Controlling escalation is an illusion. It was an illusion during the period of the Cold War, when, fortunately, the reality of controlling nuclear escalation between the superpowers never presented itself. Unfortunately, today the reality of attempting to control escalation regularly presents itself. Many contemporary belligerents are either insufficiently aware of the escalatory potential of their actions or tend to be preoccupied by short-term considerations. This book details the variety in the processes of escalation and challenges the idea of seeing escalation as an entirely rational and linear phenomenon over which control can be exercised.
In the past few years, significant scholarly attention has been devoted to the changing character of war. Experts have noted shifts in the participants in war, its driving forces, its political utility and its practices. Far less attention has been given to the changes within individual wars over time. This book aims to address those changes that occur within wars, once hostilities have started and before they terminate. It operates from the premise that, as Carl von Clausewitz, the founding father of the scientific study of war has formulated: ‘the original political objects [of war] can greatly alter during the course of war and may finally change entirely since they are influenced by events and their probable consequences’ (Reference Clausewitz, Howard and PeterClausewitz 1993, 104, italics in original). How can confrontations that appear at first sight to start small-scale and inconspicuous – a bomb attack or skirmishes by a band of rebels in the countryside – end in large-scale conflicts with huge investments in human lives and material, dragging on for years, if not decades? How does this process of escalation occur?
Clausewitz stipulated that war, in fact, possesses a natural propensity to escalate (Reference Clausewitz, Howard and PeterClausewitz 1993; Reference CimbalaCimbala 2012). War is a duel on a large scale, and the opponents seek out each other’s weak points to gain the upper hand. This process, in principle, does not possess any boundaries. War has a natural tendency to escalate into infinity. This is absolute war. The main limits to escalation, in a Clausewitzean sense, are politics and friction. The first can be related to factors affecting the will of the actor to persevere and the second mainly to the capabilities to do so. Both form the parameters along which war escalates, constituting war in practice. Political will refers to the idea and use of power in a specific context. Capabilities, in contrast to will, are often but, not exclusively, material and revolve around those instruments and resources that can be used to press the willpower.
This book starts from the premise that Clausewitzean thinking applies to rebels just as much as it applies to states: ‘Clausewitz theory of war will remain valid as long as warlords, drug barons, international terrorists, racial or religious communities will wage war’ (Reference EchevarriaEchevarria 1996, 80; Reference DuyvesteynDuyvesteyn 2005; Reference SchuurmanSchuurman 2010).
Clausewitz used an analogy of war as a chameleon, which is very illustrative for the central problem this book aims to address.
War is more than a true chameleon that slightly adapts in characteristics to the given case. As a total phenomenon its dominant tendencies always make war a paradoxical trinity – composed of primordial violence, hatred and enmity, which are to be regarded as a blind natural force, of the play of chance and probability within which the creative spirit is free to roam; and of its elements of subordination, as an instrument of policy, which makes it subject to reason alone.
The analogy of the chameleon has been subject to a series of interpretations, because it touches the heart of what Clausewitz sees as the nature of war (Reference Bassford, Strachan and Herberg-RotheBassford 2007). According to Hew Strachan, ‘war may indeed be a chameleon in that it changes its nature slightly in each individual case (its “character”) but not its nature in general, which is made up of the trinity’ (Reference Strachan, Herberg-Rothe, Strachan and Herberg-RotheStrachan and Herberg-Rothe 2007, 3). While the political logic presses war into a constraining framework, the production of violence, including the ebb and flow of the levels of violence, derives from its three immutable elements: passion, reason and chance linked to people, government and armed force, respectively, which make up the trinity of war (Reference Clausewitz, Howard and PeterClausewitz 1993). Indeed, ‘it is the interactive character of war – Clausewitz’s famous chameleon “that adapts its characteristic to a give case” – that has proven the most original avenue for analysis’ of war (Reference EvansEvans 2003, 141; Reference Duyvesteyn, Malkki, Jackson and SinclairDuyvesteyn 2012). These ideas will act as a guide in the investigations that follow. How does the trinity of people, government and armed forces contribute to, and shape, the production of violence?
The aim of this study is to think through the processes of escalation and de-escalation. The focus will be in particular on non-state actors or rebel violence, as this particular violent agent dominates in our contemporary experience of violence.Footnote 1 In fact, violent non-state actors have been a predominant actor in war for the past two centuries (Reference HolstiHolsti 1996). Furthermore, in contrast to a large part of the literature in war studies, this book will not devote much attention to the causes of war but will focus in particular on the dynamics after its initial outbreak and before its termination. There is a need to carefully think through violence in war, and its escalatory and de-escalatory workings.
The main argument of this book is that escalation in the case of rebel conflict, rather than a clearly conceptualised ladder with ever-increasing thresholds of pain, is a messy process marked by unexpected consequences of choices that were rushed into or given little prior strategic thought. The study, exploratory in nature, uses existing material to piece together the potential pathways of escalation. These will be presented in the shape of propositions, which will await further testing and refinement. Important generic thresholds can be observed, mainly with hindsight, in which situations of war gain unprecedented characteristics denoting an aggravation of conflict. Escalation ensued, for example, when the saliency of the perceived issues at stake was raised, either as a result of violence or concessions. Escalation materialised when countermeasures were enacted, new actors became involved or new weaponry was introduced. Escalation also took shape when an extremity shift occurred within the rebel group, largely unrelated to external factors or pressure. De-escalation resulted in the past when groups de-legitimised themselves through strategic mistakes, lost the support of external sponsors or a convergence of norms between belligerents took place.
The case for explaining rebel escalation and de-escalation will be set out in this chapter. We will unpack the idea of the rebel and introduce the concepts of escalation and de-escalation. In subsequent chapters, the existing literature on the rise and decline of rebel violence will be discussed and a method for measuring escalation and de-escalation will be proposed. The rest of the study is set up along thematic lines. The three key elements in war, as identified by Carl von Clausewitz – politics, military and people – will be used as a starting point for discussion. These essential elements can in some senses be seen to overlap with an imaginary trajectory of conflict from incipient, aggravated, extremely violent to a winding down, lessening of aggression and eventual resolution. The ideas about the main thresholds of escalation will be illustrated by short vignettes or brief case examples to show how these different processes have played out in the past. The vignettes will be based on experiences of armed conflict since 1945. Firstly, the idea of rebels as strategic actors, embracing a political agenda linked to means and methods to carry this out, will be presented.
