Lexico-statistical studies in Khoisan II/I: how to make a Swadesh wordlist for Proto-Tuu

FULL [-]

!Ui: *!qauq (|Xam !вьiy ~ !auiy ~ !дь:enyд, ||Ng!ke !xny, N|uu !qm-ya). 0 Not attested anywhere other than |Xam and the N|uu cluster. The root is verbal in origin (`to fill / be filled'). Protoform is approximate (*!q- reconstructed based on N|uu as well as occasional |Xam transcriptions with velar ejectives, e.g. Ik'au `to fill'; coda *-auy could also be *-aiy, since labialization frequently turns out to be of secondary origin in |Xam).

Nossob: |'Auni Цх'дп-si. 0 Cf. also Цай `to fill', which can only be related if the click efflux in one of the forms is mistranscribed. Not attested in |Haasi.

Taa: *!um (!Xoф lщhm, Kakia lum, N|u||en lum). 0 Another !Xoц equivalent is the verb l'цla, but the latter has no parallels in other varieties of Taa.

Tuu: Not reconstructible. Each subgroup has its own equivalent for this concept.

GIVE [-]

!Ui: *a (|Xam в: ~ а: ~ a: ~ a-a ~ a-a, ||Ng!ke a ~ a:, N|uu ?a:). 0 Although the verb is not attested beyond |Xam and the N|uu cluster, it is clearly the most basic equivalent for `to give' in both of these nodes and is easily reconstructible for Proto-!Ui. Nasalization is infrequent and likely secondary; quality of root vocalism is notably stable. The only other language where the main equivalent for `to give' is perfectly clear is | Xegwi, cf. ||Xegwi-Z sa, ||Xegwi-LH sa ~ s- (as in in za s-e `I will give'). It correlates with ||Ng!ke sa: `to bring, fetch' and with ||Kxau у-sa `to give' (where y- may be `me'); ultimately, all these forms can be explained away as originally causative formations from *sa ~ *si come q.v.

Nossob: Not reconstructible. 0 The situation in Nossob languages is complicated. For |Haasi, the only recorded equivalent is the monovocalic verb i; its cognacy with !Ui *a is not excluded, but given the total lack of vocalic alternations in this root in any of the !Ui languages, there is nothing to confirm it. For |'Auni, Bleek records (a) rather marginal a `to give', only attested in one or two dubious examples; may be a result of mistaken analysis or a N|uu form; (b) Ja ~ Jo ~ Jo, most often used in an imperative function and consequently comparable with |Xam Ja `let, give' (also typically an imperative). Because of this variety and the relative unreliability of Nossob data, it is better to exclude the word from comparison.

Taa: *!qha- (!Xoц !qha:, Kakia !xa: ~ !xe:). 0 Apart from this autonomous verb, the meaning `give' is also expressed in !Xoц by the auxiliary “verb-postposition” Jа:, e.g. jщa Jа: `to pass to, give to', where jщa by itself = `hold, grab, grasp'. This is probably the same word as N|u||en Ji `give', but its usage in !Xoц makes it somewhat less eligible for inclusion (and there is no way to verify if it actually displaced *!qha- in N|u||en or just accidentally happened to be the only recorded variant for give).

Tuu: Not reconstructible. 0 The only secure isogloss between more than one branch of Tuu is the auxiliary verb *JV-, commonly used in the imperative function (`give!') in some of the !Ui and Nossob languages, and in a postpositional function in !Xoц. The principal indicative forms are, however, clearly different between !Ui and Taa, and somewhat obscure in Nossob. Available data do not allow to reconstruct a precise historical scenario.

GOOD [-]

!Ui: Not reconstructible. 0 Most languages have their own equivalents, including some roots of clearly non-!Ui origin (||Xegwi-Z luga-ge, probably a Bantuism; fKho-D h'вm-зй ^ Khoekhoe *h'am `right; true') and some with very weak distribution (|Xam a:-kdn `good'; |Xam t:wв:i-п `good'; ||Ng!ke kyai ~ kyai[ N|uu <зhп-kп ^ *thai ~ *thп).

Nossob: Not reconstructible. 0 Not attested in |Haasi. For |'Auni, Bleek records the variants xwe ~ xwoi, without any etymology.

Taa: Not reconstructible. 0 !Xoц qвп has no parallels in Kakia or N|u||en. Not attested in Kakia; N|u||en Цї also has no etymology.

Tuu: Not reconstructible. 0 The meaning GOOD is clearly very unstable in Tuu; the concept itself seems rather diffuse, and the relations between all these forms and similar forms in other Khoisan groups may reflect a complex network of areal interaction (cf. Proto-Khoe *!ап `good'; fHoan qhaл `good', etc.).

