Nobody can count how many Haitians are online

Around 2003 there was no agreed figure for how many Haitians used the internet. The CIA world factbook reported thirty thousand users inside Haiti. Henri Deschamps, who ran the Haiti Global Village website, estimated four hundred thousand in the diaspora and at least a hundred thousand inside the country, a number he derived from monthly traffic to his own site. That site drew about five hundred thousand visits a month, roughly a hundred thousand of them from inside Haiti, and one person visiting repeatedly counts many times over in such a method. For the diaspora as a whole no other estimate existed. The gap matters because the same period saw the Haitian state redefine its emigrants as a tenth department alongside the nine geographic departments at home, and because every claim made about a diaspora public depends on how many people are in it.

On 5 February 2004 the FLRN seized Gonaives, the town where independence had been proclaimed two hundred years and one month before. Cap-Haitien fell on 22 February. By the end of the month the rebels stood at the edge of Port-au-Prince. Aristide resigned on 29 February and left the country, saying afterward that the American ambassador James Foley had forced the resignation and that he had been carried off against his will. Washington called it a rescue. Supreme Court Justice Boniface Alexandre became interim president and invited American, French and Canadian troops in to hold the capital. In March the United Nations authorised MINUSTAH, a peacekeeping force under Brazilian command that stayed thirteen years. Aristide settled in Pretoria. Whether his removal corrected a democracy or interrupted one has never been agreed, and the argument has the shape of most arguments about Haitian sovereignty, with Haitians on both sides of it and foreign governments holding the aircraft.

The Aristide administration collapsed on 29 February 2004, the culmination of a four-year campaign that intensified after the contested legislative elections of May 2000. Smith argues CARICOM misread the situation badly, partly because it had been the only foreign presence at the internationally boycotted 2000 elections, which opponents mistook for an endorsement and which Colin Granderson said did tremendous damage to the organisation's credibility. After the removal CARICOM took firm positions on Aristide's Jamaican exile from March to June and on the interim government, and objected that the appointment of Gérard Latortue as interim prime minister had been initiated by Washington without regional consultation. Its own decision to suspend Haiti in March 2004 has not helped, Smith writes, since the isolation coincides with the damage of tropical storm Jeanne in Gonaïves and deepens Haitian cynicism about the region. Where this account departs from the standard one: Smith argues that CARICOM's suspension of Haiti after the coup was counterproductive and that a hard line endangers future relations, a position he notes was already contested among heads of government.

The debt began with loans taken in the 1960s and 1970s for projects that never paid, and compounded when governments borrowed again to service what they already owed. Export prices fell, terms of trade worsened, and states lost the revenue to run themselves. Africa's total debt went from 9 billion dollars in 1970 to 321 billion by 1997, and interest on unpaid arrears took a growing share of national output that would otherwise have gone to schools and clinics. Western aid averaging 12 billion dollars a year through the 1990s was a stopgap, and it gave the donors an unprecedented voice in the internal policy of sovereign countries. The money came with conditions attached: liberalise, privatise, cut state spending on health and education. The campaign to cancel the debt was the newest in a line of public causes that began with the movement to abolish the slave trade, and like the missionary appeals of the nineteenth century it treated the continent as a condition to be corrected.