Rebels
The Rebel Actor
Looking at rebels, among which we count guerrillas, terrorist and insurgents, we are confronted with the fact that they form a very diverse, and according to some, even largely unstable analytical, category.Footnote 2
Not only do underground organizations differ according to their goals, they also have varying organizational models and favor different forms of action. Any attempt to develop interpretative hypotheses about “terrorism” is therefore destined to fail without a typology that can identify their range of application.
The rich material that scholars in the field of civil war studies have presented over the past years has, indeed, questioned practically every aspect of rebel violence. Rebels cannot be seen as unitary actors; their political agendas are highly changeable, as are their means and methods. Distinguishing them even from the state, that is seeing them as non-state actors is difficult in light of examples of state–rebel collusion (e.g. the discussion about militias: Reference StanilandStaniland 2012c; Reference JentzschJentzsch 2014; Reference SchneckenerSchneckener 2017; see also Reference Idler and ForestIdler and Forest 2015). Nevertheless, there continues to exist a pressing research agenda, also recognised by the many scholars involved in this enterprise. There is a need to explain the varied empirical reality of political violence in the international system today. Actors distinct from the state play a major role here. The aim of this paragraph is to unpack the concept of the rebel group and come to a workable delimitation of the phenomenon.
Rebels come in many shapes and guises. Some are well organised, hierarchical and centralised. Others are loose, flat networks or even systems, without a clear centre or leadership. Furthermore, there are several ways of looking at the development of rebels over time. Firstly, rebels can be the product of social movements, in particular social mobilisation. There is a large literature available on the establishment and growth of social movements (e.g. Reference Goodwin and JasperGoodwin and Jasper 2009). Social movements can be accompanied by a radical fringe from which violent rebel groups can spring (Reference MarsdenMarsden 2016). Examples are those groups originating from left-wing activism in the 1960s and 1970s, such as the Rote Armee Fraktion (RAF) in Germany and the Brigate Rosse in Italy but also the groups fighting independence struggles in the decolonisation period, such as in Algeria and Indochina.
Secondly, rebels can be the product of small-scale conspiracies or even individual enterprises. This is where terrorist strategies are historically derived from (Reference Miller and CrenshawMiller 1995). A small group of individuals, who adhere to a radical agenda, can decide to band together to trigger change. An example of an individual terrorist campaign is the so-called UNA bomber, Ted Kaczynski, who single-handedly terrorised the United States between 1975 until 1998 with a bombing campaign aimed at symbols of technological progress which he opposed, universities and airports (Reference ChaseChase 2003). The idea of leaderless resistance originates within right-wing extremist circles (Reference KaplanKaplan 1997; Reference MichaelMichael 2012). Also, the Focoist idea of a small dedicated cadre igniting a people’s revolution can be included here (Reference DebrayDebray 1973).
Thirdly, rebels can also be a construct from the outside with little relation to any form of an organised unit (Reference SimpsonSimpson 2012). An interesting recent example is the conflict in Afghanistan, where it has been argued that ‘the generic insurgency … is a rhetorical rather than operational construct’. More than ‘one-third of all violent attacks nationwide (and more than half in the South [of Afghanistan]) attributed to the insurgency involve local power tussles between communities and tribes – not Taliban members or insurgents – which perceive themselves as marginalized in the distribution of political power, land, water, and other government-controlled resources’ (Reference Barakat and ZyckBarakat and Zyck 2010, 197). The insurgency is thus argued to be a perception, construct or even a convenient label.
Of course, rebels can move from a small group conspiracy into a social movement and vice versa and both can be perspectives constructed by the outside rather than a factual reality. The discussion about Al Qaeda as an ideology or idea rather than an organisation is one such example (Reference SagemanSageman 2011).
Scholarship focused on civil war has in recent years moved beyond the conceptualisation of rebel groups as unitary actors. Attempts have been made to model rebels according to their level of organisation. Using the ideas of network, scholars have developed different perspectives on actor coherence (Reference Sanín and GiustozziSanín and Giustozzi 2010; Reference Bakke, Cunningham and SeymourBakke, Cunningham and Seymour 2012; Reference Pearlman and CunninghamPearlman and Cunningham 2012; Reference KrauseKrause 2014; Reference StanilandStaniland 2014: Reference BultmannBultmann 2018). By looking at the number of organisations in a social movement, the degree of institutionalisation across these organisations and the distribution of power among them, actor cohesion can be measured (Reference Bakke, Cunningham and SeymourBakke, Cunningham and Seymour 2012). The main idea is that the larger the number of organisations in the movement, combined with a weak degree of institutionalisation and a large power disparity, the greater the chances of violence. Conversely, one dominant, institutionalised and powerful actor will decrease the chances of fragmentation in the conflict. To what extent this actor will be able to escalate and act wilfully remains to be seen.
What is important at this stage is that the structure of rebel groups has been found to have important consequences for the engagement in violence (Reference StanilandStaniland 2012b; Reference CunninghamCunningham 2013). Movement structures which carry the favour of the members are more likely to withstand external pressure and violence compared to groups with contentious structures. The latter are more likely to disintegrate when outside pressure is applied (Reference PearlmanMcLauchlin and Pearlman 2012).
We have seen in recent years that violence among rebels themselves and against unarmed civilians rather than the state has increased (Reference DeRouen and BercovitchDeRouen and Bercovitch 2008; Reference Cunningham, Gleditsch and SalehyanCunningham, Gleditsch and Salehyan 2009). Some scholars have gone as far as doubting whether the state as an object of struggle is of any significance at all in contemporary armed conflict (Reference KilcullenKilcullen 2006). It cannot be denied, however, that the state often remains the referent object. Issues of contention often relate to imperfect state formation and consolidation: territorial contestation or power divisions in political systems (Reference WeinsteinWeinstein 2007; Uppsala Conflict Data Program 2020; Reference Newman and DeRouenNewman and DeRouen Jr. 2014). The warlords in Afghanistan fought for access to the state. The militias in Sudan claim that the government in Khartoum had forfeited its right to legitimate rule in Darfur. The Fuerzas Armadas Revolucionarias de Colombia, or FARC (Revolutionary Armed Forces of Colombia), apart from being involved in the drugs trade, envisioned an overhaul of the perceived unjust political and social order in the country. The Kurdistan Workers Party (PKK) fighting in Turkey aimed for the recognition of Kurdish minority rights and an independent Kurdish state. Therefore, this study has opted to predominantly focus on rebel–state interaction in particular, rather than intra-rebel conflict.
