This article was downloaded by: [College Of Charleston], [Christopher Day]On: 20 August 2014, At: 06:20Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,UK

Civil WarsPublication details, including instructions for authorsand subscription information:http://www.tandfonline.com/loi/fciv20

In Harm's Way: African Counter-Insurgency and PatronagePoliticsChristopher R. Daya & William S. Renob

a Department of Political Science, College ofCharleston, Charleston, SC, USAb Department of Political Science, NorthwesternUniversity, Evanston, IL, USAPublished online: 18 Aug 2014.

To cite this article: Christopher R. Day & William S. Reno (2014) In Harm's Way:African Counter-Insurgency and Patronage Politics, Civil Wars, 16:2, 105-126, DOI:10.1080/13698249.2014.927699

To link to this article: http://dx.doi.org/10.1080/13698249.2014.927699

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all theinformation (the “Content”) contained in the publications on our platform.However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, orsuitability for any purpose of the Content. Any opinions and views expressedin this publication are the opinions and views of the authors, and are not theviews of or endorsed by Taylor & Francis. The accuracy of the Content shouldnot be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions,claims, proceedings, demands, costs, expenses, damages, and other liabilitieswhatsoever or howsoever caused arising directly or indirectly in connectionwith, in relation to or arising out of the use of the Content.

This article may be used for research, teaching, and private study purposes.Any substantial or systematic reproduction, redistribution, reselling, loan, sub-

licensing, systematic supply, or distribution in any form to anyone is expresslyforbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

In Harm’s Way: African Counter-Insurgencyand Patronage Politics

CHRISTOPHER R. DAYa and WILLIAM S. RENOb

aDepartment of Political Science, College of Charleston, Charleston, SC, USA;bDepartment of Political Science, Northwestern University, Evanston, IL, USA

This article explains why contemporary African regimes choose different

counter-insurgency strategies and why they tend not to be population-centric.

We argue that strategies correspond to the ways in which incumbent regimes

in Africa deal with different segments of political society through patronage.

Incumbents seek varying levels of accommodation with rebel leaders, or try to

eliminate them, according to rebels’ historical position within the state. This

variation reflects differences in perceived political threats posed to

incumbents. We classify these threats as high, moderate or low, which are

associated with counter-insurgency strategies of group control, insurgent

control and insurgent elimination, respectively.

It is common among scholars and policymakers to assert that in contemporary civil

wars, states and rebels fight each other to control and govern non-combatants. In this

conventional view, rebels fight to protect their communities from external assault

and rely on civilians for support.1 Effective counter-insurgency in turn focuses on

out-governing rebels by providing basic security and social services to prevent

rebels from retaliating against people who provide essential information about rebel

membership and whereabouts. Separating rebels from non-combatants – a critical

measure of success in counter-insurgency warfare – enables states to identify and

pursue rebels to the point that they are defeated.2 This population-centric approach

to counter-insurgency guided the US military’s ‘surge’ strategy in Iraq. In 2009 the

commander of US forces in Afghanistan wrote: ‘Our strategy cannot be founded on

seizing terrain or destroying insurgent forces; our objective must be the population.’3

Guerrilla warfare’s classic treatises from Mao to Che assume that rebels seek the

sympathy and support of civilians.4 The standard vision is that successful

counterinsurgents must follow suit and ostensibly tap into non-combatant networks

for information needed to sever rebels from this essential linkage to local communities.

Counter-insurgency in Africa presents a very different picture. Most African

states lack the institutional capacity to launch extensive counter-insurgency

programs to out-govern rebels. There are numerous ‘ungoverned spaces’ in which

rebels can seek refuge, particularly in the hinterlands of Africa’s porous boundary

q 2014 Taylor & Francis

Civil Wars, 2014

Vol. 16, No. 2, 105–126, http://dx.doi.org/10.1080/13698249.2014.927699

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

regions.5 With these things in mind, Africa’s institutionally weak states should be

especially vulnerable to rebel challenge.6 Many foreign aid programs are built on the

assumption that extending effective governance is an essential element of state

security. US strategy in Africa, for example, incorporates multi-agency civilian and

military assistance to African governments to strengthen state institutions to provide

security to citizens, manage local and regional conflicts, and to exercise surveillance

over border regions.7

Yet, very few African rebels have succeeded in overthrowing governments. Only

about 10 per cent have been unambiguously victorious, like Uganda’s National

Resistance Army (NRA) and the Ethiopian People’s Revolutionary Democratic

Front (EPRDF).8 ‘Victory’ takes on more ambiguity considering the internal

fragmentation that accompanied the ‘success’ of groups like the recent Seleka

rebellion of the Central African Republic.9

The paucity of clear success is striking given the sheer numbers of rebel groups

in Africa from the past several decades. Not all of them sought to capture capitals

and install a reform agenda. Some wanted secession; others wished to carve out a

warlord fiefdom while a few fought to protect their communities from rebels and

soldiers alike.10 Yet, Africa’s myriad rebels still posed threats to incumbents and had to

be managed in one way or another. An investigation into Africa’s 150 odd cases of

rebellion shows that at least 30 per cent of them did not survive their wars, irrespective

of motivation.11 Angola’s Uniao Nacional para a Independencia Total de Angola

(UNITA) crumpled following the death of Jonas Savimbi in 2002.12 Sierra Leone’s

Revolutionary United Front (RUF) failed to overthrow a series of weak regimes.13

Either African rebels are bad fighters or rulers in institutionally weak African

states have discovered surprisingly effective alternative means of waging counter-

insurgency warfare. In many cases, regimes defeat rebels outright or make rebel

victory next to impossible. Effective counter-insurgency in Africa also fights and co-

opts rebels simultaneously, erecting few boundaries between ‘rebel surrender’ and

incorporation into the state’s authority structures. Nearly 40 per cent of Africa’s

insurgencies have ended in ceasefires or peace agreements. Some involved

international mediation, like the 1992 Rome General Peace Accords that ended the

war between Mozambique’s Resistencia Nacional Moc�ambicana (RENAMO) and

the Frente de Libertac�ao de Moc�ambique (FRELIMO) regime.14 Others like the

1988 Pece Peace Accord that ended the Ugandan People’s Democratic Army’s

(UPDA) rebellion, were home-grown affairs.15

A common factor to Africa’s counter-insurgency strategies is the striking

scarcity of population-centric approaches. An extreme example was Sudan’s

campaign in Darfur in the 2000s, which did not include efforts to out-govern rebels

by providing non-combatants with services or protection. Instead, the campaign

included civilian repression and selective mobilisation of local militias, which

developed in symmetry with increasingly fragmented rebel forces.16

We argue that regime counter-insurgency strategies in many African states are

extensions of patronage-based regime strategies for exercising authority outside of

warfare. These strategies reflect the nature of rebellion contemporary in Africa,

CIVIL WARS106

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

which tends to crowd out more ideologically committed rivals as rebels pursue goals

that have little to do with competing with the state to govern non-combatants. This

exposes rebel groups to regime strategies that do not rely on building stronger state

institutions or winning over public support. Instead, strategies aim to co-opt rebels

through amnesty and settlements, or to defeat them through the use of armed proxies

such as local militias and rebels from neighbouring countries. This logic follows the

key actors that are ‘in harm’s way’. Rather than armed forces or the civilians they

ostensibly protect, at risk here are African regimes, which match political and

military responses to the level of threat posed by different types of rebels. Rebels

that pose the greatest threat to incumbent regimes are those drawn from the very

political fabric of regime-run patronage networks. These political insiders are met

with the least coercive and most co-optative counter-insurgency strategies. The most

coercive, militarised counter-insurgency strategies are reserved for rebels that

comprise political outsiders, or those that arise from outside regime-based patronage

networks. We find that regimes in Africa are shrewd, innovative and flexible in how

they fight rebels and thus prove to be much more durable than most observers would

expect against their rivals.

In what follows, we situate our argument in the broader literatures on counter-

insurgency, African militaries and insurgent violence in Africa. We then provide a

comparative framework that explains why the political strategies for dealing with

insurgents are not population-centric, but correspond directly to how incumbents

deal with different segments of political society. Empirical evidence from Uganda

then illustrates the links between patronage politics and counter-insurgency. By

patronage politics, we refer to how individual rulers project and maintain political

authority by dispensing and withholding access to privilege, wealth and status within

state politics. This means that the goals of African rebels are often as much about

accessing these authority networks as they are about replacing them.17 This article

therefore addresses variations in how incumbents manage the violent overtures of

these actors as a strategy for achieving these goals. We conclude with the broader

implications of our findings.