GREEN [-]

!Ui: Not reconstructible. 0 Highly unstable and poorly documented meaning. In modern N|uu, expressed with H'hao-a, a borrowing from Khoe (cf. Nama Hhao `to turn green; to grow').

Nossob: Not reconstructible. 0 Not attested in |'Auni. |Haasi Hau is probably borrowed from the same Khoe source as the N|uu item.

Taa: *|air- (!Xoц /шй, Kakia /аi5}. 0 Cf. !Xoц /аim-si, pl. Iаmm-sв `dung beetle', most likely containing the same root; morphological structure of the noun suggests that -i- is an original class suffix and that the semantics of `beetle' might be primary. One might also speculate about further links with */ana (^*/a-na ?) leaf q.v.

Tuu: Not reconstructible. 0 The concept is generally unstable, not very well documented, and most of the languages have their own ways of expressing it.

36. HAIR [!Ui + Nossob + Taa]

!Ui: (|Xam jь(-ken) ~ jhь(-kdn), ||Ng!ke ju ~ jhь, fKho-M ju ~ jhu, N|uu jhu:-ke, ||Xegwi-Z

jhu-zi, ||Xegwi-LH jhь. 0 A super-stable word with fairly transparent phonology, though the aspirated articulation of the efflux tends to go unnoticed in older transcriptions.

Nossob: *|ho (|'Auni jhoo, |Haasi jo). 0 Aspirated articulation explicitly marked by Bleek for |'Auni, but noticeably absent in |Haasi.

Taa: *|qhu- ([qhuд ~ lqhuд, Kakia jwai-ni, N|u||en Ішп-te). 0 Kakia and N|u||en forms are

obviously plurals. It may be tentatively assumed that the complex voiced aspirated uvular click in !Xoц is primary, although there is variation between voiced and voiceless articulation even within !Xoц itself. Nasalized coda vowel in !Xoц is detachable as a class 2 suffix.

Tuu: *|hu-. 0 All forms are clearly related. The correspondence N|uu -h- : !Xoц -qh- is recurrent (see dog). Nossob forms are slightly closer to !Ui due to lack of nasalization, but since nasal codas are occasionally attested in !Ui as well (cf. ||Xegwi-LH lhu), this cannot be a classificatory argument.

37. HAND [!Ui + Nossob + Taa]

!Ui: *|x'a (|Xam lx'a, ||Ng!ke lx'a, (Kho-M lx'a, N|uu lx'a:, ||Ku||e lx'a:, ||Kxau lx'a, ||Xegwi-Z pl. lx'a-y). 0 Curious replacement (no external parallels) in ||Xegwi: ||Xegwi-Z kyi (with assumed suppletivism between singular and plural forms) = ||Xegwi-LH qhi:. Otherwise, a very stable item.

Nossob: *|x'a/N/ (|'Auni lx'a/nl, |Haasi n=lxay). 0 |Haasi n= is probably the 1st person possessive prefix. Nasality in the coda is either the same as the ||Xegwi plural -y or the nasal class suffix in Taa languages.

Taa: *|x'a (!Xoц lx'a:, Kakia lx'a, N|u||en lx'a).

Tuu: *|x'a. 0 One of the most stable and securily reconstructed items on the list.

HEAD [!Ui + Taa] [? + Nossob]

!Ui: *]a (|Xam Ja ~ Ja: ~ Ja:, ||Ng!ke Ja ~ Ja:, (Kho-M Ja, N|uu Ja:, ||Ku||e Ja:, ||Kxau Ja:, ||Xegwi Ja:). 0 Stable and securely reconstructed. Plural form may have been *Ja-y (as in ||Xegwi, etc.).

Nossob: (a) *]a (|'Auni Ja:); (b) *xu (|'Auni x:uu, |Haasi y=xo). 0 Difficult situation. On one hand, Common Nossob *xu is clearly the same as Proto-!Ui *xu `face' (|Xam xu, etc.) and reflects a very likely semantic shift `face' ^ `head' (especially in light of external evidence from Taa which clearly shows *Ja as the original equivalent for `head'). Since Bleek records both the older form Ja: and the innovative form x:uu in the meaning `head' for |'Auni, it might be assumed that the Proto-Nossob form was still *Ja. On the other hand, it is also possible that |'Auni had simply reinstated the original word (at least in some contexts) under the very common influence of N|uu (in fact, this scenario is explicitly advocated by D. Bleek herself, see Bleek 1937: 211). Available evidence does not allow to make a definitive decision, meaning that both items have to be counted as technical synonyms on the Proto-Nossob level.

