Hostname: page-component-cd9895bd7-hc48f Total loading time: 0 Render date: 2024-12-26T07:45:46.746Z Has data issue: false hasContentIssue false
  • English
  • Français

POTS, WORDS AND THE BANTU PROBLEM: ON LEXICAL RECONSTRUCTION AND EARLY AFRICAN HISTORY*

Published online by Cambridge University Press:  05 September 2007

KOEN BOSTOEN
KOEN BOSTOEN
Affiliation:
Royal Museum for Central Africa Tervuren, Université libre de Bruxelles
Get access Rights & Permissions [Opens in a new window]

Abstract

Historical-comparative linguistics has played a key role in the reconstruction of early history in Africa. Regarding the ‘Bantu Problem’ in particular, linguistic research, particularly language classification, has oriented historical study and been a guiding principle for both historians and archaeologists. Some historians have also embraced the comparison of cultural vocabularies as a core method for reconstructing African history. This paper evaluates the merits and limits of this latter methodology by analysing Bantu pottery vocabulary. Challenging earlier interpretations, it argues that speakers of Proto-Bantu inherited the craft of pot-making from their Benue-Congo-speaking ancestors who introduced this technology into the Grassfields region. This ‘Proto-Bantu ceramic tradition’ was the result of a long, local development, but spread quite rapidly into Atlantic Central Africa, and possibly as far as Southern Angola and northern Namibia. The people who brought Early Iron Age (EIA) ceramics to southwestern Africa were not the first Bantu-speakers in this area nor did they introduce the technology of pot-making.

Keywords

ArchaeologyBantu originslinguistics
Type
Research Article
Information
The Journal of African History , Volume 48 , Issue 2 , July 2007 , pp. 173 - 199
Copyright
Copyright © Cambridge University Press 2007

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 Estimates of the number of Bantu present-day languages vary between 440 (M. Guthrie, Comparative Bantu: An Introduction to the Comparative Linguistics and Prehistory of the Bantu Languages [4 vols.] [London, 1967–71]) and 680 (M. Mann and D. Dalby, A Thesaurus of African Languages. A Classified and Annotated Inventory of the Spoken Languages of Africa [London, 1987]), depending on how one distinguishes a language from a dialect.

2 Niger-Congo is the biggest of Africa's four language families (see J. H. Greenberg, The Languages of Africa [The Hague, 1963]). Its internal classification is still a matter of ongoing research and debate. For a recent proposal, see K. Williamson and R. Blench, ‘Niger-Congo’, in B. Heine and D. Nurse (eds.), African Languages: an Introduction (Cambridge, 2000), 11–42.

3 Vansina, J., ‘Bantu in the crystal ball, I’, History in Africa, 6 (1979), 287–333CrossRefGoogle Scholar.

4 See, for example, Oliver, R., ‘The problem of the Bantu expansion’, Journal of African History, 7 (1966), 361–76CrossRefGoogle Scholar; Vansina, J., ‘New linguistic evidence and the Bantu expansion’, Journal of African History, 36 (1995), 173–95CrossRefGoogle Scholar; M. K. H. Eggert, ‘The Bantu problem and African archaeology’, in A. B. Stahl (ed.), African Archaeology: A Critical Introduction (London, 2005), 301–26.

5 Eggert, ‘The Bantu problem’.

6 See, for example, T. C. Schadeberg, ‘Historical linguistics’, in D. Nurse and G. Philippson (eds.), The Bantu Languages (London, 2003), 143–63.

7 For a critical introduction to lexicostatistics and other historical-linguistic methods and their use for the reconstruction of African history, see Nurse, D., ‘The contributions of linguistics to the study of history in Africa’, Journal of African History, 38 (1997), 359–91CrossRefGoogle Scholar.

8 Eggert, ‘The Bantu problem’, 321.

9 Möhlig, W. J. G., ‘Sprachgeschichte, Kulturgeschichte und Archäologie. Die Kongruenz der Forschungsergebnisse als methodologisches Problem’, Paideuma, 35 (1989), 189–96Google Scholar.

