This paper expands upon material presented elsewhere, which includes more detailed citations to the literature. In particular, see “Economic Adjustments to Emancipation in the United States and British West Indies,” Journal of Interdisciplinary History, 13 (Autumn 1982), 191–220;Google Scholar and “Servants to Slaves to Servants: Contract Labor and European Expansion,” Colonialism and Migration: Indentured Labour Before and After Slavery, ed. van den Boogaart, E. and Emmer, P. C. (The Hague, forthcoming), Comparative Studies in Overseas History, vol.6. An earlier version was presented at the Caltech/Weingart/Social Science History Association Conference, March 1983.Google Scholar

¹ Lewis, W. Arthur, Growth and Fluctuations, 1870–1913 (London, 1978), pp. 181–88.Google Scholar

² The basic data and interpretations are found in Ferenczi, lmre and wilicox, Walter F., International Migrations, vol. I (New York, 1929);Google Scholar and Willcox, Walter F., ed., International Migrations, vol. 2 (New York, 1931).Google Scholar

For useful summaries of these patterns see CarrSaunders, A. M., World Population (Oxford, 1936), pp. 48–52;Google ScholarErickson, Charlotte, ed., Emigration from Europe, 1815–1914 (London, 1976);Google Scholar and Eltis, David, “Free and Coerced Transatlantic Migrations: Some Comparisons,” American Historical Review, 88 (04 1983), 251–80. For discussions of the timing and changing sources of origin of European migration to the United StatesCrossRefGoogle Scholar, see Easterlin, Richard A., “Influences in European Overseas Emigration Before World War I,” Economic Development and Cultural Change, 9 (04 1961), 331–51;CrossRefGoogle ScholarThomas, Brinley, Migration and Economic Growth (Cambridge, 1954);Google Scholar and the three-part series of articles by Gould, J. D. in Journal of European Economic History, most particularly, “European Inter-Continental Emigration, 1815–1914: Patterns and Causes,” 8 (Winter 1979), 593–679.Google Scholar

³ On this, see, most recently, Holloway, Thomas H., Immigrants on the Land (Chapel Hill, 1980).Google Scholar

⁴ The phrase is used as the title of the important study of “the export of Indian labour overseas, 1830–1920,” by Hugh Tinker. See Tinker, Hugh, A New System of Slavery (London, 1974).Google Scholar

⁵ See Engerman, “Economic Aspects.” Two Caribbean islands (Antigua and Bermuda) ended slavery without the period of apprenticeship. Slavery did not end in all the British colonies at this time, the legal ending of Indian slavery not taking place until 1843. Slavery in some areas of control within Africa continued into the twentieth century. For a useful discussion of the ending of slavery in two British East African colonies, with comparisons with the process of emancipation elsewhere, see Cooper, Frederick, From Slaves to Squatters (New Haven, 1980).Google Scholar

⁶ See Eltis, David, “The Direction and Fluctuation of the Transatlantic Slave Trade, 1821–1843: A Revision of the 1845 Parliamentary Paper,” in The Uncommon Market, ed. Gemery, Henry A. and Hogendorn, Jan S. (New York, 1979), pp. 273–301;Google Scholar and Eltis, David, “The Direction and Fluctuation of the Transatlantic Slave Trade, 1844–1867” (unpublished), for refinements to the estimates in Curtin, Philip D., The Atlantic Slave Trade: A Census (Madison, 1969). For a recent discussion of the magnitude of the slave tradeGoogle Scholar, see Lovejoy, Paul E., “The Volume of the Atlantic Slave Trade: A Synthesis,” Journal of African History, 23 (1982), 473–501.CrossRefGoogle Scholar

See also Klein, Herbert S., The Middle Passage (Princeton, 1978).CrossRefGoogle Scholar

⁷ Compare Curtin, , Atlantic Slave Trade, pp. 216 and 234, with the revision by David Eltis to appear in his forthcoming book on slavery and the slave trade in the nineteenth century. Earlier estimates by Eltis are in the works cited in footnote 6.Google Scholar

⁸ See Fogel, Robert William and Engerman, Stanley L., Time on the Cross (Boston, 1974), and the subsequent discussion of the economics of United States slavery. While the basis for expansion and the prospects for continuation in the absence of the Civil War remain debated, there seems little disagreement on the actual pre-1860 pattern. See, for exampleGoogle Scholar, Wright, Gavin, The Political Economy of the Cotton South (New York, 1978).Google Scholar