Taa: *]a- (!Xoц Jan, Kakia Ja ~ Jay, N|u||en Jay). 0 Status of the nasal coda is unclear, but

probably suffixal in light of external data.

Tuu: *]a. 0 Another highly stable Tuu root, albeit largely replaced by `face' in the Nossob group (see discussion on the |'Auni situation above).

HEAR [!Ui + Nossob] [- Taa]

* !Ui: *tu (|Xam t:u ~ t:ui, ||Ng!ke tu ~ tu:i, (Kho-M зhu ~ зhu:-wa, N|uu зu:, ||Kxau tu,

||Xegwi-Z tu). 0 This stem is attested in many different morphological variants (cf. ||Xegwi-Z past stem tu-wa, present stem tu-bi, etc.; ||Ng!ke tu: `hear' vs. tu-д: `heard' vs. tu-i `listen' in Bleek 2000: 24), but root vocalism is almost always u regardless of the morphological environment.

Nossob: l'Auni tu: ~ tu:i. ? Not attested in |Haasi. Curiously, the earliest source on l'Auni (Bleek 1929: 46) lists the forms ta:a ~ ta:a in the meaning `hear', but in Bleek 1937: 203-206, probably the same word is transcribed as tian ~ kiar (reflecting palatalized articulation of t-) with the meaning `feel'. These look like two different etyma rather than morphemic variants of each other.

Taa: *tдr (!X6ц ta:[ Kakia taa ~ tafy, N|u||en tar). ? Variable form of the !X6ц stem is ta-, but nasalization is so pervasive in all doculects that we should probably project it onto the proto-level (or even reconstruct *ta?r with an actual velar nasal coda).

Tuu: (?) *tu. ? In the !Ui branch (and possibly also in l'Auni, though the data here are limited and may also reflect !Ui influence), there is a rather clear distinction between the verbs *tu `to hear' and *ta ~ *ta `to feel', cf. |Xam ta ~ ta: ~ ta: `to feel, try, seem, beware of' with multiple text examples in Bleek 1956: 184. In Taa (or at least !X6ц), both meanings seem to have been merged in the same root *ta { Whether or not *tu and *tar are etymologically connected cannot be determined at this point, but since there is no strong evidence for grammatical Ablaut of any kind in !Ui or Taa, we should certainly treat them as two different roots, and postulate a probable lexical replacement in Taa. It should be noted that l'Auni is closer in this respect to !Ui than to Taa.

HEART [!Ui + Nossob + Taa]

!Ui: *|[,]ai (|Xam j'п:, ||Ng!ke jai ~ je, jKho-M je:-зhi, N|uu je:, ||Ku||e l'л:, ||Kxau jae ~ jai-si).

? A stable item, lost only in ||Xegwi where it is replaced by a Bantuism (||Xegwi-Z kele, ||Xegwi-LH keler). However, click efflux correspondences are unique, with a lot of variation between simple velar and glottalized articulation which cannot be fully ascribed to mistranscriptions; we are either dealing with an original root structure like *ja?i (with metathesis of glottalization) or with the unique reflexes of a rare click type (see below).

Nossob: *|[,]e (|'Auni j'e: ~ j'e:, |Haasi n=ja-je). ? Structure of the form in |Haasi, except for the usual 1st p. possessive prefix n=, is unclear (reduplication?). Note the same discrepancy in click efflux articulation (glottalic in |'Auni vs. velar in |Haasi) as in !Ui.

Taa: *|q'a- (!X6ц jq'an, pl. jq'a:, Kakia j'i:, N|u||en jar). ? Kakia j'i: is unusual here because of the vocalism, but in light of external cognates in !Ui and Nossob it may actually be seen as more archaic in that respect. Perhaps what we see are the results of morphological variation in Proto-Taa, e.g. *jq'a-i (^ Kakia j'i) vs. *jq'a-n (!X6ц jq'an, N|u||en jar).

Tuu: *|q'a(-i). ? All the forms seem related, but reconstruction of click efflux and vocalism runs into problems. There may be a correlation between the glottalized / non- glottalized effluxes seen in !Ui and Nossob, on one hand, and the glottalic uvular efflux -q'- in Taa, in which case the latter should be set up for the protoform; clear additional evidence for this correlation is, however, lacking at present. As for the root vowel, variation in Taa suggests a, but this is also inconclusive, given the prevalence of front vocalism elsewhere. Only the !X6ц-N|u||en node suggests *jq'a-n as a full stem, so this may be an innovation; on the other hand, grammatical variants *jq'a-i ~ *jq'a-n may also reflect some meaningful opposition in Proto-!Ui, with only the former variant fused and preserved in !Ui, Nossob, and Kakia.

