201African Affairs, 105/419, 201–218 doi:10.1093/afraf/adi104© The Author [2006]. Published by Oxford University Press on behalf of Royal African Society. All rights reservedAdvance Access Publication 18 January 2006AIDS, SECURITY AND THE MILITARY IN AFRICA: A SOBER APPRAISALALAN WHITESIDE, ALEX DE WAAL AND TSADKAN GEBRE-TENSAEABSTRACTThis article examines four accepted wisdoms about HIV/AIDS and Africanarmies and in each case concludes that substantial revision is necessary inthe light of emerging evidence. First, it appears that military populations donot necessarily have a higher prevalence of HIV than civilian populations.HIV levels in armies depend on many factors including the demographics ofthe army, its pattern of deployment, the nature and stage of the epidemic inthe country concerned, and the measures taken to control the disease by themilitary authorities. Second, although the epidemic has the potential toundermine the functioning of national militaries, and may have done so inisolated instances, armies in general are well placed to withstand the threat.Third, evidence that war contributes to the spread of the virus is meagreand suggests that we should be concerned primarily with specific risks thatconflict may entail including population mobility and changing sexual net-works. Lastly, the hypothesis that AIDS has the potential to disruptnational, regional, and international security remains speculative.THERE ARE MANY ACCEPTED WISDOMS ABOUT AIDS, its impact on securityand on militaries and other uniformed forces. These include the following:1. Military populations have a higher prevalence of HIV than civilianpopulations, and this is assumed to be due to the culture and behav-iour of soldiers.2. The HIV/AIDS epidemic is endangering the functioning of nationalmilitaries.3. War and conflict contribute to the spread of the virus, in partbecause of sexual violence during war.4. AIDS has the potential to disrupt national, regional and internationalsecurity.This article takes a hard look at these four shibboleths. We base our discus-sion on some new evidence contained in many recent papers, our ownAlex de Waal is a director of Justice Africa, Research Fellow, Global Equity Initiative,Harvard University, and program director at the Social Science Research Council. AlanWhiteside is the director of the Health Economics and HIV/AIDS Research Division at theUniversity of KwaZulu-Natal. Lt-Gen (retd) Tsadkan Gebre-Tensae is director of the Centrefor Policy Research and Dialogue, Addis Ababa.202 AFRICAN AFFAIRSexperience in the field, as well as our growing disquiet over the lack of criticalanalysis. We conclude that AIDS is indeed an issue for the military, but thereneeds to be a more nuanced view. We note opportunities for addressing theissue and the problems it does indeed cause. Similarly, AIDS is one securityissue among many facing poor countries hard hit by the epidemic. This alsoneeds a careful approach, and we note that there may be important AIDS-security feedback loops that have not been properly or fully understood. Weconclude by pointing to many ways in which this agenda can be taken forward.Do soldiers have higher HIV prevalence?Data on HIV prevalence in the military are notable for their scarcity. Thereare very few published studies available, and where there are data, there is agreat reluctance to release them as they are perceived as confidential or sensi-tive. Much of the literature relies on an international survey conducted by theCivil–Military Alliance (CMA) to combat HIV/AIDS in the early 1990s,which recorded some alarming prevalence figures in African militaries (mostlyreliable, some — like Angola — subsequently shown to have been incorrect).1In Cameroon, HIV prevalence in the military in 1993 was 6.2 percent com-pared to 2 percent in the general population. In Côte d’Ivoire in 1994, 12percent of recruits were infected, while in the Central African Republic, prev-alence was 22 percent among those entering military service.2 This studycontributed to the claim, still made by UNAIDS, that ‘in peacetime, STI[sexually transmitted infection] rates among armed forces are generally twoto five times higher than in comparable civilian populations. The differencecan be even greater in times of conflict.’3 While there was some evidence forthis (with respect to HIV) in the early 1990s, we will argue that it was uncriti-cally accepted and repeated and is in fact demonstrably incorrect today.We will examine three issues: prevalence among new recruits, prevalenceacross the military by age group, and lastly ask how the military fares underdifferent stages and kinds of epidemic.HIV prevalence among recruitsIt is our contention that HIV prevalence among new recruits is low andprobably even lower than in civilians of the same age. This is based on1. Roger Yeager, Craig Hendrix, and Stuart Kingma, ‘International military Human Immuno-deficiency Virus/Acquired Immunodeficiency syndrome policies and programs: strengths andlimitations in current practice’, Military Medicine 165, 2 (2000), pp. 87–92.2. S. Kingma, ‘AIDS prevention in military populations: learning the lessons of history’,International AIDS Society Newsletter, 4, March 1996, pp. 9–11.3. UNAIDS, ‘AIDS and the military: UNAIDS point of view’, UNAIDS Best Practice Collection,May 1998 (http://www.unaids.org/html/pub/publications/irc-pub05/militarypv_en_pdf.pdf,9 January, 2005).AIDS, SECURITY AND THE MILITARY IN AFRICA 203some limited data from the military, extrapolations from other data, andepidemiological logic. There are two reasons for this. First, the greatmajority of army recruits are young men in the age range 17–22. Thisdemographic category has lower HIV prevalence than young women of thesame age. This is demonstrated by data from population-based surveys. Anexample is South Africa’s 2004 Reproductive Health Research Unit(RHRU) survey of 15- to 24-year-olds.4 Among males aged 15–19, HIVprevalence was 2.5 percent; for women of the same age, it was 7.3 percent.In the 20- to 24-year-old age group, it was 7.3 percent for males and24.5 percent for females. Second, many militaries screen recruits, andthose who are found to be HIV positive will not be recruited. Informal con-sultation tells us that every armed force in Africa that is able to screen doesso and aims to exclude those who are found to be infected. Most do notpublicly admit to doing so; indeed in some countries, this