Social Science & Medicine 70 (2010) 114–120

Contents lists available at ScienceDirect

Social Science & Medicine

journal homepage: www.elsevier.com/locate/socscimed

Reframing governance, security and conflict in the light of HIV/AIDS:

A synthesis of findings from the AIDS, security and conflict initiativeq
Alex de Waal*
Social Science Research Council, HIV/AIDS and Social Transformation, Emergencies and Humanitarian Action, One Pierrepont Plaza, Brooklyn, NY 11201, United States

a r t i c l e i n f o a b s t r a c t

Article history: This paper draws upon the findings of the AIDS, Security and Conflict Initiative (ASCI) to reach conclu-
Available online 8 October 2009 sions about the relationship between HIV/AIDS, security, conflict and governance, in the areas of HIV/
AIDS and state fragility, the reciprocal interactions between armed conflicts (including post-conflict
Keywords: transitions) and HIV/AIDS, and the impact of HIV/AIDS on uniformed services and their operational
HIV effectiveness. Gender issues cut across all elements of the research agenda. ASCI commissioned 29
Security
research projects across regions, disciplines and communities of practice.
Conflict
Governance
Over the last decade, approaches to HIV/AIDS as a security threat have altered dramatically, from the
early anticipation that the epidemic posed a threat to the basic functioning of states and security
institutions, to a more sanguine assessment that the impacts will be less severe than feared. ASCI finds
that governance outcomes have been shaped as much by the perception of HIV/AIDS as a security threat,
as the actual impacts of the epidemic.
ASCI research found that the current indices of fragility at country level did not demonstrate any
significant association with HIV, calling into question the models used for asserting such linkages.
However at local government level, appreciable impacts can be seen. Evidence from ASCI and elsewhere
indicates that conventional indicators of conflict, including the definition of when it ends, fail to capture
the social traumas associated with violent disruption and their implications for HIV. Policy frameworks
adopted for political and security reasons translate poorly into social and public health policies. Fears of
much-elevated HIV rates among soldiers with disastrous impacts on armies as institutions, have been
overstated. In mature epidemics, rates of infection among the military resemble those of the peer groups
within the general population. Military HIV/AIDS control policies follow a different and parallel paradigm
to national (civilian) policies, in which armies prioritize command responsibility and operational
effectiveness over individual rights. Law enforcement practices regarding criminalized and stigmatized
activities, such as injecting drug use and commercial sex work, are an important factor in shaping the
trajectory of HIV epidemics.
Ó 2009 Elsevier Ltd. All rights reserved.

Introduction

The AIDS, Security and Conflict Initiative (ASCI) was established

q The AIDS, Security and Conflict Initiative was set up in 2006 with the objective
of providing evidence that could inform policies relating HIV/AIDS, security and
conflict. Its co-chairs are Souleymane Mboup and Georg Frerks. The research and
administration was divided between the Social Science Research Council (New
York) and the Clingendael Institute (The Hague). The project was supported by the
governments of The Netherlands, Sweden, Australia and Canada and by UNAIDS,
the Rockefeller Brothers’ Fund, Justice Africa and the Health Economics and HIV/
AIDS Research Division at the University of KwaZulu Natal, South Africa. The author
wishes to thank the following individuals within the ASCI project: Tsadkan
Gebretensae, Manuel Carballo, Tony Barnett, Steven Schoofs, Georg Frerks, Dana
Huber, Manjari Mahajan, Selma Scheewe and especially Jennifer Klot, who provided
substantial critical input into this paper.
* Tel.: þ1 617 998 0162.
E-mail address: dewaal@ssrc.org
in 2005 in order to examine evidence for the implications of the
HIV/AIDS epidemic for a range of governance, security and conflict
concerns, and also how governance crisis affects the HIV/AIDS
epidemic. Its foci included state fragility, conflict and post-conflict
transitions, the uniformed services and peacekeeping operations,
and gender. The ASCI project, jointly led by the Social Science
Research Council (New York) and the Clingendael Institute (The
Hague), was set up in the wake of a number of worrying predictions
about how HIV/AIDS could strike at the foundations of governance
in hard-hit countries. At the conclusion of the project, it is clear that
the paradigm has shifted. Research evidence from ASCI and else-
where, combined with policymakers’ experience of managing the

0277-9536/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.socscimed.2009.09.031
 A. de Waal / Social Science & Medicine 70 (2010) 114–120
epidemic and its governance impacts, have led to revised (and less
disturbing) predictions, and a more nuanced appreciation of how
HIV/AIDS affects security, conflict and governance, and vice versa
(ASCI 2009).
The issue of HIV/AIDS and security has had a brief but dramatic
intellectual life. Ten years ago, the U.S. National Intelligence Council
(NIC) prepared its first-ever assessment of the security threat posed
by infectious diseases and singled out HIV/AIDS as the gravest such
threat. The report was published in January 2000 (NIC, 2000) at the
same time as the U.S. Ambassador to the United Nations, Richard
Holbrooke, sponsored a UN Security Council debate on HIV/AIDS,1
the first time the Council had ever discussed a disease. The NIC
report sounded the alarm: ‘the persistent infectious disease burden
is likely to aggravate and in some cases, may even provoke
economic decay, social fragmentation and political destabilization
of the hardest hit countries in the developing world.’ (NIC, 2000,
p. 10).
The NIC conclusion was political as well as scientific, and was
tracked by a parallel evolution of thinking in the UN and elsewhere.
Different security frameworks converged to draw the conclusion.
One was national security establishments’ traditional focus on
‘hard’ security issues such as the threat posed by nuclear-armed
adversaries. However, in a notable shortcoming, the U.S. Central
Intelligence Agency and the Department of Defense had failed to
foresee the collapse of the Soviet Union. The analysts who emerged
from this episode with credit were, unexpectedly, demographers.
