Hostname: page-component-586b7cd67f-l7hp2 Total loading time: 0 Render date: 2024-11-24T16:02:20.906Z Has data issue: false hasContentIssue false
  • English
  • Français

The Prehistorical Implications of Guthrie's Comparative Bantu. Part II: Interpretation of Cultural Vocabulary1

Published online by Cambridge University Press:  22 January 2009

David Dalby
David Dalby
Affiliation:
School of Oriental and African Studies and International African Institute, London
Get access Rights & Permissions [Opens in a new window]

Extract

The second part of this article opens with a general caveat about the use and misuse of linguistic evidence. Guthrie, on the other hand, presented his data in a way which leaves subsequent scholars free to arrive at their own interpretations. The complexity of his data makes it difficult to achieve a necessary overview, however, and a stylized grid is therefore proposed, enabling the relative distribution of individual items of Common Bantu vocabulary to be plotted and compared. This is particularly important in the case of cultural vocabulary, where geographical distribution is normally related to levels of historical origin or diffusion. Groups of these vocabulary grids, based on Guthrie's corpus of data and his referential zones, are assembled and presented for the following semantic areas: fishing and watercraft, metal-working, pottery, livestock and cultivation (including cereals). Much of the remainder of the article is devoted to an exposition of the way in which these grids may be interpreted historically, including the need to distinguish—as far as possible—between likely cognates and likely loan-words. Attention is drawn to the possibilities (i) that Bantu languages may have begun to diverge substantially in linguistic terms before they began to move widely apart from their secondary and tertiary nucleus south of the forest, and (ii) that there may have been a differential layering, or overlapping waves, of Bantu expansion in the east.

Type
Research Article
Information
The Journal of African History , Volume 17 , Issue 1 , January 1976 , pp. 1 - 27
Copyright
Copyright © Cambridge University Press 1976

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

2 Guthrie, Malcolm, Comparative Bantu: an introduction to the comparative linguistics and prehistory of the Bantu languages (Farnborough, 1967–1971), 4 volsGoogle Scholar. Italicized references are to the chapter/paragraph number of Comparative Bantu, part I.Google Scholar

3 He also devoted a number of stylized maps or ‘topograms’ to specific items of cultural and other vocabulary within his Common Bantu corpus: see p. 3 below.

4 ‘The starred forms of Common Bantu’, part I (vol. 2).Google Scholar

5 ‘English index to probable Proto-Bantu items’, part I (vol. 2).Google Scholar

6 See part I, 61. 31–45, for an explanation of these.

7 Proto-Bantu starred forms, representing historical reconstructions, are cited by Guthrie in an upper-case italic transcription, to distinguish them from Common Bantu starred forms, cited in lower case bold, symbolizing a set of related words in a number of modern Bantu languages: see p. 484 of this article (Part I). The initial reference numbers, in the examples cited here, relate to the sequence of Guthrie's Comparative Series in part II (vols. 3 and 4), where the full data are set out for each item of vocabulary. The numbers following a nominal stem (e.g. 3/4) indicate the pair of singular/plural nominal classes to which the item belongs.

8 Although the absence of that particular item in PB-X does not necessarily imply that there was no term in PB-X with that meaning.

9 See part I (vol. 1) for his topograms on ‘horn’, ‘house’, ‘cloth’, ‘to forge’, ‘to draw (water)’, ‘skin’, ‘buffalo’, ‘chicken’, ‘iron’, ‘bee’, ‘honey’, ‘one’, ‘three’ and ‘five’; see also Guthrie, M., ‘Contributions from Comparative Bantu to the study of African prehistory’, in Dalby, D., ed., Language and History in Africa (London, 1970), 20–49Google Scholar, for topograms on ‘skin’, ‘chest’, ‘to fish with a line’, ‘fish-hook’, ‘canoe’, ‘paddle’, ‘iron’, ‘to forge’, ‘bananas’, ‘peanuts’, ‘sugar-cane’, ‘forest’.

10 Each of his zones was intended as ‘a set of groups which have a certain geographical contiguity and which display a number of common linguistic features as well’: see Dalby, D., ‘Reflections on the classification of African languages’, African Language Studies, xi (SOAS, 1970), 147–71, esp. 152–4.Google Scholar

11 For the approximate location of the zones, see Map 7 below, and for details of the divided zones, see p. 9 below.

12 Although regrettably, for apparent practical reasons, not tonal detail (except in his reconstructed starred forms): see part I, 12. 44 and 42. 51 ff.

13 See p. 22–3 below.

14 Or even necessarily, for reasons of space, all the relevant terminology included by Guthrie within his Common Bantu corpus: thus ‘livestock’ here includes only ‘goat’, ‘sheep’ and ‘cattle’ (with ‘buffalo’), while ‘cultivation’ covers only cereals among the various crops recorded by Guthrie. See part I (vol. 2), pp. 176 ff.Google Scholar, for a classified semantic index (by subject) of all items in the Common Bantu corpus.

15 Historians will readily appreciate the need to maintain a sharp distinction between actual items of vocabulary, recorded from individual languages, and hypothetical ‘common’ or ‘proto’ forms, reconstructed to cover individual items of vocabulary in a group of related or supposedly related languages. Professional linguists maintain a strict rule that reconstructed forms should never be cited without a preceding star (to distinguish them clearly from actual items of vocabulary). Historians will also recognize that a starred form has no value as prehistorical evidence unless its creator has placed on record the full set of data, and comparative methodology, by which he arrived at that starred form.

