This feature allows to generate a graphic representation of the supposed genetic relationships between the language set included in the database, in the form of a genealogical tree (it is also implemented in the StarLing software). The tree picture also includes separation dates for various languages, calculated through standardized glottochronological techniques; additionally, a lexicostatistical matrix of cognate percentages can be produced if asked for.
The tree can be generated by a variety of methods, and you can modify some of the parameters to test various strategies of language classification. The pictures can be saved in different graphic formats and used for presentation or any other purposes.
This option displays the full description for the selected database, including: (a) the complete list of primary and secondary bibliographical sources for the included languages, including brief descriptions of all titles; (b) general notes on said languages, e. g. sociolinguistic information, degree of reliability of sources, notes on grammatical and lexical peculiarities of the languages that may be relevant for the compilation of the lists, etc.; (c) details on the transcription system that was used in the original data sources and its differences from the UTS (Unified Transcription System) transliteration.
This option, when checked, uses a set of different color markers to highlight groups of phonetically similar words in different languages with the same Swadesh meaning.
Phonetic similarity between two different forms is defined in the GLD as a situation in which the aligned consonants of the compared forms (usually the first two) are deemed «similar» to each other. In order for two consonants to be «similar», they have to belong to the same «consonantal class», i.e. a group of sounds that share the same place and a similar manner of articulation. The current grouping of sounds into sound classes can be found here.
Accordingly, the aligned forms undergo a process of «vowel extraction» (all vowels are formally assumed to belong to «class H», together with «weak» laryngeal phonemes), and the individual consonants are then converted to classes, e. g. dog → TK, drink → TRNK (in comparisons, only the first two consonants will be used, so, actually TR), eat → HT (word-initial vowel is equated with lack of consonant or «weak» consonant), fly → PR (l and r belong to the same class) and so on.
If both of the first two consonants of the compared forms are found to correlate, i.e. belong to the same class, the words are deemed similar (e. g. English fly and German fliegen both have the consonantal skeleton PR). If at least one differs, the words are not deemed similar (e. g. English tooth → TT and Old Norse tɔnn → TN, although they are etymological cognates, will not pass the similarity tense because of the second position).
In most cases, checking this option will highlight phonetically similar forms that are also etymological cognates and share the same numeric cognation indexes. Occasionally, however, the checking will also yield «false positives» (accidentally phonetically similar forms that do not share a common origin) and «false negatives» (phonetically dissimilar forms, not highlighted, but actually cognate). It should be noted that one should never expect this method to yield a 100% accurate picture of etymological cognacy. Rather, the method is useful for the following goals: (a) assess the amount of phonetic change that took place between related languages; (b) give a general idea of the degree of closeness of relationship for those languages where phonetic correspondences have not yet been properly established; (c) assess the average number of «chance similarities» that may arise between different languages.
The last task is particularly instructive if the «Highlight...» option is used between two different languages from different databases, i.e. not related to each other or distantly related: in most cases, it will yield around 2-3 accidental color highlights, but occasionally, the count may go as high up as 5 or 6.
This option unfolds all of the notes that accompany the individual forms in the database. Sometimes these notes only consist of a basic reference to the bibliographical source, but at other times, they can be quite expansive, which makes browsing through the wordlist quite cumbersome. By default, the notes stay hidden (each note can also be opened separately by clicking on the sign next to the word).
Lanham & Hallowes 1956: 102, 110. Although the existence of a separate lexeme ɕʼi is mentioned, the quoted examples only include such pronominal forms as i-ɕʼi "we all", u-ɕʼi ~ u-ɕʼĩ "you all" (i = "we", u = "you (pl.)"). The entry is, therefore, somewhat dubious. Not attested in [Ziervogel 1955] or any of D. Bleek's publications.
Bleek 1937: 201. Suspiciously glossed as 'they, all' in [Bleek 1956: 13], but with at least one fitting textual example: tuku bà su !ʼʰɔbati "men shall all return". The only possible competition is represented by ǀǀani, one of the meanings of which is also glossed as 'all': cf. ku totos ǀǀani "all the people" [Bleek 1956: 557]. However, both the external evidence and one of the other contexts (ǀǀani e ǂeːi ki ɡǀǀò "much it is raining in the night") show that the main semantics of this quasi-synonym is a designation of simply large quantities ('much, many') rather than exhaustive quantities. We tentatively fill the spot in with the most uncontroversial variant, even if it has no obvious cognates in the rest of South Khoisan and, with its initial labial, suspiciously looks like a word of non-native origin.
Distribution: Preserved in ǀXam and the Nǀuu cluster. The same morpheme was also in obvious use in the same meaning in Seroa, as the enclitic -ku 'all' [Wuras 1920: 84]. Replacements: There are no reliable etymologies for Nǀuu huni-ki, ǀǀXegwi ɕʼi or ǀʼAuni bà. Precedence of *ku is hypothetically suggested only on the basis of the morpheme's relatively wide distribution; external considerations (a similar morpheme means 'all' in the Taa branch of South Khoisan); and the overall dubious nature of bà in ǀʼAuni. Phonetic shape: The root is known as ku (in ǀXam and Seroa) and ku-a (in a somewhat dubious attestation in ǀǀNg!ke); the latter could simply represent a suffixal extension.
Bleek 1956: 449. Quoted as !uːi in [Bleek 1929: 17]. Emphatic form: !ùi-tǝn ~ !u=!úi-tǝn (with partial reduplication). Transcribed as !ùːi by W. Bleek, but semantically glossed as 'meal, flour' [Bleek 1956: 449]. On the synchronic level the word is undistinguishable from the verb !uːi 'to burn (intr.); to smart, pain' [ibid.] and may be deemed a result of conversion; however, in the light of (a) typological unusualness of the situation (the meaning 'ashes' is more likely to develop from the transitive verb), (b) the additional meaning 'meal, flour' in W. Bleek's notation, (c) external parallels, all of which only confirm the nominal meaning ashes, it is quite probable that we are dealing here with graphic confusion of two phonetically similar, but etymologically different roots.
