1 Bastin, Y., Coupez, A. and de Halleux, B., ‘Classification lexicostatlstique des langues bantoues (214 relevés)’, Bulletin de l'Académic Royale des Sciences d'Outre-Mer, XXVII (1983), 173–199.Google Scholar

2 Compare annexe 9, with annexe 3 in ibid.; Heine, B., Hoff, H. and Vossen, R., ‘Neuere Ergebnisse zur Territorialgeschichte der Bantu’, Zur Sprachgeschichte und Ethnohistorie in Afrika (Berlin, 1977), 57–70, used 147 languages.Google Scholar

3 Hinnebusch, T. J., Nurse, D. and Mould, M., ‘Studies in the classification of eastern Bantu languages’ (Beiheft), Sprache und Geschichte in Afrika III (Hamburg)Google Scholar is expected. For now besides Bastin et al., see Nurse, D., ‘Bantu expansion into East Africa: linguistic evidence’, in The Archaeological and Linguistic Reconstruction of African History, eds. Ehret, C. and Posnansky, M. (Berkeley, 1982), 199–238Google Scholar; Nurse, D. and Philippson, G., ‘Historical implications of the language map of East Africa’, in L'expansion bantoue, ed. Bouquiaux, L. (Paris, 1980), III, 685–714.Google Scholar

4 Bastin, et al. , ‘Classification’, 181–183Google Scholar; cf. Heine, et al. , ‘Neuere Ergebnisse’, 58–59.Google Scholar The Heine group did not use the Swadesh list of 100 words but a similar list compiled by B. Heine.

5 The ‘group average method’ was used; cf. Bastin, et al. , ‘Classification’, 175–176.Google Scholar This differs and improves on Heine, et al. , ‘Neuere Ergebnisse’, 59.Google Scholar

6 Cf. the differentiation of Romance languages from Latin or of the West Germanic languages on the continent. No single moment in time and no single place are involved, but periods and areas.

7 Bastin, Y., Coupez, A. and De Halleux, B., ‘Statistiques lexicales et grammaticales pour la classification historique des langues bantoues’, Bulletin de l'Académie Royale des Sciences d'Outre-Mer, XXIII, iii (1979), 375–387 (68 languages).Google Scholar

8 Oliver, Roaland, ‘The problem of the Bantu expansion’, J. Afr. Hist., VII (1966), 361–376CrossRefGoogle Scholar; ‘Papers on Comparative Bantu’, African language Studies, XIV (1973)Google ScholarHeine, B., ‘Zur genetischen Gliederung der Bantu-Sprachen’, Afrika und Übersee, LVI (1973), 164–185.Google Scholar

9 The Tervuren group has over 300 languages now, including the last remaining major gaps for western Bantu. For the ‘as yet unclassified’ area of Fig. 2 lists are available since 1981. The final classifications should come from this team and, for eastern Bantu, from the Hinnebusch/Nurse/Mould team. Since the procedures of both are not identical the correspondences that are shaping up will be especially significant.

10 Bennett, P. B. and Sterk, J. P., ‘South Central Niger-Congo: a reclassification’, Studies in African Linguistics, VIII, iii (1977), 241–273.Google Scholar

11 Bastin, et al. , ‘Classification’, 176–177.Google Scholar

12 Ehret, C., ‘Linguistic inferences in early Bantu history’, in Ehret, and Posnansky, as cited in n. 3, 59–61Google Scholar, whose argument remains strictly geographical and does not take wider Bantu into account.

13 de Maret, P., ‘Preliminary report’, Nyame Akuma, XVII (1980), 10–12Google Scholar; Warmer, J. P., Asombang, R., ‘Archaeological research in the Bamenda grassfield’, Nyame Akuma, XXI (1982), 3–4Google Scholar; de Maret, P., ‘New survey of archaeological research and dates for West-Central and North-Central Africa’, XXIII (1982), 1–3.Google Scholar

14 See n. 13 and Shaw, T., Nigeria, Its Archaeology and Early History (London, 1978), 45–50Google Scholar; Andah, B. W. and Anozie, F. N., ‘Preliminary report on the prehistoric site of Afikpo’, West African Journal of Archaeology, in press.Google Scholar

15 De, Maret, ‘New Survey’, 2–3Google Scholar; personal communication, 30 Aug. 1982.

16 Data are found in Guthrie, M., Comparative Bantu, 4 vols. (Farnborough, 1967–1971), especially vol. II, indexes A and B, 116–156 and supplement, pp. 176–180.Google Scholar But distributions and reasoning for each item must be carefully checked. No faith at all should be placed in Polome, E., ‘The reconstruction of Proto-Bantu culture from the lexicon’, in L'expansion bantoue, ed. Bouquiaux, , III, 779–791.Google Scholar

17 Aymemi, A., Los Bubis en Fernando Poo (Madrid, 1942)Google Scholar; Tessmann, G., Die Bubi auf Fernando Poo (Darmstadt, 1923).Google Scholar

18 A full study of Proto-Western Bantu culture has not yet been done. For this the distribution and variability of meaning for each term involved must be more precisely plotted than it usually is and the reflexes must be regular.