⁹ See Deerr, Noel, The History of Sugar (London, 1949–1950) for a discussion of the relationship between the expansion of sugar production and the utilization of slave labor.Google Scholar

¹⁰ See the data conveniently brought together in Hanson, John R. II, Trade in Transition (New York, 1980), pp. 171, 172, 174, and 178. No estimate is given (on p. 178) for Cuban sugar exports in 1860, but it is estimated that Cuba accounted for almost one-third of world cane sugar production in that year.Google Scholar

See Fraginals, Manuel Moreno, El Ingenio (Havana, 1978), vol. 3, p. 36.Google Scholar

¹¹ See Fraginals, Manuel Moreno, Klein, Herbert S., and Engerman, Stanley L., “The Level and Structure of Slave Prices on Cuban Plantations in the Middle of the Nineteenth Century: Some Comparative Perspectives,” American Historical Review (forthcoming), for references to slave prices in Cuba, Brazil, Puerto Rico, and the United States.Google Scholar

¹² For the attack on serfdom, see Blum, Jerome, The End of the Old Order in Rural Europe (Princeton, 1978), particularly pp. 305–31.CrossRefGoogle Scholar

Blum, (p. 320) refers to the claims that “the peasant as hired laborer was two to four times more productive in a given period than was the peasant doing his compulsory labor service,” a range of estimates similar to that often given for the comparison of slave and free labor.Google Scholar

¹³ See Galenson, David, White Servitude in Colonial America (Cambridge, 1981);Google Scholar and Smith, Abbot Emerson, Colonists in Bondage (Chapel Hill, 1947).Google Scholar

¹⁴ For comparisons, see Engerman, “Servants to Slaves to Servants.”

¹⁵ For descriptions of sugar production in India and in Java, see Deerr, History of Sugar, pp. 40–62 and 208–27;Google Scholar and Geerligs, H. C. Prinsen, The World's Cane Sugar Industry: Past and Present (London, 1912), pp. 43–69 and 107–43. For a detailed analysis of the sugar industry in JavaGoogle Scholar, see Knight, G. R., “From Plantation to Padi-field: The Origins of the Nineteenth Century Transformation of Java's Sugar Industry,” Modern Asian Studies, 14 (04 1980), 177–204. Although East Indian sugar figured extensively in the British debates on the sugar duties in the first half of the nineteenth century, and exports were high in the 1840s and 1850s, Indian sugar exports started to decline after mid-century, and by the end of the century India was a net importer. By some measures India produced even more cane than did Cuba, but the quality was so low relative to costs that it was generally only for home consumption.CrossRefGoogle Scholar

¹⁶ On the movements of free Africans to the British Caribbean, see Schuler, Monica, “Alas, Alas, Kongo” (Baltimore, 1980);Google Scholar and Green, William A., British Slave Emancipation (Oxford, 1976), pp. 265–76.Google Scholar See also Laurence, K. O., Immigration into the West Indies in the 19th Century (St. Lawrence, Barbados, 1971), pp. 12–16 and 41.Google Scholar

¹⁷ On the Indian migration, see Tinker, , A New System of Slavery; on the Chinese migration within the British EmpireGoogle Scholar, see Campbell, Persia Crawford, Chinese Coolie Emigration (London, 1923); and on the Javanese migration to SurinamGoogle Scholar, see de Waal Malefijt, Annemarie, The Javanese of Surinam (Assen, 1963).Google Scholar

¹⁸ See Engerman, “Economic Adjustments.” The interrelated topics of sugar, slavery, technology, and adjustments to emancipation in the Caribbean have been the focus of the work of Sidney Mintz, and I have benefited from his insights. See, for example, his essay on “Slavery and the Rise of Peasantries,” Historical Reflections, 6 (Summer 1979), 213–42.Google Scholar

¹⁹ For a presentation of wage data, drawn mainly from the Colonial Office papers, reflecting the same relative patterns as discussed in the Parliamentary Committees of 1842 and 1847–1848 dealing with the effects of emancipation, see Riviere, W. Emanuel, “Labour Shortage in the British West Indies after Emancipation,” Journal of Caribbean History, 4 (05 1972), 1–30.Google Scholar

²⁰ See, for example, the discussion in Scott, Rebecca J., Slave Emancipation in Cuba: The Transition to Free Labor (Princeton, forthcoming).Google Scholar