Distribution: Found only in the ǀXam-Nǀuu cluster. The rest of the data are either non-existent or not very reliable. Replacements: The only other attested !Wi form for 'ashes' is ǀʼAuni !ʼʰana, whose closest relative, if it is a relative, may be the form ǀǀqʼâɲa 'dirt, rubbish' in !Xóõ; this would imply a semantic development {'dirt' > 'ashes'}. However, nothing about this form or its external connections is really reliable. Phonetic shape: The modern Nǀuu form with the uvular click efflux is automatically projected onto the proto-level, since none of the old sources recognize the presence of uvular clicks in !Wi languages.
Doke 1936: 77. Not attested in Maingard's data. The word has no parallels in the rest of !Kwi and is most likely a relatively recent borrowing from Central Khoisan (the root *ǀǀxʼũ 'bark' is safely reconstructible there).
Not reconstructible due to lack of proper attestation. The only dialect cluster where the word 'bark' was consistently recorded is Nǀuu-ǂKhomani, and it seems to be just a recent borrowing from Central Khoisan. Regardless of whether ǀʼAuni ǀǀõː is also a (poorly transcribed?) borrowing or a native word, it is not sufficient to come up with a proper reconstruction.
Bleek 1956: 416. Quoted as !auːtuː in [Bleek 1929: 21]. Attested only in W. Bleek's notes (absent from L. Lloyd's materials); meaning glossed as 'belly, stomach'. The word is morphologically complex; the suffix -tu frequently appears in body part terms as well as other nouns.
Bleek 1956: 602. Highly dubious, since the meaning is glossed as 'stomach, inside'. However, the only confirming textual example is ŋ ǀǀxʼãː cí "my stomach aches", with a non-diagnostic context, and modern data from the closely related Nǀuu dialects does show the polysemy 'belly / stomach' for this word. No other candidates are attested for ǀǀNg!ke.
Sands et al. 2006. Dubious (acc. to B. Sands, the meaning should rather be 'stomach', whereas 'belly' is rather kunĩ, but this information probably needs additional confirmation; cf. a similarly questionable situation for Bleek's ǀǀNg!ke).
Not attested; cf., however, ǀʰugaː, pl. ǀʰuga-le 'stomach' [Ziervogel 1955: 43]. This is quite possibly the same word as ǀubwa 'stomach', quoted in [Bleek 1956: 323]; the discrepancy in the intervocalic consonant is disturbing, but in either case, the structure of the word is not typical for ǀǀXegwi or the South Khoisan family as a whole, indicating that this may be a loan from some unidentified source.
Not reconstructible: each dialect cluster, in which the word 'belly' is attested, displays a different root. It is interesting that at least several languages seem to lexically distinguish between 'belly' and 'stomach', e. g. ǀXam has !áu-tu 'for belly' and ǀoˤa [Bleek 1956: 317] for 'stomach' (at least, such a distinction is a valid hypothesis based on inspection of textual examples). The word for 'stomach' is actually better reconstructible for Proto-!Wi than 'belly' (since ǀXam ǀoˤa is the etymological equivalent of ǀǀXegwi ǀʰugaː).
Bleek 1956: 450. Quoted as !uiː-ya, pl. !uiː!uiː-ta in [Bleek 1929: 22]. Transcribed by W. Bleek as !úiːya and glossed as 'big, stout' in [Bleek 1956: 450]. The forms are morphologically complex and easily derivable from the verbal stem !ui 'to grow' [Bleek 1956: 449]. Secondary synonym: !érri ~ !érri-tǝn, pl. !ett=!étten [Bleek 1956: 422] 'old / big / grown-up / great'. Attested contexts are insufficient to determine the exact semantic difference between the two words, but only the first one is given as the translation equivalent of English 'big' in [Bleek 1929], and most contexts for !érri seem to confirm the semantics of 'grown-up / old' better than simply 'big' (in size).
Bleek 1929: 22. For some reason, the word is not attested at all in [Bleek 1956], making the corresponding entry in the early source [Bleek 1929] somewhat dubious.
Ziervogel 1955: 41, 55. Meaning glossed as 'large' or 'big'. Quoted as ǀǀxʼeː ~ ǀǀxʼe-xo 'be big' in [Lanham & Hallowes 1956: 105]. In [Bleek 1929: 22] the meaning 'big' is rendered as ǀaː and "confirmed" by the textual example in [Bleek 1956: 267]: ha kwe la ǀaː "that man is big". However, this is most likely an incorrect glossing: data from other sources clearly show that ǀaː really means 'long / tall' q.v.
Bleek 1937: 208; Bleek 1956: 249. This is the most statistically frequent word for which the meaning 'big' may be suggested unambiguously, cf.: ʘoša te úši "the jackal is big", ǂʼe ki e, se ǀǀʼʌn ús "that person's hut is big" [Bleek 1956: 249]. Two other words are also glossed as meaning 'big': (a) ǀʼã́si 'big, long, tall' [Bleek 1937: 208], which, as in the case with ǀǀXegwi, refers only to 'big' as 'tall, elongated' (cf. !e tari ǀʼãsi "that person is big" = "that person is tall"), see 'long'; (b) kéi 'big' [Bleek 1937: 203], which is phonetically similar to kái 'to grow, to swell' [ibid.] and could mean 'grown (up)', cf. the example: ǀʼa kéi, a ki úki ǀǀʼʌn ǂé "the girl is big, sits on the hut floor" [Bleek 1956: 86]. It also looks suspiciously similar to Proto-Central Khoisan *káí 'big' and could be a borrowing from that family.
