Victims, Perpetrators, or Both?
Militias during Civil War in Peru and Sierra Leone
Witold Mucha
Institute for Development and Peace (INEF), Duisburg, Germany
Paper presented at 7 ECPR General Conference,
Bordeaux, September 4 to 7, 2013
Abstract
Peace and conflict research has assessed people’s motivation to take up arms predominantly
by the distinction between ‘justice-seeking’ and ‘loot-seeking’ incentives. Moreover, the majority of studies have based their assumptions on a classic conflict dyad perspective constituted by either the state fueling escalation processes on the one hand or an armed insurgent
group on the other. These two assumptions combine a research gap that will be focused on in
this paper: namely, the academic neglect of those societal forces that are neither the state nor
the rebel group, which however become entangled in internal armed conflict – for reasons
beyond justice or loot alone. This paper will fill this academic void by taking a comparative
look into the pivotal role of self-defense militias during civil war in Peru (1980-1995) and
Sierra Leone (1991-2002). Both civil wars have been discussed as ‘grievance’-driven insurgencies evolving into ‘greed’-based conflicts. Covering fueling as well as inhibiting factors
of internal armed conflict, two issues will be of particular interest: first, the state’s cooptation strategy will be analyzed with respect to the means (e.g. military training) and the
overall impact on the escalation onset (i.e. termination vs. prolongation). Second, the unintended negative effects on the security situation on the ground will be looked into as well. It
will be shown that the state’s instrumentalization of self-defense militias helped defeating the
insurgent group and thus ending civil war in the long run. However, the empowerment of
militias also added fuel to violence in short term as the security vacuum on the ground was
taken advantage of by settling old scores with neighboring communities which were identified as ‘state enemies’.
Work in Progress. Please do not cite without author’s permission.
Contact: [email protected]
1. INTRODUCTION
More than two years after the beginning of the heralded ‘Arab Spring’, policymakers and
academics have begun to critically assess the different outcomes of popular uprisings. While
the supposed success stories from Tunis and Cairo of early 2011 initially led decision-makers
(re-)believe in the one-directional power of democratization, civil wars in Libya and Syria
brought these hopes back down to earth. From a peace and conflict studies perspective, the
different paths shed light on the prevailing difficulties among policy-makers and researchers
to grasp the timing and intensity of violence as well as the multitude of stakeholders driving
the escalation processes (Gause III 2011; Anderson 2011). In particular, two conflict actor
types stand out in that regard: while the state has seldom been looked at as force that directly
intensifies violence on the ground (Davenport and Inman 2012; Lyall 2009; Herreros 2006;
Mucha 2013a), militias mostly went under the radar in contemporary peace and conflict research (Barter 2013; Carey, Mitchell and Lowe 2013; Ahram 2011).1 These academic voids
are striking because of two things: first, particularly the course of uprisings during the ‘Arab
Spring’ has demonstrated that at the end of the day it is the state that decides upon the intensity and duration of violence escalation. Each of the mentioned uprisings started out as peaceful protest driven by a heterogeneously composed movement. The respective states’ reaction
– i.e. negotiation or repression – eventually determined whether the opposition radicalized
their agendas or not. Second, as particularly the cases of Libya and Syria showed the longer
and intense the escalation proceeded the more tangled up the setting on the ground became;
especially when the clashes occurred far off the capital city areas. For instance, the volatile
post-conflict transition in Libya is set back by a number of autonomous militias that had
emerged out of the revolt against Qaddafi in 2011 (Kirkpatrick 2013; Pelham 2012).
In short, peace and conflict studies have struggled grasping internal armed conflict settings
(ex ante and ex post) because of an incomplete assessment of the state as escalating force on
the one hand and the blurred actor constellation on the ground on the other. This paper will
attempt to account for both shortcomings by taking a comparative look into the pivotal role of
militias during civil war in Peru (1980-1995) and Sierra Leone (1991-2002). The focus will
be on those societal forces that are neither the state nor the rebel group, which however become entangled in internal armed conflict – for reasons beyond justice or loot alone. The cases of Peru and Sierra Leone are particularly useful due to three circumstances: first, despite
their different political, economic, ethnic and cultural preconditions, both civil wars have
been discussed as ‘grievance’-driven insurgencies evolving into ‘greed’-based conflicts. In
both contexts, militias became the feather that broke the balance towards the state. Second,
before acquiring that pivotal relevance, by the means of military assistance and training the
very militias were de facto integrated by the state into civil war. Third, from a methodological point of view, the case selection is useful given the huge empirical material provided by
the respective truth and reconciliation commissions (CVR 2003; TRC 2004).
1
In this paper, the ‘state’ will not be dealt with as a homogenous entity alone. Constituted by different ‘bits of
state’, such as for instance the government, different intra-government factions, the army leadership and different factions within, the police forces and so on, the term ‘state’ is all-encompassing. It will be further differentiated in the empirical analysis below (see section 4).
1 This paper will be divided in five sections. Following the introduction, second, a literature
review will present the academics’ blurred notion of militias. A few words will be also given
on the cases of Sierra Leone and Peru and the way these have been discussed in peace and
conflict studies (section 2). Third, tentatively distinguishing from prominent concepts by
Weinstein or Kalyvas, a framing model for analysis will be presented that grasps conflict
actors beyond the classic dyad thinking of either the state or the rebel group (section 3).
Fourth, at the core of this paper, the cases will be summarized with regard to the conflict genesis and outcome. On the one hand, the commonalities of state instrumentalization will be
shown; while on the other, the unintended effects of this policy on the ground will be illustrated. Applying a dual escalation perspective that covers fueling as well as inhibiting factors
of internal armed conflict, two issues will be of major interest: first, the state’s co-optation
strategy will be analyzed with respect to the means (e.g. military training) and the overall
impact on the escalation onset (i.e. termination vs. prolongation). Second, the indirect effects
on the local security situation will be looked at (section 4). At last, the findings will be concluded and critically discussed with regard to the generalization of results (section 5). It will
be shown that the state’s instrumentalization of self-defense militias helped defeating the insurgent group and thus ending civil war in the long run. However, the empowerment of militias also added fuel to violence in short term as the security vacuum on the ground was taken
advantage of by settling old scores with neighboring communities which were identified as
‘state enemies’.
2. LITERATURE REVIEW
Militias are predominantly associated with as government proxies. This is not surprising given that non-state actors that support the state’s opposition (e.g. recruits, shelter, intelligence)
are quickly pigeonholed as insurgents, terrorists or else. The most general understanding of
militias is that of an irregular armed group that is at least tolerated by the governing power.
Despite its sometimes spontaneous genesis it operates on behalf of the regime. However, in
the course of conflict militias might make themselves independent and pursue an own agenda.
Schneckener (2007) summarizes: “Their task is to fight rebels, to threaten specific groups or
to kill opposition leaders. These militias are often created, funded, equipped and trained in
anti-guerilla tactics (counter-insurgency) by state authorities” (Schneckener 2007: 11). Beyond the close relationship with the government, Schultz (2005) points to the amorphous
character of militias: “[The] members often come from the under classes and tend to be composed of young males who are drawn into this milieu because it gives them access to money,
resources, power, and security. Not infrequently, members are forced to join” (Shultz 2005:
17). Militias have been identified to intensify conflict settings given their often unpredictable
agenda and changing loyalty. Hence, governments and external actors have frequently tried to
win militias over for their purposes; be it by giving economic incentives or simply by training
so-called self-defense units (Schneckener 2007: 16-18).
Defining Militias
While that definition helps to specify the notion of what militias are supposed to look like –
e.g. ‘paramilitary’, ‘death squads’, ‘irregular force’, or ‘vigilante’; however, it does not say
2 much about the relationship towards the conflict actors on the ground such as the state and/or
the rebel group(s). Given the focus of this paper on the instrumentalization of militias by the
state, two definitional challenges need to be met: first, in what ways can militias be distinguished from rebel groups and second, in how far are states per se capable of influencing the
agendas of militias? Barter (2013) discusses the conceptual difficulties:
“While by no means straightforward, identifying rebel groups presents fewer challenges,
in part because rebel forces may be addressed in terms of a single relationship: they are
armed groups which oppose the state. Defining anti-rebel forces is more difficult, not only because political biases run high, but also because there are two key relationships involved: the state/militia relationship and the rebel/militia relationship. While the behavior
of rebel groups is largely conditioned by the behavior of state forces, anti-rebel groups
will be influenced by the behavior of states as well as rebels. (…) [Viewing] militias as
proxies is misleading, principally because the very weak states which rely on militias are
those which are least capable of determining their agendas” (Barter 2013: 76-77).
Given the variety of definitions, understandings, and synonyms on the one hand and the interest in the instrumentalization of militias on the other, this paper will use a broad definition:
militias are non-state armed groups with some level of organization that apply violence
against perceived enemies. The moment they are sponsored by the government without being
part of the regular security forces, they become pro-government militias (PGMs) (Barter
2013: 77; Carey, Mitchell and Lowe 2013: 250; Ahram 2011: 17).2 This broad notion does
not only facilitate the comparison of the diverse empirical contexts of Peru and Sierra Leone.
Vigilantes, paramilitaries, or self-defense organizations all fit well into the overarching category. More important, it also allows for tracing the change of militias over time. As the cases
of Peru and Sierra Leone will show, self-defense militias long existed before the state’s
instrumentalization attempt. The more they were co-opted and armed the more autonomous
and unpredictable they became. Looking into the change of militias over time (i.e. prior, during and post-conflict), this paper will differ from the majority of studies that primarily deal
with PGMs alone. Barter (2013) elaborates further on the blurred nature of militias:
“Such groups typically blur public/private as well as military/civilian distinctions, undermine state (and rebel) efforts to monopolize violence, and are rarely accountable for
their actions. Their ranks include groups created by the state, groups recognized by the
state, and groups which may resist both state and rebel forces” (Barter 2013: 77).
State-centrist Perspective on Militias
Much has been written on the impact of structural causes on internal armed conflict (e.g. natural resources, horizontal inequality, or ethnicity) (Collier and Hoeffler 2004; Fearon and
Laitin 2003; Hegre and Sambanis 2006). However, only little attention has been paid to the
agency that drives escalation processes (Davenport and Inman 2012; Lyall 2009; Herreros
2006; Mucha 2013a; Mucha 2013b). In particular, quantitative research has been struggling
2
This focus does not imply that militias utilizing violence against state troops is turned a blind eye on. Likewise
the fact will not be ignored that militias take up arms in some areas due the lack of statehood and provision of
security on the ground.
3 in grasping the actors on the ground. Although recent studies have analyzed the fragmentation of rebel groups, for the sake of feasibility by econometric means, the government has
been grasped as a unitary stakeholder alone (Wood 2010; Humphreys and Weinstein 2008;
Bakke, Cunningham and Seymour 2012; Cederman and Gleditsch 2009; Cunningham 2011).
As qualitative research suggests, this static conceptualization is often misleading (Reno 2007,
2009, 2010; Wood 2010; Carey, Mitchell and Lowe 2013). For instance, the difficulties of
understanding post-conflict Libya well illustrates the shortcomings in contemporary peace
and conflict studies that predominantly rely on large-n frameworks. Carey, Mitchell and
Lowe (2013) conclude: “Despite the effect that pro-government militias can have on the political, economic, and social stability and security of civilians, (…) there is little quantitative
research on these groups. This research lags behind the case study literature in analyzing the
impact of these groups (…)” (Carey, Mitchell and Lowe 2013: 250).
Indeed, a lot of case studies have been done on militias; particularly by researchers working
on Latin America (Stanley 1996; Centeno, 2002; Mazzei, 2009). More recent idiographic
studies have shifted the regional focus to Kenya (Nyabola 2009), Nigeria (Ikelegbe 2006),
Sierra Leone (Hoffman 2007; Mitton 2012; Vincent 2013), Sudan (Kwaja 2010; Maitre 2009),
Somalia (Bakonyi 2013; Verhoeven 2009), Uganda (Mkutu 2008), and Indonesia (Bertrand
2004; Giblin 2007; Wilson 2011). Given the academic community’s increased interest in the
‘Arab Spring’ aftermath, a few recent studies have been also done on militias in Libya (Pelham 2012; Lutterbeck 2013; Chivvis et al. 2012) and Yemen (Brandt 2013). However, as
Carey, Mitchell and Lowe (2013) find: “[there] are currently no systematic measures of these
informal violent organizations that act on behalf of the government” (Carey, Mitchell and
Lowe 2013; 250).
