¹ Nieboer, H. J., Slavery as an Industrial System (The Hague, Revised Edition, 1910), 383–8.Google Scholar Modern discussions include Domar, E. D., ‘The Causes of Slavery or Serfdom: A Hypotheses’, Journal of Economic History, XXX 1 (1970),Google Scholar and Boserup, E., The Conditions of Agricultural Growth: The Economics of Agrarian Change under Population Pressure (Chicago, 1965), 73–5.Google Scholar

² Nieboer, H. J., op. cit. 385.Google Scholar

³ Domar, E. D., op. cit. 31.Google Scholar Such a criticism may be unduly harsh since Nieboer's concluding chapter does explicitly recognize the role of ‘external causes’ in allowing for the development of slavery. ‘But though, where motives for keeping slaves fail, no external causes will give rise to slavery—even when there are such motives, slavery will not exist, if there are no external causes rendering it possible, i.e. if there is no opportunity of procuring and retaining slaves. Where neither capture or purchase of aliens, nor enslavement of members of the tribe is practicable, or where the slaves can very easily escape, slaves cannot be kept, though there may be much use for them.’ Nieboer, H. J., op. cit., 417–18.Google Scholar

⁴ Such an hypothesis accords with Nieboer's fuller discussion of ‘internal and external causes’ of slavery. Plantation demands for labour represent an internal cause while African supply responses are the external. It also conforms to the emphasis which Robert Evans, Jr., has placed on the reaction of the labour-supplying sectors. ‘The essential element which is required in order for a labour shortage to lead to the use of coerced labour is the failure of other sectors of the economy, those from whom the workers will be taken, to protest.’ Evans, Robert Jr, ‘Some Notes on Coerced Labor’, Journal of Economic History, XXX, 4 (1970), 862.Google Scholar

⁵ Though attempts were frequently made to enslav1e the Indian populations, labour from this source is of lesser historical importance. With the partial exception of Brazil and Mexico, Indian slave labour did not represent a long-term solution to the problem of New World labour demand. Among the reasons were insufficiency of numbers and the continual prospects of escape. See Nieboer, H. J., op. cit. 410–16;Google ScholarRich, E. E., ‘Colonial Settlement and Its Labor Problems’, The Cambridge Economic History of Europe, IV (Cambridge, 1967) Chapter vi;Google ScholarWyndham, H. A., The Atlantic and Slavery (London, 1935), Part II.Google Scholar The emphasis on West Africa is the more acceptable because the majority of all slaves shipped to the Americas probably came from the Lower Guinea Coast, between the present day Ivory Coast and Cameroon. See de Barenne Kilson, Marion Dusser, ‘West African Society and the Atlantic Slave Trade, 1441–1865’, in Huggins, Nathan I.Kilson, Martin, and Fox, Daniel M., Key Issues in the Afro-American Experience (New York, 1971), 1, 40.Google Scholar Further, the destination for a very large proportion of slaves shipped from Central African ports was Brazil and other South American points, an area outside the general frame of reference of this essay.

⁶ For an excellent review of definitions and concepts applicable, see Beckford, G. L., Persistent Poverty (London, 1972), Chap. 2.Google Scholar For detail on the plantation pattern, see Pitman, F. A., ‘Slavery on the British West India Plantations in the Eighteenth Century’, The Journal of Negro History, II, 1 (07 1926).Google Scholar

⁷ Conrad, A. H. and Meyer, J. R., ‘The Economics of Slavery in the Antebellum South’, 43–92, in The Economics of Slavery (Chicago, 1964).Google Scholar The original article by Conrad and Meyer spawned a subsequent debate which is perceptively reviewed by Engerman, S. L., ‘The Effects of Slavery upon the Southern Economy: A Review of the Recent Debate’, in Explorations in Entrepreneurial History (Second Series) IV, 2 (Winter, 1967), 71–97.Google Scholar Much of the original writing bearing on the debate appears in a collection edited by Aitken, H. G. J., Did Slavery Pay? (Boston, 1971).Google Scholar

