Years ago, I met this guy from Cameroon. I asked him what language he spoke, and, besides Italian and French (and maybe some English), he said he spoke Kwa, a Bamileke language from his home country. I looked it up on Wikipedia, and I thought I'd found it as Kwa', but he said no, the apostrophe shouldn't be there, because apostrophes mark "fat" (his term, probably means "glottalized", will be discussed at post end) vowels, and the name of his language doesn't have a fat vowel in it. But "Kwa" with no apostrophe is either the Kwa of Nigeria aka Baa, or one of the Kwa languages aka New Kwa, none of which are from Cameroon.

So what language is this? Is my friend being confused by the apostrophe or is there really an apostrophe-free Kwa in Cameroon? (Also, what does that apostrophe denote?)


I tried to study the phonology of the language by asking him specific questions, and the following are excerpts from an article I submitted for creation on Wikipedia which will soon be deleted for lack of information.


The vowels of Kwa are five, the same as in Italian: /a/, /i/, /u/, /e/ and /o/. I don't yet know if the /e/ and /o/ phonemes are [e] and [o] or [ɛ] and [ɔ], I will investigate as I go on. I haven't yet investigated the various possible phonations. Anyway the sounds are similar to Italian and - most of all - French, as - as my consultant says - "Kwa is assimilated to French". Apparently it isn't written, and the way to write it is to approximate it with French sounds. However, there are what he calls "fat vowels", indicated with an apostrophe (e.g. "a'"), which are probably glottalized, as they're marked with the same mark that marks glottalized consonants, which he terms as «assorbite nella gola» (absorbed in the throat). Investigating this is part of the vowel phonation investigation I still have to conduct.
What I have to investigate is: nasalized, epiglottalized*, glottalized*, pharingealized, palatalized, velarized, breathy voiced, slack voiced*, creaky voiced*, stiff voiced*, faucalized*, harsh*, strident* and ballistic*. * means I can't pronounce them, except for: - glottalized, where Wikipedia says «Glottalization of vowels and other sonorants is most often realized as creaky voice (partial closure)», so I'll merge the two; - creaky voiced, where I can obtain what I define creaky voice only on low notes; - faucalized, where I'm not sure about how I make them, and I make them the way I would if I were yawning and speaking at the same time; - strident and ballistic, where besides being unable to make them I think I can infer from Wikipedia that they aren't present in Kwa.


As with the vowels, I still need to investigate the phonations - voiceless, voiced and voiceless aspirated apart.
According to my consultant, there is a three-way contrast among voiceless stops: plain, aspirated, and "absorbed in the throat", which are probably glottalized or ejective. Indeed he approved my ejective p'. Now if we define "glottalized" as "coarticulated with a /ʔ/, glottalized and ejective are two sides of one coin: if the stops are released simultaneously, you get what we could call a "tense" stop, if you release the oral stop before the glottal, the opening of the mouth rarefies the air, causing the glottis to shoot up and you to pronounce an ejective consonant. That's the conclusion I reached by trying to pronounce glottalized stops as defined above, and observed that with /p/ I ended up on the ejective almost always. Therefore, I guess one can choose either of the two. However, since my consultant approved the ejective /p'/, I recommend ejectives.
For now, voiced plosives are, two me, in a two-way contrast by glottalization. Again, the above definition of glottalization brings to something that, to me, sounds like implosives, though I'm not sure about my implosives. I'll try to investigate, but it will be rather difficult.
As for other consonants, they can't be glottalized. At least, none of those I found were "absorbable in the throat" according to my consultant.
With that, the chart of consonants - for now - reads thus:
enter image description here
As you can see, the chart is incomplete and some zones need further investigation.
Glottalized fricatives are absent as labial, but need investigation outside that column.
It's curous that Kwa lacks ʃ while having ʒ, and that nasals don't assimilate: Nzinkomba, a name, is not /nziŋ'komba/, but /nzin'komba/. That's how I excluded both the velar and the labiodental nasals.
/pf/ and /bv/ are apparently absent, both as affricates and as plosive-fricative sequence.
Palatals may be the result of velar palatalization, i.e. /c/ may actually be /kj/, since the two are hard to tell apart for me and impossible for my consultant.
Retroflex, Alveolo-palatal and Palato-alveolar sounds, outside fricatives and affricates, are to me indistinguishable from their alveolar equivalents, so I can't investigate them. I excluded all sounds of these kinds which didn't exist as alveolar, assuming they don't exist because the alveolars don't, and I don't think he'd distinguish the various kinds to the point of saying "no" to alveolars but "yes" to other articulation places.
The /t/ and /d/ are alveolar, a thing which I inferred from his pronunciation of Italian, with the evident alveolar stops that go outside a standard Italian accent which would have dentals. Since the alveolarity is constant, which it isn't in French, I assumed it came from Kwa. The other interesting thing is that /t/, in the /nt/ sequence, becomes dental: [ntɛ], as he noticed himself, sounds different. He said it's "a different letter" or something like that. I must investigate if this works with /d/ as well.


