² Pierre Bayle, Dictionnaire historique et critique (3rd rev. ed., Rotterdam, 1720), ii, 1291, s.v. Goulu, note F.

³ Nevertheless, I agree emphatically with the point made by Cordell, Dennis and Gregory, Joel, ‘Historical demography and demographic history in Africa: theoretical and methodological considerations’, Canadian Journal of African Studies, xiv (1980), 391–3Google Scholar, that historical demography intrinsically, if not always in practice, is concerned with much more than the numbers.

⁴ For instance, I have felt obliged to ignore the question of possible changes in the population structure of West African societies resulting from the slave trade. Such studies can be regarded as unexceptionable to the extent that they do not require allegiance to any particular range of population figures but only to the idea that, if left to their own devices, West African populations would have shown structures similar to those of other pre-industrial societies (and perhaps to certain African societies during the colonial period, when labour migration must have replicated some of the conditions during the slave trade).

⁵ To my chagrin it has not proved possible here to attempt more than a series of observations which claim to do no more than emphasize the tenuous nature of the various beliefs about almost every aspect of the slave trade. I have not been able to confront many particular arguments and I have generally ignored (although not forgotten) the work being done for other parts of Africa beyond the extent to which it might throw light on the issues considered here for West Africa. For this I am grateful to be able to refer to Paul Lovejoy's recent and most useful summing up of the current status questionis, Lovejoy, P., ‘The volume of the Atlantic slave trade: a synthesis’, J. Afr. Hist, xxiii (1982), 473–501.CrossRefGoogle Scholar Among other things, Lovejoy captures nicely the diversity of opinion.

⁶ Campbell, George, Philosophy of Rhetoric (London, 1776), 169.Google Scholar

⁷ Most of these points, of course, were raised in Curtin, Philip D., The Atlantic Slave Trade: A Census (Madison, 1969), passim.Google Scholar

⁸ As Lovejoy, ‘Atlantic slave trade’, 494–5, points out, as parts of b became parts of a during the past fifteen years, it transpired that overestimates and underestimates tended to cancel each other out, so that a has actually grown less than the estimated shrinkage of b. Abstractly of course this cannot happen and this emphasizes the need to keep both the abstract and the actual firmly in mind in historical enquiry.

⁹ On the basis of which (to cite but one example) L.-M. Diop has expressed the belief that ‘une forte proportion’ of the Atlantic slave trade operated through clandestine traders. See her ‘Le sous-peuplement de l'Afrique noire’, Bulletin de l'lnstitut Fondamental de l'Afrique Noire, sér. B, xl (1978), 737.

¹⁰ As simply the numerical difference between the number who boarded ship in West Africa and the number who debarked in the New World, this figure would not take into account at what point during the voyage slaves died (or tended to die) or whether they died from shipboard circumstances or from conditions acquired while still in Africa. This is discussed by, among others, Miller, Joseph C., ‘Mortality in the Atlantic slave trade: statistical evidence on causality’, J. Interdisciplinary History, xi (1983/1984), 385–423.Google Scholar

¹¹ Curtin, , Atlantic Slave Trade, 275–86Google Scholar; Postma, Johannes, ‘Mortality in the Dutch slave trade’ in Gemery, Henry A. and Hogendorn, Jan S., eds. The Uncommon Market: Essays in the Economic History of the Atlantic Slave Trade. (New York, 1979), 253–5Google Scholar; Manning, Patrick, ‘The enslavement of Africans: a demographic model’, Canadian Journal of African Studies, xv (1981), 509Google Scholar; Morgan, W. B. and Pugh, J. C., West Africa (London, 1969),424Google Scholar; Thornton, John K., ‘The demographic effect of the slave trade on western Africa, 1500 to 1850’ in African Historical Demography II (Edinburgh, 1981),709Google Scholar; Gemery, and Hogendorn, , ‘The economic costs of West African participation in the Atlantic slave trade: a preliminary sampling for the eighteenth century’, in Uncommon Market, 155Google Scholar; Person, Yves, ‘La population de l'Afrique noire durant les XVIIIe et XIXe siecles’, Culture et Sociétée, iii (1980), 35Google Scholar; Rodney, Walter, How Europe Underdeveloped Africa (Washington, 1972), 96.Google Scholar

¹² Postma, ‘Mortality’, 242–3, does distinguish and suggests a range of 3–5 per cent for deaths while waiting in Dutch trading forts.