The difficulties to grasp militias in general and PGMs in particular is not solely because of
the disproportionate reliance on quantitative analysis. Qualitative research has been likewise
confronted with challenges in understanding why militias form and behave in one way or the
other. Similar to the aforementioned shortcomings of studies focusing on structural causes of
internal armed conflict, the state-centrist perspective on militias has been criticized most in
literature; primarily, for viewing militias as state proxies alone rather than autonomous and
distinct actors. Accordingly, the motivations of militias are not dealt with in particular or at
all (Barter 2013: 80). For instance, Mason and Krane (1989) found that death squads are organized by weak states in order to put down popular uprisings. Simlarly, Mazzei (2009) argues that death squads are used strategically to “counter reform efforts” (Mazzei 2009: 5).
According to both authors, the state’s rationale is that launching death squads against potential rebellions is cheaper than dealing with structural grievances by conventional means (e.g.
economic reforms, redistribution of wealth etc.) (Mason and Crane 1989; Mazzei 2009).
Ahram (2011) identifies external threats or their absence as crucial factors for the genesis and
initiation of militias. Regimes confronted with hostile governments in adjacent countries are
more likely to rely on conventional armed forces. In turn, PGMs are established by governments when there is no imminent threat by regional neighbors. The economic costs of maintaining militias are brought forward as major argument (Ahram 2011).
4 It is not surprising that this state-centrist view on militias has predominantly pilloried the
government as perpetrator of human rights abuses. Only few studies have been published so
far that cover systematic violence committed by militias. For instance, these include reports
by Human Rights Watch (HRW) on the situation in Afghanistan, Côte d’Ivoire, the Philippines, and most recently in Libya (Human Rights Watch 2011a, 2011b, 2012, 2013; Barter
2013: 80). Perceiving militias as state proxies alone makes it difficult to understand formations that are driven by different motivations. Carey, Mitchell and Lowe (2013) point to
the informal and semi-official nature of militias:
“In some instances the government tries to keep the group at arm’s length, while in other
cases governments openly create, train, and pay such groups. For example, village defense forces that have been created by governments fall under this second category,
where the link to the PGM is far more open and institutionalized than in the case of the
Janjaweed” (Carey, Mitchell and Lowe 2013: 251).
The variety of militias is well illustrated by studies on Latin America. For instance, contrary
to PGMs in Chiapas or El Salvador, Mazzei (2009) found that paramilitaries in Colombia
partly emerged in response to rebel violence. Huggins (1991) points to the unpredictable
character of militias. In the case of Guatemala, PGMs became so autonomous that they eventually harmed the state forces (see also Mitchell 2004). The Colombian onset seems even
more blurred. Against the backdrop of a fifty-year-long civil war, to some extent observers
can no longer track whether PGMs serve the state’s interests, their own, or the drug cartels
(Gutiérrez Sanín and Jaramillo 2004; Palacio Castanâneda 1991). The analysis is even more
complicated the more local and remote the dynamics on the ground get. In particular, the case
of Peru well demonstrates the blurred and constantly changing militia environment. In short,
so-called rondas campesinas (i.e. peasant rounds) existed long before the expansion of the
guerilla; primarily serving as local security providers for instance against cattle rustlers. With
both the guerilla and the state forces consolidating in the affected areas, the rondas
campesinas were enticed but also threatened to join either of the two stakeholders. Juxtaposing militias in the North and the South, Starn (1999) argues that the latter were co-opted by
the state in the war against the guerilla while Fumerton (2001) underlines the communal
ownership of self-defense as crucial variable for mobilization (see also Krujit 1999). As the
empirical analysis will show, not all forces opposing rebel organizations necessarily serve the
state.3 In other words, “[being] anti-rebel is not the same as being pro-state, an assumption
which dominates the literature” (Barter 2013: 79).
The ambivalent outcome of militias in Peru and Sierra Leone has not been systematically
assessed by contemporary peace and conflict studies. Except for some studies (Degregori
1996; Fumerton 2001; Mason and Campany 1995; Starn 1995; Zahar 2000; Boas and Hatloy
2008; Ferme and Hoffman 2004; Hoffman 2003, 2004, 2006, 2007; Millar 2012; Moeller
2006; Richards 2002), the cases were either related to the ‘greed’ debate or in the case of Sierra Leone to post-conflict reintegration measures, security sector reforms in general and
child soldiers recruitment in particular (Felbab-Brown 2005; Cornell 2005, 2007; Dreyfus
3
Even more so when rebel movements share more ethnic affiliation with local militia-like organizations than
state forces do (Fielde and Nilsson 2012; Bakonyi 2013).
5 1999; Weinstein 2003; Hyland 2008; Van Dun 2013; Zack-Williams 1999, 2006; Ross 2004a,
2004b; Collier, Hoeffler and Rohner 2009; Lujala, Gleditsch and Gilmore 2005; Berdal and
Malone 2000; Le Billon 2001; Murphy 2003; Shepler 2005; Maclure and Denov 2006;
Denov and Maclure 2007; MacMullin and Loughry 2004; Williamson 2006).
3. FRAMING MODEL
Having presented the manifold nature of militias, the following section will present a framing
model that specifies the understanding. It will be the goal to systematize militias along their
motivations in joining the state’s fight against the oppositional irregular armed group. The
premise is that the question of state instrumentalization can be better identified on the basis of
this model (in terms of incentives, threats etc.). Given the shortcomings of state-centrist perspectives (i.e. ‘militias as proxies’), this paper will diverge from common approaches. Understanding militias as actors on their own, they will be qualitatively assessed along four interdependent and overlapping indicators: initiation, resources, direction, and accountability.
Initiation
Partly following prominent work by Weinstein (2005, 2007) and Kalyvas (2004, 2006), the
initiation (or synonymously origin) of militias is predominantly driven by economic rationale.
As such, before deciding whether or not to organize armed struggle against the state’s enemy,
militias and their respective leaders will run through the options first. What is effectively in
for the specific community the militia is based on: is the state’s demand for cooperation tied
to military and economic assistance (i.e. exogenous initiation) or do we have to defend ourselves against rebel attacks (i.e. endogenous initiation)? Assuming the state is capable of
providing services, the threshold to join the state forces will be low. While this onedirectional relationship represents the majority of cases, particularly the Peruvian case has
been determined by a strong guerilla initially providing more services to local communities
than the national government.
Resources
In order to fight against rebels on a sustained and not merely self-reliant basis, militias depend on a set of resources that goes beyond military equipment alone. Partly applying
Ohlson’s (2008) Triple-R concept, resources can be divided in three dimensions: first, these
include ‘military capabilities’ such as weapons as well as financial means to sustain the actions. Second, ‘organizational capabilities’ such as networking are found necessary to mobilize collective actors. In addition to those two traditional resource dimensions, third, Ohlson
(2008) adds the so-called ‘opportunity structure’. Accordingly, resources can be determined
by external constraints the conflicting parties cannot directly affect; that is, for the good or
the bad. For instance, these might be a mountainous terrain, sudden political instabilities,
foreign powers or diasporas supporting affiliated groups in one way or the other. Hence, diverging from the state-centric view on militias, particularly the organizational dimension of
anti-rebel activities seems relevant: to what extent are militia efforts coordinated and expanded beyond the mostly local level?
6 Direction
Closely related with the resource and initiation categories, similar to rebel groups, militias
can pursue different directions to sustain their anti-rebel activities. Depending on the respective context, the analysis will have to show whether the militia “(…) is an anti-rebel militia
primarily offensive, searching for enemies across a given region, or [whether it is] primarily
defensive, securing communities from rebel attacks” (Barter 2013: 80). The distinction between ‘offensive’ and ‘defensive’ militias is borrowed from Weinstein’s (2005) duality of
‘predatory’ and ‘popular’ rebel forces. According to the author, resource availability determines whether non-state actors pursue a predatory/offensive or popular/defensive strategy.
Mobile militias will be more offensive and less keen to provide for a certain local community
than stationary ones. Given the distance and thus lack of local support, the threshold to make
use of abusive means against local people gets lower. In turn, defensive and stationary militias are more likely to be supported by local communities; including logistics, intelligence etc.
While offensive militias pursue material benefits from their anti-rebel activities, defensive
militias are primarily interested in the security of local communities which they are often part
of themselves.
Accountability
The more detached (physically and/or psychologically) the state is perceived from the militia’s local context, the less accountable the militia is expected to be to the national government. Instead, the direct constituency – i.e. local village, family, tribe etc. – will provide legitimacy and justification for whatever violence is exercised in the wake of conflict. The
question of accountability and legitimacy becomes even more pivotal the more the state forces commit human rights abuses against those constituents. Although this indicator is probably
most difficult to assess, it helps to differentiate the state’s success (or the lack of it) in coopting militias in remote areas far off the capital city.
Despite Barter’s (2013) focus on PGMs only, at this stage of analysis his four-fold typology
is useful to generally distinguish militias. To what extent it can be applied to the Peruvian and
Sierra Leonean context will have to be discussed further below.
Table 1: Four Types of Militia Organizations Initiation & Resources Direction & Accountability Offensive Defensive State Created Death Squads Village Guards Locally Created Private Armies Self‐Defense Communities Source: Own illustration based on Barter 2013: p. 80.
With the help of this framing model, the following section will analyze the cases of Peru and
Sierra Leone. It is the understanding that these four indicators shape the militias’ willingness
to support (or not) the state’s counterinsurgency efforts. Empirically speaking, four directive
questions will lead the analysis: How and why were militias created during that specific peri7 od? What were the militias’ primary resources for sustaining the armed struggle against the
rebel groups? Did the militias operate in an offensive or defensive way on the ground? What
were the driving factors for that strategic approach (e.g. income opportunities etc.)? Who
were the militias accountable to during the civil war? It will be the overarching goal to understand in how far the militias changed during the co-optation process. Particular light will be
shed on the unintended effects on intracommunal violence.
4. EMPIRICS
This section is divided into four parts. At first, a summary of civil war genesis in Peru and
Sierra Leone will be presented. This overview will help to understand the pivotal role of militias for the conflict outcome. Second, the commonalities between these two settings will be
presented along the four indicators elaborated above. A particular focus will be on the state’s
instrumentalization efforts. Based on these findings, third, the parallels regarding the unintended effects of militarization and autonomization of militias on local level will be presented.
At last, a cursory overview of the state’s efforts to recapture these dynamics during and after
conflict will end the empirical analysis.
Civil War in Peru, 1980-1995
Peru’s first steps towards democratic rule were accompanied by guerrilla attacks launched by
the revolutionary Maoist SL. On the eve of the first presidential elections allowed by the military government on 17 May 1980, ballot boxes were burned in the town of Chushi in the
Ayacucho region. This incident has been identified as first act of war (McClintock 1989: 62–
64). Throughout the 1980s, the guerrilla group grew in territory, organization and popular
support. By 1991, the group had control of much of the rural areas of the center and southern
parts of Peru. 12 September 1992 marks the downfall of the guerrilla group, when its leader,
Abimael Guzmán, was captured by the police. The organization fractured into splinter groups
and guerrilla activities diminished to a minimum extent (Taylor 1998: 35–58). However, that
decline encouraged the state forces to increase their counterinsurgency actions, which in turn
would intensify the attacks by the remaining SL. These escalatory incidents would last until
1995 (Koc-Menard 2006: 332–337). According to the Peruvian CVR at least 23,969 people
were killed between 1980 and 2000. However, due to the huge number of disappearances
during the war, the CVR has declared that estimates are more likely around 69,280 people
(Comisión de la verdad y reconciliación 2003).