HORN [!Ui + Nossob + Taa]

* !Ui: *||вп (|Xam jjhл: ~ jj4i:, ||Ng!ke Цаї, fKho-M Цеї, N|uu jjqhoe-si, ||Xegwi-LH Ці:). ? In most old sources, the word is hopelessly confused with tooth q.v., but N|uu and ||Xegwi data clearly show that these are two different (albeit phonetically similar) etyma. However, there is additionally a serious incongruence between N|uu //qhoe- and the rest of !Ui forms which rather go back to something like *flm; the worst problem is the vocalism, since the other languages (as well as external cognates in Nossob and Taa) show no signs of labial vowels. It is possible that modern N|uu //qhoe- is not related (although in that case, its provenance is a mystery); in any case, the reconstruction is primarily based on the correlation between |Xam and ||Xegwi as the most distant members of the !Ui branch.

Nossob: |'Auni /лп. 0 Not attested in |Haasi.

Taa: *||д- (!Xoц Цаё, Kakia /nn-sa, N|u||en /а). 0 Kakia /nn-sa = !Xoц pl. /an-sв.

Tuu: *||д-. 0 Nasalization of the vowel tentatively projected onto the proto-level due to its presence almost everywhere. Different codas most likely represent old morphology (e.g. sg. */a-i vs. pl. */a-n, as in Taa, with generalization of the sg. form in !Ui and Nossob?).

I [!Ui + Nossob + Taa]

!Ui: (|Xam q, ||Ng!ke q ~ n, jKho-M q ~ pa ~ n ~ na, N|uu q, ||Ku||e q, ||Kxau q ~ n,

||Xegwi-Z ?q ~ ?n ~ ?in ~ ?iq ~ ?m ~ ?im ~ 7Ї). 0 The primary and most common form of the root is that of a syllabic velar nasal; everything else is the result of contextual assimilations or combinations with various emphatic particles.

Nossob: (|'Auni q ~ n ~ na ~ m, |Haasi q). 0 See notes on !Ui.

Taa: (!Xoф n, Kakia n ~ na ~ q ~ qa ~ nya, N|u||en q ~ n ~ na).

Tuu: *q. 0 The complete original paradigm (including stressed forms, clitical forms, emphatic forms, assimilated variants, etc.) is hard to reconstruct, but the monoconso- nantal core of the Proto-Tuu 1st p. pronoun was undoubtedly a velar nasal, as still preserved in modern N|uu.

KILL [-]

!Ui: *|ha ~ *|hi (|Xam jв(:) ~ jhв(:) ~ ji:, |Ng!ke ja(:) ~ ji: ~ jhi, fKho-M jx'a, N|uu jha). 0 Reconstructible on the |Xam-N|uu level. Vowel gradation is similar to what is observed in several other cases, but difficult to explain based on extant data (for modern N|uu, only the a-grade form of the root is attested, suggesting analogical leveling in recent times). In ||Xegwi, the equivalent is Aiq ~ Aeq `hit, strike; kill' (Z) = A'euq `to hit' (LH); etymology is unclear, but the attested polysemy suggests semantic innovation (common semantic shift `hit' ^ `kill').

Nossob: |Haasi !au. 0 Clearly the same word as |'Auni !au `to beat; to knock down', but not `to kill', although the word `kill' is not attested in |'Auni at all, so it cannot be excluded that the meaning `kill' was expressed by the same form (*!au) in Proto-Nossob.

Taa: (?) *qa- (!Xoф qвi, var. form qa-JV, Kakia H'a:, N|u||en //wan). 0 It is unclear if the lateral click in Kakia and N|u||en is a real reflex of Proto-Taa *q- or if it represents a failed attempt to transcribe uvular articulation, but similar examples exist (e.g. !Xoф qво `pretty' = N|u||en IIxai id.; !Xoф qаla `to dig' = Kakia jkalaa id., where j is clearly a typo for //), so we tentatively list all these forms as cognate. Note that !Xoф qвi also conveys the meaning of `forceful downward movement' (e.g. qвi /'um `stamp', etc.), which may be a case of homonymy but may also reflect an original meaning similar to `hit', as in | Xegwi or Nossob languages.

