Annotated Swadesh wordlists for the Taa group (Peripheral Khoisan family).

Languages included: !Xóõ [taa-xoo], Kakia (Masarwa) [taa-kak], Nǀuǀǀen [taa-nue].
Reconstruction: Preliminary version available.


I. !Xóõ

Traill 1994 = Traill, Anthony. A !Xóõ dictionary. Köln: Rüdiger Köppe Verlag. // A large dictionary of the !Xóõ language, including detailed paradigmatic information; based on one of the most detailed and professional investigations of a Peripheral Khoesan language, and an "anchor-type" source for the study of the South Khoisan family in general.

Maingard 1958 = Maingard, L. F. Three Bushman Languages. Part II: The Third Bushman Language. In: African Studies, 17, pp. 100-115. // A brief sketch of !Xóõ phonology and grammar with some illustrative material; suffers from rather poor quality of transcription, but still preserves some historical interest.

II. Kakia, Nǀuǀǀen

Bleek 1929 = Bleek, Dorothea F. Comparative Vocabularies of Bushman Languages. Cambridge University Press. // (A collection of mid-size vocabularies from 12 "Bushman" dialects (several North, South, and Central Khoisan idioms are represented), with most of the data collected by D. Bleek herself. Not as thorough as Bleek 1956, and even less reliable in regards to data transcription, but the English-Bushman data organization principle makes it a useful source to consult in the preparation of Swadesh wordlists.)

Bleek 1956 = Bleek, Dorothea F. A Bushman Dictionary. American Oriental Society: New Haven, Connecticut. // (A huge (almost 700 pages) collection of comparative data on Khoisan that includes both Dorothea F. Bleek's own collection and data from numerous other researchers published up until the 1930s (W. Bleek, L. Lloyd, etc.). Transcription quality varies in between all the different sources, but is generally unreliable, quite typical of all Khoisan data published before the second half of the XXth century. Nevertheless, the edition still contains a wealth of priceless data, particularly on extinct North and South Khoisan languages.)


I. !Xóõ


The following modifications have been made during the transliteration of A. Traill's system for the transcription of !Xóõ into the UTS:

A. Click effluxes (illustrated with a dental click):

ǀg ɡǀ Voiced accompaniment.
gǀx ɡǀx Prevoicing + velar (phonetically uvular) fricative.
ǀkxʼ ǀxʼ Velar affricate (phonetically realised as ejective).
gǀkxʼ ɡǀxʼ Prevoicing + velar affricate.
ǀɢ ǀɢ Voiced uvular efflux; phonetically, also includes prenasalization = ɳǀɢ.
gǀqh ɡǀqʰ Prevoicing + uvular aspirated stop.
ɢǀqh ɡǀqʰ Prevocing + prenasalization + uvular aspirated stop.
ǀn̥ ǀn̥ Voiceless nasalization accompaniment.
ǀn ɳǀ Nasalized accompaniment.
ǀʼ ǀʼ Glottal stop accompaniment.

B. Simple consonants and clusters:

dtx dx "Prevoiced" cluster.
dth "Prevoiced" aspirated dental stop.
tʼkxʼ txʼ Since the affricate kx is always ejected, ejective articulation is not marked.
dtʼkxʼ dgxʼ "Prevoiced" cluster.
ts c
dz ʒ
dtsh ʒʰ "Prevoiced" aspirated affricate.
tshx cx
dtshx ʒx
tsʼkxʼ cxʼ
dtsʼkxʼ ʒgxʼ
gkh "Prevoiced" aspirated velar stop.
kxʼ This velar affricate is always ejective in nature.
gkxʼ gxʼ "Prevoiced" velar affricate.

C. Vowels:

Nasalized vowels.
Vh Breathy vowels.
Pharyngealized vowels.
VV Phonetically (but not phonologically) long vowels.
V̰Ṽ Ṽ̰ː Double vowels marked with different diacritics are generally combined into one vowel symbol, marked with the length diacritic (although length has no phonological value in !Xóõ).

The wordlist incorporates most of the paradigmatic information given in [Traill 1994]. This means: For nouns - (a) the plural form, more often than not unpredictable based on the singular; (b) information on concordial classes (1/4, 2, or 3) and tonal classes (I or II); for verbs this includes giving the "variable form", i. e. the stem variant that is used in conjunction with class markers. For detailed explanations of these traits of !Xóõ morphology, the introduction to [Traill 1994] may be consulted.

II. Kakia, Nǀuǀǀen

1. General.

Kakia is a presumably extinct variety of Taa, very close to !Xóõ, but reflecting at least a small handful of lexicostatistical discrepancies that indicate a significant period of individual development (it is hard to trust the results of lexicostatistics, however, since the unreliable character of D. Bleek's semantic glossing may result in several "false" discrepancies). In [Bleek 1929] and [Bleek 1956], this idiom is defined with the Bantu term Masarwa, no longer in general use among Khoisanologists.

Nǀuǀǀen is another Taa variety, described by D. Bleek in 1920 at Tsachas, Uichenas and Aminuis (Namibia). Another name for this dialect is Nǀusan. Approximately the same minor number of lexical discrepancies between Nǀuǀǀen and !Xóõ as between Kakia and !Xóõ are detectable, but it is not clear how many of these reflect genuine lexical divergence and how many reflect poor data quality.

