Father-Bother distinction

<<No, that's not what I said - although I could have made myself clearer. As I've just indicated, I use the term "low back merger" to refer to the merger of /A:/, /Q/ and /O:/. So according to what you think I said, the low back merger depends on the phonetic quality of the vowels? If one speaker has two vowels [A:] and [Q:], he doesn't have the low back merger, but if [A:] is fronted to [a:] - even while preserving the same phonemic distribution - he suddenly has it? No, that makes no sense. If that's the case - that the term conflates phonemics and phonetics - then it should be abandoned because it's useless to us in any discussion of dialects. Indeed, you've certainly misunderstood me, because even though I use [A:] and [Q:] in my General-American influenced idiolect, my fellow New Englanders with a more traditional accent would use [a:] and [Q:] for the phonemes in question. They have the merger, but I don't, just on account of a slight difference of fronting? No, if that's what I thought the term meant, I wouldn't have used it.>>

Well I have [a] for "cot" and [A] for "caught". They're not really long vowels as you seem to indicate. Do you actually use long vowels in these cases?

There is a bit of allophonic vowel length allophony in my dialect (e.g., the vowel in "cot" is a bit shorter than the one in "cod", the one in "bet" is a bit shorter than the one in "bed"), but my free vowels - like /A:/, /Q:/ - tend to be consistently longer in any given situation than my checked vowels - like /E/, /I/. If I wanted to use a more narrow transcription, I could transcribe the vowel in "cod" as long, the vowels in "cot" and "bed" as half-long, and the vowel in "bet" as short.

Another reason why I use the length marks is that I like to make clear the difference between checked and free vowel phonemes. For example, transcribing the New England "cot, caught" vowel as /Q/ or [Q] often leads to the misconception that it's the same phoneme as /Q/ in RP "cot", which it is definitely not. RP /Q/ is a checked vowel phoneme that doesn't have any counterpart in New England (or almost anywhere in North America); ENE /Q:/ in "cot", even though it's qualitatively similar, is a free vowel phoneme that corresponds to RP /O:/ - opened in quality and expanded in distribution.

Sorry, strike the first "allophonic". :)

<<Another reason why I use the length marks is that I like to make clear the difference between checked and free vowel phonemes. For example, transcribing the New England "cot, caught" vowel as /Q/ or [Q] often leads to the misconception that it's the same phoneme as /Q/ in RP "cot", which it is definitely not. RP /Q/ is a checked vowel phoneme that doesn't have any counterpart in New England (or almost anywhere in North America); ENE /Q:/ in "cot", even though it's qualitatively similar, is a free vowel phoneme that corresponds to RP /O:/ - opened in quality and expanded in distribution.>>

I don't really have a [Q] in my dialect as I have the NCVS. [Q] is unrounded to [A]. [A] doesn't really need length marks as RP /A:/ is a free vowel.

The above Guest was me.

<<Indeed, you've certainly misunderstood me, because even though I use [A:] and [Q:] in my General-American influenced idiolect, my fellow New Englanders with a more traditional accent would use [a:] and [Q:] for the phonemes in question.>>

Just curious Lazar, but is the distinction between having "ah" as [a:] and having it as [A:] the distinction between r-dropping and r-pronouncing ENE English? Do all ENE r-droppers have [a:] and all ENE r-pronouncers have [A:]?

<<I don't really have a [Q] in my dialect as I have the NCVS. [Q] is unrounded to [A].>>

But in interdialectal phonemics, I would still write it as /Q:/, just as I would write the [a:] of a Bostonian or Worcesterite as /A:/.

<<[A] doesn't really need length marks as RP /A:/ is a free vowel.>>

No no no, it's not that I apply some arbitrary standard of RP non-ambiguity, where I would only use a length mark when it's necessary to avoid confusion with RP phonology - that would be crazy and RP-centric. No, it's just that I choose to systematically mark free vowels with a length mark. I only gave the example with RP to show why I think the length marks are useful. For your dialect, I could just as easily give an example of a Southern Irish dialect where [Q] has been unrounded to short [A] while remaining distinct from the "father" class.

In your speech, can you detect any difference in length between the vowels in pairs like "cod" and "bed"? If not, then I would not suggest that you use length marks. (It's partly for phonemic reasons that I like to use them in phonetic transcription of my own speech, but I wouldn't use them if they had no basis in reality.) But if you do detect a difference, then I think it may be useful (but by no means necessary) to put them in.

<<Just curious Lazar, but is the distinction between having "ah" as [a:] and having it as [A:] the distinction between r-dropping and r-pronouncing ENE English? Do all ENE r-droppers have [a:] and all ENE r-pronouncers have [A:]?>>

Yes. In my experience, non-rhotic speakers generally use [a:], and rhotic speakers like me generally use [A:].

<<Yes. In my experience, non-rhotic speakers generally use [a:], and rhotic speakers like me generally use [A:].>>

I have [a:] for some instances of my "ride" vowel contrasted with [A:] for my "father" vowel. For me, "ice" is [aIs] while "eyes" is [a:z].

I actually have some odd thing with my short o vowel. In certain words it gets a glide for instance "hog" [hAUg], "long" [lAUN], "soft" [sAUft] contrasted with, for instance, "lot" [lAt]. I wonder why this is.

You seem to have the "lot-cloth" split, in which certain instances of historical /Q/ have shifted to historical /O:/ - mostly before voiceless fricatives (/f/, /s/) and velars (/g/, /N/). It can also occur in "gone", and especially in the South, in "on". This split, although its distribution varies, seems to be practically universal among North Americans who have the cot-caught distinction. In General American, those words would be:

hog [hQ:g]
long [lQ:N]
soft [sQ:ft]
lot [lA:t]

This vowel [AU] - is this the same vowel that you use in "caught"?

<<This vowel [AU] - is this the same vowel that you use in "caught"?>>

Yes. I'm from Alabama.

What about "paws" and "pause"? I have a distinction with [pO:z] and [pO@z]. "pause" is homophonous with "Paul's".

Whoops. I got the order wrong. It's [pO@z] "paws" and [pO:z] "pause".

<<Are you from London? That looks a lot like the split that occurred there. Where RP has /O:/ word finally, London English has /O@/. The diphthong remains when suffixes are added, so you get minimal pairs like paws-pause, bored-board, etc.>>

Yes, I'm from London.