²¹ For discussions of Australian sugar, see, most recently, Saunders, Kay, Workers in Bondage (St. Lucia, Queensland, 1982);Google ScholarGraves, Adrian, “The Nature and Origins of Pacific Island Labour Migration to Queensland, 1863–1906”, forthcoming in Studies in International Labour Migration, ed. Richardson, P. and Marks, S.;Google Scholar and Shlomowitz, Ralph, “Melanesian Labor and the Development of the Queensland Sugar Industry, 1863–1906,” Research in Economic History, 7 (1982), 327–61. For a discussion of NatalGoogle Scholar, see Richardson, Peter, “The Natal Sugar Industry, 1849–1905: An Interpretive Essay,” Journal of African History, 23 (1982), 515–27; on FijiCrossRefGoogle Scholar, see Moynagh, Michael, Brown or White? A History of the Fiji Sugar Industry, 1873–1973 (Canberra, 1981); and, on HawaiiGoogle Scholar, still useful is Coman, Katharine, The History of Contract Labor in the Hawaiian Islands (New York, 1903). In the cases of Fiji and Hawaii the initial European settlement did have a negative demographic impact on the native population, but the declines were not as severe as in the case of European movement into the Americas.Google Scholar

²² For a useful discussion of policies in the British Caribbean, see Green, , British Slave Emancipation.Google Scholar

See also Nwulia, Moses D. E., The History of Slavery in Mauritius and tile Seychelles, 1810–1875 (Rutherford, New Jersey, 1981);Google Scholar and Mookherji, S. B., The Indenture System in Mauritius, 1837–1915 (Calcutta, 1962).Google Scholar

²³ For comparisons, see Lewis, , Growth and Fluctuations, pp. 181–88; Lovejoy, “Volume of the Atlantic Slave Trade,” p. 497;Google Scholar and Eltis, “Free and Coerced Migrations.” A discussion of Indian outflows by receiving area is in Davis, Kingsley, The Population of India and Pakistan (Princeton, 1951), pp. 98–106;Google Scholar and the intracontinental movement of Chinese labor is discussed in Purcell, Victor, The Chinese in Southeast Asia (London, 1951). It is estimated that “nearly 162,000” convicts were sent from the United Kingdom to Australia between 1788 and 1868.Google Scholar

See Shaw, A. G. L., Convicts and the Colonies (London, 1966), pp. 147–48, and 361–68.Google Scholar

²⁴ See sources cited in Table I and in Engerman, “From Servants to Slaves to Servants.”

²⁵ In the numerous hearings of Parliament in regard to the difficulties of the West Indies (1842, 1847–1848, and 1897) and of the role of contract labor (1871, 1910, and 1915) the most frequently claimed advantage of contract labor was that it was “continuous” and “dependable.”

²⁶ These terms varied over time and differed among the receiving areas, dependent upon various factors, including distance between sending and receiving areas. There are useful summaries of these provisions and their changes in parliamentary reports, of which the following were the most useful: Report of the Commission on Treatment of Immigrants in British Guiana (1871); Copy of Mr. Geoghegan's Report on Coolie Emigration from India (1874); Report of the Committee on Emigration from India to the Crown Colonies and Protectorates (1910); and Report to the Government of India on the Conditions of Indian Immigrants in Four British Colonies and Surinam (1915).Google Scholar

See also Grierson, George A., Report on Colonial Emigration from the Bengal Presidency (Calcutta, 1883). In generalGoogle Scholar, see the discussions in Tinker, , A New System of Slavery;Google Scholar and Kondapi, C., Indians Overseas, 1838–1949 (New Delhi, 1951).Google Scholar

²⁷ This point was made by several witnesses in the 1842 parliamentary hearings, and recurred throughout the period of contract labor. The induced effect on domestic labor, plus the impact (direct or indirect) of permanently settling excontract labor, became the basis of the argument that the planters hiring contract labor should be subsidized out of general revenues.