The Peruvian civil war has often been simplified as conflict driven by rebels waging war in
order to make profit from the cocaine trade (Dreyfus 1999; Kay 1999; Felbab-Brown 2005;
Mason and Campany 1995; Cornell 2005; Harmon 1992). However, the narco-dollar argument is incomplete. A more differentiated view is necessary in order to understand the emergence and impact of SL. The Maoist guerrilla group was able to implicate the state into a 15year-long civil war due to an opportune conjuncture of background and actor-related circumstances. First, the impact of the failed agrarian reforms of 1969 on the isolated Ayacucho area
needs to be highlighted. In short, the land reforms resulted in benefits to a number of people
that were better situated than they had been prior to those policies. However, the most disadvantaged and deprived segment of Peru’s peasantry, the sierra comuneros, were rendered
8 even more vulnerable to economic crisis (Kay 2001: 170; see also Kay 1982). The traditionally marginalized and isolated rural-based Ayacucho department was particularly affected
(Palmer 1986: 137). Due to the government’s incapacity to foster sustainable development in
this region, the SL filled that void by providing services the central state was not capable of
or willing to do (Degregori 2007: 8–12). In return, the province became the major retreat and
recruitment area for SL. Beyond these two interdependent background causes, the state has
been criticized for underestimating the guerrilla at the beginning of the first operation in the
early 1980s. Indeed, it was not until late 1982 that the government deployed armed forces
into the emergence zone of Ayacucho. Given the repressive nature of the initial counterinsurgency and counterviolence by SL, the death toll shot up significantly between 1983 and 1984.
However, in the late 1980s, the state forces regained legitimacy after adjusting their approach
towards a more bottom-up oriented strategy and less indiscriminate violence (Palmer 2005:
96–100). At the same time, SL began to expand their operations to urban areas around Lima
and increased their activities in the coca-abundant Upper Huallaga area (Cornell 2005: 757–
758). In summary, the high conflict escalation in Peru between 1980 and 1995 can be best
explained by the conjuncture of five factors: the impact of the failed agrarian reforms on the
isolated Ayacucho department, the initially inadequate counterinsurgency by the state and the
criminalization of coca by the SL. Furthermore, the capabilities of the SL that had begun to
organize its armed struggle already in the late 1960s should not be underestimated as a conflict-fueling factor. A majority of scholars have stressed the influence of charismatic leader
Abimael Guzmán in this regard (McClintock 1986: 131–153).4
Civil War in Sierra Leone, 1991-2002
According to the Sierra Leone Truth and Reconciliation Commission (TRC), an estimated
70.000 people were killed during civil war between 1991 and late 2001 (TRC 2004). Similar
to the Peruvian and many other conflict settings, large-scale escalation in Sierra Leone was
caused by a conjuncture of structural background and actor-related preconditions. Against the
backdrop of the second lowest living standards of any country worldwide in 1991 (United
Nations 1993), for two decades Sierra Leone had been ruled by dictatorial and kleptocratic
regimes primarily enriching themselves through illicit diamond deals (Reno 1995). Given the
popular discontent with the ruling power, an initially small rebel group, that had entered the
country from Liberia in 1991, the so-called Revolutionary United Front (RUF), was successful in recruiting particularly youth to take up arms against the status quo (Bellows and Miguel
2009: 1145; Hoffman 2006: 5-6). At the latest by early 1992, the RUF had managed to spread
the conflict throughout the entire country. While most of the literature has identified the ac 4
Limited relevance has been attributed by scholars to external factors. Against the backdrop of Cold War, the
acquiescence by the US towards human rights violations committed by the Peruvian military has been identified
as a minor conflict-fueling element. Based on bilateral economic aid, conservative presidents, such as Belaúnde
(1980–1985) and initially Fujimori (1990–2000), were supported in their fight against left-oriented guerrillas by
the United States – generally, looking the other way in cases of human rights abuses by state forces. Thus, the
affected indigenous communities were less immune to SL propaganda. Probably more influential in this regard,
the guerrilla’s dogma was substantiated by the US counternarcotics initiatives in the coca-abundant Upper Huallaga Valley. Enforcing harsh eradication and failed crop substitution programs, these initiatives had the opposite
effect: faced with an increasingly destroyed economic livelihood, the coca farmers were further alienated from
the government that was perceived as a henchman of US interests. As a result, this facilitated the recruiting of
future members by SL (Obando 2006: 173–174).
9 cess to Sierra Leone’s diamond wealth as the rebel group’s primal motivation, particularly
Richards (1996, 2002, 2005, 2006) has stressed political grievances as initial reasons for joining the RUF. However, in line with the majority of scholars, Bellows and Miguel (2009)
identify ‘greed’ as primary motivation for the RUF: “Mining diamonds in Sierra Leone requires no heavy machinery or technology, since these alluvial stones sit close to the surface in
dried riverbeds, so any armed group that controlled a diamond-rich area could extract and
then sell the diamonds for large profits (Bellows and Miguel 2009: 1146).” According to
Keen (2005), eventually there was no armed group left that would not be involved in diamond smuggling: “Any battles were largely restricted to the areas with the richest diamond
deposits” (Keen 2005: 212). Quantitative research further substantiated the relationship between political instability in Sierra Leone and incentives for armed groups to protract the war
(Keen 2005: 20).5
According to Conibere et al. (2004), similar to the Peruvian onset, ethnicity did not play a
major role during civil war in Sierra Leone (Conibere et al. 2004; TRC 2004). Humphreys
and Weinstein (2006) confirm the lack of ethnic targeting in showing that abuse of civilians
by the RUF or other armed factions was not related to a specific ethnic affiliation (Humphreys and Weinstein 2006; TRC 2004). In addition to the importance of ‘blood diamonds’
on the one hand and the irrelevance of ethnicity on the other, civil war in Sierra Leone has
been controversially discussed with respect to the disproportionate share of violence perpetrated by the RUF. According to figures by TRC (2004), at least 70% of all war atrocities had
been committed by RUF members during the conflict (TRC 2004; Smith et al. 2004;
Conibere et al. 2004). Similar attention has been given by scholars to the cooperation between RUF rebels and the Sierra Leone Army (SLA) (Zack-Williams 2001; Reno 2003; Davies 2000). According to Keen (2005), direct clashes were deliberately avoided by coordinating the respective movements in the diamond abundant areas. Thus, both groups mutually
benefitted from the revenues. Bellows and Miguel (2009) illustrate: “Some soldiers apparently fought for the SLA by day and the RUF by night. As a result, the main victims of the violence were civilians, who were terrorized not only by the RUF but also by the army that was
supposed to protect them” (Bellows and Miguel 2009: 1146). Against this backdrop of local
insecurity, self-defense militias would become crucial stakeholders on the ground; both in
Sierra Leone as well as in Peru.
Militia Existence prior to Conflict
While the sometimes tacit and sometimes blatant cooperation between rebels and the armed
forces did not occur in Peru to a striking extent, still local communities were confronted with
threats by both of these groups. In particular, the more remote the areas were situated. This
section will briefly present the pre-instrumentalization status quo of self-defense militias in
5
Several studies have shown that diamonds played a key role in financing the activities of the RUF. However,
these have been controversially discussed as root causes of conflict. The absence of democratic opportunities,
patrimonialism, clientelistic networks, inefficiency etc. have been rather identified as cause. Excellent studies on
Sierra Leone are provided by Kabia 2008: pp. 93-105; Davies 2000: pp. 349-369; Keen 2003: pp. 67-94; Richards 2003; Ndumbe 2001: pp. 90-105. A regional comparative perspective of Sierra Leone, Liberia, Côte
d'Ivoire and Ghana is provided by Ware/ Ogunmola 2010: pp. 71-87.
10 Peru and Sierra Leone. This reference is important given that local militias existed in both
countries long before they were co-opted by the state.
Historically, rondas campesinas originated in northern Peru’s department of Cajamarca in the
mid-1970s. These self-defense and vigilance groups emerged as a reaction to the centerstate’s incapacity to execute justice in remote areas in general and to protect local communities against organized rustler gangs in particular. The communities were confronted with two
security challenges: first, with national economy’s crisis hitting local ground particularly in
the rural areas, cattle rustlers increasingly cooperated with local authorities (judges, police,
and district attorneys). Second, the less security was provided by the center-state representatives the more suspicious the villagers became towards each other. This in turn led to a rise of
intracommunal robbery and rustling. According to a majority of scholars, years before conflict outbreak in 1980 the organization of rondas campesinas had become a success story
which would rapidly expand throughout the country’s remote areas (Degregori 1996; Zahar
2000; Fumerton 2001; Starn 1995).
In Sierra Leone, the Mende Kamajoisia (or Kamajors) represented the largest of the various
militias that fought on behalf of the government against the RUF in the mid-1990s. Similar to
the concept of rondas campesinas, the Kamajors had been originally established in order to
protect villages against intruders (Richards 2002: 268). Kamajors were primarily located in
the south and east of the country; mostly in the Bo district. According to Hoffman (2007),
“(…) the Mende term referred to specialized hunters empowered to use both firearms and
occult ‘medicines’ (…) in the pursuit of big game and, more importantly, the various animal,
human, and extra-human forces of the forest that threaten rural villages” (Hofmann 2007: 642;
see also Zack-Williams 1999: 152; Møller 2006: 16; Ferme and Hoffman 2004: 74-76; Richards 2002: 268).
Given the pre-civil war existence of self-defense militias in Peru and Sierra Leone, two implications in terms of exogenous or endogenous ‘initiation’ need to be taken into consideration: first, instrumentalizing militias by the state would be less difficult given the existing
organizational structures and institutions on local and sometimes even regional level. Second,
the impact of military training and assistance on communal dynamics would be more unpredictable as militias might become more autonomous than initially expected.
In both cases, the government’s decision to actively integrate self-defense organizations into
the armed struggle was based on strategic considerations. Given the lack of success of conventional counterinsurgency approaches (i.e. mostly top-down operations, use of heavy weaponry etc.), political and military leaders decided to co-opt militias in the attempt to defeat
the armed groups that had spread out through the countries. In the case of Sierra Leone, this
decision was further substantiated by President Ahmad Tejan-Kabbah’s distrust with the SLA
leadership (Richards 2002: 269). The following section will juxtapose the ways the regimes
co-opted self-defense militias for their purposes.
Co-optation of Militias in Peru and Sierra Leone
Similar to the Sierra Leonean onset, authorities in Peru very late decided to bring rondas
campesinas into play. Although local self-defense organizations had been heard of already in
the late 1970s and despite some military leaders arguing in favor of rondas campesinas dur11 ing the early 1980s, the counterinsurgency strategy was formally readjusted as of 1989.6 Two
imperatives were promoted: selective operations against Senderistas and the systematic organization of so-called Comités de Defensa Civil (Civil Defense Committees, CDC). The
argument went as follows: “En estos dominios no propiamente militares de la guerra interna
se hacen los avances más significativos, tanto para ganar el apoyo de la población como para
neutralizar la capacidad de acción de los subversives” (CVR 2003: p. 285). This rethinking
process was based on the Manual de Ejército Guerra no Convencional Contrasubversión ME
41-7 (Countersubversive Manual, ME 41-7) presented in June 1989. This manual promoted a
strategy based on political, economic, and psychosocial as well as military aspects of counterinsurgency. One major pillar was to systematically support training and arming of local peasant organizations. Although rondas campesinas had emerged prior to 1989, the formal acquiescence was given in the early 1990s. For instance, on 9 December 1989, in front of national
and international reporters, President García himself presented the CDC of Rinconada Baja,
Apurímac Valley, with two hundred shotguns. Although the congress did not pass respective
legislation until 1990, this direct CDC-support facilitated the autonomization of local militias.
The pivotal relevance of rondas campesinas to the conflict is summarized by Kay (2001):
“(…) [El] cambio de estrategia del gobierno y los militares hacia las rondas contribuyó
también a lograr la derrota de Sendero. En vez de considerar a las rondas como
organizaciones potenciales de terroristas que simpatizaban con Sendero, el gobierno
empezó a darse cuenta que se trataba de verdaderas asociaciones de base que intentaban
defender la supervivencia de sus miembros y compensar las fallas del Estado en su
incapacidad para protegerlos contra los robos y crímenes, administrar la justicia según su
paracer y proporcionarles los servicios más esenciales. (...) [El] actor principal para la
derrota de Sendero Luminoso es el de las comunidades campesinas que habían sido
también las principales víctimas de Sendero, así como las víctimas iniciales de las
fuerzas de seguridad” (Kay 2001: X).
