A System Built Out of Ledgers

We know a great deal about the Atlantic slave trade because it was a business, and businesses keep accounts.

The Trans-Atlantic Slave Trade Database has assembled records for some 36,000 voyages: ports of departure and arrival, numbers embarked and disembarked, dates, owners, captains, and often the goods traded. It is one of the most complete quantitative pictures we have of any pre-modern activity, and it exists because the people running the system needed to know their margins.

The headline figures are these. Roughly 12.5 million people were embarked from Africa between the sixteenth century and the nineteenth. Around 10.7 million survived to disembark, meaning something in the region of 1.8 million died during the crossing: an overall mortality of about 15%, falling over time as voyages got faster and insurers got stricter. These figures exclude deaths during capture, deaths on the march to the coast, and deaths in the barracoons before embarkation, which were substantial and very poorly documented, as well as the very high mortality in the first years after arrival.

The second figure surprises almost everybody. Of those 10.7 million, fewer than 400,000 (under 4%) arrived in the territory that became the United States. Brazil received close to five million. The Caribbean received most of the rest.

Which raises the obvious question: why was the enslaved population of the United States nonetheless around four million by 1860, while the Caribbean, having received more than ten times as many people, held far fewer?

The answer is the grimmest fact in this chapter. Sugar plantations in the Caribbean and Brazil ran negative natural increase for over a century: more people died each year than were born, by a wide margin. The population existed only because it was continuously replaced by purchase. The North American plantation population, by contrast, grew by natural increase.

This is not a story about kindness. It is arithmetic about relative costs, and it is the system in one sentence: where replacement by purchase was cheaper than the food, rest and medical care that would let a population reproduce itself, owners bought replacements. Where the trade was more expensive or more distant, they did not. Everything else follows from that calculation, including why abolition of the trade, as distinct from abolition of slavery, was such an urgent question for the islands.

What made this system different from earlier slavery is worth stating precisely, because slavery in some form existed in most complex societies and the differences are what produced the specific legacy.

Three things. Status was hereditary and passed through the mother, so the condition was permanent across generations. It was progressively racialised in law, until slave and African became for legal purposes the same category, which no earlier system had systematically done. And its purpose was the production of export commodities (sugar, tobacco, coffee, cotton) for a world market, at industrial scale, on a schedule set by the crop.

That combination created something new: a commercial enterprise with an inherited caste attached, which is why its aftermath was racial in a way that the aftermath of Roman or Ottoman or West African slavery was not.

One more correction, and it is uncomfortable in both directions. Europeans very rarely went inland to capture people. They died of malaria and yellow fever at rates that made the interior effectively closed, and they bought at coastal forts and markets from African states and merchant networks who conducted the capture, the marches and the sale.

This does not distribute the moral responsibility evenly, and it is not offered as though it might. European demand created the market, sustained it for three centuries, and supplied the firearms that flowed the other way and fed the wars that produced captives. But the mechanism matters for understanding the system, and for understanding what it did to West and Central Africa: not simply a loss of population, though the loss was enormous and concentrated in young adults, but the reshaping of states around the business of raiding, and a currency of firearms that made the alternative to participating extremely dangerous.

That is the part the shipping records cannot show you. They are very good at tonnage, mortality, insurance and margins. They contain almost nothing about the people they counted, and nothing at all about what was happening two hundred miles inland to produce them. An archive built by a business answers the questions the business had.