COUNTERINSURGENCY, AFRICAN ARMIES AND AFRICAN REBELS

Scholarship on African counter-insurgency as a distinct category is scarce and both

traditional and contemporary literature on the subject has limited relevance to

Africa’s political context. Scholarly work on African militaries and insurgent

violence focus on the means available to Africa armies and the motives of rebels and

do not provide a full picture of why regimes choose different approaches or why

civilian populations are relatively marginal to the wars between these actors, even if

civilians experience significant consequences of these wars.

The dominant approach of ‘out-governing the rebel’ evolved from a doctrine

based on experiences from past wars and across armies.18 Military officers from the

late 19th and early 20th centuries wrote foundational works on counter-insurgency

in contexts where their armies fought and helped set up government administrations

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 107

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

in the places they controlled.19 Other authors drew from the experiences of Britain

and France in colonial territories such as Malaysia, Algeria, Kenya and Northern

Ireland.20 The works that inform current policies also reflect US experiences in

Vietnam and the Philippines. John Nagl refers to these rebellions as a ‘special kind

of revolutionary war’ and stresses the importance of military ‘organizational

culture’ in separating rebels from non-combatants.21

David Kilcullen critiques these classical approaches in light of a shift in

contemporary insurgency.22 He and others recognise that while the more classic

forms of rebellion persist – those that seek to overthrow incumbent regimes – a new

approach must deal with localised threats that are potential extensions of an

ideologically driven ‘global insurgency’.23 This requires widening the political

scope of counter-insurgency into regional and global contexts, utilising propaganda

networks, and for longer periods.24 Kilcullen also notes that ‘counter-insurgency

mirrors the state’ insofar as it reflects regime type.25 Yet, his approach still assumes

that civilians matter and that rebels are insulated from the regime politics.

In both classic and contemporary approaches, standard prescriptions include

population-centric ‘statebuilding’ measures such as training local armies, building

up state institutions to provide services to non-combatants and limiting government

corruption.26 Even scholarly analysis that downplays the autonomous role of

ideological agendas ties violence to variations in spatial control and degrees of

success in competitive statebuilding projects.27

A recent RAND study supports this observation.28 It tests 20 practices – both

classic and contemporary – against 30 cases studies and finds that the most effective

strategies use a combination of approaches that recognise the relevance of

population-centrism. Even when insurgents do not depend on civilians for tangible

support, the authors call for ‘additional areas of emphasis’, like ‘hearts and minds’

strategies to enhance government legitimacy.29 These recommendations embrace

contemporary practices that favour the ‘beat cop’ and ‘cultural awareness’, which

are unmistakably population-centric.30 But the study does not make any statements

about why regimes choose one combination of approaches over others.

Thus, classic and contemporary approaches fail to consider that rebels and

incumbent regimes engage with one another in a significant fashion beyond

battlefield clashes to control territory and people, and beyond the reach of outside

interventions. Dominant approaches devote little attention to how the strategies that

regimes use to keep themselves in power are extended to dealing with rebel

challenges, or consider why otherwise rational leaders would resort to patronage or

other devices that would seem to undermine their institutional capabilities to address

rebel threats. They also do not consider the latitude that really exists for regimes that

are otherwise extremely dependent on foreign patrons to act against rebels. The

preoccupation with non-combatants as the focus of conflict ignores the real political

relationships that can exist between incumbent regimes and rival sources of power

situated in individual rebel leaders or within entire groups. Moreover, the large

number of African rebel groups incorporated into incumbent regimes through

political settlements suggests an alternative dynamic is at play.

CIVIL WARS108

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

At the level of African militaries, few studies account for how they have

strategically engaged with rebels and non-combatants as extensions of regime

survival. Much of the literature attributes institutional weaknesses to historical

antecedents such as pre-colonial social and political structures.31 Others consider

colonial armies and anti-colonial insurgencies as providing templates for modern

African militaries.32 And the most common work on African militaries has focused

on their role in regime politics.33 By the 1980s, the problem of coups d’etats became

almost a caricature of African politics as armies displaced democratic regimes to

‘restore order’ to the state, seldom stepping down once in power. Individuals such as

Idi Amin and Sani Abacha loomed large as indicators of a broader pattern of what

Samuel Decalo called an ‘authoritarian syndrome’.34 African militaries came to be

viewed as weak, incompetent and unable to formulate, much less execute coherent

counter-insurgency strategies on behalf of regimes that feared their own military

forces as much as they feared rebels.35

A more sanguine view holds that African militaries are modestly strengthening

their institutional capacities by participating in regional cooperation and

peacekeeping operations.36 But there still have been few attempts to explain

counter-insurgency effectiveness or the large-scale failure of rebels in the African

context. Why African regimes are able to use a variety of tactics against rebels with a

high degree of success is seldom explained. Empirical evidence finds that even

severely disorganised and impoverished regimes in Africa have a surprisingly solid

record of either defeating or incorporating rebels.

On the rebel side of the story, scholarship on rebel governance shows how these

actors erect rudimentary political structures to control territory and locals in order to

satisfy basic needs such as food supply and recruitment.37 This builds on previous

literature that tied rebel organisation to non-combatants through shared social

structures, mass sentiments and norms.38 A look at historical cases shows that

African rebels devoted a great deal of attention to governing non-combatants. In the

1960s and 1970s, Amilcar Cabral’s Partido Africano da Independencia da Guine e

Cabo Verde (PAIGC) developed a population-centric strategy that used cadres of

educated urban youth to mobilise the rural population for revolution, carving out

extensive ‘liberated zones’ rooted in local culture.39 In the 1980s, Uganda’s NRA

cultivated grassroots loyalties and at times developed dependency upon civilians.40

Yet, by the 1990s, mass mobilisation and governance diminished as strategies to

control civilians and gain their acceptance. Rebel-controlled zones more often

became generators of mass civilian flight rather than sites of reciprocity

arrangements between rebel leaders and communities. Sierra Leone’s RUF and the

NationalPatriotic Front ofLiberia (NPFL)were emblematic of such ‘newwars’ – those

without ideological narratives, lacking popular support, resource-driven and highly

predatory, situating rebels at the juncture of state weakness, globalisation and

increasingly privatised violence.41 Political economy models of conflict followed

suit, seeing private gain as the prime mover of war and central to how rebels chose

their strategies.42 In this view, natural resources gave these rebels the income that

they needed to attract followers and to fight without developing supportive links to

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 109

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

non-combatants. Jeremy Weinstein used this insight to explain variation in rebel

violence against civilians, arguing that when rebels have access to outside resources,

rebel ‘consumers’ crowd out ‘investors’, truncating a leader’s need for ideological

conditioning of fighters and permitting indiscipline in order to maintain the

membership of low-commitment members, or ‘opportunists’.43

This literature on civil wars and insurgent violence dovetails with the prevailing

logic of counter-insurgency – that civilian support and cooperation is required for

rebel success, irrespective of rebel motivations. But it is not clear just how much

support is required for success.44 Equally, while this work does address relations

between rebels and non-combatants, it tells us little about what to expect from

regime counter-insurgency strategies. It does not take account of what happens when

rebel and government forces commit atrocities and do not seek material or political

support from non-combatants as part of their war fighting strategies. Non-combatant

motivations to support either side of a conflict vary. Their choices can reflect how

well they feel they are being ‘governed’. But they can also reflect a profit motive, the

lack of an alternative, coercion or fear of reprisal. Many non-combatants simply flee

to avoid these choices, a prominent feature of Africa’s wars without rebel-governed

‘liberated zones’.45

The decline in importance of civilians for rebels reflects distinct patterns of

Africa’s civil wars and African politics more generally. Rebel governance, or lack

thereof, is a marker for more salient structural issues that historically situate

rebellions within Africa’s political society and its institutions. The argument that

follows recognises the tendency for both rebels and armies to downgrade the

importance of civilians, which contrasts with the dominant counter-insurgency

literature. But it also diverges from the work on civil wars and insurgent violence in

identifying the principal cause of decisions to forego population-centric strategies.

Rather than military means or rebel motives, what matters here is the political

character of regimes in states with very weak or failing institutions. This argument

privileges the role of pre-war political networks in determining regime strategies. It

starts with the simple observation that rebels in these states tend to confront armies

that are highly fragmented and may be politically disinclined to pursue population-

centric strategies when fighting rebels.

