4

As a teacher of languages, it has struck me how English vowels love not just diphthongs, but even triphthongs, and this tendency presents itself in how native English speakers generally tend to pronounce non-native languages and loanwords. This example from Hans H. Ørberg’s Lingua latina per se illustrata: Student’s manual (p. 4) exemplifies this:

Long vowels
ā as in ‘father’: ālā, pānis
ē as in Scottish ‘late’ (no diphthong!): mē
ī as ee in ‘feet’: hīc, līberī
ō as in Scottish ‘go’ (no diphthong!): pōnō
ū as in ‘fool’: ūna, tū
ȳ as French u in ‘pur’: Lȳdia
― Ørberg, Hans H.: Lingua latina per se illustrata : pars I : Latine disco : Student’s Manual, Focus, 2001: 4.

What is it about English phonology or phonotactics that lends these two long vowels, e and o, particularly prone to being diphthongised?

Canned Man
  • 804
  • 7
    In English, we don't have e and o in our language (except if we're from Scotland or a few other regions). The closest vowels, eɪ and oʊ, are diphthongs. So we naturally diphthongize them when we try to speak foreign languages. This shouldn't be any more surprising than the fact that French people pronounce "th" as "z" and "s", because "th" doesn't exist in French. We don't generally diphthongize /ɑː/ or short vowels. – Peter Shor Dec 09 '22 at 13:30
  • 4
    Or are you asking why these long vowels are diphthongs in English? They didn't use to be, so the answer to that undoubtedly lies in the history of English phonology. – Peter Shor Dec 09 '22 at 13:30
  • 1
    I'd guess they're easily diphthongized because they're intermediate in height and thus more prone to wander in both directions, compared to vowels at extremes which can only go in one direction. Although long /ɑ:/ is prone to wander or be shortened, /u:/ became /ju:/ in many cases, and middle English /i:/ became a diphthong, so I'm not sure the claim is true. – Stuart F Dec 09 '22 at 14:37
  • 1
    'Why' is difficult to answer in phonology (though that question does motivate a lot of research). I vaguely remember that there was a famous article by Charles Hockett that tried to explain why some languages have an 'f' and others don't, with the explanation being 'agriculture vs hunter-gatherer'. Eating bread made labiodental fricatives more likely? less? I can't remember? The paper was famous because it was a stand alone explanation... there have been no -other- supportable 'why' explanations of phonology. The best 'explanation' is usually more of a de facto history of what happened. – Mitch Dec 09 '22 at 15:48
  • 4
    Part of the Great Vowel Shift and its aftershocks was a general tendency to distinguish vowels by quality (i.e, tongue position) instead of length (i.e, the time the vowel was held). The GVS changed the quality of ME long vowels, and the perception of length persisted in some dialects by diphthongizing the moved (now tense) "long" vowels, resulting in [iy, ey, ow, uw] as normal allophones. These are not phonemic diphthongs like /aw, ay, oy/, but rather inevitable sources of palatalization or labialization. – John Lawler Dec 09 '22 at 17:25
  • @JohnLawler I think you are the closest one of the above to approaching an answer to the why. I find it odd that it is just these two vowels that so badly wants a vowel friend to join hands with. – Canned Man Dec 11 '22 at 00:32
  • Well, you and @StuartF. the height hypothesis is interesting. Is there a phonological and/or phonotactic theory behind it? – Canned Man Dec 11 '22 at 00:32
  • 4
    @CannedMan - it's not just /e/ and /o/ -- they're the most obvious, but /i/ and /u/ are also diphthongized; the tongue moves further toward the high vowel point. With a little attention one can hear it clearly. And those are the 4 vowels that are the outcome of the Great Vowel Shift, as they have fallen out in Modern English. – John Lawler Dec 11 '22 at 01:59
  • 1
    @PeterShor Monophthong /o/ and /e/ both famously occur in the Upper Midwestern / North Central speech of the Dakotas, Minnesota, Wisconsin, and northern Michigan. You'll also find these in many speakers from the Inland North, especially its western reaches, and in Canadian speakers across much of Ontario (eg, [toˈɹɒɾ̃o], [təˈɹɒɾ̃o], [ˈtɹɒɾ̃o], [ˈtɹɒɾ̃ə]) See John Lawler’s comment here. – tchrist Dec 11 '22 at 02:53
  • 4
    @JohnLawler I wonder if you could please post an answer for us. It doesn't have to be long and involved, just the basics like you've already outlined in comments. Your point that it’s not just [ej] and [ow] but *all* the tense vowels that have semi-consonantal off-glides *including* [ij] and [uw] is an important one. Why this key point is so often lost or omitted from most such presentations, I have no idea. It explains pre-L/R breaking and much more. I'd really love to see this question answered. – tchrist Dec 11 '22 at 03:12
  • 2
    @CannedMan This video of Dr Geoff Lindsey's about how we write and say the English "long" vowels might help. It explains that the way we these days most often write these "consonantal diphthongs" (vowel + semi-consonant) is extremely misleading and confusing, and that we used to write them more clearly. I think this is part of the picture that now seems so confusing. – tchrist Dec 11 '22 at 03:30
  • @tchrist: OK, delivered. – John Lawler Dec 11 '22 at 22:13
  • @tchrist Wow. I mean, who doesn’t like Dr Lindsey, but this is something I have longed to learn for years: All the things that just didn’t make sense to me, but I was unable to explain both to myself and others why that was. Now I just need to rewatch it a couple of times to be properly able to digest it. I think, maybe, that you could write an excellent answer from this perspective as well, not as a counter to the already good answer by Lawler, but as an addition. – Canned Man Dec 12 '22 at 18:44

1 Answers1

11

By request, I'll unpack what I said in the comment above.