⁸ Harlow, V. T., A History of the Barbados (Oxford, 1926), 307.Google Scholar

⁹ For prices on indentured servants, see Smith, A. E., Colonists in Bondage (Chapel Hill, 1947), 31–4.Google Scholar It should be noted that these prices represent ‘support prices’ resulting from various island governments' attempts to attract white labour. It would not be correct to assume that the removal of such supports would have produced a price equalization between slave and indentured labour. Ship captains with a cargo of servants would simply move north to the Middle Colonies where demand for indentured servants was greater. Slave prices appear in the Report of the Lords of the Committee of Council Appointed for the Consideration of all Matters Relating to Trade and Foreign Plantations (1789) unpaged, part iv, no. 25, ‘A Chronological Account of the Prices…’. The data refer to slaves purchased on the coast of West Africa and sold in the British West Indies.

¹⁰ Rees, J. F., ‘Mercantilism and the Colonies’, Chap. xx in Rose, H. H. et al. (ed.), The Cambridge History of the British Empire, I (N.Y., 1929), 588.Google Scholar For detail on the ‘deficiency laws’ and the point cited, see Wyndham, H. A., The Atlantic and Slavery, op. cit. 270–I.Google Scholar

¹¹ Note the discussion of the valuation question in Phillips, U. B., American Negro Slavery (N.Y., 1952), chap. xix.Google Scholar For the modern theory and calculations for the American South, see Conrad, A. H. and Meyer, J. R., op. cit. 43–92.Google Scholar

¹² Phillips, op. cit. 47.

¹³ Conrad and Meyer, op. cit. The calculations are explicit in this regard.

¹⁴ Behavioural traits associated with slave tribal origins served to raise or lower the prices of slaves depending on the positive or negative value of the presumed trait. See Pitman, F. W., op. cit. 589–92.Google Scholar

¹⁵ Kirkland, Edward C., A History of American Economic Life (N.Y., 1951), 570.Google Scholar

¹⁶ Ibid. 171.

¹⁷ Curtin, P. D., The Atlantic Slave Trade (Madison, 1969),Google Scholar Table 77 and Fig. 27. The comparability of data from the seventeenth century with those of the eighteenth is subject to reservation as Curtin notes.

¹⁸ Report of the Lords, op. cit., unpaged, part iv, no. 25.

¹⁹ Prado, C. Jr, The Colonial Background of Modern Brazil (Berkeley, 1967), 19.Google Scholar

²⁰ Largely applicable to the period after 1660. During the earlier period, the doctrine that colonies should serve as outlets for surplus population had considerable currency. See Knorr, E. E., British Colonial Theories 1570–1850 (Toronto, 1994), 41–8, and 68–81.Google Scholar

²¹ Kidnapping and ‘trepanning’ as methods of acquiring indentured servants are described in Harlow, V. T., A History of the Barbados (Oxford, 1926), 294–301.Google Scholar Continuing through the eighteenth century, coercion was employed to obtain several types of scarce labour—soldiers for the Prussian army, seamen for the Royal Navy, etc. Such coercion served to depreciate the price of military labour, and thereby lessened the burden falling on taxpayers.

²² For example, quoting Harlow: ‘Before the Restoration, the usual period of service was from five to seven years; after that date it became reduced to three or four years’, op. cit. 301.

