority that invented
the typrewriter-style vowel markers also invented a new consonant letter
to represent the 6th nasal phoneme?
Here from Mohanan in *The World's Writing Systems* p. 420: The symbol n
represents dental [n[] when initial, alveolar [n] elsewhere; but its
geminate form <same description as above but 3 arches, the two central
posts vertical> represents both dental [n[n[], as in [pan[n[i] 'pig',
and alveolar [nn], as in [kanni] 'unmarried girl'."
M. Ranjit Mathews
Peter T. Daniels wrote:
> I daresay there is NOT ONE Indic script in which the labials (e.g. m)
> come before the dentals (e.g. n).
The Tamil alphabet is k N c n^ t. n. t[ n[ p m j r<trl> l v<blb> ^l l. r. n
The last 4 letters are not in the Hindi alphabet. Any letters after n that you might see in
some Tamil alphabets are Sanskrit sounds that are not used by purists.
> I wouldn't be able to distinguish spoken Malayalam from Telugu, Tamil,
> or for that matter Sinhalese if I heard all of them together, but I can
> look at Bill's phonetic description and know exactly what the
> distinction between [n] and [n_] is. Both of them are apical (involving
> the tip of the tongue); [n] is dental, meaning the tonguetip goes
like in corinth korIn[t[ or pan[n[i (pig) or hin[d[iCome to think of it, the last letter of
Corinth might be theta.
> against the front teeth, as in most languages of Europe; [n_] is
like the n in /nest/, I presume. I didn't know the _ meant alveolar/pin_n_e/ (next) is a
Malayalam word where the ns are alveolar
If there is no separate consonant for the two ns, perhaps, the first is rendered with a
modifier to the p. I can't imagine pan[n[i and pin_n_e spelt with the same n.
> alveolar, meaning the tonguetip goes against the gumridge, as in
> English. Malayalam is unique, or nearly so, among the languages of the
> world in having such a phonemic contrast. It's not that any one of the
> sounds is unusual; having six different ones is.
Hindi has the same 6 nasals rendered by 5 letters and one nasalizing diacritic, so this is not
unusual. Malayalam is unusual, however, in having an alveolar t/d, which other Indian languages
don't have eg. /patti/ (happened) is different from /pat.t.i/ (dog), but other Indians can't
pronounce these two Malayalam words differently.
> You don't say whether you're a native speaker of Malayalam; I suspect
> not, since [nj] is closer to the pronunciation of [n^] (palatal nasal)
I am, but I write Malayalam in either Tamil, Devanagari or Roman script. I haven't used the
Malayalam script since I was 4 or 5.
> than to [n_], but if you come from a language that has both (e.g. Hindi
> or Spanish or French), you might hear [n_] as [n^].
>
> > There is a distinct Malayalam letter for each of N, n^, n. n[, m and n .
>
> Sorry, but there just isn't. In the usual way of laying out an Indic
> script chart, there are 5 rows (k c t. t p), and the last column is the
> nasals; the letter Bill is talking about is the 4th one down, a pair of
> arches with in-curved outsides and vertical middle post.
In all Indic and Dravidian languages, the last column is N n^ n. n[ m. In addition to these
nasals, Tamil has an n_ at the end of the alphabet.
> What does the 6th letter look like, and why isn't it in even the most
> recent materials on the language? Has the same authority that invented
> the typrewriter-style vowel markers also invented a new consonant letter
> to represent the 6th nasal phoneme?
I found a Malayalam alphabet on the web, and am surprised to find that there is indeed no /n/
at the end after /r./ as there is in Tamil.
> Here from Mohanan in *The World's Writing Systems* p. 420: The symbol n
> represents dental [n[] when initial, alveolar [n] elsewhere; but its
> geminate form <same description as above but 3 arches, the two central
> posts vertical> represents both dental [n[n[], as in [pan[n[i] 'pig',
> and alveolar [nn], as in [kanni] 'unmarried girl'."
I know the alveolar n_n_. I just didn't know that the underscore meant alveolar. It is used in
/pin_n_e/ (next), /pin_n_i/ (braided), and /t[en_n_i/ (slipped).
I don't know the word kanni but there is a place called ko:n[n[i :-) Kumari is girl;
Kanyakumari is "unmarried girl" and is the English spelling of the Indian name for Cape
Comorin. I haven't seen it spelt in Malayalam but I've always heard it as kanya as in
"kanyadhanam" (dowry) unless it is spelt in Malayalam as "kanniadhanam". That would be odd,
however, because it isn't possible or kosher to follow a consonant with a vowel; the consonant
and the vowel together make one character. There is therefore a "ni" character but following
this with an 'a' seems unlikely.
I cant imagine how Malayalam writers convey the dental and alveolar n's to the reader if /n[/
and /n/ are not separate characters like in Tamil.
pan[n[i (pig), pin_n_e (next), t[o:n[n[i (felt), t[en_n_i (slipped)
Like I said before the 6 nasals can be used in similar contexts to get 6 words with 6 meanings.
pan[n[i (pig), pin_n_e (next), pan.n.i (done), b^haNNi (splendor), pan^n^i (cotton) and pammi
(quiet demeanor).
Perhaps some Malayali who CAN write Malayalam can tell us how one can tell the first two double
'n's apart. Perhaps the n[n[ is rendered by a nasalizing modifier to the p followed by a single
n whereas the n_n_ is rendered by a double n.
Markus Laker
Larry Phillips <lar...@home.com> wrote:
['blancmange']
> When I have heard a Brit say
> it, it is more like a 'bl&maanj'
We say /bl@'mA.ndZ/.
> (is that '&' the way to show a schwa?).
No, that's /@/. /&/ is the vowel we use in 'cat'. However, I can't
recommend mixing ASCII IPA and ad-hoc pronunciation systems: it seems
like an even more certain recipe for confusion than the ad-hoc system on
its own.
['restaurant']
> Many Brits even out the stress
> and nasalize (huh?) the 'nt'.
Yes, that's right. Prestigious accents tend to do this, resulting in
/'rEst(@)rA.~/. Other accents are more likely to pronounce the word as
/'rEst(@)rA.nt/.
> > As for "Agincourt", you could be right. I don't pronounce the Ontarian
> > "Agincourt" in the same way as I pronounce the French one, but perhaps
> > most North Americans do use the anglicized version for both whereas
> > the British do not.
>
> I think so.
I've always heard something between /'&ZinkU@/ and /'&dZINkO:/,
depending on the linguistic skill of the speaker. I'd consider any
version with a final /t/ to be an error. I didn't know there was an
Agincourt in Ontario; I'm talking about one closer to home.
While I'm writing, let me give the usual RP pronunciations of some other
words mentioned in this thread: 'Aunt' is /A:nt/, 'ant' is /&nt/, 'saws'
and 'sores' are both /sO:z/, 'pasta' is /'p&st@/, 'pastor' is /'pA:st@/,
'spa' and 'spar' are both /spA:/, and the others are /'m&zd@/,
/tO'j@Ut@/ (or perhaps /tOI'@Ut@/), /plA:nt/, /A:sk/, /fl&p/, /t&p/,
/tS&p/, /sl&p/, /hA.t/, /gA.d/, /flO:nt/, /tO:t/, /bO:t/, /klA.t/,
/kA.p/, /'t&k@U/, /'g&rA:Z/ or /'g&rIdZ/, /'nAtsi/ (not /'n&tsi/, as
asserted elsewhere), /'fA:D@/, /'bA.D@/, /t@'mA:t@U/, /S@'rA:d/, /v&n
'gA.x/ or /v&n 'gA.f/, /'It@li/ (and not /'aIt@li/ or /'It@laI/),
/sI'nA:tr@/, /stA:f/, /klA:s/, /pA:T/, /hA:f/ and /dA:ns/ (note that
these last five words use /A/ and not /a/).
I can't comment on words from the subthread crossposted to
soc.culture.indian.
Markus
a1a5...@bc.sympatico.ca
>adi...@commschool.org (Aaron J. Dinkin) wrote:
>> I think no dialect of English has a phoneme /i:/. I'm not speaking of
>> allophones here, merely phonemes, and to say that there's a phoneme /i:/
>> implies that there's a distinct phoneme /i/ that can be contrasted with
>> /i:/. There isn't.
>
>'Puttee' /'pVti:/; 'putty' /'pVti/.