During the 1980s, American demographers debated whether
evidence for a rising infant mortality rate in the USSR reflected
a real trend or not, and if it did, what this trend might portend
(Davis & Feshbach, 1980; Eberstadt, 1988). Those who argued that it
was a real health crisis which indicated a crisis of the Soviet system,
were proven right. The professional standing of demographers as
intelligence seers was reinforced by the findings the State Failure
Task Force (SFTF) (Esty, Goldstone, Gurr, Surko, & Unger, 1995),
which was set up following concerns that anarchy, as in Somalia,
Haiti or Yugoslavia, threatened American interests. It was an exercise
in quantitative political science, and when it processed its data to
generate the best models for predicting state failure, the SFTF found
three variables from among its list of 75 to be the most important:
openness to trade, democracy, and infant mortality (Esty et al.,1995).
Health and demographic indicators were validated.
The second strand was the paradigm of ‘human security’
(Commission on Human Security 2003; UNDP 1994). This emerged
as an American concern during the Clinton Administration, which
in 1996 reoriented intelligence policy towards a new order of
security threats including humanitarian crises, climate change and
emerging and re-emerging infectious diseases. The twin ‘hard’ and
‘soft’ approaches to security converged in the issue of HIV/AIDS: it
was a threat to human security which also posed dangers to mili-
taries and seemed to threaten state failure. Human security was
also promoted by the emergent multilateral approach to HIV/AIDS,
notably the human rights-focus of the World Health Organisation’s
Global Programme on AIDS headed by Jonathan Mann and subse-
quently UNAIDS under Peter Piot.
In 2002, a follow-up NIC estimate projected a ‘second wave’ of
epidemics in China, India and Russia (Eberstadt, 2002; NIC, 2002).
Already alarmed by the speed with which HIV had taken hold
among general populations in eastern and southern Africa, the NIC
projected that a host of other countries would also proceed towards
hyperendemic HIV. UNAIDS sounded similar warnings. These
1
UN Security Council. ‘The situation in Africa: the impact of AIDS on peace and
security in Africa,’ 10 January 2000. The Security Council subsequently passed
Resolution 1308 in July, focusing on HIV/AIDS and peacekeeping operations.
115
warnings helped create an intellectual climate in Washington DC in
which the President’s Emergency Plan for AIDS Relief (PEPFAR)
could be presented as simultaneously a philanthropic, diplomatic
and security initiative. Reflecting this, PEPFAR reports to the State
Department and not the US Agency for International Development.
In 2003, UN Secretary General Kofi Annan warned of the ‘triple
threat’ of HIV/AIDS, food insecurity and governance in southern
Africa and established the Commission on HIV/AIDS and Gover-
nance in Africa.
The NIC reports and UN fears generated publications projecting
the damage that HIV/AIDS would bring. At best these were
deductive extrapolations from models or from history (Bell,
Devarajan, & Gershbach, 2004; Cheek, 2001; De Waal, 2004; Price-
Smith & Daly, 2004) and at worst they were lists of hypothetical
outcomes strung together (International Crisis Group, 2001; Neil-
son, 2005). The rationale for ASCI was that if these alarming
scenarios were even partly correct, they warranted closer scrutiny,
and therefore a more solid empirical basis was needed both to
substantiate the links between HIV/AIDS and security and to
identify what measures might be effective in response.
Today, the consensus has shifted. The evidentiary base gener-
ated during the last few years does not support the more pessi-
mistic predictions. The findings of ASCI are summarized in this
paper. More importantly, the predicted disasters did not become
part of policymakers’ experience, and the more extreme forecasts
were gradually discounted. Scholarly publications and official
reports mark this shift (ASCI 2009; CHGA 2008; De Waal 2006; NIC
2008; Spiegel et al., 2007). The NIC’s 2008 report re-orients U.S.
concern to the economic impacts of disease and ill-health and
suggests that the governance impacts of health are ‘less pervasive’.
Most significantly, the 2008 NIC report concludes with a section
‘Health as Opportunity: A new look at a successful paradigm.’
Global health, it suggests, is a fruitful field for U.S. influence,
including effective engagement with rising powers, reconstruction
and stabilization, smoothing relations with adversaries, easing
north-south tensions, and advancing economic development. It
highlights ‘health diplomacy’, variously interpreted as using health
assistance to promote U.S. national interests and using U.S.
governmental budgets to promote global public health. This illus-
trates how the security response to HIV/AIDS has influenced the
governance of global public health.
The diversity of the posited linkages between HIV/AIDS and
conflict, security and governance (in both directions), the range of
the relevant kinds of evidence and analysis (from quantitative
multi-country comparisons to ethnographic study to operational
and policy research), and the multiple interests of the policy
community, led ASCI to sponsor a range of research projects with
heterodox research questions and methods. All research projects
were assessed for method, rigour and ethics by the coordinators of
the four research tracks. Resource constraints dictated that ASCI
focus on enhancing existing research projects by scholars in the
field, so that the research is investigative and illustrative rather
than exhaustive. Not every possible link between HIV/AIDS
and security has been chased down and there are significant
geographical gaps (for example central Asia). But the diverse
methods and cases chosen provide multiple perspectives on the
central issues, providing a multi-disciplinary overview. ASCI suc-
ceeded in tackling the principal theoretical and policy questions
that have driven concern over the last decade, and its findings pose
as many questions as they illuminate answers.
ASCI’s conclusions are in line with the changing consensus
revealed in recent reports on related topics (CHGA 2008; NIC 2008).
States and security institutions have proven resilient in the face of
the epidemic, but HIV/AIDS has important effects, usually specific
to sectors or localities, and many of them framed by gender
116 A. de Waal / Social Science & Medicine 70 (2010) 114–120
relations. Governance has shaped the epidemic, including through
sectoral policies, local government reform and law enforcement.
HIV/AIDS and state fragility
The findings of the ASCI research on HIV/AIDS and state fragility
are synthesized by Barnett (2009). These papers are thematically,
methodologically and geographically diverse. The case studies and
approaches were selected by Barnett to represent a broad and
illustrative range of issues and locations.