²³ Empirical estimates of slave labour supply curve elasticities made by E. P. LeVeen support the contention that these curves were relatively elastic. See LeVeen, E. P., British Slave Trade Suppression Policies 1821–1865 (unpublished University of Chicago Ph.D. dissertation, 1971), 166–8.Google Scholar

²⁴ See the discussion in Basil Davidson, ‘Slaves or Captives? Some Notes on Fantasy and Fact’, in Huggins, Kilson, and Fox, op. cit. 54–73; Fisher, A. and Fisher, H., Slavery and Muslim Society in Africa (London, 1970), especially chaps. 5, 6, and 7;Google ScholarLugard, F. D., ‘Slavery in All its Forms’, Africa, 1933;CrossRefGoogle Scholar and Cohen, Ronald, ed., ‘Slavery in Africa’, Trans-Action, IV, Special Supplement (07.–02. 1967) which surveys the institution of slavery among the Kanuri, the Ibo, and the Ashanti.Google Scholar

²⁵ Cf. Fage, J. D., ‘Slavery and the Slave Trade in the Context of West African History’, J. Afr. Hist. X, 3 (1969).Google Scholar Fage employs a definition of slaves almost identical to the one used in the text. This is not to argue that slavery was indigenous to Africa; the institution might have been an import from the Mediterranean basin where it had been practised for over a millennium by Greece, Rome and their successors including the Muslim Barbary States. Contact between these states and Africa south of the Sahara might have introduced the idea to the latter areas.

²⁶ A catalogue of source material is in Mauny, Raymond, Tableau Géographique de l'ouest Africain au moyen âge (Dakar, 1961).Google Scholar There is a discussion of the early evidence in Fisher and Fisher, op. cit., chap. 3.

²⁷ Fage, J. D., A History of West Africa (Cambridge, 1969), 94.Google Scholar

²⁸ Ibid. However, this development does not appear to have taken place everywhere. Dr Walter Rodney has shown for the area from what is now Gambia to Liberia, that neither slavery nor slave trading seem to have been encountered until social changes occurred following the arrival of European traders. Rodney, Walter, ‘African Slavery and Other Forms of Social Oppression of the Upper Guinea Coast in the Context of the Atlantic Slave-Trade’, J. Afr. Hist., VII, 3 (1966).Google Scholar See also Rodney's, A History of the Upper Guinea Coast 1545–1800 (Oxford, 1970), 260–3.Google Scholar

²⁹ Meillassoux, Claude, ed., The Development of Indigenous Trade and Markets in West Africa (London, 1971), 54. The parenthetical passages are footnotes 1 and 2 in the original.Google Scholar

³⁰ Wyndham, H. A., The Atlantic and Slavery (London, 1935), 225.Google Scholar

³¹ Fage, History of West Africa, op. cit. 94. There is supporting evidence in Meillassoux, op. cit. 9–10, and Hair, P. E. H., ‘The Enslavement of Koelle's Informants’, J. Afr. Hist., XI, 3 (1965), 193–203.CrossRefGoogle Scholar Enslavement for various crimes seems to have been more common on the Upper Guinea Coast than in most other areas. There was also more raiding against fellow countrymen in this region. See Rodney, History of the Upper Guinea Coast, op. cit. 106–9, 117. The situation was just the opposite on the Gold Coast, where enslavement and overseas sale of criminals was rare. Daaku, K. Y., Trade and Politics on the Gold Coast 1600–1720 (Oxford, 1971), 29.Google Scholar

³² Davidson, Basil, Black Mother (Boston, 1961), 106–7.Google Scholar For a description of several strong West African Governments and their acquisition of captives late in the slave trade see Webster, J. B. and Boahen, A. A., History of West Africa (New York, 1967),Google Scholar part two. The strong state/weak state situation is explored thoroughly in the regional essays contained in Ajayi, J. F. A. and Crowder, Michael, eds., History of West Africa, I (New York, 1972).Google Scholar Note the map showing fragmented polities in Daaku, op. cit. 199.

³³ Mannix suggests perhaps only one or two out of 100 slaves ‘were free Africans kidnapped by Europeans’. See Mannix, D. P., Black Cargoes: A History of the Atlantic Slave Trade, 1518–1865 (New York, 1962), XI.Google Scholar

³⁴ Wyndham describes the tenuous position in which Europeans often found themselves on the coast. Wyndham, op. cit., chap. iii, 60–7.

