Crops moved east and herds moved west across Africa's own networks

The origins of Africa's thirteen exports are scattered across the continent, and the scatter is itself the finding. By 5000 to 3000 BCE Africa was already crossed by regional connections that shaped what people grew and what they traded. Crops first domesticated in West Africa moved east along the Sudan belt, and in the easternmost stretch they joined the outward flow into Eurasia. Traffic ran the other way at the same time. Word histories reconstructed from Niger-Congo, Nilo-Saharan and Afrasian languages show goats and cattle spreading from the eastern Sudan far across West Africa in the fifth and fourth millennia BCE, and the donkey, domesticated in the far northeast, may have arrived in the western and central Sudan just as early. Crops going one way, animals going the other, each transfer altering the economy that took it in. This is a continent generating its own momentum thousands of years before anyone in Europe had heard of it.

Barrera-Mordan, in the Dominican Republic, is one of the sites that forced archaeologists to move the date of human arrival in the northern Caribbean. The material there is a stone-working complex built on flaked tools struck from cores, and it dates to around 4000 BCE. The people who made it belong to what is labelled the Lithic age. They had no pottery, and they should not be confused with the ceramic-producing groups who reached the islands thousands of years afterwards and are far better known. What the assemblage records is a specialised adaptation, worked out for a set of islands with particular stone, particular coastlines and particular food. Sites like this are how the initial migration routes from the surrounding mainland get mapped, because the tool types are the trail. Hispaniola, the island Haiti shares, was inhabited about six thousand years ago by people who knew exactly what they were doing with rock.

Hor-Aha, first or second king of the First Dynasty, sent his forces south and destroyed what remained of the Qustul state. The southern frontier of the Old Kingdom was fixed by that campaign and held for centuries afterwards. He did it wearing the White Crown that a Qustul ruler had worn before him, attended by the Horus falcon that had attended a Nubian king, carrying an idea of sacred kingship developed upriver in the country he was burning. The Old Kingdom that followed went on displaying its inheritance in plain sight. Its political institutions and its ritual observances kept the marks of the wider Middle Nile world they came out of, in the ceremonies around the king, in the handling of royal death, and in the claim that a ruler holds divine authority. Conquest ended the kingdom. It left the kingdom intact inside the conqueror's own practice, where it is still legible.

The First Dynasty royal tombs at Abydos hold more than kings. Around them lie the graves of retainers and high-status people put there to accompany the ruler and look after him afterwards. Hor-Aha has thirty-six such burials. His successor Djer has as many as three hundred and eighteen. Djet has one hundred and fifty-four, and Den, the fifth ruler, has one hundred and thirty-five. The practice was neither confined to Egypt nor started there. It appears equally early further south, it was still going in the second millennium BCE at the height of the Kerma kingdom on the Upper Nubian Nile, and it recurs across a long line of Nilo-Saharan-speaking or Nilo-Saharan-influenced states in the Sudan belt, from pre-Christian Nubia to the Wagadu empire of the Ghana in the ninth century CE. Egypt is one branch of that tradition. It kept retainer burial for a few generations and then let it go.