>
>I could quote more common words -- bargee, (argy) bargy; bailee, bailey;
>bumpee, bumpy -- but in each case the stress is thrown on to the last
>syllable by the final 'ee'.
You do not seem to have marked "puttee" thus, though this may be
a misunderstanding of AIPI on my part.
"Puttee" has the same final vowel as "fee"; and that final vowel
is, as you say, stressed. This is long vowel equivalent of the
short "ee" sound found in the first syllable of "reaction" and
this latter is NOT the short form of the "i" sound heard in
"putty", "city" and in "sit".
I think ehat AJD is saying is that you are comparing apples and
oranges. And your pronunciation of "putty", from what you say,
is non-standard to my ear.
>We could discuss until kingdom come whether
>it's the stress or the length of the [i] that distinguishes these pairs
>of words.
>
>Markus
Not with me! The stress has nothing to do with it.
Markus Laker
a1a5...@bc.sympatico.ca wrote:
> On Sun, 05 Jul 1998 21:31:09 GMT, lakerSki...@tcp.co.uk
> (Markus Laker) wrote:
[About Aaron's assertion that there was no /i:/ phoneme:]
> >'Puttee' /'pVti:/; 'putty' /'pVti/.
> >
> >I could quote more common words -- bargee, (argy) bargy; bailee, bailey;
> >bumpee, bumpy -- but in each case the stress is thrown on to the last
> >syllable by the final 'ee'.
>
> You do not seem to have marked "puttee" thus, though this may be
> a misunderstanding of AIPI on my part.
No, you're right. The stess in 'puttee' falls on the first syllable, as
it does in 'putty'. Those words are the only minimal pair I could find.
All the pairs that are more common differ also in stress.
> "Puttee" has the same final vowel as "fee"; and that final vowel
> is, as you say, stressed.
Not according to OED2. I refer to the word 'puttee' /'pVti:/ that names
a long strip of cloth wound spirally around the leg, not the word that
names something putted by a golfer; that would presumably be /pV'ti:/.
> This is long vowel equivalent of the
> short "ee" sound found in the first syllable of "reaction" and
> this latter is NOT the short form of the "i" sound heard in
> "putty", "city" and in "sit".
Therein lies the confusion. In the RP of a hundred years ago, 'city'
was /'sItI/. These days it's /'sIti/. That final vowel is the one in
'seat', not in 'sit'.
For evidence, I refer you to the Oxford University Press. OED2 and (I'm
told) COD8 render 'city' as /'sItI/; COD9, reflecting changes in RP, has
/'sIti/. I've made this suggestion before and been rebuffed, but I
really think you should buy a large, well-respected and, above all,
modern dictionary, and use it frequently. There's no point in waiting
for history to rewind itself and the RP of the 1950s to become the
dominant standard once more. It will never happen.
> I think ehat AJD is saying is that you are comparing apples and
> oranges.
His posting about /i/ and /i:/ wasn't in response to any of mine, and I
haven't yet seen a response to my mention of 'puttee'. I await one with
interest, because Aaron knows how to handle himself on a.u.e.
> And your pronunciation of "putty", from what you say,
> is non-standard to my ear.
In 1950, it would have been. Now, it's the norm.
Peter T. Daniels
M. Ranjit Mathews wrote:
>
> Peter T. Daniels wrote:
>
> > I daresay there is NOT ONE Indic script in which the labials (e.g. m)
> > come before the dentals (e.g. n).
>
> The Tamil alphabet is k N c n^ t. n. t[ n[ p m j r<trl> l v<blb> ^l l. r. n
Of course ... I don't think of Tamil because Tamil has jettisoned
virtually its entire Indic script heritage.
But we're talking about Malayalam script, not Tamil script.
>
> The last 4 letters are not in the Hindi alphabet. Any letters after n that you might see in
> some Tamil alphabets are Sanskrit sounds that are not used by purists.
The Grantha letters are used by purists who need to write Sanskrit
words!
>
> > I wouldn't be able to distinguish spoken Malayalam from Telugu, Tamil,
> > or for that matter Sinhalese if I heard all of them together, but I can
> > look at Bill's phonetic description and know exactly what the
> > distinction between [n] and [n_] is. Both of them are apical (involving
> > the tip of the tongue); [n] is dental, meaning the tonguetip goes
>
> like in corinth korIn[t[ or pan[n[i (pig) or hin[d[iCome to think of it, the last letter of
> Corinth might be theta.
>
> > against the front teeth, as in most languages of Europe; [n_] is
>
> like the n in /nest/, I presume. I didn't know the _ meant alveolar/pin_n_e/ (next) is a
> Malayalam word where the ns are alveolar
>
> If there is no separate consonant for the two ns, perhaps, the first is rendered with a
> modifier to the p. I can't imagine pan[n[i and pin_n_e spelt with the same n.
You seem to have as much trouble reading English as Malayalam. Bright
and Mohanon, in the passages I quoted, both clearly state that the two
nasals, the dental and the alveolar, are written with *the same*
character. There is *no* distinction between them in the script. Just as
in English, the phonemes /T/ and /D/ are both written with <th>.
>
> > alveolar, meaning the tonguetip goes against the gumridge, as in
> > English. Malayalam is unique, or nearly so, among the languages of the
> > world in having such a phonemic contrast. It's not that any one of the
> > sounds is unusual; having six different ones is.
>
> Hindi has the same 6 nasals rendered by 5 letters and one nasalizing diacritic, so this is not
> unusual.
Where are you GETTING this stuff? Accordinig to Yamuna Kachru (in *The
World's Major Languages*), the nasal phonemes of Hindi are: m n (n.)
(n^) (N), where "the phonemes that occur only in the highly Sanskritised
or highly Persianised varieties are given in parentheses."
Malayalam is unusual, however, in having an alveolar t/d, which other
Indian languages
> don't have eg. /patti/ (happened) is different from /pat.t.i/ (dog), but other Indians can't
> pronounce these two Malayalam words differently.
What you wrote here indicates a contrast between dental (or alveolar)
stops and retroflex stops, a contrast familiar in most languages of
India. Perhaps the Mal. t is alveolar rather than dental, so that part
of having a Hindi accent in Mal., for instance, would be using a dental
instead of an alveolar; but they're not missing a contrast.
>
> > You don't say whether you're a native speaker of Malayalam; I suspect
> > not, since [nj] is closer to the pronunciation of [n^] (palatal nasal)
>
> I am, but I write Malayalam in either Tamil, Devanagari or Roman script. I haven't used the
> Malayalam script since I was 4 or 5.
Then what business do you have telling us what the Malayalam script does
and does not have?
>
> > than to [n_], but if you come from a language that has both (e.g. Hindi
> > or Spanish or French), you might hear [n_] as [n^].
> >
>
> > > There is a distinct Malayalam letter for each of N, n^, n. n[, m and n .
> >
> > Sorry, but there just isn't. In the usual way of laying out an Indic
> > script chart, there are 5 rows (k c t. t p), and the last column is the
> > nasals; the letter Bill is talking about is the 4th one down, a pair of
> > arches with in-curved outsides and vertical middle post.
>
> In all Indic and Dravidian languages, the last column is N n^ n. n[ m. In addition to these
> nasals, Tamil has an n_ at the end of the alphabet.
That's Tamil. Not Malayalam.
>
> > What does the 6th letter look like, and why isn't it in even the most
> > recent materials on the language? Has the same authority that invented
> > the typrewriter-style vowel markers also invented a new consonant letter
> > to represent the 6th nasal phoneme?
>
> I found a Malayalam alphabet on the web, and am surprised to find that there is indeed no /n/
> at the end after /r./ as there is in Tamil.
*Tamil* is the one that's out of step with the rest of India.
>
> > Here from Mohanan in *The World's Writing Systems* p. 420: The symbol n
> > represents dental [n[] when initial, alveolar [n] elsewhere; but its
> > geminate form <same description as above but 3 arches, the two central
> > posts vertical> represents both dental [n[n[], as in [pan[n[i] 'pig',
> > and alveolar [nn], as in [kanni] 'unmarried girl'."
>
> I know the alveolar n_n_. I just didn't know that the underscore meant alveolar. It is used in
> /pin_n_e/ (next), /pin_n_i/ (braided), and /t[en_n_i/ (slipped).