Barnett discounts any direct link between HIV/AIDS and state
fragility and instead searches for putative indirect linkages,
whereby for example the impacts of the epidemic may play out
through stresses on the labour force. One line of inquiry is large-n
quantitative cross-country comparisons. Two exercises were con-
ducted by Dutta and Barnett (2007) and Sato (2008) using a range
of governance indicator variables (as used by the OECD and World
Bank) to determine whether there is a measurable association
between HIV prevalence and state fragility. Governance in this
context includes measures of state capacity and the quality of
government. Dutta and Barnett compared ‘fragile’ and non-‘fragile’
countries and Sato asked whether differences among fragile states
could be accounted for by different levels of HIV prevalence.
Neither study found any such association, although they both hint
that a differently-framed inquiry might yield some links. For
example, Dutta and Barnett found that Tuberculosis was associated
with state weakness. The two studies used diverse datasets, indi-
cators and methods, giving further reason for confidence that the
negative finding is robust.
Several case studies underlined that national-level impacts have
not materialized. Feshbach (2008) investigated Russian demog-
raphy, focusing on the health of recruits into the army. He paints an
alarming picture of the health of the population and documents
how the military is struggling to find sufficient healthy young
recruits for its ranks. Among the health problems are drug addic-
tion (accounting for the ‘unfit’ diagnosis of 21,000 recruits), mental
disorders (130,000), Tuberculosis (3000) and HIV (2100). HIV/AIDS
is a serious second-order problem.
In south-east Asia and the Pacific, Altman (2008) and O’Keefe
(2008) caution against the view that HIV in the region will develop
into southern African-style generalized epidemics. O’Keefe chal-
lenges the ‘Africanization’ hypothesis for the region, arguing that
the epidemic has markedly different characteristics to Africa
(concentrated rather than generalized), the nature of states is
different, and the picture of a convergent crisis of HIV/AIDS and
governance in sub-Saharan Africa is also oversimplified.
The lack of association between HIV prevalence and state
fragility is an important negative finding. The ASCI studies examine
three possible explanations.
The first explanation is methodological: country-based datasets
used are at too high a level of aggregation. The limited longitudinal
datasets for HIV prevalence and the diversity of the countries
classified as ‘fragile’ mean that any associations are submerged.
This directs us towards sub-national impacts. It is plausible to argue
that the impacts of HIV/AIDS are felt primarily at the household
level, and from there the community and district levels, before any
national-level effects unfold. Pursuing this hypothesis, three
studies examined the impacts of HIV/AIDS on local government.
Chirambo and Steyn (2009) studied the impacts of HIV/AIDS on
local government in South Africa. They discovered increased
absenteeism and burn-out of councilors, higher costs and workload
consequent on more by-elections for vacant posts, and the chal-
lenge of maintaining an updated electoral roll with higher levels of
adult mortality. They found stigmatization of HIV/AIDS, with just
one councillor openly living with HIV from among a sample of
3895. They concluded that local government is showing ‘cracks’
that contribute to poor service delivery to the general public.
Jacob (2008) investigated the impact of HIV/AIDS in the Indian
states of Manipur and Nagaland, which have the highest prevalence
of HIV in the country. He found no evidence for governance impacts
of the epidemic. At most, his data contain inconclusive hints that
the epidemic, in conjunction with the insurgency and a crisis of
injecting drug use, may be contributing to a higher rate of attrition
among civil servants. Jacob counsels that even if HIV/AIDS were to
have an appreciable impact in these two states, India is so vast that
we can discount the chance that a local crisis would undermine
national stability.
In the Indonesia province of Papua (Irian Jaya), which has the
highest HIV prevalence in the country, Smith (2008) found no
indication that the epidemic was causing human resource losses
sufficient to undermine service delivery or local government
accountability. However, local government reforms were helping to
create localized environments of increased HIV risk. The creation of
new local administrative centres led to boom towns which attrac-
ted migrants and commercial sex workers, in a context of extreme
gender inequalities and high rates of sexual and gender-based
violence. Decentralization of government authority to provincial
level contributed to the misappropriation of HIV funds by local
elites to support electoral campaigns.
The hypothesis that the national level of aggregation fails to
capture important local-level impacts is confirmed.
A second hypothesis is that the relationship between HIV/AIDS
and governance is non-linear and certain combinations of factors
can combine to create HIV-related state fragility. An example
of a limiting case, a ‘perfect storm’ of these factors could be
Swaziland, a small country which has the world’s highest national
HIV prevalence alongside weak governance and other social and
economic stresses. Whiteside and Whalley (2007) describe the
Swazi epidemic as a new kind of humanitarian ‘emergency’, affec-
ting society over a generation. Naysmith, de Waal, and Whiteside
(2009) investigated whether the combination of hyperendemic HIV
and other stresses make the economic life of the country unsus-
tainable, concluding that it is plausible to argue that Swazi house-
holds have lost so much resilience on account of HIV/AIDS that they
are facing a long-term emergency including the threat of ‘new
variant famine’ (De Waal & Whiteside, 2003). This protracted crisis
has no clear exit and warrants new policy responses.
However, the Swaziland case is an extreme and limiting case
that does not readily transfer.2 Another country with hyperendemic
HIV, Botswana, shows governance resilience, most likely associated
with its wealth and commitment to a high quality of public
administration. Elsewhere, such as hard-hit localities in larger
countries including South Africa, Malawi and Tanzania, the local-
ized conjuncture of HIV/AIDS with other stresses has contributed to
vortices of crisis, but national governments have been less affected.
A third possible interpretation of the negative result focuses on
the opposite causal pathway – from governance to the socio-
economic factors that influence HIV transmission. State fragility is
a composite concept comprised of diverse shortcomings related to
violent conflict, state capacity and political stability, each of which
taken separately may have different impacts on population risk of
HIV. For example, a number of studies show that internal armed
conflict often is associated with lower HIV prevalence (Spiegel et al.,
2007). It is also plausible that transitions from authoritarianism to
democracy bring new freedoms which in turn increase HIV risks.
On the other hand, low state capacity constrains effective HIV/AIDS
2
The possible exception is Lesotho, a similar small and landlocked southern
African nation.