COUNTERINSURGENCY IN AFRICA

Population-centric strategies are not often used nor are necessary to achieve the aims

of African regimes. African counter-insurgency strategies tend to be rebel-centric,

in which the relations with non-combatants vary considerably according to the

on-going relationships between a rebel group and state authority. Non-combatants

are still at times targets for control or co-optation. They can be recruited to fight on

behalf of the state. Or they can be ignored as the incumbent regime pursues rebels

directly. It is important to note that none of these strategies presupposes a

government or a rebel effort to protect and administer these populations as part of

war-fighting strategies. These regime strategies build on the politics of patronage

CIVIL WARS110

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

that characterises the on-going exercise of political authority in these countries, and

is now applied to the task of countering rebel challenges. Counter-insurgency in this

kind of regime thus is an extension of existing strategies of rule, albeit in more

violent and less certain circumstances.

At first glance, many African militaries pose considerable threats the regimes

that they are supposed to protect. Even so, military power in the hands of patronage-

based regimes in Africa is capable of undertaking counter-insurgency campaigns.

But as extensions of the politics of incumbent regimes, their behaviour reflects the

dependence on non-institutional means of exercising power, and reliance on

controlling channels of patronage and the distribution of these resources. Counter-

insurgency strategies of African states are therefore endogenous to regime politics.

This means that the ways in which rulers manage domestic rebellion correspond to

the ways in which they deal with threats to their authority from different segments of

the political society that are not necessarily at war with their regimes. This kind of

politics reflects pre-war strategies of the incumbent regime to control its own

members and the population at large. Counter-insurgency in this context is not about

military doctrine or winning wars in purely military campaigns. It is about bringing

recalcitrant rebel leaders and non-combatants into the orbit of state control through

co-optation, for example, by drawing specific groups back into political society

through restoring linkages to local notables. A key point is that rebellions that

involve groups with formal and informal historical linkages to regime networks will

be treated differently from rebel groups that do not. Incumbent regimes will seek

varying levels of accommodation with rebel leaders, or try to eliminate them,

according to these historical institutional positions within the state. This variation in

strategies reflects differences in degrees of perceived political threats they pose to

incumbent rulers. We classify these perceived threats as high, moderate, and low.

These categories correspond to the counter-insurgency strategies of group control,

insurgent control, and insurgent elimination, respectively (see Table 1). The cases

that follow in the next section illuminate distinct sets of tactics associated with each

strategy.

Rebel groups with close ties to state authority often possess intact, complex,

state-interdependent networks. These groups pose a higher threat to incumbents, and

are costly to eliminate. Often these rebels represent entire ethnic, regional or

religiously affiliated groups within political society. Group control therefore

involves bargaining and negotiation with rebel leaders in order to co-opt them, as

well as demobilising any potentially broad support they may have from civilians.

TABLE 1

OVERVIEW OF THREATS AND STRATEGIES

Threat level Counter-insurgency strategy

Low Insurgent eliminationModerate Insurgent controlHigh Group control

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 111

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

While involving non-combatants, this strategy nevertheless remains rebel-centric

and not population-centric. Alternatively, rebel groups that contain previous elites

but are historically disconnected from the incumbent regime pose a threat for the

same organisational reasons as groups with direct ties. But because these groups are

several steps removed from state authority, there is no salient overlap between the

rival political networks of rebel and incumbent, and threats are only moderate. These

moderate threats are therefore dealt with through a strategy of insurgent control,

where rebels are targeted for co-optation and civilians are largely ignored. Finally,

political outsiders pose low threats and present few risks to pursuing their military

elimination, which involves military action that targets insurgent resources and

vulnerabilities on the battlefield, while ignoring civilians.

Strikingly, it is rare to find contemporary African regimes competing with rebels

to out-govern non-combatants as part of a counter-insurgency strategy, because non-

combatants are not central to struggles over the control of patronage networks and

their resources. Most African regimes are only concerned with rebel ties to non-

combatants insofar as it means rebels will seize resources that regimes need to

maintain their political networks. These regimes usually face factionalised and often

violently divided domestic political networks that leaders manipulate to assert

authority. Dissatisfied regime insiders may decide that rebellion is a good way to

negotiate an elevated position in existing political networks. Effective counter-

insurgency strategies in most wars in Africa focus on limiting these challenges to

patronage networks and the material resources that rulers need to make these

networks run. As noted above, this terrain of politics is the underlying causal factor

that leads to the formation of rebel groups that forego population-centric strategies.

This type of political strategy weakens and disorganises state agencies, including

armies and security forces, imposes serious limitations on these states’ capacities to

pursue population-centric strategies when they fight rebels. Since their armies and

militias are organised and fight much like the rebels, all sides are prone to

committing atrocities against non-combatants. Violence of this sort is an indicator of

a patronage network-centric rather than a population centric form of rebel war.

Regimes treat true political outsiders differently. Such rebels pose a different

kind of threat to regimes because they actually do try to administer non-combatants

in areas that they control and offer ideological programs in which there is no role for

the incumbent political establishment. This kind of rebel is surprisingly rare in

Africa, in part due to the difficulties of finding the political space to organise these

alternatives in the context of dense patronage networks. But when these outsider

rebels do appear, regimes adopt a strategy of elimination, as co-optation is more

costly and difficult when confronted with hostile newcomers to the political system.

Given the weak military capacities of these regimes, they will sometimes seek

external military assistance, claiming that the ideologues are dangerous anti-system

rebels, such as Islamic extremists, secessionists or in league with drug traffickers and

others to attract support from powerful non-African backers.

More typical are rebels that form when a dominant, established group

experiences a sudden status reversal within the state’s inter-group political

CIVIL WARS112

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

hierarchy. Their leaders articulate the perception that ‘one’s group is located in an

unwarranted subordinate position on a status hierarchy’ based on that group’s

historically perceived ‘right of control’ over the state apparatus.46 If successful, this

kind of rebel will now ‘be the group perceived as farthest up the. . . status hierarchy

that can be most surely subordinated through violence’ in competition among

political factions.47 This political historical proximity of the group to state privilege

puts these rebels in a distinctly hostile relationship to the incumbent regime.

Regimes thus respond to this kind of rebel through strategies that prioritise

management of rival sources of power and authority.48 Even regimes in fragmented

states will seek to organise and dominate existing networks of social control in order

to assert authority, gain more security and ensure survival. In this context where

regimes have long lacked direct command and discipline over state security

agencies or the capacity to mobilise the population, they continue to pursue the

politics of negotiation, selective targeting and playing local power brokers off one

another to dominate and incorporate rebels when they can and eliminate other rebels

if they must.49

In African states, this political realm is not structurally distinct from the rest of

society.50 Authority rests upon the ‘informalisation of politics’, where rulers regulate

access of society to resources and distribute them down vertical networks in

exchange for personalised political support.51 The ‘instrumentalisation of disorder’

ensures power can be extended within the weak institutionalisation of political

practices.52 This strategy stands in contrast to global norms that prescribe the

management of conflict as involving the strengthening of formal institutions, a

version of population-centric counter-insurgency. In the informalised political

context, disenfranchised members of previous and current regimes determine most

rebels’ relationships to state authority. Such actors will have political access,

knowledge and military skills that serve as leadership conduits in the patronage-

based political system, and this contributes to the structure and organisation of their

rebel movements, factors well known to incumbent state militaries.

The capacity of pre-war regime politics to shape rebel behaviour lies in the

tendency of the patronage-based political strategy to co-opt and shape the social

networks and relationships into which rebel groups normally would try to integrate

as part of population-centric strategies. The Maoist concept of guerrilla war saw

these social networks as isolated from regime politics, an insulation that protects

rebels from the military and security forces of the state as leaders recruit and

organise followers. These networks eventually became the framework for

establishing liberated zones in China as well as in Africa’s anti-colonial struggles.

But in contemporary states where regimes rule through dense patron–client

networks in lieu of effective state institutions, these social networks are incorporated

into day-to-day politics. Among the many who are subordinate members of these

political networks, toppling this kind of regime and seizing state power is best

accomplished through removing one’s superiors in this network. Conversely, the

pursuit of the classic rebel guerrilla strategy through mobilising non-combatants in

population-centric rebellion becomes exceedingly difficult when so many social

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 113

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

relationships upon which guerrilla forces presumably rely are dominated by patron–

client networks with ties to the capital.