²⁸ See, for example, the discussions of the provisions and arguments for different cost-sharing arrangements in the Report of the West India Royal Commission (1897). In general, for British Guiana about one-third of the costs of immigration were paid for out of general revenues, the remainder charged to the planter class (via an acreage tax) and to the specific hiring planter (who paid in installments over the period of indenture).Google Scholar

See Report (1897), Appendix C, vol. 2, pp. 118– 19. For the similar provisions in Trinidad, see pp. 313–15 and 238. For Jamaica, however, the costs were borne by those employing the migrants.Google Scholar

²⁹ For example, after mid-century there was a legal requirement for some areas that 30 percent of the contract laborers be female, which then became the observed ratio. In contrast were the high proportions of males in the Indian out-migration prior to the legislation (introduced in the 1850s), and the extremely high proportions of males in the outflows from China and the Pacific islands. it is estimated that only about 1.2 percent of the outflow from China to Latin America was female. See Meagher, Arnold Joseph, “The Introduction of Chinese Laborers to Latin America; the ‘Coolie Trade,’ 1847–1874” (Ph.D. diss., University of California, Davis, 1975), p. 96. The estimates for the Pacific islanders are generally under 10 percent female.Google Scholar

See Corns, Peter, Passage, Port and Plantation (Melbourne, 1973), p. 46.Google Scholar

³⁰ See, for example, the summary table in Grierson, Report on Colonial Emigration, Appendix 3, as well as the various sources listed in footnote 26.

³¹ See the discussions in Tinker, , A New System of Slavery; in Campbell, Chinese Coolie Emigration; and in Meagher, “The Introduction of Chinese Laborers.”Google Scholar

³² See Tinker, , A New System of Slavery, pp. 236–366, for a discussion of the attack on Indian contract labor emigration.Google Scholar

³³ See, for example, the articles by Graves and Shlomowitz cited in footnote 21.

³⁴ For annual estimates of the outflow to (and return from) Ceylon in the period 1839–1870. see Jayaraman, R., “Indian Emigration to Ceylon: Some Aspects of the Historical and Social Background of the Emigrants,” Indian Economic and Social History Review, 4 (12 1967), 319–59.CrossRefGoogle Scholar

³⁵ See, for example, the graph of the annual movements of emigration and rice prices in Grierson, , Report on Colonial Emigration; and the comments in Report of the Com, nittee on Emigration from India (1910) (Cd. 5192), p. 10.Google Scholar

³⁶ The outflow to the British colony, Mauntius, was longer in time and greater than that to French Réunion, Mauritius accounting for over four-fifths of the Indian Ocean emigration, according to the estimates of Table 1. A considerably greater share for Mauritius in indicated by the estimates in Ferenczi, and Willcox, , International Migrations, vol. I, pp. 904–5, the discrepancy being due in part to early “smuggling” of Indian laborers into the French colony (p. 146), and also to the failure to record the Indian migration to Reunion in the 1850s in the official sources.Google Scholar

³⁷ The British Caribbean received about four-fifths of the Caribbean movement, about 45 percent of the total going to British Guiana, and one-quarter to Trinidad. See Table 1.

³⁸ See the discussion in Green, British Slave Emancipation; Laurence, Immigration into the West Indies;Google Scholar and Adamson, Alan H., Sugar Without Slaves (New Haven, 1972). It is possible that a decline in the costs of passage had an influence on this shift in destination, but the magnitude of any change and its impacts were relatively minor. For the Atlantic crossing it appears that the major decline in passenger rates occurred by the 1830s, even though freight rates continued to decline in the second half of the nineteenth century. For data on nineteenth-century passenger fares, see David W. Galenson, “The Rise and Fall of Indentured Servitude in the Americas: An Economic Analysis” (unpublished);Google Scholar and Gould, , “European Inter-Continental Emigration,” pp. 611–14. For data on freight ratesGoogle Scholar, see North, Douglass C., “The Role of Transportation in the Economic Development of North America,” in Les Grandes Voies Maritimes dans le Monde, XV-X1X Siècles (Paris, 1965), pp. 209–46;Google Scholar and Harley, Charles K., “The Shift from Sailing Ships to Steamships, 1850–1890: A Study in Technological Change and Its Diffusion,” in Essays on a Mature Economy, ed. McCloskey, Donald N. (London, 1971), pp. 215–34 (which indicates a sharp decline in the freight rate between Bombay and London over the years 1873 to 1890). Nevertheless the discussions within the Colonial Office suggest that the delay in large-scale migration of contract labor from India to the Caribbean reflected political factors, not restricted planter demand.Google Scholar

³⁹ Nineteenth-century sugar output in Mauntius hit peak levels in the 1860s, and then remained at about the same average until the burst of expansion after the 1890s, with the rise of cane farming. After 1865 there was a sharp drop in the arrivals of Indians in Mauritius and an increase in the ratio of departures (to return to India) to the numbers arriving, with a dramatic slowing down in the rate of population growth as well as sugar production.