¹³ E.g. Manning, ‘Enslavement’, 506–16.

¹⁴ Wrigley, C. C., ‘Historicism in Africa: slavery and state formation’, African Affairs, lxx (1971), 115.Google Scholar

¹⁵ Inikori, J. E., ‘Under-population in nineteenth-century West Africa’ in African Historical Demography ii, 301–2.Google Scholar

¹⁶ Morgan, and Pugh, , West Africa, 434.Google Scholar

¹⁷ Curtin, Philip D., Economic Change in Precolonial Africa: Senegambia in the Era of the Slave Trade (Madison, 1975), i, 182.Google Scholar

¹⁸ Rodney, , How Europe Underdeveloped Africa, 96.Google Scholar

¹⁹ Thornton, ‘Demographic effect’, 709; Manning, ‘Enslavement’, 509. Cf. Gemery and Hogendorn, ‘Economic costs’, 155–6.

²⁰ Mauny, Raymond, Tableau géographique de l'Ouest africain au Moyen Age (Dakar, 1961), 377–9.Google Scholar

²¹ Idem. Les siècles obscurs (Paris, [1971]), 240–1, 279.

²² Austen, Ralph A., ‘The trans-Saharan slave trade: a tentative census’ in Uncommon Market, 23–76.Google Scholar

²³ Ibid. 66–7.

²⁴ Ibid. 42–3.

²⁵ Lewicki, Tadeusz, ‘Arab trade in Negro slaves up to the end of the XVIth century’ as cited in ‘African Studies symposium in Cracow’, Africana Bulletin, vi (1967), 111.Google Scholar

²⁶ Hopkins, A. G., An Economic History of West Africa (London, 1973),83Google Scholar; Austen, ‘Trans-Saharan slave trade’, 30 n. My intuition agrees with Austen's but in both cases the intuition stems from the suspected magnitude of the entire slave trade rather than from any relationship between Lewicki's claim and any evidence for it.

²⁷ Austen, ‘Trans-Saharan slave trade’, 66.

²⁸ With the single exception of estimates of a that might be lower than the available figures.

²⁹ Johnson, Samuel, Shakespeare [1765] in Johnson, Samuel, Works of Samuel Johnson (New Haven, 1958–85), vii, 109.Google Scholar

³⁰ A recent survey, particularly useful for its bibliography, is Robert Dewar, E., ‘Environmental productivity, population regulation, and carrying capacity’, American Anthropologist, lxxxvi (1984), 601–14.CrossRefGoogle Scholar

³¹ An example is Inikori's misinterpretation of Thurstan Shaw's argument; see Inikori, , ‘Introduction’ to Inikori, J. E. ed., Forced Migration: The Impact of the Export Slave Trade on African Societies, (New York, 1982), 29–30Google Scholar. Occasionally, too, when an area's population outstrips its ability to feed itself, it relies on imports, as does Japan today.

³² However, Caldwell, J. C., ‘Two comments on Manning’, Canadian Journal of African Studies, xvi (1982), 127Google Scholar, does argue for the ultimate efficacy of the carrying capacity approach.

³³ Buikstra, Jane E. and Konigsberg, Lyle W., ‘Paleodemography: critiques and controversies’, American Anthropologist, lxxxvii (1985), 316–33Google Scholar, provides some flavour of the recent debate.

³⁴ For comments on one such effort see Henige, David, ‘If pigs could fly: Timucuan population and Native American historical demography’, Journal of Interdisciplinary History, xvi (1985/1986) 701–20.Google Scholar

³⁵ Diop, ‘Sous-peuplement’, 780–803; Kea, Ray A., Settlement, Trade, and Polities in the Seventeenth-Century Gold Coast (Baltimore, 1982), 137–46, 379–82Google Scholar; Johnson, Marion, ‘Census, map, and guesstimate: the past population of the Accra region’ in African Historical Demography i (Edinburgh, 1977), 272–93Google Scholar. Cf. Austen, ‘Trans-Saharan slave trade’, 51–8.