 A. de Waal / Social Science & Medicine 70 (2010) 114–120
responses, as shown by studies of Mozambique (Høg, 2008), Haiti
(Weigel, 2008) and Myanmar (Khin, 2008). Policies such as
decentralization can have unexpected and complex implications for
HIV, as illustrated by Smith’s (2008) study.
In line with the injunctions ‘know your epidemic’ and ‘and act
on its politics,’ (Buse, Dickinson, & Sidibé, 2008; UNAIDS 2007)
ASCI’s findings point to the diversity of governance and security
contexts, and point to the need to adopt policies and programmes
best suited to each. This includes attending to the ways in which
diverse manifestations of state crisis may affect HIV and AIDS.
Armed conflicts and their aftermath
In the study of the relationship between armed conflict and HIV
vulnerability, an influential revisionist critique has argued, first that
refugee populations generally have lower HIV prevalence than their
hosts (Spiegel, 2004) and, second, that conflict-affected pop-
ulations in general are commonly less at risk of HIV (Spiegel et al.,
2007). The factors that influence population level risk of HIV
include their pre-conflict level of HIV, level of movement including
labour migration and trade, and the degree of interaction between
a displaced population and its host. Conflict as such does not prove
to be a significant predictor of HIV vulnerability for a population,
and indeed insofar as conflict leads to the isolation of communities,
it can help to keep rates low.
ASCI commissioned three studies to address different aspects to
the question of how armed conflict and post-conflict transitions
affect HIV vulnerability. Davenport and Loyle (2009) undertook
a cross-country quantitative analysis across all countries for which
data are available (n ¼ 197) to assess the relationships (if any)
between conflict-years at a national level and national HIV preva-
lence (using data for 2007). They augmented this with a sub-
national analysis of HIV prevalence and conflict in Rwanda, using
data for killings during the 1994 genocide. Their model has three
factors: individual behavioural disruption associated with conflict,
resources available for HIV/AIDS programmes, and interaction
between conflict-affected civilian populations and others, including
refugee flows and the movements of belligerent armies and
peacekeepers.
Davenport and Loyle found that measurements of conflict
(magnitude, duration and geographical scope) had no correlation
with national HIV prevalence, both for the global dataset and for
sub-Saharan Africa. They found an association between national
resource availability and HIV (increased resources imply lower HIV
and vice versa), and a positive correlation between armed conflict
and the reduction of resources. They found that refugees from high
prevalence countries tend to increase the prevalence among their
hosts while those from medium and low prevalence countries do
not. They also found that international peacekeepers have a nega-
tive correlation with HIV prevalence in the country of deployment,
and speculate that this may be due to the aid resources provided to
a country which hosts a peacekeeping mission.
Davenport and Loyle’s case study of Rwanda tests the different
components of the framework at a subnational level. The study
uses data for the magnitude and duration of violence perpetrated
during the 1994 genocide (measured using estimates for numbers
of people killed and the length of time the killing continued) and
civil war disaggregated at the prefecture and HIV prevalence in
2005. It also uses data for the numbers of people who were dis-
placed during and after the genocide, for the movements of the
belligerent armed forces and the presence of peacekeeping
missions. While reliable data for the overall extent of sexual
violence during the genocide are not available, the evidence points
to an exceptionally high number of rape survivors among the
genocide survivors, and very high levels of HIV among those rape
117
survivors. In the eleven years between the genocide and the 2005
survey, most of those infected with HIV died, helping to explain
why HIV prevalence declined over those years. Meanwhile, post-
genocide Rwanda has established a highly-regarded health system
including an HIV/AIDS programme. The resource factor was not
included in the study because it was not possible to disaggregate it
by prefecture.
Davenport and Loyle found that the duration of the violence is
a significant predictor of HIV prevalence eleven years later, but no
effect of the magnitude of the violence (numbers killed) on HIV
prevalence. They found (contrary to their general conclusion on this
issue) a positive relationship between the presence of UN peace-
keeping troops and HIV prevalence, possibly skewed by the urban
bias of the peacekeeping presence. The location and movement of
refugees was associated with higher HIV prevalence.
A second study, by Seckinelgin, Bigirumwani and Morris (2008)
used ethnographic method to study the gendered impacts of the
conflict and post-conflict transition in Burundi. The study shows
how conflict and post-conflict transition are both associated with
patterns in the spread of HIV, in complex ways. The customary
patrilineal and patrilocal nature of Burundian society, in which
family is structured around men, property is owned by men, and
women have subordinate status, defines the context in which
women and girls seek status, livelihood and security. During the
years of conflict, most practices revolving around gender and the
patriarchal social hierarchy have remained unchanged, but young
people’s sexual behaviour is reported to have become freer.
The authors question the characterization of the Burundi
conflict as ‘low intensity.’ Despite the often low levels of armed
engagement between the fighting groups, the social implications of
the conflict have been intense and far-reaching irrespective of how
levels of violence have fluctuated. The relationship between HIV/
AIDS and conflict is a function of how armed groups mobilize and
sustain themselves (with rebel forces dispersed across the country),
the displacement experienced by the population, and the long-
term nature of these disruptions, which in turn have implications
for social organization. Over time, armed groups became social
formations, with their members establishing relationships ranging
from casual sexual partnerships to marriage.
The mobilization of armed groups took many men away from
the domestic environment, leaving women vulnerable. Women’s
weak legal rights became a serious constraint on survival, pushing
many into transactional sexual relationships. The commodification
of women’s bodies – already pronounced in the customary social
system – was accentuated. Male hierarchies were restructured with
authority accruing to military commanders. This had many rami-
fications, including in some cases orders given to junior soldiers to
rape women, and the differential protection given by men to their
partners during the conflict – officers provided more security to
their partners than junior ranks. Due to the protracted nature of the
conflict, ‘normal’ marital arrangements could not simply be sus-
pended for the duration and then resumed. Rather, the pressures of
conflict changed social mores and practices in enduring ways.