However, in a large part of the academic literature, there appears to be a disconnect between the activities of the rebel group and the state. As noted by some critical security scholars, rebel violence can only be understood in conjunction with the role of the state (Reference BlakeleyBlakeley 2007; Reference JacksonJackson 2007). Most of the theories attempting to understand rebel violence are ‘ultimately socially constructed in opposition to state authority and so there is an inescapable sense in which the state itself must play a role in their creation’ (Reference ParkerParker 2007, 156–157). Overall, existing theories award ‘some regular importance to the actions of governments. In particular, official “coercion” –sometimes designated “repression” or “retribution” – is generally considered to affect the frequency, magnitude and intensity of violent action’ (Reference SnyderSnyder 1976, 278, emphasis in original). The state is, therefore, logically part of the phenomenon of rebel violence rather than outside of it. Still, many studies into terrorist or insurgent campaigns accept the role of the state uncritically; the ‘conflict management approach conceptually mistreats violence by ignoring the state’s role in it’ (Reference SnyderSnyder 1976, 283).
This state-centred perspective leads to a very fundamental disconnect between rebel violence in the shape of terrorism and insurgency and the countermeasures taken against these strategies. Few of the existing studies, perhaps with the exception of specific case studies, treat the strategy in conjunction with countermeasures. Countermeasures are highly dependent on the policy perspective the state maintains. In the case of the United States, ‘counterterrorism policy is not just a response to the threat of terrorism, whether at home or abroad, but a reflection of the domestic political process’ (Reference CrenshawCrenshaw 2001, 329). When terrorism is seen as a criminal act, counterterrorism is a law enforcement problem. When terrorism is seen as a security problem, counterterrorism becomes a police and security services problem. Counterterrorism and counter-insurgency are often taken unjustifiably, as distinct and separate from terrorism and insurgency.
Despite the challenges to the rebel concept highlighted in this paragraph, the rebel group in all its different guises remains an important subject for investigation. For the purposes of this study, a rebel group will be defined as a sub- or non-state actor which has mounted a violent challenge against state power. It is identifiable as an actor through its threats and acts of violence. Furthermore, the rebel group is a political actor. The reasons for ascribing political agency to the rebel group will now be outlined.
Rebel Ends
According to some notable recent assessments, rebels are non-strategic and non-political actors. Scholars putting forward this point of view have questioned the validity of an instrumental approach to rebel violence, and they have doubted the existence of a means–ends relationship in rebel conflict. These claims have been based both on theoretical and empirical arguments. These ideas are part of a wider discussion about what war is about, which started in the early 1990s with the publication of Martin van Creveld’s book Transformation of War (Reference Van CreveldVan Creveld 1991). Van Creveld argued that war is not a product of politically guided actors seeking the attainment of goals. Rather, war is pursued for its own sake, for personal recognition, prowess and honour. Subsequently, others pointed at the wilful targeting of civilians, barbarity, ethnic factors, culture and greed to argue that war was beyond the political reins that Clausewitz had argued, kept it in check (Reference KeeganKeegan 1993; Reference KaplanKaplan 1994; Reference KaldorKaldor 2001; Reference Collier and HoefflerCollier and Hoeffler 2004; Reference AbrahmsAbrahms 2008, Reference Abrahms2011, Reference Abrahms2018). Several scholars have in fact argued that rebel violence should not be seen as possessing any kind of strategic attributes at all but as primarily geared towards communication, theatre and performance (Reference CrelinstenCrelinsten 1987, Reference Crelinsten, Wilkinson and Stewart2002).
Nevertheless, there is plentiful evidence in social science investigations that rebel violence does bear witness to larger means–ends logics. For the case of terrorist groups, for example, Ted Robert Gurr has concluded that violence is a conscious choice made by groups in conflict (Reference Gurr and RichardsonGurr 2006). To illustrate this point, on more than one occasion, substitution behaviour by groups using terrorism has been witnessed, which will be further elaborated in Chapter 6 (Reference Sandler, Tschirhart and CauleySandler, Tschirhart, and Cauley 1983; Reference Cauley and ImCauley and Im 1988; Reference Enders and SilkeEnders 2004; Reference Arce M. and SandlerArce and Sandler 2005). Substitution, or the waterbed effect, points to a shift or refocus of activities. If an attack is made more difficult in one area or with one particular means, a shift can be observed to other targets or instruments. This can be interpreted as a sign of collective rationality. Some have described these activities as a ‘collectively rational strategic choice’ (Reference Crenshaw and ReichCrenshaw 1990, 9; Reference Kydd and WalterKydd and Walter 2006). Other studies have also hinted at the strategic rationale behind ostensibly a-strategic phenomena, such as suicide terrorism, which fits into a pattern of nationalist campaigns (Reference PapePape 2005) or barbarous warfare, which possesses a measure of strategic logic in poverty-stricken areas such as West Africa (Reference RichardsRichards 1996). Yet others have described rebels as strategic calculators when it comes to alliance behaviour in often highly complex conflicts (Reference ChristiaChristia 2012) or compliant with international law and regulations in warfare (Reference JoJo 2015). Also based on interviews with rebel group leaders and cadres, the evidence points to rational and deliberate policy development (Reference DudouetDudouet 2012, 96). All these studies indicate and demonstrate that rebel groups are political and strategic operators.
A potentially more significant challenge than proving that rebels are strategic and political actors comes from investigations of micro-level conflict. On the individual level, interviews with individual combatants and polls among populations involved in political violence have shown a diverse set of reasons why people engage in violence. Self-preservation, peer pressure, social bonds and self-betterment are often referred to issues in these studies (Reference Peters and RichardsPeters and Richards 1998; Reference ArgoArgo 2009; Reference LadburyLadbury 2009; Reference AlexanderAlexander 2012). Remarkable is that categories are similar for very different conflict locales with different rebel groups espousing different political agendas. Furthermore, some scholars have claimed that ‘people participate in terrorist organizations for the social solidarity, not for their political return’ (Reference AbrahmsAbrahms 2008, 94).Footnote 3 This series of explanations is notable for the absence of politics or ideology as a motivating factor.
These insights are part of a challenge, which social science has tried to grapple with for many decades; explanations on the individual level about engagement in violence are often difficult to translate into explanations about group behaviour and strategic effect, especially in an interactive fashion with the state. While valuable in terms of dissecting the development and logic of war, explanations focusing on individual worth, social meaning and honour possess little all-encompassing explanatory value for either the empirical phenomenon that is the focus of this study or the escalation of rebel violence. When individuals continue to engage in violence out of peer pressure, a quest for self-worth or social solidarity, this does explain primarily individual motivation but says little about the behaviour of the larger group, its leadership and the actual employment of force, let alone explain sudden spikes in the level or spread of violence, that is escalation. Still, we will return to the topic of individual engagement in Chapter 7.