10 For a discussion of the origins of the method, see Y. Malkiel, Etymology (Cambridge, 1993). For more information on its application for the use of African history, see Nurse, ‘The contributions’, or the introductory chapter of K. Bostoen, Des mots et des pots en bantou. Une approche linguistique de l'histoire de la céramique en Afrique (Frankfurt am Main, 2005).

11 More complex historical scenarios should of course be reckoned, since both sources of lexical resemblance interact easily. A word borrowed at a certain point in time, for instance, may get integrated into the language's lexicon and be transmitted subsequently as part of the inherited vocabulary. Words may also pass from one language to the other when a community shifts to a new language, but maintains part of its original vocabulary.

12 R. M. W. Dixon, The Rise and Fall of Languages (Cambridge, 1997), 45.

13 See, for example, Ehret, C., ‘Cattle-keeping and milking in eastern and southern African history: the linguistic evidence’, Journal of African History, 8 (1967), 1–17CrossRefGoogle Scholar; D. L. Schoenbrun, A Green Place, a Good Place: Agrarian Change, Gender and Social Identity in the Great Lakes Region to the 15th Century (Oxford, 1998); J. Vansina, How Societies Are Born: Governance in West Central Africa Before 1600 (Charlottesville, 2004).

14 Klieman, K., ‘Comments on Christopher Ehret, “Bantu history: re-envisioning the evidence of language”’, International Journal of African Historical Studies, 34 (2001), 48–51Google Scholar.

15 Vansina, How Societies, 45.

16 Guthrie, Comparative Bantu.

17 The term omumi is an agent noun derived from the verb *-ma, which is attested in reduplicated form in Nyaneka: -mama ‘modelar com barro (ou outro material aplicável)’ (A. J. Da Silva, Dicionário Português–Nhaneca [Lisbon, 1966], 360). Even if Proto-Bantu (PB) *b before *U may regularly become zero in Nyaneka, e.g. PB *-bbì, (spider) >e-uvi, this verb cannot possibly be derived from *-b mb-, since *U and *mb remain respectively u and mb in Nyaneka, e.g. PB *-btò (seed) >om-buto; *-jímb- (to sing) >-imba.

18 In Bostoen, Des mots, 11–15, or Bostoen, K., ‘What comparative Bantu pottery vocabulary may tell us about early human settlement in the Inner Congo basin’, Afrique & Histoire, 5 (2006)Google Scholar, I discuss in more detail why this simple example is a symptom of a deeper problem. See also R. Klein-Arendt, ‘Pre-colonial non-Bantu influence on Savannah Bantu vocabulary. The case of the Chaga (E62) iron terminology’, in K. Bostoen and J. Maniacky (eds.), Studies in African Comparative Linguistics with Special Focus on Bantu and Mande (Tervuren, 2005), 147–64. For that matter, not only linguists question the methodological validity of the way certain historians approach language data; there also exist serious debates amongst historians themselves. See for instance Vansina, J., ‘Linguistic evidence and historical reconstruction’, Journal of African History, 40 (1999), 469–73Google Scholar, in which he heavily criticizes C. Ehret, An African Classical Age: Eastern and Southern Africa in World History, 1000 bc to ad 400 (Charlottesville, 1998).

19 See for instance Y. Bastin et al., Reconstructions lexicales bantoues 3/Bantu Lexical Reconstructions 3 (Tervuren, 2003), online database (http://linguistics.africamuseum.be/BLR3.html), or Guthrie, Comparative Bantu. Given the vastness of their enterprise, these studies can only give us a rough idea of both the actual distribution and the semantics of lexical items. That is why their valuable data need to be complemented in order to be appropriate for reliable historical reconstruction.

20 A comparative series is a set of terms occurring in different languages with similar phonological form and a related meaning.

21 For a good introduction to the comparative method, see Nurse, ‘The contributions’, 361–3; for more extensive information, see for instance T. Crowley, An Introduction to Historical Linguistics (Auckland, 1992).