³⁵ The British Government fully appreciated the disruption to trade that could occur by this means. It established a fine of £100 on any captain found guilty of kidnapping. Earlier, the Royal African Company also strongly discouraged the practice. Ibid. 225.

³⁶ Meilassoux, op. Cit. 66, with the qualification that captives were sometimes used to supply the needs of court.

³⁷ Philip D. Curtin, ‘The Slave Trade and the Atlantic Basin: Intercontinental Perspectives’, in Huggins, Kilson, and Fox, op. cit. 76. Almost all the cases of slave revolts in West Africa were fomented by marketed slaves. Meillassoux, op. cit. 65.

³⁸ See Curtin, ‘The Slave Trade…’, op. cit. 86. Little evidence is available on the extent of the purely internal trade, where slaves were purchased for use within West Africa itself. Curtin thinks the internal trade was more important than exports across the Sahara. ‘The Atlantic Slave Trade 1600–1800’ in Ajayi and Crowder, op. Cit. 245.

³⁹ For Muslim legal and customary views of slavery in Africa, see Fisher and Fisher, op. cit., chap. 2. See also Bennett, N. R., ‘Christian and Negro Slavery in Eighteenth Century North Africa’, J. Afr. Hist., 1, 1 (1960).CrossRefGoogle Scholar Muslim slave use is noted in Gann, Lewis H. and Duignan, Peter, Africa and the World (San Francisco, 1972), 220.Google Scholar

⁴⁰ There is fragmentary evidence to show that the price of slaves in the Mediterranean area was not less than twice the price in West Africa. See Fisher and Fisher, loc. cit. 68–9. The difference was probably much greater, as shown by the selection of high North African prices in Davidson, ‘Slaves or Captives’, op. cit. 58–9.

⁴¹ Fage notes an estimate that as many as two slaves in five may have died on the desert crossing. See History of West Africa, op. cit. 81. It might appear that profitable possibilities would thus exist for shipping slaves by sea from the West African coast to the Barbary States, via Cape Verde and the Straits of Gibraltar. But the fleets of Algiers, Tunis, etc., were mostly small feluccas, xebecs, and row galleys unsuited to the Atlantic and to long voyages. Detail on the question of why galleys could not cope with the Atlantic is in DeGregori, Thomas R., Technology and the Economic Development of the Tropical African Frontier (Cleveland, 1969), 105.Google Scholar The small size of the sailing vessels used by the North African states is noted by Bennett, op. cit. 65, 76. Furthermore, after the early 1400s their Portuguese and Spanish neighbours, with whom they were frequently at war until the nineteenth century and whose masted vessels were fully at home in the open ocean, maintained a state of naval supremacy which would have made such a coastal traffic extremely risky at best. The long-standing antipathy between the North African Muslims and the Iberian countries also explains why no very extensive slave trading developed between them, although for a time such trade was on the increase. Cf. Edinburgh University, The Transatlantic Slave Trade from West Africa (Edinburgh, 1965), 77.Google Scholar

⁴² Costs were so high that the trans-Saharan commerce has been called primarily a luxury trade. See Davidson, ‘Slaves or Captives’, op. cit. 58–9. Mauny's estimate of 10,000 slaves shipped annually across the desert (see Mauny, op. cit. 379), is probably very high.