Tuu: Not reconstructible. 0 |Xam-N|uu *jha ~ *jhi, with its seemingly archaic vowel gradation, looks like a good candidate, but has no recognizable cognates outside of that cluster. On the whole, the etymon `kill' looks unstable and easily replaceable by words with the semantics of `hit'.

KNEE [-]

!Ui: *]u- (|Xam Joay ~ Juay, ||Ng!ke Jц: ~ Jx, N|uu Jь:-si, ||Xegwi-Z, ||Xegwi-LH Jo-ma).

0 Same root in all three languages, but with different suffixation (f\u-laly in !Ui-N|uu, *Ju-ma in ||Xegwi; we can still see the pure root form in |Xam pl. Ju-Jua-d:e, with reduplication).

Nossob: |'Auni Uwe-Uwe (Bleek 1937); sg. Цое, pl. jjwe:-lwe (Bleek 1956). 0 Not attested in |Haasi.

Taa: *||xь (!Xoц flxь:-Jдn, Kakia jц-Jay, N|u||en Цй-Ji). 0 All Taa doculects yield a compound in the meaning `knee', where the second part is clearly head q.v. (so it may be assumed that the original meaning of the compound was rather `knee-cap'). The root is tentatively reconstructed with a voiced velar fricative efflux based on !Xoц data (Kakia and N|u||en transcriptions are unreliable).

Tuu: Not reconstructible. 0 Superficially, |'Auni joe resembles Taa *||xь, and a common origin is not excluded (assuming inadequate click transcription and different suffixation). But it is even more similar to Proto-Kalahari Khoe *joe `knee' (Vossen 1997: 457), meaning that borrowed origin is more likely here than inherited. If the Nossob form is excluded, !Ui and Taa forms may reflect the original Proto-Tuu `knee' with comparable probability.

KNOW [-]

!Ui: (?) *||xae (||Ng!ke jai, N|uu jxae). 0 This seems to be the main, if not only, equivalent of the required meaning in the N|uu cluster, cognate with |Xam jai ~ jaf `to take notice, be(come) aware of smth.' (Bleek 1956: 550). In |Xam itself, the meaning `know' is usually correlated with the root f'en(n) ~ f'л `to know, to think', further related to ||Ng!ke f'л, N|uu f'i: `to think', still further to !Xoф f'вn `to think' and ultimately to Khoe *f'an `to think (^ to know)', from which this root may have been diffused into different branches of Tuu. ||Xegwi-LH ci:, ||Xegwi-Z ci-ya remains without any etymology. In light of all this, the N|uu root with its semantically similar |Xam cognate remains the optimal, if still weak, candidate for Proto-!Ui `to know'.

Nossob: (a) |'Auni jxai ~ jx'e-ki; (b) |Haasi jьma. 0 The |'Auni form is clearly connected with N|uu jxae, but it may be a borrowing rather than a genetic cognate. The |Haasi form is just as clearly related to !Xoц (below), yet this could also be interpreted as an areal isogloss. Unclear.

Taa: !Xoц jьma (var. form ju-BV). 0 Kakia ja `to know', published only in the early source Bleek 1929: 51, is somewhat dubious.

Tuu: Not reconstructible. 0 Although there is a clear isogloss between |'Auni and !Ui, on one hand, and between |Haasi and !Xoц, on the other, both may have areal rather than genetic interpretations, and given the overall unstable nature of this concept in Tuu (cf. the Khoe root fan with its wide diffusion over Tuu territory), we should probably exclude this word from comparison for safety reasons.

LEAF [-]

!Ui: Not reconstructible. 0 `Leaf' is not a basic concept for !Ui speakers, and the word is usually borrowed (N|uu bla:r-si ^ Afrikaans, ||Xegwi-LH li=khasi-zi ^ Swazi), not attested, semantically questionable (e.g. |Xam jumm: `leaf, stick'), or completely isolated (||Ng!ke xerro: `leaves, foliage' without any external cognates).

Nossob: Not attested.

Taa: (?) *|ana (!Xoц jana, Kakia jama). 0 Formally reconstructible for Proto-Taa; however, the root is completely identical with Khoe *jana `leaf, grass' (Vossen 1997: 424), which strongly suggests a borrowed origin (which agrees well with the general unstability of this concept in Tuu). Cf. also N|u||en Jabu `leaf' (Bleek 1929: 52; incorrectly listed as SIV = |'Auni in Bleek 1956), a different root without any external etymology.

Tuu: Not reconstructible. All correlated items may be of secondary origin.

