THE DEMOGRAPHY OF SLAVERY

The winding road to Codrington College climbs the steep cane-covered hills to the oldest Anglican seminary in the New World. Within earshot of the eighteenth-century college buildings, Atlantic waves crash into Conset Bay on Barbados’ rugged eastern coastline. Christopher Codrington bequeathed his Barbadian estates in 1710 to the Society for the Propagation of the Gospel. At his death, he empowered the Society to establish a missionary seminary on the Codrington estates, and to manage his large sugar plantations and the 300 enslaved people upon them. Codrington’s wishes, however, were only partly fulfilled. While the buildings rose, the slave population of the Codrington plantations underwent massive decline. Low birth rates, infant and adult death, and widespread disease so debilitated the slave population that, despite purchasing 450 additional slaves from 1712 to 1761, the total number of slaves on the Codrington plantations numbered 190 in 1760. Every year, the Society bought dozens of Africans from slave traders in Bridgetown, but the number of births and new acquisitions failed to keep pace with death on the plantation. Young and old, African and creole succumbed to tropical disease and to the physical rigors of New World slavery (Bennett, 1958: 1-3, 52, 61). The experience of the enslaved population on the Codrington plantations was

replicated on thousands of estates throughout the Caribbean from the seventeenth to nineteenth centuries. Death defined plantation America, and its omnipresence encouraged slaveholders to be callous about life. Planters and slaves lived on the “threshold of death” and, as Vincent Brown observes, “in one of history’s greatest episodes of creative destruction”, the vibrant, profitable economies of New World slavery ultimately “consumed its inhabitants”. On the Codrington estates, as elsewhere in tropical America, frequent death winnowed the slave population, particularly among the newly arrived, but the slaveholders’ predilection for males also ensured an unnaturally small population of women. There were accordingly fewer babies, and those infants faced a range of infections that killed regularly. Children who survived their infant years embarked on a life where the slaveholder’s lash drove them to toil long hours in desolate conditions. That labor was undertaken with a diet inadequate in nutrients, minerals, and calories to ward off infection or to fuel the basic energy requirements of cane work. Death, disease, and destruction dictated life, and only frequent acquisitions from the Atlantic slave trade sustained the population. With mortality ranging between 5 and 7.5 per cent on mideighteenth-century sugar estates, Barbadian planter Henry Drax calculated that to maintain an estate of 200 slaves required the acquisition of between 10 and 15 new

hands annually. In no small measure, it was the persistent annual and daily fatalities within the slave populations that spurred the volume and longevity of the Atlantic slave trade (Brown, 2008: 57-58; Amussen, 2007: 96). By 1761, the Society recognized the crippling financial cost of replacing the slave

population with new acquisitions. Having purchased hundreds of enslaved Africans, the slave population was more than a third smaller than that bequeathed by Codrington. As Archbishop Thomas Secker observed, “I have long wondered and lamented that the Negroes in our plantations decrease, & new Supplies become necessary continually. Surely this proceeds, from some Defect, both of Humanity, & even of good policy.” Secker’s point was well made; the Society’s policy of acquiring two males for every female was culpable for the slave population’s incapacity to grow by natural means, and the reliance on “unseasoned” African imports ensured that, on average, one in every two new recruits died within the first three years of Caribbean slavery. Slaveholders recognized the expense of child-rearing (the reduction in physical labor when relatively more women were acquired, the loss of field-labor by mothers, and the cost of raising children to maturity, many of whom would die in childhood), though once the price of slaves began to rise sharply in the 1760s, planters saw a potential cost advantage in encouraging child birth among the enslaved rather than buying new African slaves. From 1780 to 1807, the cost of purchasing a male slave within the British West Indies rose by 141 per cent, almost three times the relative increase in the price of the primary export, raw sugar. Planters responded in kind, ameliorating conditions and encouraging population growth. As the Surveyor General William Senhouse observed of the planters’ new-found enlightened self-interest, “the owner’s humanity will in a few years be amply rewarded by a valuable increase in his property”. Senhouse proved prescient in his commentary. Slightly improved food, medical treatment, and care of the new-born on the Codrington estates soon paid dividends as the number of infant mortalities decreased and the gender balance of slave population gradually stabilized. By 1792 women were in the majority, and by the turn of the century births surpassed deaths. After 70 years of Society management, the slave population on the Codrington estates was finally growing by natural means (Bennett, 1958: 89, 96; Eltis et al., 2005: 680; Handler and Lange, 1978: 84). The demographic transition experienced on the Codrington plantations was early

within the context of West Indian slavery. It was not, however, exceptional. Across the Caribbean, though at different times and rates, population dynamics shifted from low fertility and high mortality toward relative demographic stability from the late eighteenth to mid-nineteenth centuries. At the same time, the slave populations became increasingly creolized or American-born. This in turn had a positive long-term effect on birth rates, as the creole born gradually equalized the imbalanced sex ratios of the eighteenth century. Since American-born slaves tended to have a relatively stronger immunological response to New World tropical diseases than newly arrived Africans, mortality rates also began to fall. In Barbados, this demographic transition was well advanced by 1800, but the abolition of British slave trading from 1808 forced West Indian planters to pay greater attention to child-births and natural increase. Progress, however, advanced unevenly and sporadically. In fact, between 1807 and 1834, the slave population of the British West Indies declined from 770,000 to 665,000. The expansion of cane sugar plantation agriculture through the first three decades of the nineteenth century in Jamaica, and particularly in the newly acquired cane colonies of Trinidad,