>
> I don't know the word kanni but there is a place called ko:n[n[i :-) Kumari is girl;
> Kanyakumari is "unmarried girl" and is the English spelling of the Indian name for Cape
> Comorin. I haven't seen it spelt in Malayalam but I've always heard it as kanya as in
> "kanyadhanam" (dowry) unless it is spelt in Malayalam as "kanniadhanam". That would be odd,
> however, because it isn't possible or kosher to follow a consonant with a vowel;
Confusing speech and writing again.
the consonant
> and the vowel together make one character.
Funny use of "character." (And in the revised "typewriter" script, the
vowels don't deform the consonants, they simply follow them.)
There is therefore a "ni" character but following
> this with an 'a' seems unlikely.
>
> I cant imagine how Malayalam writers convey the dental and alveolar n's to the reader if /n[/
> and /n/ are not separate characters like in Tamil.
See above.
>
> pan[n[i (pig), pin_n_e (next), t[o:n[n[i (felt), t[en_n_i (slipped)
>
> Like I said before the 6 nasals can be used in similar contexts to get 6 words with 6 meanings.
>
> pan[n[i (pig), pin_n_e (next), pan.n.i (done), b^haNNi (splendor), pan^n^i (cotton) and pammi
> (quiet demeanor).
>
> Perhaps some Malayali who CAN write Malayalam can tell us how one can tell the first two double
> 'n's apart. Perhaps the n[n[ is rendered by a nasalizing modifier to the p followed by a single
> n whereas the n_n_ is rendered by a double n.
See, once again, above.
Justin B Rye
> Aaron J. Dinkin <adi...@commschool.org> wrote:
>> I think no dialect of English has a phoneme /i:/. I'm not speaking of
>> allophones here, merely phonemes, and to say that there's a phoneme /i:/
>> implies that there's a distinct phoneme /i/ that can be contrasted with
>> /i:/. There isn't.
I agree there's no /i/-vs-/i:/ distinction; but there's an established
tradition in British textbooks of standardising on /i:/ for the "ee
phoneme". The RP phoneme in question is characteristically longer
than (eg) /I/, and is a member of the phonologically interesting long
vowel family - why insist on disguising this fact? Yes, I suppose /i/
is simpler (but would you simplify /EI/ to /e/ and /@U/ to /o/?); and
yes, I suppose it would be nice to have a standard convention for
a.u.e purposes. But as long as we agree on the system, the precise
labels we choose for these phonemes aren't worth a Holy War.
Markus Laker <lakerSki...@tcp.co.uk> wrote:
> 'Puttee' /'pVti:/; 'putty' /'pVti/.
>
> I could quote more common words -- bargee, (argy) bargy; bailee, bailey;
> bumpee, bumpy -- but in each case the stress is thrown on to the last
> syllable by the final 'ee'. We could discuss until kingdom come whether
> it's the stress or the length of the [i] that distinguishes these pairs
> of words.
I could equally well quote [U]-[u]-[,u:] cases... well, actually the
best I can find is "Dagwood rescued Ermintrude" - [Ud ud ,u:d].
Would you call that a three-way phonemic distinction? I wouldn't.
We're probably going to get more wittering from you-know-who about
['pVti] being a lazy Cockney fad, but the joke's on him: the slurring
of the pitted-vs-pitied distinction was itself originally a Londonism,
and its main stronghold now seems to be south of the Thames.
JBR - mail to "xibalba", not "nospam"
Ankh kak! (Ancient Egyptian blessing)
Larry Phillips
Markus Laker wrote:
>
> Larry Phillips <lar...@home.com> wrote:
>
> ['blancmange']
>
> > When I have heard a Brit say
> > it, it is more like a 'bl&maanj'
>
> We say /bl@'mA.ndZ/.
>
> > (is that '&' the way to show a schwa?).
>
> No, that's /@/. /&/ is the vowel we use in 'cat'. However, I can't
> recommend mixing ASCII IPA and ad-hoc pronunciation systems: it seems
> like an even more certain recipe for confusion than the ad-hoc system
> on its own.
That wasn't meant to be a mixing. I suppose I could have written
"bl'maanj", in an attempt to show the schwa, but I saw no reason to
avoid the IPA schwa. Note that I did avoid all the other IPA
stuff in my ad hoc transliteration.
>>> As for "Agincourt", you could be right. I don't pronounce the
>>> Ontarian "Agincourt" in the same way as I pronounce the French one,
>>> but perhaps most North Americans do use the anglicized version for
>>> both whereas the British do not.
>>
>> I think so.
>
> I've always heard something between /'&ZinkU@/ and /'&dZINkO:/,
> depending on the linguistic skill of the speaker. I'd consider any
> version with a final /t/ to be an error. I didn't know there was an
> Agincourt in Ontario; I'm talking about one closer to home.
The one in Ontario is pronounced 'ayjin court' (ayjin would be like the
worg 'aging', with the terminal 'g' dropped'. The residents don't,
of course, consider it an error.
Aaron J. Dinkin
(Markus Laker) wrote:
> While I'm writing, let me give the usual RP pronunciations of some other
> words mentioned in this thread: 'pasta' is /'p&st@/, 'pastor' is /'pA:st@/
Oh dear. So, rhoticity excepted, UK "pastor" equals US "pasta" /'past@/ and
UK "pasta" equals US "pastor" /'p&st@r/.
Are there any other pairs like this?
Aaron J. Dinkin
(Markus Laker) wrote:
> a1a5...@bc.sympatico.ca wrote:
>
> > On Sun, 05 Jul 1998 21:31:09 GMT, lakerSki...@tcp.co.uk
> > (Markus Laker) wrote:
>
> [About Aaron's assertion that there was no /i:/ phoneme:]
>
> > >'Puttee' /'pVti:/; 'putty' /'pVti/.
> > >
> > >I could quote more common words -- bargee, (argy) bargy; bailee, bailey;
> > >bumpee, bumpy -- but in each case the stress is thrown on to the last
> > >syllable by the final 'ee'.
> >
Here you go. (I haven't responded before because I haven't seen the
original message yet.)
I'm not familiar with the word "puttee". I personally can't quite imagine
an English pronunciation for something spelled <puttee>, stressed on the
first syllable, which differs from "putty". The AHD gives two
pronunciations for it, which (translated into ASCII IPA) are /pVt'i/ and
/'pVti/, which is identical to the pronunciation it gives for "putty". The
AHD, of course, is no authority on British pronunciation.
But still - you claim that the distinction between "puttee" and "putty"
demonstrates that British English has phonemic vowel length for [i]. Can
you produce a minimal pair between what you call /i/ and /I/? If there
isn't one, I'll hold that [i] is an allophone of /I/, and [i:] is just /i/.
You can still call it /i:/, if you like, but I think that's redundant -
it's something of the same reason I use /o/ for the phoneme realized as
[oU].
Aaron J. Dinkin
<j...@nospam.demon.co.uk> wrote:
> > Aaron J. Dinkin <adi...@commschool.org> wrote:
> >> I think no dialect of English has a phoneme /i:/. I'm not speaking of
> >> allophones here, merely phonemes, and to say that there's a phoneme /i:/
> >> implies that there's a distinct phoneme /i/ that can be contrasted with
> >> /i:/. There isn't.
>
> I agree there's no /i/-vs-/i:/ distinction; but there's an established
> tradition in British textbooks of standardising on /i:/ for the "ee
> phoneme". The RP phoneme in question is characteristically longer
> than (eg) /I/, and is a member of the phonologically interesting long
> vowel family - why insist on disguising this fact?
What "long vowel family"? Do you mean the tense vowels? Length plays no
role here. For instance, the tense /i/ in "happy" /'h&pi/ is probably
_shorter_ than the lax /I/ in "pit" /pIt/. But the last syllable in "happy"
is the same phoneme as the longer vowel in "beet" /bit/. There is no "long
vowel family"; the same vowel phoneme can occur with different lengths.
> Yes, I suppose /i/ is simpler (but would you simplify /EI/ to /e/ and /@U/ to
> /o/?);
I would and do (except it's not [@U] in my dialect; it's [oU]).
Markus Laker
> But still - you claim that the distinction between "puttee" and "putty"
> demonstrates that British English has phonemic vowel length for [i]. Can
> you produce a minimal pair between what you call /i/ and /I/?
Okay, but it's as contrived as 'puttee' was obscure: 'minimiser'
/'mInImaIz@/ vs. 'mini miser' /'mIni maIz@/, a small person who dislikes
spending money. As I said, this pair is contrived, but I tried it out
on Sue and she guessed the intended meanings without being prompted
('that button on the screen' and 'small Scrooge' were her answers).