Part of the Great Vowel Shift and its aftershocks

which occurred between about 1400 and 1700, depending on where (and what variety of) English was being spoken. These things take a loooong time.

was a general tendency to distinguish vowels by quality (i.e, tongue position)

Like the vowels in beat and bit are different in quality, and are also different from the vowels in bait, bet, bat, bot, bought, boat, but, boot, and foot. They're all distinguished by where in the mouth the tongue is; that's vowel quality.

instead of length (i.e, the time the vowel was held).

Old and Middle English had, like German and French, long and short vowels. The terms "long vowel" and "short vowel" are often used in grammar school to teach English reading, but they're not really long or short.

If you're a native English speaker, you were probably taught that there were 5 vowels, and they came in long and short varieties. This theory describes Middle English, which did have a, e, i, o, and u occurring both short and long (i.e, held twice as long as short), though Middle English also had a long and short æ vowel that got lost during the GVS. However, it does not describe Modern English, which has no distinctive vowel length, but does have around 13 qualitatively distinct vowels. Middle English was the language that English spelling was designed for, so this explains a great deal.

The GVS changed the quality of ME long vowels, but the perception of length persisted in some dialects by diphthongizing the moved "long" vowels, now phonemic tense /i, e, o, u/, resulting in [iy, ey, ow, uw] as normal allophones of the tense vowels.

[w] is the labializing offglide for tense rounded vowels /o/ and /u/ (also for back vowels, because English back vowels are rounded, whereas front vowels aren't); and [y] is the palatalizing offglide for tense front vowels /e/ and /i/. These offglides affect what comes after the vowels.

This can be perceived easily by observing the behavior of tense vowels before resonants, especially /l/ (/r/ has other issues after vowels). Many English speakers use an epenthetic schwa in words like fool, feel, foal, fail.

I.e, in these words, for these speakers, there are two syllables -- the first stressed syllable with a tense vowel -- and the second, short syllable, consisting of a reduced vowel ([ə] or [ɨ]) and the final /l/. In essence, a syllabic resonant.

  • In fail and feel, there's a /y/ between the two syllables: ['fiyɨl] and ['feyɨl]
  • In fool and foal, there's a /w/: ['fuwɨl] and ['fowɨl].

Indeed, the phenomenon of tense vowel diphthongs is one of the major features of an English accent in other languages. Any native English speaker who can learn to say pure tense vowels [i:, e:, o:, u:] at normal speech rates, without diphthongizing their vowels, can improve their accent greatly in any European language, for instance, because none of them have this feature, but rather use pure vowels throughout, except for phonemic diphthongs like /ay/ and /oy/. And /ow/ and /ey/ are likely to occur and be phonemically distinct from /o/ and /e/ in those languages, anyway.

John Lawler
  • 107,887
  • 1
    I might add that, in the same piece that identifies the start of the Great Vowel Shift (on May 5, 1403), Jim McCawley also mentions May 19 as Diphthong Day, a public holiday in Australia. – John Lawler Dec 12 '22 at 17:04
  • 1
    This is a very good answer. If you at the end can connect what you have laid out closer to the question at hand, it will be a great answer. I am not surprised to read the GVS as the answer, or, said another way: I am not surprised to read the replacement of vowel quantity for quality to be the answer (allowing the diphthongising tendency to be understood as part of the vowel’s quality (the type of diphthongisation is, after all, closely tied to the vowel’s inherent quality)). – Canned Man Dec 12 '22 at 18:38
  • Also, reading the comment from @tchrist somewhat clarified your answer (https://english.stackexchange.com/questions/599686/why-are-long-e-and-o-most-prone-to-be-diphthongised-by-english-speakers?noredirect=1#comment1510095_599686). Some minor additions would make your answer more self-contained. – Canned Man Dec 12 '22 at 18:41
  • Couldn’t bought and boat have the same monophthong in some Northern English and Scottish dialects? – Canned Man Dec 15 '22 at 14:55
  • @CannedMan I can't imagine how monophthong /o/ and /ɔ/ would ever be neutralized in that position where they fall in boat and bought. This isn't the story case. – tchrist Dec 15 '22 at 22:48
  • By the way, if you're looking for a canonic answer, nihil obstat. Imprimatur. – John Lawler Dec 17 '22 at 00:02