Much has been written on the eventual success of CDC in contributing to SL’s defeat in rural
Peru (Fumerton 2001; Starn 1995; Degregori 1996). The following episode well illustrates
the impact of CDC in that regard: on 13 April 1988, around forty heavily armed SL
guerrilleros accompanied by hundreds of Senderista peasants attacked the village of
Pichiwillca, the basis of the Defensa Civil Antisubversiva (Counterinsurgency Civil Defense,
DECAS), in the Apurímac River Valley. This valley was economically of high value to the
SL as well as the indigenous communities living there due to its agricultural richness. Helped
by ronderos from neighboring villages and a navy infantry unit, the SL attack was fought off
leaving more than ten guerilleros dead. According to CVR testimonies, this ‘success’ marked
the beginning of a large-scale counter-offensive organized by the DECAS. In the following
two years the ronderos were able to regain control over the entire Apurímac River Valley.
However, the control over the Valley proved to be preliminary step. The DECAS began to
expand its activities in the Sierra. On 9 August 1989, around 200 ronderos left Santa Rosa to
6
Although the formal integration of rondas campesinas happened in 1989, first coordinated initiatives had been
launched during the mid-1980s. Military commander General Adrián Huamán advocated a new approach emphasizing on humanitarian aid and development – i.e. ‘winning hearts and minds’ – rather than armed repression.
He had been installed in 1984. See Fumerton 2002: p. 92.
12 Tambo in order to reorganize the existing but weak CDC in La Mar province. In contrast to
previous years, these developments illustrated a new character of rondas campesinas in two
regards. First, in shifting towards offensive operations against the guerilla the CDC began to
professionalize their cause. For instance, ronderos were more and more often paid by neighboring communities to participate in the battle against the SL. This was certainly due to the
material support though not less relevant due to the legitimacy provided by the state. Second,
contrary to first initiatives in the early 1980s, this time the CDC themselves chose not to back
down against the guerilla. “Esta vez, la proliferación de las rondas no se debe solamente a la
imposición militar, sino que expresa también la decisión del campesinado” (CVR 2003: p.
445).
Only at a late stage of conflict, as largest and most powerful paramilitary actor the Kamajors
were co-opted by the state under the umbrella of the so-called Civil Defense Forces (CDF) in
1997 (International Crisis Group 2001: 11). Similar to the genesis of self-defense communities in Peru, the Kamajors initially numbered merely one or two villages in the southeastern
Mende chiefdoms. However, the more ‘success’ they had in resisting against SLA and RUF
factions the more attention was given to the Kamajors (Ferme and Hoffman 2004: 89-90).
Similar to Peru, the number of Kamajors grew substantially the more the center-state proved
incapable of providing security services in remote areas (Ferme and Hoffman 2004: 83). After President Tejan-Kabbah formally established the CDF umbrella under which the
Kamajors quickly stood out most, the armed forces leadership considered the militia as imminent threat. Hoffman (2006) described the power shifts as follows:
“[the] CDF was a collection of irregular units of “hunters” and semiprofessional veterans
of the Liberian conflict, some trained by Executive Outcomes, a South African security
outfit. These hunter-militias achieved a certain international notoriety for their practice of
initiating members with rites designed to make their bodies bulletproof. Although active
throughout the conflict, bands of organized hunter-irregulars achieved enough prominence under the SLPP [i.e. Tejan-Kabbah’s Sierra Leone People’s Party] to be considered a threat by the military, which in May 1997 staged a coup and invited the RUF to
join in a junta government. The alliance between the Armed Forces Revolutionary Council (AFRC) and the RUF lasted until early 1998, when a combined force of CDF and
West African peacekeepers (Economic Community of West Africa Ceasefire Monitoring
Group) reinstated Kabbah and his government” (Hoffman 2006: 5).
The affiliation with the state had been best illustrated by Tejan-Kabbah appointing the
Kamajor leader Chief Sam Hinga Norman as Deputy Minister of Defense in 1996. Along
with cuts in military personnel and food subsidies, a military coup by factions of the armed
forces had eventually become a rational option to consider (Zack-Williams 1999: 152). The
close relationship with the SLPP and the increasing leverage of the Kamajors was further
demonstrated during the short period under the AFRC-RUF alliance between May 1997 and
February 1998. With the SLPP rulers exiling into Guinea, the militia eventually expanded its
activities on regional level. Recruits were increasingly enlisted in Liberia and Guinea thereby
particularly focusing on mercenaries living in the region’s refugee camps (Hoffman 2004:
215). The professionalization of Kamajors was intensified during that period given that they
“(…) found themselves among the main targets of the brutal AFRC-RUF regime” (Hoffman
13 2006: 6). Hunters were increasingly trained in the use of firearms in the attempt to make them
replacing their traditional arrows, bows and knives with guns (Hoffman 2006: 15).
Applying the four framing indicators elaborated above, the brief overview of state
instrumentalization of militias in Peru and Sierra Leone demonstrates several points that deserve further analysis:
Initiation: In both cases, self-defense militias emerged as a reaction against the state’s incapability to provide services, such as foremost security, on local ground. However, this primarily endogenous initiation was later accompanied by exogenous incentives by the state the
more the rebel groups consolidated in the respective areas. Facing an increasingly competitive and thus vulnerable context in-between the two major adversaries, self-defense militias
eventually sided with the center-state that seemed more promising in the long run.
Resources: Closely related with the initiation, militias in Peru and Sierra Leone particularly
benefitted from the military equipment they were provided with by state forces. Ironically,
the early success of rondas campesinas or Kamajors in defeating rebel factions by plain
weapons such as machetes or arrows helped them to attract national attention in the short
term. However in the long run, this attention further led to the state authorities increasingly
providing the militias with better military equipment.7
Direction: Militias in Peru and Sierra Leone were defensive at the beginning. The more the
rebels or state forces (or both) committed human rights abuses against local communities the
more the militias would fight back. Despite this commonality, one striking difference between CDC in Peru and the Kamajors in Sierra Leone was the gradual expansion of the latter
in the later stages of conflict in the late 1990s. While militias were not forced to cross national borders in Peru, the regionalization of civil war in Sierra Leone made the Kamajors become a quasi-regular state force under the formal umbrella of the exile SLPP. Yet another
variance was the exploitation of natural resources. While coca was (and still is) cultivated by
peasants and SL sympathizers in Peru, diamond smuggling was a primarily illicit endeavor in
Sierra Leone. According to Weinstein’s logic, the Kamajors’ wide expansion throughout the
country and region was more inclined to become predatory and offensive.
Accountability: As the militias initially emerged on local level, the respective communities,
tribes, and families represented the primary constituency. The more the militias were professionalized and integrated into the CDC and CDF the less accountable they became to those
constituents. The case of Sierra Leone is more obvious in that regard given that Kamajor
leader Chief Sam Hinga Norman was appointed as Deputy Minister of Defense in 1996.
Based on these preliminary findings, Barter’s four-fold typology proves incapable of covering the changing character of militias in Peru and Sierra Leone. For instance, while rondas
7
For instance, after the community of San José de Secce had stood up for themselves against Senderistas and
killing seven by knives and machetes in 1984, President Belaúnde was quickly to praise this act of self-defense
in a press conference: “[Sin] armas, se impusieron gallardamente y los echaron. Por ese camino espero que se
restablezca la tranquilidad y que la presencia y valor de estos pueblos sea suficiente y no se requiera acudir a la
fuerza para acabar con el terrorismo” (Belaúnde quoted in DESCO 1988: p. 93). Poorly equipped with spears,
machetes or other plain weapons originally used against cattle rustlers, General Huamán’s militarization initiative led to the moderate arming of peasant communities. Mostly, the armed forces provided them with longrange rifles. Although an outdated weapon from the First World War, given that the firepower was comparable
to the assault rifles used by the SL, rondas campesinas were glad to have them (Fumerton 2001: p. 489).
14 campesinas started out as self-defense communities in Peru, however, they would evolve into
village guards with the state militarizing their organizational bases. The same holds true for
the Kamajors that had originated as self-defense units in the Bo district. Applying Barter’s
typology the table below shows that Peruvian and Sierra Leonean militias would match several types of militia organizations over time.
Table 2: Four Types of Militia Organizations: Peru and Sierra Leone Direction & Accountability Initiation & Resources State Created Locally Created Offensive Defensive Death Squads Village Guards •
•
•
•
‐ ‐ Comités de Defensa Civil (CDC) (1990s‐) Civil Defense Forces (CDF) (1997‐) Private Armies Self‐Defense Communities •
•
•
•
Defensa Civil Antisubversiva (DECAS) (1988/89‐) CDF and Kamajors (1997‐) Rondas campesinas (1970s‐90s) Kamajors in Bo district (early 1990s) Source: Own illustration based on Barter 2013: p. 80.
Before delving into the unintended effects of state instrumentalization of militias, there are
three issues worth touching upon that call Barter’s typology into question: First, militias can
be very unpredictable and changing forces. For instance, starting out as self-defense organization they can evolve into a (regional) network of army-like forces striving for material gains.
As the Sierra Leonean context shows, this can be particularly true when retrievable natural
resources come into play. This volatility is also illustrated by the militias’ changing loyalties
towards the rebels and/or the state forces. Second, despite this paper’s implicit assumption of
the state instrumentalizing militias, the brief overview has shown that militias are foremost
actors on their own; basing their collective decisions on strategic rationales, too. Hence, particular in the later stages of conflict, both CDC in Peru and CDF in Sierra Leone benefitted
from the cooperation with the national government. Third, despite the exhaustive overview of
militias in the literature, it remains difficult to grasp the transnational perspective. As the case
of Sierra Leone shows, with the conflict spilling over national borders and militias being recruited in regionally dispersed refugee camps, they are not easy to grasp as unitary and cohesive actor.
Irrespective of these issues, the preliminary analysis has shown that militias were co-opted by
the state in Peru and Sierra Leone in order to support the regular forces in the armed struggle
against the rebels. The term ‘co-optation’ is more adequate as the militias likewise benefitted
from the cooperation. The following section will present the unintended effects of the state’s
co-optation of militias on local ground.
15 Unintended Effects in Peru
The previous analysis has implied that militias became the feather that broke the balance towards the state. In the case of Peru, the strategic shift of counterinsurgency in the late 1980s
crucially helped to defeat the SL. Instead of perceiving rondas campesinas as likely sympathizers of the guerilla, the government eventually realized their potential as ally in rural areas;
both in terms of intelligence and as armed force (Kay 2001). While their increasingly offensive participation in direct battles with the SL certainly contributed to higher violence escalation, their partaking proved a crucial factor in ultimately defeating the SL. This was particularly due to two developments. First, organizationally, the SL increasingly lost recruitment
opportunities, shelter, and simply access to food. Second, the SL was confronted with yet
another frontline besides the police and military forces. The pivotal nature of rondas
campesinas both to the SL as well as to the state forces has been illustrated by Mason (2004):
“Shining Path rebels are subject to the same dilemma as the state: If they (and not the
state) are seen as the perpetrators of violence, then peasants will turn away from them
and to the state for protection. Rebel violence against civilian supporters can erode their
base of support, just as government violence erodes popular support for the government.
Sendero’s near defeat was as much a function of its own harsh treatment of peasants as
the government’s tactical agility” (Mason 2004: 258).
In short, if the CDC had not actively fought against the SL, civil war would have probably
not halted in 1995.
While much of the literature has predominantly discussed the crucial role of CDC in ending
civil war in Peru (Degregori 1989), only a few studies have pointed to the negative unintended effects of militarization on local ground (CVR 2003: 454). Given the limited scope of the
conference paper, the following overview will only cursorily cover three of these unintended
effects in the Peruvian context: very much interdependent and overlapping these were the rise
of accidental killings, the creation of a local security vacuum, and the disruption of traditional
hierarchies and consensual conflict resolution mechanisms in exchange for a militarized local
culture and living together.
Looming in late 1984, the increasingly better organized rondas campesinas became more
involved with violence on the ground. Primarily defending their communities, in some cases
the attacks and counter-attacks resulted in accidental killings of noninvolved people. For instance, on 20 January 1985, four youth were stoned to death by a peasant group close to Ayacucho city after mistaking them for SL members (DESCO 1988: 119). A similar incident
took place in Huánuco on 26 September 1985 when rondas campesinas confused
plainclothed police members with Senderistas and killed three of them by machetes (DESCO
1988: 133). These accidental killings hold two implications: on the one hand, they are striking given that the state forces had been criticized by the communities for the arbitrary nature
of counter-insurgency. On the other hand, the rise of ‘collateral damage’ illustrated the increasingly opaque battle field on the ground – not to forget, the window of opportunity to
settle old scores.