> If there
> isn't one, I'll hold that [i] is an allophone of /I/, and [i:] is just /i/.
> You can still call it /i:/, if you like, but I think that's redundant -
> it's something of the same reason I use /o/ for the phoneme realized as
> [oU].
What would be really useful for questions like this is a wordlist with
transcriptions in ASCII IPA or any other suitable system. Does anyone
know of one?
> -Aaron J. Dinkin
> Dr. Whom
Markus
Dr. 'What's going on?'
[Posted and mailed, because my articles aren't getting out well at the
moment. Mind the spamtrap.]
Markus Laker
> In article <35a2a56b...@news.tcp.co.uk>, lakerSki...@tcp.co.uk
> (Markus Laker) wrote:
>
> > While I'm writing, let me give the usual RP pronunciations of some other
> > words mentioned in this thread: 'pasta' is /'p&st@/, 'pastor' is /'pA:st@/
> Oh dear. So, rhoticity excepted, UK "pastor" equals US "pasta" /'past@/ and
> UK "pasta" equals US "pastor" /'p&st@r/.
Nonononono. RP doesn't use /a/; RP-speaking pastors call themselves
/'pA:st@/. For reference below, ASCII IPA uses /A:/ for the vowel in RP
'car' or Californian 'caught', /&/ for the vowel in RP or Californian
'cat', and /a/ for the vowel in French 'chat', which isn't used as a
pure vowel in RP.
Anyway, what we have here is worse than a difference between RP and
general American. Notice that I gave transcriptions for RP, not for the
UK in general. In various parts of the UK, the first vowel in 'pastor'
could be [A:], [&], [&:] or possibly [a]. In some of those places, the
local accent is rhotic, in others not. I can also imagine seriously
pretentious accents in which 'pasta' would come out as [pA:st@].
Confusion reigns, and as usual it's the A's that cause the trouble.
Markus
Matola
>In article <6nrpf7$306$1...@xibalba.xibalba.demon.co.uk>, Justin B Rye
><j...@nospam.demon.co.uk> wrote:
>
>> > Aaron J. Dinkin <adi...@commschool.org> wrote:
>> >> I think no dialect of English has a phoneme /i:/. I'm not speaking of
>> >> allophones here, merely phonemes, and to say that there's a phoneme /i:/
>> >> implies that there's a distinct phoneme /i/ that can be contrasted with
>> >> /i:/. There isn't.
>>
>> I agree there's no /i/-vs-/i:/ distinction; but there's an established
>> tradition in British textbooks of standardising on /i:/ for the "ee
>> phoneme". The RP phoneme in question is characteristically longer
>> than (eg) /I/, and is a member of the phonologically interesting long
>> vowel family - why insist on disguising this fact?
>
>What "long vowel family"? Do you mean the tense vowels? Length plays no
>role here. For instance, the tense /i/ in "happy" /'h&pi/ is probably
>_shorter_ than the lax /I/ in "pit" /pIt/. But the last syllable in "happy"
>is the same phoneme as the longer vowel in "beet" /bit/. There is no "long
>vowel family"; the same vowel phoneme can occur with different lengths.
Maybe I can clear things up a little. Vowel length is phonetic in US English
and phonemic in UK English, and that is part of the source of confusion in this
thread.
In US English, or at least for speakers of US English with pronunciation
systems similar to mine, vowels are lengthened according to a phonological
rule: that is, they are lengthened when followed by a voiced obstruent in the
same syllable. Obstruents are those consonants that are not glides, liquids, or
nasals. The voiced obstruents are /b/, /v/, /D/, /d/, /z/, /Z/, /dZ/, and /g/.
Aaron is correct that for US English it doesn't make sense to talk about a long
vowel family because all the phonemic vowels have short and long (and nasal)
allophones. (For example, /i/ can show [i] or [i:] (or [i~]).) And for US
English it doesn't make sense to try to equate long vowels with tense vowels
because all permutations of tense/lax and short/long are possible at the
phonetic level.
Compare:
bit /bIt/ [bIt<unx>]
beet /bit/ [bit<unx>]
bid /bId/ [bI:d<unx>]
bead /bid/ [bi:d<unx>]
(Here's where somebody steps in points out that [t<unx>] and [d<unx>] are
identical or nearly identical and it's really vowel length differentiating
bit/bid and beet/bead.)
In UK English (UK English speakers please correct me if I'm wrong) vowel length
is a feature of individual phonemes (some vowels are just longer than others)
and, as Justin has pointed out, various notation systems do capture this. Thus,
/I/ contrasts with /i:/. (There are [I] and [i:], but no *[I:] or *[i].)
Whether there are two simultaneous contrasts (both short/long and tense/lax) or
only one (short/long) is a question I'll leave for the experts.
Compare:
bit /bIt/ [bIt]
beet /bi:t/ [bi:t]
bid /bId/ [bId]
bead /bi:d/ [bi:d]
From a US English standpoint, where long vowels are just positional variants of
the basic short vowels, I can understand how it might not seem to make sense to
have a /i:/ without any */i/ and propose an */i/ that is somehow "more basic".
But given that the difference between /I/ and /i:/ *is phonemic* and that
length is (or is one of) the contrasting feature(s), it does make sense to
represent that feature (even if it is probably redundant).
I'm sure that for both languages you could come up with phonological rules to
predict all sorts of minute variation in vowel length such as happy/pit/beet
discussed above, but I think such variations are secondary to the ones I've
described. In my own speech, the length difference between the [i:] in bead and
the [i]s in happy and beet (and the [I] in pit) is much greater than any
difference between the [i]s in happy and beet (and the [I] in pit).
Michael Matola
Posting from zip code 20852, in the southern (somewhat less godforsaken) part
of Rockville, Maryland (north of Where Wisconsin Avenue Turns Into Rockville
Pike, south of White Flint Mall, east of Interstate 270, west of Grosvenor
metro station), known to some (not me!) as "North Bethesda"---surely a
fictitious geopolitical entity (or a geopolitical entity wannabe).
Aaron J. Dinkin
mat...@aol.com (Matola) wrote:
> In UK English (UK English speakers please correct me if I'm wrong) vowel
> length is a feature of individual phonemes (some vowels are just longer than
> others) and, as Justin has pointed out, various notation systems do capture
> this. Thus, /I/ contrasts with /i:/. (There are [I] and [i:], but no *[I:] or
> *[i].) Whether there are two simultaneous contrasts (both short/long and
> tense/lax) or only one (short/long) is a question I'll leave for the experts.
>
> From a US English standpoint, where long vowels are just positional variants
> of the basic short vowels, I can understand how it might not seem to make
> sense to have a /i:/ without any */i/ and propose an */i/ that is somehow
> "more basic". But given that the difference between /I/ and /i:/ *is
> phonemic* and that length is (or is one of) the contrasting feature(s), it
> does make sense to represent that feature (even if it is probably redundant).
Wow, a phonologist who actually knows what he's talking about. Glad to have
you aboard, Matola.
How can there be two simultaneous contrasts? If there are no distinctions
between tense amd lax vowels except where there is also a distinction
between long and short vowels, and vice versa, surely only one is phonemic
and the other merely phonetic. You could find out which is which by using
[I:] or [i] in a word, for example, and asking a British speaker whether
they perceive it as /I/ or as /i:/, couldn't you? Or investigating whether
other vowels display tense/lax or long/short distinctions? (Of course, if
other vowels display both, for instance if there exist /o/, /O/, and /O:/,
then /i:/ versus /I/ is justified.)
Aaron J. Dinkin
(Markus Laker) wrote:
> adi...@commschool.org (Aaron J. Dinkin) wrote:
>
> > But still - you claim that the distinction between "puttee" and "putty"
> > demonstrates that British English has phonemic vowel length for [i]. Can
> > you produce a minimal pair between what you call /i/ and /I/?
>
> Okay, but it's as contrived as 'puttee' was obscure: 'minimiser'
> /'mInImaIz@/ vs. 'mini miser' /'mIni maIz@/, a small person who dislikes
> spending money. As I said, this pair is contrived, but I tried it out
> on Sue and she guessed the intended meanings without being prompted
> ('that button on the screen' and 'small Scrooge' were her answers).
I'd argue that there's at least also a stress difference between those two
- primary stress is on the first syllable of "minimizer" and on the third
of "mini-miser".