Distribution: Only attested in the basic meaning 'big' in Nǀuu, but preserved in ǀXam as !xoː 'upright, tall' [Bleek 1956: 500]. Possibly preserved in ǀHaasi as well, but could also be interpreted as a re-borrowing from ǂKhomani into that language. Replacements: (a) ǀXam !ui-ya, morphologically derived from !ui 'to grow' {'grow' > 'big'}; (b) ǀǀXegwi ǀǀxeya, of unknown origin; (c) Lower Nǂossob *u-si ~ *o-si, also of unknown origin. It is theoretically possible that this is the main !Wi root for 'big', but lack of parallels in the much better described ǀXam and Nǀuu make this dubious. Phonetic shape: Correspondences between Nǀuu and ǀXam are fairly straightforward.
Story 1999: 21. Semantic difference between the two words is unclear. Judging by textual examples, both may be used in free variation, cf.: ǀǀʰasá kʼa ɔ̂ː-si "the child is big", matabab kʼa !xwaː "Matabab is big" [Story 1999: 24, 25]. It is not excluded that !xwaː is a Nǀuu word used alongside the authentic ǀHaasi equivalent, but there is no way of certifying that.
Proto-!Wi:*u-si ~ *o-si #
The Lower Nǂossob equivalent, also potentially of Proto-!Wi origin; see notes on *!xo.
Bleek 1956: 119. Plural form: xʼexʼenn ~ xʼaxʼann. Transcribed as xʼarri ~ xʼar̃r̃i, pl. xʼɛːxʼenn ~ xʼennxʼenn by W. Bleek; quoted as xʼar̃i, pl. xʼexʼen in [Bleek 1929: 22]. All of these forms are always glossed in the meaning 'little bird'. For 'large bird', the correct word seems to be ɡǀǀɛrritǝn-ti, pl. ɡǀǀɛrritǝn-de ([Bleek 1956: 530]; [Bleek 1929: 22]), transparently derived from the word ɡǀǀɛ́rri(ya) 'feather' [ibid.]. It is not quite clear which one is more frequent statistically, but xʼãnni is quoted in several contexts where the semantics 'bird' is clearly more important than 'small', indicating that this is probably the more "generic" word of the two in ǀXam.
Bleek 1956: 334. Quoted as ǀwíː in [Bleek 1929: 22]. Plural form: ǀwi-ŋ ~ ǀwi-nyǝn. Judging by textual examples, the word can also be used in the meaning 'vulture': ǀwiŋ eː müri "vultures eat goats" [Bleek 1956: 334].
Maingard 1937: 240. Plural form: ǀwi-ke ~ ǀwi-ɕe. Cf. ǀʼɔ̄ʔȭ-sī 'small bird' [Doke 1936: 73] (probably the same lexeme, although the click efflux does not match with Maingard's variant).
Sands et al 2006. Phonetically transcribed as [ǀqʰǝi-si] in Miller et al. 2009: 152. Polysemy: 'bird / vulture', although in the latter meaning the word seems to be attested without the singulative suffix -si, i. e. as ǀqʰui [Miller et al. 2009: 155].
Bleek 1937: 205; Bleek 1929: 22; Bleek 1956: 170. The prefixal element si= is attested in several other ǀʼAuni words as well (e. g. si=!ã 'kaross'), but its meaning remains unclear. It must, however, be a separate morpheme due to the general laws of the word structure in ǀʼAuni. Cf. also ɡǀoː 'bird' ([Bleek 1937: 209]; [Bleek 1956: 280]), which could be, despite the difference in click efflux articulation, a prefix-less variant of the same root.
Story 1999: 21. The prefixal element si= here is the same as in ǀʼAuni.
Proto-!Wi:*ǀqʰu-
Distribution: Preserved everywhere except in ǀXam. Replacements: In ǀXam, replaced by xʼãnni, clearly related to Central Khoisan *xʼani 'vulture; (> bird)'; the quirkiness of the situation is in that the original !Wi word for 'bird' was retained in ǀXam, but in the meaning 'vulture': ǀwiː ~ ǀu̯iː (W. Bleek) [Bleek 1956: 334]. Perhaps the original meaning was narrowed down {'bird' > 'vulture'}, while the new word was borrowed from Khoekhoe, although neither the source nor the very fact of borrowing can be ascertained. Phonetic shape: The click efflux is reconstructed based on the attested reflexation in Nǀuu (none of the earlier sources consistently mark uvular effluxes), but is not very certain. "Narrow !Wi" consistently reflects the bivocalic stem *ǀqʰui, but comparison with the Lower Nǂossob languages shows that -i is most likely a fossilized class marker.
Distribution: Preserved in all languages. Reconstruction shape: Glottalized articulation of the affricate is well supported by data from all primary branches.
Bleek 1956: 464. Attested only in L. Lloyd's materials and therefore absent in [Bleek 1929]. A very close synonym exists in ǀʰoá-ka ~ ǀʰóaː-ka ~ ǀʰóa-kǝn ([Bleek 1956: 289]; [Bleek 1929: 22]); however, in [Bleek 1956] the word is glossed as 'dark; black; used for any dark colours', and analysed as derived from the noun ǀʰóˤː 'darkness' [Bleek 1956: 288]. The latter comparison is not certain (vocalic structure of the two words is quite different), but for ǀʰoá-ka several examples with the meaning 'dark blue', etc., are actually quoted, and, with the addition of external parallels that seem to confirm the archaic nature of !weːn in the meaning 'black', the latter is currently the more eligible term for this position of the two.
Bleek 1956: 439, 463. Quoted as !weː in [Bleek 1929: 22]. The latter source adds a secondary synonym: ɡǀɔː, confirmed in the same orthography in [Bleek 1956: 280]. However, textual examples are provided only for !we (even if they are not entirely diagnostic, cf.: ǀxʼaːse ku !oe "a snake which is black" [Bleek 1956: 439]); !we is also much better confirmed by external data.
Ziervogel 1955: 39, 40, 58. The n-prothetic form is, most likely, a samdhi variant. Quoted as ǯwaː ~ ǯwãː 'be black' in [Lanham & Hallowes 1956: 108] (the authors note that, as a noun, the word means 'pot', but this is probably just a case of homonymy).