Although the legitimizing and militarizing of communities enabled these to be less vulnerable
to guerilla attacks, at the same time, existing tensions between neighboring communities were
16 heightened by the forced cooperation: “En la mayoría de casos, la medida provocó
resentimientos y una resistencia pasiva entre los comuneros de estos pueblos, debido (…) a la
profunda rivalidad que desde tiempo existía muchas veces entre comunidades que ahora
estaban obligadas a convivir” (CVR 2003: 43. See also Degregori 1996: 51-56; CVR 2003:
271-272.). Hence, violent clashes between communities were first reported on 20 February
1983 in the communities of Huaychao, Carhuaran, and Uchuraccay in the province of Huanta
(CVR 2003: x).8 Not surprisingly, after similar confrontations in this province, the communities began to demand security guarantees by the state.9 The early rondas campesinas often
did not have a choice but to become a CDC. As the armed forces usually demanded the organization of CDC under the umbrella of the military, most communities chose to comply
rather than being accused of secret collaboration with the SL (CVR 2003: 454).
The disruption of former conflict resolution mechanisms probably poses the least studied
unintended effect of CDC creation in Peru. According to a majority of testimonies gathered
by the CVR (2003), a military logic was imposed in the villages by establishing CDC. Following this logic, everyday life was organized along war rationales (e.g. incessant military
work-out, vigilante activities etc.). In particular, this included penalizing of petty offenses
instead of making use of consensual community-based institutions. The CVR quotes a testimony that well illustrates the impact of the militarization:
“Bueno, ese tiempo no había un buen arreglo, porque ese tiempo todo era castigo. Ese
tiempo no había ningún juez, sólo los comandos hacían respetar las leyes, también había
un teniente, sólo ellos castigaban al que se comportaba mal en la comunidad, lo
golpeaban con la culata del arma. Teniente y comando eran los únicos que había en la
comunidad, no había ni presidente; por lo tanto, no había ningún tipo de arreglo, sino
castigo” (CVR 2003: 450).
In most cases the focus on military training constrained the local communities to go along
with their regular work such as for instance managing their estates. The overall militarization
of everyday life deeply impacted on the social fabrics within the local communities. Given
the command exercised by the armed forces, usually younger ronderos became CDC leaders
as most of them were graduates of the Peruvian military. Significantly, these were chosen by
the communities according to bellicose qualities alone: “al más valiente, al que tenía más
mando, a él le nombrábamos nosotros” (CVR 2003: 451). In turn, the traditionally meritocratic ruling in Andean communities was pushed aside; with it an entire generation of elder authorities. Naturally, these young leaders frequently made use of their autonomy towards their
local communities by liberating themselves from social control. In a few cases, the CVR has
even identified warlord-like commandos. Accordingly, yet another testimony illustrates their
exceptional position: “Eran como dioses. (…) [Ellos] actuaron como reyes ya, prácticamente
8
For instance, ronderos of Acos Vinchos, district of Huamanga, repeatedly attacked the neighboring community
of Quinua. Eventually, the Quinua villagers decided to form a CDC on their own in order to fight back (Del
Pino 1992: 507). The CVR lists dozens of testimonies stating the increase of intracommunal human rights
abuses after the emergence and formal approval of CDC by the national government (CVR 2003: 454).
9
Indeed, already in late 1984, first accusations by national as well as international third parties demanded the
state to regain some form of monopoly of violence in the remote areas in order to prevent intracommunal massacres (CVR 2003: 453-455).
17 ellos, con el nombramiento que han obtenido, ya eran rey ya, a una persona que no le
computaba ya le decían, ya, a la tortura” (CVR 2003: 451).
In sum, the creation of CDC in Peru did not only serve to defeating SL and thus ending civil
war. The militarization also led to the emergence of local security vacuums and the disruption
of traditional conflict resolution mechanisms. As the CDC were formally established in the
early 1990s (de facto in the mid-1980s) and dissolved in the early 2000s (see below), deeprooted suspicions would be difficult to tackle by a national reconciliation process.
Unintended Effects in Sierra Leone
The onset was similar in Sierra Leone. A majority of scholars have identified the CDF or
rather the Kamajors as important force in defeating the RUF and related armed factions (Bellows and Miguel 2009; Ferme and Hoffman 2004; TRC 2004; Møeller 2006; Richards 2002;
Hoffman 2003, 2004). Accordingly, the TRC (2004) stated: “[the] Commission concludes
that the CDF played a vital role in defending the nation from the predatory actions of rebel
forces and renegade troops” (TRC 2004: 76). Although TRC-data conveys a minor responsibility of human rights abuses and intracommunal violence to CDF – i.e. 6 % compared to at
least 60 % attributed to RUF –, some studies point to the unexpected autonomization of the
Kamajors in the wake of conflict (TRC 2004: 38; Bellows and Miguel 2009). For instance,
despite the overall admiration, Keen (2005) argues that some CDF units eventually made use
of their authority by abusing civilians as well as entering illicit diamond trade (Keen 2005:
268; see also Møller 2006: 16). In order to make a comparison with the Peruvian onset more
feasible, the following section will focus on accidental killings, the emergence of local security vacuums, and the disruption of indigenous conflict resolution mechanisms.
It is striking that ‘irregular’ killings in general and human rights violations in particular have
not been discussed as ‘accidental’ phenomenon in the militia-context of Sierra Leone. As
primary source of post-conflict investigation, the TRC (2004) has blamed parts of CDF units
for selectively killing SLA soldiers in the attempt to take over the military’s role as protector
of the people and property (e.g. in Kenema District): “Kamajors carried out multiple acts of
torture and killings against soldiers. In many cases, corpses of soldiers were dismembered
and parts of them eaten. Kamajors also committed similar violations and abuses against civilians whom they perceived to be affiliated to the Army” (TRC 2004: 77). Furthermore, CDF
were accused of human rights abuses against civilians during their direct clashes with the
AFRC-RUF and RUF factions. In particular the “Operation Black December” in late 1997
and early 1998 as well as the Freetown massacre on 8 May 2000 have been highlighted by
the Commission in that regard:
“[Operation Black December] was purposely designed to debilitate the strongholds of the
AFRC junta in the Southern and Eastern Provinces. In the process, it caused immense
suffering to the civilian populations of many communities in these Provinces. It also led
to massive and systematic human rights abuses including summary killings, torture and
looting at checkpoints established by the Kamajors. (…)The President authorised Chief
Samuel Hinga Norman to undertake a largescale mobilisation of members of the Civil
Defence Forces in Freetown on 8 May 2000. (…) Kamajors carried out multiple violations (…), including killing civilians, plundering vehicles and properties and torturing
18 captives. Kamajors carried out arrests of persons subsequently detained in state facilities
as “Protective Custody” prisoners. The Kamajors, together with other militant elements,
initiated the attack on Foday Sankoh’s residence [i.e. founder and leader of the RUF] on
8 May 2000 and used the occasion for large-scale attack and abuse of perceived RUF
sympathizers” (TRC 2004: 79-80).
While the selective killings of ‘state enemies’ was certainly part of the war logic on both
sides of the conflict and thus cannot really be identified as ‘unintended effect’, the human
rights abuses can indeed.
In terms of local security vacuums, the rise and expansion of CDF certainly contributed to
lack of control and oversight on the ground (Møller 2006: 16). However, contrary to the
mostly locally-oriented and thus decentrally monitored rondas campesinas in Peru, power
and leadership was highly centralized in the CDF (TRC 2004: 78). As such, the supervision
of Kamajors was difficult in two ways. On the one hand, the national government was incapable of overseeing the CDF leaders, while on other hand, these were incapable of monitoring the fighters on the ground. Hence, local insecurity was not due to the militarization of
former self-defense militias but rather due to the lack of control of militias from above. The
TRC (2004) blamed the respective authorities as follows:
“The Commission finds that ambiguity in the institutional character of the CDF precipitated a persistent lack of coherence, cohesion and co-ordination in its operations, which
led to the commission of many human rights violations and abuses in the enforcement of
what the massed ranks saw as the “law”. The lack of effective oversight over the National Co-ordinator, the High Priest and the ground commanders of the CDF ultimately led to
grave human rights violations being committed. In particular, the mandate that was given
to Kamajors to monitor the movements of the populace at strategic gateways and checkpoints was poorly defined and effectively licensed those charged with security to dispense summary justice against perceived miscreants, “strangers” and “collaborators”.
While power and leadership was highly centralised in the CDF, effective control of the
fighting forces was vested in the commanding officers at the lower level. (…) [By] staying in a movement that had become a systematic violator of human rights, (…) [the authorities] lent legitimacy and their implicit endorsement to the atrocities committed by
the Kamajors” (TRC 2004: 78).
Given this lack of oversight particularly in the heartlands of the Kamajors, Districts of the
South and South-East, inhabitants were selectively targeted by the Kamajors along ethnic
lines (TRC 2004: 79). This is striking given that the majority of scholars, including the TRC
itself, have denied that ethnicity played a relevant role during the conflict (Conibere et al.
2004). Most likely, ethnic hatred was based on regional rivalries within the Kamajordominated CDF movement:
“The Kamajors were intensely protective of their territories and their movement against
perceived infiltration by Northerners. They held the prejudice that Northerners might be
inclined to display allegiance to the leadership of the AFRC junta, largely because Johnny Paul Koroma was a member of the Limba ethnic group, which originates from the
North” (TRC 2004: 79).
19 Compared to findings presented by the CVR (2003), the aforementioned disruption of traditional conflict resolution mechanisms is not explicitly dealt with by the TRC report (2004).
Instead, primary focus is put on the responsibility for human rights abuses by the initiators of
the Kamajor Society; foremost Kamajor High Priest Allieu Kondewah. Apparently, the initiation policy was used to extort and exploit the membership of the movement in order to benefit economically (TRC 2004: 77). In particular, two features were heavily criticized by the
TRC: First, Kondewah’s and others’ initiation rituals falsely made recruits believe in superhuman abilities on the battlefield. Second, cannibalistic ceremonies were imposed which
eventually disrupted the social and cultural fabric of local communities. The TRC (2004)
summarized accordingly:
“The Commission finds that personal greed and ambition as well as avarice dictated the
initiation policy of the Kamajor High Priest, Allieu Kondewah. (…) The benefits bestowed upon Kamajors by their participation in initiation ceremonies were fictional. Initiation did not protect the subject from harm nor endow him with superhuman ability. (…)
A proportion of warfront casualties incurred by the Kamajors was attributable to the misplaced gusto with which their combatants went into battle. (…) [The initiators] subverted
the sacred and long-standing traditions of initiation and rites of passage that exist peacefully in Sierra Leone. (…) The Initiators of the CDF forced men who joined the
Kamajors to eat human body parts during the initiation ceremony. (…) Civilians from
communities surrounding the initiation site and even would-be recruits were in many instances killed for the express purposes of “sacrificing them to the cause”. The Initiators
of the CDF also carried out or ordered varying degrees of violence and intimidation
against those they enlisted. They tortured and killed initiates” (TRC 2004: 77-78).
In summary, killings of soldiers and civilians were committed by the CDF. Contrary to the
primarily locally situated rondas campesinas in Peru, indiscriminate Kamajor violence seems
to have been due to its expanding and offensive character evolved in the wake of civil war.
Local security vacuums were caused by the lack of supervision within a highly centralized
command structure (CDF leadership and SLPP government) rather than by decentral militarization of single communities. At last, traditional institutions of conflict resolution have not
been affected given the ‘un’-local character of the CDF movement; particularly after 1998.
Altogether, the major commonality between CDF in Sierra Leone and CDC in Peru remains
the low death-toll attributed to militias in both conflict onsets. While CDF has been blamed
for six per cent of the total number of battle-related deaths of roughly 70.000 in Sierra Leone
(1991-2002), less than 4 per cent has been attributed to CDC in Peru (1980-1995) for the similar amount of 69.000 people killed (TRC 2004: 38; CVR 2003: 21). As such, both CDF and
CDC ‘behaved’ far better than the state forces and the rebel groups. The next section will
briefly summarize the situation of militias in Peru and Sierra Leone after the termination of
civil war.
