vowel pronunciations

Another Guest   Sun Nov 16, 2008 9:29 pm GMT
I've been trying to figure out what the "standard" pronunciations of the vowels are, and I haven't had an easy time of it. I went on youtube, and found this video http://www.youtube.com/watch?v=XlSeYt6PN6s , and the pronunciation of D (which I assume is suppsed to be the same as ɒ) is given at 3:32. But it seems to me that the sound given when she says the letter alone is different from when she pronounces in the word "on". Alone, it's like a longer version of the "a" in saw, but then she changes it to the "o" in "on". And when I went to wikipedia, I got even more confused. They say that "ɔ" is the vowel sound in "caught", but it sounds like the vowel sound in "coat". Their "ɑ", they claim, is the sound in "cot", but it sounds like the one "long saw" to me. I don't distinguish between "cot" and "caught", so I figured I was cot-caught merged, but they claim that in the cot-caught merger, these sounds get merged, and they sound COMPLETELY different to me.

I also checked out MW: http://www.merriam-webster.com/cgi-bin/audio.pl?cot00001.wav=cot

That's, again, the long version of "saw".

Anyone have any other resources or comments?
Lazar   Mon Nov 17, 2008 12:57 am GMT
<<and the pronunciation of D (which I assume is suppsed to be the same as ɒ) is given at 3:32.>>

Well I think that's just the way she chose to write ɒ.

<<Alone, it's like a longer version of the "a" in saw, but then she changes it to the "o" in "on".>>

I'm not sure what you mean here. The sound in "saw" is [ɔ:], which is both longer and higher than the sound in "on" [ɒ]. If the [ɒ] doesn't sound exactly the same when she says it alone and when she says it in "on", that's understandable, because /ɒ/ is a checked vowel phoneme in RP (i.e. must always be followed by a consonant), so it's unnatural to produce [ɒ] in isolation. But in any case, her isolated [ɒ] and her [ɒ] in "on" seem basically the same: a short open-back rounded vowel.

<<They say that "ɔ" is the vowel sound in "caught", but it sounds like the vowel sound in "coat".>>

You mean her [ɔ:] sounds like the vowel sound in "coat"? If you're American, that may be because RP uses a tenser vowel in "caught", at or above cardinal [ɔ]. The American English vowel sound in (caught-caught unmerged) "caught" is often defined as [ɔ], but it's usually an opener vowel between cardinal [ɔ] and [ɒ]. Longman's pronouncing dictionaries transcribe the British and American pronunciations as [ɔ:] and [ɒ:], respectively, to convey this distinction. In turn, American English tends to use a back vowel in "coat" - [ɤʊ] or [oʊ] or [o:] - whereas RP uses a more centralized vowel - [əʊ] or [ɜʊ].

<<Their "ɑ", they claim, is the sound in "cot",>>

In American English, not in British English.

<<but it sounds like the one "long saw" to me.>>

Sorry, I'm not sure I get what you're saying.

<<I don't distinguish between "cot" and "caught", so I figured I was cot-caught merged,>>

Well if you don't distinguish between them, you're definitely merged. (Just out of curiosity, do you distinguish between the vowels in "father" and "bother"? This is a New England feature which might complicate things a bit; most cot-caught merged Americans use one vowel for all of those words.)

<<but they claim that in the cot-caught merger, these sounds get merged, and they sound COMPLETELY different to me.>>

Well the fact that a certain phonemic distinction doesn't exist in your dialect doesn't mean that you can't distinguish between given sounds phonetically.

<<That's, again, the long version of "saw".>>

The vowel in the m-w.com sound file is [ɑ].
Another Guest   Mon Nov 17, 2008 11:01 pm GMT
<<You mean her [ɔ:] sounds like the vowel sound in "coat"? >>
Wikipedia's [ɔ] sounds like the vowel sound in "coat". And as far as I can tell, all of Wikipedia's vowel audio samples are the same regardless of whether they have the colon after them or not.