³⁶ Gardiner, Alan H., The Kadesh inscription of Ramesses II (Oxford, 1960), 13,41Google Scholar; Kadry, Ahmed, ‘Some comments on the Qadesh battle’, in Bulletin du centenaire [Supplement of the Bulletin de l'lnstitut français d'archéologie orientate, 81] (Cairo, 1981), 47–55.Google Scholar

³⁷ Is it not odd, for instance, that estimates of Fante army sizes decreased as that state became more powerful during the early eighteenth century? See Kea, Settlements, 138.

³⁸ Niane, D. T., ‘Mali and the second Mandingo expansion’, in UNESCO General History of Africa IV (Berkeley, 1984), 156Google Scholar, 206n. Niane's discussion embodies the common belief that there is a necessarily correlative relation between dense populations and imperial status so that demonstrating one is seen to enjoin proving the other as well.

³⁹ A particular ambitious effort along these lines is Thornton, ‘Demographic effect’.

⁴⁰ Caldwell, J. C., ‘Major questions in African demographic history’ in African Historical Demography i, 7Google Scholar; Willcox, Walter F., International Migrations, II, Interpretations (New York, 1931), 53–4, 640–2Google Scholar; Carr-Saunders, A. M., World Population: Past Growth and Present Trends (London, 1964), 34–5Google Scholar; Clark, Colin, Population Growth and Land Use (New York, 1967), 64–6Google Scholar. All derive from Giovanni Baptista Riccioli, Geographiae Hydrographiae Reformatae (Venice, 1672), 679–80.

⁴¹ Caldwell, ‘Major questions’, 18.

⁴² Manning, ‘Enslavement’, 499–526. Further studies by Manning are forthcoming.

⁴³ As does Caldwell, ‘ Two comments’, 128, whereas Inikori, ‘ Two comments’, 135, considers the premise ‘ quite close to historical reality’. If nothing else, the idea that most captured females were assimilated by other African societies seems to compromise the notion that the Atlantic slave trade was a radical departure from existing mores unless we accept the idea that integrating extra-societal females occurred to African societies only on seeing them pass on their way to the coast. This is possible, of course, but perhaps more probable (again, if the premise is accepted) is that slave raiding was an age-old feature of African societies, at least with females as the objective.

⁴⁴ Manning, P., ‘Contours of slavery and social change in Africa’, American Historical Review, lxxxviii (1983), 850.Google Scholar

⁴⁵ Inikori, ‘Introduction’, 29–31; idem. ‘Under-population’, 296–301.

⁴⁶ Inikori, ‘Introduction’, 32; idem. ‘Under-population’, 303, however, prefers to draw the comparison with blacks in the antebellum American South.

⁴⁷ This contretemps is discussed in Henige, ‘ If pigs could fly’.

⁴⁸ Diop, ‘Sous-peuplement’, 730–3; idem. ‘Méthode et calculs approximatifs pour la construction d'une courbe répresentative de l'évolution de la population de l'Afrique noire, du milieu du XVIe siècle au milieu du XXe’, in African Historical Demography II, 147–9. While I do not suggest that the work of Diop and Inikori necessarily typifies the work being done on West African historical demography, I chose to present their arguments precisely because of their extravagance, in order better to underscore how difficult it is to refute any given line of reasoning in the state of the evidence.

⁴⁹ Diop, ‘Sous-peuplement’, 734–5

⁵⁰ Ibid. 738.

⁵¹ Ibid. 744. Perhaps Diop overplays her hand when she argues in support of her thesis that seventeenth-century maps of Africa featured no ‘lacunae’, or refers to ‘the agriculture practices and mentality of black people’, or rhapsodizes about the nutritional superiority of yams, sorghum, and millet. For more, see now Population XL (1985), 855–9.

⁵² Inikori, ‘Under-population’, 303–4. For all of Africa Inikori (‘Introduction’, 33) has recently suggested a figure of 112 million for potential losses.

⁵³ Inikori, ‘Under-population’, 298–300; Diop, ‘Sous-peuplement’, 726–8, believes that droughts, famines, and diseases were probably less common in pre-colonial Africa than in contemporary Europe.