A subsequent question is whether there is indeed a link between group behaviour and strategic effect:
[g]roups may use violence to pursue both organizational and strategic ends, but the link between the two is not well understood. Is the achievement of one necessary for the achievement of the other? Are organizational and strategic goals complementary or contradictory, and under what conditions?
There are a few studies that have been successful in explaining the interaction between individual disposition towards continued engagement in political violence in conjunction with the interests and agenda of the rebel group leadership and specifically strategic output (Reference Della PortaDella Porta 1995b; Reference McCormickMcCormick 2003). One suggestion has been that organisational considerations take centre stage when rivalries exist among the social movement family from which the rebel derives (Reference KrauseKrause 2014). When there is a strong and hegemonic organisation representing the specific agenda, strategic considerations have free reign. When there is rivalry, organisational survival and infighting play out, which preclude a concentration of generating strategic effect. While insightful, it does not solve the puzzle of the generation of strategic effect as a result of diverse individual participation in rebel groups.
Other avenues to link the distinct sets of explanations are as follows: firstly, the war systems ideas, which stress economic self-betterment as an important force overlapping with the interest of a continuing existence of a war economy (Reference Reno, Berdal and MaloneReno 2000; Reference WeinsteinWeinstein 2007; Reference KeenKeen 2012). Secondly, socialisation and rebel culture could also act as a transmission mechanism among the leadership, group and individual levels (Reference WoodWood 2003; Reference SagemanSageman 2004; Reference MittonMitton 2012; Reference BeevorBeevor 2017). Rebel culture, for example, in the case of Sierra Leone, benefited the strategic necessities of the rebel leadership, in this case the Rebel United Front (RUF), which also links the two sets of explanations (Reference MittonMitton 2012).Footnote 4
This investigation will not solve the fundamental research problem. We deem it justifiable to continue based on the means–ends presumption, awaiting further investigation. In cases where ideas about group dynamics do offer causal explanations for escalation, this will be addressed separately (see again Chapter 7 in particular).
While this study takes as a starting point the nature of war as essentially political, following Clausewitzean thought, the black box of politics can and should be pried open more. Some scholars have argued that politics can only be time and place specific; therefore, any endeavour to investigate the specifics of politics will end up demonstrating the limits of the social science investigator’s toolkit (Reference SmithSmith 2012). The question why violence is used to further political goals might indeed demonstrate these limitations and be highly context dependent and indeed unique in every case. However, the question how politics, in its different shapes and guises, is instrumental in the process of violence and where the use of violence runs into limitations is scientifically productive and very pressing.
Previous explanations of political violence have importantly focused on the causes and the termination of conflict. We know, through elaborate studies, that both grievance and opportunity can lead to conflict (Reference Berdal and MaloneBerdal and Malone 2000; Reference Fearon and LaitinFearon and Laitin 2003; Reference Collier and HoefflerCollier and Hoeffler 2004; Reference Collier and SambanisCollier and Sambanis 2005). The literature about grievance places emphasis on psychological processes among a disaffected or disadvantaged population. Deprivation, injustice or resentment plays a role as conflict-generating factors (Reference GurrGurr 1970). Despite a long pedigree in conflict studies, there is little solid evidence that grievance in and of itself causes armed conflict. How do groups organise to express grievances violently, and at what point in time does mobilisation occur? This literature emphasises cost–benefit calculations within a particular opportunity structure (Reference Fearon and LaitinFearon and Laitin 2003; Reference WeinsteinWeinstein 2007). The political opportunity structure sees the personal calculation of expected benefits and rewards with an organisational capacity and political opportunity as the combination of factors that holds explanatory value for non-state actor violence. While there is evidence to support this point of view, the question why some groups with the same opportunities take up arms and others do not remains an important puzzle.
These grievance theories face several obstacles at present; significant for this study are the following issues: firstly, the existing dichotomy of grievance versus opportunity cannot account for overlapping and/or changeable motivations; secondly, the explanations do not possess the ability to explain the dynamic nature and strategic interaction between the state and the non-state actor, which is the essence of conflict; thirdly, these explanations focus on the causes of non-state actor violence, they are often based on large-n studies, and it is generally expected that by extrapolation these ideas will be illuminating to understand the continuation of conflict. This can and should be questioned. This study is specifically aimed at moving away from the preoccupation in the study of war from its causes and labelling and explaining motivation of warring parties with often ex ante claims. Rather, it asks attention for the dynamics after the outbreak of violence and the highly changeable character of conflict during its course.
These existing ideas explaining organised political violence, such as opportunity and grievance explanations, rely heavily on bargaining theory and utilitarian explanations, which inform the core of this field of research (Reference PowellPowell 2002; Reference ReiterReiter 2003; see also discussion in: Reference AbrahmsAbrahms 2011). Bargaining theory originates from scholarship into economics. In essence there is a bargaining process going on in war where one actor desires something and another fails to deliver it. It is presumed that both are subject to rational calculations, weighing costs and benefits of particular behaviour in the course of their exchanges. The necessary prerequisites in this interaction process are political will and capability. By exercising power and reducing the courses of action open to an opponent, compliant behaviour can be elicited. We know from previous studies that force is an instrument that is often and quite easily referred to in international affairs (Reference HironakaHironaka 2005; Reference Regan and AydinRegan and Aydin 2006). When the expected utility of resistance of the actor subject to coercion exceeds the worth of the prize, the latter actor will start to show compliance with the demands of the coercer.
Framing this political will via ideological categories is a very common approach. A large number of academic studies place centre stage the ideological agendas of rebels. Ideologies, such as nationalism, anarchism, Marxism, political Islam and ethnicity, have individually, or in combination, been used as labels for many conflicts throughout modern history. Different opportunity structures but also issues of contention can produce different conflict dynamics. David Rapoport, for example, has identified four waves of modern terrorism since the late nineteenth century, which have been distinguished according to their ideological content (Reference RapoportRapoport 2001b). The first wave from 1880 until roughly 1920 was concerned with social reform movements, which used violent means, and anarchism, which had adopted the ‘propaganda by deed’ philosophy in the 1880s. The second wave from 1920 until 1960 was focused on national self-determination and nationalism, exemplified by violent decolonisation struggles in Asia and Africa. From 1960 until 1979, Rapoport identified a left-wing revolutionary wave. The fourth and last wave started in 1979 after the Iranian revolution and lasts till today. It is defined by political ideologies based on religion.
