It's clear that in the production of many vowels air isn't free to go to every available space in the oral cavity once it leaves the pharynx. Our tongue even touches the molars in [i]. That's just an irrefutable proof, yet again and again I read the definition I presented at the title.

Without parting from the subject I can even mention the "central oral resonant" name use. Some authors use it when referring to vocoids but it's not like we could apply the laterality paramater in all of them, the same for orality (they could be nasalized).

3 Answers 3


I don't know to what extent the term "obstruction" is elevated to technical status in phonetics, but it is traditional for phoneticians to speak of a consonant obstruction and a vowel constriction. The distinction, as I understand it, pertains to how radical the narrowing is, and the fact that the oral tract is still aerodynamically open with vowels. In the case of a consonant, the narrowing is sufficient that it hinders or stops airflow.

  • 1
    Oh thank you, I suspected they perceived the term "obstruction" different but wanted a confirmation. Sep 12, 2018 at 17:06

I don't think any linguists would define vowels as strictly being produced with absolutely no obstruction whatsoever. If any do, it's an exaggeration. There aren't any such clear cut distinctions - there's a gradual opening of the mouth from stops through fricatives, approximants, and vowels. This is what is called the Sonority hierarchy. Wikipedia gives a simple table there showing how various terms relate to each other. One thing it points out is that vowels are syllabic: they form the centre of syllables. But this isn't a hard distinction. While most consonants are non syllabic, in some languages some consonants are analysed as being syllabic, including in English.

The mouth does have to actually produce different sounds, so anything other than the most open vowel ([a]) has to be less open because of the physical limitations. What distinguishes the vowels from the approximants is largely a matter of convention. We consider both how sonorant and turbulent the sound is, as well as the phonotactics of each language. The cross-language generalisations we produce are only that, generalisations, not hard rules.

One last thought: some languages probably have situations in which allophones cross the divide between vowels and consonants (though I can't remember of any examples.) Analysis of allophones and which is considered the 'default' is tricky at the best of times, so it's just another thing which you can expect to see in individual language's phonologies. But these edge cases don't mean that it's not useful to still have the basic categories of vowel and consonant.

  • What did you mean with allophones cross the divide between vowels and consonants? Also, could you clarify that "central resonant" point? Sep 13, 2018 at 3:03
  • Here's an example of a language where a vowel is analysed as being a glide consonant in certain contexts. I don't know what would be meant by "central resonant", it's not terminology I'm familiar with.
    – curiousdannii
    Sep 13, 2018 at 3:11
  • In the page it's written "some vowels become glides in diphtongs". A glide in this situation is still a vocalic sound, the difference is that it isn't syllabic. Sep 13, 2018 at 14:43

It's described like that because there is no significant difference in intra-oral pressure with vowels, from with normal breathing. However you shape the tube of the oral tract, or narrow, enlarge or squash it, there is no stricture narrow enough to create audible turbulence (at least whilst there is vocal fold vibration). There is also no characteristic obstruction diverting the air from leaving centrally through the mouth. Whether any air leaves through anywhere else, i.e. the nasal cavity, is not relevant.

Your Answer

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

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