Six-vowel system for English

[Rounin Ryuuji]

Pull wool full bull for me all have a rounded nucleus and a rounded offglide (from vocalized /l/) or even just a rounded overlong monophthong (where vocalized /l/ merges with the nucleus), so those cannot simply be cases of stressed schwas (since schwas followed by /l/ are unrounded for me).

Do (m)any people rime them with gull dull hull cull?

I certainly don't.

As far as I go, pull, wool, full, bull rhyme with each-other, [pʰʊɫ], [wʊɫ], [fʊɫ], [bʊɫ], but do not rhyme with with gull, dull, hull, cull, [gʌɫ], [dʌɫ], [hʌɫ], [kʰʌɫ], or with ghoul, fool, rule, [guwɫ], [fuwɫ], [ɹuwɫ] which don't rhyme perfectly with "tool, stool" [tʰuw.wəɫ], [stuw.wəɫ], now that I think on it.

[Alces]

My Lancashire dialect can be seen as basically a five-vowel system, due to the merger of the FOOT and STRUT lexical sets. There are only five monophthongs which are always pronounced short and are permitted to occur in stressed syllables. However there are a number of complications:

- Schwa can only occur in unstressed syllables and minimal pairs contrasting it with the other short monophthongs are rare or nonexistent; however, all five of the other short monophthongs can appear in unstressed syllables, and their appearance vs schwa is not phonologically predictable. So it's really a five-vowel system in stressed syllables and a six-vowel system in unstressed syllables.

- In terms of quality, the two back vowels are only weakly rounded, and are much lower than the corresponding cardinals. That is, the /o/ of LOT-CLOTH has more or less the same pronunciation as in SSBE, while the /u/ of FOOT-STRUT is something between the FOOT and STRUT vowels of SSBE (though closer to FOOT).

- To the two back vowels there correspond two vowels which are always pronounced long, and are close, but slightly different, in quality. The most prominent differences (I'm sure there'd probably also be minor height/backness distinctions if I measured them precisely) are that the long vowel corresponding to LOT-CLOTH, which is that of PALM-START, is completely unrounded, and the long vowel corresponding to FOOT-STRUT, which is that of THOUGHT-NORTH-FORCE-CURE, is more fully rounded. However these quality differences appear to be superseded by length perceptually: if I pronounce "part" short, it clearly sounds like "pot", and vice versa; if I pronounce "bought" short, it clearly sounds like "but", and vice versa. So there's a reasonable argument for seeing these long vowels as long counterparts of the corresponding short vowels. However, they were not counterparts in any sense historically, since of course PALM-START historically comes mostly from /a/ rather than /o/, and THOUGHT-NORTH-FORCE-CURE historically comes mostly from /o/ rather than /u/. Also, there's a third long vowel, the one in NURSE-SQUARE, which complicates the picture: when pronounced short, it sounds like schwa (though slightly more open), and schwa is not a proper member of the five-vowel system since it never occurs in stressed syllables. Admittedly, closer to Liverpool you can hear people with more or less the same accent as me except that they pronounce this vowel as [E:], thus making it a long counterpart of the DRESS vowel (and that one does actually work historically to some extent).

- Most of the diphthongs could be regarded as short monophthong + /j w/ sequences on perceptual grounds: /ej/, /uj/ (not /oj/: the first element clearly sounds much more like FOOT-STRUT than LOT-CLOTH, since LOT-CLOTH is so low), /@w/ (a bit awkward, like NURSE-SQUARE, since its first element is closest to schwa, although its first element is also quite rounded and to my ears this makes it sound quite distinct from schwa) /aj/, /aw/, /ij/, /uw/. However, the last four diphthongs are subject to a lengthening rule which seems to make most sense if they are unital segments, rather than monophthong + /j w/ sequences: /ij uw/ become [i: u:] (hardly diphthongized as far as I can tell, compared to when they're pronounced short) and /aj aw/ become [a:e] and [a:o] (with noticeable lowering of the second segment) before voiced consonants, with the exception of /awl/ which is still pronounced short oddly enough. The /@w/ diphthong is also backed to [ow] (with even the [w] sounding more back) before coda /l/, which is perhaps less problematic but it still seems neater to analyse this as happening to a unitary segment rather than a sequence /@w/.

[Nortaneous]

Also for some reason the word but seems to be losing its strong form, and can be /bət/ even when stressed.

I can't imagine that making any difference, so I guess I don't have distinct weak vowels. (And "Goethe" seems like [ˈgɝˌtʰʌ].) But writing English with distinct weak vowels makes the phonological representations nicer - if I were to write the weak vowels as /ʌ i/ instead of /ə ɨ/, I'd have to mark stress.

[Travis B.]

The only stressed "weak" vowels in my lect are /ə/ before /r/ in stressed syllables (as otherwise a syllabic rhotic phoneme would be needed) and KIT/COMMA, where in many positions there is no distinction made between KIT and COMMA (where they are phonetically identical and differ only in stress).

[anteallach]

Doesn't require contrastive syllabification, only requires relevance of morpheme boundaries.

No, because there are examples where the ones with the long vowels are monomorphemic, most famously Mary as opposed to merry. If not using contrastive syllabification you might have to make them /merri(j)/ and /meri(j)/ respectively, which amusingly is the reverse of where the orthography puts the double r.

