----- Original Message -----
From: Muke Tever
To: phoNet@egroups.com
Sent: Wednesday, May 10, 2000 9:27 AM
Subject: Re: [phoNet] (unknown)

Please make sure that you read this posting in UTF-8 encoding.

Clicks are stops produced with two articulatory closures in the oral cavity. The pocket of air enclosed between the two closures is rarefied by a ‘sucking’ action of the tongue. The release of the more forward closure produces a loud and extremely salient transient. This so-called velaric airstream mechanism is always ingressive (the air is sucked in) and can only be used for stops and affricates. Clicks are inherently stop-like or affricate-like depending on their place of articulation: clicks involving an alveolar or palatal closure are acoustically like plain stops, while bilabial, dental and lateral ones sound more like affricates.

As far as I know, the fullest practically accessible overview of the phonetic properties of clicks is the relevant chapter (‘Clicks’) in Ladefoged & Maddieson’s Sounds of the World’s Languages (1996, Oxford: Blackwell), henceforth LM. I’ll follow LM’s transcription:


bilabial click


dental click


alveolar click


palatal click

|| lateral click

Clicks are used throughout the Khoisan language family of southern Africa and in the neighbouring Nguni languages (Zulu, Xhosa, etc.), which borrowed them from Khoisan. They also occur in Sandawe and Hadza, two languages (or rather language groups, believed to be distantly related branches of Khoisan) in Tanzania, and in Dahalo, a South Cushitic language spoken in Kenya. The only non-African language known to employ clicks as regular speech sounds is Damin, an ‘alternative code’ used by speakers of Lardil (Australia) – actually representing a very elaborate kind of language game. Of course, diverse ‘kiss-kiss’, ‘tut-tut’ and ‘gee-up’ noises can be used as meaningful interjections worldwide.

‘As a class clicks are probably the most salient consonants found in a human language. They are easier to identify than non-click consonants, and are virtually never confused with non-click consonants’ (LM: p.250). In other words, clicks are PERCEPTUALLY OPTIMAL consonants, in which phonetic enhancement is pushed to the extreme. The question why their geographical distribution should be as limited as it is remains one of the most intriguing puzzles of general phonetics and of the theory of phonological universals.

As noted above, clicks necessarily involve two closures: an anterior one, which is regarded as primary in the sense that it determines the click’s place of articulation, and a posterior one, which is typically velar or (less commonly) uvular. This posterior ‘accompaniment’ can be transcribed as a velar or uvular oral or nasal stop (k, , , ŋ, , , q, ɢ, etc.). It’s fairly easy to pronounce a nasalised click once you realise that while maintaining the double oral closure you can freely breathe through the nose.

‘Plain’ clicks (i.e. those that involve [k] as the posterior accompaniment) are among the easiest sounds to pronounce, but the addition of various complex accompaniments produces more challenging ‘special effects’ . For example, an aspirated dental click transcribed [k|ʰ] in Nama k|ʰo ‘play music’ contrasts with a nasalised dental click [ŋ|] in ŋ|o ‘measure’ and a voiceless aspirated nasalised dental click [ŋ̥|ʰ] in ŋ̥|ʰo ‘push into’. There are some really hair-raising combinations of elements making up a click accompaniment, as in Zhul’hōasi gk!ˣˀàrú ‘leopard’ which features a prevoiced affricated glottalised alveolar click [gk!ˣˀ]. There are more than 100 known ways of beginning a word with a click, so the size of Khoisan click-phoneme systems ranges from 20 (Nama) to as many as 83 (!Xóõ). In the latter language about 70% of words begin with a click (it’s worth noting that with the exception of Sandawe and Hadza click languages permit ONLY word-initial clicks).

All of the above is a long but necessary preamble to a brief attempt at answering your question: do clicks change? (and how do they change if they do?).

The very phonetic salience of clicks guarantees their relative stability in a language that has acquired them, although the stop-like alveolar and palatal clicks are known to be replaced with velar and palatal consonants respectively in some Khoisan languages. The fact that different though genetically related click languages have different click inventories is sufficient proof that at least some features of clicks do change historically. It seems to me that while the general character of a click (as represented by the velaric airstream mechanism) doesn’t change easily, the secondary features of the accompaniment (voice, glottalisation, aspiration, affrication, etc.) are quite prone to mutation. There is also some room for modification as regards the point of articulation of either closure: for example, an alveolar click may shift towards a palatal articulation, and a velar accompaniment may be retracted and become uvular.

I’d love to be able to show some actual examples of historical sound change involving clicks, but I’ve found none so far in my reference books. If I come across any nice reconstructions, I’ll share them with you. Or perhaps somebody on the list knows where to find them, eh?


Muke wrote:

Ok, I have a question!
I have a vague understanding of how many sounds can change in a language
over time... a stop /t/ could become a fricative /T/, or an unvoiced /t/
becoming a voiced /d/ for example...

...but I don't know how this kind of change affects the more 'exotic' sounds
such as clicks.  Do clicks mutate?  (I'm sure they must).  Do they mutate
into other clicks, or can they become regular consonants?  Do clicks often
mutate _from_ consonants?

That's all my question.  ;p