The irony of this situation in many parts of Africa is that the politics that makes

these states so weak in institutional terms – through granting privileges, including

protection from prosecution for misdeeds on the basis of political loyalties, the

tolerance of corruption, the appointment of officials without regard for experience or

training – also makes these states particularly effective at inhibiting the

development of population-centric rebels. This is true in spite of the vulnerability

of this kind of regime to replacement by its own agents, whether from subordinate

officials or by people who hold no official title but play important roles in the non-

bureaucratic networks of patronage and control. This kind of politics, rather than

scrambles for lootable resources, plays the crucial role in shaping rebel and counter-

insurgency strategies. As the examples below show, the bulk of contemporary rebel

groups in Africa reflect the political strategies of the regimes that they fight,

including in how they interact with non-combatants. It is fairly common, for

example, for wars in Africa to feature state forces and rebels that are equally

factionalised and that prey upon non-combatants in similar ways.

COUNTERINSURGENCY IN UGANDA

Recent conflicts in Africa show a significant degree of variation in counter-

insurgency strategies. Consider the case of Uganda. Of the country’s 29 rebellions

recorded by the Uganda Amnesty Commission Report, not a single group seized

power.53 During the 1990s, Yoweri Museveni’s National Resistance Movement

(NRM) government fought several of these groups simultaneously but deployed a

range of approaches. Against the Lord’s Resistance Army (LRA), the government

pursued a modal strategy of group control. Alternatively, the Allied Democratic

Front (ADF) prompted insurgent elimination. For the West Nile’s rebellions, the

Ugandan government opted for a strategy of insurgent control.

Methodologically, Uganda provides a defined geographic area and relatively

short time span. Holding the state and regime constant (Uganda and the NRM since

1986), these cases provide a high degree of variation across rebel groups’ political

profiles and corresponding counter-insurgency strategies. Each group received

backing from the Sudanese government, claimed similar grievances against the

NRM, and were highly predatory. These controls address potential explanations

rooted in rebel motivations to govern civilians. And while the capacity of the army, the

Ugandan People’s Defense Force (UPDF), may have been limited, it nevertheless

demonstrated relatively stable and sophisticated forms of fightingmultiple rebel groups

at the same time, but did so differently depending on the group.

Group Control

The campaign to defeat the LRA in northern Uganda came closest to the population-

centric form, but not out of concern for the protection of civilians. Non-combatants

were caught between both violent rebels that unleashed waves of human rights

CIVIL WARS114

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

abuses and widespread brutality on the part of Uganda’s military.54 Rebel fighters

preyed on upon ethnic Acholi non-combatants and relied upon kidnapping to acquire

recruits. Violence against non-combatants was also designed to embarrass and

discredit the regime. This was arguably a population-centric rebellion, but one

designed to communicate to incumbent elites rather that demonstrate legitimacy to

civilians and gain popular support. On the government’s part, rather than trying to

win the loyalty of citizens to undercut support for rebels, the strategy of group

control sought to demobilise non-combatants to demobilise rebels.

From the perspective of the regime in Kampala, the LRA posed a high threat as

an extension of a firmly established ‘lineage’ within Ugandan political society.

Precursors to the LRA rebellion reflected the dissatisfaction of ethnic Acholi leaders

over their removal from historically dominant position in state politics. Displaced

from the capital by Museveni’s forces in 1986, Acholi members of the former

regime fled north and regrouped as a rebellion called the Ugandan People’s

Democratic Army (UPDA). True to prediction, this group, which contained top

elites from the recently expelled regime, quickly entered into a political settlement

with the new regime once its resources ran dry. More significant, however, was the

decision of intransigent fragments of this group to form a series of rebel groups.

Alice Lakwena’s Holy Spirit Mobile Forces were viewed as bizarre outsiders and

met with a strategy of insurgent elimination.55 By the early 1990s, Joseph Kony’s

LRA emerged as the dominant rebel group.

Over the course of this transformation, rebellion in northern Uganda was

perceived by those in the capital as a problem that had not been solved by co-opting

the UPDA. Also, by the mid-1990s, the Sudanese government began supporting the

LRA to undermine Uganda’s support for the Sudan People’s Liberation Army

(SPLA). Complicating matters further was the changing profile of the LRA’s

organisation and leadership. While initially containing a number of cadres from the

former regime, Joseph Kony, a primary school dropout, was emblematic of a

political outsider with negligible connections to the country’s political establish-

ment. This status meant that the usual patronage tools to reintegrate rebel leaders

into the political establishment were less helpful for addressing this challenge. Still,

because the LRA fought in Acholiland, the group continued to reflect the broad,

historically contentious political relationship northern and southern elites.56 To

many in Museveni’s circle, these ‘outsider rebels’ seemed to aggravate a ‘northern

problem’ insofar as they posed the threat of attracting support from a broader

population that felt itself to be historically marginalised from the country’s politics,

an arena where Acholi elites were once dominant.

The Ugandan government pursued a strategy of group control against the LRA,

which included both peace talks and fighting them in the bush. Talks in 1993–94

showed that Museveni was willing to accept former government soldiers as

negotiators.57 A general amnesty bled the LRA of fighters. But Museveni was always

reluctant to deal with Kony in particular, whom he viewed as a political outsider.

The parallel approach sought to weaken LRA forces by militarising and isolating

northern Uganda with media blackouts and the process known locally as panda gari,

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 115

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

where soldiers swept the north engaging in arbitrary arrests, torture and extrajudicial

killings.58 Government officials also recruited local ‘Arrow Boys’ to augment the

army, a tactic that was abandoned as their recruits proved difficult to control. These

militias prompted LRA reprisals against non-combatants, drew the army into intra-

Acholi disputes, and Uganda’s government realised that the LRA’s persistence was

related to these political divisions within Acholi society.59

Following Museveni’s electoral victory in 1996, things took a population-centric

turn when the army began herding Acholi civilians into ‘protected villages’.60 This

tactic attempted to demobilise segments of Acholi society that were also weakening

government control over the area. Removing non-combatants from areas of LRA

operation made sense in this situation whereas in the other cases, including those

below, counter-insurgency strategies were best described as ‘political network-

centric’ in approach. But ‘protected villages’ were not a classic population-centric

strategy of providing security and services in order to win non-combatant support. In

moving Acholi civilians into camps and making services available through

international relief agencies, the strategy was designed to deprive LRA rebels of

sources of supply and give a free hand for the military to pursue rebels.61 This

drained the countryside of non-combatants and allowed the government to keep

careful watch over these people under the guise of protecting them.62 Interestingly,

in areas of LRA operations outside Gulu District, the UPDF did not forcibly displace

non-combatants. Outside Acholiland the army actively recruited local youth as

auxiliary forces, or ‘Local Defence Units’ (LDUs), to provide internal security while

the army pursued the rebels, which reflected the less politically ‘problematic’ nature

of non-Acholi areas. In summary, the strategy of group control meant that the

government’s war against the LRA provided a justification for ongoing state

oppression against the Acholi as a whole.

Insurgent Elimination

Of Uganda’s simultaneous conflicts, the one between the UPDF and the ADF in

Southwest Uganda bore the fewest traces of population-centrism. Based in the

mountainous borderlands with the Democratic Republic of Congo and backed by the

Sudanese government, the ADF was not required to cultivate ties to non-combatants

for support. Violence was as predatory as the LRA’s, but was interpreted by some as

a short-term signal of commitment to its patron.63 In response, the Ugandan

government went after the ADF with a full strategy of insurgent elimination. This

meant military operations that bypassed non-combatants and maintained an

orientation towards rebels that offered no patronage and no room for

accommodation.

For the Ugandan government, the ADF should have posed a high threat as

Sudan’s government sought to expand its strategy to destabilise Uganda by opening

up a new front from DRC.64 But the group’s organisational anatomy and the political

status of its members determined otherwise. The ADF was made from the detritus of

three rebellions that had no meaningful ties to Uganda’s political establishment.

First, the National Army for the Liberation of Uganda (NALU) grew in part from

CIVIL WARS116

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

disenfranchised veterans of 1950s-era Ruwenzururu revolt against colonial rule –

outsiders to Museveni’s NRM.65 Although former government official Amon Bazira

led the group, he had severed his political networks and was refused amnesty.66

Second, the National Democratic Army (NDA) was associated with a small crop of

army deserters and the urban ‘Bwaise’ gang culture.67 Finally, the Ugandan Muslim

Liberation Army (UMLA) drew from radicalised, urban Muslims linked to the

Tabliq Youth Movement.68 This group’s attack against the Buseruka police station in

February of 1995 was met with a swift military response.69

Remaining fighters scattered into Zaire and Sudan where they found anti-

Museveni networks orchestrated by the Sudanese Army Security Service.70

Although ADF operations were extensive, its leadership and agenda were never

clear.71 Critical in this analysis, these rebels did not present an alternative political

program to non-combatants. Rather than fighting to gain access to existing political

networks and the resources associated with them, the rebellion operated at the behest

of these outside interests. And although the group’s different factions may have

harboured varied grievances against the NRM regime, infighting was common-

place.72 The ADF was beholden to Khartoum’s regime’s political priorities, which

left these rebels without the capacity to develop their own local powerbase.