19 This notation refers to Guthrie, , Comparative Bantu, III, IV.Google Scholar

20 For a recent summary cf. Huffman, T. N., ‘Archaeology and ethnohistory of the African Iron Age’, Annual Review of Anthropology, XI (1982), 133–138.CrossRefGoogle Scholar Current research is listed in Nyame Akuma.

21 As on Fig. 2 for Ubangian, Gbaya and Central Sudanic enclaves in Bantu territory or Western Bantu enclaves on the Mbomu River and in Bahr el Ghazal. At a less remote level and between Bantu language clusters examples of Cameroun (the splitting of the group labelled A 80) are known. The location of Bobangi also betrays later movement.

22 The nomenclature adopted is new and may not become standard. Bastin et al., ‘Classification’, do not provide a nomenclature while Heine et al., Neuere Ergenbisse, does not quite fit. Labelling by letters or numbers is confusing because the standard referential classification (not genetic!) using letters and numbers varies in different classifications.

23 Bastin et al., ‘Classification’, does not include Bubi yet. But it is known (unpublished data, Tervuren) that there are several Bubi languages. We follow tentatively Heine, et al. , Neuere Ergebnisse, 61.Google Scholar

24 This means that the tenacious notion in Gabonese historiography that Bantu speakers only entered the country in the centuries after A.D.1200 is totally erroneous. It flows from the speculations of Avelot, R., ‘Recherches sur l'histoire des migrations dans le bassin de l'Ogooué et la region littorale adjacente’, Bulletin de gèographic historique et descriptive, XX (1905), 357–412.Google Scholar

25 Farine, B., Sites prehistoriques gabonais (Libreville, 1963), fig. 31, 56Google Scholar; Pommeret, Y., Civilisations prehistoriques au Gabon (Libreville, 1966)Google Scholar; Farine, B., ‘Le néolithique de Moanda’, Bulletin de la société préhistorique et protohistorique gabonaise', V (1966), 79–94Google Scholar; de Maret, P., ‘Gabon’, in The Archaeology of Central Africa, ed. Van Noten, F. (Graz, 1982), 62–63.Google Scholar

26 Ibid., 57–9. The association between Uelian artefacts, slag and tuyeres at one site may be fortuitous. Not all archaeologists can accept Van Noten's conclusion that the Uelian is entirely an Iron Age phenomenon. The file remains open.

27 de Maret, P., ‘Lower Zaire, Congo, Angola’, in Archaeology of Central Africa, ed. Van Noten, , 63–65.Google Scholar

28 Van Noten, , op. cit. p. 65.Google Scholar

29 Flux was therefore much greater than it would be later when demographic pressures from already settled nuclei led to ‘outflow’. Such later movements are a consequence of stabilization. The settlement of the Americas by Europeans shows similar dynamics.

30 Phillipson, D. W., The Later Prehistory of Eastern and Southern Africa (London, 1977), 141Google Scholar, dates Chondwe between the fourth and sixth centuries A.D. For Shaba, cf. De Maret, P., ‘The south-east and the copperbelt’, in Archaeology of Central Africa, ed. Van Noten, , 86–88; cf. also Huffman, as cited in n. 20, 136.Google Scholar

31 De Wet, J. M. J., ‘Domestication of African cereals’, African Economic History, III (1977), esp. maps, 19, 21, 24Google Scholar; Harlan, J. R. and Stemler, A., ‘The races of sorghum in Africa’, in Origins of African Plant Domestication, eds. Harlan, J. R., De Wet, J. M. J. and Stemler, S. (The Hague, 1976), 465–478.CrossRefGoogle Scholar

32 Guthrie, , Comparative Bantu, II, 177, then pp. 27–28Google Scholar; and check each CS where ‘X’ is given in vols.

33 CS 293½–294. CS 260 may be the most likely source with a skewed reflex constituting the innovation but see also CS 288–290 and ps 73.

34 Bastin, et al. , ‘Statistiques’, 380.Google Scholar

35 Ibid. 381.

36 Guthrie, Comparative Bantu, I, 97–110, esp. 100 and topograms 1–2.Google Scholar

37 Simmonds, N. W., Bananas (London, 1959)Google Scholar; McMaster, D. N., ‘Speculations on the coming of the banana to Uganda’, Journal of Tropical Geography, XVI (1962), 57–69Google Scholar; Barrau, J. as cited by Bouquiaux, J. and Thomas, J., ‘Le Peuplement Oubanguien’, in L'expansion bantoue, ed. Bouquiaux, , III, 816.Google Scholar

38 Barrau as cited in ibid., fn 9: line AAB-ABB from Asia via the Near East. Such letters designate groups of genetic varieties of banana. All AAB varieties are usually known as ‘plantain’, but not all ‘plantain’ are AAB, and ‘plantain’ is not a valid genetic category.