Militias after War Termination
After the defeat of SL in the mid-1990s, it still took some more years until 2002 when rondas
campesinas were officially recognized by President Toledo. By law, they are defined as autonomous and democratic form of community organization. Particular importance is given to
20 their conflict resolution capacities that are supposed to be pacific only. Besides a set of duties,
the rights to democratic participation, to respect for the rights of children and youth are emphasized along the banning of discrimination against women, senior citizens, and disabled
people (Langdon and Rodriguez 2007: 98). Despite the seemingly peaceful reintegration of
CDC into post-conflict Peru, the formal recognition has not necessarily pacified the local
settings:
“While some people consider this bill a triumph for the peasant and indigenous movement, others consider it a governmental strategy to keep the rondas within their control.
In other words, by institutionalizing the rondas, their grassroots power may be limited.
Even though the analysis of the Law No. 27908 is mixed, it seems clear that the passage
of this bill demonstrates the social significance of rondas campesinas in the last thirty
years of Peruvian history” (Langdon and Rodriguez 2007: 98-99).
After the formal ending of civil war in January 2002, the Kamajors lost their recognized status in Sierra Leone. However, many combatants still identified themselves with the movement. The formal end deepened the divide between the demobilizing segments and the professionalized parts of the CDF. Particularly the latter moved across the Liberian and Guinean
borders to support the rebels’ armed struggle against Charles Taylor (Ferme and Hoffman
2004: 78).10 Given the disproportionate share of younger ex-combatants and the lack of income opportunities in post-conflict Sierra Leone, Ferme and Hoffman (2004) have pointed to
the indirect displacement of Kamajors to a volatile context in the regional neighborhood:
“Given the extensive networks of trade and migration (voluntary and forced) throughout
the Mano River region, [the influx of Kamajors] amounted to a substantial number of
young men who expressed some degree of personal connection to communities across
the national borders—an identification underscored by ethnic and linguistic allegiances
that do not map onto official state boundaries. The largely unmet expectations of peacetime employment and education opportunities through Disarmament, Demobilization,
and Reintegration programs (DDR), as well as pressure from patrons with a vested interest in overthrowing the Taylor regime, have led many within this highly mobile demographic toward participation in the next phase of the now longrunning Mano River War”
(Ferme and Hoffman 2004: 78).
Apparently, the transnational character of the CDF constrained the Sierra Leonean state in
formally integrating the militias into the post-conflict status quo. While rondas campesinas
took up their former role as local self-defense units against cattle rustlers, parts of the
Kamajors movement opted for making use of their military capabilities in neighboring conflict zones.
10
Most often these combatants had family ties on both sides of the border. Given the regional expansion of the
CDF in general and the Kamajors in particular, some of these units had actually spent time in Liberia working
and/or fighting with one of the factions in the earlier Liberian civil war (Ferme and Hoffman 2004).
21 5. CONCLUSION
Militias in Peru and Sierra Leone were both: victims and perpetrators of violence. They became victims as they were violently forced into cooperation with the SL in Peru or the RUF
in Sierra Leone as well as with the respective state forces. They became perpetrators as they
either decided to defend their communities against external intruders (i.e. rebels and/or state
forces) or they were co-opted by the state in the armed struggle against the rebel forces. In
particular, the co-optation of militias – i.e. military training, provision of equipment – had an
ambivalent outcome. On the one hand, rondas campesinas in Peru and Kamajors in Sierra
Leone advanced as crucial ally of the state in defeating the rebel groups; simply by either
constraining the rebels’ recruitment opportunities, shelter, and access to food or serving as
yet another frontline besides the police and military forces. On the other hand, based on the
state’s legitimization and militarization, the militias became autonomous and often unbound
forces. Particularly in rural and remote areas, militias would de facto usurp local monopoly of
violence.
In the case of Peru, the state’s co-optation strategy led to three major unintended effects. First,
set in a general atmosphere of suspicion, the number of accidental killings and human rights
violations by indigenous communities increased. Closely related second, existing tensions
between neighboring communities were heightened by the forced cooperation. The security
vacuum on the ground was taken advantage of by settling old scores. Third, by imposing a
military logic in the indigenous communities traditional conflict resolution mechanisms were
disrupted (i.e. penalizing instead of consensual settlement of intracommunal disputes). The
case of militias in Sierra Leone was different. First, selective killings of soldiers and civilians
were committed by the CDF. Second, contrary to the locally operating rondas campesinas in
Peru, indiscriminate Kamajor violence happened due to the movement’s offensive character
evolved in the wake of civil war. Local security vacuums were caused by the lack of supervision within a highly centralized command structure rather than by decentral militarization of
single communities. Although the autonomization of Kamajors certainly led to human rights
abuses foremost by the initiators of the Kamajor society, the TRC has not specifically identified the disruption of traditional conflict resolution mechanisms.
Beyond the empirical findings on the unintended effects of state instrumentalization of militias in Peru and Sierra Leone, the analysis holds important theoretical implications for future
research. Contrary to the majority of studies that perceive militias as state proxies alone, this
paper has shown that a state-centrist view fails to grasp the manifold and constantly changing
nature of militias on the ground. By applying Barter’s (2013) four-fold typology, it became
apparent that neither the rondas campesinas nor the Kamajors fit into one category only.
They changed over time from local self-defense communities to autonomous village guards –
often on regional level. The Kamajors ultimately even spilled their operations over national
borders to Liberia and Guinea.
In summary, future studies will have to be aware of two issues in order to better grasp the
impact of militias on the ground. First, militias are more than state proxies alone. They act on
their own and are driven by specific motivations which are not necessarily consistent with the
22 state’s interests. Second, militias are not fixed entities. Adapting to the respective conflict
environment they might change over time.
Although the analysis of two cases only is based on a narrow empirical scope, the results hold
implications for policy-makers. In Peru and Sierra Leone, the co-optation of militias crucially
helped the state in defeating the rebel groups. However at the same time, the unintended negative effects were not anticipated (i.e. accidental killings, security vacuums etc.). For instance,
irrespective of the national state’s willingness or capacity to monitor dynamics on local
ground, recent conflicts in Libya or Syria illustrated the necessity to take the long-term impact of militias into consideration before militarizing them. While transitional authorities
have struggled to integrate militias in Benghazi to the national reconciliation process in Libya
(Pelham 2012; Mayer 2013), the blurred militia context in Syria has prevented Western governments from taking military action against the Assad regime (Carpenter 2012; Mucha
2013b).
In short, the arming of militias might be a strategically right decision by the state in the short
term. However, this legitimizing and military autonomization might likewise lead to local
insecurity and intracommunal violence as well as seriously hamper the post-conflict transitional process.
LITERATURE
Ahram, Ariel I. (2011a): Proxy Warriors: The Rise and Fall of State-Sponsored Militias.
Stanford: Stanford University.
Ahram, Ariel I. (2011b): “Origins and persistence of state-sponsored militias: Path dependent
processes in third world military development”. In: Journal of Strategic Studies 34:4, pp. 531555.
Anderson, L. (2011): “Demystifying the Arab Spring. Parsing the Differences between Tunisia, Egypt, and Libya”. In: Foreign Affairs 90:3, pp. 2–7.
Baines, Erin and Emily Paddon (2012): “'This is how we survived': Civilian agency and humanitarian protection”. In: Security Dialogue 43:3, pp. 231-247.
Bakke, Kristin M., Kathleen Gallagher Cunningham and Lee JM Seymour (2012): “A plague
of initials: Fragmentation, cohesion, and infighting in civil wars”. In: Perspectives on Politics
10:2, pp. 265-283.
Bakonyi, Jutta (2013): “Authority and administration beyond the state: local governance in
southern Somalia, 1995–2006”. In: Journal of Eastern African Studies 7:2, pp. 272-290.
Barter, Shane Joshua (2012): “Unarmed Forces: Civilian Strategy in Violent Conflicts”. In:
Peace & Change 37:4, pp. 544-571.
Barter, Shane Joshua (2013): “State Proxy or Security Dilemma? Understanding Anti-Rebel
Militias in Civil War”. In: Asian Security 9:2, pp. 75-92.
23 Bellows, John and Edward Miguel (2009): “War and local collective action in Sierra Leone”.
In: Journal of Public Economics 93, pp. 1144-1157.
Berdal, Mats and David M. Malone (2000): Greed and grievance: economic agendas in civil
wars. London: Lynne Rienner Publishers.
Bertrand, Romain (2004): ““Behave Like Enraged Lions”: Civil Militias, the Army and the
Criminalisation of Politics in Indonesia”. In: Global Crime 6:3-4, pp. 325-344.
Blakely, Ruth (2009): State Terror and Neoliberalism: The North in the South. New York:
Routledge.
Bøås, Morten and Anne Hatløy (2008): “‘Getting in, getting out’: militia membership and
prospects for reintegration in postwar Liberia”. In: The Journal of Modern African Studies
46:1, pp. 33-55.
Brandt, Marieke (2013): “Sufyān's “Hybrid” War: Tribal Politics during the Ḥūthī Conflict”.
In: Journal of Arabian Studies: Arabia, the Gulf, and the Red Sea 3:1, pp. 120-138.
Campbell, Bruce and Arthur Brenner (2000): Death Squads in Global Perspective: Murder
with Deniability. New York: Palgrave Macmillan.
Carey, Sabine C., Neil J. Mitchell and Will Lowe (2013): “States, the security sector, and the
monopoly of violence: A new database on pro-government militias”. In: Journal of Peace
Research 50:2, pp. 249-258.
Carpenter, Ted Galen (2012): “The U.S. Should Be Wary of Arming Syrian Rebels”. In: US
News and World Report, 1 November 2012.
Cederman, Lars E. and Kristian Skrede Gleditsch (2009): “Introduction to special issue on
‘Disaggregating civil war’”. In: Journal of Conflict Resolution 53:4, pp. 487–495.
Centeno, Miguel Angel (2002): Blood and Debt: War and the Nation State in Latin America.
University Park, PA: Penn State University Press.
Chivvis, Christopher S., Keith Crane, Peter Mandaville and Jeffrey Martini (2012): Libya’s
Post-Qaddafi Transition. The Nation-Building Challenge. Santa Monica: RAND Corporation.
Clutterbuck, Richard (1995): “Peru: Cocaine, Terrorism and Corruption”. In: International
Relations 12, pp. 77-92.
Collier, Paul and Anke Hoeffler (2004): “Greed and grievance in civil war”. In: Oxford Economic Papers 56:4, pp. 563-595.
Collier, Paul, Anke Hoeffler and Dominic Rohner (2009): “Beyond greed and grievance: feasibility and civil war”. In: Oxford Economic Papers 61:1, pp. 1-27.
Comisión de la verdad y reconciliación [Truth and Reconciliation Commission] (2003):
Anexo Estadístico. Informe Final [Final report]. Lima: Comisión de la verdad y
reconciliación.
Conibere, R. et al. (2004): Statistical appendix to the report of truth and reconciliation commission. Report of Sierra Leone, Human Rights Data Analysis Group, The Benetech Initiative.
24 Cornell, Svante E. (2005): “The Interaction of Narcotics and Conflict”. In: Journal of Peace
Research 42:6, pp. 751-760.
Cornell, Svante E. (2007): “Narcotics and Armed Conflict: Interaction and Implications”. In:
Studies in Conflict & Terrorism 30:3, pp. 207-227.
Costa, Gino (2006): “Two Steps Forward, One and a Half Steps Back: Police Reform in Peru,
2001–2004”. In: Civil Wars 8:2, pp. 215-230.
Crabtree, John (2002): “The impact of neo-liberal economics on Peruvian peasant agriculture
in the 1990s”. In: The Journal of Peasant Studies 29:3-4, pp. 131-161.
Cunningham, Kathleen Gallagher (2011): “Divide and conquer or divide and concede: How
do states respond to internally divided separatists?”. In: American Political Science Review
105:2, pp. 275-297.
Davenport, C. and M. Inman (2012): “The State of State Repression Research since the
1990s”. In: Terrorism and Political Violence 24:4, pp. 619–634.