<<Sorry, I'm not sure I get what you're saying. >>
I meant that it sounds like a long version of the vowel in "saw".

<<Well if you don't distinguish between them, you're definitely merged. >>
I might be cot-caught merged, but I'm definitely not [ɑ]-[ɔ] merged.

<<(Just out of curiosity, do you distinguish between the vowels in "father" and "bother"?>>
I don't think so.

<<Well the fact that a certain phonemic distinction doesn't exist in your dialect doesn't mean that you can't distinguish between given sounds phonetically. >>
When I say that they're completely different, I mean that I hear them as separate vowels. To me, [kɑt] sounds like a word somewhere between "cat" and "cot", and [kɔt] sounds like "coat", albeit with a funny accent. The idea of [kɑt] and [kɔt] being homophones is completely foreign to me.
Westerner   Mon Nov 17, 2008 11:42 pm GMT
>> When I say that they're completely different, I mean that I hear them as separate vowels. To me, [kɑt] sounds like a word somewhere between "cat" and "cot", and [kɔt] sounds like "coat", albeit with a funny accent. The idea of [kɑt] and [kɔt] being homophones is completely foreign to me. <<

Here's how I pronounce those:

cat /kęt/
cot /kɑt/
caught /kɑt/
coat /koUt/

on /ɑn/
saw /sɑ/

/ɑ/ can be either [ɑ]or [ɒ]
Another Guest   Tue Nov 18, 2008 2:24 am GMT
Here is M-W's audio file for caught: http://www.merriam-webster.com/cgi-bin/audio.pl?catch002.wav=caught

It doesn't sound at all like Wikipedia's. According to WP:

"For merged speakers in Canada and most of the United States, the two sounds [ɑ] and [ɔ] are allophones; they often do not perceive differences in their usage, hear neither of them as a separate phoneme, and hear the distinct vowels used by speakers whose dialects do distinguish them as variations on the same vowel."

That doesn't describe me at all.
Point   Tue Nov 18, 2008 4:36 am GMT
Conservative General American (bother-father merged)
cat /kęt/
cot /kɑt/
caught /kɔt/ (/ɔ/ is realized as [ɒ])
coat /koUt/

on /ɑn/
saw /sɔ/ (/ɔ/ is realized as [ɒ])

Canadian/Western US (Bother-father-cot-caught merged)
cat /kęt/
cot /kɑt/
caught /kɑt/
coat /koUt/

on /ɑn/
saw /sɑ/

/ɑ/ can be either [ɑ] or [ɒ]

Irish-English
cat /kat/
cot /kɑt/
caught /kɑt/
coat /koUt/

on /ɑn/
saw /sɑ/
Point   Tue Nov 18, 2008 4:39 am GMT
Northern Cities vowel shift (e.g. Chicago, Detroit, and other cities in the Northern US (but not Canada))

cat /kięt/
cot /kat/
caught /kɔt/ (/ɔ/ is realized as [ɑ])
coat /ko(U)t/

on /an/
saw /sɔ/ (/ɔ/ is realized as [ɑ])
Pronunciation   Tue Nov 18, 2008 4:50 am GMT
Here is the official pronunciation of those words:
http://www5.zippyshare.com/v/30748296/file.html
Lazar   Tue Nov 18, 2008 4:59 am GMT
<<And as far as I can tell, all of Wikipedia's vowel audio samples are the same regardless of whether they have the colon after them or not.>>

Well I put the length mark there because as an English vowel, it's generally considered to be rendered long. But it's true that the length is immaterial to the quality.

<<Wikipedia's [ɔ] sounds like the vowel sound in "coat".>>

Yes, I found the sound sample in question and listened to it. As I said in my previous post, your perception of this vowel as being more like the "coat vowel" is related to the fact that cardinal [ɔ] is tenser than the vowel that's generally rendered as [ɔ] in English. Reread what I wrote about the "caught" vowel.

