Map 4 West Africa
There were several reasons why Abidjan and Côte d’Ivoire should have become the focus of West Africa’s HIV-1 epidemic. Neglected until late in the colonial period but endowed with vast areas of virgin tropical forest, Côte d’Ivoire experienced rapid development during the first two decades of independence, with a 6.8 per cent annual growth rate of real Gross Domestic Product between 1965 and 1980.15 Sparsely populated, its prosperity attracted immigrants both from economically faltering neighbours like Ghana and from the poorer savanna countries to the north. By the late 1980s some two million migrants from Burkina, over one million from Mali, and large numbers from Niger were present in Côte d’Ivoire at any time. Although many migrants worked in agriculture, over half lived in cities, especially in Abidjan, whose development as a major port increased its population between 1955 and 1984 from 120,000 to nearly 1,800,000. In 1975 some 40 per cent were non-Ivoirian immigrants. In older West African cities the control of retail trade by women fostered a rough equality of numbers between the sexes, but Abidjan, alone in West Africa, had the large male majority among adults that in East African cities like Nairobi led to highly commercialised sex, although in Abidjan it led also to more sophisticated forms of courtesanship, owing to the greater economic independence of women in West Africa and the region’s less constrained sexual traditions.16 Like Nairobi, Abidjan was a primate city on which the whole of Côte d’Ivoire’s excellent transport system focused. And as Vinh-Kim Nguyen has shown,17 two other features of Abidjan helped to make it an epicentre of HIV infection. One was an aspiration to modernity that bred individualistic choice, extreme differences of wealth, sexual adventurism – the median age of sexual debut was fifteen18 – and complex, disassortative networks through which HIV could pass. In 1994, 51 per cent of Abidjan’s men aged 20–24 said they had casual sex and 56 per cent never used a condom.19 The other circumstance favouring an epidemic was the economic crisis that struck Côte d’Ivoire during the 1980s as the world economy faltered and the easy growth opportunities of the 1970s were exhausted. This bred unemployment, sexual commercialisation, weakened health services, and resort to Abidjan’s 800 informal dispensaries ‘that sprout like mushrooms after rain’.20
When HIV-1 prevalence was first measured in Abidjan in 1985, the city was on the verge of an epidemic more explosive than those in Kinshasa or even New York, with an annual incidence of new infections of over 3 per cent in 1989.21 The core were the city’s sex workers and their male clients. Between 1986 and 1993 HIV prevalence among sex workers rose from 38 to 86 per cent; at the latter date 50 per cent had HIV-1, 2 per cent HIV-2, and 34 per cent both. Studies showed that contact with sex workers was the chief risk factor for men, largely explaining why in 1988 men outnumbered women by nearly five to one among HIV-positive patients admitted to city hospitals and why 83 per cent of the 24,735 people estimated to have died of Aids-related diseases in the city between 1986 and 1992 were men. Deaths were most common among informal sector workers in the older working-class quarters.22 By 1991, however, as the Minister of Health put it, the epidemic ‘is in the process of passing from populations at risk to the general population,’ as HIV-positive men infected their wives and other partners, creating a second peak of incidence. By 1993, the ratio of men to women infected had fallen to less than two to one.23 Antenatal prevalence rose between 1986 and 1989 from 3.3 to 9 per cent. During the mid 1990s it fluctuated around 15 per cent. As in southern Africa, good transport and high levels of mobility ensured an unusually narrow difference between urban and rural prevalences. In 1994 an estimated 41 per cent of all West Africa’s Aids cases were in Côte d’Ivoire.24
Abidjan was not only the place where HIV-1 and HIV-2 met, it was also the epicentre of infection for the entire eastern half of West Africa. This infection spread along two routes. One was the network of migrant sex workers who left their rural homes for a few years to work in the cities of neighbouring countries, seeking to bring home enough to set up a small business or finance their siblings’ schooling, without revealing their occupation to their families or potential future husbands. Like sex workers everywhere in the continent, these women were invariably blamed for expanding the epidemic, although almost all must themselves have contracted the disease from infected men resident in the towns where they came to work. West African sex workers were extraordinarily mobile. Of those attending a clinic in Abidjan in 1992, 82 per cent were from Ghana, 9 per cent from Côte d’Ivoire, and 2 per cent from Nigeria, but by 1998 only 9 per cent were from Ghana, 29 per cent from Côte d’Ivoire, and 56 per cent from Nigeria. Recovery in the Ghanaian economy and recession in Côte d’Ivoire and Nigeria probably shared the explanation with numerous Aids deaths among Ghanaian women and violence towards the Ghanaian community in Abidjan following a soccer match in 1993.25
Ghana was the first country to which Abidjan’s epidemic spread. Testing facilities became available there late in 1985 and were immediately deployed on sex workers. Of those tested in Accra early in 1986, only 5 of 236 were found HIV-positive, but when attention switched to women returning from Abidjan, 74 of 151 were found infected and many already gravely ill. At the end of 1987 the doctor in charge reported that Ghana had 276 known HIV cases, of whom 242 were women, 199 were sex workers returned from Côte d’Ivoire, and 145 came from Ghana’s Eastern Region, where the patrilineal Krobo people allowed women no rights over land and young women had long been engaged in commercial sex. ‘There is no work here,’ a woman from the area explained at that time. ‘In Abidjan I can earn 10,000 CFAs a day. . . . I have about 12 men a day. Since I heard about AIDS I always make them use condoms . . . I don’t know anyone who has it.’26 Although Ghanaians habitually blamed HIV on these women, it was plainly an oversimplification, for they had been singled out for testing and their predominance among those with HIV demonstrated that they had seldom transmitted the virus, which many were probably too sick to do. Transmission was clearly more diffuse. Nevertheless, by 2001, as national adult prevalence hovered around 3 per cent, Eastern Region was still the most heavily infected area and commercial sex was still central to the epidemic. HIV prevalence in Accra at that time was 5.9 per cent among men who bought sex and 0.5 per cent among those who did not. Among men aged 15–19, 84 per cent of cases were attributable to commercial sex.27
This combination of relatively low general prevalence and high infection rates among mobile sex workers and their clients was widespread within the region of West Africa focused around Abidjan. In Benin, for example, HIV prevalence among pregnant women in Cotonou rose slowly from 0.4 per cent in 1990 to 3.4 per cent in 1997–8, while prevalence among the city’s commercial sex workers rose from 3.3 per cent in 1986 to 58.0 per cent in 1997–8. It was calculated in the early 2000s that 76 per cent of male HIV infection in the city was contracted through commercial sex. Benin was unusual in that HIV prevalence in the general population was higher in some provinces than in the capital city, partly because commercial sex, a long-established practice there, was also widely dispersed, with a close correlation between infection in sex workers and in the general public.28 The remarkable point, as in Ghana, was that high infection among commercial sex workers did not precipitate the explosive epidemic seen in Kigali, Nairobi, and Abidjan. One reason was probably the equal gender balance in West African cities other than Abidjan. Another was that condoms had come to be quite widely used in commercial sex: by 54 per cent of clients in Cotonou in 1997–8, so they claimed, and by 90 per cent in Accra in 2001. In Cotonou the age at first sex was relatively high and women in the general population reported few sexual partners. Most important, perhaps, were the two contrasts emphasised by a study in 1997–8 that compared Cotonou and Yaounde in western Africa with Kisumu and Ndola in the east: the high levels of male circumcision in West African cities (almost 100 per cent in Cotonou) and the relatively low levels of HSV-2 in the general population (12 per cent among men and 30 per cent among women in Cotonou).29
Away