See Deerr, , History of Sugar, pp. 203–4;Google Scholar and Kuczynski, R. R., Demographic Survey of the British Colonial Empire (London, 1949), vol. 2, pp. 779 and 796.Google Scholar

⁴⁰ The most accessible breakdown of world sugar production between cane and beet between 1839 and 1940 is in Deerr, , History of Sugar, pp. 490–91. After World War I there was a relative increase in cane production, cane becoming about two-thirds of overall sugar production in the interwar period. Even with the declining share of world production, in the period 1839–1843 to 1889–1893 (before the war-induced Cuban decline), world cane sugar production rose at an annual rate of about 2.6 percent per year.Google Scholar

⁴¹ On the readjustments of the Louisiana sugar industry after the Civil War, see Sitterson, J. Carlyle, Sugar Country (Lexington, 1953). For a recent comparison of one of the British West Indies, Jamaica, with patterns in the American SouthGoogle Scholar, see Holt, Thomas C., “‘An Empire over the Mind’: Emancipation, Race, and Ideology in the British West Indies and the American South,” in Region, Race, and Reconstruction, ed. Kousser, J. Morgan and McPherson, James M. (New York, 1982), pp. 283–313.Google Scholar

See also Beachey, R. W., The British West Indies Sugar Industry in the Late 19th Century (Oxford, 1957), for a discussion of subsequent developments. The literature on the British West Indies is cited and discussed in Engerman, “Economic Adjustments.”Google Scholar

⁴² See Engerman, “Economic Adjustments.”

⁴³ For an examination of the impact of the changing sugar duties on British sugar imports, see Roberta M. Delson, “Sugar Production for the Nineteenth-Century British Market: Rethinking the Roles of Brazil and the British West Indies” (unpublished).

⁴⁴ See the reports and testimony in the Select Committee on the West Indian Colonies (1842) and in the Report from the Select Committee on Sugar and Coffee Planting (1847–1848).Google Scholar

⁴⁵ The declines were not only on sugar plantations. In Jamaica there was also a substantial outflow of labor from the coffee plantations, despite the expectations of some that coffee, since it was produced on smaller plantations than was sugar, would not be affected to the same extent as sugar. See the testimony of Peter Borthwick in the Fifth Report from the Select Committee on Sugar and Coffee Planting, and the data from the Jamaica Assembly Committee in the Appendix to the Seventh Report. For data on Jamaica coffee exports, see Hall, Douglas, Free Jamaica (New Haven, 1959), p. 186.Google Scholar

⁴⁶ See, for example, the detailed 1833 paper entitled, “Heads of a Plan for the Abolition of Negro Slavery, and for securing the continued Cultivation of the Estates by the Manumitted Slaves,” reprinted in “The Standing Committee of West India Planters and Merchants to Lord Goderich,” in the Appendix to the Fourth Report from the Select Committee on Sugar and Coffee Planting. See also the discussions of tax policy in Grey, Earl, The Colonial Policy of Lord John Russell's Administration (London, 1853), vol. 1, pp. 75–79.Google Scholar

⁴⁷ The data from the “Blue Books” printed in MacGregor, John, Commercial Statistics (London, 1850), vol. 5, p. 128, show that Indians were over 95 percent of the “population employed on sugar estates in Mauritius” in 1846.Google Scholar For a discussion of the disappearance of the “old population” from employment on sugar plantations, see the testimony of John George Raymond in the First Report of the 1847–1848 Select Committee. In 1870 it was estimated that Creoles accounted for only about 6 percent of sugar estate laborers. See Copy of Mr. Geoghegan's Report, p. 84. For an 1858 report relating to Trinidad that shows that less than one-quarter of the laborers on sugar estates were “Trinidad Creoles,” with a similar share of “Immigrant Creoles,”Google Scholar see Green, British Slave Emancipation, p. 292. In the 1847–1848 Select Committee Report (Appendix to Seventh Report), it was estimated that exslaves represented only 30 percent of the large sugar and cocoa estate workers (“Copy of a Despatch from Governor Lord Hams to Earl Grey”), and by 1897 it was estimated that the share of resident Creole laborers in the sugar plantation work force was “not quite 13 per cent.”Google Scholar See the Report of the West India Royal Commission (1897) (c. 8655), p. 100. There was also a “considerable amount of what is described as ‘outside labor’ employed” but this included both “free immigrants” and “Creoles.” In that year over 50 percent of the resident immigrants were still under indenture. Estimates of the ethnic composition of the sugar plantation labor force in British Guiana, drawn from census reportsGoogle Scholar, are in Mandle, Jay R., The Plantation Economy (Philadelphia, 1973), p. 21.Google Scholar By 1871 less than 30 percent of the estate population was non- Asian. There are many data on the number of immigrants in the plantation labor force, with details as to whether they are under indenture or not, in the various reports of the Immigration Agent- General. It seems clear that while the percentage of immigrants under indenture declined over time, migrants continued to represent the bulk of the plantation labor force. For the shift in British Guiana, from over three-quarters of the “Immigrant labor force resident on plantations” under indenture in 1872 to about 27.6 percent in 1903–1904, see Adamson, Sugar Without Slaves, p. 106.Google Scholar