Distribution: Preserved (at least) in the ǀXam-Nǀuu branch and, quite possibly, in ǀǀXegwi. Replacements: ǀHaasi ǀǀe has no etymology and is unlikely to correspond to ǀXam-Nǀuu *!oe (the clicks are incompatible). The word could theoretically reflect something archaic (distribution-wise, this is possible), but, being attested only in a single, not very reliable source, and being completely devoid of internal and external parallels (for now), should not be eligible for proto-status as a primary candidate. Reconstruction shape: Correspondences between ǀXam and Nǀuu are straightforward, with the exception of unpredictable aspiration in the latter (this could be a transcriptional error in one or more of the sources). Final nasal in ǀXam is most likely suffixal. A particular problem concerns the form found in ǀǀXegwi. If we analyze it as a separate root, then ǀǀXegwi čwa ~ nčwa finds no suitable parallels in any other !Wi languages and, especially in the light of the variant with the initial nasal, could look suspiciously like a potential borrowing from one of the click-dropping Kalahari Khoe languages (cf., for instance, ǀXaise nʓú 'black'); in fact, the click-dropping could have taken place within ǀǀXegwi itself, since this language regularly dispenses with the palatal click along the same lines as East Kalahari Khoe languages (see 'wind', etc.). However, mass borrowings into ǀǀXegwi basic lexicon from Khoe are not a norm, and this scenario is no more likely than, with certain reservations, an attempt to regard ǀǀXegwi čwa as a potentially regular development from Proto-!Wi *!oe (> *koe > čoe > čwa). At least one similar, if not completely identical, example exists that could also reflect the same palatalization of a former alveolar click or velar stop (see 'stone'), and we also have evidence for palatalization before *e in ǀǀXegwi in the case of the palatal click (cf. the reflexation of *ǂ- in such items as 'rain', 'short', 'wind'). For this reason, we tentatively select the etymological decision that the ǀǀXegwi equivalent for 'black' continues the old !Kwi stem.
Bleek 1956: 634. Emphatic form: ǀǀxau-ka-kǝn. Transcribed by W. Bleek as ǀǀxáu-kǝn ~ ǀǀxáu-ki, emph. ǀǀxau-ka-kǝn; quoted as ǀǀxaukn in [Bleek 1929: 22]. The form is morphologically complex, but only suffixal forms are attested. Possible secondary synonym: ɡ!aːuː, emph. ɡ!aːuː-kǝn ~ ɡ!aːuː-gǝn [Bleek 1956: 378]; attested only in L. Lloyd's records and not confirmed in any way by external data.
Distribution: Preserved in all languages. Reconstruction shape: The most difficult question here is whether ǀǀXegwi ƛʼẽũ (a form attested only in the sparse data of Lanham and Hallowes) can go back to *ǀǀxau(-N), since normally the lateral affricate in ǀǀXegwi develops out of a palatal click in Proto-!Wi (see 'dog', 'moon', etc.). Very tentatively we accept this as part of the same etymology, due to the possibility of transcriptional error or a unique development of the lateral influx before a velar fricative efflux (no other examples in the corpus). Apart from ǀǀXegwi, most of the other forms correspond to each other on a trivial level. Semantics and structure: The stem is either used on its own or with a nominal extension (-ke ~ -ken), typical of many other basic nominal items as well.
Bleek 1956: 457. Plural form: !waː-gǝn ~ !wáː-ka-kǝn ~ !wáː-kn. Quoted as !wa, pl. !aːgn in [Bleek 1929: 23] (the plural form is probably a misprint for *!waːgn). The word is very similar in form to 'leg, root' [Bleek 1956: 457], but subtle differences such as a short vowel in 'bone' (!wắ) vs. long vowel in 'leg' (!wáː), or the fact that only 'leg' forms its plural stem with the aid of reduplication (!wá=!wá-gǝn) indicate that these words are not even complete homonyms. Cf. also !ã́ 'bone of arm or foreleg' [Bleek 1956: 401], similar in form but probably a different root nonetheless.
Bleek 1956: 548. Polysemy: 'bone / bone knife'. Quoted as ǀǀaba, pl. ǀǀaǀǀa in [Bleek 1929: 23]; ǀǀaba, pl. ǀǀaǀǀa ~ ǀǀɛŋ in [Bleek 2000: 19]. Comparison of sg. and pl. forms shows that *ǀǀa is likely to have been the original root, although this particular word-formation model is very unusual.
Distribution: Preserved in ǀXam and ǀǀXegwi; replaced in the entire Nǀuu cluster; not attested in Lower Nǂossob languages. Replacements: Common Nǀuu *ǀǀaba 'bone' regularly corresponds to ǀXam ǀǀabba 'a piece of eland's bone that forms part of the completed arrow' [Bleek 1956: 548]; this allows to suppose a broadening of the original meaning: {'piece of animal bone' > 'bone (gen.)'}. The opposite development is not excluded, but contradicts the general distribution of *ǂ2a. Reconstruction shape: The correspondence of ǀXam !- to ǀǀXegwi !- is very rare, but all instances of ǀǀXegwi !- go back to a special phoneme that we tentatively mark as *ǂ1-, which is also reflected in ǀXam as !-, so essentially the correspondence seems to be regular (see 'one' for another example). Labialization in ǀXam, as in many other similar cases, seems to be secondary (see 'foot', 'liver', etc.), although its causes and conditions have not yet been established.