22 W. K. Barnett and J. W. Hoopes (eds.), The Emergence of Pottery: Technology and Innovation in Ancient Societies (Washington, 1995).

23 N. Barley, Smashing Pots. Feats of Clay from Africa (London, 1994); Gosselain, O. P., ‘In pots we trust. The processing of clay and symbols in Sub-Saharan Africa’ Journal of Material Culture, 4 (1999), 205–30CrossRefGoogle Scholar; Jacobson-Widding, A., ‘Pits, pots and snakes. An anthropological approach to ancient African symbols’, Nordic Journal of African Studies, 1 (1992), 5–27Google Scholar.

24 Gosselain, O. P., ‘Materializing identities: an African perspective’, Journal of Archaeological Method and Theory, 7 (2000), 190CrossRefGoogle Scholar.

25 O. P. Gosselain, Poteries du Cameroun méridional. Styles techniques et rapports à l'identité (Paris, 2002), 7 (my translation).

26 See, for example, T. N. Huffman and R. K. Herbert, ‘New perspectives on Eastern Bantu’, Azania, 29–30 (1994–5), 27–36.

27 Bostoen, Des mots.

28 Vocabulary was collected not only in linguistic, but also in ethnographic, literature. See, for instance, the online available repertory of the Centre de Recherches Archéologiques (CReA) of the Université libre de Bruxelles, set up by O. Gosselain (June 2002), including more than 800 bibliographical sources referring to some 650 populations of sub-Saharan Africa: www.ulb.ac.be/philo/crea/pdf/sources_poterie_contempo.pdf.

29 J. Hagendorens, Dictionnaire français–otetela (Leuven, 1984) (my translation).

30 D. Senasson, ‘Approche ethno-archéométrique des céramiques actuelles de la région de Mubuga (Burundi)’ (unpublished Master's thesis, Paris, 1993); Bostoen, K. and Harushimana, G., ‘Parole et savoir-faire populaires: conversations à propos de la poterie des Twa au Burundi’, LPCA Text Archives, 4 (2003), 1–39Google Scholar (www2.fmg.uva.nl/lpca/textarchives/vol4/manwerika_sinabajije.html) (my translation).

31 This does not mean that no other verbs with the same meaning are attested. In some North East Coast Bantu languages, such as Digo, Bondei, Ruguru, Kami and Kutu, for example, a verb related to the Swahili verb kufinyanga (to make pottery) occurs. However, such verbs are few and most of them are limited to one particular language where they often coexist with one of the major verbs.

32 G. Tessmann, Die Pangwe. Völkerkundliche Monographie eines Westafrikanischen Negerstammes (Berlin, 1913); Césard, E., ‘Le Muhaya (L'Afrique Orientale)’, Anthropos, 31 (1936), 97–114Google Scholar; S. Galley, Dictionnaire fang–français et français–fang, suivi d'une grammaire fang (Neufchâtel, 1964); S. Kaji, A Haya Vocabulary (Tokyo, 2000).

33 These examples were taken, in order of appearance, with my translation, from D. E. C. Stirke and A. W. Thomas, A Comparative Vocabulary of Sikololo–Silui–Simbunda (London, 1916); V. Guerreiro, Rudimentos de língua maconde (Lourenço Marques, 1963); E. Van Avermaet and B. Mbuya, Dictionnaire kiluba–français (Tervuren, 1954), The White Fathers' Bemba–English Dictionary (London, 1954) and Guia de conversação olunyaneka (Huilla, 1908).

34 These examples were taken, in order of appearance, with my translation, from K. Felberg, Nyakyusa–English–Swahili and English–Nyakyusa Dictionary (Dar es Salaam, 1996); R. Moser, Aspekte der Kulturgeschichte der Ngoni in der Mkoa wa Ruvuma, Tanzania (Vienna, 1983); J. Busse, Die Sprache der Nyiha in Ostafrika (Berlin, 1960); Y. Turner, Tumbuka–Tonga English Dictionary (Blantyre, 1952); A. Charmoille, Dictionnaire kifipa–français (Rome, 1902). The Lambya example stems from my own field notes, as does the meaning ‘to make pottery’ in Tumbuka, which is not mentioned by Turner, Tumbuka.