Did you ask her about "putty" and "puttee"?
Aaron J. Dinkin
(Markus Laker) wrote:
> adi...@commschool.org (Aaron J. Dinkin) wrote:
>
> > In article <35a2a56b...@news.tcp.co.uk>, lakerSki...@tcp.co.uk
> > (Markus Laker) wrote:
> >
> > > While I'm writing, let me give the usual RP pronunciations of some other
> > > words mentioned in this thread: 'pasta' is /'p&st@/, 'pastor' is
> > > /'pA:st@/
>
> > Oh dear. So, rhoticity excepted, UK "pastor" equals US "pasta" /'past@/ and
> > UK "pasta" equals US "pastor" /'p&st@r/.
>
> Nonononono. RP doesn't use /a/; RP-speaking pastors call themselves
> /'pA:st@/. For reference below, ASCII IPA uses /A:/ for the vowel in RP
> 'car' or Californian 'caught', /&/ for the vowel in RP or Californian
> 'cat', and /a/ for the vowel in French 'chat', which isn't used as a
> pure vowel in RP.
I'm sorry - for the US pronunciations I gave I gave the representation for
my own dialect. But now I'm talking about what, for lack of a better name,
I call "supradialectal phonemes". When I hear someone whose accent I know
to be English say [A:], the phonological parser in my mind interprets it as
/a/. Thus, I can say that the RP pronunciation of "father" is "essentially
the same" as mine: although you say /'fA:D@/ and I say /'faD@r/, they're
both "f" plus "ah" plus "th" plus "er". Similarly, although the RP
pronunciation of "clerk" is _phonetically_ closest to my pronunciation of
"clock" /klAk/, I hear it as /klark/. What I call /a/ and what you call
/A:/ are, in this analysis, really allophones of the same "supradialectal
phoneme", which I, somewhat misleadingly, called /a/ above. Thus, when you
say /'pA:st@/, I hear /'past@/, 'coz they're really the same thing.
Justin B Rye
I've worn puttees, as it happens. Does that make me an expert?
> I personally can't quite imagine
> an English pronunciation for something spelled <puttee>, stressed on the
> first syllable, which differs from "putty". The AHD gives two
> pronunciations for it, which (translated into ASCII IPA) are /pVt'i/ and
> /'pVti/, which is identical to the pronunciation it gives for "putty". The
> AHD, of course, is no authority on British pronunciation.
Well, for me (using /i/ for the /i(:)/ phoneme) the paste is /'pVti/,
a cloth strip is a /'pV,ti/ and only a golf ball is a /pV'ti/. Or is
that last one /p@'ti/? I've mentioned before that I'm not sure my
/V/-vs-/@/ distinction survives this kind of factoring-out of stress.
Syllable boundaries might prove to be significant as well.
> But still - you claim that the distinction between "puttee" and "putty"
> demonstrates that British English has phonemic vowel length for [i]. Can
> you produce a minimal pair between what you call /i/ and /I/?
There are plenty of minimal pairs for what I'd call /I/ and unstressed
/i(:)/ - blooded/bloodied, taxes/taxis, and indeed putted/puttied
("hit with a putter" versus "smeared with putty"). The fact that many
other accents render those -/Iz/ and -/Id/ suffixes with /@/ helps to
keep this whole topic confusing.
> If there
> isn't one, I'll hold that [i] is an allophone of /I/, and [i:] is just /i/.
The other day I found a book with a whole section on what it called
"the happY vowel" (it was the Gimson memorial phonology collection,
whatever that's called). A couple of dialectological points -
* The late-C18 English prescriptivists Sheridan and Walker, whose
lives were dedicated to the suppression of Cockneyisms, agree that
in their day "easy" involved a repeated vowel and "city" didn't.
The centring of the unstressed [i] didn't start till the C19.
* Even in old RP there was always wide variation, from the accents in
which "Newbury" was all but indistinguishable from "Newborough" to
equally prestigious accents in which the final vowel was [i]. For
examples of the latter, listen to Noel Coward or Sir John Gielgud.
But this whole thing gets more monstrously complicated the more I look
into it. For a start, some RPondian speakers have [i] in "pitied"
*and* "pitted", and for these accents your [i]-as-an-allophone-of-/I/
interpretation makes a lot of sense.
> You can still call it /i:/, if you like, but I think that's redundant -
> it's something of the same reason I use /o/ for the phoneme realized as
> [oU].
Exactly: it's not nearly so tempting for me to transcribe [@Uk<h>EI]
as /oke/. I'd rather skip the procedural debate - now I've made the
point that /i:/ is a British tradition I'd be perfectly happy to stick
to /i/.
a1a5...@bc.sympatico.ca
(Markus Laker) wrote:
[ ]
>For evidence, I refer you to the Oxford University Press. OED2 and (I'm
>told) COD8 render 'city' as /'sItI/; COD9, reflecting changes in RP, has
>/'sIti/. I've made this suggestion before and been rebuffed, but I
>really think you should buy a large, well-respected and, above all,
>modern dictionary, and use it frequently. There's no point in waiting
>for history to rewind itself and the RP of the 1950s to become the
>dominant standard once more. It will never happen.
>
Perhaps one of those defines "puttee" as something stricken on a
golf course? Your humour is enchanting, but I think I'll stick
to the OED and use my ear for what's new (and usually
rejectable).
>> I think ehat AJD is saying is that you are comparing apples and
>> oranges.
>
>His posting about /i/ and /i:/ wasn't in response to any of mine, and I
>haven't yet seen a response to my mention of 'puttee'. I await one with
>interest, because Aaron knows how to handle himself on a.u.e.
>
He likes hyphens fer sure fer sure.
>> And your pronunciation of "putty", from what you say,
>> is non-standard to my ear.
>
>In 1950, it would have been. Now, it's the norm.
>
>Markus
>
I keep telling you and you do not listen: RP is RP and it does
not change. It is FIXED. It is the pronunciation equivalent of
a lingua franca and knows no geographic or (theoretically) social
boundries. Actors learn it, even Gordonstoun seems to perpetuate
it -- though only Spencer knows where Diana and her pals went to
school.
I believe your protestations are like those of the chap who not
too long ago tried to tell me that RP ridiculously prescribed
"balCONEy" as the pronunciation of "BALconey". Absurd.
Matola
>In article <199807082129...@ladder03.news.aol.com>,
>mat...@aol.com (Matola) wrote:
<snip>
>> Whether there are two simultaneous contrasts (both short/long and
>> tense/lax) or only one (short/long) is a question I'll leave for the
experts.
<snip>
>> But given that the difference between /I/ and /i:/ *is
>> phonemic* and that length is (or is one of) the contrasting feature(s), it
>> does make sense to represent that feature (even if it is probably
>redundant).
>Wow, a phonologist who actually knows what he's talking about.
Nonononono. I plead innocent on both counts.
>How can there be two simultaneous contrasts?
In general that's not a problem. /i/ (front, high, unrounded, tense) and /E/
(front, mid, unrounded, lax) contrast two features, height and tense/lax.
>If there are no distinctions
>between tense amd lax vowels except where there is also a distinction
>between long and short vowels, and vice versa, surely only one is phonemic
>and the other merely phonetic.
To argue that one distinction is phonemic and the other is phonetic, you'll
have to come up with some minimal pairs to show the phonemic distinction and a
phonological rule that predicts the phonetic distinction.
>You could find out which is which by using
>[I:] or [i] in a word, for example, and asking a British speaker whether
>they perceive it as /I/ or as /i:/, couldn't you?
I don't think such an experiment would reveal anything useful. The listener
might even conclude that the speaker is a nonnative speaker, or at least a
speaker of a different dialect, because systematically using the wrong
allophone for a given phoneme or an altogether foreign phone is part of what
makes up a "foreign accent."
But on the other hand, this kind of experiment is what we're doing any time we
have a conversation.
>Or investigating whether
>other vowels display tense/lax or long/short distinctions? (Of course, if
>other vowels display both, for instance if there exist /o/, /O/, and /O:/,
>then /i:/ versus /I/ is justified.)
I just checked the various transcription systems (for UK English) in David
Crystal's *Cambridge Encyclopedia of the English Language* (that's where I
think I was channeling some of my original post from). The Gimson system (1962)
does give /i:/ versus /I/. The Jones system (1956) has /i:/ versus /i/, and
Crystal writes about it (page 237): "Gimson (a student of Jones) modified this
system in an attempt to show vowel qualities more accurately."