Davies, Victor A.B. (2000): “Sierra Leone: Ironic Tragedy“. In: Journal of African Economies 9:3, pp. 349-369.
Davis, Diane E. (2009): “Non-State Armed Actors, New Imagined Communities, and Shifting Patterns of Sovereignty and Insecurity in the Modern World”. In: Contemporary Security
Policy 30:2, pp. 221-245.
Degregori, Carlos Iván (1996): Las rondas campesinas y la derrota de Sendero Luminoso.
Lima: IEP/Universidad Nacional de San Cristóbal de Huamanga.
Denov, Myriam and Richard Maclure (2007): “Turnings and Epiphanies: Militarization, Life
Histories, and the Making and Unmaking of Two Child Soldiers in Sierra Leone”. In: Journal
of Youth Studies 10:2, pp. 243-261.
DESCO (Centro de Estudios y Promoción del Desarollo) (1989): Violencia Política en el
Perú: 1980–1988. Vol. I & II. Lima: DESCO.
Dokubo, Charles (2000) “‘An Army for rent’, private military corporations and civil conflicts
in Africa: The case of Sierra Leone”. In: Civil Wars 3:2, pp. 51-64.
Dreyfus, Pablo G. (1999): “When all the Evils Come Together. Cocaine, Corruption, and
Shining Path in Peru's Upper Huallaga Valley, 1980 to 1995”. In: Journal of Contemporary
Criminal Justice 15:4, pp. 370-396.
Fearon, James D. and David D. Laitin (2003): “Ethnicity, insurgency, and civil war”. In:
American Political Science Review 97:1, pp. 75-90.
Felbab-Brown, Vanda (2005): “The Coca Connection: Conflict and Drugs in Colombia and
Peru”. In: The Journal of Conflict Studies 25:2, pp. 104-128.
Ferme, Mariane C. and Danny Hoffman (2004): “Hunter Militias and the International Human Rights Discourse in Sierra Leone and Beyond”. In: Africa Today 50:4, pp. 73-95.
Fielding, David and Anja Shortland (2012): “The dynamics of terror during the Peruvian civil
war”. In: Journal of Peace Research 49:6, pp. 847-862.
25 Fjelde, Hanne and Desirée Nilsson (2012): “Rebels against Rebels: Explaining Violence between Rebel Groups”. In: Journal of Conflict Resolution 56:4, pp. 604-628.
Francis, David J. (1999): “Mercenary intervention in Sierra Leone: Providing national security or international exploitation?”. In: Third World Quarterly 20:2, pp. 319-338.
Francis, David J. (2000): “Torturous Path to Peace: The Lomé Accord and Postwar
Peacebuilding in Sierra”. In: Security Dialogue 31:3, pp. 357-373.
Fumerton, Mario (2001): “Rondas Campesinas in the Peruvian Civil War: Peasant Selfdefence Organisation in Ayacucho”. In: Bulletin of Latin American Research 20:4, pp. 470497.
Gause III, F. G. (2011): “Why Middle East Studies Missed the Arab Spring. The Myth of
Authoritarian Stability”. In: Foreign Affairs 90:4, pp. 81–90.
Giblin, Susan (2007): “Violent Conflicts in Indonesia”. In: The International Journal of Human Rights 11:4, pp. 517-527.
Gitlitz, John S. and Telmo Rojas (1983): “Peasant Vigilante Committees in Northern Peru”.
In: Journal of Latin American Studies 15:1, pp. 88-92.
Gleditsch, Nils Petter, Peter Wallensteen, Mikael Eriksson, Margareta Sollenberg and Håvard
Strand (2002): “Armed Conflict 1946–2001: A new dataset”. In: Journal of Peace Research
39:5, pp. 615-637.
González-Cueva, Eduardo (2000): “Conscription and Violence in Peru”. In: Latin American
Perspectives 27:3, pp. 88-102.
Harmon, Christopher C. Harmon (1992): “The purposes of terrorism within insurgency: Shining path in Peru”. In: Small Wars & Insurgencies 3:2, pp. 170-190.
Hegre, Håvard and Nicholas Sambanis (2006): “Sensitivity analysis of empirical results on
civil war onset”. In: Journal of Conflict Resolution 50:4, pp. 508-535.
Herreros, F. (2006): “‘The Full Weight of the State’: The Logic of Random State-Sanctioned
Violence”. In: Journal of Peace Research 43:6, pp. 671–689.
Hills, Alice (2011): “‘War don don’: stability, normalcy and Sierra Leone, Conflict”. In: Security & Development 11:1, pp. 1-24.
Hoffman, Danny (2003): “Like beasts in the bush: synonyms of childhood and youth in Sierra Leone”. In: Postcolonial Studies 6:3, pp. 295-308.
Hoffman, Danny (2004): “The Civilian Target in Sierra Leone and Liberia: Political Power,
Military Strategy, and Humanitarian Intervention”. In: African Affairs 103, pp. 211-226.
Hoffman, Danny (2006): “Disagreement: Dissent Politics and the War in Sierra Leone”. In:
Africa Today 52:3, pp. 3-22.
Hoffman, Danny (2007): “The meaning of a militia: understanding the civil defence forces of
Sierra Leone”. In: African Affairs 106:425, pp. 639-662.
26 Huggins, Martha (1991): Vigilantism and the State in Modern Latin America: Essays on Extralegal Violence. New York: Praeger Publishers.
Human Rights Watch (1996): Colombia’s Killer Networks: The Military-Paramilitary Relationship Partnership and the United States. Washington: Human Rights Watch.
Human Rights Watch (2011a): “Just Don’t Call It a Militia”. Impunity, Militias, and the “Afghan Local Police”. New York: Human Rights Watch.
Human Rights Watch (2011b): Côte d’Ivoire: Violence Campaign by Security Forces, Militias. New York: Human Rights Watch.
Human Rights Watch (2012): Philippines: No Progress Disarming, Demobilizing Militias.
New York: Human Rights Watch.
Human Rights Watch (2013): For Libya to end the violence, it needs to shut down the militias. New York: Human Rights Watch.
Humphreys, M. and J. M. Weinstein (2006): “Handling and manhandling civilians in civil
war: determinants of the strategies of warring factions”. In: American Political Science Review 100:3, pp. 429-447.
Humphreys, M. and J. M. Weinstein (2007): “Demobilization and Reintegration”. In: Journal
of Conflict Resolution 51:4, pp. 531-567.
Humphreys, M. and J. M. Weinstein (2008): “Who Fights? The Determinants of Participation
in Civil War”. In: American Journal of Political Science 52:2, pp. 436-455.
Hyland, Frank (2008): “Peru’s Sendero Luminoso: From Maoism to Narco-Terrorism”. In:
TerrorismMonitor 6:23, pp. 1-3.
Ikelegbe, Augustine (2006): “Beyond the threshold of civil struggle: youth militancy and the
militia-ization of the resource conflicts in the Niger Delta region of Nigeria”. In: African
Study Monographs, 27:3, pp. 87-122.
International Crisis Group (2001): Sierra Leone: Managing Uncertainty. Africa Report No.
35. Freetown/Brussels: International Crisis Group.
International Crisis Groups (2002a): Sierra Leone after Elections: Politics as Usual? Africa
Report No. 49. Freetown/Brussels: International Crisis Group.
International Crisis Group (2002b): Sierra Leone's Truth and Reconciliation Commission: A
Fresh Start? Africa Briefing No. 12. Freetown/Brussels: International Crisis Group.
International Crisis Groups (2003): Sierra Leone: The State of Security and Governance. Africa Report No. 67. Freetown/Brussels: International Crisis Group.
International Crisis Group (2004): Liberia and Sierra Leone: Rebuilding Failed States. Africa
Report No. 87. Freetown/Brussels: International Crisis Group.
Kabia, John M. (2008): “Greed or Grievance? Diamonds, Rent-Seeking and the Civil War in
Sierra Leone (1991-2002)”. In: Omeje, Kenneth (ed.): Extractive Economies and Con-flicts
in the Global South. Aldershot: Ashgate, pp. 93-105.
27 Kalyvas, S. N. (2004): “The Paradox of Terrorism in Civil War”. In: Journal of Ethics 8:1, pp.
97–138.
Kalyvas, S. N. (2006): The Logic of Violence in Civil War. Cambridge: Cambridge University Press.
Kalyvas, S. N., and M. A. Kocher (2007): “How ‘Free’ Is Free Riding in Civil Wars: Violence, Insurgency, and the Collective Action Problem”. In: World Politics 59:2, pp. 177–216.
Kay, B. (1999): “Violent Opportunities: The Rise and Fall of ‘King Coca’ and Shining Path.”
In: Journal of Interamerican Studies and World Affairs 41:3, pp. 97-127.
Kay, C. (1982): “Achievements and contradictions of the Peruvian agrarian reform”. In: The
Journal of Development Studies 18:2, pp. 141-170.
Kay, C. (2001): “Estructura Agraria, Conflicto y Violencia en la Sociedad Rural de América
Latina.” In: Revista Mexicana de Sociología 63:4, pp. 159-195.
Keen, David (2003): “Greedy Elites, Dwindling Resources, Alienated Youths. The Anatomy
of Protracted Violence in Sierra Leone”. In: International Politics and Society 2, pp. 67-94.
Keen, David (2005): Conflict and collusion in Sierra Leone. New York: Palgrave.
Kirkpatrick, David D. (2013): “Violence Against Libyan Protesters Threatens to Undercut
Power of Militias”. In: New York Times, 9 June 2013.
Koc-Menard, S. (2006): “Switching from Indiscriminate to Selective Violence: The Case of
the Peruvian Military (1980–95).” In: Civil Wars 8:3-4, pp. 332–354.
Kruijit, Dirk (1999): “Exercises in State Terrorism: The Counter-Insurgency Campaigns in
Guatemala and Peru”. In: Koonings, Kees and Dirk Kruijt (ed.): Societies of Fear: the Legacy
of Civil War, Violence, and Terror in Latin America. London: Zed Books, pp. 35–43.
Kucera, Michal and Miroslav Mares (2013): “Vigilantism during democratic transition”. In:
Policing and Society: An International Journal of Research and Policy, forthcoming.
Kwaja, Chris M. A. (2010): “The feasibility of security sector reform and access to justice in
Sudan: Challenges and prospects”. In: African Security Review 19:1, pp. 6-19.
Langdon, Jennifer and Mery Rodriguez (2007): “The Rondas Campesinas of Peru”. In: Hancock, Landon and Christopher Mitchell (ed.): Zones of Peace. Bloomfield: Kumarian Press,
pp. 91-104.
Le Billon, Philippe (2001): “The political ecology of war: natural resources and armed conflicts”. In: Political Geography 20:5, pp. 561-584.
Lujala, Päjvi, Nils Petter Gleditsch and Elisabeth Gilmore (2005): “A Diamond Curse? Civil
War and a Lootable Resource”. In: Journal of Conflict Resolution 49:4, pp. 538-562.
Lutterbeck, Derek (2013): “Arab Uprisings, Armed Forces, and Civil–Military Relations”. In:
Armed Forces & Society 39:1, pp. 28-52.
Lyall, L. (2009): “Does Indiscriminate Violence Incite Insurgent Attacks? Evidence from
Chechnya”. In: Journal of Conflict Resolution 53:3, pp. 331–362.
28 Maclure, Richard and Myriam Denov (2006): ““I Didn't Want to Die So I Joined Them”:
Structuration and the Process of Becoming Boy Soldiers in Sierra Leone”. In: Terrorism and
Political Violence 18:1, pp. 119-135.
Macmullin, Colin and Maryanne Loughry (2004): “Investigating Psychosocial Adjustment of
Former Child Soldiers in Sierra Leone and Uganda”. In: Journal of Refugee Studies 17:4, pp.
460-472.
Maitre, Benjamin R. (2009): “What Sustains ‘Internal Wars’? The dynamics of violent conflict and state weakness in Sudan”. In: Third World Quarterly 30:1, pp. 53-68.
Manwaring, Max G. (2002): “Non-State Actors in Colombia: Threats to the State and to the
Hemisphere”. In: Small Wars & Insurgencies 13:2, pp. 68-80.
Marshall, Monty G., Ted Robert Gurr and Keith Jaggers (2010): Polity IV Project: Dataset
Users’ Manual. Center for Systemic Peace: George Mason University.