The approach of placing centre stage ideological aims or goals is a prevalent one in the investigation of rebel activities. Ideology can be understood as
a more or less systematic set of ideas that includes the identification of a referent group (a class, ethnic, or other social group), an enunciation of the grievances or challenges that the group confronts, the identification of objectives on behalf of that group (political change – or defense against its threat), and a (perhaps vaguely defined) program of action.
The focus on ideology has given rise to a large body of scholarship. For example, political Islam, as a defining feature of many contemporary conflicts, has been offered as an explanation for conflicts as diverse as Afghanistan, Algeria, the Central African Republic, Iraq, Nigeria, Somalia, Sudan and Yemen. Arguments in the debate have focused on the local/national and global level interconnecting in these conflicts (Reference CliffordClifford 2005; Reference KilcullenKilcullen 2009; Reference MackinlayMackinlay 2009) and the idea of political Islam as a franchise with a label that can easily be adopted to attract support, fighters and capital.
Ethnicity, another motivational label, has also gained a large body of scholarship (Reference BrownBrown 1993; Reference SambanisSambanis 2001; Reference DuyvesteynDuyvesteyn 2005). Ethnicity can be seen as an ideology and can define the agenda of rebels (Reference KaufmannKaufmann 1996a, Reference Kaufmann1996b). For African civil wars, for example, ‘[t]he ethnic group is the natural component of a rebellion against the state, as the many links that exist among its members provide an efficient way of overcoming the free-rider problems involved in mobilizing a rebellion or insurgency’ (Reference AzamAzam 2001; Reference PosenPosen 1993; Reference KaufmannKaufmann 1996b, 430). Within certain political structures, ethnicity can play a constitutive role and define the boundaries of networks.
Ideology can motivate individuals and collectives; act as an instrument of socialisation; provide an organisational template and a doctrine for strategy and tactics, for weapon and target selection; provide justification and legitimation of the use of violence, as well as inform the timing of violence. These aspects have mostly come to light in the debate about jihadist violence and the attempts to explain it. The jihadist political agenda informs a specific way of fighting. The types of weapons that are selected are linked to the acceptance of (in)discriminate violence in the rebel ideology. Jihadist attacks have, for example, been linked to high and increasing casualty rates per attack compared to other ideologically motivated fighters (Reference HoffmanHoffman 1998; Reference Asal and RethemeyerAsal and Rethemeyer 2008; Reference Jäckle and BaumannJäckle and Baumann 2017). Suicide terrorism has been both prevalent among nationalist groups (Sri Lanka) and adopted by jihadist groups (Reference PapePape 2005). In the case of Chechnya, the occurrence of indiscriminate violence has been explained based on the lack of reliance on local support bases (Reference Toft and ZhukovToft and Zhukov 2015). Not only was their agenda defined by jihadist ideology, their way of organising themselves as well as the selection and justification of targets, that is, (in)discriminate killings, were all connected (Reference MoghadamMoghadam 2008; Reference Sanín and WoodSanín and Wood 2014). Moreover, ideology also informed the organisational fissures and factionalisation within rebel movements (Reference HafezHafez 2020). All these factors related to the ideological focus create a certain measure of path dependency.
However, ideological labels are not without problems (Reference LyallLyall 2010). While very prevalent, they run the risk of obfuscating many important conflict dynamics. Firstly, ideological issues that are claimed to inform and drive conflict are highly changeable. Shifts in ideological content can occur. For example, both the IRA and ETA not only espoused nationalism but also adopted social revolution and Marxism in the course of their existence (Reference ZirakzadehZirakzadeh 2002; Reference DuyvesteynDuyvesteyn 2004; Reference Parker and SitterParker and Sitter 2015). Also, ideological interpretations can change over time: ‘it is difficult to use ideology as the critical variable that explains the resort to or continuation of terrorism. The group, as selector and interpreter of ideology, is central’ (Reference CrenshawCrenshaw 1985, 471). Ideology is therefore flexible rather than static.
Secondly, the existence of political entrepreneurs who capitalise on the political opportunity structure and opportunistically adopt ideological cloaks as rallying mechanism put into question the motivational claims of conflict (Reference TillyTilly 2003). Political entrepreneurs are individuals who try to affect the course of politics (Reference Schneider and TeskeSchneider and Teske 1992). They attempt to engage with the issues in the political opportunity structure and press a specific agenda. In the conflict in Afghanistan, for example, political entrepreneurs have played a central role (Reference ThruelsenThruelsen 2010). The conflict in Liberia in the 1990s can also be seen through the lens of political entrepreneurship (Reference DuyvesteynDuyvesteyn 2005) as well as the many conflicts in the Caucasus after the demise of the Soviet Union (Reference ZurcherZurcher 2007, 6). Political entrepreneurs will attempt to overcome the so-called free rider problem, trying to actively enlist the engagement of those hoping to benefit from the positive effects of collective action without wanting to make the effort. Political entrepreneurs dealing with free riders can offer selective incentives to these potential supporters. Rebel groups, in order to function, need to provide selective incentives to group members. Selective incentives can include status incentives but also material incentives. Some scholars have pointed out that it is ‘the on-going provision of such collective and selective goods, not ideological conversion in the abstract that has played the principal role in solidifying social support for guerrilla armies’ (Reference Goodwin and SkocpolGoodwin and Skocpol 1989, 494).
Third, apart from changeability of ideological content, and the role of opportunistic political entrepreneurs, rebels have a tendency to pragmatically incorporate local grievances and discontent. They have been known to incorporate local grievances to propel their struggle forwards (Reference KrigerKriger 1992; Reference KalyvasKalyvas 2003, Reference Kalyvas, Strachan and Sibylle2006; Reference DuyvesteynDuyvesteyn 2005; Reference WeinsteinWeinstein 2007; Reference KilcullenKilcullen 2009). Competition, control and local considerations offer explanatory value when it comes to witnessing more or less violence in insurgencies. Among others, Charles Tilly has suggested that there are repertoires of contention and violence and pre-existing histories of violent exchange that can shape and define a struggle (Reference TillyTilly 2003). Culture has also been found to influence rebel violence; collectivist, as opposed to individualist, cultures have been alleged to stand out for attacks on out-groups, foreigners and indiscriminate violence (Reference Weinberg and EubankWeinberg and Eubank 1994). This interesting debate about the precise role of ideology notwithstanding, these findings overall confirm the rebel group as a political actor, pursuing explicit political ends.