⁴³ Cf. Hla Myint, ‘The Classical Theory of International Trade and the Underdeveloped Countries’, Economic Journal, 68 (June, 1958), 317–37; his Economic Theory and the Underdeveloped Countries (London, 1971), chap. 5; and Brown, Wilson B. and Hogendorn, Jan S., ‘Agricultural Export Growth and Myint's Model: Nigeria and Peru’, Agricultural History, XLVI, 2 (04, 1972), 213–24.Google Scholar

⁴⁴ See the short summary of the model in Eicher, Carl K. and Liedholm, Carl, Growth and Development of the Nigerian Economy (East Lansing, 1970), 6–7.Google Scholar The version given above is modified slightly to include some elements of Adam Smith's ‘Productivity Theory’, so named by Myint, Economic Theory and the Underdeveloped Countries, op. cit. 120. This modified version contains some elements which can be used as a criticism of Myint's original formulation. For example, one focus of this essay will be on firearms as an instrument of technical change, but this could be termed real capital formation, not predicted by the original model. Our similar consideration of money and credit could alternately have been treated as the amassing of the working capital of the trade. The growth of indigenous entrepreneurial ability is in the same category: included here, but not predicted by the original. Given the applicability of so much of the model, however, we have attempted to modify rather than criticize it.

⁴⁵ With the exception of transport, internal changes are little emphasized by Myint.

⁴⁶ Cf. Davidson, , Black Mother, op. cit. 225–6,Google Scholar who emphasizes ideological and protonational exclusiveness as an explanation for the willingness to enslave fellow blacks.

⁴⁷ As where serfs bound to the land might be required to provide a period of military service to their lord after the harvest was in.

⁴⁸ Fragmentary evidence for the position of the watershed may be deduced from portuguese efforts to change the direction of the Saharan traffic at Wadan on the westernmost Sahara caravan route. See Edinburgh University, op. cit. 77. The demand shift and the resulting alteration in supply patterns from the Saharan trade to the coast is depicted in Hargreaves, John D., West Africa: the Former French States (Englewood Cliffs N.J., 1967), 35, 37,Google Scholar and Gann and Duignan, op. cit. 313.

⁴⁹ It was imperative that this area be widened over time whenever the slave supply was exhausted in the immediate proximity of the entrepreneurs' territory. The geographical widening involved innovative exchange mechanisms, with slaves and trade goods often handled several times by middlemen. Curtin in Ajayi and Crowder, op. cit. 245, 254; Gann and Duignan, op. cit. 324. State entrepreneurship manifested itself in the armed patrols devoted to protection and convoy on various long-distance trade routes. R. A. Adeleye, ‘Hausaland and Bornu 1600–1800’ in Ajayi and Crowder, op. cit. 524.

⁵⁰ A discussion of these tactics in the Nupe Kingdom is in Mason, Michael, ‘Population Density and “Slave Raiding”—the Case of the Middle Belt of Nigeria‘, J. Afr. Hist., X, 4 (1969), 551–64.CrossRefGoogle Scholar

⁵¹ Meillassoux, op. cit. 63–7 and Daaku, op. cit. 29, 31.

⁵² For specific examples see Kilson, M. D. d. B., op. cit. 45–51.Google Scholar Strong states were often particularly anxious to acquire ports. Hargreaves, op. cit. 38. In addition, they might impose tolls on inland commerce headed coastward, and block imports of firearms to states further inland. Fynn, J. K., Asante and its Neighbours 1700–1807 (London, 1971), 25.Google Scholar

⁵³ Economic theory predicts that the monopolist will restrict supply and raise price so as to maximize total profit.

⁵⁴ One of the clearest cases of this was in the area of the Niger Delta. See Davidson, ‘Slaves or Captives’, op. cit. 69.

⁵⁵ Curtin, ‘The Slave Trade and the Atlantic Basin’, op. cit. 92.

⁵⁶ Kea, R. A., ‘Firearms and Warfare on the Gold and Slave Coasts from the Sixteenth to the Nineteenth Centuries’, J. Afr. Hist., XII, 2 (1971), 188;Google Scholar and Fisher, Humphrey J. and Rowland, Virginia, ‘Firearms in the Central Sudan’, J. Afr. Hist., XII, (1971), 218.Google Scholar Kea says (p. 194) that the demand for firearms was ‘almost insatiable’ on the coasts in the late seventeenth and early eighteenth centuries.

