LINGUIST List 16.941

Mon Mar 28 2005

Sum: Click Origins

Editor for this issue: Jessica Boynton <>

To post to LINGUIST, use our convenient web form at


        1.    Mark Jones, Click Origins

Message 1: Click Origins

Date: 28-Mar-2005
From: Mark Jones <>
Subject: Click Origins

Fund Drive 2005 is now on! Visit to donate now!
Regarding query:

Dear all,

Some time ago I posted a query asking for details on hypotheses that click
consonants in Khoisan languages may have originated from labial-velar stops
(LINGUIST List: Vol-16-481. Wed Feb 16 2005).

Both labial-velar stops and labial clicks involve velar and labial constriction.
A superficial articulatory parallel is clear, though it should be noted that
labial clicks are apparently the rarest of all clicks in terms of places of
articulation. I attributed the hypothesis that there might be a connection to
Klaus Kohler (1998: 267), who speculates positively about it, and Peter
Ladefoged and Ian Maddieson (1996: 340), who take pains to draw a clear
distinction between the phonetic realisation of labial-velars and similar
articulatory manoeuvres in labial clicks.

My query asked whether anyone had any further information about this hypothesis,
and also about the possible origins of the labial-velars themselves, including
fortition of a labial-velar approximant.

Marie-Lucie Tarpent at the Department of Modern Languages, Mount Saint
Vincent University, Halifax, Nova Scotia, Canada commented on her impressions of
strongly articulated /w/ as an idiosyncrasy in some speakers of English, an
articulation which she imagined was close to that required for a labial-velar
stop. Marie-Lucie also pointed out that when Romance languages borrowed words
such as 'war' from Germanic speaking tribes (possible *werra), the borrowing
involved a /gw/ cluster, e.g. Spanish 'guerra', French 'guerre'(now both simply
initial /g/) and Italian 'guerra' (still /gw/).

More Canadian responses, this time from Dr. Bruce Connell at York University,
Toronto, who set out the orthodox view of the origins of labial-velar stops.
These originate from increasing overlap of sequential velar and labial gestures:
kua > kwa > kpa. This developmental sequence, as well pw, bw > kp, gb, is also
mentioned by Dr Mike Cahill at SIL International in Dallas, Texas, USA. More
details can be found in:

Cahill, M. (1999). "Aspects of the phonology of labial-velar stops." Studies in
African Languages 28.2: 155-184

Connell B (1994). "The structure of labial-velar stops." Journal of Phonetics,
22: 441-476

Connell B (1998/9). "Feature geometry and the formation of labial-velars: a
reply to Mutaka and Ebobissé." Journal of West African Languages and
Linguistics, 27.1: 17-32.

Ponelis, F. (1974). "On the dynamics of velarization and labialization: Some
Bantu evidence." Studies in African Linguistics, 5.1: 27-58.

Bruce Connell mentioned evidence for this process in Igbo dialect variation, and
also occasional realisations of /w/ as the reflex of /gb/. On the hypothesis
that clicks might arise from labial-velar stops, Bruce pointed out that if this
were the case, we might expect to find that clicks were more common than they are.

Mike Cahill commented that /kp/ and /gb/ may both merge as /gb/, or become /p/
and /b/. A development to velars, i.e. /kp/ > /k/ etc., is apparently less
frequent. Mike also pointed out that voiced labial-velars may be produced with
an implosive airstream (larynx lowering). On the click-labial-velar hypothesis,
Mike noted that no languages have both, but that if labial clicks do originate
from some kind of articulatory mistiming of the labial and velar gestures, a
shift from labial click to other click types is then required.

The origin of labial-velars seems agreed as originating from increasing overlap
between labial/velar stops and a following /u/ or /w/. It should be noted that
this gestural overlap by and of itself would not result in two simultaneous
closures - some 'gestural enhancement' must also occur to result in complete
labial occlusion in /kp, gb/ from labial approximation for /w/ in /kw, gw/, and
to result in complete velar occlusion in /kp, gb/ from velar approximation for
/w/ in /pw, bw/.

The point made in the query and echoed by Mike Cahill - that any clicks
resulting from labial-velar stops would be labial, and that labial clicks are
themselves rare - remains a problem, though one which can be countered in two ways:

1) Perceptual reanalysis of click place of articulation - this does happen in
attested cases of variation with click places of articulation ("click
replacement" see Traill and Vossen 1997);

2) Triple closures with labial clicks, producing posterior and anterior lingual
closures at the same time as labial closure (noted in Ladefoged and Maddieson
1996: 352).

The rarity of labial clicks could then be due to them being the source of other
click types. But then we are still faced with the recurring problem, noted again
by Bruce Connell, of why clicks are not more widespread.

Many thanks to the respondents for their assistance. Watch this space for more

Mark J. Jones


Kohler, Klaus J. (1998). "The development of sound systems in human language."
In James R. Hurford, Michael Studdert-Kennedy, and Chris Knight (eds.)
Approaches to the Evolution of Language. Cambridge, UK: CambridgeUniversity
Press, pp. 265-278.

Ladefoged, Peter, and Ian Maddieson (1996). Sounds of the World's Languages.

Traill, Anthony, and Rainer Vossen. (1997). "Sound change in the Khoisan
languages: new data on click loss and click replacement." Journal of African
Languages and Linguistics 18: 21-56.

Linguistic Field(s): Historical Linguistics
Respond to list|Read more issues|LINGUIST home page|Top of issue