Bleek 1956: 623. Quoted as ɳǀǀɔein-tu in [Bleek 1929: 28]. Transcribed by W. Bleek as ɳǀǀwaíŋ-ttu ~ ɳǀǀwéin-ttu [Bleek 1956: 623]. The element -tu is a standard suffix for body parts. A possible synonym is ǀǀaːxu [Bleek 1929: 28], ǀǀaxu [Bleek 1956: 564], but the word is only found in W. Bleek's early (less certain) records; examples of contexts given in [Bleek 1956] yield both the meaning 'chest' and 'women's breasts'. This may, in fact, be simply one specific usage of the word ǀǀaxu ~ ǀǀãxu 'side' [ibid.]. The principal word for 'chest' is quite distinct from the main word for 'female breast' (also 'milk'): !ʰwai ~ !ʰwei-tǝn [Bleek 1956: 431].
Bleek 1956: 624. Quoted as ɳǀǀweːntu in [Bleek 1929: 28]. Possible secondary synonym: ǀǀaːtu [Bleek 1929: 28], reproduced in [Bleek 1956: 560] as ǀǀatyu ~ ǀǀãtu, pl. ǀǀate-ŋǝn, with no textual examples (for some reason, the word is marked there as "SI", i. e. ǀXam rather than ǀǀNg!ke, but this is almost certainly a mistake, since the form is credited to D.B., i. e. Dorothea Bleek, whose research was on ǀǀNg!ke rather than ǀXam).
Ziervogel 1955: 43. Quoted in the phrase haː-n-šagu 'it is my chest' in [Lanham & Hallowes 1956: 98]. Distinct from ɕʰa-zi 'female breast' [Lanham & Hallowes 1956: 102]. The same word is quoted as kaː-zi in [Bleek 1929: 28] and [Bleek 1956: 84] (with the palatal consonant is transcribed as a velar), but the meaning is erroneously glossed as 'chest' (male).
Distribution: Preserved in the entire ǀXam-Nǀuu cluster, probably replaced elsewhere. Additionally, cf. ǀǀKuǀǀe ɡǀǀɔintu 'breast' [Bleek 1956: 532]. Replacements: (a) ǀǀXegwi ša-gu regularly corresponds to Nǀuu ǂqʰaː 'sternum / breastbone' (see notes on Nǀuu; the correspondence is exactly the same as in the word for 'wind' q.v.); it is likely that we deal here with a broadening of the original meaning ('breastbone' > '(male) chest'), provided, of course, that the ǀǀXegwi item is accurately glossed as far as semantics is concerned; (b) ǀʼAuni ǂan 'chest' has no etymology; because of this, it is a serious contender for Proto-!Wi 'breast', but is not technically reconstructible to the same level of chronological depth as *ɳǀǀoiŋ. Reconstruction shape: Reconstruction of the click efflux is approximate (nasalization, marked in early transcriptions by W. and D. Bleek, could technically reflect the influence of nasalized vowels), as is the reconstruction of the stem diphthong. It is quite probable that the stem incorporates the same nominal suffix -iŋ as in other items on the Swadesh list (e. g. 'dog' q.v.), in which case one could think of a morphological segmentation into *ɳǀǀo + *-iŋ, with subsequent assimilative processes (*-oi- > -oe- or *-oi- > -ui-) depending on the dialect. Semantics and structure: All of the discussed items are strictly limited to the semantics of 'male chest'; the meaning 'female chest' in !Kwi is usually expressed by the same stem as 'milk' and represented by other lexical roots.
Bleek 1956: 544, 566. Quoted as ǀǀe ~ ǀǀa in [Bleek 1929: 25]. Transcribed as ǀǀa ~ ǀǀaː ~ ǀǀɛː by W. Bleek. The vocalic variation is probably due to contraction with different class markers, although this is hard to verify based on available descriptions and examples. It is important to stress that, even though [Bleek 1929] quotes this stem under 'burn (intr.)', most textual examples in [Bleek 1956] explicitly confirm its transitive usage; conversely, 'burn (tr.)', for which she gives the equivalent !uːi, is the intransitive stem 'to burn / smart / pain' [Bleek 1956: 449], so there must have been some confusion.
Bleek 1956: 545, 566. Quoted as ǀǀa in [Bleek 1929: 25] and [Bleek 2000: 21]. Vocalic gradation is a typical feature of short monosyllabic verbal stems in ǀǀNg!ke. As in the case of ǀXam, [Bleek 1929: 25] glosses ǀǀa as intransitive, but textual examples clearly confirm its transitive use: cf. ǀǀʼõĩ ɳǀe sa, ha ǀǀa ŋ "the sun comes, it burns me" [Bleek 1956: 545], etc. In the meaning 'to burn (transitive)', [Bleek 1929: 25] yields the equivalent kukúru, which, in this particular form, is not confirmed in [Bleek 1956]; the closest parallel is kurúke 'burn' [Bleek 1956: 107], which is only attested as intransitive (!): ǀʼi kurúke "the fire burns" [ibid.].
Bleek 1937: 209; Bleek 1956: 294. Meaning glossed as 'to burn, light a fire, roast'. The accompanying example is ǀá n ǀʼi "light the fire" (typologically, "light the fire" frequently = "burn the fire" in this region, so this may indeed have been the default verb for 'burn' in ǀʼAuni).
Distribution: Preserved in ǀXam, ǀǀNg!ke, possibly also ǀHaasi. Additionally, cf. ǀǀKuǀǀe ǀǀaː 'to burn' [Bleek 1956: 545]. Replacements: (a) Nǀuu !xao 'to burn' is comparable with ǀXam !xãũa 'to cook' [Bleek 1956: 498], but it is not possible to determine the original semantics based on this contrast, although distribution of the various terms for 'burn' shows that Nǀuu !xao is clearly a semantic innovation; (b) ǀʼAuni ǀá is an isolated entry with no parallels in sight. The root *ǀǀa is found in [Bleek 1956: 545] for both ǀǀXegwi and ǀʼAuni, but only in the meaning 'to cook', which presupposes an areal semantic isogloss: {'to burn' > 'to cook'}. Reconstruction shape: We select *ǀǀa as the basic (original) variant of the root; *ǀǀe seems to be the result of contraction with a suffixal marker. The issue of why the ǀHaasi equivalent was recorded with a labialized vowel is even more obscure; however, we do not find this a sufficient reason to exclude the word from the etymology, since it does not have a better one, and complicated, poorly understood vowel gradation in verbal roots is a commonality in all !Wi languages.