Crystal makes no mention of a tense/lax distinction in his discussion of (UK)
English vowels, and writes elsewhere (page 239): "A particularly important
factor is length (symbolized by [:]). When we listen to the 12 pure vowels, it
is evident that five of them are relatively long in duration and seven are
relatively short.... The contrast between long and short vowels is not just one
of length (*quantity*); a different place of articulation (*quality*) is
involved. This is why Gimson, for example, in his transcription gives different
symbols to these pairs of vowels (/i:/ vs. /I/, etc.)---drawing attention to
the quality differences between them (p. 237). If length were the only factor,
a transcription of /i:/ vs. /i/ would suffice."
The five "long vowels" are (in my attempt to represent Gimson's system in ASCII
IPA) /i:/, /A:/, /A.:/, /u:/, and /V":/.
In sum, I'm guessing that UK English (again, all corrections are welcome) just
doesn't have the tense/lax distinction that US English does, and, as I
mentioned in my previous post, the vowel length distinctions in the two
languages behave very differently. Two UK English long/short pairs, /i:/ vs.
/I/ and /u:/ vs. /U/, do happen to match up nicely with two US English
tense/lax pairs, /i/ vs. /I/ and /u/ vs. /U/.
Michael Matola
Markus Laker
> I keep telling you and you do not listen: RP is RP and it does
> not change. It is FIXED.
I listen. I understand. I reject the notion on three grounds: first,
RP has changed noticeably since the old Pathe Newsreels were recorded.
A pensioner with a 1940s RP accent, if he can carry it orff, sounds
educated and dignified. A teenager with the same accent sounds
ridiculously pretentious.
Second, RP isn't even unified. You can tell RP speakers apart by their
accents, and there differences wide enough to be recorded in as crude an
instrument as IPA. How can RP be fixed?
Third, all other accents are changing. If RP were fixed, it would
eventually become unintelligible to speakers of all other accents.
> It is the pronunciation equivalent of
> a lingua franca
Yes, I agree.
> and knows no geographic or (theoretically) social
> boundries.
Speak 1940s RP in some of the housing estates around here and see how
long you last.
> Actors learn it, even Gordonstoun seems to perpetuate
> it
Yes. But all you're doing is to list some of the ways in which RP is
used and perpetuated. None of this proves that RP doesn't change.
> -- though only Spencer knows where Diana and her pals went to
> school.
>
> I believe your protestations are .... Absurd.
And I yours. Now where does that get us?
Aaron J. Dinkin
mat...@aol.com (Matola) wrote:
> Aaron J. Dinkin writes:
>
> >In article <199807082129...@ladder03.news.aol.com>,
> >mat...@aol.com (Matola) wrote:
>
> <snip>
>
> >> Whether there are two simultaneous contrasts (both short/long and
> >> tense/lax) or only one (short/long) is a question I'll leave for the
> experts.
>
> <snip>
>
> >> But given that the difference between /I/ and /i:/ *is
> >> phonemic* and that length is (or is one of) the contrasting feature(s), it
> >> does make sense to represent that feature (even if it is probably
> >> redundant).
>
> >How can there be two simultaneous contrasts?
>
> In general that's not a problem. /i/ (front, high, unrounded, tense) and /E/
> (front, mid, unrounded, lax) contrast two features, height and tense/lax.
I suppose I worded that poorly. How can there be two simultaneous contrasts
that only appear simultaneously? That is, if there's no minimal pair for
either contrast (two vowels that differ only in length, or only in
tenseness), then you can't say the contrast is phonemic. If there are a
pair of vowels that differ only in both length and tenseness, then you can
say that one contrast is phonemic and the other merely allophonic (thus
generating a minimal pair) or you can say that what our notation or
terminology calls two separate contrasts is really a single contrast, and
we shouldn't confuse the issue by using symbols that differ in two
respects.
> >If there are no distinctions
> >between tense amd lax vowels except where there is also a distinction
> >between long and short vowels, and vice versa, surely only one is phonemic
> >and the other merely phonetic.
>
> To argue that one distinction is phonemic and the other is phonetic, you'll
> have to come up with some minimal pairs to show the phonemic distinction and
> a phonological rule that predicts the phonetic distinction.
A phonological rule? Easy: whenever the phoneme /i/ appears, it's
phonetically [I]. A minimal pair? "Beet" and "bit".
> >You could find out which is which by using
> >[I:] or [i] in a word, for example, and asking a British speaker whether
> >they perceive it as /I/ or as /i:/, couldn't you?
>
> I don't think such an experiment would reveal anything useful. The listener
> might even conclude that the speaker is a nonnative speaker, or at least a
> speaker of a different dialect, because systematically using the wrong
> allophone for a given phoneme or an altogether foreign phone is part of what
> makes up a "foreign accent."
Well, maybe. I'm becoming more and more convinced that what I've called /U/
and /@/ are actually allophones of the same phoneme. The reason I think
this is because when I pronounce as word that has /U/ in it with [@], or
vice versa, it doesn't sound wrong. For instance, I recognize [p@t] as
"put".
> In sum, I'm guessing that UK English (again, all corrections are welcome)
> just doesn't have the tense/lax distinction that US English does, and, as I
> mentioned in my previous post, the vowel length distinctions in the two
> languages behave very differently.
But if it doesn't have the tense/lax distinction, the phonemes are /i/ and
/i:/, not /I/ and /i:/.
Aaron J. Dinkin
> On Mon, 06 Jul 1998 19:26:51 GMT, lakerSki...@tcp.co.uk
> (Markus Laker) wrote:
>
> >> I think ehat AJD is saying is that you are comparing apples and
> >> oranges.
> >
> >His posting about /i/ and /i:/ wasn't in response to any of mine, and I
> >haven't yet seen a response to my mention of 'puttee'. I await one with
> >interest, because Aaron knows how to handle himself on a.u.e.
> >
> He likes hyphens fer sure fer sure.
Huh? I think I use too many parentheses, but this about hyphens is new to me.
Justin B Rye
after being so effectively rescued, but here goes anyway:
Aaron J. Dinkin <adi...@commschool.org> wrote:
> How can there be two simultaneous contrasts? If there are no
> distinctions between tense amd lax vowels except where there is also a
> distinction between long and short vowels, and vice versa, surely only
> one is phonemic and the other merely phonetic.
How would you apply this to French [r"], for instance? The language
has no other uvulars, no other trills, and no voicing distinction in
sonorants; indeed, /r"/ often becomes a voiceless fricative. Clearly,
it's the uvularity that's important. But is this a good justification
for simplifying the phoneme label to /X/?
> You could find out which is which by using [I:] or [i] in a word, for
> example, and asking a British speaker whether they perceive it as /I/ or
> as /i:/, couldn't you?
I routinely hear [bid] from Scots as "bead", but [bI:d] might sound
like a nonrhotic "beard".
> Or investigating whether other vowels display tense/lax or long/short
> distinctions?
Well, here's a list of the "short" vowels: /I E & V A. U/ (*1). These
vowels are not only distinctly different in quantity from /i A V" O u/
and the diphthongs (*2); they also form the class of vowels that are
never word-final (*3).
* Weaselling footnotes:
1) I'm leaving the unstressable /@/ out of this.
2) given the same context; and /&/ is often semilong.
3) this isn't true of conservative RP, which permits /I#/ in "happy".
> (Of course, if other vowels display both, for instance if there exist
> /o/, /O/, and /O:/, then /i:/ versus /I/ is justified.)
Actually, in my own speech "bed" and "bared" are [E] and [E:], but the
traditional transcription makes the latter /E@/. Come to think of it,
my /E/ is a bit East Anglian; conservative RP uses a closer variant
(as does real Cockney!)... which is why this vowel is often rendered
as /e/ in British sources. Yes, /e/ as in "bed".
a1a5...@bc.sympatico.ca
(Markus Laker) wrote:
>> I believe your protestations are .... Absurd.
>
>And I yours. Now where does that get us?
>
>Markus
Well, for me, certainly not on your "housing estate". Thanks for
the warning.
a1a5...@bc.sympatico.ca
J. Dinkin) wrote:
> I'm becoming more and more convinced that what I've called /U/
>and /@/ are actually allophones of the same phoneme. The reason I think
>this is because when I pronounce as word that has /U/ in it with [@], or
>vice versa, it doesn't sound wrong. For instance, I recognize [p@t] as
>"put".