Due to these factors the ADF was, in contrast to the LRA, considered a low threat

to the NRM regime, which pursued a heavy-handed operation to eliminate the

group.73 Unlike the LRA campaign in Acholiland, the UPDF recruitment of LDUs

was systematic and militias were deeply integrated.74 Moreover, the conflict’s

175,000 internally displaced were not corralled into UPDF-run camps.75

Government-forced population movements were not seen as a distinct strategy to

deprive the ADF of a civilian support base or to keep watch over a specific ethnic

group that posed a threat to the government. What is more is that the ADF’s titular

leader and key outside organiser, Jamil Mukulu, exposed the group’s linkages to

transnational networks associated with Al Qaeda, enough to place the ADF on the

US Terrorist Exclusion List after 9/11.76 With the ADF now branded as a terrorist

organisation, Museveni had a public face to foreclose any accommodation with the

rebellion. The UPDF pursued the ADF across Uganda’s western border onto

Congolese soil until the group was declared defeated in 2007.77 More recently the

group has resurfaced in Eastern DRC but as an adjunct to Congolese rebels.78

Insurgent Control

Rebels from Uganda’s West Nile Region posed a moderate threat and were met with

the strategy of insurgent control. Both the West Nile Bank Front (WNBF) and the

Ugandan National Rescue Front II (UNRF II) were residual extensions of West

Nile’s fleeting dominance in Ugandan politics, albeit historically ‘two regimes

removed’ from the NRM and complicated by their relationship with their Sudanese

sponsor, who preferred malleable proxies. Operating out of Sudan they were outside

both the UPDF’s reach and the orbit of potential civilian support. They had little

concern for the day-to-day affairs of governing, and the refusal of ordinary West

Nilers to support them led to a general pattern of violence against non-combatants,

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 117

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

including forced recruitment of fighters. Insurgent control, this middle ground

between group control and insurgent elimination, targeted rebels militarily

alongside accommodation strategies that had a limited role for key non-combatants,

but not for West Nile’s general civilian population.

Although the West Nile region was politically peripheral to the NRM, its

rebellions posed a moderate threat as their leaders were drawn from Amin-era

politics that predated Museveni’s regime, and had a potential power base.79 During

Amin’s rule, the military was stocked with West Nilers and became the primary tool

of his personal authority.80 Shortly after Milton Obote overthrew Amin in the late

1970s, many of these officials fled the reconstituted government army, which

exacted reprisals on West Nile soldiers and civilians alike, making nearly half of the

region’s population refugees in Southern Sudan and Eastern Zaire.81 Out of these

camps grew rebellions that represented different ethnic factions of West Nilers: the

Kakwa and Madi-dominated Former Uganda National Army (FUNA) and the

Ugandan National Rescue Front (UNRF), comprised largely of Aringa. Both groups

fought back against Obote’s reinstated regime in the early 1980s, and also against

one another.82 As FUNA disintegrated, UNRF leader Moses Ali negotiated a power-

sharing arrangement with Museveni before he came to power.83 Thus, by the time

the NRM established itself in Kampala, there were ex-officials remaining in Sudan

and those that regained access to state institutions. In the 1990s, these two groups of

West Nilers coalesced around the WNBF and the UNRF II, respectively.

The WNBF’s leader was former Amin official Juma Oris, who after 1978 had

become alienated from other West Nile elites and fought as a mercenary for the

Sudanese army.84 As a rebel leader, Oris was more a middleman for the Sudanese

government, which fought the SPLA by pooling resources and shuffling fighters

between all of its proxies, including the LRA and ADF.85 The bulk of WNBF’s

fighters were unemployed youth that had spent years as refugees.86 Many upper rebel

cadres were former military officers and aside from shared experiences in Amin’s

regime or as refugees, ethnic factions from the Amin era persisted within the

WNBF’s organisation.87 These divisions manifested as confusion over the group’s

goals, ranging from reinstating Amin to power to declaring West Nile and

independent state.88 This tendency to factionalise was not lost on counterinsurgents,

which capitalised on the ability to peel off those seeking clemency despite Oris

having severed his own access to Uganda’s political networks.89

The UNRF II arose independently of the WNBF but still contained a small knot

of Amin-era functionaries that had closer political ties to the NRA by virtue of the

earlier absorption of the original UNRF. The failure of the NRA to honour certain

provisions of this settlement prompted a group of ex-rebels to decamp to Zaire and

organise the new rebellion.90 This time it was led by former army official Ali

Bamuze, who surrounded himself with Aringa functionaries drawn from Kampala’s

professional bourgeoisie.91 As one of the many Khartoum-backed rebellions, the

UNRF II also operated out of bases in South Sudan.92 Yet, its members explicitly

sought to negotiate themselves back into the state on better terms for themselves and

for West Nile as a whole, which eventually was reciprocated by Museveni.

CIVIL WARS118

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

The dynamics of the WNBF and UNRF II played into the hands of

counterinsurgents, which sought to stop both rebellions while ensuring their power

bases were properly managed within the context of Uganda’s ruling elite coalition.

Control of both WNBF and UNRF II focused on dealing with rebel leaders and their

options through both coercion and co-optation. This did not necessarily involve

winning the widespread support of non-combatants despite the region’s ongoing

political marginalisation and ripe conditions for political grievances against the

state. Still, the UPDF did not militarise the region nearly as much as they had in

neighbouring Acholiland at that time. Despite establishing a military presence in and

around West Nile’s main displaced camps, population movements were not

orchestrated by the army like they were in Gulu District. On the battlefield, the

rebels took advantage of the porous border with Sudan and used sanctuaries in

Eastern Equatoria from which to launch their attacks. In response the UPDF enlisted

the help of Museveni’s neighbouring allies, the anti-Mobutu Banyamulenge in Zaire,

and the SPLA in Sudan.93 An offer of a blanket amnesty supplemented this military

pressure, which haemorrhaged disgruntled fighters that were short of food and

medicine despite Sudanese resources.94 This amnesty provided opportunities for

fighters to join the UPDF and maintain rank.95 Defection left the WNBF vulnerable

to a rout by the SPLA in Yei by March of 1997.96

The control strategy made only limited use of LDUs.97 The UPDF granted some

requests from civilians for arms, but with some scepticism given the history of the

region and the disorganised nature of local militias.98 Even so, this approach

reflected the government’s ability to co-opt young men in its war effort, which had

the added benefit of being able to balance and control the potential flow of recruits

into rebel hands. On the civilian side, traditional West Nile authorities were hostile

to both rebellions and accepted the UPDF strategy.99 Having languished in refugee

camps throughout the early 1980s, this war-fatigued population simply did not wish

to participate directly, as they had been the targets of state repression at the hands of

Museveni’s predecessor. Thus, insurgent control meant leveraging collective

resistance to armed insurgency into brokering deals with rebel leaders, exemplified

by the Aringa-Obongi Peace Initiative Committee (AROPIC) that brought the

UNRF II into dialogue with the NRM.100

In summary, these cases illustrate the Ugandan government’s tendency to

downgrade the role of civilians in fighting rebels, an empirical reality that stands in

contrast to the dominant view of much of the counter-insurgency literature. If

civilians had mattered, one would expect to have seen a common population-centric

strategy irrespective of rebel group, but this was not the case. Not only were

population-centric strategies not critical to defeating Uganda’s rebels, the LRA case

demonstrated unsuccessful attempts to recruit local militias. Different tactics

reflected different strategies, a close look at which shows what mattered most to

Ugandan counterinsurgents was not non-combatants, but how useful patronage

could be depending on the rebels they fought. Some individual tactics may have

tangentially involved non-combatants, but strategies varied between groups (see

Table 2). The differences between these cases offer compelling evidence that the

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 119

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

threat level posed to incumbents, seen as a function of a rebellion’s political profile,

was most salient to counter-insurgency in Uganda. Strategies and the level of

patronage made available were associated with varying degrees of political

connections of rebel leaders to the NRM and to previous regimes.