Bleek 1956: 593. Plural form: ǀǀu-ǀǀúttǝn (with reduplication and suffixation). Quoted as ǀǀuru, pl. ǀǀuǀǀutǝn in [Bleek 1929: 60]. Transcribed as ǀǀuru, pl. ǀǀuǀǀúddi ~ ǀǀuǀǀúti by W. Bleek. Plural forms show that -ru in the sg. forms should be judged a detachable class suffix (at least, synchronically).
Ziervogel 1955: 43. Plural form: !ʼelo-le. Cf. also ǀǀɔla 'fingernail' in [Bleek 1929: 60] and [Bleek 1956: 586]. The forms in [Ziervogel 1955] and D. Bleek's records are compatible only if we assume that at least one of them has been significantly mistranscribed.
Story 1999: 22. The form is probably plural (kʼa= is a productive plural prefix). Cf. ǂü̃ 'finger' [Story 1999: 21]; despite the difference in click effluxes, this might be the same root.
Proto-!Wi:*ǀǀqor-
Distribution: Preserved in all branches of the group, but possibly replaced in ǀǀXegwi and ǀHaasi. Replacements: (a) ǀǀXegwi !ʼelo-loŋ is a dubious entry in Ziervogel's materials: it contains a rare case of the alveolar click (attested either as a reflexation of the rare phoneme *ǂ1- or in borrowings) and contradicts the materials of D. Bleek, which clearly show that the old word for 'fingernail' was being preserved in at least some dialects of ǀǀXegwi. Consequently, this may be a pseudo-replacement, particularly if the word was mistranscribed or its semantics was inaccurately glossed; (b) ǀHaasi kʼa=ɳǂü is formally a plural from ǂü̃ 'finger'; it is not understood whether R. Story simply misglossed the semantics of the 'word' or whether the meanings 'finger' and 'fingernail' were genuinely merged in ǀHaasi. In the former case, this is yet another pseudo-replacement. Reconstruction shape: The uvular efflux is tentatively set up for this word based on the accurately transcribed form in Nǀuu (earlier sources do not recognize the existence of uvular effluxes in most !Wi languages). The stem is always bisyllabic, but the second vowel is hard to reconstruct: reflexes of *ǀǀqoru, *ǀǀqoro, *ǀǀqori, and *ǀǀqora are all attested, with the variation reflecting either old morphological gradation (e. g. different suffixes for sg. and pl. numbers) or the results of vocalic assimilation (both with the first syllable vowel and the vocalism of additional suffixes, e. g. *ǀǀqoro-sa > ǀǀora-sa in ǀʼAuni, etc.).
Bleek 1956: 329. Emphatic form: ǀwáˤ-ka-kǝn. Transcribed as ǀwaː-gǝn by W. Bleek. Quoted as ǀwa-tn ~ ǀwaː-gn in [Bleek 1929: 29]; the first variant here is, however, most likely a misunderstanding - in [Bleek 1956: 331] the same lexeme is glossed as 'star, cloud', even though all the textual examples quoted from W. Bleek's and L. Lloyd's records exclusively convey the meaning 'star', never 'cloud'. Even if both words are formed from the same root (typologically, very dubious), they are clearly distinguished through different suffixes (velar -kǝn for 'cloud', dental -tǝn for 'star' q.v.). As a possible secondary synonym, cf. also ɡ!ùru 'white clouds' [Bleek 1956: 389] (apparently a more rarely encountered word than ǀwaːˤ-gǝn, but analysis of contexts does not exclude the possibility that the latter is really 'raincloud' and the former is 'white cloud'; were this to be confirmed, we would have to swap the forms around).
Bleek 1956: 203; Bleek 1929: 29. In the former source, the form is defined as plural ('clouds': !àˤ kãũ tiɔːke "rain falls from the clouds"), which makes sense, since -ke is one of the productive plural markers in ǀǀNg!ke. Secondary synonym: ɡ!um ([Bleek 1956: 388]; [Bleek 1929: 29]), with no textual examples.
Bleek 1929: 29. Not attested in [Ziervogel 1955] or Lanham & Hallowes' papers. The existence of the word is confirmed in the English indexes in [Bleek 1956: 705], but the main section of the dictionary only lists ǀǀxeːŋ 'woman' [Bleek 1956: 635] - probably a typographic error ("merger" of two entirely different words). Still, a somewhat dubious entry.
Bleek 1937: 209; Bleek 1956: 290. Meaning glossed as 'cloud' (sg.) in the former source and as 'clouds' (pl.) in the latter; according to D. Bleek's observation, the suffix -sa generally marks singulative forms (opposed to -si in the plural). The word is not found at all in [Bleek 1929: 29], which yields a different equivalent, ǀǀkʼani, not confirmed in later sources.
Story 1999: 21. Meaning glossed as 'clouds' (pl.). There are not enough data to decipher this compound properly.
Proto-!Wi:
Not reconstructible. Each language has its own equivalent for this Swadesh meaning, and the accuracy of semantic notation may be placed under doubt in most cases. Considering that the meaning 'cloud' is generally very unstable in the Khoisan area, we prefer to abstain from any attempts at Proto-!Wi reconstruction here.
Bleek 1956: 119. Emphatic form: xʼaːoˤ-wa. Quoted as xʼàoˤ in [Bleek 1929: 29]. Transcribed as xʼáo ~ xʼaːoː by W. Bleek. The word is listed as the first equivalent for 'cold' in [Bleek 1929], said to be the lexical opposite of kau 'to be warm' in [Bleek 1956] and illustrated with several examples in which it is found in such noun phrases as 'cold wind' etc. Possible synonyms include: (a) sérri ~ ssèrri-tǝn ~ ssérri-tǝn 'cool, cold' [Bleek 1956: 167], also found as a noun: ssérreː ~ ssèrreya ~ ssérri-tǝn 'cold wind' [ibid.]; (b) ǀǀxwèː ~ ǀǀxwéː 'to be cold, become cold', ǀǀxwéː-tǝn 'cold (n.)' [Bleek 1956: 639]. Attested examples and descriptions do not allow to estimate these words' real chances at filling the primary slot for 'cold'.
