I'm from Alabama and I pronounce "water" as /wAt@`/. I have however heard older people in my region have various different pronunciations for it, while people my age here pronounce it as I do.
"water"
I've heard a lot of Americans pronounce it /wO4@'/. I pronounce it /wA4@'/, but I'm cot-caught merged, so it wouldn't make a difference either way. Are you c-c merged too?
I myself pronounce "water" as [wQ4R=:] (that's really analogous to GA [wO4@`] but with a lower main vowel and a different rhotic).
One note: you guys are using // here, which actually indicates phonemic transcription and not phonetic transcription, while your transcriptions clearly indicate that you mean to transcribe sounds, due to [4] and [@`] not being distinct phonemes within North American English dialects. For transcribing sounds rather than phonemes, what you really should be using here instead of // is [].
One note: you guys are using // here, which actually indicates phonemic transcription and not phonetic transcription, while your transcriptions clearly indicate that you mean to transcribe sounds, due to [4] and [@`] not being distinct phonemes within North American English dialects. For transcribing sounds rather than phonemes, what you really should be using here instead of // is [].
I pronounce "water" as ["wQ4@`] (although I'm debating whether I should transcribe it as ["wQ:4@`]).
<<I've heard a lot of Americans pronounce it /wO4@'/. I pronounce it /wA4@'/, but I'm cot-caught merged, so it wouldn't make a difference either way. Are you c-c merged too?>>
I pronounce "cot" and "caught" the same way. Both are /kAt/. I've heard /wot@`/ and /wOrt@`/ for "water" among older speakers in my region.
<<due to [4] and [@`] not being distinct phonemes within North American English dialects.>>
Well, [4] isn't, but [@`] clearly is. I distinguish "tuna" and "tuner" after all. My speech is not nonrhotic.
I pronounce "cot" and "caught" the same way. Both are /kAt/. I've heard /wot@`/ and /wOrt@`/ for "water" among older speakers in my region.
<<due to [4] and [@`] not being distinct phonemes within North American English dialects.>>
Well, [4] isn't, but [@`] clearly is. I distinguish "tuna" and "tuner" after all. My speech is not nonrhotic.
>>Well, [4] isn't, but [@`] clearly is. I distinguish "tuna" and "tuner" after all. My speech is not nonrhotic.<<
There are reasons why [@`] in North American English (the same goes for my own [R=]) should be analyzed as the sequence of phonemes /@r/ and not a single phoneme /@`/. Such are that there is a hole in the distribution of vowels before /r/ where the sequence [@r\] never appears except when followed by a stressed vowel, and then such is always interchangeable with [@`] and never contrasts with it.
Due to the complete lack of any contrast between [@r\] and [@`] combined with [@r\] still appearing optionally in pre-stress positions, it would make sense to posit the underlying form here to be /@r/ rather than to posit some extra phoneme /@`/ which happens to optionally get shifted to [@r\] while leaving a hole in phoneme distribution for /@r/. Consequently, the rule would be that /@r/ would be shifted to [@`] in all positions except before stressed vowels, where is remains in free variation with [@`] as the unshifted [@r\].
There are reasons why [@`] in North American English (the same goes for my own [R=]) should be analyzed as the sequence of phonemes /@r/ and not a single phoneme /@`/. Such are that there is a hole in the distribution of vowels before /r/ where the sequence [@r\] never appears except when followed by a stressed vowel, and then such is always interchangeable with [@`] and never contrasts with it.
Due to the complete lack of any contrast between [@r\] and [@`] combined with [@r\] still appearing optionally in pre-stress positions, it would make sense to posit the underlying form here to be /@r/ rather than to posit some extra phoneme /@`/ which happens to optionally get shifted to [@r\] while leaving a hole in phoneme distribution for /@r/. Consequently, the rule would be that /@r/ would be shifted to [@`] in all positions except before stressed vowels, where is remains in free variation with [@`] as the unshifted [@r\].
One note: the above also makes sense that other vowels before /r/ will also at times reduce along with /r/ when unstressed to make [@`] in NAE dialects indicating that [@`] can come about through the reduction of other vowels and not just the inherently reduced vowel /@/ before /r/. Consequently, it would not make sense to make a special case exception for [@`] as a single phoneme /@`/ while leaving a whole distribution-wise for /@r/ if other vowel-/r/ sequences can bring about [@`] as well.
That should be "Consequently, it would not make sense to make a special case exception for [@`] not associated with alternation with another unreduced vowel as a single phoneme /@`/ while leaving a hole distribution-wise for the inherently reduced /@r/ if other vowel-/r/ sequences which can undergo vowel reduction can bring about [@`] as well."