See also Rodney, Walter, A History of the Guyanese Working People, 1881–1905 (Baltimore, 1981), p. 231.Google Scholar

⁴⁸ See the estimates, for the end of the century, of the costs of immigration and its allocation, in the sources cited in footnote 28. The costs of getting immigrants to the West Indies were generally estimated to fall in the range of £15-£25, with an additional cost if return fares were required. See 1897 Report, Appendix C, vol. 2, p. 315, for the details of the Trinidad fund. An estimate for British Guiana is in that volume, p. 49. Also compare the details of the fund on p. 119, with the estimated annual inflowGoogle Scholar, in Nath, Dwarka, A History of Indians in British Guiana (London, 1950), p. 180. For a discussion of the financing issueGoogle Scholar, see Adamson, , Sugar Without Slaves, pp. 137–42. The minimum daily wage rate in Trinidad and in British Guiana for immigrants in 1883 was one shilling, with the planter responsible for housing and medical care, as well as for food for varying periods after arrival. See Grierson, Report on Colonial Emigration, Appendix 3. See also the range of “daily wages for estate labor,”Google Scholar in Rodney, , History of the Guyanese Working People, p. 232, in excess of the minimum. Although there were additional costs from illness after arrival, and from the time required to gain the experience necessary to increase labor productivity, the allocation of the costs and the availability of credit suggest that the planters using indentured workers got their labor force at a reasonable, for them, premium.Google Scholar

⁴⁹ For data on the shift in the ethnic composition of the labor force in Hawaii with the growth of the sugar industry in the late nineteenth century, see Coman, , History of Contract Labor, p. 64; and the discussion and sources in Ronald Takaki, “The Making of a Multiethnic working Class in Hawaii” (unpublished). For the discussion of a similar change, from native-born workers to the attraction of migrants under contract in FijiGoogle Scholar, see Moynagh, , Brown or White?, pp. 34–39. The Australian case is discussed in the works of Shlornowitz, Saunders, and Graves, cited in footnote 21. There is a distinction in that in the Australian cases the resident labor force was white.Google Scholar

⁵⁰ See Sitterson, , Sugar Country; and Joseph P. Reidy, “The Development of Central Factories and the Rise of Tenancy in Louisiana's Sugar Economy, 1880–1910” (unpublished). For 1898 in Louisiana it was estimated that of the cane handled by “central factories and large plantations,” 60.8 percent was grown by owners, 7.5 percent by tenants, and 31.7 percent purchased from others. U.S. Census OfficeGoogle Scholar, Twelfth, Agriculture, Part II (Washington, D.C., 1902), p. 459. Most of the cane farms were owned and operated by whites, and blacks worked predominantly as plantation laborers. For the ethnic breakdown of the operators of sugar (and other) farms in 1900, see U.S. Census OfficeGoogle Scholar, Twelfth, , Agriculture, Part I (Washington, D.C., 1902), pp. lv, xcvii, and civ. The early postbellum attempts to use migrant labor—Chinese, Italian, or exslave from the Upper South—had been of very limited success.Google Scholar

⁵¹ See Geerligs, Prinsen, The World's Cane Sugar Industry, for measures of the shift in Trinidad, where in 1909, about one-quarter of cane was grown by cane farmers, of whom only 53 percent were East Indians—a quite different ratio from that of the plantation estate labor force (p. 218).Google Scholar