35 L. F. Dos Santos, Dicionário Português–Chope e Chope–Português (Lourenço Marques, 1949); K. Kavutirwaki, Lexique nande–français et français–nande (Kinshasa, 1978); for Copi pottery, see Lawton, A. C., ‘Bantu pottery of southern Africa’, Annals of the South-African Museum, 49 (1967), 1–440Google Scholar; for Nande pottery, see Bergmans, L., ‘Kruiken en potten’, Ontwakend Afrika, 58 (1955), 21–30Google Scholar.

36 A. Coupez et al., Dictionnaire kinyarwanda–kinyarwanda et kinyarwanda–français (Tervuren, 2005) (my translation).

37 A similar semantic evolution from a general to a more restricted technical sense was observed for the common Bantu verbs for forging, e.g. *-t d- and *-pònd-, whose basic meanings are ‘to hammer, to beat’ and ‘to pound, to beat’, respectively. See de Maret, P. and Nsuka, F., ‘History of Bantu metallurgy: some linguistic aspects’, History in Africa, 4 (1977), 43–65CrossRefGoogle Scholar; Klein-Arendt, R., ‘The iron crafts of the Swahili from the perspective of historical semantics’, Afrikanistische Arbeitspapiere, 64, Swahili Forum, 7 (2000), 153–204Google Scholar.

38 As pointed out earlier, the verbs *-gìng- and *-màt- are confined to the Inner Congo Basin and the Lake Corridor area, respectively. They will not be discussed further in this paper. The verb *-gìng- plays a prominent role in a lengthy article I wrote on the pottery vocabulary of the Inner Congo Basin. See Bostoen, ‘Comparative Bantu pottery vocabulary’.

39 Guthrie, Comparative Bantu. In Guthrie's terms, PB-X approximately equates to Proto-Bantu.

40 See Vansina, ‘New linguistic evidence’.

41 For more details on the internal Bantu classification, see for instance D. Nurse and G. Philippson, ‘Towards a historical classification of the Bantu languages’, in D. Nurse and G. Philippson (eds.), The Bantu Languages (London, 2003), 164–81; or Vansina, ‘New linguistic evidence’, which is based on the later-published Y. Bastin et al., Continuity and Divergence in the Bantu Languages: Perspectives from a Lexicostatistic Study (Tervuren, 1999). The subgroups and their designations referred to in this article follow the aforementioned Vansina classification.

42 The verb -me- (to make pottery) occurs in the Budu language (N. Asangama, ‘Le budu: langue bantu du nord-est du Zaïre, esquisse phonologique et grammaticale’ [2 vols.] [Ph.D. dissertation, Paris, 1983]), whilst kumaja (to make pottery) is found in the Lengola language (Stappers, L., ‘Esquisse de la langue lengola’, Africana Linguistica, 5 [1971], 255–307CrossRefGoogle Scholar).

43 K. Williamson and K. Shimizu, Benue-Congo Comparative Wordlist (2 vols.) (Ibandan, 1968); R. Harguindéguy, Premiers éléments pour un dictionnaire adja–français (Azové, 1969).

44 Other lexical evidence pointing in the same direction is the reconstruction of two Proto-Bantu nouns referring to types of pottery, i.e. *-bìgá (pot), and *-j ng (cooking pot). As regards the potter's clay, the noun *-mà is the most likely candidate for reconstruction to Proto-Bantu. However, *-b mbà (potter's clay) cannot be excluded.

45 Roset, J. P., ‘Les plus vieilles céramiques du Sahara’, Archéologia, 138 (1983), 43–50Google Scholar.

46 Mohammed-Ali, A. S. and Khabir, A. R. M., ‘The wavy line and the dotted wavy line pottery in the prehistory of the Central Nile and the Sahara-Sahel belt’, African Archaeological Review, 20 (2003), 25–58CrossRefGoogle Scholar.