It's not that Wikipedia is mistaken, and the vowel in the sample is not quite the same as the "coat" vowel for most English speakers, but it's rather close to it (from your perspective as an AmEng speaker), so it makes sense that you perceive it that way. (And I think the Wikipedia sound sample might be just a bit tenser than true cardinal [ɔ], analagous to how [ɔ] is often rendered in British English. I fear that I'm needlessly complicating things, though.)

<<I meant that it sounds like a long version of the vowel in "saw".>>

You mean, a long version of the vowel that you use in "saw"? A British English speaker, for example, would use a different vowel from what you use there.

<<I might be cot-caught merged, but I'm definitely not [ɑ]-[ɔ] merged.>>

I'm not sure what you mean; are you saying that you have a distinct phoneme /ɔ/ that occurs in words natively in your dialect? (If that's the case, then you might be a New Englander like me; do you use separate vowels in "father" and "bother"?) Or are you just saying that you can perceive a difference between those two sounds phonetically?

<<When I say that they're completely different, I mean that I hear them as separate vowels. To me, [kɑt] sounds like a word somewhere between "cat" and "cot", and [kɔt] sounds like "coat", albeit with a funny accent. The idea of [kɑt] and [kɔt] being homophones is completely foreign to me.>>

I think you're mixing up phonemics and phonetics; let me try to clarify.

<<It doesn't sound at all like Wikipedia's.>>

Again, you need to reread what I wrote about the "caught" vowel in my previous post. The reason why the MW sound in "caught" doesn't sound at all like Wikipedia's [ɔ] is because it's not [ɔ]; I would render the MW sound as [ɒ:]. When Wikipedia talks about a vowel /ɔ/ in American English, they're referring to a specific phoneme (a contrastive unit of sound) using a conventional (but in my opinion, flawed) representation. This vowel phoneme - to the extent that it occurs as a distinct phoneme in American English - is usually rendered in real phonetic terms as something like [ɒ:], not as [ɔ:].

<<"For merged speakers in Canada and most of the United States, the two sounds [ɑ] and [ɔ] are allophones; they often do not perceive differences in their usage, hear neither of them as a separate phoneme, and hear the distinct vowels used by speakers whose dialects do distinguish them as variations on the same vowel."

That doesn't describe me at all.>>

I'm sorry, but that's just sloppiness and ineptitude on the part of Wikipedia. (An example of why I've basically given up trying to improve Wikipedia, and why I have little respect for it as a source of accurate information.) Whoever wrote that passage is, themself, conflating phonemics and phonetics and making crude assumptions. The only substantive thing being said here is that some speakers have merged two phonemes. The assertion that most merged speakers would truly use [ɑ] and [ɔ] as allophones is just nonsense; those vowels are far too far apart phonetically. What the writer is trying (and failing) to convey is that once the phonemes have merged, there's room for some allophonic variation, and merged speakers might render the same phoneme as [ɑ] or [ɒ] in a state of free or conditioned variation. The bit about merged speakers not perceiving the unmerged phonemes as distinct might apply in some situations, if the unmerged speaker being listened to was an American who realized the two phonemes as [ɑ:] and [ɒ:], respectively, because those two sounds can sometimes be difficult to distinguish; but to suggest that a merged speaker magically becomes deaf to the phonetic difference between [ɑ] and [ɔ] is just utter nonsense. Any English speaker could perceive the phonetic difference between RP "car" and RP "core", the former using [ɑ:] and the latter using a rather tense [ɔ:].
Lazar   Tue Nov 18, 2008 5:01 am GMT
Sorry, this sentence:

<< As I said in my previous post, your perception of this vowel as being more like the "coat vowel" is related to the fact that cardinal [ɔ] is tenser than the vowel that's generally rendered as [ɔ] in English. >>

should read:

<< As I said in my previous post, your perception of this vowel as being more like the "coat vowel" is related to the fact that cardinal [ɔ] is tenser than the vowel that's generally rendered as [ɔ] in American English. >>
Travis   Tue Nov 18, 2008 5:39 am GMT
>>
Northern Cities vowel shift (e.g. Chicago, Detroit, and other cities in the Northern US (but not Canada))

cat /kięt/
cot /kat/
caught /kɔt/ (/ɔ/ is realized as [ɑ])
coat /ko(U)t/

on /an/
saw /sɔ/ (/ɔ/ is realized as [ɑ])<<

Note that there can be differences in places with respect to such, as shown by the above as pronounced here in Milwaukee, WI:

cat [kʰɛ̞ʔ(t)] or sometimes [kʰe̯ę̝ʔ(t)] when stressed in progressive speechc
cot [kʰaʔ(t)]
cought [kʰɒʔ(t)]
coat [kʰoʔ(t)]

on [ãːn]
saw [sɒː]

Note that I am using phonetic transcription here as with respect to these kinds of matters trying to just transcribe phonemes is generally of little use with respect to actually indicating how things are pronounced.
Trawicks   Tue Nov 18, 2008 4:43 pm GMT
<<Any English speaker could perceive the phonetic difference between RP "car" and RP "core", the former using [ɑ:] and the latter using a rather tense [ɔ:].>>

Agreed. I know many stage actors from the Western US who make no distinction between the two in their OWN dialect, but certainly have no problem separating them when they're doing, say, a character who speaks RP. So the idea that their brains are unable to comprehend the difference is sheer nonsense.

It IS true that the human brain has a hard time not filtering another person's dialect through the structure of their own. A classic example is the way that the "classic" New York pronunciation of "thirty third" street is rendered as "toity toid street." Now, in reality, the "oy" in words like that is actually pronounced something like [ɜɘ]--a far cry from [ɔɪ]. But because "oy" is the only non-frontal dipthong in GenAm, most Americans tend to perceive the centralized dipthong in Old New York "third" as "oy." Americans also do the same with the classic Irish [əɪ] and Cockney [ɑɪ] renderings of the /ai/ phoneme.
Milton   Tue Nov 18, 2008 7:32 pm GMT
-'saw [sɑ], law [lɑ], doll [dɑl], dollar ['dɑlər], caller/collar ['kɑlər], cot/caught [kɑt], don/dawn [dɑn], Rexall [reksɑl], wrong [rɑŋ]'- are preferred pronunciations in Traditional California dialect, and in urban Atlantic Canada (Halifax, St. John's), but they are also found in many areas of the West, such as in Tuscon, Arizona and are still used by some Southern Ontarians (mainly those from Windsor Ontario).

Valley Girlese, rural Washington state and Western Canada (BC, Alberta) like rounding [ɑ] to [ɒ]: saw [sɒ], doll [dɒl], dollar ['dɒlər], don/dawn [dɒn]. Canadian Oxford dictionary uses the rounded vowel, reflecting the Western Canadian preference.

Cambridge Advanced learner's dictionary [available online], sticks with the Traditional Californian pronunciation, and gives the unrounded [ɑ] in all these words, even in the newest edition (released a month ago).
Travis   Tue Nov 18, 2008 7:52 pm GMT
>>Cambridge Advanced learner's dictionary [available online], sticks with the Traditional Californian pronunciation, and gives the unrounded [ɑ] in all these words, even in the newest edition (released a month ago).<<

I do not get why an dictionary that is supposed to be for "American English" insists only only providing cot-caught-merged pronunciations, when a slight majority of NAE-speakers still maintain the cot-caught distinction (at least from what I remember when there was a discussion about just how widespread such is). Why impose cot-caught-merged pronunciation upon individuals who would use such dictionaries, such as non-native learners of English, when they could still always easily find the merged pronunciations from unmerged specified pronunciations if they wanted to?
feati   Tue Nov 18, 2008 8:37 pm GMT
My Longman DCE gives /ɔ:/ for British English and /ɒ:/ for American English in words like "caught".

Why do people write confusing things like "/kɔt/ (/ɔ/ is realized as [ɒ])" and not simply "/kɒt/" or "[kɒt]"? It's not like [ɒ] is an allophone of /ɔ/ in American English... Or is it?