Burundi’s post-conflict disarmament, demobilization and rein-
tegration (DDR) process was not designed with gender relations of
HIV/AIDS in mind. Women and girls attached to the armed groups
were excluded from the process including its educational and
material components. The national policy was that men and
women should be treated equally within the DDR process, but that
was only possible for a restricted category of female combatants
and only on the condition that they went to the DDR camps, which
most did not want to do. Female ex-combatants and other women
who had borne children during the conflict fathered by combatants
were stigmatized. In all cases, HIV/AIDS sensitization was cursory,
and the brief training modules had no impact on behaviour change.
118 A. de Waal / Social Science & Medicine 70 (2010) 114–120
Delays in the process caused numerous problems. The research
highlights the inadequate conceptualization of gender vulnerabil-
ities and HIV/AIDS in DDR programmes and policies.
The third research activity on this question was focus group
discussions on HIV/AIDS issues in four post-conflict countries:
Bosnia, the Democratic Republic of Congo, Haiti and Liberia (ASCI,
2009) These discussions identified programmatic neglect of the
transition from conflict to post-conflict. Too often when hostilities
end, emergency relief agencies pull out and move elsewhere, but
development agencies only begin their assessments for rehabili-
tation and reconstruction, and do not actually initiate their pro-
grammes. Donor funding for emergency work dries up while
development assistance takes a while to come on-stream, leading
to a hiatus of external assistance. Also, most ongoing programmatic
activity is focused on groups or communities, which have been
identified as ‘at risk’ during the conflict phase, rather than grap-
pling with risk environments and the ways in which they may
change very rapidly when conflict comes to an end. A single group
can have very different risk profiles as conditions change.
Post-conflict DDR policies are often designed as ambitious and
generously-funded exercises to allow former combatants to be
retrained and reintegrated into civilian life with a sense of self-
worth, but end up as hasty and under-resourced exercises in
meeting quotas. This shortcoming may be compounded by
extremely limited employment alternatives, insufficient funding
for survivors of conflict, and the often conflictual relationships
across these groups, including returnees.
A case study of southern Sudan following the 2005 peace
agreement (D’Awol, 2008) illustrates these points. The southern
Sudan DDR process has been delayed, downsized and truncated.
HIV sensitization has been ad hoc, inconsistent and difficult to
monitor and evaluate. The DDR process involves numerous actors
with differing mandates, working methods, capacities and objec-
tives, and has gaps such as police and prison officers. The pro-
grammes are not sufficiently gender-sensitive despite the presence
of large numbers of women combatants, who are often the first in
line for demobilization. D’Awol concludes that efforts are incom-
mensurate with need, and the relatively fragmented and isolated
initiatives do not add up to the necessary holistic approach. She is
concerned that southern Sudan remains in danger of a post-conflict
HIV/AIDS epidemic.
Uniformed services
The third focus of ASCI research was on HIV/AIDS in the uni-
formed services. ASCI commissioned a series of research reviews
and policy studies. An early outcome was a sharp revision of the
conventional wisdom that HIV prevalence among armed forces is
several times higher than in civilian populations and that the
epidemic is undermining the functioning of militaries (Whiteside,
de Waal, & Gebretensae, 2006). This review found that HIV preva-
lence among militaries was usually within a similar range to that of
civilian populations, sometimes lower among new recruits
(because of the demographic from which they are drawn) and
commonly higher among middle-ranking officers – comparable to
peers in comparable mobile, well-paid occupations. It found that
while HIV/AIDS poses a threat to armies’ operational capability,
most armed forces are also well-placed to manage these threats,
because of their built-in capacity to withstand losses of personnel.
Lastly, the review concluded that military HIV/AIDS policies have
contributed to this (limited) success.
Several factors have led militaries to develop HIV/AIDS
responses in parallel to and isolation from their civilian counter-
parts. One reason is the introversion of military establishments.
Habitual secrecy was not helped by publicity that blamed soldiers
for being a vector of HIV transmission. In some countries, including
South Africa and Ethiopia, armies were ahead of national govern-
ments in formulating and implementing HIV/AIDS programmes,
giving a low profile to their efforts partly to minimize any public
controversy. However, the most important reason for policy
seclusion is that militaries prefer to implement policies of
compulsory HIV testing and the selective exclusion of those who
test positive, from recruitment, promotion or deployment. Armies
defend this practice on the grounds of operational imperatives
including the necessity of maintaining full-strength, good health
units, which is particularly important for resource-constrained
armies. This contradicts best practice in the civilian sector and is
contrary to human rights principles and, in some countries, the
constitution. In Namibia and South Africa, courts have ruled that
national defence forces must conform to laws that prohibit
employment discrimination based on HIV status. Rather than pre-
senting counter-arguments based on rival principles, such as mili-
tary necessity, proportionality and emergency requirements,
military establishments have preferred to ignore the rulings.
The South African High Court case of May 2008,3 is a landmark
in this controversy. The Security Forces’ Union and three individual
soldiers sued the South African National Defence Force (SANDF) on
the grounds that its policy of mandatory HIV testing and exclusion
and non-promotion of HIV positive individuals was discriminatory
and unconstitutional. The court ruled that the blanket ban was not
medically justified and ignored government policy of non-
discrimination at the workplace. It ordered the army to develop
a new policy by November 2008. In response, the SANDF Defence
Chief stuck to his guns and argued that he should not be compelled
to compromise on his mandate ‘to prepare forces for war and other
exigencies’ and should not therefore give way to humanitarian
concerns including the rights of individuals living with HIV. The
litigants interpreted this as foot-dragging.
Most of ASCI’s research work in this field focused on developing
policy frameworks that can be used by military planners, including
the peacekeeping departments of the UN and African Union (AU).
Kershaw (2008) identified the adverse impacts on unit cohesion as
the principal route whereby HIV/AIDS undermines the effective-
ness of armies. Impacts are felt at the operational unit level, at the
level of the general staff and among specialist units. Because HIV/
AIDS is a long-wave event, strategic planning for these impacts is
required. Kershaw developed a ‘Force Capability Analytical
Framework’ for examining the effects of HIV/AIDS and a ‘Military
Institutional Audit’ for investigating how these impacts are felt in
specific instances.
