I was surprised to find in Zsiga (2020: 120, 125) a claim (by Donegan & Stampe 2009) that vowels in Hawaiian, as well as oral vowels in French, are always oral.

Unfortunately Donegan & Stampe are a phonology, not phonetics, paper and don't reveal where or how they got such data. I also found Parker Jones (2018), a more descriptive phonetics paper, who transcribes all vowels following nasal consonants as nasalized, though he is describing a native speaker who has non-native-speaking parents and acquired the language in immersion schools so this disagreement may be due to a dialectal or diachronic variation. Or have I misinterpreted Donegan & Stampe or Zsiga?

Is it true that (studies have found that) vowels in Hawaiian are never nasalized (in other words, nasal consonants do not cause adjacent vowels to be nasalized)? Are there any other languages where this is the case?

Is it true that oral vowels in French are never nasalized? Is it common for languages with phonemic nasal vowels to never nasalize phonologically oral vowels?

2 Answers 2


Cohn 1990 in her UCLA dissertation (Phonetic and Phonological Rules of Nasalization) compares English, Sundanese and French w.r.t. airflow. Nasal airflow during vowel production in these languages are (1) entirely predictable from local physical context, "partial" and quite gradient, (2) the result of an abstract rule and (3) underlying as well as rule-derived, respectively.

You can take French to be a language which definitionally has "oral" and "nasal" vowels, and from observation of such vowels infer limits on what an oral or nasal vowel must or may do, in human language. Observation of nasal airflow data demonstrates a number of points. First, transition from preceding oral to following nasal, or vice versa, always involves overlap – lag, so that nasal vowels after oral vowels or consonants are initially less nasal that they are at the midpoint. Second, there is a tendency for oral vowels to anticipate the nasality of the following nasal consonant or vowel bu some about (around the last 20% of the oral vowel), but the anticipation may be earlier in the case of /VnC/ as in "bonne deux".

However, if you integrate nasal airflow over the entire segment duration (essentially what auditory judgment does), then nasal vowels remain nasal and oral vowels remain nasal.


I cannot speak for Hawaiian. But as far as French is concerned, there is no doubt that some vowels like aN, eN, oN are fully nasalized and the non-nasalized vowels are fully oral. Otherwise this would create a confusion between say for example: aN and a. In other words, the intrinsic nasality of consonants like m or n does not spread onto neighboring vowels, contrary to what frequently happens in a number of languages, like English among others.

  • 3
    Thanks for the answer, but can you name a source? Just because the oral-nasal contrast is maintained doesn't automatically indicate the oral vowels are always completely oral, since phonetic nasality is not a binary, and there can be other means (like duration and phonation) reinforcing the contrast.
    – Nardog
    Mar 8, 2021 at 7:31

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.