See also Johnson, Howard, “The Origins and Early Development of Cane Farming in Trinidad, 1882–1906,” Journal of Caribbean History, 5 (11 1972), 46–74. In Australia, the expansion of cane farming by whites, with tariff protection, shifted the percentage of sugar produced with white labor from an estimated 15.7 percent in 1902 to 90.7 percent by 1910.Google Scholar

See Geerligs, Prinsen, The World's Cane Sugar Industry, p. 335.Google Scholar

See also Shlomowitz, Ralph, “The Search for Institutional Equilibrium in Queensland's Sugar Industry, 1884–1913,” Australian Economic History Review, 19 (09 1979), 91–122. In Fiji, however, the transformation to small-scale farming still meant that Indian labor dominated sugar production.CrossRefGoogle Scholar

See Moynagh, , Brown or White?, pp. 69–119. Similarly the importance of Indian labor continued with the shift to cane farming in Mauritius.Google Scholar

See Coombes, Alfred North, The Evolution of Sugarcane Culture in Mauritius (Port Louis, Mauritius, 1937), p. 38.Google Scholar

⁵² For an excellent discussion of the optimum scale of production in sugar (and cotton), see Shlomowitz, Ralph, “Plantations and Smallholdings: Comparative Perspectives from the world Cotton and Sugar Cane Economies, 1865–1939,” Agricultural History (forthcoming), and, for an analysis of the United States, see also his “‘Bound’ or ‘Free’? Freedman Labor in Cotton and Sugar Cane Farming, 1865–1880,” Journal of Southern History (forthcoming). See alsoGoogle Scholar, Hirschman, Albert O., “A Generalized Linkage Approach to Development, with Special Reference to Staples,” Economic Development and Cultural Change, 25 (Supplement 1977), 67–98.Google Scholar

⁵³ On the unusual Brazilian case in the seventeenth and eighteenth centuries, see Schwartz, Stuart B., “Free Labor in a Slave Economy: The Lavadores de Cana of Colonial Bahia,” in Alden, Dauril, ed., Colonial Roots of Modern Brazil (Berkeley, 1973), pp. 147–97. For a later periodGoogle Scholar, see his “Patterns of Slaveholding in the Americas: New Evidence from Brazil,” American Historical Review, 87 (02 1982), 55–86.CrossRefGoogle Scholar

⁵⁴ This, of course, was the initial adjustment after the Civil War for cotton production in the United States South. Why there were different developments for these two crops—sugar and cotton—and the possible impact of varying technologies upon differences in the adjustments made by exslaves after emancipation in the United States and the West Indies remain questions of interest. See, for example, Engerman, “Economic Adjustments”; Shlomowitz, “‘Bound or ‘Free’”;?; and Holt, “‘An Empire Over the Mind.’” For an earlier view, see Root, J. W., The British West Indies and the Sugar Industry, (Liverpool, 1899), pp. 18–20.Google Scholar

⁵⁵ See Timoshenko, Vladimir P. and Swerling, Boris C., The World's Sugar (Stanford, 1957), pp. 63–89; and, on the CaribbeanGoogle Scholar, Hagelberg, G. B., The Caribbean Sugar Industries: Constraints and Opportunities (New Haven, 1974).Google Scholar

⁵⁶ See the works of Shlomowitz and Graves cited in footnotes 21 and 51. What is interesting in this case is that there had been discussions on the need for a new technology to permit the transition to white labor. For example, in the Weekly Statistical Sugar Trade Journal, 25 (10 3, 1901), there is a quote from an Australian report on the desirability of technical changes that would permit substitution of “lower by higher forms of labor,” and even earlier similar points had been raised. For another facet of the changes in the Australian sugar industry, teams working on a collective piece-work basisGoogle Scholar, see Shlomowitz, Ralph, “Team Work and Incentives: The Origins and Development of the Butty Gang System in Queensland's Sugar Industry, 1891–1913,” Journal of Comparative Economics, 3 (03 1979), 41–55.CrossRefGoogle Scholar