47 D. W. Phillipson, African Archaeology (3rd ed., Cambridge, 2005), 151–60; McIntosh, S. K. and McIntosh, R. J., ‘Recent archaeological research and dates from West Africa’, Journal of African History, 27 (1986), 413–42CrossRefGoogle Scholar.

48 Huysecom, E.et al., ‘Ounjougou (Mali): a history of Holocene settlement at the southern edge of the Sahara’, Antiquity, 78 (2004), 579–93CrossRefGoogle Scholar; MacDonald, K., ‘Kourounkorokale revisited: the Pays Mande and the West African microlithic technocomplex’, African Archaeological Review, 14 (1997), 143–60CrossRefGoogle Scholar.

49 Breunig, P.et al., ‘New research on the Holocene settlement and environment of the Chad Basin in Nigeria’, African Archaeological Review, 13 (1996), 111–45CrossRefGoogle Scholar.

50 Chenorkian, R., ‘Ivory Coast prehistory: recent developments’, African Archaeological Review, 1 (1983), 127–42CrossRefGoogle Scholar.

51 York, R. N., ‘Excavations at Dutsen Kongba, Plateau State, Nigeria’, West African Journal of Archaeology, 8 (1978), 139–63Google Scholar.

52 Stahl, A. B., ‘Reinvestigation of Kintampo 6 rockshelter, Ghana: implications for the nature of change’, African Archaeological Review, 3 (1985), 117–50CrossRefGoogle Scholar.

53 Smith, A. B., ‘Radiocarbon dates from Bosumpra Cave, Abetifi, Ghana’, Proceedings of the Prehistoric Society, 41 (1975), 179–82CrossRefGoogle Scholar.

54 Shaw, T., ‘Holocene adaptations in West Africa: the Late Stone Age’, Early Man News, 3–4 (1978–9), 51–82Google Scholar.

55 Lavachery, P., ‘The Holocene archaeological sequence of Shum Laka rock shelter (Grassfields, Cameroon)’, African Archaeological Review, 18 (2001), 240CrossRefGoogle Scholar.

56 de Maret, P., ‘Pits, pots and the far west streams’, Azania, 29–30 (1994–5), 318–23Google Scholar.

57 See Lavachery, ‘The Holocene archaeological sequence’, where the author discusses the possible appearance of pottery around 7000 bp. The evidence is still weak however, since only four potsherds were found.

58 F. A. Hassan, ‘Abrupt Holocene climatic events in Africa’, in G. Pwiti and R. Soper (eds.), Aspects of African Archaeology (Harare, 1996), 83–9; Lavachery, ‘The Holocene archaeological sequence’.

59 Maley, J. and Brenac, P., ‘Vegetation dynamics, palaeoenvironments and climatic changes in the forests of West Cameroon during the last 28,000 years bp’, Review of Palaeobotany & Palynology, 99 (1998), 157–87CrossRefGoogle Scholar.

60 Since the appearance of Neolithic attributes, such as pottery and ground stone tools, does not coincide with the advent of food production in much of Africa, certain archaeologists avoid the term ‘Neolithic’. I adopt here the designation ‘Stone to Metal Age’, preferred by de Maret and Lavachery, knowing that many African archaeologists still use the old term ‘Neolithic’ as a synonym of SMA. Others, like Stahl, have settled on ‘ceramic Late Stone Age’. See Stahl (ed.), African Archaeology.

61 Lavachery, ‘The Holocene archaeological sequence’.

62 K. Williamson and R. Blench, ‘Niger-Congo’; D. Nurse and G. Philippson, ‘Introduction’, in Nurse and Philippson (eds.), The Bantu Languages, 1–12.

63 K. Williamson, ‘Benue-Congo overview’, in J. Bendor-Samuel (ed.), The Niger-Congo Languages. A Classification and Description of Africa's Largest Language Family (Lanham, 1989), 247–75.

64 de Maret, P., ‘New survey of archaeological research and dates for West-Central and North-Central Africa’, Journal of African History, 23 (1982), 1–15CrossRefGoogle Scholar.

