Can the modal verb "can" be pronounced as /ŋ(k)/ ?

It may be the context of a following /k/, as in "we can come and see", but I have also notice it being reversed /ŋ(k)/.

Therefore, I'd like to know whether this is true, in phonetic terms, and how widespread it is. Some academic reference to this process would be clarifying.

For example, 5:12 'and you can come over' https://youtu.be/sriYmFBvOsE?t=312

  • 2
    What is your question? What do you mean by "revesed /ŋ/?" Please make sure you are asking a question. I hear what you are referring to in the video, but you haven't actually asked a question to be answered. Dec 20, 2016 at 0:53
  • I think this is an interesting question, and much clearer after the edit. Wouldn't the /k/ you're hearing after the nasal just be the start of the following word "come", though? Rather than a reversal, I think it's more likely that the initial /k/ is simply elided...
    – herisson
    Dec 20, 2016 at 1:11
  • 1
    What I hear (and do in my own casual speech in that phrase) is that the initial /k/ of "can" becomes a glottal stop, and the remainder of the word becomes a syllabic nasal that assimilates to the place of articulation of the following /k/, thus becoming [ŋ]. As @sumelic suggeted, the [k] you hear after the [ŋ] is simply the beginning of the next word ("come). I spent some of my formative years in Texas (where the video seems to be taking place), and the phrase in the video sounds natural to me. Dec 20, 2016 at 1:16
  • 1
    @GJC, No, I'm not set up here to make audio files. The details are: (1) syllable offset voiceless stops are glottalized before stops -- meaning that they are pronounced with closure of the glottis (they are not ejective). (2) alveolar stops (t/d/n) assimilate in position regressively to a stop. (3) In a homorganic stop cluster, the oral articulation can be delayed, leaving behind just a glottal stop, in the case of consonants that were glottalized in step (1), For instance in "pop bottle", /pb/ >(1) [p'b] >(3) [ʔb], and in "hot mama", /tm/ >(1) [t'm] >(2) [p'm] >(3) [ʔm].
    – Greg Lee
    Dec 20, 2016 at 14:38
  • 1
    To clarify: you've used the notation for phonemic transcription, /.../; but it sounds like you're actually interested in a phonetic realization of <can> as [ŋ(k)]?
    – ruakh
    Feb 25, 2017 at 6:05

1 Answer 1


Yes, it exists but more prominently when the modal, preposition or any word is unstressed and takes a schwa sound (ə). It has the ability to transform to [ŋ] as a bridge to the K in the next word. This is because there is resistance to articulate the consonant n if your following word is more strongly stressed (a nasal is a closer bridge to a glottal stop and the schwa sound invites it for linking purposes). . If you stress the modal 'can' [kæn] the [ŋ] disappears. I notice the same effect with unstressed prepositions eg "in Canada." and relative pronouns eg "when crying...".

  • This is because the effects of regressive assimilation (the upcoming [k] changing the previous nasal to [ŋ]) are more notable in the reduced unstressed syllables of connected speech, and less noticeable when carefully separated out through emphasis.
    – tchrist
    Feb 28, 2017 at 15:29
  • It appears also before a /t/: We can tell youtube.com/watch?v=9y_uAv_kJUs
    – GJC
    Feb 28, 2020 at 11:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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