Tobago, and Demerara-Essequibo, hampered the prospects of demographic growth in the British Caribbean. No other crop was produced under such physically exhausting conditions as sugar. Long shifts, gang labor, and night work characterized life on a sugar plantation and distinguished it from the relatively less demanding regimes associated with coffee, cotton, pimento, cocoa, or food provisioning. Although the rate of natural increase rose on those estates where material or laboring conditions improved, the erosive effect of high adult and infant mortality on the cane estates (particularly in the “new” sugar colonies) undermined the positive effect of amelioration long into the nineteenth century. Thus, despite the introduction of pro-natalist policies and the increased number of women within the slave population, it was not until the 1840s (a full decade following emancipation) that the black population of Jamaica finally began to show positive growth. The demographic histories of Jamaica and Barbados differed, but the gradual rise in the number of women and creole born within the slave labor forces created a pattern of marginal (though positive) natural growth for the nineteenth-century slave populations (Higman, 1984: 72-78, 307-78; 1995: 134-35). If gradual demographic stability characterized the last decades of British West Indian

slavery, the most striking aspect of the eighteenth-century slave systems was their failure to create self-sustaining populations. The combination of low natality and high mortality bequeathed a devastating demographic legacy throughout tropical America. Almost one in four Africans sold into slavery, nearly four million people, were destined for Brazil from 1700-1850, yet Brazilian planters counted just 1.5 million slaves in the 1872 census. Cuba, an island economy that converted to cane sugar and coffee monoculture in the 1790s, imported some 780,000 slaves between 1790 and 1867, but in 1862 just 370,000 slaves resided on the island. Neighboring Jamaica and Saint Domingue fared no better. Jamaica imported over one million slaves, mostly between 1700 and 1807, but by 1832 just 312,876 were registered on the colonial returns. In Saint Domingue, almost 800,000 slaves were imported from 1680 to 1777, though the slave population numbered no more than 290,000 by the end of that period (Bergad, 2007: 96-97; Bergad et al., 1995: 38; Higman, 1995: 54; Hall, 1971: 14). The United States, however, stood apart. Before Congress prohibited American

involvement in the Atlantic slave trade in 1808, some 360,000 slaves were imported to the United States and its colonial antecedents. Less than 4 per cent of the entire volume of the Atlantic slave trade entered the United States, but by 1860, 3.95 million slaves lived in the American South. The rate of population growth was exceptional. In 1810, some 1.1 million slaves resided on American soil, but by 1825, the US was the largest slaveholding nation on Earth. Slavery in the American South was exceptional on several levels: the nineteenth-century slave population grew swiftly (without reliance on the international slave trade), practically every slave in 1860 was creole born, and the growth of US slavery underpinned the rapid emergence of American cotton, cultivated by native-born African-American slaves from the Carolinas to Texas. It was, in turn, the value of their cotton and slaves that made American slaveholders stalwart defenders of chattel bondage. The American slave population boomed because of even sex ratios, comparatively low infant and adult mortality, a substantially higher birth rate than death rate, and high levels of per capita food consumption (a scenario shared with the British Bahamas, where the slave population also expanded naturally). Conditions elsewhere in the tropics nonetheless remained dismal, irrespective of the

pro-natalist policies exhibited in the British West Indies and in Cuba (Eltis, 2001: 45; Fogel, 1989: 123-32). As Charles Pennell, the British Consul in Salvador, Bahia noted in 1827, “the annual

mortality on many sugar plantations is so great that unless their numbers are augmented from abroad the whole slave population would become extinct in the course of about twenty years”. The proprietors, Pennell concluded, “act on the calculation that it is cheaper to buy male slaves than to raise Negro children”. By 1832, the rate of decline within the Brazilian slave population stood at 5 per cent per year, ensuring that if no additional purchases were made, the slave force would be reduced to half its size in seven years. The factors eroding the slave population’s capacity to expand were not dissimilar to those on the Codrington estates a century earlier. There was a high proportion of males to females, exacerbated by male-orientated gender ratios within the international and inter-regional slave trade. In addition, crude birth rates hovered at about 34 per 1000, while death rates of between 47 and 65 per 1000 were not uncommon in the plantation belts of Bahia, a prime location for newly arrived Africans for the first half of the nineteenth century. Infant mortality and the death of recently imported Africans, furthermore, contributed to the demographic phenomenon Pennell described. Indeed, so high were the prospects of morbidity among Brazilian slaves that life expectancy at birth was 18.3 years for males, almost half of that of US slaves (35.5) in 1850 (Schwartz, 1985: 365-71; Eltis and Engerman, 1993: 310-12; Graham, 2004: 299).