65 A. Assoko Ndong, ‘Synthèse des données archéologiques récentes sur le peuplement à l'Holocène de la réserve de faune de la Lopé, Gabon’, L'Anthropologie, 106 (2002), 135–58; Clist, B., ‘Archaeology in Gabon, 1986–1988’, African Archaeological Review, 8 (1989), 59–85CrossRefGoogle Scholar; Clist, B., ‘Le site d'Okala, Province de l'Estuaire, Gabon et son importance pour la compréhension de la sédentarisation en Afrique Centrale’, Comptes-rendus de l'Académie des Sciences de Paris, 325 (1997), 151–6Google Scholar; B. Clist, ‘Des premiers villages aux premiers européens autour de l'estuaire du Gabon. Quatre millénaires d'interactions entre l'homme et son milieu’ (Ph.D. dissertation, Université libre de Bruxelles, 2005).

66 de Maret, P., ‘The Ngovo group: an industry with polished stone tools and pottery in Lower-Zaïre’, African Archaeological Review, 4 (1986), 103–33CrossRefGoogle Scholar.

67 M. K. H. Eggert, ‘The Central African rainforest: historical speculation and archaeological facts’, in I. Glover (ed.), The Humid Tropics (London, 1992), 1–24.

68 R. de Bayle de Hermens, Recherches préhistoriques en République Centrafricaine (Paris, 1975).

69 De Maret, ‘Pits, pots’.

70 See, for example, N. David, ‘Early Bantu expansion in the context of Central African prehistory: 4000–1 bc’, in L. Bouquiaux (ed.), L'expansion bantoue. Actes du Colloque International du CNRS, Viviers (France), 4–16 avril 1977, vol. iii (Paris, 1980), 609–47; P. de Maret, ‘Le contexte archéologique de l'expansion bantu en Afrique centrale’, in T. Obenga, Actes du Colloque international ‘Les peuples bantu. Migrations, expansion et identité culturelle’ Libreville 1–6 avril 1985, vol. i (Libreville, 1989), 118–38; Vansina, J., ‘Western Bantu expansion’, Journal of African History, 25 (1984), 129–45CrossRefGoogle Scholar.

71 P. Lavachery, ‘De la pierre au métal. Archéologie des dépôts holocènes de l'abri de Shum Laka (Cameroun)’ (3 vols.) (Ph.D. dissertation, Université libre de Bruxelles, 1998).

72 With regard to dating, this is several millennia later than the period Kairn Klieman claimed, i.e. 5000-4000 bc (see K. Klieman, ‘The Pymies Were Our Compass’: Bantu and Batwa in West Central Africa, Early Times to c. 1900 c.e. [Portsmouth, 2003], 35–65). Her very early Bantu expansion hypothesis is mainly founded on glottochronological calculations, a lexicostatistics-based method commonly rejected by linguists (see, e.g., Nurse, ‘The contributions’, 366), and a number of early C14 dates for ceramics from the La Sablière site in Gabon, which are problematic (for the most critical appraisal of those dates, see Clist, Des premiers villages).

73 B. Clist, ‘Synthèse régionale du Néolithique’, in R. Lanfranchi and B. Clist, Aux origines de l'Afrique centrale (Libreville, 1991), 181–3; Clist, B. and Lanfranchi, R., ‘Contribution à l’étude de la sédentarisation en République Populaire d'Angola’, Leba, 7 (1992), 245–67Google Scholar.

74 Dos Santos, J. R. and Everdosa, C. M. N., ‘A estação arqueológica de Benfica, Luanda’, Revista da Faculdade de Ciencias da Universidade Luanda, 5 (1970), 33–51Google Scholar; R. Lanfranchi and B. Clist, ‘Néolithique: Angola’, in Lanfranchi and Clist, Aux origines, 179–80.

75 Phillipson, D. W., ‘An archaeological reconsideration of Bantu expansion’, Muntu, 2 (1985), 69–84Google Scholar.

76 B. Clist and R. Lanfranchi, ‘Age du Fer Ancien: Angola’, in Lanfranchi and Clist, Aux origines, 219–23.

77 Vogelsang, R.et al., ‘Holocene human occupation and vegetation history in Northern Namibia’, Die Erde, 133 (2002), 113–32Google Scholar.