LESSONS FROM AFRICA

African regimes like Uganda’s NRM consider themselves ‘in harm’s way’ to

varying degrees when considering counter-insurgency strategies. Rebels pose not

just military threats but fundamental political threats that vary in tandem with their

historical position within political society. This approach to understanding counter-

insurgency in contemporary Africa stands in contrast to conceptualisations of rebel

challenges as emanating from leaders and followers who are marginal to and distinct

from the regime politics that they seek to overthrow. While political outsiders are

found among Africa’s contemporary rebels (like the LRA), many groups contain key

former insiders from the prevailing political establishment. This tendency for rebels

to emerge from and fight to control patronage networks exposes them to regime

counter-insurgency strategies.

These strategies do not require strong state institutions and regimes often forego

efforts to protect and administer non-combatants. Instead, regime strategies focus on

selective uses of tactics to segregate, coerce, co-opt and demobilise rebel rivals and

in some cases, entire groups ostensibly represented by these rebels. Tactical moves

and strategic directions thus trace pre-war patron–client relationships, albeit

reshaped as regime and rebel forces fight for control over the resources and the

significant relationships that make up these networks. Therefore, the main contribution

of this study shifts attention away from civilians as the ‘centre of gravity’ of these

conflicts and towards the bare bones of regime politics in such countries. This

consideration illuminates an important dimension to counter-insurgency that is often

eclipsed by disproportionate attention paid to contemporary cases where third party

occupiers like the USA dominate the scene.

TABLE 2

UPDF COUNTERINSURGENCY APPROACHES

LRA ADF WNBF/UNRF II

Threat High Low ModerateCOIN strategy Group control Insurgent elimination Insurgent controlCOIN tactics Heavy regional

militarizationForced relocation ofnon-combatants

Use of militiaslimited to non-Acholi

Ongoing negotiationsAmnesty

Light regionalmilitarization

Limited use of militiasCross-border pursuit

Moderate regionalmilitarization

Use of militiasOngoing negotiationsthrough local leadersAmnesty

CIVIL WARS120

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

A key lesson of this examination is that effective counter-insurgency need not

focus on providing services or even protecting non-combatants, although there are

many good reasons to do these things if resources and political will permit.

Organisationally capable militaries also are not essential components of successful

counter-insurgency strategies that employ the political tools of patronage politics,

provided that the overall context of elite coalition splits and of rebel leaders who

fight for improved positions within this incumbent coalition occupy the attentions of

rebel recruits and crowds out would-be outsider challengers. Understanding how

counter-insurgency works under these particular conditions requires taking seriously

the centrality of regime politics, and in particular, the relationship between regime

insiders and their political networks, with rebel commanders and fighters. It is hard

to derive specific recommendations from ‘classics’ of counter-insurgency that are

based on experiences from other regions and from conflicts from decades earlier in

Africa. Specific knowledge of the personal backgrounds and connections of

individual rebel commanders and the real networks of political authority in weakly

institutionalised states provides more guidance.

This survey of the different kinds of rebel threats and corresponding counter-

insurgency strategies is built upon these important variables while also relying on

comparative case studies from Uganda to illuminate particularities in each case.

Further research should cast a wider net around a larger set of African cases. In

addition to gauging the infrequency of population-centric counter-insurgency

strategies, this future work ought to pay particular to attention to ‘off-the-path’

regime strategies that to do not necessarily correspond seamlessly with the political

profile of rebellions. For instance, the Khartoum regime’s war against the SPLA, a

rebel group that was by many measures comprised establishment Southern

Sudanese, was a protracted affair that often bordered on touching another category

of counter-insurgency – that of group elimination.101 The obverse was true in Sierra

Leone, where liberal internationalist peacemakers did not understand the nature of

the RUF rebellion as political outsiders.102 When attempts to shoehorn them into the

prevailing political establishment through a settlement failed, it became clear that

the preference of the regime in Freetown had always been insurgent elimination.103

Scholars and policymakers can apply such observations of the pitfalls of bad

strategic choices to contemporary efforts to manage rebellions in countries like Mali

and the Central African Republic, and in places like Somalia where the relationship

between regime politics and the organisation and behaviour of militias is less clear.

Lessons from Africa should be relevant to other places where state institutions

are weak and political struggle focuses on control over patronage networks.

Elements of these lessons appear in the evolution of counterinsurgent warfare in Iraq

and Afghanistan as field commanders learned that at least some segments of the

rebels that they faced were better understood in terms of their links to the existing

political establishment.104 Afghanistan’s President Ahmed Karzai repeatedly uses

his patronage relationships with local families to control areas and co-opt

challengers. American commanders frequently find that government appointees

undermine US military operations and are less than enthusiastic about building

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 121

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

strong state institutions. Karzai and his local political clients negotiate with rebels,

using promises of patronage to exploit the ambitions of some of their leaders and

cultivating these to exacerbate rivalries in rebel ranks.105

As the evidence in this essay shows, African regimes fight rebels in this way too, as

offers of patronage to attract andmanipulate the ambitions of rebel leaders is one of the

few tools that are within their grasp. Their success is surprising, given the numbers of

regimes in very weak and failing states that manage to survive rebel challenges in

contextswhere itwould seem tobe easy to convince groupsof non-combatants that they

should take up arms and overthrow their corrupt and violent rulers.

NOTES ON CONTRIBUTORS

Christopher Day is an Assistant Professor of Political Science at the College of

Charleston. He earned his PhD from Northwestern University in 2012. His research

focuses on civil wars, insurgent violence and the institutional role of armed actors in

African politics. Email: dayc@cofc.edu

William Reno is a Professor of Political Science and Director of the Program of

African Studies at Northwestern University. His current research focuses on the

organisational strategies and choices of armed groups that operate in socially

fragmented environments in Somalia and Afghanistan. Email: reno@northwestern.

edu

NOTES

1. David Kilcullen, The Three Pillars of Counterinsurgency, U.S. Government CounterinsurgencyConference, Washington, DC, 28 Sep. 2006; Robert Thompson, Defeating Communist Insurgency:Experiences from Malaya and Vietnam (New York: Praeger 1966); David Galula, Counter-insurgency Warfare: Theory and Practice (New York: Praeger 1964).

2. US Army & Marine Corps, Counterinsurgency Field Manual [FM3-24] (Chicago: University ofChicago Press 2007), esp. pp.34–52.

3. Stanley A. McChrystal, Commander’s Initial Assessment [‘The McChrystal Report’], (Kabul:International Security Assistance Force 30 Aug. 2009) I-1.

4. Mao Zedong, The Red Book of Guerrilla Warfare (El Paso: El Paso Norte Press 2010) pp.42–45;Ernesto Guevara, Guerrilla Warfare (New York: Vintage Books 1969) pp.32–35.

5. Boas Atzili, ‘When Good Fences Make Bad Neighbors: Fixed Borders, State Weakness, andInternational Conflict’, International Security 31/3 (Winter 2006/07) pp.139–173.

6. See Angel Rabasa, Steven Boraz, Peter Chalk, Kim Cragin, Theodore W. Karasik, Jennifer D. P.Moroney, Kevin A. O’Brien, John E. Peters, Ungoverned Territories: Understanding and ReducingTerrorism Risks (Santa Monica: RAND 2007).

7. Barack Obama, US Strategy toward Sub-Saharan Africa, (Washington, DC: White House Jun.2012), p.5.

8. Eric T. Young, ‘The Victors and the Vanquished: The Role of Military Factors in the Outcome ofModern African Insurgents’, Small Wars and Insurgencies 7/2 (Autumn 1996) p.178.