2. Transcription.

All of the UTS transliteration details for Kakia and Nǀuǀǀen follow the standard conventions that we have chosen for D. Bleek's Khoisan data. The main ones are as follows:

ǀ, ǂ, !, ǀǀ, ʘ ǀʼ, ǂʼ, !ʼ, ǀǀʼ, ʘʼ "Lone" click signs normally denote a click with a glottal stop accompaniment.
ǀk, ǂk, !k, ǀǀk, ʘp ǀ, ǂ, !, ǀǀ, ʘ Clicks with a following k (or the variant p after the labial click) are analyzed as clicks with "zero" accompaniment (k and p are deleted during transliteration).
ǀg, ǂg, !g, ǀǀg, ʘb ɡǀ, ɡǂ, ɡ!, ɡǀǀ, ɡʘ Voiced accompaniment.
ǀn, ǂn, !n, ǀǀn, ʘm ɳǀ, ɳǂ, ɳ!, ɳǀǀ, ɳʘ Nasalized accompaniment.
ǀh, ǂh, !h, ǀǀh, ʘh ǀʼʰ, ǂʼʰ, !ʼʰ, ǀǀʼʰ, ʘʼʰ Denoted in the transliteration as glottal stop + aspiration accompaniment (although in reality these clicks may have been phonetically realized without the glottal stop).
ǀkh, ǂkh, !kh, ǀǀkh, ʘph ǀʰ, ǂʰ, !ʰ, ǀǀʰ, ʘʰ Denoted in the transliteration as aspirated accompaniment (in reality, at least some of these clicks may have been phonetically realized with an additional uvular efflux, e. g. = ǀqʰ, etc.).
ǀx, ǂx, !x, ǀǀx, ʘx ǀx, ǂx, !x, ǀǀx, ʘx Velar fricative accompaniment.
ǀk", ǂk", !k", ǀǀk", ʘk" ǀxʼ, ǂxʼ, !xʼ, ǀǀxʼ, ʘxʼ Ejective velar affricate accompaniment. In [Bleek 1929], this type of clicks is frequently transcribed as ǀkʼ, etc.; we preserve this detail in the transliteration, although it is almost certainly not significant.
ts, dz c, ʒ Alveolar affricates.
tš, dʒ, š, ʒ č, ǯ, š, ž Postalveolar affricates and fricatives.
Ejective dental stop.
ɛ, ɔ ɛ, ɔ These vowels are most probably just positional variants of the regular e, o; however, we preserve this phonetic distinction in our transliteration.
y ü Usually a positional variant of u, preserved in our transliteration.
ˉV High tonal register.
̱V Low tonal register.

It must be noted that all of D. F. Bleek's data have to be approached with extreme caution, both in regard to the preciseness of the phonetic notation and correctness of the semantic glossing. The earlier source, [Bleek 1929], is particularly unreliable, and its data can only be trusted to a certain extent if they are confirmed in the later, revised and much expanded comparative dictionary [Bleek 1956]. However, since both of them are the only source of data on certain varieties of Taa, there is no justification to completely exclude them from comparison.

NB: In [Bleek 1956], Kakia (Masarwa) entries are regularly marked as SV, and Nǀuǀǀen entries as SVI. However, there are a few cases when, by way of typographic error, SIV (normally marking the !Kwi language ǀʼAuni) is substituted for SVI. Fortunately, most of these mistakes are detectable through comparison with [Bleek 1929] (where these typos are absent), analysis of textual examples accompanying the entries, or external comparison.

III. Proto-Taa

A proper Proto-Taa reconstruction is virtually impossible, given the huge discrepancy in data quantity and quality between Lone Tree !Xóõ, represented by Traill's massive dictionary, and all other varieties of Taa, represented mostly by phonetically and semantically inaccurate data from D. Bleek's collections, as well as a handful of papers by L. F. Maingard and E. Westphal. As of now, "Proto-Taa" reconstructions are essentially !Xóõ forms, projected onto the proto-level with a little extra morphophonological analysis. In a very small handful of cases ('ashes', 'breast', 'full', 'snake'), there is some evidence (internal or external) that the !Xóõ equivalent of the Swadesh meaning may be a recent lexical replacement, but for the most part, expert judgement is inevitably biased towards !Xóõ simply because it is so much better described than everything else.

The issue of regular phonetic correspondences between !Xóõ and earlier described dialects is quite complex, mainly because it is inextricably linked to the issue of transcriptional accuracy; most of the differences between these varieties of Taa are presumably due to errors in notation than actual phonetic change. For this reason, we do not provide here a list of phonetic correspondences, but comment on each situation individually in the "Reconstruction shape" section of the comments.

Database compiled and annotated by: G. Starostin (March 2011; preliminary reconstruction added, April 2015).
Reconstruction sources: G. Starostin, personal notes (partially published in: Starostin, George. From Modern Khoisan Languages to Proto-Khoisan: the Value of Intermediate Reconstructions. In: Aspects of Comparative Linguistics III [Aspekty komparativistiki III]. Orientalia et Classica, vol.. XIX. Мoscow, RSUH Publishers, pp. 337-470).