78 J. Richter, ‘Archaeology along the Kavango river / Namibia’, Southern African Field Archaeology, 11–12 (2002–3), 78–104.

79 J. Kinahan, Pastoral Nomads of the Central Namib Desert (Windhoek, 1991).

80 Smith, A. B. and Jacobson, L., ‘Excavations at Geduld and the appearance of early domestic stock in Namibia’, South African Archaeological Bulletin, 50 (1995), 3–14CrossRefGoogle Scholar.

81 See, for instance, Richter, ‘Archaeology along the Kavango’, 81.

82 Vansina, ‘New linguistic evidence’; Denbow, J., ‘Congo to Kalahari: data and hypotheses about the political economy of the western stream of the Early Iron Age’, African Archaeological Review, 8 (1990), 139–76CrossRefGoogle Scholar.

83 J. Vansina, ‘A slow revolution: farming in subequatorial Africa’, Azania, 29–30 (1994–5), 15–26.

84 See, for example, Smith and Jacobson, ‘Excavations at Geduld’, 9; R. Vogelsang, ‘Migration oder Diffusion? Frühe Viehhaltung im Kaokoland’, in M. Bollig, E. Brunotte and T. Becker (eds.), Interdisziplinäre Perspektiven zu Kultur- und Landschaftswandel im ariden und semiariden Nordwest Namibia (Cologne, 2002), 141.

85 Robinson, K. R., ‘Bambata ware: its position in the Rhodesian Iron Age in the light of recent research’, South African Archaeological Bulletin, 21 (1966), 81–5CrossRefGoogle Scholar.

86 Denbow, ‘Congo to Kalahari’.

87 Mitchell, P. and Whitelaw, G., ‘The archaeology of southernmost Africa from c. 2000 bp to the early 1800s: a review of recent research’, Journal of African History, 46 (2005), 209–41CrossRefGoogle Scholar. See also P. Mitchell, The Archaeology of Southern Africa (Cambridge, 2002) (particularly ch. 9, ‘Taking stock: the introduction and impact of pastoralism’).

88 Denbow, J., ‘A new look at the later prehistory of the Kalahari’, Journal of African History, 27 (1986), 3–28CrossRefGoogle Scholar.

89 Walker, N., ‘The significance of an early date for pottery and sheep in Zimbabwe’, South African Archaeological Bulletin, 28 (1983), 88–92CrossRefGoogle Scholar.

90 Phillipson, D. W., ‘The first South African pastoralists and the Early Iron Age’, Nsi, 6 (1989), 127–34Google Scholar; Denbow, ‘A new look’.

91 Mitchell and Whitelaw, ‘The archaeology of southernmost Africa’, 216.

92 An analogous scenario has been suggested for the verb *-t d- (to forge), being a specialization of the meaning ‘to hammer, to beat’, and several other metallurgy-related terms, e.g. *-tádè (stone) adopting the meaning ‘iron ore’. See de Maret and Nsuka, ‘History of Bantu metallurgy’, or P. de Maret and G. Thiry, ‘How old is the Iron Age in Central Africa?’, in P. R. Schmidt (ed.), The Culture and Technology of African Iron Production (Gainesville, 1996), 29–39.

93 This author tends to see East-Bantu as a primary branch of Proto-Bantu, as it emerges from several internal Bantu classifications (see, for instance, P. Piron, Classification interne du groupe bantoïde [2 vols.] [Munich and Newcastle, 1997]; or Bastin et al., Continuity). There is no space here to set out the reasons at length, but unlike Ehret, for instance, I do not see East-Bantu as a sub-sub-branch of Savannah-Bantu. See Ehret, ‘Subclassifying Bantu’; Ehret, C., ‘Bantu expansions: re-envisioning a central problem of early African History’, International Journal of African Historical Studies, 34 (2001), 5–41CrossRefGoogle Scholar. Moreover, like Nurse and Philippson, I am rather hesitant about the validity of a Savannah branch of Bantu and consider the linguistic traits shared by these languages, as opposed to the Forest Bantu languages, as the result of contact ‘across the Savannah communities, once they had formed a more or less continuous chain from southwest to northeast’ (Nurse and Philippson, ‘Towards’, 180). As I will explain further on, this Savannah continuum might account for the presence of *-b mb- in the West-Bantu languages.