>
Golf anyone?
a1a5...@bc.sympatico.ca
J. Dinkin) wrote:
>In article <35a4e818...@news.bctel.ca>, a1a5...@bc.sympatico.ca wrote:
>
>> On Mon, 06 Jul 1998 19:26:51 GMT, lakerSki...@tcp.co.uk
>> (Markus Laker) wrote:
>>
>> >> I think ehat AJD is saying is that you are comparing apples and
>> >> oranges.
>> >
>> >His posting about /i/ and /i:/ wasn't in response to any of mine, and I
>> >haven't yet seen a response to my mention of 'puttee'. I await one with
>> >interest, because Aaron knows how to handle himself on a.u.e.
>> >
>> He likes hyphens fer sure fer sure.
>
>Huh? I think I use too many parentheses, but this about hyphens is new to me.
>
>-Aaron J. Dinkin
>Dr. Whom
You do not recall Clinton's bashing and Clinton bashing and your
"Clinton-bashing"? First time I heard that superfluous mark
called a parenthesis.
Compartments many hath 'Titanic': never sail she down the nick.
By the way, in RP it's 'puttEE' for the Indian loan word; 'PUtti'
for the goo & the verb 'putty' ('u' as in 'son', 'i' as in
'sit'); 'PUTTid' for the ppl. of 'putty'; 'PUTTed' for the ppl.
of the golfing 'put' ('u' as in 'son', 'e' as in 'ten'; 'poot'
for the placing verb 'put' ('u' as in 'book').
How on earth can AIPA make things clearer than that? One has to
have a very aberrant delivery of those common English vowels to
go wrong.
When I said I used 'TT' to indicate a glottal stop I was, of
course, only teasing the Plurality. (PlewRALiti)
Matola
>In article <199807092151...@ladder03.news.aol.com>,
>mat...@aol.com (Matola) wrote:
>
>> Aaron J. Dinkin writes:
>How can there be two simultaneous contrasts
>that only appear simultaneously?
/i/ (front, high, unrounded, tense) and /u/ (back, high, rounded, tense).
>That is, if there's no minimal pair for
>either contrast (two vowels that differ only in length, or only in
>tenseness), then you can't say the contrast is phonemic.
I don't understand what you mean here.
>If there are a
>pair of vowels that differ only in both length and tenseness, then you can
>say that one contrast is phonemic and the other merely allophonic (thus
>generating a minimal pair)
You can't just say that one has to be phonemic and the other phonetic.
Again, if you claim that some distinction is phonemic, you have to give a
minimal pair to prove it. If you claim that some distinction is allophonic, you
have to posit a phoneme, establish complementary distribution for the phones in
question, and write a phonological rule that predicts the distribution.
This is essentially what I did for tense/lax vs. short/long in US English in my
original post.
>or you can say that what our notation or
>terminology calls two separate contrasts is really a single contrast,
>and we shouldn't confuse the issue by using symbols that differ in two
>respects.
<snip and rearrange>
>But if it doesn't have the tense/lax distinction, the phonemes are /i/ and
/i:/, not /I/ and /i:/.
I presume that Jones would agree with you and Gimson wouldn't. The Gimson
system, I believe, is well established in UK tradition.
>> >If there are no distinctions
>> >between tense amd lax vowels except where there is also a distinction
>> >between long and short vowels, and vice versa, surely only one is phonemic
>> >and the other merely phonetic.
>>
>> To argue that one distinction is phonemic and the other is phonetic, you'll
>> have to come up with some minimal pairs to show the phonemic distinction
>and
>> a phonological rule that predicts the phonetic distinction.
>
>A phonological rule? Easy: whenever the phoneme /i/ appears, it's
>phonetically [I]. A minimal pair? "Beet" and "bit".
1. I'm not clear which language we're talking about here---US or UK English?
What's the phonemic transcription for that pair?
2. As far as phonological rules go, the one you propose is not a very useful
one. If /i/ is always [I], then why have different symbols? Why have a rule at
all? See discussion above and below for the kind of evidence it would take to
convince me.
>Well, maybe. I'm becoming more and more convinced that what I've called /U/
>and /@/ are actually allophones of the same phoneme. The reason I think
>this is because when I pronounce as word that has /U/ in it with [@], or
>vice versa, it doesn't sound wrong. For instance, I recognize [p@t] as
>"put".
For me, /U/ and /@/ are independent phonemes. I have /pUt/ and /p@t/, for
example. (Both are spelled "put.") If you think [U] and [@] are allophones of
the same phoneme, (1) establish complementary distribution, (2) come up with a
rule to predict the distribution, and (3) give a name to the phoneme.
Michael Matola
Aaron J. Dinkin
mat...@aol.com (Matola) wrote:
> Aaron J. Dinkin writes:
>
> >In article <199807092151...@ladder03.news.aol.com>,
> >mat...@aol.com (Matola) wrote:
> >
> >> Aaron J. Dinkin writes:
>
> >How can there be two simultaneous contrasts
> >that only appear simultaneously?
>
> /i/ (front, high, unrounded, tense) and /u/ (back, high, rounded, tense).
I would argue that American English does not have phonemic roundedness
because there is no minimal pair distinguished by roundedness. For
instance, there is (back, high, rounded, tense) but no (back, high,
unrounded, tense). There is, however, phonemic front/back position: /&/
(front, low) versus /A/ (back, low). The distinctions made in US English
vowel phonology are position front/back, position high/low, and tenseness.
The fact that some vowels happen to be rounded is not relevant for
identifying phonemic contrasts. The only phonemic contrast between /i/ and
/u/ in US English is position.
> >That is, if there's no minimal pair for
> >either contrast (two vowels that differ only in length, or only in
> >tenseness), then you can't say the contrast is phonemic.
>
> I don't understand what you mean here.
For instance, US English doesn't have phonemic roundedness because there
are no two vowels that differ only in roundedness. Similarly, if you assume
length is phonemic in UK English, there is no minimal pair for tenseness,
or if you assume tenseness is phonemic, there is no minimal pair for
length. They can't both be phonemic.
> >If there are a
> >pair of vowels that differ only in both length and tenseness, then you can
> >say that one contrast is phonemic and the other merely allophonic (thus
> >generating a minimal pair)
>
> You can't just say that one has to be phonemic and the other phonetic.
>
> Again, if you claim that some distinction is phonemic, you have to give a
> minimal pair to prove it. If you claim that some distinction is allophonic,
> you have to posit a phoneme, establish complementary distribution for the
> phones in question, and write a phonological rule that predicts the
> distribution.
The UK phonemes conventionally called /i:/ and /I/ differ only in length.
The difference in tenseness is similar to the difference in roundedness
between US /i/ and /u/. It is _phonemically_ more accurate to use /i/
instead of /I/, because the phonemic distinction is length, not tenseness.
It is _phonetically_ more accurate to call it [I], but /I/ does not reflect
the nature of the phonemic distinctions. It's the same reason, I've
probably said, as I use /o/ to refer to the phoneme in my dialect usually
realized as [oU].
> >or you can say that what our notation or
> >terminology calls two separate contrasts is really a single contrast,
> >and we shouldn't confuse the issue by using symbols that differ in two
> >respects.
>
> <snip and rearrange>
>
> >But if it doesn't have the tense/lax distinction, the phonemes are /i/ and
> /i:/, not /I/ and /i:/.
>
> I presume that Jones would agree with you and Gimson wouldn't. The Gimson
> system, I believe, is well established in UK tradition.
That doesn't make it accurate.
> >> >If there are no distinctions
> >> >between tense amd lax vowels except where there is also a distinction
> >> >between long and short vowels, and vice versa, surely only one is
> >> >phonemic and the other merely phonetic.
> >>
> >> To argue that one distinction is phonemic and the other is phonetic,
> >> you'll have to come up with some minimal pairs to show the phonemic
> >> distinction and a phonological rule that predicts the phonetic
> >> distinction.
> >
> >A phonological rule? Easy: whenever the phoneme /i/ appears, it's
> >phonetically [I]. A minimal pair? "Beet" and "bit".
>
> 1. I'm not clear which language we're talking about here---US or UK English?
> What's the phonemic transcription for that pair?
UK English. "Beet and "bit" are /bi:t/ and /bit/, respectively.
> 2. As far as phonological rules go, the one you propose is not a very useful
> one. If /i/ is always [I], then why have different symbols?