Binega (2008) undertook a comparative study of intervention
packages on HIV/AIDS in armed forces in Africa, paying particular
attention to the development of policy and practice within the AU
and its growing involvement in peace support operations. His study
spelled out challenges for policy development and implementation.
One of these is that troops on AU or UN peacekeeping missions
remain under the command of and in the employ of their respective
countries, which have differing policies and practices on HIV/AIDS.
While education and prevention are assigned to the UN, treatment
remains the responsibility of the troop-contributing country,
making it difficult to standardize approaches. Binega advocates the
wider adoption of a ‘command-centered approach’ to HIV/AIDS
control in militaries (supported by medical and health units) and
argues in support of mandatory HIV testing and screening within
an overall package that makes the health of military units the
responsibility of their commanding officer.
3
South African Security Forces Union and Others v. Surgeon General and Others
(16 May 2008), Case 18683/07.
 A. de Waal / Social Science & Medicine 70 (2010) 114–120
Military approaches to HIV/AIDS are developed to meet
nationally defined priorities rather than the requirements of
peacekeeping operations. Most militaries are seeking guidance in
developing their policies. Many troop-contributing countries
implement mandatory pre- and post-deployment HIV testing. This
contradicts international norms but UN policy is to accept national
practices in this regard. Peacekeeping operations face abiding
problems with stigma and discrimination, education and aware-
ness, continuing risk behaviour by soldiers, poor quality of data and
surveillance systems, variability in treatment regimes across
countries, and divergent approaches to female peacekeepers.
ASCI undertook research into law enforcement services and
HIV/AIDS, an important but remarkably neglected field (Pearce,
2007; Roderick, 2007). Case studies of policing and HIV/AIDS were
carried out in South Africa (Masuku, 2007), Benin (Gabriel-Nelson,
2007), and Sierra Leone (Thompson, 2007). One finding from these
reviews was that police services are highly sensitive on HIV/AIDS
issues, underling the importance of building trust between the
police and the research and HIV/AIDS policy communities. As with
other institutions heavily reliant on human resources, police
services in sub-Saharan Africa are institutionally impacted by
a high prevalence of HIV within their ranks. The same issues that
affect soldiers on peacekeeping operations also impact the civilian
police contingents that are increasingly a central part of those
missions.
Outside sub-Saharan Africa, where HIV epidemics are concen-
trated among groups that exist at the margins of the law and society,
engaging in activities that are both high risk for HIV and also
criminalized, the major concern is how law enforcement practices
influence the trajectory of the epidemic. These groups include
injecting drug users (IDUs), commercial sex workers (CSWs), men
who have sex with men, individuals with diverse sexual orienta-
tions, street children, survivors of sexual and gender-based violence
(SGBV), and illegal migrants and trafficked women. Harm reduction
practices such as needle exchange and methadone substitution
programmes are effective measures for minimizing HIV risk among
IDUs, but in many countries they are prohibited so that participation
in a harm reduction programme may put an individual at risk of
arrest and prosecution. Gender-related attitudes and practices
among police and prison officers are also a critical determinant of
the vulnerability of individuals who come into contact with the law,
including CSWs and survivors of SGBV. Law enforcement officers are
the first line of defence and response for those exposed to SGBV. In
some countries, it is police services that are the de facto regulators of
sex work. Scanning the list of the most at-risk groups in concen-
trated HIV epidemics, it is not difficult to conclude that law
enforcement practices will be an important determinant of the
future trajectory of those epidemics. This field remains neglected,
and the consultation on the police and HIV/AIDS convened by ASCI
in September 2007 was, remarkably, the first-ever global consulta-
tion of its kind. The principal conclusion of ASCI’s research in this
field is that this research agenda is in its infancy.
Conclusions
In each of the fields addressed by ASCI research, the initial
questions were significantly revised by the findings. Reflecting the
expansive and multi-disciplinary nature of the research questions,
ASCI’s methods and subjects of study were heterogenous. The
agenda allowed for diverse analytical and methodological paths to
be followed. These included a tour d’horizon of key issues, quanti-
tative investigations, selective in-depth case studies, and reflections
on the last decade of policy and practice. This approach is eclectic
but has the virtue of providing a range of perspectives on the central
119
cluster of problems, with a richness of insight that could not have
been provided by investigations confined to a single discipline.
Based on the received wisdom of the turn of the Millennium, the
starting hypotheses were that HIV/AIDS was having far-reaching if
indirect impacts on the viability of states, the functioning of mili-
taries and the extent of conflict and insecurity, and that in turn the
weakening of governance and security institutions would intensify
HIV/AIDS epidemics. Such ‘downward spirals’ (cf. Price-Smith &
Daly, 2004) were not substantiated by the research findings, with
the possible exception of Swaziland. To the contrary the main
finding was that the HIV/AIDS epidemic has become one among
many stresses on affected societies and institutions, and that its
impacts are appreciable at the level of local and sectoral governance
rather than national security.
The ASCI research agenda points to several specific areas for
policy attention. One is that HIV risks may increase during post-
conflict transitions in comparison to the phase of active hostilities.
Programmes for DDR have an under-utilized potential to control
HIV rates, and also run the risk that, if mishandled, they could
contribute to HIV risks. A second area is that uniformed services’
testing policies should be developed in ways that can manage the
competing demands of operational responsibilities versus indi-
vidual rights. A third focus is that HIV prevention and response
should address the entirety of the security sector, including
customs, prisons and immigration. The agenda of HIV and police
services is important and neglected, especially in view of the
potential of policing practices for influencing the trajectory of HIV
epidemics concentrated among marginalized or criminalized
groups. The research finds that sexual violence has yet to be
adequately addressed as a bio-social risk factor for HIV, and that it
would be unwise to minimize the role of rape in HIV transmission
on the basis of aggregate prevalence data.
The ASCI findings point to the need for better quality and more
finely disaggregated data. Current indices of state fragility are not
adequate to demonstrate any statistically significant association
with HIV at the country level, pointing towards the need for
subnational, sector and gender-specific measurements. The domi-
nant explanatory frameworks for analyzing societal HIV vulnera-
bility need to be developed, both in terms of finer categorization of
the epidemic types and risk factors, and also the changing social
roles, identities and environments.
