How are the true diphthongs of English distinguished from each other and other vowels with traditional distinctive features theory?

  • 1
    What is meant here by "true" diphthongs? For example, do you mean "phonemic" diphthongs? Commented Oct 26, 2014 at 5:13
  • Also the various Englishes have different repertoires of diphthongs. Commented Oct 26, 2014 at 6:38
  • Yes, by "true" diphthongs, I mean those that are distinctive/phonemic, i.e. /aɪ/, /aʊ/, and /oɪ/.
    – k8i
    Commented Oct 26, 2014 at 13:13
  • I'm thinking of standard North American English.
    – k8i
    Commented Oct 26, 2014 at 13:13

1 Answer 1


In American English, the "true" diphthongs are considered to be

  • /ay/, as in I, by, buy, try, five, tie, lye, hi, high, and jai alai
  • /aw/, as in ow, bough, how, brown, snout, out, trowel, and sauerkraut
  • /oy/, as in soil, toy, coin, Troy, ploy, oink, and hoi polloi

All of these are complex vowel clusters, starting at a phonemic low vowel (/o, a/), and continuing with a gesture toward a high front vowel /i/, represented as a semivowel /y/, or towards a high back vowel /u/, represented as a semivowel /w/.

The "untrue" diphthongs are the tense vowel phonemes /i, e, o, u/, which are predictably terminated by a gesture toward /i/ (for /e, i/) and toward /u/ (for /u, o/). These are not marked as diphthongs in Kenyon and Knott's transcription of American English phonemics, because they're predictable.

As for distinctive features, the precise representation depends on which version of "distinctive feature" theory one wishes to use, and what its constraints and presuppositions are. Consult your phonology instructor for details.

  • I should have been more precise in my question. I'm thinking of traditional Sound Pattern of English (SPE) features. I don't think the contrast can be captured in this framework. But even in a non-linear framework where the vowel can be represented as occupying two slots within the nucleus, if one were to try to use features to represent this, would it be [+low],[+high,-back] for /aɪ/ and [+low],[+high,+back] for /aʊ/?
    – k8i
    Commented Oct 26, 2014 at 13:22
  • A close reading of SPE will reveal that they actually do have an extensive treatment of the English diphthongs. Non-linear representations (as you propose) aren't different from SPE representations in this case. The SPE analysis of these diphthongs as deriving from /ī, ū, œ̄/ is compatible with nonlinear theory, and the only reason we wouldn't say that now is that we better understand the flaws in the original argument.
    – user6726
    Commented Apr 2, 2015 at 4:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.