All I know about this is in the message my consultant sent me via facebook: «en effet c est une communauté très refermé sur elle mème , qui a tous vecu sur les montagnes et trés jalouse de sa culture et un peu trop conservatrice !!» (actually it's a community very closed in itself, which has lived everything on the mountains, and very jealous of its culture and a bit too conservative!!)


Actually, my only reference is my mothertongue consultant. That's why I submitted this for review, in order to get more precise information and clear the mystery of the name, for according to Wikipedia Kwa is Nigerian, while Kwa' is Cameroonian, and my consultant says that there's no apostrophe in his language, because the vowel is not «grassa» (fat) as the apostrophe would make it. All I have is him and what I know from Wikipedia about sounds. Please help me.

Screenshot of another section under Phonology

  • 4
    Typical of this site: Someone asks a really interesting question based on original research, and everyone wants to close it.
    – fdb
    Sep 10, 2017 at 22:55
  • 1
    This is not the kind of question termed "language identification" where usually an written artifact or a sound bit without further searchable identification is presented. I voted "leave open" on this one. Sep 11, 2017 at 10:33
  • 2
    @MickG, perhaps you can reword your question to make it clear what you are asking, if it's not "What language is this?". I will say that the facts imputed to the language don't match any language of Cameroon or Delta Nigeria.
    – user6726
    Sep 11, 2017 at 15:48
  • 1
    @MickG: Yes, you do. Sep 11, 2017 at 16:21
  • 1
    It appears that there is disagreement about the meaning of the "language identification" guidance. I don't see that that guidance is only about "what language questions" supported by written or recorded data.
    – user6726
    Sep 11, 2017 at 17:07

1 Answer 1


I will re-interpret the question so that it is clearly not off topic, as follows:

I met a gentleman who claims to speak the Bamileke language Kwa in Cameroon. He denied that it was the language Kwa', which is known to be a dialect of Bamileke. I discerned the following features of the language... Since he denies that his language is Kwa', I would like to know based on the properties of the language what language he speaks.

The consonant chart indicates a very rich set of consonants, including aspiration, glottalization, and glottalization of voiced stops, and a fairly limited vowel inventory. Bamileke languages have a relatively small consonant inventory (voiced / voiceless stops), and a rich vowel inventory (i ʉ u e o ə a ɑ, or similar arrangements). No Grassfields langauge is recorded as having glottalization. In fact, no Cameroonian language is recorded as having glotalized consonants. There are some Bantu languages spoken in Mozambique and South Africa which have glottalization, implosives, and a 5-vowel inventory, but even then, none has anything like the full inventory of segments reported in the present language. Glottalized consonants are found in Afro-Asiatic languages in the area of Ethiopia, but they do not have an aspiration contrast or any contrast among voiced consonants.

Ejectives are also found in Khoisan languages of southern Africa, and !Xu is reputed to include [p t k ph th kh t' k' b d g b' d' g'] in its consonant inventory – however, it also has a huge number of clicks. Sandawe (Central Tanzania) has among others [p t k ph th kh b d g p' t' k'], so lacking one series, but it also has clicks (not as many as !Xu). Because implosives are rare and ejectives are rare, it's not difficult to sift through the language families of Africa to determine that no language has this inventory.

If you eliminate various segments from the inventory, you widen the possibilities. If you reduce the set of consonants to [p,b], then it would well be any Grassfields language, including Kwa'. Without knowing which segments are felt to be most dubious, it's not possible to be more specific.

If you go with just the name and country, it might be the Adamawa language Kuo, the Bamileke language Kwa', or the related Ring language Oku also known as Kuɔ. If you go with the name, country and reputed area of Cameroon, it must be Kwa', in which case you have to discount the gentleman's rejection of that identification. It is very likely that his rejection is based on confusion of local orthography versus general Wiki-type language name reports. Since Bamileke languages also have complex tone systems, it is relatively likely that apostrophe is used in a practical orthography (perhaps for tone, perhaps for representing vowel contrasts). Glottal stop is a common final consonant in the area, and we may presume that the spelling "Kwa'" means that it is called [kwaʔ]. Since we don't have any further information on the Bamileke dialect bko, all one can say at this point is that if we believe the geographical and naming claims, it is most likely bko. The linguistic claims have to be totally discounted.

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