From: rh@rudhar.com

Sat, 22 Mar 2025 20:30:24 +1300: Ross Clark 
scribeva:

>On 22/03/2025 7:59 p.m., Ruud Harmsen wrote:
>> Sat, 22 Mar 2025 09:46:53 +1300: Ross Clark 
>> scribeva:
>>> Here's how the phonemic analysis of AmEng that I was taught many years
>>> ago treats this:
>>> The vowel of the -ton syllable is [?]; it occurs only unstressed.
>>
>> Shwa. (My crappy old Agent program can see nor post IPA (although it
>> can post in UTF8). But I easily guessed what you posted, and confirmed
>> it by looking under the hood, in the data file. Linux IS fully Unicode
>> enabled.)
>>
>>> It's in complementary distribution with the phonetically similar [?] in
>>
>> Upside down v.
>>
>>> "gun" and "one", which occurs only stressed.
>>> So the two are allophones of one phoneme.
>>> (In the current pronunciation regime of OED, all three of these vowels
>>> are written as .*)
>>
>> Yes, I understand that’s the explanation. But I still think it’s a
>> weird rhyme, because of the stress difference,
>
>I would say it's a weird pronunciation of "Galveston", with an extra
>stress that shouldn't be there. But given that pronunciation, there's
>nothing wrong with the rhyme.

Yes, I can agree with that. For AmEng, that is. Even with that stress,
still largely unthinkable in South-Brit, I would think. But I cannot
speak for them, being a non-native speaker.

>and because in my view
>> (which is not mainstream and is not scientifically based, I know),
>> they are not the same phoneme. Being in complementary distribution
>> isn’t enough of a criterion for that.  and  are also in
>> complementary distribution, but clearly not the same phoneme, and they
>> couldn’t ever rhyme.
>
>But we know the answer to that one is that they are not phonetically
>similar. Whereas [?] and [?] certainly are.
>Personally, as a speaker of NAmEng, I consider the theory intuitively
>plausible. It also accounts for why, for many speakers, the stressed
>forms of words like "of" and "from" have [?].

Perhaps, yes.

>> I also consider the history of the language and the phonemes. I know
>> very well that according to any phonemic theory, and PTD, I shouldn’t,
>> but I do it anyway. The BUG vowel has an unrounded [o] realisation in
>> Northern England, which shwa could never have.  (when stressed)
>> and  and ,  and  have the same vowel there. The
>> origin and sound of shwa in English, as in Galveston, is totally
>> different and unconnected.
>
>A phonological version of the Etymological Fallacy?

(Had to look that up, didn't know the term, do recognise the
phenomenon.) Probably, yes, except that here of course I don’t
consider it a fallacy.

>> This also reminds me of a discussion we had years ago, about Memphis
>> sounding like Memphus, in a song sung by Cher. Unthinkable in
>> South-Brit. The THIS and THUS vowels are always distinct there.
>
>They're quite distinct for me, too. What you mean is that in S-B (RP?)

SB = anything not Canadian or US, inclusing Australian and
New-Zealandish, perhaps also South-African. What about Irish English?

>the (various) unstressed vowels have sorted themselves into just two
>groups, where as for me (and I guess most NAmEng speakers, and others)
>there's only one.

--
Ruud Harmsen, https://rudhar.com

--- SoupGate-DOS v1.05
 * Origin: you cannot sedate... all the things you hate (1:229/2)