39 Blakney, C. P., ‘On banana and iron: linguistic footsteps in African history’ (M.A. thesis, Hartford Seminary, 1983), 54–94Google Scholar, and my own distribution map.

40 Wainwright, G. A., ‘The coming of the banana to Uganda’, Uganda Journal, XVI (1952), 145–147Google Scholar; Munro-Hay, S., ‘The Foreign Trade of the Aksumite port of Adulis’, Azania, XVII (1982), 107–125.CrossRefGoogle Scholar Axumites being familiar with the ensete, a relative of the banana, could very well have brought the plant to their home from India where, according to Pliny, soldiers of Alexander the Great found it well established.

41 David, N., ‘Prehistory and historical linguistics in Central Africa: points of contact’, in Ehret, and Posnansky, as cited in n. 3, 81, 89.Google Scholar For the oil-palm see Saxon, D. E., ‘Linguistic evidence for the eastward spread of Ubangian peoples’Google Scholar, ibid. 76, and CS 140 (western Bantu).

42 Bouquiaux, and Thomas, in L'Expansion ban tone, ed. Bouquiaux, , 807–824.Google Scholar

43 Calvocoressi, D. and David, N., ‘A new survey of radiocarbon and thermoluminescence dates for West Africa’, J. Afr. Hist, XX (1979), 10.Google Scholar

44 Van Grunderbeek, M. C., Roche, E. and Doutrelepont, H., Le premier age du fer au Rwanda et au Burundi: Archèologie et environnement (Brussels, 1983), 48–50.Google Scholar

45 de Maret, P., ‘The Iron Age in the west and the south’, in Archaeology in Central Africa, ed. Van Noten, , 77–79Google Scholar; idem, in litt., 30 Aug. 1982.

46 See fn. 30.

47 Nsuka-Nkutsi, F. and De Maret, P., ‘Etude comparative de quelques termes métallurgiques dans les langues bantoues’, in L'expansion bantoue, ed. Bouquiaux, , III, 731–742Google Scholar; De Maret, P. and Nsuka, F., ‘History of Bantu metallurgy: some linguistic aspects’, History in Africa, IV (1977), 43–65CrossRefGoogle Scholar; Guthrie, , Comparative Bantu, I, 132, 138–140Google Scholar, for a distribution of some of these terms.

48 Seitz, S., Die Zentralafrikanischen Wildbeuterkulturen (Wiesbaden, 1977), 26–48Google Scholar; P. 29 gives the best map to date on the occurrence of ‘pygmies’ in which most of the groupings are indicated. At present the San do not range further north than southern Angola and south-westernmost Zambia. Formerly, according to writers and traditions, ‘pygmy’ hunters extended further south, being found on the middle Kasai and in the kingdom of Kongo (in the north), whilst traditions about San-like hunters are found further north in central Angola and in northern Zambia.

49 Hiernaux, J., The People of Africa (New York, 1974), 113–125.Google Scholar

50 Lee, R. B., The !Kung San. Men, Women and Work in a Foraging Society (Cambridge, 1979), 309–322Google Scholar is the fullest discussion of birth spacing and nomadic life. His findings seem to be true for all described hunter-gatherers in Central Africa, although the information is often fragmentary.

51 Hiernaux, , People of Africa, 118–125Google Scholar, esp. 122 for biological information; Hulstaert, G. H., ‘Nordkongo-Der Zentrale Teil’, in Baumann, H., ed., Die Völker Afrikas und Ihre Traditionellen Kulturen (Wiesbaden, 1975), I, 741Google Scholar, for the absorption through marriage of Jofe ‘pygmies’ by Boyela ‘negroes’ and similar cases: Hart, J. A., ‘Nomadic hunters and village cultivators’ (M. A. thesis, Michigan State University, 1979), 34–35, 71–75Google Scholar, for bakbala relationships between Mbuti and Pakombe, whereby a number of Mbuti women were married by Bira or Pakombe in each generation. For Joset, P. E. (‘Cinquante annees d'histoire du territoire de Beni (1889–1939)’, MS. in author's possession, 74–75)Google Scholar, the Pakombe were thought to have lost their original language to adopt ‘pygmy’ and Bira speech, whose women they married. Bira is spoken both by some Mbuti and by farmers. Cases of biological ‘pygmy’ turning to farming have been reported from the Mayi Ndombe area, while the Kele of Gabon were in the nineteenth century abandoning their farms to become professional hunters, some fusing with ‘pygmies’, Deschamps, H., Traditions orales et archives au Gabon (Paris 1962), 128–133.Google Scholar

52 Hart, , ‘Nomadichunters’, 34, 114Google Scholar; Vansina, J., Degeschiedenis van de Kuba (Tervuren, 1963), 108, 124, 195–197, 203–204, 305.Google Scholar

53 Thomas, J. M. C., ‘Emprunt ou parenté?’, in Bahuchet, S., ed., Pygmées de Centrafrique (Paris, 1939), 141–169Google Scholar; Letouzey, R., Contribution de la botanique au problème d'uneevenutelle langue pygmee (Paris, 1975).Google Scholar