So this video explains clearly how to do the 3 Xhosa clicks at the same time as each vowel sound. The Wikipedia page also shows clearly how to produce those 3 clicks as well, independent of any vowel sounds. So just c [ǀ], x [ǁ], and q [ǃ].

However, then I listen to the alphabet here in this video, and they get to "c" and it sounds nothing like the first two links above as they sound out the [ǀ]. I notice that the Wikipedia for Xhosa lists 18 clicks (instead of just the 3 demonstrated in the first two links):

[kǀʼ] c 
[kǀʰ] ch    
[ɡ̊ǀʱ] gc     
[ŋǀ]  nc 
[ŋǀʱ] ngc   
[ŋǀˀ] nkc 

[kǁʼ] x  
[kǁʰ] xh 
[ɡ̊ǁʱ] gx
[ŋǁ]  nx  
[ŋǁʱ] ngx 
[ŋǁˀ] nkx 

[kǃʼ] q   
[kǃʰ] qh 
[ɡ̊ǃʱ] gq
[ŋǃ]  nq   
[ŋǃʱ] ngq 
[ŋǃˀ] nkq   

Connecting [kǀʼ] c to the alphabet video, it seems like that sound is a [g] and the click [ǀ] at the same time or something, but I can't quite tell. Maybe they are doing a [ɡ̊ǀ] sound. For the first part of the question, wondering if one could identify what sound that is, and if I am correct in my understanding that they are to be pronounced at the same time (not sure how this is possible atm so probably not doing it right yet).

For the second part, I'm wondering in general how to make the rest of the click sounds. Basically there are 3 prefixes in the IPA version, [k], [ɡ̊], and [ŋ]. But I'm not sure if they should actually occur before or during the subsequently annotated click sound, e.g. [ŋǀ]. Wondering if one could clarify if it goes before or during the click. So then I just want to know if it works like this:

[kǀʼ] c     = k + click sound at same time
[kǀʰ] ch    = k + click sound at the same time, aspirated at the end
[ɡ̊ǀʱ] gc    = silent g + click sound at same time, aspirated at the end
[ŋǀ]  nc    = humming ŋ + click sound at same time
[ŋǀʱ] ngc   = humming ŋ + click sound at same time, aspirated at the end
[ŋǀˀ] nkc   = humming ŋ + click sound at same time, followed by glottal stop before next sound

The tricky part about these 18 clicks is that the k/g/ŋ sounds are all produced by closing the back of the mouth, while the clicks occur by closing the front of the mouth, so it doesn't seem possible at first, it's like "I can only close my mouth in one place not two" sort of thing. I can do the clicks with the prolonged vowel sounds no problem, just not what the IPA is showing with the other 18 versions.

  • 1
    Coarticulated consonants are not altogether rare in sub-saharan Africa, and the most common one cross-linguistically is the co-articulated labial-velar stop /k͡p/, also found in Vietnamese.
    – Michaelyus
    Commented Oct 18, 2018 at 15:13

1 Answer 1


The first video is the best exemplar of the three, The wiki page isn't even Xhosa, it's hypothetical articulations by an unknown person who doesn't speak those languages, and lacks the professional phonetician cred of John Esling and Peter Ladefoged. It primarily produces real words of Xhosa, which is a good model for how the clicks sound in that language. He does produce a few samples of "just the click", but does add something unnatural (from the perspective of Xhosa) in his isolated sample of q, which is that he nasalizes the release vowel. The alphabet video would have been better if the sound has be produced with a following vowel [a].

The problem with [ŋǀˀ] and so on is that there are not symbols for each separate click, analogous to how we notate p,t,b,d,m,n. The convention has been adopted that the manner of articulation (apart from aspiration) is notated with the initial letter. You have to consult a real physiological study of a particular language to get the dynamics of click articulation – the articulation of clicks in Khwe languages is different from that of the Nguni languages. The pre-letters k,g,ŋ carry the information that the click is voiced or nasal, but these are not the independent phonemes [k], [g], [ŋ]. The choice to use these letters is based on the fact that clicks have a non-distinctive back (velar) closure as well as a front (lingual or labial) closure, and from an aerodynamic-acoustic perspective, only the back closure has acoustic consequences until the front closure is released. This study (an MA thesis) has ample scholarly citations, lays out the issues, and provides some new acoustic data (duration measurements). Physiological studies are, unfortunately, not easy to carry out.

Usually, the most effective way to control production of these clicks is to ignore IPA spelling, and listen to and mimic a native speaker. There are some commercial courses with audio CD produced for Zulu and Xhosa – I don't have any of those materials so I can't make a recommendation.

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.