The etymological database for the West Central Khoisan language subgroup, linked directly to the general Central Khoisan database.
!NB!: It is not yet transparently clear whether West Central Khoisan actually exists as a separate, historically valid taxon, because there is currently not enough evidence to state that the splitting of Proto-Non-Khoekhoe was really binary (into West and East Central Khoisan). In terms of lexicostatistics, the distance between, e. g., the Kxoe subgroup (//Ani, Kxoe, Buga, |Ganda) and the |Gwi-//Gana subgroup of WCK is quite comparable to the one between either of these subgroups and East Central Khoisan. In essence, the term "West Central Khoisan" may just be a blanket grouping for everything Non-Khoekhoe that is NOT East Central Khoisan (the latter is definitely a compact and separate grouping, characterized, among other things, by such an important phonetic isogloss as the affricativization of the palatal click). Nevertheless, until a proper lexicostatistical assessment of the data has been conducted (this is hampered by insufficient data in the wordlists), it is reasonable to suspend final judgement and to provisionally accept "West Central Khoisan" as a single taxon.
Data sources: Unlike in the case of ECK, for WCK several detailed language dictionaries are available. The best of these are Kilian-Hatz 2003 (KH) for Kxoe and Visser (2001) for Naro; Tanaka (1978) for |Gwi and //Gana contains a lot of material, but, unfortunately, with poor transcription quality. Most of the other data are drawn from Vossen 1997 and various articles by R. Vossen; the quality is excellent, but the amount of available information on //Ani, Buga, |Ganda, and #Haba is, unfortunately, quite small.
The database consists of the following fields:
1. Proto-West Khoe: the hypothetical protoform for PWCK. (Even if WCK is really two or three separate subgroups, this does not constitute much of a problem, since most of the protoforms would be phonetically identical anyway).
2. Meaning: the general meaning(s) of the reconstructed stem.
3. //Ani: the default source is Vossen 1997 and other articles by R. Vossen.
4. Kxoe: the default source is KH; where available, forms in Oswin Köhler's transcription (Kö.) are also cited (taken either from Vossen 1997 or Köhler's original articles as marked in the references section).
5. Buga: the default source is Vossen 1997.
6. |Ganda: the default source is Vossen 1997.
7. Naro: the default source is Visser's dictionary; the most phonetically reliable transcriptions, however, belong to Vossen (V.; 1997 and var. articles). Alan Barnard's data from Barnard (1985) is also given where possible (Ba.), but some of it seems to be rather sloppy in regard to click transcription.
8-9. |Gwi, //Gana: the default source is Vossen 1997. Tanaka 1978 (Ta.) is very useful as a data source, but seriously unreliable in terms of transcription.
10. #Haba: the default source is Vossen 1997.
11. Tsaukwe: some data on an obviously WCK idiom, collected by S. Passarge ("Die Buschmänner der Kalahari", Berlin 1907) and included in Dorothea Bleek's dictionary (1956). The transcription is obviously of a very low quality.
12. Notes: additional comments and considerations.
13. References: bibliographical links.
Notes on transcription:
Clicks: | = dental click, ǂ = palatal click, ! = alveolar click, || = lateral click. For Tsaukwe, the sign $ denotes an unidentified click influx in Passarge's notes (only "guessable" through comparison with external data).
Click effluxes include zero (no special marking), voiced articulation (ɡ|, etc.), nasalisation (ɳ|, etc.), aspiration (|h, etc.), velar fricative (|x, etc.), uvular stop (|q, etc.), and the glottal stop (|ʔ, etc.). For most of the discussed idioms, R. Vossen distinguishes between two nasalized clicks: simple nasalized (marked here as ɳ|) and "voiced nasalized ("nasal-stimmhaft"), marked here as ɳ|n. The same distinction is made for Kxoe by Kilian-Hatz, who transcribes the two as ɳ| and ɳ|g, respectively. Judging by the correspondences, though, this opposition does not seem to be too consistently marked in the data in order to be reflected in the reconstructions, so for now it is ignored on the higher levels.
Non-click consonants: c, ʒ = hissing voiceless and voiced affricates (= ts and dz in Vossen's original transcription); ś = ç (Köhler) = palatal fricative (only in Kxoe); ć = tç (Köhler) = palatal affricate (only in Kxoe); q = uvular voiceless stop; ʔ = glottal stop; ɣ = velar voiced fricative (tentatively set up in a few reconstructions, not attested in the actual languages).
Vowels: ɛ, ɔ = open correlates to closed e, o. The tilde sign denotes nasalisation; á, à, ǎ, â = tonal markings.
As in all other Khoisan databases, original transliteration is mostly retained for all sources older than 1980 (also for Barnard 1985) for fear of incorrect interpretation of the data. Particular caution should be paid to Tanaka's data, where, e. g., ! is frequently substituted for ǂ, and the zero efflux is regularly marked as -t- instead of the usual -k- (or, in fact, zero).