9. International Crisis Group, Central African Republic: Better Late Than Never, Africa Briefing no.96 (2 Dec. 2013).

10. William Reno, Warfare in Independent Africa (New York: Cambridge University Press 2011);Morten Bøas and Kevin Dunn, African Guerrillas: Raging against the Machine, (Boulder: LynneRienner 2007)

11. Data collected from Uppsala University’s Department of Peace and Conflict Research; http://www.pcr.uu.se/data/

CIVIL WARS122

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

12. Ton Hodges, Angola from Afro-Stalinism to Petro-diamond Capitalism (London: James Currey

2001).13. Lansana Gberie, A Dirty War in West Africa: The RUF and the Destruction of Sierra Leone,

(Bloomington: Indiana University Press 2005).14. See Richard Synge, Mozambique: UN Peacekeeping in Action, 1992-94 (Washington, DC: United

States Institute of Peace 1997).15. Okello Lucima (ed.), Protracted Conflict, Elusive Peace: Initiatives to end the violence in northern

Uganda (London: Conciliation Resources 2002) pp.29–31.16. Alex de Waal, ‘Counter-Insurgency on the Cheap’, London Review of Books 26/15 (5 Aug. 2004)

pp.25–27; Martin Shaw, ‘Darfur: Counter-insurgency, Forced Displacement and Genocide’, British

Journal of Sociology 62/1 (Mar. 2011) pp.56–61.17. See Pierre Englebert, Africa: Unity, Sovereignty & Sorrow (Boulder: Lynne Rienner 2009).18. Robert M. Cassidy, Counterinsurgency and the Global War on Terror: Military Culture and

Irregular War (Westport: Praeger Security International 2006).19. C.E. Callwell, Small Wars: Their Principles and Practice (London: General Staff – War Office,

1906): United States Marine Corps, Small Wars Manual, 1940; Charles Gwynn, Imperial Policing

(London: Macmillan 1936).20. Robert Thompson, Defeating Communist Insurgency: Experiences from Malaya and Vietnam

(London: Chatoo & Windus 1966); Frank Kitson, Low Intensity Operations: Subversion,

Insurgency, Peacekeeping (London: Faber and Faber 1971).21. John Nagl, Learning to Eat Soup with a Knife: Counterinsurgency Lessons from Malaya and

Vietnam (Westport: Prager 2002) pp.24–29.22. David Kilcullen, ‘Counterinsurgency Redux’, Survival 48/4 (Winter 2006/07) p.111.23. David J. Kilcullen, ‘Countering Global Insurgency’, The Journal of Strategic Studies 28/4 (Aug.

2005) pp.597–617; Frank G. Hoffman, ‘Small Wars Revisited: The United States and

Nontraditional Wars’, The Journal of Strategic Studies. 28/6 (Dec. 2005) pp.913–940.24. Kilcullen (note 22) pp.121–125.25. David Kilcullen, Counterinsurgency (Oxford: Oxford University Press 2012), pp.5–13.26. US Army & Marine Corps (note 2).27. Stathis Kalyvas, The Logic of Violence in Civil Wars, (New York: Cambridge University Press,

2006) esp. pp.198–206.28. See Christopher Paul, Colin P. Clarke, and Beth Grill, Victory Has a Thousand Fathers: Sources of

Success in Counterinsurgency (Santa Monica, CA: RAND Corporation, MG-964-OSD, 2010).29. Ibid., p.xxii.30. Ibid., pp.32–81.31. Richard Reid, War in Pre-Colonial Eastern Africa (Kampala: Fountain 2007).32. Robert B. Edgerton, Africa’s Armies: From Honor to Infamy. A History from 1791 to Present

(Boulder: Westview Press 2002).33. John Harbeson (ed.), The Military in African Politics (New York: Praeger 1988).34. Samuel Decalo, Psychoses of Power: African Personal Dictatorships (Gainsville: Florida Academic

Press, 2nd edition, 1998).35. George Klay Kieh, Jr. and Pita Ogaba Agbese, The Military and Politics in Africa: From

Engagement to Democratic and Constitutional Control (Burlington: Ashgate 2004) pp.1–16.36. Herbert Howe, Ambiguous Order: Military Forces in African States (Boulder: Lynne Rienner 2001);

Benedikt Franke, Security Cooperation in Africa: A Reappraisal (Boulder: Lynne Rienner 2009);

Paul Williams, The African Union’s Conflict Management Capabilities, Council on Foreign

Relations Working Paper (Oct. 2011).37. Zachariah Mampilly, Rebel Rulers: Insurgent Governance and Civilian Life During War (Ithaca:

Cornell University Press 2011); Claire Metelits, Inside Insurgency: Violence, Civilians, and

Revolutionary Group Behavior (New York: New York University Press 2009).38. Roger D. Petersen, Resistance and Rebellion: Lessons from Eastern Europe (New York: Cambridge

University Press 2001); Elisabeth Jean Wood, Insurgent Collective Action and Civil War in

El Salvador (New York: Cambridge University Press 2003); Norma J. Kriger, Zimbabwe’s

Guerrilla War: Peasant Voices (New York: Cambridge University Press 1992); Timothy

P. Wickham-Crowley, Guerrillas and Revolution in Latin America: A Comparative Study of

Insurgents and Regimes Since 1956 (Princeton: Princeton University Press 1992).

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 123

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

39. Amilcar Cabral, ‘Brief Analysis of the Social Structure in Guinea’, Richard Handyside (trans. & ed.)

Revolution in Guinea: Selected Texts by Amilcar Cabral (New York: Monthly Review Press 1969)

p.59.40. Nelson Kasfir, ‘Guerrillas and Civilian Participation: The National Resistance Army in Uganda,

1981-86’, The Journal of Modern African Studies 43/2 (Jun. 2005) pp.271–96.41. Mary Kaldor, New & Old Wars: Organized Violence in a Global Era (Stanford: Stanford University

Press 2001); Mark Duffield, Global Governance and the New Wars: The Merging of Development

and Security (London: Zed Books 2001).42. Paul Collier and Anke Hoeffler, ‘Greed and Grievance in Civil War’ Oxford Economic Papers 56/4

(2004) pp.563–95; Mats Berdal and David M. Malone,Greed and Grievance: Economic Agendas in

Civil Wars (Boulder: Lynne Rienner 2000); Paul Collier, Economic Causes of Civil Conflict and

Their Implications for Policy, (Washington, DC: World Bank 2000); Michael Ross, ‘What do we

know about Natural Resources and Civil Wars’, Journal of Peace Research 41/3 (Summer 2000)

pp.337–56; David Keen, ‘The Economic Functions of Violence in Civil Wars’, Adelphi Paper 303

(Oxford: Oxford University Press for the International Institute for Strategic Studies 1996).43. Jeremy Weinstein, Inside Rebellion: The Politics of Insurgent Violence. (New York: Cambridge

University Press 2007) p.10.44. Nathan Leites and Charles Wolf, Jr., Rebellion and Authority: An Analytic Essay on Insurgent

Conflicts (Chicago: Markham 1970).45. Kimberly A. Maynard, Healing Communities in Conflict: International Assistance in Complex

Emergencies (New York: Columbia University Press 1999), pp.107–122.46. Roger Peterson, Understanding Ethnic Violence: Fear, Hatred and Resentment in Twentieth-

Century Eastern Europe (New York: Cambridge University Press 2002), p.40.47. Ibid., p.25.48. Christopher Clapham, Africa and the International System: The Politics of State Survival

(Cambridge: Cambridge Univeristy Press 1996), p.59.49. See Edward L. Gibson, ‘Boundary Control: Subnational Authoritarianism in Democratic Countries’,

World Politics 58/1, (Oct. 2005) pp.101–32; Catherine Boone, Political Topographies of the

African State: Territorial Authority and Institutional Choice (New York: Cambridge University

Press 2003); Joel Migdal (ed.), State in Society: Studying How States and Societies Transform and

Constitute One Another (New York: Cambridge University Press 2001).50. Patrick Chabal and Jean-Pascal Daloz, Africa Works: Disorder as Political Instrument

(Bloomington: Indiana University Press 1999) pp.4–16.51. Michael Bratton and Nicholas van de Walle, ‘Neopatrimonial Regimes and Political Transitions in

Africa’, World Politics 46/4 (Jul. 1994), p.459.52. Chabal and Daloz (note 50) p.13.53. Republic of Uganda, The Amnesty Commission: Peace and Reconciliation in Action. Report 2007–

2008 (Kampala: The Amnesty Commission), p.14.54. For a comprehensive overview of the conflict, see Tim Allen and Koen Vlassenroot, The Lord’s

Resistance Army: Myth and Reality (London: Zed Books 2010); Sverker Finnstrom, Living With Bad

Surroundings: War, History, and Everyday Moments in Northern Uganda [The Cultures and

Practices of Violence] (Durham: Duke University Press 2008); Zachary Lomo and Lucy Hovil,

Behind the Violence: The War in Northern Uganda, Institute for Security Studies Monograph Series,

no. 99 (Mar. 2004); Ruddy Doom and Koen Vlassenroot, ‘Kony’s Message: A New Koine? The

Lord’s Resistance Army in Northern Uganda’, African Affairs 98 (1999) pp.5–36.55. See Heike Behrend, Alice Lakwena and the Holy Spirits: War in Northern Uganda 1985-1997

(Athens: Ohio University Press, 2000); Tim Allen, ‘Understanding Alice: Uganda’s Holy Spirit

Movement in Context’, Africa 61/3 (1991) pp.370–399.56. See A.G.G. Ginyera-Pinycwa, Northern Uganda in National Politics (Kampala: Fountain, 1992); ‘Is

There a “Northern Question”?’ in Kumar Rupesinghe (ed.) Conflict Resolution in Uganda (London:

J. Currey, 1989).57. Lucima (note 15) p.35.58. Doom and Vlassenroot (note 54) pp.23–34; ‘NRA Mops Up in the North’, New Vision 18 Mar.