16 The term ‘skewed’ is not ideal, as it prejudges the historical relationship between variant forms.

17 The same device is used where two related Comparative Series, with the same shape but different meanings, are combined on the same vocabulary grid. Thus *-dub- ‘to fish (with net/basket)’ is not recorded with this meaning in Zones A, B and H, where it occurs only with the general (and presumably earlier) meaning ‘to dip’. On the other hand, o-gònό ‘fish-trap’ is recorded in Zone S with this meaning but also with the meaning ‘wicker bird trap’.

18 Indicated by ‘ps’ before the reference number.

19 See part I, 71, and cf. the osculant Comparative Series 871/905/907, with the common meaning ‘bellows’ but with slightly different shapes (combined below on one grid) or the osculant Comparative Series 119/120/134 ‘pot’ (shown on separate grids).

20 For a more detailed map of Guthrie's Zones, see his Classification of Bantu Languages (London, 1948)Google Scholar, but as amended for North-west Bantu in The Bantu Languages of Western Equatorial Africa (London, 1953).Google Scholar

21 The four zones have been subdivided in each case largely on criteria of geographical distribution. It should be noted, however, that for the purposes of the comparative Bantu studies currently being undertaken at the Musée Royal de l'Afrique Tropicale at Tervuren, the groups D40–60 have been separated from the remainder of Guthrie's Zone D and combined with his groups E10–20, to form a new Zone J. Although this is a useful modification, to have followed it here would have been too confusing in terms of the original Guthrian classification on which Comparative Bantu is based. Note also that groups S10, S50 and S60 together formed Guthrie's original Zone, T (Classification, 70–73)Google Scholar, which he afterwards combined with his Zone S.

22 An item occurring in A, Á, C and Ć, for example, would have been excluded from the Common Bantu corpus since it would not have met the criterion of occurrence in three of Guthrie's original zones.

23 See Classification, 9–19.

24 A discussion of this question will be included in the subsequent article ‘Bantu: the West African connection’, to be published in this Journal.

25 I.e. Tiv and the Ekoid cluster.

26 E.g. ‘fish-hook’ from ‘to fish with hook and line’, or ‘iron’ from ‘stone’.

27 See examples under ‘knife’ and ‘pot’. For the purpose of the following vocabulary grids, note also that the Bantu noun-class numbers have been omitted in the case of nominal stems, and that the stroke has also been omitted from the second letter of the four subdivided zones.

28 For the use of Proto-Bantu 1, 2 and 3, see p. 498 of this article (Part I).

29 Cf. the modern semantic development of the words ‘car’ and ‘engine’ in English, for example.

30 In part I, 73, 75, he refers the reader for these details to his Topogram 22 ‘buffalo’, in the same volume, although information on *-gόmbό (etc.) = ‘buffalo’ is not in fact included there.

31 Which could explain a loss of the term but scarcely its phonetic ‘skewing’.

32 This term for ‘cattle’, among others, has been discussed by Christopher Ehret: see p. 24 below.

33 For Guthrie's slightly different view on the prehistory of this term, see part II, commentary on Comparative Series 905.

34 Also ‘person’ or ‘place’, with membership of the relevant noun-classes.

35 For a West African example of the highly complex semantic range covered by such environmental terms, see Dalby, David, ‘The concept of settlement in the West African savannah’, in Oliver, Paul, ed., Shelter, Sign and Symbol (London, 1975), 197–205.Google Scholar

36 There may be a link with *-pi ‘to become burnt’, but see Guthrie's note, part II, CS 1506/1507.

37 Suspicion about this item would also be aroused by the variability of the noun classes involved: four alternative pairs are cited among the nine ‘unskewed’ examples in the comparative series.

38 See Ehret, C., ‘Cattle-keeping and milking in Eastern and Southern African history: the linguistic evidence’, J. Afr. Hist., viii, 1 (1967), 1–17CrossRefGoogle Scholar; ‘Sheep and Central Sudanic Peoples in Southern Africa’, J. Afr. Hist., IX, 2 (1968), 213–22Google Scholar; ‘Patterns of Bantu and Central Sudanic Settlement in Central and Southern Africa’, Transafrican Journal of History, III, 1/2 (1973), 1–71.Google Scholar

39 See ‘Cattle-keeping…’, 1 (note 2).Google Scholar

40 See ‘Sheep and Central Sudanic…’, 220.Google Scholar

41 See ‘Patterns of Bantu and Central Sudanic…’, 1.Google Scholar

42 See ‘Sheep and Central Sudanic…’, 220.Google Scholar

43 Note also the comparable distribution of the synonymous term *-méémé and perhaps also *-páηgá.

44 See Hair, P. E. H., ‘The contribution of early linguistic material to the history of West Africa’, in Dalby, Language and History… 50–63, esp. note 19.Google Scholar

45 See Phillipson, D. W., ‘The chronology of the Iron Age in Bantu Africa’, J. Afr. Hist., xvi, 2 (1975).Google Scholar

46 Beginning with Brusciotto's grammar of Kongo, published in Rome in 1659.

47 The only other coherent group of terms, with similar distribution, is a set of four relating to fishing: *-dób- ‘to fish with hook and line’, *-dóbό ‘fish-hook’, *-gόnό ‘fish-trap’ and *-d**b- ‘to fish with net/basket’.