Bleek 1956: 289; Bleek 1929: 29. The latter source also mentions a special verbal stem siːya 'to be cold', but it is not confirmed anywhere in [Bleek 1956].
Maingard 1937: 243. Attested in the phrase ŋ ɕa ǀʼʰu "I am cold". Entirely different stem, not confirmed by external sources, is found in [Doke 1936: 63]: kāɾīʔī.
Not reconstructible; every language has its own equivalent for this Swadesh meaning. Curiously, the best candidate for Proto-!Wi 'cold' is a word whose reflexes cannot be easily defined as the basic equivalent for 'cold' in any attested !Wi language: Proto-!Wi *ǀǀxoe, marginally attested both in ǀXam and ǀǀNg!ke (ɡǀǀaː ǀǀxweː "night's coolness, evening" [Bleek 1956: 639]) and possibly of the same origin as ǀʼAuni ǀǀxau [Bleek 1929: 29] (although the coda correspondences would be quite irregular) and ǀǀXegwi !xoa [Bleek 1929: 29], under the condition that Bleek's ! in this case mistranscribes the lateral click ǀǀ. This choice is also indirectly supported by external parallels in Taa. Nevertheless, judging from a formal perspective, attested data are too scarce and ambiguous to postulate a lexical replacement from Proto-!Wi to all of its modern day descendants.
Bleek 1956: 165. Quoted as sʼi ~ ši ~ sʼe in [Bleek 1929: 30]. Transcribed as se ~ sːe ~ ssʼe by W. Bleek. The same root is also encountered with different vocalism, transcribed as sːaː ~ ssʼaː ~ sːa by W. Bleek and sːaː by L. Lloyd [Bleek 1956: 161]; also saŋ ~ sːaŋ ~ sːaˤŋ (W. Bleek), sːaˤŋ (L. Lloyd) [Bleek 1956: 163]. The variations are typical of verbal roots (especially statistically frequent ones) and may indicate contractions with various class markers. Some of the variants, as attested in the accompanying examples, may also have the causative meaning 'to bring'.
Bleek 1956: 161, 165, 166, 168. Quoted as si ~ se ~ saː in [Bleek 1929: 30]; sa ~ sieya ~ saː in [Bleek 2000: 20, 22]. Vocalic variations are typical of simple verbal stems and may indicate contractions with various class markers or verbal particles.
Maingard 1937: 245, 251. Transcribed as sīyā [Doke 1936: 75]. Secondary synonym: ǀǀŋ̄ [Doke 1936: 69] - a highly dubious form, considering a complete lack of external parallels. Perhaps the real meaning is slightly different ('to arrive'?), or imperative ('come!').
Ziervogel 1955: 51; Lanham & Hallowes 1956: 102. Past tense form is quoted as seya, future tense - as se [ibid.]. Quoted as se ~ sa in [Bleek 1929: 30] and [Bleek 1956: 161, 165].
Bleek 1937: 205; Bleek 1956: 161, 165, 168. Quoted as sɛː in [Bleek 1929: 30]. Vocalic variants may indicate contractions with class markers or verbal particles.
Story 1999: 21. Secondary synonyms include: ǀi ~ ɳǀi ~ ɳǀi-sa. These forms may be the same as cʼi, provided the affricate could sometimes be misheard by Story as a dental click; in any case, unlike cʼi, they have no external links.
Proto-!Wi:*sa ~ *si
Distribution: Preserved in all daughter languages. Additionally, cf. ǀǀKxau saː ~ seː id., ǀǀKuǀǀe sa ~ si id. [Bleek 1956: 161, 165]. Reconstruction shape: Initial consonant is a sibilant (as opposed to 'bite' q.v.), as preserved in the majority of reflexes. Occasional attestations of a glottalized sibilant (sʼ-) or even a glottalized affricate (cʼ-) most likely reflect the result of stem contraction: *sV-/ʔ/a ~ *sV-/ʔ/i (combinations with different suffixes) > *sVʔa ~ *sVʔi > *sʼa ~ *sʼi. As in many similar cases, original root vocalism is difficult to reconstruct because of contractions with various vocalic suffixes; both *a and *i are at least equally probable.
Bleek 1956: 267. Transcribed as ǀʼaː, emphatic f. ǀʼaː-kǝn, pl. ǀʼáǀʼa-kǝn by W. Bleek. Quoted as ǀʼaː in [Bleek 1929: 33]. Secondary synonym: ǀúː-kǝn ~ ǀù-kǝn ~ ǀù-ka (L. Lloyd), ǀuː-ka ~ ǀúː-kǝn (L. Lloyd) [Bleek 1956: 324]. Relations between these two stems are complex. The first root is glossed as 'to fight, die, be killed; n. fight, harm, curse' in [Bleek 1956]; this means, almost for certain, that we are dealing here with at least several phonetically close, but confused stems, but the really interesting detail is that, of the quoted examples, very few actually refer to 'death' (although at least one phrase from W. Bleek's records is very explicitly translated: ŋ ǀʼaː 'I die'). The second stem, possibly derived from ǀu 'to be ill' [Bleek 1956: 322], is translated as 'to die, be dead, faint', and may rather express the stative meaning '(to be) dead' (cf.: ha xʼauki ɲ!aunko ǀuːka 'he has not yet died', etc.), but it is also possible that the word was on its way to replace the older root ǀʼa (a chance which it never got due to the extinction of the ǀXam language).
