The head of the electoral council flees after a palace summons

Leon Manus, president of Haiti's provisional electoral council, had at first defended the way the senate votes of May 2000 were counted. After studying the law he concluded the method was illegal and that only five senators had genuinely been elected in the first round rather than the nineteen the council claimed. He refused to certify the result. He was summoned to the national palace, where by his own account President Rene Preval threatened him in person and Jean Bertrand Aristide did the same by telephone, giving him an ultimatum and warning that the capital and the provincial cities would be engulfed in fire and blood unless the results were announced at once. Manus left secretly for the Dominican Republic and then the United States with American diplomatic help. From New York he wrote to the president of the senate listing stolen ballot boxes, destroyed tally sheets, and police who watched torn ballots cover the streets of Port-au-Prince.

By the time Parham finished her fieldwork, an email list known as the Corbett List had been running for eight years and included some of the most influential leaders of the Haitian diaspora. Human rights advocates used it to circulate petitions and to gather support for changes in Haiti. Word of the list travelled along ties that already joined advocacy workers, academics and missionaries, and those ties held it together. She sets it beside a website forum that produced an organisation which folded inside a year. Her conclusion is that online space for a scattered nation serves expression well and organising badly, unless the people in it are already connected somewhere on the ground. About half the respondents in one Haitian forum she studied worked with advocacy organisations, and thirty of forty Haitian community groups she catalogued had an email address, a website, or both.

Seymour Drescher published a study of Jews and New Christians in the Atlantic slave trade in 2001 and attached a figure to the question. It is unlikely, he wrote, that more than a fraction of one percent of the twelve million Africans enslaved and relayed across the ocean were bought or sold by Jewish merchants even once. At no point along the chain of that trade were Jews numerous, wealthy or powerful enough to affect its structure or its flow, or to lessen the suffering of the people carried in it. The number matters because a very large claim had been made in the opposite direction and had spread widely. One percent of twelve million would be a hundred and twenty thousand people, and the estimate is that the real share fell well below even that. Schorsch reaches the same conclusion from a different direction, finding Jewish slave owning and slave trading minimal in the Atlantic and the Mediterranean alike.

After independence most African governments put money into schools, primary enrolment rose sharply, and there was deliberate effort to get girls into classrooms. The gains were uneven and then partly reversed. Adult illiteracy south of the Sahara is higher than in any other region of the world, the economic crisis that began in the 1970s cut state education budgets, and the gap between boys and girls is widest in the countryside. Health followed a similar curve. Malaria remains the heaviest disease burden on the continent and sleeping sickness was spreading again by the 1990s. More than two thirds of the world's HIV cases are African, with about 34 million people infected since the early 1980s and over 11 million dead, concentrated in the east and the south. Illness tracks poverty, thin food, bad sanitation and dirty water. Clinics decayed and trained staff left for better pay abroad. Vaccination programmes still cut infant mortality sharply after the 1960s and life expectancy rose, which shows what small sums can buy.