Twenty ports handled ninety three percent of the slave trade

Of nearly thirty five thousand documented slaving voyages, twenty ports organised ninety three percent. The database records departures from a hundred and eighty eight places around the Atlantic basin, and the concentration in so few of them is the striking result. Rio de Janeiro and Salvador da Bahia stood at the top of the list. The ranking moved over time, with Lisbon and London prominent early and Rio and Havana taking the lead in the nineteenth century as the business shifted to the American side of the ocean. Between them the top twenty carried off more than 8.3 million people. Concentration on that scale means the trade depended on a small number of places where capital, insurance, shipwrights, credit and captains were all available at once. It also explains why abolition worked once it came. A traffic running through twenty harbours can be strangled by naval patrols and law in a way a dispersed one could not.

Nicolas de Ovando reached Hispaniola in 1502 as governor, bringing two thousand five hundred colonists, among them the first European families to settle anywhere in the Americas. Families change what a colony is. Single men on a gold expedition can leave. Households with children, tools and livestock intend to stay, and they need land, which has to come from somebody. Ovando's answer was policy. Rouse records that he set out to eliminate the independent caciques so the expanding settlements would be secure, and over the next two years he carried it through with a thoroughness that ended Taino political life on the island. This is the administration that fixed the shape of Spanish colonisation everywhere it went afterwards: a royal governor, a settler population holding land grants, a subject population assigned to work those grants, and the removal of any indigenous authority capable of saying no.

Anacaona ruled Xaragua in the southwest of Hispaniola, the country around what is now Leogane. She had been the wife of one cacique and the sister of another, and after both were destroyed she governed in her own name, known for the poetry and the ceremonial dances of her court. In 1503 Spanish soldiers came into her territory with an invitation to talk. She received them with a feast. They seized her, killed the chiefs who had come with her, and hanged her. Dubois offers the episode as the emblem of what happened to indigenous leadership under Spanish rule, and it earns the role. There was no battle. The Spanish worked the protocol of hospitality that Taino politics ran on, the same protocol Guacanagari had extended to Columbus eleven years earlier. Her name survives in Haiti as the memory of a sovereign the island lost.

The Code Noir was drafted under Jean-Baptiste Colbert, the minister who built Louis XIV's navy and his colonial administration, and it reads like the work of a capable bureaucrat. Owners had to be Roman Catholic. Every enslaved person had to be baptised. Sunday work was prohibited. Torture was prohibited. Masters owed their people food, clothing and care in sickness and in age. Anyone freed acquired the standing of a colonial subject. The same code expelled all Jews from French territory overseas. In Saint-Domingue the protective clauses were treated as advice. The planters there worked from an arithmetic later called the planter's policy: an adult brought from Africa cost less to replace than to keep alive, so the cheaper course was to exhaust people and buy more. France could hold up its statute book as evidence that it governed like a civilised power while men holding its commissions ran the harshest slave regime in the hemisphere. The distance between the printed article and the cane field is where colonial power actually operates.