⁵⁷ See Scott, , Slave Emancipation in CubaGoogle Scholar, drawing upon data in the U.S. War Department, Report on the Census of Cuba, 1899 (Washington, D.C., 1900), pp. 403–05, 555–56, and 560. In 1917, based on a sample of more than one-half of the producers of its sugaroutput, it was estimated that “nearly 80 percent of the cane produced on the plantation lands of Cuba is planted, cultivated, brought to maturity and harvested by a large body of farmers called colonos.” U.S. Department of Commerce, Bureau of Foreign and Domestic CommerceGoogle Scholar, The Cane Sugar Industry, Miscellaneous Series, No. 53 (Washington, D.C., 1917), p. 360. Although not all colonos were small-scale, this seems to have been a general pattern of operations. This report also gives an estimate based on nearly three-quarters of sugar-producing establishments in Puerto Rico that “62.64 percent of the cane harvested for the crop in 1913–14 was purchased from colonos” (p. 254). The average acreage in cane of colonos in Puerto Rico was less than that in Cuba. For Hawaii, factories purchased only II. 5 percent of cane from “outside planters” (p. 177).Google Scholar

⁵⁸ See the sources cited in footnote 51.

⁵⁹ On British Guiana, see Adamson, , Sugar Without Slaves; and Rodney, A History of the Guyanese Working Class. For Hawaii in the first decade of the twentieth century, see the reports of the Commissioner of Labor on Hawaii published in the Bulletin of the Bureau of Labor. There were various arrangements utilized in Hawaii for payment, including time and piece wages, and contracts made with groups of varying size for cultivation and harvesting. Although this latter point could mean that the actual production unit, after planting, was smaller than the plantation size would suggest, the contracts covered only a part of the labor force (35 percent in 1915) and the workers under these contracts were apparently of the same ethnicity as the remainder of the plantation labor force. See Dept. of CommerceGoogle Scholar, The Cane Sugar Industry, pp. 106–10Google Scholar, and “Fourth Report of the Commissioner of Labor on Hawaii,” in Bulletin of the Bureau of Labor (Washington, D.C., 1911), vol. 22, pp. 684–89. This report also contains data on the ethnic breakdown of various occupations on Hawaiian sugar plantations, demonstrating the higher frequency of managerial and skilled occupations among the Caucasian, Hawaiian, and Portuguese than the Asian immigrant labor, as well as their higher wages in given occupations (pp. 684–92). For a discussion of the link between the availability of Asian labor and the possible future need for a shift to small farming if this labor supply were to be restricted or if “the Asiatic approaches the white man in his demands as a laborer,”Google Scholar see “Third Report of Commissioner of Labor on Hawaii,” in Bulletin of the Bureau of Labor (Washington, D.C., 1906), vol. 13, pp. 441–47.Google Scholar

⁶⁰ That such an outcome is true of the material consumption by exslaves in the United States remains one of the unsettled issues in the recent controversies about the economics of the postbellum South. See, for example, the estimates in Seagrave, Charles Edwin, “The Southern Negro Agricultural Worker: 1850–1870” (Ph.D. diss., Stanford University, 1971);Google Scholar and the discussion of its use by Fogel and Engerman in Time on the Cross, vol. 2, pp. 160–61;Google Scholar and in David, Paul et al. , Reckoning With Slavery (New York, 1976), pp. 224–25. It should be noted that the existence of possible differentials among rewards from different types of labor after emancipation is an issue separable from the question of whether emancipation led to higher or lower levels of material income for the exslaves. While similar questions of alternative incomes from plantation work as opposed to small-scale farming arose throughout the West Indies, it is still not possible from the presently available data to make the appropriate comparisons.Google Scholar

⁶¹ See, for example, the discussion in Hagelberg, , The Caribbean Sugar Industries, pp. 76–85, which points to a number of critical factors in trying to compare the private (and social) costs of production on large and small farms. In most earlier discussions it was believed that small farms were less productive in terms of output per acre, but that these developed because mill operators would otherwise have been unable to obtain adequate quantities of cane. See, for exampleGoogle Scholar, Moynagh, , Brown or White?, p. 96, on Fiji; and Coombes, The Evolution of Sugarcane Culture in Mauritius, pp. 80 and 86, on Mauritius. In 1913–1914 in Puerto Rico output of cane per acre on estates was in excess of that on land harvested by colonos, although there was little difference in the estimates of the cost per ton to the mill. See Dept. of CommerceGoogle Scholar, The Cane Sugar Industry, pp. 252–55. The Cuban pattern was more mixed, the report concluding that on plantations “the increased tonnage of cane per acre was not sufficient to outweigh the increased labor and material costs” (pp. 374–87), although the range of variation by region and by establishment makes a clear conclusion uncertain. These differences in productivity per acre are, of course, consistent with either (or both) lower labor input (in terms of time or intensity) or lower compensation on smaller farms.Google Scholar