Rebel Means and Methods
In order to fight, a rebel group has several approaches available to conduct its armed engagements, among which terrorism and insurgency predominate (Reference Kalyvas, Strachan and SibylleKalyvas 2011). There is a limited number of cases where non-state actors have used direct or conventional strategies to conduct a war, such as the Spanish civil war, the Sri Lankan and Chechen wars (Reference Kalyvas, Strachan and SibylleKalyvas 2011). In recent years, conventional confrontations were also visible in the conflicts in Syria, South Sudan and Ukraine.
A strategic approach can be investigated based on the presence or absence of the levels of strategy (Reference LuttwakLuttwak 2001; Reference FreedmanFreedman 2013). Several levels can be distinguished: tactics or tactical attacks serve as a conduit to provoke the opponent, for rebel groups usually the state. The tactical use of force relates to the operational plan of drawing out the opponent or provoking him or her into overreacting. The provocation and confrontation, on the operational level, comes together with a form of coercion, with recruitment and propaganda aims. As with all strategic approaches, the levels work both up and down: the tactical level attacks bring closer the operational level coercion in order to effect political change. Conversely, the political program infuses the operational level coercion and intimidation campaigns and dictates the shape of the tactical level attacks.
Before discussing the tenability of this strategic framework to understand rebel groups, it should be noted, first, that there is large-scale conceptual and definitional confusion in the debate about rebel approaches. In particular, the use of the terms terrorism and insurgency often create misunderstandings. At some point in the past, in fact, the terms were used interchangeably denoting the same phenomenon (Reference KilcullenKilcullen 2005, 604). This problem still persists in some treatments in which categories are conflated, resulting in a conceptual muddle (e.g. Reference AbrahmsAbrahms 2008; Della Reference Della PortaPorta 2018; see for a critical discussion: Reference Chenoweth, Miller, McClellan, Frisch, Staniland and AbrahmsChenoweth et al. 2009). Moreover, historically there has also been a tendency to emphasise insurgency at the cost of terrorism, as the less significant social phenomenon, in particular until the 1970s (Reference BeckettBeckett 2005, 24). Finally, guerrilla is also an important term often used interchangeably with insurgency, and conceptualised as a tactic of warfare.
Analytically, terrorism and insurgency are not the same and need to be separated (Reference Fumerton, Duyvesteyn, Holmqvist-Jonsater and CokerFumerton and Duyvesteyn 2009; Reference CroninCronin 2015, 87). While modern terrorism has roots in the nineteenth century and important predecessors in the shape of regicide and tyrannicide (Reference RapoportRapoport 2001a), insurgency has more recent foundations. Originally, small-scale ambushes and hit-and-run attacks were called guerrilla or small war, deriving from the Spanish resistance against Napoleonic occupation of the Iberian peninsula. These activities were warfare or tactical activities and part of the larger conventional war effort, as also conceptualised by Clausewitz (Reference Daase, Strachan and Herberg-RotheDaase 2007, 182; Reference HeuserHeuser 2011).Footnote 5
The linking of the guerrilla tactics with political ideology and a strategic agenda at the time of Mao’s Great March gave rise to the idea of revolutionary war or insurgency (Rich and Reference Duyvesteyn, Malkki, Jackson and SinclairDuyvesteyn 2012). Revolutionary war, or insurgency, was first conceptualised by Mao Zedong as an independent strategic approach (Reference Tse-Tung and GuevaraTse-Tung and Guevara 1961; Rich and Reference Duyvesteyn, Malkki, Jackson and SinclairDuyvesteyn 2012). Mao identified three phases of revolutionary war: the first phase consisted of hit-and-run attacks to liberate territory in the countryside and contest government control. This is in essence defensive. The second phase aimed at extending the liberated areas into liberated zones. The third phase consisted of a conventional confrontation against weakened government forces. This is in essence offensive.
Important arguments have been raised against the claim that rebels use carefully considered means and methods to realise their political aims (Reference EpprightEppright 1997; Reference Hoyt, Cronin and LudesHoyt 2004; Reference AbrahmsAbrahms 2006; Reference SimpsonSimpson 2012). Rebel activities are argued to be limited to tactical skirmishes with a preference for the spectacular, and devoid of any strategic logic giving witness to a process of aligning ends, ways and means (Reference Crelinsten, Wilkinson and StewartCrelinsten 1987, Reference Crelinsten2002; Reference Post and ReichPost 1990).
However, others have put forward that rebel violence tends to be mainly tactical (Reference Ucko, Rich and DuyvesteynUcko 2012). When rebels are lucky, their activities can translate directly into political effect. A useful distinction can be made between intended and actual strategic effect (Reference RobertsRoberts 2005). The effect might be accidental. Some speak of collapsing levels of strategy (Reference EpprightEppright 1997). An example of tactical attacks possessing political strategic effect are of course the 9/11 attacks. Even though intended strategic effect might be hard to realise when applying rebel strategies, it cannot be denied that actual effect is present.
As for operational effect, it is true that insurgents often find it hard to generate operational output with the limited means they have available (Reference SimpsonSimpson 2012). According to some analyses, the global jihad possesses a clear operational level: ‘the essence of jihadist “operational art” is the ability to aggregate numerous tactical actions, dispersed across time and space, to achieve an overall strategic effect’ (Reference KilcullenKilcullen 2005, 609). Even highly fragmented insurgencies have the potential to produce strategic effect (Reference JardineJardine 2012).
This study departs from the more mainstream view that terrorism and insurgency cannot be seen as strategies. While some experts have claimed that insurgency is not a strategy, and has never been one (Reference HammesHammes 2012), others have argued: ‘[i]t may be that the single most important similarity between terrorism and traditional warfare is in its inherently strategic nature’ (Reference ArquillaArquilla 2007, 377; Reference ThompsonThompson 2014). Indeed, terrorism and insurgency constitute two strategies that can be employed by non-state groups –but also states for that matter – guided by a variety of political agendas. An advantage of such an instrumental approach is that it shifts attention away from the actor and its motivation, towards the act of force itself, avoiding among others the terrorist versus freedom fighter discussion. Terrorism and insurgency are acts of coercion to change the behaviour of an opponent.
