Labio-velarization is a feature of accents of Kabyle in some area(s). For example, the word aseggas could be both pronounced [asəɡɡas] or [asəɡɡʷas].

I think there is a difference between hearing [ʷ] and a consonant followed by [w]. I mean, [asəɡɡʷas] does not seem to sound the same as [asəɡɡwas].

Since I'm not familiar with this accent, it's hard for me to tell the difference. Could this be explained, or better, shown with audio samples?

(This question isn't specific to Kabyle language, it's focused on labialization/labio-velarization in comparison with [w].)

  • 1
    May be tangential, but in English /t/ is normally not labialized before /w/, while in Korean it is - the difference is noticeable enough that English words like twin are borrowed into Korean with an extra vowel between t & w (i.e., /tɨwin/) to keep /t/ from becoming /tʷ/.
    – jick
    Commented Jan 22, 2020 at 3:39
  • @jick That makes little sense. [tʷin] is much closer to English twin than [tɨwin]. I bet the vowel is inserted only because otherwise there would be no way to differentiate it from tin.
    – Nardog
    Commented Jan 23, 2020 at 4:56
  • @Nardog Korean does distinguish /tin/ and /twin/, but apparently we feel that /tɨwin/ is the closest approximation of English twin.
    – jick
    Commented Jan 23, 2020 at 5:20

2 Answers 2


There are many key differences between [ʷ] and [w]. The most important is that [ʷ] is a secondary articulation on another sound, meaning it is a simultaneous modification, not a separate following sound. For the difference between [k] and [kʷ] is that in one, the lips are flat, and in one, they are rounded. The difference between [k] and [kw] is that in both, the lips are flat for the [k], but then a second sound with rounded lips is produced.

Another difference, even in the languages where they are pronounced more similarly, is that they are treated differently phonologically. /kʷ/ is treated as one consonant and /kw/ as two, and some languages have rules about the size or placement of clusters.

  • 1
    Thanks, it confirms what I thought. Since [u] also rounds the lips, will the difference between [kʷu] and [ku] be perceptible? I can ask a new question about it if it's more appropriate.
    – Amessihel
    Commented Jan 21, 2020 at 15:06
  • 4
    <k> doesn't necessarily mean it's articulated with unrounded lips, it just means it is unmarked for—the transcriber did not feel it important to mark—labialization.
    – Nardog
    Commented Jan 21, 2020 at 16:32
  • 3
    IPA Principles, ¶ 5: "The use of symbols in representing the sounds of a particular language is usually guided by the principles of phonological contrast ... The three k-sounds of the English words keep, cart, cool can be heard and felt to be different, but from the linguistic or phonological point of view the differences are not distinctive and all may be represented by the same [k] symbol. The same applies to the French k-sounds in qui, cas, cou, though these differ phonetically from the corresponding English ones."
    – Nardog
    Commented Jan 21, 2020 at 19:37
  • @Nardog so as far as I can tell, cart is pretty much your standard k, whereas cool has rounding, and in keep, the contact with the roof of the mouth is further forward. Is that right, or is there more to it (especially in terms of secondary articulation)?
    – JD2000
    Commented Jan 23, 2020 at 14:47

If you are looking for a phonetic basis for thinking that you have [gw] versus [gʷ], you can listen for an effect on the preceding segment, where the end of the previous vowel is more likely to show signs of partial rounding before [gʷ], compared to if you have a sequence [gw]. I said "likely", so you can't use this as an absolute distinguishing feature.

Generally, you have to rely on phonological tests. One is sonority-sequencing in the syllable – the tendency for obstruents and sonorants to be ordered so that the sonorant is nearest to the syllable peak (vowel). A word like [bəkw], qua consonant-glide sequence, is less likely than a single consonant at the end of the syllable i.e. [bəkʷ]. Again, this is a tendency, and one would be hard-pressed to insist that syllable-final consonant plus glide sequences are universally impossible.

Another argument could be pattern of reduplication. If a language has consonant plus glide sequences = rounded consonants, and also consonant plus liquid, plus also has a pattern of CV reduplication, then the application of reduplication to supposed Cw could be a good diagnostic. Thus if /plam/ → [pa-plam] and /kwat/ → [kakwat], you have evidence for a two-segment analysis as velar plus glide, but [kʷakʷat] would be evidence for a single round segment. Again, this is a (strong) "tend" argument (there is an alternative analysis invoking contrastive structure to the onset, if you believe in such things).

In Berber and Ethiopian Semitic languages (and maybe Moroccan Arabic, details are not clear to me), the "kw" sequences are more clearly secondary articulations on consonants, based on phonological pattern.

  • Thank you. For "kw" sequences in Berber languages, it depends. For example, one of the plural forms of tikkelt is tikwal. Here, this is really a [kw] sequence, and not [kʷ]. (Source)
    – Amessihel
    Commented Jan 21, 2020 at 19:25
  • I wrote "really", but at second glance, maybe not. Nothing says the lips must not be rounded before the w of tikkal.
    – Amessihel
    Commented Apr 14, 2021 at 23:33

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.