I spent some time on a research project examining spectrograms and coding vowels for speakers of American English from a few rural regions in the state of Oklahoma.

I noticed that certain speakers seemed to consistently target 3 separate peaks (or articulator positions) on certain vowels, such as the /ɛ/ in a word like /hɛn/ (and to my ear such vowels sounded noticeably different than the English varieties from my AE region). While I expected to find diphthongs among the data sets, it really surprised me to find a hint of triphthongs in an English variety. I didn't find much research out there on whether that is actually some sort of a diphthong-glide combination or a "real" triphthong (where the speaker has 3 distinct target areas for the vowel), and I suspect that it is a very limited situation in English, and not a clear example of triphthongs.

What I really wondered after the project was if there are languages where triphthongs are a regular feature? Could someone please tell me an example of a language or dialect that contains triphthongs? I would love to know. Thank you!

  • There are many, indeed: Thai เขียว [kʰǐːau] (green), Chinese 牛 [niú] (cow), off the top of my mind. Commented Dec 2, 2014 at 17:20
  • 1
    What are you counting as "the vowel"? Most of the examples given by others thus far wouldn't count as triphthongs in my book because I would either consider them to cover multiple syllables or I would consider the glide at the beginning of the purported triphthongal sequence to be part of the onset and thus not part of the nucleus of the syllable. But of course it all depends on one's model. Commented Dec 2, 2014 at 17:55
  • 2
    Can you provide your best attempt at the phonetic transcription for the realization of hen that you are talking about? As an American I'm pretty sure I know what you are talking about but it would probably be helpful for others. I mention this because I could understand arguments for considering that kind of sequence, as opposed to the glide-vowel sequences that others are offering as examples, to be a true triphthong. Commented Dec 2, 2014 at 18:05
  • @musicallinguist You have a point there. I'll be deleting the answer for now and looking more into it if I have some time.
    – Alenanno
    Commented Dec 2, 2014 at 18:18
  • @musicallinguist by "the vowel," I am really only referring to examples where the segment in question is clearly just covering one syllable. As for whether or not there is a glide effect involved at the onset or coda of the syllable and the vowel in the segment is somehow not limited to the nucleus, I don't know. I guess that would be the big question, and like you said, depends on the model. Commented Dec 3, 2014 at 20:35

3 Answers 3


A Vietnamese word like khiếu contains a core vowel [i] followed by two off-glides [ɜ] and [w]. The first is definitely not an on-glide. Words of this type are very common in Vietnamese and in other South-East-Asian languages.

  • Interesting, I wish I could hear how khiếu is pronounced. Question: since [w] is an approximate, can you really consider this word to be an example of a triphthong? Commented Dec 3, 2014 at 21:21
  • 1
    A word with the same triphthong: forvo.com/word/thi%E1%BA%BFu/#vi
    – fdb
    Commented Dec 3, 2014 at 21:54
  • 1
    Or this one (the lady in Saigon dialect, the man in standard Hanoi pronunciation): forvo.com/word/ng%C6%B0%C6%A1%CC%80i_me%CC%A3
    – fdb
    Commented Dec 3, 2014 at 22:09
  • 1
    +1. @tgerardot14 transcribing it with a [w] is just a notational choice. In fact, as a phonologist I prefer using approximant symbols for ends of diphthongs/triphthongs because it emphasizes the fact that we are really dealing with a single syllable and the "core vowel" of the nucleus, as fdb put it, comes earlier on in the sequence. The last part is an "offglide", so why not use a "glide" symbol for it? Commented Dec 4, 2014 at 0:51

Portuguese certainly has triphtongs:

Paraguai - /para'gwai/

luau - /lwaw/

Since the former is an import from Tupi-Guarani via Castillian, I would assume that Castillian has triphtogs too (I would not assume that of Tupi-Guarani because the /gw/ probably originates in a glotal stop, /para'ʔajʔ/).

Isn't "why" an example of a Standard English triphtong, /hwaj/?

Or what we call tritongos in Portuguese is not the same as what you call triphtongs?


From Bernese German:

  • [t͡siəu̯] 'goal'
  • [kfyəu̯] 'feeling'
  • [ʃtuə̯u̯] 'chair'
  • [myəi̯] 'effort,struggle'

Since OP asked for regular features: The ones ending in [u̯] are all results of a sound change where coda-/l/ became [u̯] or [u]. So these are simply combinations of earlier diphthongs + /l/.

[myəi̯] on the other hand is morphologically complex. The glide [i̯] is a non-syllabic realisation of the female nominaliser -i. It is usually syllabic, e.g. [hœːχ-i] 'height' or [tʏmː-i] 'stupidity'. Its occurrence after diphthongs seems fairly rare to me; then again it's not that common overall. Come to think of it, a similar behaviour can be found for the subjunctive 1 ending -ig, which can be realised as [i̯(g)], e.g. in [tːyə̯i̯(g)] 'was (SUBJ)'.

So yeah triphthongs arise due to somewhat recent diachronic processes and morphological combinations. They should in no case be considered part of the phonemic inventory, but at least phonologically, if not morphologically complex.

Also I should mention that the middle target can sometimes be elided (e.g. [myi̯]).

  • 1
    The examples are good, but your answer would be more informative if you gave some more context (e.g. is this a regularly occuring feature in Bernese German or are these words exceptions, are there variances in pronounciation, what are the relations to similar dialects, etc). Commented Oct 25, 2016 at 16:34

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.