Bleek 1937: 208; Bleek 1956: 267. Meaning glossed as 'dead'; cf. kʰora ǀʼe kie ǀʼã "Khora is dead". In [Bleek 1937: 208], the word is actually transcribed as ǀǀʼã; the lateral click is clearly a misprint, but it is curious that in D. Bleek's earlier records, the word 'to die' is transcribed as ǀʼaǀʼan [Bleek 1929: 33] - possibly a reduplicated variant of the verbal stem? Cf. also presumably suffixal derivates: ǀʼáuo ~ ǀʼáu 'dead' [Bleek 1937: 208]. There is also a strange expression heri ɳ!uːba ([Bleek 1937: 202]; [Bleek 1956: 60]), which D. Bleek translates as 'has died'; probably some sort of idiomatic euphemism.
Distribution: Preserved in all daughter languages except for ǀHaasi. Additionally cf. ǀǀKxau ǀʼa, ǀǀKuǀǀe ǀʼa id. [Bleek 1956: 267]. Replacements: In ǀHaasi, replaced by !ʰo of unclear origin. Reconstruction shape: Correspondences are generally quite straightforward. Lack of click glottalization in Ziervogel's and Lanham / Hallowes' ǀǀXegwi data is not easy to explain; possibly the result of contamination with the old !Wi verb 'to kill' q.v.
Bleek 1956: 433, 467. Emphatic form: !úíŋ-yaŋ. Plural form: !úíŋ-!úíŋ ~ !wìŋ-!wiŋ ~ !ʰwíŋ-!ʰwíŋ. Quoted as !wiŋ in [Bleek 1929: 34]. Also transcribed as !wíŋ by W. Bleek.
Miller et al. 2007: 58. The two variants allegedly reflect dialectal variants (first one is Western dialect, second one is Eastern). Quoted as ǂʰou in [Westphal 1965: 141].
Ziervogel 1955: 37, 39, 44. Plural form: ƛu-me. Quoted as ƛʰwiŋ in [Lanham & Hallowes 1956: 112]. The plural form is quoted as ƛʰu-miŋ in [Lanham & Hallowes 1956: 104]. Quoted as ǀǀwi in [Bleek 1929: 34] and [Bleek 1956: 600]. The latter source also quotes the form !xẽ as a synonym [Bleek 1956: 499]. It is not highly likely that ǀǀwi and !xẽ are phonetic variants, or even side effects of mistranscription; !xẽ is probably a different word (which, furthermore, could have been misglossed).
Bleek 1937: 219; Bleek 1956: 663. Quoted as ǀʼʰaŋ in [Bleek 1929: 34] (could be the same root with a misheard click, cf. the transcription ǂʰaŋ in the closely related ǀHaːsi language).
Story 1999: 21, 30. Plural form: ǂʰaaŋ ~ ka=ǂʰaaŋ.
Proto-!Wi:*ǂʰu-
Distribution: Preserved in all daughter languages. Additionally, cf. ǀǀKxau ǂʰuni id., ǀǀKuǀǀe !ʼwiŋ id., !Gã!ne !ʼinyi id., Seroa kuenia id. [Bleek 1956: 104, 400, 495, 662]. Reconstruction shape: The original palatal click is correctly reflected as alveolar (!) in ǀXam and Bleek's ǀǀNg!ke, and expectedly develops into a lateral affricate in ǀǀXegwi. The aspirated click efflux is tentatively reconstructed based on accurately transcribed Nǀuu data. Semantics and structure: The stem displays a significant number of suffixal variants: *ǂʰu-iŋ (ǀXam, ǀǀNg!ke; modern Nǀuu *ǂʰuɲ probably goes back to the same variant, with palatalization of the nasal and subsequent contraction of the stem), *ǂʰu-aŋ (Lower Nǂossob, with subsequent contraction in both dialects), *ǂʰu-e ~ *ǂʰu-i (ǀǀXegwi). Their functions in Proto-!Wi and/or subsequent stages of language development remain unclear (unfortunately, where some of these variants are attested in the same language, as in Ziervogel's ǀǀXegwi records, their respective functions are unknown).
Bleek 1956: 126. Transcribed as xʼwã ~ xʼwãŋ ~ xʼwã́ ~ xʼwẽː ~ xʼwɛ̃ by W. Bleek. Quoted as xʼwã ~ xʼwẽ in [Bleek 1929: 34]. Vocalic gradation, as in many other simple verbal stems, probably reflects results of merger with different class markers.
Bleek 1956: 117, 121, 601. Quoted as xʼã ~ xʼẽĩ in [Bleek 1929: 34]. Occurrence of the lateral click in the variant ǀǀxʼã is a unique idiosyncrasy that is most probably negligible (no etymological data can confirm click articulation for this stem).
Miller et al. 2009: 157. Transcribed phonetically as [kχʼǝ̃ĩ]. In [Sands et al. 2006], additional (morphologically determined?) variants xʼa and xʼã are given.
Ziervogel 1955: 39. This is the present tense stem; the past tense is glossed as xʼaː [ibid.]. Quoted as xʼẽĩ in [Lanham & Hallowes 1956: 115]. Quoted as xʼã ~ xʼẽ in [Bleek 1929: 34]; as xʼaː ~ ǀǀxʼẽ in [Bleek 1956: 117, 604].
Distribution: Preserved in all daughter languages. Additionally, cf. ǀǀKuǀǀe kwã ~ ǀǀxʼwãĩ, Seroa xʼã, !Gã!ne xʼaː id. [Bleek 1956: 109, 116, 609]. Reconstruction shape: Nasalization is such a persistent feature of this stem's vocalism that it is tentatively included in the reconstruction, although it is also true that there are non-nasal reflexes as well (ǀǀXegwi, ǀHaasi), and that nasalization could be a secondary feature, generalized to the root from contracted variants with verbal suffixes. Labialization in ǀXam is clearly secondary (no other language has this feature, and "superfluous" labialization, particularly after velars, is a very common feature in this language).
Build treei