1991; ‘NRA Launches Search Operation in the North’, New Vision 3 Apr. 1991.59. Confidential Interview with UPDF Brigadier General, Dec. 2013.60. ‘Saleh Announces North offensive’,New Vision 20 Jun. 1996; ‘Gov’t to Establish Protected Villages

in Acholi’, Sunday Vision 29 Sep. 1996.

CIVIL WARS124

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

61. See Adam Branch, Displacing Human Rights: War and Intervention in Northern Uganda (Oxford:Oxford University Press 2011); Christopher Dolan, Social Torture: the case of northern Uganda,1986-2006 (Berghahn Books 2009).

62. ‘Army to Deply All Over Acholi’, New Vision 20 Jul. 1996.63. Lucy Hovil and Eric Werker, ‘Portrait of a Failed Rebellion: An Account of Rational, Sub-Optimal

Violence in Western Uganda’, Rationality and Society 17/1 (Feb. 2005) pp.5–34.64. ‘Why Does Zaire, Sudan Team Up against Uganda?’ New Vision 4 Dec. 1996.65. ‘Kasese Insecurity: A Historical View’, New Vision 21 Sep. 1992; also see A. Syahuka-Muhindo,

The Rwenzururu Movement and the Democratic Struggle, Center for Basic Research Working Paperno. 15, Jun. 1991; Bamusede Bwambale, The Faces of the Rwenzururu Movement (Kampala:Unknown binding 2001).

66. ‘Bazira Can’t Get Amnesty’, New Vision 13 May 1993.67. ‘Who Is Maj. Itongwa the “Federo Rebel Leader”?’ Sunday Vision 5 Mar. 1995.68. Sallie Simba Kayunga, Islmamic Fundamentalism in Uganda: A Case Study of the Tabligh Youth

Movement, Centre for Basic Research Working Paper no. 37, Sep. 1993, p.71.69. ‘Thugs Attack Hoima Police’, New Vision 23 Feb. 1995; ‘NRA Kills 63 Hoima Rebels’, The

Monitor, 27 Feb. 1995.70. Gerard Prunier, From Genocide to Continental War: The ‘Congolese’ Conflict and the Crisis of

Contemporary Africa (London: Hurst 2009) p.86.71. ‘ADF Command: Who is Who’, New Vision 5 Oct. 1998; ‘The ADF: Rebels without a Cause’, The

Monitor 10 Dec. 1999.72. ‘ADF Factions Clash’, New Vision 11 Jan. 1999; ‘ADF Rebels Execute 20’, New Vision 19 Feb. 1999.73. ‘Using Hammer on Western Rebels, but Small Stick on Kony & Co.’, The Monitor 21 Nov. 1996.74. Interview, UPDF Brigadier General, Dec. 2013.75. African Rights, Avoiding an Impasse: Understanding the Conflict in Western Uganda (Dec. 2001) p.1.76. ‘Bin Laden Trains ADF’, New Vision 18 Oct. 2001; ‘Osama’s Man’, The Sunday Monitor 28 Oct.

2001.77. Christopher R. Day, ‘The Fates of Rebels: Insurgencies in Uganda’, Comparative Politics 43/4 (Jul.

2011) p.447.78. Kristof Titeca and Koen Vlassenroot, ‘Rebels without borders in the Rwenzori borderland?

A biography of the Allied Democratic Forces’, Journal of East African Studies 6/1 (Feb. 2012)pp. 154–76. See also International Crisis Group, Eastern Congo:The ADF-NALU’s Lost Rebellion,Africa Briefing no. 93 (19 Dec. 2012).

79. See Mark Leopold, Inside West Nile: Violence, History & Representation on an African Frontier(Oxford: James Curry 2005).

80. Amii Omara-Otunnu, Politics and the Military in Uganda, 1890-1985 (NewYork: St. Martin’s Press1987) pp.125, 130, 133–35.

81. Barbara Harrell-Bond, Imposing Aid: Emergency Assistance to Refugees (Oxford: OxfordUniversity Press 1986).

82. See Bernard Rwehururu, Cross to the Gun: The Fall of Idi Amin and the Ugandan Army (Kampala:Netmedia 2008).

83. Refugee Law Project, Negotiating Peace: Resolution of Conflicts in Uganda’s West Nile Region,Refugee Law Project Working Paper no. 12 (Jun. 2004), pp.6–7; ‘UNRF – Brigadier Moses Aigiven Min. of Tourism & Wildlife’, Focus 29 Aug. 1986.

84. ‘Who is behind the West Nile rebellion?’ Sunday Vision 17 Sep. 1995.85. ‘Oris Rebel Officer Reveals Sudan Role’, New Vision 3 May 1996; ‘West Nile Sold Off – Oris

Rebels’, New Vision 12 Feb. 1996.86. ‘Rebels Recruit in West Nile’, Sunday Vision 28 May 1995; ‘Why West Nile Youth Join Rebels’,

The Crusader 28 Jun. 1996.87. ‘Oris Men: Tribal Split’, New Vision 28 Aug. 1996.88. ‘Rebels Want Idi Amin president’, New Vision 29 May 1995; ‘Oris Rebels Want West Nile State’,

New Vision 10 Jun. 1995.89. ‘Oris Deputy Appeals to Museveni for clemency’, New Vision 8 Apr. 1997.90. ‘Museveni’s Dishonesty Caused War, Says Former Rebel’, The Crusader 17 Oct. 1996.91. Interview with former UNRF II functionary, Kampala, Aug. 2011.92. ‘Oris, Kony Forces Form United Front’, New Vision 5 Apr. 1997.93. ‘SPLA Battles Over Oris HQ’, New Vision 30 Dec. 1995; ‘SPLA Pound Oris Rebels’, Sunday Vision

4 Feb. 1996; ‘Banyamulenge Eject Oris’, New Vision 3 Mar. 1997.

AFRICAN COUNTER-INSURGENCY AND PATRONAGE POLITICS 125

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014

94. ‘160 Oris Rebels Surrender in Arua’, New Vision 17 Apr. 1996; ‘Oris Men Surrender’, New Vision 7Feb. 1997.

95. ‘Museveni to Recruit Oris Rebels’, New Vision 15 Oct. 1996; ‘UPDF Absorbs 500 Rebels as 70 getAbducted’, The Monitor 6 Nov. 1996; ‘517 Oris Rebels to Join UPDF’, New Vision 8 Nov. 1996.

96. ‘2,000 Killed in Bloody Ambush’, New Vision 17 Mar. 1997; Prunier (note 70) p.133.97. Confidential Interview with UPDF Brigadier General, Dec. 2013.98. ‘Arua Residents Get Guns’, New Vision 18 Oct. 1996; ‘Villagers to Fight Oris Rebels’, New Vision 4

Dec. 1996; ‘Civilians Capture WNBF Commander, The Monitor 7 Jan. 1997.99. Confidential interview with son of Arua elder, Jan. 2008.100. Refugee Law Project, pp.20–25; also see Government of Uganda, The Peace Agreement Between

The Government of the Republic of Uganda and The Uganda National Rescue Front II, 24 Dec.2002.

101. See Douglas Johnson, The Root Causes of Sudan’s Civil Wars: Peace or Truce (London: JamesCurrey 2012).

102. See Paul Richards, Fighting for the Rainforest: War, Youth, and Resources in Sierra Leone(Bloomington: Indiana University Press 1996).

103. Government of Sierra Leone, Witness to the Truth: Report of the Sierra Leone Truth &Reconciliation Commission, 3A (Freetown 2004) pp.364-456.

104. An early example of this is Lt Col Chris Gibson, ‘Battlefield Victories and a Strategic Success: ThePath Forward in Iraq’, Military Review (Sep./Oct. 2006) pp.47–59. One of the better books is BingWest, Strongest Tribe (New York: Random House 2008).

105. Carter Malkasian, War Comes to Garmser: Thirty Years of Conflict on the Afghan Frontier (NewYork: Oxford University Press 2013).

CIVIL WARS126

Dow

nloa

ded

by [

Col

lege

Of

Cha

rles

ton]

, [C

hris

toph

er D

ay]

at 0

6:20

20

Aug

ust 2

014