Because /i/, as I said, represents the nature of the phonological
distinction made, while [I] represents the phonetic realization of that
phonological distinction. That is, is there are two phonemes that are
(high, front, long) and (high, front, short), why confuse the issue by
introducing a distinction to the notation that's not present in the
phonology?
> >Well, maybe. I'm becoming more and more convinced that what I've called /U/
> >and /@/ are actually allophones of the same phoneme. The reason I think
> >this is because when I pronounce as word that has /U/ in it with [@], or
> >vice versa, it doesn't sound wrong. For instance, I recognize [p@t] as
> >"put".
>
> For me, /U/ and /@/ are independent phonemes. I have /pUt/ and /p@t/, for
> example. (Both are spelled "put.")
What do they mean?
> If you think [U] and [@] are allophones of the same phoneme, (1) establish
> complementary distribution, (2) come up with a rule to predict the
> distribution, and (3) give a name to the phoneme.
Gladly. (1) I've tried to find a minimal pair for quite some time now with
no success. If I ever find one, of course, I'll have to revise my
diagnosis. (2) It's what I usually call /@/ in unstressed syllables or
before /r/ and in "children"; it's what I usually call /U/ otherwise. (The
rule is more complicated than that, of course, because there are several
allophones of "what I usually call /@/".) (3) If I have to, I'll say it's
/U/.
Justin B Rye
> You do not recall Clinton's bashing and Clinton bashing and your
> "Clinton-bashing"? First time I heard that superfluous mark
> called a parenthesis.
Superfluous? How much would you pay to see my dog eating?
> Compartments many hath 'Titanic': never sail she down the nick.
Do you generate this stuff by Markov chaining, or what?
> By the way, in RP it's 'puttEE' for the Indian loan word; 'PUtti'
> for the goo & the verb 'putty' ('u' as in 'son', 'i' as in
> 'sit'); 'PUTTid' for the ppl. of 'putty'; 'PUTTed' for the ppl.
> of the golfing 'put' ('u' as in 'son', 'e' as in 'ten'; 'poot'
> for the placing verb 'put' ('u' as in 'book').
You're claiming that you (and every well-educated BBC newsreader)
would pronounce the word "putted" with an /E/? Oh, Clarence. I can
understand how you come to be so abysmally ignorant; I can even cope
with your belligerent refusal to learn; but why do you constantly butt
in and demand to be given precisely the information you then have to
go to such lengths to avoid understanding? Questions such as:
> How on earth can AIPA make things clearer than that?
Well, even within England, "'u' as in 'son'" could mean anything from
Cockney [a] to Yorkshire [U]; the only people your method conveys
anything to reliably are the people who speak with the same accent as
you - that is, the people who don't need you to tell them. When your
audience is a mixed bag of Singaporeans, Aberdonians, and San
Franciscans, ad hoc transcriptions are worse than useless.
IPA is also more concise (no need for a key every time), more general
(it allows for discussion of Italian /putti/), and more precise (my
own conservative-RP-speaking informant renders both "putted" and
"puttied" as ['p<h>VtI"d] in narrow transcription). Its main drawback
is that it requires the kind of honest attention to acoustic and
articulatory details that would make your delusions harder to sustain.
> One has to have a very aberrant delivery of those common English
> vowels to go wrong.
Sorry, Clarence, but your alleged /pVtEd/ is itself a deeply aberrant
delivery, and has never been recorded in any accent of English I've
heard of.
a1a5...@bc.sympatico.ca
<j...@nospam.demon.co.uk> wrote:
>Clar@nce <a1a5...@bc.sympatico.ca> wrote:
>> You do not recall Clinton's bashing and Clinton bashing and your
>> "Clinton-bashing"? First time I heard that superfluous mark
>> called a parenthesis.
>
>Superfluous? How much would you pay to see my dog eating?
>
That, dear boy, was the whole point. Your Clinton-eating would
be worth something to me.
>> Compartments many hath 'Titanic': never sail she down the nick.
>
>Do you generate this stuff by Markov chaining, or what?
>
The pretty ring of torture as the English nevertheless gets its
message home is not appealing? Oh well. May I have a free
ticket to your lemon-sucking show?
>> By the way, in RP it's 'puttEE' for the Indian loan word; 'PUtti'
>> for the goo & the verb 'putty' ('u' as in 'son', 'i' as in
>> 'sit'); 'PUTTid' for the ppl. of 'putty'; 'PUTTed' for the ppl.
>> of the golfing 'put' ('u' as in 'son', 'e' as in 'ten'; 'poot'
>> for the placing verb 'put' ('u' as in 'book').
>
>You're claiming that you (and every well-educated BBC newsreader)
>would pronounce the word "putted" with an /E/?
Are golfers different? They make no differentiation between
"ied" and "ed"? Do you not? Say "storied", "lurid" and "rooted"
together. There is no difference, for you, in the final
syllable?
>....................................................................Oh, Clarence. I can
>understand how you come to be so abysmally ignorant; I can even cope
>with your belligerent refusal to learn; but why do you constantly butt
>in and demand to be given precisely the information you then have to
>go to such lengths to avoid understanding? Questions such as:
>
>> How on earth can AIPA make things clearer than that?
Perhaps it is because by using actual words one avoids the
tiresome endless disputes you seem to enjoy. One arrives
sometimes at such ludicrous positions as your telling me that RP
demands the pronunciation 'balCONEy' for 'BALlcony'.
>
>Well, even within England, "'u' as in 'son'" could mean anything from
>Cockney [a] to Yorkshire [U]; the only people your method conveys
>anything to reliably are the people who speak with the same accent as
>you - that is, the people who don't need you to tell them. When your
>audience is a mixed bag of Singaporeans, Aberdonians, and San
>Franciscans, ad hoc transcriptions are worse than useless.
>
Come. This particular exchange, though you may have forgotten,
arises purely and simply from one on the pronunciation lingua
franca which is RP. These other dialects are of no concern.
>IPA is also more concise (no need for a key every time), more general
>(it allows for discussion of Italian /putti/), and more precise (my
>own conservative-RP-speaking informant renders both "putted" and
>"puttied" as ['p<h>VtI"d] in narrow transcription). Its main drawback
>is that it requires the kind of honest attention to acoustic and
>articulatory details that would make your delusions harder to sustain.
>
Hmm. If one discards the final 'd' of that transcription does the
final vowel turn out to be the one with which you claim to
pronounce that of 'city'? Or am I displaying my belligerent
abysmal ignorance again? If the vowel is the same then it seems
right for 'putted' but not for 'puttied'. Not in RP anyway. How
did Italian get in here, by the way?
>> One has to have a very aberrant delivery of those common English
>> vowels to go wrong.
>
>Sorry, Clarence, but your alleged /pVtEd/ is itself a deeply aberrant
>delivery, and has never been recorded in any accent of English I've
>heard of.
>
Well, you have a point: it is not quite the vowel of 'ten'. But
it it isn't that of 'puttied' either. I do love the way you
pounce, by the way. Diet is it?
Brian J Goggin
[...]
>[...] the pronunciation lingua franca which is RP.
Are you suggesting that RP is a lingua franca? If so, could you
explain "lingua franca" to me please?
bjg
M. Ranjit Mathews
It is to people around the world who listen to BBC. This is not to say
that most of the world speaks with RP, but rather that it's well
understood.
a1a5...@bc.sympatico.ca
Goggin) wrote:
>On Sun, 12 Jul 1998 14:34:21 GMT, a1a5...@bc.sympatico.ca wrote:
>
>[...]
>
>>[...] the pronunciation lingua franca which is RP.
>
>Are you suggesting that RP is a lingua franca? If so, could you
>explain "lingua franca" to me please?
>
>bjg
>
I suppose in Ireland you might call it the Lowest Common
Denominator, crossing yourself of course.
a1a5...@bc.sympatico.ca
<ran...@austin.ibm.com> wrote:
>Brian J Goggin wrote:
>
>> On Sun, 12 Jul 1998 14:34:21 GMT, a1a5...@bc.sympatico.ca wrote:
>>
>> >[...] the pronunciation lingua franca which is RP.
>>
>> Are you suggesting that RP is a lingua franca? If so, could you
>> explain "lingua franca" to me please?
>
>It is to people around the world who listen to BBC. This is not to say
>that most of the world speaks with RP, but rather that it's well
>understood.
>
I think Mr Goggin knows that -- but it is hot in Ireland at the
moment.