94 Designations adopted from Vansina, ‘New linguistic evidence’.

95 This is the assumption I favoured in Bostoen, Des mots.

96 Cf. Nurse and Philippson, ‘Towards’, 173–6.

97 Vansina, How Societies, 45.

98 Phillipson, ‘An archaeological reconsideration’.

99 Phillipson, African Archaeology, 249.

100 Clist, B., ‘A critical reappraisal of the chronological framework of the early Urewe Iron Age industry’, Muntu, 6 (1987), 35–62Google Scholar.

101 For an overview of eastern and southern African sites linked to the Chifumbaze complex, see Phillipson, African Archaeology, 249–65.

102 Ibid. 264.

103 Phillipson, ‘An archaeological reconsideration’, 76–8.

104 Huffman, T. N., ‘Ceramics, settlements and Late Iron Age migrations’, African Archaeological Review, 7 (1989), 155–82CrossRefGoogle Scholar.

105 Bastin, Y., ‘Essai de classification de quatre-vingts langues bantoues par la statistique grammaticale’, Africana Linguistica, 9 (1983), 11–108CrossRefGoogle Scholar; for more details on this East-Coastal-Bantu, see also K. Bostoen and C. Grégoire, ‘La question bantoue: bilan et perspectives’, Mémoires de la Société de Linguistique de Paris (forthcoming).

106 Bostoen, Des mots, 406–13, 427–32. The lexical innovation in question is °-k&adango (frying pan).

107 See, for instance, Nurse and Philippson, ‘Towards’, 175.

108 This concerns a bowl Van Grunderbeek regards as an Urewe innovation indicating a change in subsistence economy. Similar bowls were found in other ‘eastern stream’ sites. See Grunderbeek, Van, ‘Essai d’étude typologique de céramique urewe de la région des collines au Burundi et Rwanda’, Azania, 13 (1988), 11–55CrossRefGoogle Scholar; Soper, R. C., ‘Early Iron Age pottery types from East Africa: comparative analysis’, Azania, 6 (1971), 39–52CrossRefGoogle Scholar.

109 Bostoen, Des mots, 406–13.

110 Phillipson, D. W., ‘Archaeology and Bantu linguistics’, World Archaeology, 8 (1976), 65–82CrossRefGoogle Scholar.

111 Phillipson, ‘An archaeological reconsideration’, 78.

112 P. de Maret, Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaïre: Sanga et Katongo, 1974 (Tervuren, 1985) and Fouilles archéologiques dans la vallée du Haut-Lualaba, Zaïre: Kamilamba, Kikulu et Malemba-Nkulu, 1975 (Tervuren, 1992).

113 Phillipson, ‘An archaeological reconsideration’, 78.

114 Phillipson, ‘The first South African pastoralists’, 131.

115 Katanekwa, N., ‘Upper Zambezi Iron Age research project phase ii: a preliminary report’, Archaeologia Zambiana, 20 (1981), 12–14Google Scholar.

116 Lanfranchi and Clist, ‘Néolithique: Angola’, 179.

117 Clist and Lanfranchi, ‘Age du Fer Ancien: Angola’, 220.

118 Soper, R. C., ‘A general review of the Early Iron Age of the southern half of Africa’, Azania, 6 (1971), 5–38CrossRefGoogle Scholar.

119 Van Grunderbeek, M.-C., ‘Essai de délimitation chronologique de l'Age du Fer Ancien au Burundi, au Rwanda et dans la région des Grands Lacs’, Azania, 27 (1992), 53–80CrossRefGoogle Scholar.

120 Phillipson, African Archaeology, 251.