Mason, T. David and C. Campany (1995): “Guerrillas, Drugs, and Peasants: The Rational
Peasant and the War on Drugs in Peru.” In: Terrorism and Political Violence 7:4, pp. 140-170.
Mason, T. David and Dale A. Krane (1989): “The Political Economy of Death Squads: Toward a Theory of the Impact of State-Sanctioned Terror”. In: International Studies Quarterly
33:2, pp. 175-198.
Mayer, Ann Elizabeth (2013): “Essay: Building the New Libya: Lessons to Learn and to Unlearn”. In: University of Pennsylvania Journal of International Law 34:2, pp. 365-387.
Mazzei, Julie (2009): Death Squads of Defense Forces? How Paramilitary Groups Emerge
and Challenge Democracy in Latin America. Chapel Hill, NC: University of North Carolina
Press.
McClintock, C. (1986): “Sendero Luminoso: La Guerilla Maoista Del Perú”. In: Revista Occidental 3:2, pp. 131-153.
McClintock, C. (1989): “The Prospects for Democratic Consolidation in a ‘Least Likely’
Case Peru”. In: Comparative Politics 21:2, pp. 127-148.
Millar, Gearoid (2012): “'Our brothers who went to the bush': Post-identity conflict and the
experience of reconciliation in Sierra”. In: Journal of Peace Research 49:5, pp. 717-729.
Mitchell, Neil (2004): Agents of Atrocity: Leaders, Followers and Violation of Human Rights
in Civil Wars. New York: Palgrave Macmillan.
Mitton, Kieran (2009): “Reconstructing Trust in Sierra Leone”. In: The Round Table: The
Commonwealth Journal of International Affairs 98:403, pp. 461-471.
Mitton, Kieran (2012): “Irrational Actors and the Process of Brutalisation: Understanding
Atrocity in the Sierra Leonean Conflict (1991–2002)”. In: Civil Wars 14:1, pp. 104-122.
Mkutu, Kennedy (2008): “Disarmament in Karamoja, Northern Uganda: Is This a Solution
for Localised Violent Inter and Intra-Communal Conflict?”. In: The Round Table: The
Commonwealth Journal of International Affairs 97:394, pp. 99-120.
29 Møller, Bjørn (2006): The role of militias and other paramilitaries in African (un)civil wars.
Copenhagen: DIIS.
Mucha, Witold (2013a): “Does counterinsurgency fuel civil war? Peru and Syria compared”.
In: Critical Studies on Terrorism 6:1, pp. 140-166.
Mucha, Witold (2013b): “Prone to Conflict, but Resilient to Violence. Why Civil Wars
Sometimes Do Not Happen: Insights from Peru and Bolivia”. In: Journal of Peace, Conflict &
Development 20, pp. 97-114.
Murphy, William P. (2003): “Military Patrimonialism and Child Soldier Clientalism in the
Liberian and Sierra Leonean Civil Wars”. In: African Studies Review 46:2, pp. 61-87.
Ndumbe, J. Anyu (2001): “Diamonds, Ethnicity, and Power: The Case of Sierra Leone”. In:
Mediterranean Quarterly 12:4, pp. 90-105.
Nilsson, Desirée and Mimmi Söderberg Kovacs (2013): “Different Paths of Reconstruction:
Military Reform in Post-War Sierra Leone and Liberia”. In: International Peacekeeping 20:1,
pp. 2-16.
Nyabola, H Nanjala (2009): “The legal challenge of civil militia groups in Kenya”. In: African Security Review 18:3, pp. 89-102.
Obando, E. (2006): “U.S. Policy toward Peru: At Odds for Twenty Years.” In: Loverman, B.:
Addicted to Failure. U.S. Security Policy in Latin America and the Andean Region. Lanhman:
Rowman & Littlefield Publishers, pp. 169–196.
Ohlson, T. (2008): “Understanding Causes of War and Peace”. In: European Journal of International Research 14:1, pp. 133–160.
Pasára, Luis (1991): “Ambivalencia en los nuevos actores sociales: La experiencia Peruana”.
In: Nueva Sociedad 115, pp. 56-68.
Pelham, Nicolas (2012): “Libya in the Shadow of Iraq: The ‘Old Guard’ Versus the Thuwwar
in the Battle for Stability”. In: International Peacekeeping 19:4, pp. 539-548.
Penfold, Peter (2005): “Faith in resolving Sierra Leone's bloody conflict”. In: The Round
Table: The Commonwealth Journal of International Affairs 94:382, pp. 549-557.
Peters, Krijn and Paul Richards (1998): “'Why We Fight': Voices of Youth Combatants in
Sierra Leone”. In: Africa: Journal of the International African Institute 68:2, pp. 183-210.
Raleigh, Clionadh (2012): “Violence Against Civilians: A Disaggregated Analysis”. In: International Interactions: Empirical and Theoretical Research in International Relations 38:4,
pp. 462-481.
Reno, W. (1995): Corruption and state politics in Sierra Leone. New York: Cambridge University Press.
Reno W. (2003): “Political Networks in a Failing State. The Roots and Future of Violent
Conflict in Sierra Leone”. In: International Politics and Society 2, pp. 29-43.
Reno, W. (2007): “Patronage Politics and the Behavior of Armed Groups”. In: Civil Wars 9:4,
pp. 324-342.
30 Reno, W. (2009): “Explaining Patterns of Violence in Collapsed States”. In: Contemporary
Security Policy 30:2, pp. 356-374.
Reno, W. (2010): “Transforming West African Militia Networks for Postwar Recovery”. In:
Troubled Regions and Failing States: The Clustering and Contagion of Armed Conflict Comparative Social Research 27, pp. 127-149.
Richards, Paul (1996): Fighting for the Rainforest: War, Youth and Resources in Sierra Leone. London: James Currey.
Richards, Paul (2002): “Militia Conscription in Sierra Leone: Recruitment of Young Fighters
in an African War”. In: The Comparative Study of Conscription in the Armed Forces 20, pp.
255-276.
Richards, Paul (2003): The Political Economy of Internal Conflict in Sierra Leone. Working
Paper 21. The Hague: Netherlands Institute of International Relations ‘Clingendael’ Conflict
Research Unit.
Richards, Paul (2005): “To Fight or to Farm? Agrarian Dimensions of the Mano River Conflicts (Liberia and Sierra Leone)”. In: African Affairs 104/417, pp. 571-590.
Richards, Paul (2006): “An Accidental Sect: How War Made Belief in Sierra Leone”. In:
Review of African Political Economy 33:110, pp. 651-663.
Richards, Paul (2009): “Dressed to Kill. Clothing as Technology of the Body in the Civil War
in Sierra Leone”. In: Journal of Material Culture 14:4, pp. 495-512.
Ross, Michael L. (2004a): “What Do We Know about Natural Resources and Civil War?”. In:
Journal of Peace Research 41:3, pp. 337-356.
Ross, Michael L. (2004b): “How Do Natural Resources Influence Civil War? Evidence from
Thirteen Cases”. In: International Organization 58:1, pp. 35-67.
Sanchez, W. Alejandro (2003): “The Rebirth of Insurgency in Peru”. In: Small Wars & Insurgencies 14:3, pp. 185-198.
Sanín, Francisco Gutiérrez and Ana María Jaramillo (2004): “Crime, (counter-)insurgency
and the privatization of security - the case of Medellín, Colombia”. In: Environment and Urbanization 16:2, pp. 17-30.
Schneckener, Ulrich (2007): “Armed Non-State Actors and the Monopoly of Force”. In:
Bailes, Alyson/ Schneckener, Ulrich/ Wulf, Herbert: Revisiting the State Monopoly on the
Legitimate Use of Force. Policy Paper No. 24. Geneva: Geneva Centre for the Democratic
Control of Armed Forces (DCAF), pp. 10-18.
Shepler, Susan (2005): “The Rites of the Child: Global Discourses of Youth and Reintegrating Child Soldiers in Sierra Leone”. In: Journal of Human Rights 4:2, pp. 197-211.
Shultz, Richard H. (2005): “The Era of Armed Groups”. In: Berkowitz, Peter (ed.): The Future of American Intelligence. Stanford: Hoover Press, pp. 1-39.
Sluka, Jeffrey (2000): Death Squad: The Anthropology of State Terror. Philadelphia: University of Philadelphia Press.
31 Smith, L. Alison, Catherine Gambette, and Thomas Langley (2004): Conflict Mapping in
Sierra Leone: Violations of International Humanitarian Law from 1991 to 2002. New York:
No Peace Without Justice.
Stanley, William (1996: The Protection Racket State: Elite Politics, Military Extortion, and
Civil War in El Salvador. Philadelphia, PA: Temple University Press.
Starn, Orin (1995): “To Revolt against the Revolution: War and Resistance in Peru's Andes”.
In: Cultural Anthropology 10:4, pp. 547-580.
Starn, Orin (1999): Nightwatch: The Politics of Protest in the Andes. Durham, NC: Duke
University Press.
Sullivan, John P. (2010): “Counter-supply and counter-violence approaches to narcotics trafficking”. In: Small Wars & Insurgencies 21:1, pp. 179-195.
Truth and Reconciliation Commission of Sierra Leone (2004): Witness to Truth: Final Report
of the TRC. Accra: Graphic Packaging.
Utas, Mats and Magnus Jörgel (2008): “The West Side Boys: military navigation in the Sierra
Leone civil war”. In: The Journal of Modern African Studies 46:3, pp. 487 511.
Van Dun, Mirella (2013): ““It’s Never a Sure Deal”: Drug Trafficking, Violence, and Coping
Strategies in a Peruvian Cocaine Enclave (2003-2007)”. In: Journal of Drug Issues 43:3, pp.
1-17.
Verhoeven, Harry (2009): “The self-fulfilling prophecy of failed states: Somalia, state collapse and the Global War on Terror”. In: Journal of Eastern African Studies 3:3, pp. 405-425.
Vincent, James Bibi Maiah Vincent (2013): “A Village-Up View of Sierra Leone’s Civil War
and Reconstruction”. In: IDS Bulletin 44:1, pp. 30-43.
Ware, Helen/ Ogunmola, Dele (2010): “Probing the Roles of Governance and Greed in Civil
Strife in West Africa“. In: Goldsmith, Benjamin E./ Brauer, Jurgen (ed.): Economics of War
and Peace: Economic, Legal, and Political Perspectives. Bingley: Emerald, pp. 71-87.
Weinstein, Jeremy (2003): “A New Threat of Terror in the Western Hemisphere”. In: SAIS
Review 23:1, pp. 1-17.
Weinstein, Jeremy (2005): “Resources and the Information Problem in Rebel Recruitment”.
In: Journal of Conflict Resolution 49:4, pp. 598-624.
Williamson, John (2006): “The disarmament, demobilization and reintegration of child soldiers: social and psychological transformation in Sierra Leone”. In: Intervention 4:3, pp. 185205.
Wilson, Chris (2011): “Provocation or Excuse? Process-Tracing the Impact of Elite Propaganda in a Violent Conflict in Indonesia”. In: Nationalism and Ethnic Politics 17:4, pp. 339360.
Wlodarczyk, Nathalie (2009): “Politically Enfranchising the Non-political: Safeguarding
Peace through Civic Education and Inclusion? The Civil Defence Forces in Sierra Leone”. In:
Civil Wars 11:2, pp. 200-214.
32 Wood, Reed M. (2010): “Rebel capability and strategic violence against civilians”. In: Journal of Peace Research 47:5, pp. 601-614.
Zack-Williams, Alfred B. (1999): “Sierra Leone: The political economy of civil war, 199198”. In: Third World Quarterly 20:1, pp. 143-162.
Zack-Williams, Alfred B. (2001): “Child soldiers in the civil war in Sierra Leone”. In: Review of African Political Economy 28:87, pp. 73-82.
Zack-Williams, Tunde B. (2006): “Child Soldiers in Sierra Leone and the Problems of Demobilisation, Rehabilitation and Reintegration into Society: Some Lessons for Social Workers in War-torn Societies”. In: Social Work Education 25:2, pp. 119-128.
Zahar, Marie-Joëlle (2000): “Protégés, clients, cannon fodder: Civilians in the calculus of
militias”. In: International Peacekeeping 7:4, pp. 107-128.
33