LIE [!Ui + [Nossob + Taa]]

!Ui: *ta (|Xam t:a: ~ ta: ~ t:л ~ t:e:n ~ te:rj, ||Ng!ke tia ~ kia:, fKho-D зв, N|uu зa:, ||Kxau ta ~ da ~ tn). 0 Despite some phonetic variation in the coda (most likely reflecting various morphologic variants), the most common and probably original shape of the root should be reconstructed with final *-a. The situation in ||Xegwi is unclear, with Bleek and LH contradicting themselves and neither of the variants (Bleek ja: `lie', LH ir=0ipe `I lie down') having external connections.

Nossob: |'Auni toa. 0 Cf. also tьa `to lie curled up' and transitive toa-a `to lay down, to bury'. Not attested in |Haasi.

Taa: *tu (!Xoф tы:, Kakia tu(:) ~ ta: ~ ta:, N|u||en tu:). 0 Note the variation in Kakia, unclear on its own but instructive in light of external comparison.

Tuu: *ta ~ *tu. 0 Although the dominant variant of this root in !Ui is clearly *ta and in Taa clearly *tu, scant evidence for the opposite also exists: Bleek 1929: 53 records the variant tu for ||Ng!ke, and Bleek 1956 has ta: for Kakia. The situation is slightly reminiscent of hear (except that the distribution of vocalizations is reversed), but in that case there was additional evidence to argue in favor of two original roots (`hear' vs. `feel'). Here, it is rather advisable to treat both forms as morphological variants of a single original root. The Nossob form is morphologically closer to Tuu, but could actually represent an “intermediate” variant (to-a ^ *tu-a, while !Ui languages go further and contract *tu-a ^ *ta?)

LIVER [!Ui + Taa]

!Ui: *flaN (|Xam fwar ~ fuar, ||Ng!ke fai:n, N|uu fan ~ far ~ flap, ||Kxau rara). 0 Not at

tested in ||Xegwi. Variants fluctuate between *far/a/ and *fa-ni.

Nossob: Not attested.

Taa: *flam (!Xoц fam, N|u||en flam). 0 Not attested in Kakia.

Tuu: *flaN. 0 It can hardly be doubted that all listed forms belong together, but reconstruction of the coda is somewhat problematic, given all the variety between !Ui and Taa. It is possible that the original root was simply *fa-, particularly in light of !Xoц plural forms (possessive fа:, alienated fa) which seem to drop the labial nasal as a suffix. On the other hand, nasality is such a persistent feature for all reflexes that it is hard to believe it was not, in some way or other, an intrinsic part of the root. The provisional reconstruction *faN reflects that uncertainty.

LONG [!Ui + Nossob?]

!Ui: *|'a (||Ng!ke j'a:, N|uu j'a:, ||Kxau j'a:, ||Xegwi-Z ja, ||Xegwi-LH Ja:). 0 The isogloss between N|uu, ||Kxau, and ||Xegwi clearly identifies *j'a as the optimal candidate for Proto-!Ui `long', although there are some phonetic problems -- in N|uu, the coda has a nasalized vowel, and in ||Xegwi we see unpredictable variation between ja (no glottali- sation) and Ja (nasal efflux) depending on the doculect. This may be due to a more complex protoform (e.g. something like *ja?a with different types of contractions) or to several original morphological variants (e.g. *j'a vs. *j'a-/a/N); the issue requires a better understanding of !Ui adjectival morphology. Curiously, the root is not at all attested in |Xam, where the semantic definition `tall / long / high' is instead attributed to the form !xф:-wa, pl. !xф-!xф-ka -- transparently derived from the verb !xo: `to grow up, climb up; make upright, make tall'.

Nossob: |'Auni j'a-si. 0 Glossed as `big, long, tall', but `big' is probably incorrect (the proper equivalent for this meaning in |'Auni is us/i/, see big). Not attested in |Haasi.

Taa: (a) !Xoф !'am; (b) Kakia !um. 0 Not reconstructible (the two forms are clearly not related).

Tuu: Not reconstructible. 0 The |'Auni form is clearly the same as the !Ui form, although it is impossible to determine if it is inherited or historically borrowed from N|uu.

LOUSE [!Ui + Taa]

!Ui: *0u- (|Xam Quirj, ||Ng!ke Ooin-ya, N|uu Ou-si, ||Xegwi-LH Oe-zi). 0 Root vowel re

construction is provisional (labial variant is the most common, but it could be assimilated to the labial click).

Nossob: Not attested.

Taa: !Xoц Ou:?. 0 Plural: Oa:-tк. Not attested in Kakia or N|u||en. Another synonym is !Xoф llxфni `louse'; semantic differences between the two forms are unclear.