Terrorism can be conceptualised as a strategy with the ambition of realising political aims with the use of violent means. The strategies of terrorism and insurgency can be distinguished in three respects: their political-strategic logic, organisational and relational characteristics (Reference Fumerton, Duyvesteyn, Holmqvist-Jonsater and CokerFumerton and Duyvesteyn 2009; Cf. Reference KhalilKhalil 2013). As a strategy, insurgency aims for political power. All insurgent actions are geared towards the ultimate take-over of power, whereas terrorism’s goal is to merely provoke political change, which has to materialise through other means, either through a full-blown insurgency or through the de-legitimising acts of the state itself. In organisational terms, insurgents systematically organise and prepare as an alternative centre of power. Terrorists, in general, make few, if any, preparations in this direction. The relationship with the population is therefore also different. The terrorists remain highly secretive in order to increase the effectiveness of their strategy. Insurgents aim to mobilise and organise the population in order to establish an alternative form of social order. We should be aware that a strategy of terrorism can develop into an insurgency strategy. Furthermore, insurgency struggles can turn to the strategy of terrorism to further their cause (Reference DixonDixon 2020). On top of that, the insurgency strategy also tends to use terrorism on a tactical level (Reference Arreguín-ToftArreguín-Toft 2012).
As for the levels of strategy and their application to insurgency, the strategy relies on the tactical level similarly, on hit-and-run attacks using a variety of weaponry and material against unwitting but symbolic targets. On the operational level, insurgents also aim to provoke and coerce the opponent into changing its course of action. Furthermore, the highly visible acts are also geared towards recruitment and building a supportive environment. The operational plan includes the establishment of an organisation that can act as a shadow state or alternative political order. This base organisation needs to propel the struggle towards strategic political success. This comes in the shape of political control over population and territory as ultimate victory of an insurgent strategy.
As already noted, recent departures in the debate have questioned the central role of the state in insurgencies (Reference KilcullenKilcullen 2006, 112) and the role of territory (Reference SalehyanSalehyan 2007; Reference Smith and JonesSmith and Jones 2015; Reference ToftToft 2014). The state is argued to have lost its relevance because of a lack of pre-occupation of rebels with obtaining state power. Rather, rebels have been claimed to now espouse a global perspective fighting against an unjust global order. These post-territorial arguments are closely linked to researchers posing the fundamental question: “To what extent is territory still a factor in a conflict over values?” We contend that even warlords and insurgent leaders prefer to exert power through less costly means such as legitimacy and authority, which can come in the shape of some form of social contract, rather than through costly means, such as continued coercion and repression, which require substantial and consistent enforcement power. Therefore, the role of the population and territory remain important. We saw, for example, how the Islamic State in Iraq and Syria clearly identified a need to focus both on population and territorial control to build a Caliphate (Reference WhitesideWhiteside 2016).
The choice for a terrorist or insurgent strategy is not set in stone. Rebels can and do change between strategies. When circumstances allow and means become available, rebels have been known to shift from terrorism to insurgency strategies. Insurgents, when repression is fierce and life in the open becomes dangerous, have in the past reverted to terrorism. Furthermore, within an insurgency struggle, terrorism can be used as a complimentary tactic to achieve desired ends. Both terrorist and insurgency strategies exhibit the inherent feature of strategy, raising the price of further resistance by the opponent via the crossing of thresholds, that is escalating the confrontation.
Summarising, the rebel group is a viable analytical category representing an important empirical phenomenon that begs for further explanation. The rebel group is a political actor capable of using means and methods to further its political goals. Their armed conflicts tend to display recurrent rising and declining levels of violence. How can we explain these?
Escalation
Escalation is a surprisingly under-investigated concept in social science.Footnote 6 At the same time, the word is heavily used in the study of international affairs, often with a very imprecise meaning. The term originally emerged in the field of strategic studies in the 1960s in the context of the superpower confrontation and the risk of nuclear escalation (Reference SmokeSmoke 1977, 35; Reference SchellingSchelling 2008; Reference KahnKahn 2012). The use of the term has been strongly linked to the Cold War, state actors and international confrontations (Reference CarlsonCarlson 1995). Escalation has been defined, in the Cold War context, as ‘an increase in the level of conflict’ (Reference KahnKahn 2012, 3). The basic idea of escalation is that an actor deliberately steps up the level of conflict or seeks its spillover to demonstrate resolve. By using the credible threat of violence, it aims to bring his or her desired outcome closer.Footnote 7
In contrast to the wide-ranging debate about rebels, detailed earlier, the concept of escalation is relatively uncontested. Escalation is, according to Herman Kahn in his seminal work on the subject, a ‘competition in risk taking or at least resolve’ and stops when ‘the fear that the other side may react, indeed overreact’ manifests itself. This ‘is most likely to deter escalation, and not the undesirability or costs of the escalation itself’ (Reference KahnKahn 2012, 3). Escalation has furthermore, been conceptualised as
an action that crosses a saliency which defines the current limits of a war, and that occurs in a context where the actor cannot know the full consequences of his action, including particularly how his action and the opponent’s potential reaction(s) may interact to generate a situation likely to induce new actions that will cross still more saliencies.
The essence of escalation, going back to the pioneering work of Thomas Schelling, is that it signals to a defender that a challenger is both capable and willing to inflict harm in case the defender does not comply with the challenger’s wishes (Reference SchellingSchelling 2008). In a process of coercive bargaining, there are deliberate steps that an actor can take to demonstrate resolve. This challenger has information that it shields from the defender about its measure of resolve to be victorious in the dispute. The measure of resolve reveals to what extent it is able and willing to pay the price in blood and treasure. These ‘sunk costs’ are fundamental in the process of escalation, as conceptualised by Schelling. The coercive bargaining process, moving from a crisis towards conflict, escalation steps can be used. It is a process of ‘brinkmanship’ with an uncertain outcome, and according to Schelling, importantly, ‘leaves something to chance’ (Reference SchellingSchelling 1980).
These approaches to escalation are characterised by two main features. Firstly, escalation is conceptualised as largely linear and seen as a step-by-step process leading from peace to war (Reference HolstiHolsti 1972; Reference KahnKahn 2012). A very common way of conceptualising escalation is by using the analogy of a ladder. Herman Kahn’s escalation ladder, developed in the 1960s, had forty-four rungs. The upper half of the ladder involved nuclear weapons, leading to a highest rung of ‘insensate war’ (Reference KahnKahn 2012, 38).
Secondly, not only is the linear approach dominant, escalation is seen as a rational phenomenon with state actors making choices to either escalate or de-escalate. Both courses of action are based on cost–benefit calculations. The expectation is that the opponent acts based on a same or similar calculus. The ideal is to possess escalation dominance, a situation in which any action of the opponent can be matched or surpassed, making the exercise of further manoeuvring by the opponent futile. A second-best option would be to create escalation asymmetry, a situation in which the opponent becomes unable or unwilling to follow suit. Thresholds or rungs on the ladder focus preferably on weaknesses of the opponent that are not easily remedied or patched up.