<<There are reasons why [@`] in North American English (the same goes for my own [R=]) should be analyzed as the sequence of phonemes /@r/ and not a single phoneme /@`/. Such are that there is a hole in the distribution of vowels before /r/ where the sequence [@r\] never appears except when followed by a stressed vowel, and then such is always interchangeable with [@`] and never contrasts with it.>>
That's interesting with regard to your comment about [@`] in North American English. I'm from Jamaica and my speech is partially nonrhotic where historical /@r/ has become /@/ for me finally and before consonants. For instance, I pronounce "cover" as /kVv@/.
However, I do have [@`] in some words such as "ignorant" [Ign@`n] and "separate" (adjective) [sEp@`t]. Would [@`] be a phoneme for me?
That's interesting with regard to your comment about [@`] in North American English. I'm from Jamaica and my speech is partially nonrhotic where historical /@r/ has become /@/ for me finally and before consonants. For instance, I pronounce "cover" as /kVv@/.
However, I do have [@`] in some words such as "ignorant" [Ign@`n] and "separate" (adjective) [sEp@`t]. Would [@`] be a phoneme for me?
>>That's interesting with regard to your comment about [@`] in North American English. I'm from Jamaica and my speech is partially nonrhotic where historical /@r/ has become /@/ for me finally and before consonants. For instance, I pronounce "cover" as /kVv@/.
However, I do have [@`] in some words such as "ignorant" [Ign@`n] and "separate" (adjective) [sEp@`t]. Would [@`] be a phoneme for me?<<
I would probably guess that what happened is the elision of /r/ having occurred in a pattern similar to with other non-rhotic dialects, then some vowels after the surviving instances of /r/ having been elided so that /@r/ with /r/ in a syllable coda became possible again, and finally a rule was introduced so that said /@r/ would be realized as [@`], similar to North American English. However, I do not know enough about the specifics of English dialects in the Caribbean to be certain here.
(Note that when I speak of North American English, I am specifically referring to the English dialect continuum covering the continental US and Canada, and not including the Caribbean or Honduras, which are not part of said dialect continuum.)
However, I do have [@`] in some words such as "ignorant" [Ign@`n] and "separate" (adjective) [sEp@`t]. Would [@`] be a phoneme for me?<<
I would probably guess that what happened is the elision of /r/ having occurred in a pattern similar to with other non-rhotic dialects, then some vowels after the surviving instances of /r/ having been elided so that /@r/ with /r/ in a syllable coda became possible again, and finally a rule was introduced so that said /@r/ would be realized as [@`], similar to North American English. However, I do not know enough about the specifics of English dialects in the Caribbean to be certain here.
(Note that when I speak of North American English, I am specifically referring to the English dialect continuum covering the continental US and Canada, and not including the Caribbean or Honduras, which are not part of said dialect continuum.)
Thanks Travis, you're right: I should have used square brackets instead of slashes. Just sloppiness on my part. In regards to Megabox's comment, I'm wondering where you stand in the scale of speech from basilect to mesolect to acrolect. From my understanding, basilectal speakers are completely (or nearly) non-rhotic, and post-vocalic r was re-introduced into mesolectal varieties and increases in frequency towards the acrolect. I don't think that there is a separate phoneme /@'/, in this situation, since it's occurence can be predicted. If we have a sequence /@r/ (I'll just use r for r\ since we're only dealing with English), we can predict three realizations: [@], [@r], and [@']. The first occurs at the end of a word or before a consonant; the second, before another vowel (possibly only stressed vowels?). The third is something I've never noticed in Jamaican speech (though you would know better than me!). It seems that when the following vowel is elided, [@r] coalesces into [@']. So an underlying /@r/ +V (a vowel) becomes [@']. Also, just to clarify, in "ignorant", do you have a syllabic nasal at the end or just [@'n] as one syllable? (like the last syllable in 'cistern')
<<Also, just to clarify, in "ignorant", do you have a syllabic nasal at the end or just [@'n] as one syllable? (like the last syllable in 'cistern')>>
"ignorant" is two syllables for me. There's no syllable "n" at the end. "cistern" on the other hand, is /sIst@n/ for me, and doesn't really rhyme with my "ignorant".
"ignorant" is two syllables for me. There's no syllable "n" at the end. "cistern" on the other hand, is /sIst@n/ for me, and doesn't really rhyme with my "ignorant".
When you use [@'], don't you really mean [@`] not [@']? I don't know what [@'] would be.