Except here the geminate /r/ is ad-hoc, since there are no other geminates. I think you can still say "Mary" is /ˈmejri/ with /ej/ > [ɛː] before /r/, however.

But I don't think you can do anything analogous with sari. I think modern SSBE-type dialects do need the long vowels to be distinct somehow.

[anteallach]

- To the two back vowels there correspond two vowels which are always pronounced long, and are close, but slightly different, in quality. The most prominent differences (I'm sure there'd probably also be minor height/backness distinctions if I measured them precisely) are that the long vowel corresponding to LOT-CLOTH, which is that of PALM-START, is completely unrounded, and the long vowel corresponding to FOOT-STRUT, which is that of THOUGHT-NORTH-FORCE-CURE, is more fully rounded. However these quality differences appear to be superseded by length perceptually: if I pronounce "part" short, it clearly sounds like "pot", and vice versa; if I pronounce "bought" short, it clearly sounds like "but", and vice versa. So there's a reasonable argument for seeing these long vowels as long counterparts of the corresponding short vowels. However, they were not counterparts in any sense historically, since of course PALM-START historically comes mostly from /a/ rather than /o/, and THOUGHT-NORTH-FORCE-CURE historically comes mostly from /o/ rather than /u/. Also, there's a third long vowel, the one in NURSE-SQUARE, which complicates the picture: when pronounced short, it sounds like schwa (though slightly more open), and schwa is not a proper member of the five-vowel system since it never occurs in stressed syllables. Admittedly, closer to Liverpool you can hear people with more or less the same accent as me except that they pronounce this vowel as [E:], thus making it a long counterpart of the DRESS vowel (and that one does actually work historically to some extent).

What about NEAR? Bisyllabic sequence of FLEECE and schwa?

[Alces]

Yeah, "near", "kneel", "fire" and "pile" are bisyllabic for me, although when a vowel follows the liquid in the same morpheme, as in "Kieran", "pirate" or "pylon", I have monosyllabic long /i/ [i:] or /aj/ [a:e].

When I was younger I used to pronounce "Ireland" as ["a:el@nd], exactly homophonous with "island", but I've since deliberately started saying ["aj@l@nd] so as not to confuse people.

[KathTheDragon]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

[Richard W]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

But is that English?

[alice]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

But is that English?

No, and it is contrary to the will of the LORD, who sanctions only /mek.dər/.

[quinterbeck]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

As a non-programmer* I would have confidently pronounced this /ˈməkˌdɪː/, but yes, definitely not STRUT.

*I write a lot of SQL code for my job but it doesn't count

[KathTheDragon]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

But is that English?

Does it matter? It's being filtered through my idiolect's phonology.

[Travis B.]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

But is that English?

No, and it is contrary to the will of the LORD, who sanctions only /mek.dər/.

To me it's emphatically [ˈme̞ʔk̚tʁ̩ˤ(ː)].

[bradrn]

It’s not a word I need to say regularly, but I’d use [ˈmæ͡ikʰ.dɪɻʷ]. (Interestingly, this is rhotic, even though my dialect usually isn’t.)

I do, however, have at least one genuine example of a stressed weak vowel: Sydney, which I regularly pronounce [ˈsɨᵈni].

[Travis B.]

I do, however, have at least one genuine example of a stressed weak vowel: Sydney, which I regularly pronounce [ˈsɨᵈni].

I have a similar pronunciation, [ˈsɘːnːi(ː)], but for me this is one position where KIT and COMMA are pronounced exactly the same way, so it still counts as KIT.

[Alces]

My instinctive pronounciation for "mkdir" would be ["mejkd3:] (combination of bringing out the underlying "make" and spelling-pronouncing the "-ir"), but it's not a word I've ever used or heard in speech.

I would normally pronounce a stressed schwa in the names of speech sounds---I can say things like "C sounds like [s@] before I and E, but [k@] elsewhere".

Maybe the best example of a stressed schwa is "gonna". Obviously it's normally a word which is unstressed in the sentence, but I think it can end up with contrastive stress, maybe in a sentence like:

Nah, you're not gonna just try to do it, you're gonna do it---I believe in you!

[Vijay]

While at work today I discovered that I pronounce "mkdir" as /ˈməkˌdəː/ with a stressed ə that is definitely not the STRUT vowel.

But is that English?

Yes.

[Nortaneous]

Development of a new vowel phoneme through restressing of unstressed forms without reversal of vowel reduction probably happened in late Tocharian B, so it's not implausible.

[bradrn]

Development of a new vowel phoneme through restressing of unstressed forms without reversal of vowel reduction probably happened in late Tocharian B, so it's not implausible.

Also in Kalam, Yimas, Abelam, Selau, Tashlhiyt and probably Komnzo (Blevins 2016), so it’s not even uncommon if you look in the right place.

[Travis B.]

Maybe the best example of a stressed schwa is "gonna". Obviously it's normally a word which is unstressed in the sentence, but I think it can end up with contrastive stress, maybe in a sentence like:

Nah, you're not gonna just try to do it, you're gonna do it---I believe in you!

Gonna for me is a clear STRUT word, even when reduced to a clitic. Trying to pronounce it with COMMA for me makes it end up sounding like a KIT word.