⁵⁴ Miller, J. C., ‘The significance of drought, disease, and famine in the agriculturally marginal zones of west-central Africa’, J. Afr. Hist, xxii (1982), 17–61.CrossRefGoogle Scholar

⁵⁵ Nicholson, Sharon E., ‘Saharan climates in historic times’, in Williams, Martin A. J. and Faure, Hugues, eds, The Sahara and the Nile (Rotterdam, 1980), 178–85Google Scholar; Brooks, George E., Western Africa to ca. 1860 A.D.: A Provisional Historical Schema Based on Climate Periods (Bloomington, 1985), 154–96.Google Scholar

⁵⁶ Brooks, , Western Africa, 154–67Google Scholar; Kaba, Lansiné, ‘Archers, musketeers, and mosquitos: the Moorish invasion of the Sudan and the Songhay resistance (1591–1612)’, J. Afr. Hist, xxii (1981), 457–75.CrossRefGoogle Scholar

⁵⁷ Inikori, ‘Under-population’, 300–1, 305–6; idem. ‘Introduction’, 46–51.

⁵⁸ This state of endemic warfare is conveniently summarized in Claude Meillassoux, ‘The role of slavery in the economic and social history of Sahelo-Sudanic Africa’ in Inikori, Forced Migration, 76–83.

⁵⁹ Inikori, ‘Under-population’, 297–8.

⁶⁰ Inikori, ‘Introduction’, 29; idem. ‘Two comments’, 131; Suret-Canale, Jean, ‘La Sénégambie à l'ére de la traite’, Canadian Journal of African Studies, xi (1977), 125–34.Google Scholar

⁶¹ Becker, Charles and Martin, Victor ,’ Kayor et Baol: royaumes sénégalais et traite des esclaves au XVIIIe siècle’, Revue française d'histoire d'Outre-Mer, lxii (1975), 271.Google Scholar Cf. Becker, Charles, ‘La Sénégambie à l'époque de la traite des esclaves’, Revue française d'histoire d'Outre-Mer, lxiv (1977), 203–24Google Scholar; Becker develops these arguments in ‘Les effets démographiques de la traite des esclaves en Sénégambie: esquisse d'une histoire des peuplements du 17e à la fin du 19e siècle’, Colloque International sur la Traite des Noirs, Nantes, July, 1985.

⁶² Becker and Martin, ‘Kayor et Baol’, 272.

⁶³ Ibid. 273.

⁶⁴ Becker, and Martin, , ‘Les teeñ du Baol: essai de chronologie’, Bulletin de l'Institut Fondamental de l'Afrique Noire, sér B, xxxvin (1976), 464–80.Google Scholar

⁶⁵ Ibid.

⁶⁹ Becker, ‘Sénégambie’, 218.

⁶⁷ Butler, Samuel, Further Extracts from the Note-books of Samuel Butler, ed. Bartholomew, A. T. (London, 1934), 158.Google Scholar

⁶⁸ Commensurability governs many aspects of historians' work, whether quantitative or not. This is appropriate to the extent that it is necessary, but it bears emphasizing that commensuration is basically another form of interpretation, governed more by the historian's perceptions than by further and independent evidence.

⁶⁹ A good survey of the premises of pyrrhonism is Arne Naess, Scepticism (London, 1968). See also Popkin, Richard H., ‘The high road to Pyrrhonism’, American Philosophical Quarterly, ii (1965), 18–32Google Scholar; Burnyeat, M. F., ‘The sceptic in his time and place’, in Rorty, Richard, Schneewind, J. B. and Skinner, Quentin, eds. Philosophy in History(Cambridge, 1985), 225–54.Google Scholar

⁷⁰ J. E. Inikori, ‘The slave trade and the African economies, 1571–1870’, in The African slave trade from the fifteenth to the nineteenth century. Reports and papers of the meeting of experts organized at Port-au-Prince, Haiti, 31 January to 4 February 1978 (Paris, 1979), 57.

⁷¹ ‘Nearly all the [historical] evidence that will ever appear is accessible now’. Lord Acton, Lectures in Modern History (London, 1907), 315.Google Scholar

⁷² I would like to thank Bruce Fetter, Paul Hair, Paul Lovejoy, Joseph Miller, and Donald Wright for providing their own devil's advocacy on an earlier draft of this paper, without necessarily implying that they would agree with the residue.

⁷³ Thomson, William, Kelvin, Baron, ‘Electrical units of measurement’, in his Popular Lectures and Addresses (London, 1889–1894), I, 73.Google Scholar