zompist bboard

My dialect has something that is best called the gulf-golf merger, namely /ʌl/ > /ɔl/ before fortis obstruents in a limited set of forms, namely gulf, ultra-, multi-, multiply, bulk*, and hulk*. At the same time, in my dialect at least, the forms mulch and sulk have /ʌl/.

* I have /ʌl/ here, but both of my parents have /ɔl/ therein.

The thing that I am wondering about this is the diachronics. Namely, /l/ in m dialect does not normally round vowels before it, but rather assimilates in rounding to that before and that after it. OTOH this is an example of /l/ triggering backing, and /l/ does tend to encourage backing in my dialect. Yet this appears to indeed be a case of /l/ triggering both backing and rounding. (I presume this sound change precedes, ɔ > ɒ in my dialect, which would explain why /l/ seemingly lowers the vowel, unlike how /l/ normally behaves in my dialect.) So this is making me wonder whether this was a sound change that originally occurred in some other variety if English, one where /l/ does encourage rounding, and was only later borrowed into my dialect - or whether /l/ formerly encouraged rounding in my dialect or its predecessors for that to only be lost in more recent times.

Isn't your use of /ɔ/ instead of /ɑ/ in "golf" an example of /l/ triggering rounding on a preceding vowel? I believe the vowel of "golf" was historically in the LOT set, or at least, that that is the pronunciation that is ancestral to the modern standard pronunciations in Britain and North America. Most, if not all, speakers of American English seem to diachronically have ɒl > ɔl / _{#,C}. (I haven't observed much about the frequency of this because I have the cot-caught merger, and I don't know many people who don't have the merger.)

Sumelic wrote:Isn't your use of /ɔ/ instead of /ɑ/ in "golf" an example of /l/ triggering rounding on a preceding vowel? I believe the vowel of "golf" was historically in the LOT set, or at least, that that is the pronunciation that is ancestral to the modern standard pronunciations in Britain and North America. Most, if not all, speakers of American English seem to diachronically have ɒl > ɔl / _{#,C}. (I haven't observed much about the frequency of this because I have the cot-caught merger, and I don't know many people who don't have the merger.)

That is only if you treat ɒ > ɑ as preceding (ɒ or ɑ) > ɔ, which I have seen no evidence for.

As I recall reading, there has never been a sound change "ɒ > ɑ" in English to begin with — the LOT vowel instead went Middle English /ɔ/ > Early Modern English /ɑ/ > RP /ɒ/. (This would have been preceded by the CLOTH vowel going /ɑ/ > /ɑː/ > /ɔː/.)

Anyway, you mention that this "is an example of backing", so should I infer that your STRUT vowel is something like [ɜ] rather than [ʌ]? If so, it's phonologically entirely reasonable to suppose that ul gets first retracted to [ʌɫ], and this [ʌ] would then proceed to spontaneously labialize into [ɔ].

Tropylium wrote:As I recall reading, there has never been a sound change "ɒ > ɑ" in English to begin with — the LOT vowel instead went Middle English /ɔ/ > Early Modern English /ɑ/ > RP /ɒ/. (This would have been preceded by the CLOTH vowel going /ɑ/ > /ɑː/ > /ɔː/.)

I have heard a variety of things about the low back vowels in English, but I have never heard that the LOT vowel unrounded across the board to only re-round later.

Tropylium wrote:Anyway, you mention that this "is an example of backing", so should I infer that your STRUT vowel is something like [ɜ] rather than [ʌ]? If so, it's phonologically entirely reasonable to suppose that ul gets first retracted to [ʌɫ], and this [ʌ] would then proceed to spontaneously labialize into [ɔ].

The STRUT vowel in most English dialects aside from northern English English ones and Inland North ones is [ɐ] not [ʌ], and I presume my dialect had the same prior to the NCVS. I am positing that the sound change I describe here occurred before the NCVS because the GULF reflex lowers along with my reflex of /ɔ/, implying that the change was ɐ > ɔ and later ɔ > ɒ

If it helps at all, I've got the gulf-golf merger and I assume it's throughout Australian English. I remember talking to a friend of mine when I was young about how it's funny that we write "mulch" instead of "molch". For some reason, it didn't strike me as one of the many cases where an "o" is pronounced like a "u" (like in "love") but as something that we were all pronouncing "wrongly". The merger doesn't feel complete and I sometimes undo it but I think that's spelling pronunciation ... it doesn't feel natural to me, and the more I think about it, the more I don't even know what's more natural for me to say now.

I also appear to have a form of this merger, though the merged vowel can vary from /ʌ/ to /ɑ/. depending on how strongly the word is stressed.

TaylorS wrote:I also appear to have a form of this merger, though the merged vowel can vary from /ʌ/ to /ɑ/. depending on how strongly the word is stressed.

Are you cot-caught merged?

I noticed this merger a lot on American TV shows. I particularly remember on Star Trek Enterprise, T'Pol was always talking about the Volcans.

Travis B. wrote:

TaylorS wrote:I also appear to have a form of this merger, though the merged vowel can vary from /ʌ/ to /ɑ/. depending on how strongly the word is stressed.

Are you cot-caught merged?

Yes, my single low-back vowel varies from [ɑ] to [ɔ] depending on environment.

My girlfriend's dad, who is age 66, white, and from central Michigan (not too far from Travis B.) has merged gulf/golf, pronounces ultra "all-tra," multiply like "malt-iply," and the rest. This has been preserved in the younger generation, as my girlfriend (age 29) does it too. All of us are cot/caught unmerged; all of us suffer from Great Lakes Vowel Shift, especially in the back vowels.

There's something peculiar about his use of L's. On the one hand, he doesn't pronounce the L in palm or always, but he'll say "wheelbarrel," "yulp" instead of "yup" (yes), and generally merge unstressed /ow/ and /əl/. More data are needed. Has anyone heard about intrusive L in Michigan English? Or lateralization of back vowels? On an unrelated note, he also pronounces always /ɔweʃ/ or maybe even /ɔwɛʃ/.