At the macro policy level, one main area of interest is designing
effective HIV/AIDS under widely divergent systems of governance.
Another is how the HIV/AIDS response has itself influenced the
nature of national and global governance and inter-state relations,
including U.S. ‘medical diplomacy.’ In the field of HIV/AIDS, gover-
nance and security, the focus is upon micro-level environments and
how they change.
At a national level, in sub-Saharan African countries, HIV/AIDS
programmes and the associated resources have become a political
prize and an area of policy contestation (Høg, 2008). Meanwhile, HIV/
AIDS best practice, promoted and funded by international institu-
tions, has generally been a mechanism in support of civil society and
human rights norms (De Waal, 2006) – with exceptions such as
PEPFAR’s promotion of faith-based norms. In the least capacitated
states such as Haiti (Weigel, 2008) or the most repressive ones such as
Myanmar (Khin, 2008), such internationally-endorsed standardized
blueprints are not applicable. Rather, HIV/AIDS policies must be
designed around the exigencies of circumstance, informed by the
ethnographic contours of institutional collapse and the demands of
micro-political management. In addition, many military and law
enforcement establishments remain excluded from the international
policy consensus, with their concerns neglected.
HIV/AIDS led the way in a massive decade-long expansion in
funding for international health. While the performance of
120 A. de Waal / Social Science & Medicine 70 (2010) 114–120
international aid with respect to poverty reduction is coming under
renewed critical scrutiny, the effectiveness of assistance in
promoting health remains widely applauded. The debate on
international aid for health is not whether it is worthwhile, but
whether assistance should focus on vertical disease-specific efforts,
on health systems, or on some combination of the two. The ‘AIDS
and security’ scare of a decade ago played an important role in
galvanizing this response, particularly in the U.S. The international
debate on HIV/AIDS and security arose from a convergence of
conceptual and policy frameworks during the 1990s. Normative
assumptions about statehood combined with research methodol-
ogies and specific historical conjunctures (emergent infectious
disease epidemics, several state collapses in Africa, and the collapse
of the USSR) to produce a set of fears, focused upon the vulnera-
bility of governance structures to collapse on account of epidemic
disease. Although the macro-level fears were largely misplaced, the
fear was real, and the actions taken accordingly have had conse-
quences. Those normative frameworks and intellectual paradigms
drove global responses, and in turn influenced the trajectory of the
HIV/AIDS pandemic and shaped the landscapes of international
security and governance.
A negative finding invites us to discount the importance of HIV/
AIDS, conflict and security. That would be a mistake. Adjusting our
focus, we see important concerns and major challenges. These
include mitigating the impacts of HIV/AIDS on local government
and communities; enhancing service delivery in weak, failing or
repressive states; addressing a host of issues in post-conflict tran-
sitions and peacekeeping environments; and the role of law
enforcement services in shaping the trajectory of HIV/AIDS
epidemics. This is a challenging agenda for the next decade.
References
Altman, D. (2008). The political dimensions of responses to HIV/AIDS in Southeast
Asia. ASCI Research Report No. 5. http://asci.researchhub.ssrc.org/the-political-
dimensions-of-responses-to-hiv-aids-in-southeast-asia.
ASCI. (2009). AIDS, security and conflict: New realities, new responses. New York:
Social Science Research Council.
Barnett, T. (2009). ASCI synthesis paper: HIV/AIDS and state fragility. ASCI.
Bell, C., Devarajan, S., & Gershbach, H. (2004). Thinking about the long-run
economic costs of AIDS. In M. Haacker (Ed.), The macroeconomics of HIV/AIDS.
Washington DC: International Monetary Fund.
Binega, M. (2008). Comparative study on intervention packages on HIV/AIDS in the
uniformed services in Africa. ASCI Research Report No. 20. http://asci.researchhub.
ssrc.org/working-papers/ASCI%20Research%20Paper%2020-CPRD.pdf.
Buse, K., Dickinson, C., & Sidibé, M. (2008). HIV: know your epidemic, act on its
politics. Journal of the Royal Society of Medicine, 101, 572–573.
CHGA.. (2008). Securing our future: The report of the commission on HIV/AIDS and
governance in Africa. Addis Ababa: UN Economic Commission for Africa.
Cheek, R. (2001). Playing God with HIV: rationing HIV treatment in Southern Africa.
African Security Review, 10(4).
Chirambo, K., & Steyn, J. (2009). AIDS and state fragility: the challenge of increased deaths
among councillors in local government in South Africa. ASCI Research Report No. 25.
http://asci.researchhub.ssrc.org/working-papers/ASCI%20Research%20Paper%
2025-IDASA.pdf.
Commission on Human Security. (2003). Human security now. New York: United
Nations.
D’Awol, A. (2008). HIV/AIDS policies and programming in DDR and SSR activities in
Southern Sudan: a post-conflict analysis. ASCI Research Report No. 16. http://
asci.researchhub.ssrc.org/working-papers/Anyieth%20DAwol.pdf.
Davenport, C., & Loyle, C. (2009). The conflict-HIV/AIDS nexus: an empirical assess-
ment. ASCI Research Report No. 21. http://asci.researchhub.ssrc.org/working-
papers/ASCI%20Research%20Paper%2021-Davenport%20and%20Loyle.pdf.
Davis, C., & Feshbach, M. (1980). Rising Soviet infant mortality. INTERCOM, 8(17),
12–14.
De Waal, A. (2004). How will HIV/AIDS transform African governance? African
Affairs, 102, 1–24.
De Waal, A. (2006). AIDS and power: Why there is no political crisis – yet. London: Zed
Books.
De Waal, A., & Whiteside, A. (2003). ‘New variant famine’: AIDS and food crisis in
Southern Africa. The Lancet, 362, 1234–1237.
Dutta, I., & Barnett, T. (2007). HIV and state fragility: is HIV a security risk?. ASCI
Research Report No. 7. http://asci.researchhub.ssrc.org/hiv-and-state-failure-is-
hiv-a-security-risk.