Tuu: *0u-. 0 The lexical match between !Ui and !Xoц is transparent and allows to reliably project the root onto the Proto-Tuu level, even despite relatively scant attestation of the word in both branches (and a complete lack of attestation in Nossob). It should be noted that the alternate !Xoц synonym jjxфni bears an uncanny resemblance to the common Kalahari Khoe term for `louse', *fa'uni, which in term is somewhat irregularly connected with Khoekhoe *fa'uri id. (Vossen 1997: 462); however, presence of a lateral click in !Xoц is somewhat befuddling, since it does not allow to explain the word as a (quite common) relatively recent borrowing from Kalahari Khoe. Could this be another piece of evidence from a “pre-Tuu / pre-Khoe substrate”?



References

1. Bleek, Dorothea F. 1929. Comparative vocabularies of Bushman languages. Cambridge University Press.

2. Bleek, Dorothea F. 1937. Grammatical Notes and Texts in the |Auni Language. In: J. D. R. Jones (ed.). Bushmen of the Southern Kalahari: 195-220. Johannesburg: Witwatersrand University Press.

3. Bleek, Dorothea F. 1956. A Bushman Dictionary. American Oriental Series, vol. 41. New Haven, Connecticut: American Oriental Society.

4. Bleek, Dorothea F. 2000. The UNlke or Bushmen of Griqualand West (ed. by Tom Gьldemann). Khoisan Forum Working Paper No. 15. Kцln: University of Cologne.

5. Bleek, Wilhelm H. I., Lucy Lloyd. 1911. Specimens of Bushman Folklore. London: George Allen & Co.

6. Collins, Chris, Levi Namaseb. 2011. A Grammatical Sketch of Njuuki with Stories. Kцln: Rьdiger Kцppe Verlag.

7. Crawhall, Nigel. 2004. !Ui-Taa language shift in Gordonia and Postmasburg Districts, South Africa. PhD thesis: Faculty of Humanities, University of Cape Town.

8. Doke, Clement M. 1936. An outline of fKhomani Bushman phonetics. Bantu studies 10: 433-461.

9. Gьldemann, Tom. 2005a. “Tuu”: A New Name for the Southern Khoisan Family. In: Tom Gьldemann (ed.). Studies in Tuu (Southern Khoisan). University of Leipzig Papers on Africa, Languages and Literatures 23: 2-9. Leipzig: Institut fьr Afrikanistik, University of Leipzig.

10. Gьldemann, Tom. 2005b. Tuu as a language family. In: Tom Gьldemann (ed.). Studies in Tuu (Southern Khoisan). University of Leipzig Papers on Africa, Languages and Literatures 23: 11-30. Leipzig: Institut fьr Afrikanistik, University of Leipzig.

11. Gьldemann, Tom. 2014. The Lower Nossob varieties of Tuu: !Ui, Taa or neither? In: Tom Gьldemann, Anne-Maria Fehn (eds.). Beyond `Khoisan': Historical relations in the Kalahari Basin: 257-282. Amsterdam: John Benjamins Publishing Company.

12. Gьldemann, Tom. 2017. Casting a Wider Net over N||ng: The Older Archival Resources. Anthropological Linguistics 59(1): 71-104.

13. Gьldemann, Tom. 2018. Historical linguistics and genealogical language classification in Africa. In: Tom Gьldemann (ed.). African Languages and Linguistics: 58-444. Berlin: DeGruyter Mouton.

14. Hastings, Rachel. 2001. Evidence for the genetic unity of Southern Khoesan. Cornell working papers in linguistics 18: 225-246.

15. Kassian, Alexei, George Starostin, Anna Dybo, Vasiliy Chernov. 2010. The Swadesh wordlist. An attempt at semantic specification. Journal of Language Relationship 4: 46-89.

16. Kцhler, Oswin. 1981. Les langues khoisan. In: G. Manessy (ed.). Les langues dans le monde ancien et moderne. I: Les langues de l'Afrique subsaharienne: 455-615. Paris: Editions du CNRS.

17. Kцnig, Christa, Bernd Heine. 2001. The !Xun of Ekoka. A demographic and linguistic report. Khoisan Forum Working Paper No. 17. Kцln: University of Cologne.

18. Lanham, Leonard, D. P. Hallowes. 1956a. Linguistic relationships and contacts expressed in the vocabulary of Eastern Bushman. African studies (Johannesburg) 15(1): 45-48.

19. Lanham, Leonard, D. P. Hallowes. 1956b. An outline of the structure of eastern Bushman. African studies (Johannesburg) 15(3): 97-118.