⁵⁷ Kea, op. cit. 209, and Fisher and Rowland, op. cit. 232. The importance of firearms is emphasized by Goody, Jack, Technology, Tradition, and the State in Africa (London, 1971), 28–9, 53;Google Scholar and by Hunwick, J. O., ‘Songhay, Bornu, and Hausaland in the sixteenth century’ in Ajayi and Crowder, op. cit. 208, 237–8.Google Scholar

⁵⁸ Davidson notes that at the peak period of the eighteenth century, Birmingham alone was exporting 100,000 to 150,000 muskets a year to Africa. Black Mother, op. cit. 242. See also his analysis in Africa in History (London, 1968), 193, and in ‘Slaves or Captives’, op. cit. 69–70. The importance of this ‘gun–slave cycle’ has proven to be unexpectedly controversial in recent years. Some authors apparently feel that bows and arrows, swords, lances, etc., were more effective in African hands than flintlock firearms. Why this should be so in Africa when it was not in Europe and the Americas is not entirely clear. It is true, however, that muskets in the West African trade were notorious for their poor state of repair. Cf. Kea, op. cit. 203–4.

⁵⁹ Rottenberg, Simon, ‘The Business of Slave Trading’, The South Atlantic Quarterly, LXVI, 3 (summer 1967), 413.Google Scholar These figures are generally corroborated by Report of the Lords of the Committee of Council Appointed for the Consideration of all Matters Relating to Trade and Foreign Plantations (1789) unpaged, part iv, no. 1, ‘An Account of the Number of ships with their tonnage…’. It must be recalled that even small coastal vessels could carry 200 or so slaves. See Rodney, History of the Upper Guinea Coast, op. cit. 97. Conditions were very competitive in ocean shipping during most of the slave trade, so that it is unlikely that monopoly profits were earned on any scale. After 1750, a few ships were specially designed for the trade with resulting economies.

⁶⁰ Mannix, Black Cargoes, op. cit. 106. The ‘loose’ versus ‘tight’ controversy is discussed on pp. 105–7. There is also evidence to show that death rates among slaves in transit decreased during the eighteenth century. See Curtin, The Atlantic Slave Trade, op. cit. 276–9.

⁶¹ Rottenberg, op. cit. 424.

⁶² Johnson, Marion, ‘The Cowrie Currencies of West Africa, Parts land II’, J. Afr. Hist., XI, 1 (1970), and XI, 3 (1970), 332.Google Scholar

⁶³ Ibid. Part I, 18.

⁶⁴ Ibid.. Part I, 21–2.

⁶⁵ Newbury, C. W., ‘Credit in Early Nineteenth Century West African Trade’, J. Afr. Hist., XIII, 1 (1972), 84–6, 94–5.Google Scholar

⁶⁶ A discussion of credit in a local setting is in Latham, A. J. H., ‘Currency, Credit, and Capitalism on the Cross River in the Pre-Colonial Era’, J. Afr. Hist., XII, 4 (1971).Google Scholar Use of credit to finance slave traders in the eastern Sudan is discussed by O'Fahey, R. S., ‘Slavery and the Slave Trade in Dar Für’, J. Afr. Hist., XIV, 1 (1973), 33.Google Scholar Credit on the Gold Coast is discussed by Daaku, op. cit. 41. Sometimes credit took a remarkably direct form: loans of guns and ammunition by white traders with the understanding that repayment would be in slaves. Ibid. 30.

⁶⁷ Mauny, Op. Cit. 377.

⁶⁸ Ibid. 338.

⁶⁹ The market was not always easy to read, and ‘goods that were highly prized at one part of the coast had less value at another’. Rottenberg, op. cit. 416. Also see Fage, History of West Africa, op. Cit. 89.

⁷⁰ See, for example, Nurkse, Ragnar, Problems of Capital Formation in Underdeveloped Countries (New York, 1953).Google Scholar

⁷¹ Fage, History of West Africa, op. cit. 91.