Eberstadt, N. (1988). The poverty of communism. New York: Transaction Publishers.
Eberstadt, N. (2002). The future of AIDS. Foreign Affairs, 81(6), 22–42.
Esty, D., Goldstone, J., Gurr, T. R., Surko, P., & , & Unger, A. (1995). State Failure Task
Force Report. Washington DC.
Feshbach, M. (2008). Russian demography, health and the military: current and
future issues. ASCI Research Report No. 8. http://asci.researchhub.ssrc.org/
russian-demography-health-and-the-military-current-and-future-issues.
Gabriel-Nelson, D. (2007). HIV and paramilitary officers in the Republic of Benin.
Justice Africa. www.justiceafrica.org/wp-content/uploads/2007/10/policing-
benin.pdf.
Høg, E. (2008). States of HIV fragility: capacity, vulnerabilities, and epidemic
evolution in Mozambique. ASCI Research Report No. 11. http://asci.researchhub.
ssrc.org/states-of-hiv-fragility-capacity-vulnerabilities-and-epidemic-evolution-
in-mozambique.
International Crisis Group. (June 2001). HIV/AIDS as a security issue. Issues Report
No. 1. Washington DC.
Jacob, H. (2008). Impact of HIV/AIDS on governance in Manipur and Nagaland. ASCI
Research Report No. 6. http://asci.researchhub.ssrc.org/impact-of-hiv-aids-on-
governance-in-manipur-and-nagaland.
Kershaw, R. (2008). The impact of HIV/AIDS on the operational effectiveness of
military forces. ASCI Research Report No. 4. http://asci.researchhub.ssrc.org/
the-impact-of-hiv-aids-on-the-operational-efficacy-of-military-forces.
Khin, O. (2008). ‘First, what is harm?’ The political dilemmas of humanitarian aid to
Burma (Myanmar). ASCI Research Report No. 14. http://asci.researchhub.ssrc.
org/working-papers/ASCI%20Paper%2014-Ohnmar%20Khin.pdf.
Masuku, T. (2007). AIDS in the South African police service: a threat to the state?
Justice Africa. www.justiceafrica.org/wp-content/uploads/2007/10/policing-
south-africa.pdf.
National Intelligence Council. (January 2000). The global infectious disease threat and
its implications for the United States. Washington DC: NIC, NIE. 99–17D.
National Intelligence Council. (September 2002). The next wave of HIV/AIDS: Nigeria,
Ethiopia, Russia, India and China. Washington DC: ICA. 2002–04D.
National Intelligence Council. (December 2008). Strategic implications of global
health. Washington DC: ICA. 2008–10 D.
Naysmith, S., de Waal, A., & Whiteside, A. (2009). Revisiting New Variant Famine:
The Case of Swaziland. Food Security, 1(3), 251–260.
Neilson, T. (January 2005). AIDS, economics and terrorism in Africa. Global Business
Coalition on AIDS.
O’Keefe, M. (2008). State fragility and AIDS in the South Pacific. ASCI Research
Report No. 9. http://asci.researchhub.ssrc.org/state-fragility-and-aids-in-the-
south-pacific.
Pearce, H. (2007). The police and HIV/AIDS: a literature review. ASCI Research Report No.
1. http://asci.researchhub.ssrc.org/the-police-and-hiv-aids-a-literature-review.
Price-Smith, A., & Daly, J. (2004). Downward spiral: HIV/AIDS, state capacity and
political conflict in Zimbabwe. Washington DC: US Institute of Peace. Peaceworks
No. 53.
Roderick, A. (October 2007). HIV/AIDS, democracy and governance in Africa:
a special edition on policing. Justice Africa, GAIN Issues Brief No. 11. www.
justiceafrica.org/wp-content/uploads/2007/10/gain-brief-october-2007.pdf.
Sato, A. (2008). Is HIV/AIDS a threat to security in fragile states?. ASCI Research
Report No. 10. http://asci.researchhub.ssrc.org/is-hiv-aids-a-threat-to-security-
in-fragile-states.
Seckinelgin, H., Bigirumwani, J., & Morris, J. (2008). HIV/AIDS, conflict and the
gendered implications of transition in Burundi. ASCI Research Paper No.13. http://
asci.researchhub.ssrc.org/gender-hiv-aids-and-ddr-in-burundi/project_view.
Smith, C. (2008). Indigenous welfare and HIV/AIDS risks: the impacts of government
reform in the Papua region, Indonesia. ASCI Research Report No.
12. http://asci.researchhub.ssrc.org/working-papers/ASCI%20Paper%2012-Claire%
20Smith.pdf.
Spiegel, P. (2004). HIV/AIDS among conflict-affected and displaced populations:
dispelling myths and taking action. Disasters, 28(4), 322–339.
Spiegel, P., Bennedsen, A. R., Claass, J., Bruns, L., Patterson, N., Yiweza, D., et al.
(2007). Prevalence of HIV infection in conflict-affected and displaced people in
seven sub-Saharan African countries: a systematic review. The Lancet, 369,
2187–2195.
Thompson, A. (2007). Rebuilding institutions, reinforcing risk: The police and HIV/AIDS
in Sierra Leone. Justice Africa.
UNAIDS.. (2007). UNAIDS annual report: Knowing your epidemic. Geneva: UNAIDS/
08.21E/JC1535E.
UNDP.. (1994). Human development report 1994. New York: Oxford University Press.
Weigel, J. (2008). Neither will nor capacity: a review of the challenges of responding
to AIDS in fragile states. ASCI Research Report No. 15. http://asci.researchhub.
ssrc.org/working-papers/ASCI%20Paper%2015-Jon%20Weigel.pdf.
Whiteside, A., & Whalley, A. (2007). Reviewing ‘emergencies’ for Swaziland: Shifting
the paradigm for a new era. Durban: Health Economics and HIV/AIDS Research
Division.
Whiteside, A., de Waal, A., & Gebretensae, T. (2006). AIDS, security and the military
in Africa: a sober appraisal. African Affairs, 105, 201–218.