20. Maingard, Louis F. 1937. The fKhomani dialect of Bushman: its morphology and other characteristics. In: J. D. Rheinallt Jones, Clement M. Doke (eds.). Bushmen of the southern Kalahari: 237-275. Johannesburg: Witwaters- rand University Press.

21. Meinhof, Carl. 1929. Versuch einer grammatischen Skizze einer Buschmannsprache. Zeitschrift fьr EingeborenenSprachen 19: 161-188.

22. Miller, Amanda, Johanna Brugman, Bonny Sands, Levi Namaseb, Mats Exter, Chris Collins. 2009. Differences in airstream and posterior place of articulation among N|uu clicks. Journal of the International Phonetic Association 39(2): 129-161.

23. Naumann, Christfried. 2014. Towards a genealogical classification of Taa dialects. In: Tom Gьldemann, Anne- Maria Fehn (eds.). Beyond `Khoisan': Historical relations in the Kalahari Basin: 283-302. Amsterdam: John Benjamins Publishing Company.

24. Sands, Bonny. 2014. Adoption, Maintenance and Loss of Click Contrasts. Talk presented at: Sound Change in Interacting Human Systems, 3rd Biennial Workshop on Sound Change, University of California, Berkeley, May 2831, 2014.

25. Sands, Bonny, Amanda Miller, Johanna Brugman. 2007. The Lexicon in Language Attrition: The Case of N|uu. In: Doris L. Payne, Jaime Pena (eds.). Selected Proceedings of the 37th Annual Conference on African Linguistics: 5565. Somerville: Cascadilla Proceedings Project.

26. Starostin, George. 2008. From Modern Khoisan Languages to Proto-Khoisan: the Value of Intermediate Reconstructions. In: Ilya Smirnov (ed.). Aspekty komparativistiki III [Aspects of comparative linguistics Ш]. Orientalia et Classica, Vol. XIX: 337-470. Moscow: Russian State University for the Humanities.

27. Starostin, George. 2015. Annotated Swadesh wordlists for the !Wi group (Peripheral Khoisan family). Ms.

28. Starostin, George. 2016. From wordlists to proto-wordlists: reconstruction as “optimal selection”. Faits de langues 47/1: 177-200.

29. Starostin, George (ed.). 2011-2021. The Global Lexicostatistical Database. Moscow: Higher School of Economics / Santa Fe, NM: Santa Fe Institute.

30. Starostin, Georgij. 2013. Jazyki Afriki: opyt postrojenija leksikostatisticheskoj klassifikacii. Tom I: Metodologija. Kojsanskije jazyki. Moscow: Jazyki slav'anskoj kul'tury.

31. Story, Robert. 1999. K'ujha:si manuscript (MS collections of the Kijhazi dialect of Bushman, 1937). Ed. by Anthony Traill. Khoisan Forum Working Paper No. 13. Kцln: University of Cologne.

32. Traill, Anthony. 1994. A IXФ5 dictionary. Quellen zur Khoisan-Forschung/Research in Khoisan studies, Bd 9. Kцln: Rьdiger Kцppe Verlag.

33. Traill, Anthony. 1995. Interpreting |Xam phonology: the need for typological cleansing. In: Anthony Traill, Rainer Vossen, Megan Biesele (eds.). The complete linguist: papers in memory of Patrick J. Dickens (Afrikanische Sprachen und Kulturen): 509-523. Kцln: Rьdiger Kцppe Verlag.

34. Traill, Anthony. 2018. A Trilingual !XФ5 dictionary: IXфф - English - Setswana. Quellen zur Khoisan-Forschung/Research in Khoisan studies, Bd 37. Kцln: Rьdiger Kцppe Verlag.

35. Vossen, Rainer. 1997. Die Khoe-Sprachen: Ein Beitrag zur Erforschung der Sprachgeschichte Afrikas. Kцln: Rьdiger Kцppe Verlag.

36. Westphal, Ernst O. J. 1962. On classifying Bushman and Hottentot languages. African Language Studies 3: 30-48.

37. Westphal, Ernst O. J. 1971. The Click Languages of Southern and Eastern Africa. In: Jack Berry, Thomas A. Sebeok (eds.). Linguistics in Sub-Saharan Africa: 367-420. The Hague / Paris: Mouton.

38. Ziervogel, Dirk. 1955. Notes on the language of the eastern Transvaal Bushmen. In: E. F. Potgieter (ed.). The disappearing Bushmen of Lake Chrissie. Hidding-Currie publications of the University of South Africa, no. 1: 34-64. Pretoria: J L van Schaik.

Размещено на Allbest.ru