The criticism that has been levelled against the Cold War scholarship on escalation has not only challenged the moral foundations of the ladder approach and the theorising of the ‘unthinkable’. Other aspects, such as the overriding ideas of rationality and linearity of escalation, also received criticism. In a reflection on the Cuban missile crisis of 1962, Albert and Roberta Wohlstetter, for example, raised pressing questions regarding the supposed linearity of escalation:
The aspect of ‘escalators’ that inspired its use in this connection, we suspect, is the fact that moving stairways carry a passenger on automatically without any effort of his will. However, as we have suggested there are down-escalators as well as up-escalators, and there are landings between escalators where one can decide to get off or to get on, to go up or down, or to stay there; or take the stairs. Just where automaticity or irreversibility takes over is an uncertain but vital matter, and that is one of the reasons a decision maker may want to take a breath at a landing to consider next steps. It is apparent from President Kennedy’s own descriptions of the Cuban crisis … that he gave enormous value to the cautious weighing of alternatives made possible by the interval of almost a week; to the five or six days mentioned for hammering out the first decision. And the decision made was precisely one that left open a variety of choices.
They conclude that the variety of choice in case of Cuba was far larger than initially conceptualised. More recently, scholars have taken issue with the deliberate nature of escalation (Reference Morgan, Mueller, Medeiros, Pollpeter and CliffMorgan et al. 2008). This present study will clearly echo these earlier reflections on escalation.
Few of these classical ideas about escalation have found their way into discussions about rebel violence. Still, a need has been recognised for more research into escalation in the case of civil unrest (Reference TarrowTarrow 2007, 595; see also Reference Collier and SambanisCollier and Sambanis 2005, Volume I 318–319, Volume II 314–315; Reference Davenport, Armstrong and LichbachDavenport, Armstrong, and Lichbach 2008, 22–25). These calls have stayed very close to the Cold War conceptualisation of escalation as emphasising the development towards war in crisis situations, rather than the dynamics of violence after the outbreak (see also Della Reference Della PortaPorta 2013). For the period of the Cold War, this was of course understandable in light of the theoretical exercise to think through a confrontation involving nuclear weapons. In the post–Cold War world, this emphasis is more surprising because of prevailing conflict patterns.
Some authors have used the concept of escalation in the context of rebel violence without paying much attention to the concept itself (Reference Daase, Strachan and Herberg-RotheDaase 2007, 194). Other experts, omitting the use of the word ‘escalation’ altogether, have looked at its properties in practice. One example focusing on enduring conflicts notes that the longer a confrontation lasts, the more protracted it tends to become; ‘protracted conflict keeps creating derivative issues, factionalizes opponents, destroys trust, invites outside intervention, and brings to power hard-liners and extremists’ (Reference OberschallOberschall 1993, 104). Duration is linked here to the rising and falling levels of violence.
The contemporary prevalence of rebel violence in the international system makes it an important and necessary subject of investigation. What happens when state and non-state actors engage in violent confrontations? We know from previous research that non-state actor violence tends to show many recurrences of escalation and de-escalation. These confrontations usually end in an indeterminate fashion, rather than in a clear-cut victory, defeat or ceasefire (Reference KreutzKreutz 2010). Furthermore, there is an important ‘spectrum between phases of escalation and de-escalation of violence … [which] accounts for the many situations of “neither peace nor war,” which nowadays constitute a specific form of social order in many areas and regions’ (Reference Bakonyi and Bliesemann de GuevaraBakonyi and Bliesemann de Guevara 2009, 407). There is much room for furthering the explanations of the occurrence and re-occurrence of violence between a state and non-state actor since a systematic investigation or ‘empirically based explanation of the escalation and de-escalation of political conflict’ to date is lacking (Della Reference Della PortaPorta 1995b , xvi; Reference OberschallOberschall 1993, 104). Furthermore, the dynamic interaction between rebels and the state deserves further investigation; the role of the state, rather than the rebel, has received scant attention in recent studies into the logic of violence (Reference PierskallaPierskalla 2010). This is the challenge the present study takes up.
Before we proceed, it needs to be noted that escalation should not be interpreted a priori as a phenomenon that is always preferably avoided. Lewis Coser argued in the 1950s that ‘[c]ommitment to the view that social conflict is necessarily destructive of the relationship within which it occurs leads … to highly deficient interpretations’ (Reference CoserCoser 1956, 8). Conflict can act ‘[a]s a stimulus for the creation and modification of norms, conflict makes the readjustment of relationships to changed conditions possible’ (Reference CoserCoser 1956, 128). Furthermore,
[i]n the short term, a polarization in the political spectrum follows; in the long term, new forms of collective action become part of the accepted repertoires. This means that, from a historical perspective, social movements do influence even the more stable institutions and deep-rooted political cultures. These changes are, however, not produced by the social movements alone, but by their interactions with other actors both allies and opponents.
Radicals, conflicts and escalations perform functions in shaping political power and social relations; ‘one of the creative functions of conflict resides in its ability to arouse motivation to solve a problem that might otherwise go unattended’ (Reference DeutschDeutsch 1973, 360); moreover, ‘the creation of essentially disciplined forces from bands of people who are, or act like, criminals and thugs has been at the center of much state building’ (Reference MuellerMueller 2004, 23).
This caveat should not be read as a categorical endorsement of rebel agendas. However, it is important to stress that recent research has put forward the unrecognised positive role of rebel groups as successful in democratic transformation and economic development (Reference ToftToft 2010). Furthermore, the capacity of violent groups to provide rebel government (Reference MampillyMampilly 2011; Reference Arjona, Kasfir and MampillyArjona, Kasfir and Mampilly 2015) and legitimacy (Reference DuyvesteynDuyvesteyn 2017) has been notable. There is a clear and urgent need to further these understandings.
This study will aim to explore in more detail the rebel collective utilitarian model of strategy in conjunction with group and individual level explanations. It will, by looking at the available material, try to think through some of the processes that can be witnessed. There is substantial evidence that the rebel collective utilitarian model of strategy is relevant, exists and operates, and there are indications that this can be linked to escalation and de-escalation processes (Reference DuyvesteynDuyvesteyn 2005). The focus will be on how the process of escalation occurs. The rest of this study will proceed as follows. In Chapter 2, the concepts of escalation and de-escalation will be discussed based on the existing literature on the subject. The aim of the chapter is to dissect the existing approaches and to develop a methodology for assessing escalation and de-escalation. Chapter 3 and subsequent chapters will discuss the more detailed propositions relating to processes of escalation and de-escalation.
