Whether in self-defence or as acts of aggression, rebels kill, injure and sometimes terrorise opponents to achieve more or less clearly defined political objectives. The idea that unarmed civilians can resist rebels and restrain their use of violence seems counterintuitive, if not naïve. One could think that rebels do not take kindly to criticism and use their means of repression to quell any form of resistance. Yet nonviolent political campaigns have proven capable of bringing down repressive states. Some scholars even suggest that nonviolent resistance is more likely to trigger political change than violent revolutions. Drawing on a statistical analysis of 323 nonviolent and violent campaigns waged between 1900 and 2006, Maria Stephan and Erica Chenoweth found that nonviolent resistance was more successful on average than violent rebellion and insurgency (53 per cent versus 26 per cent).Footnote 1
There is no immediately apparent reason to assume that nonviolence is less effective against rebel groups than against formally recognised states. A growing number of scholars have, in fact, begun to invoke the concept of resistance to examine the dynamics of unarmed mobilisation under rebel rule.Footnote 2 Labelled the ‘new frontier’ in resistance studies,Footnote 3 nonviolent action against non-state armed groups is an emerging research field that has attracted growing scholarly attention over the past decade. Existing research has helped to enhance our understanding on a number of key questions: why and when do civilians mount specific forms of resistance (violent vs nonviolent; individual vs collective) or place different types of demands on their rebel rulers (moderate vs maximalist)? To date, however, research on the effectiveness of civilian resistance remains in its infancy. We know little as to whether and when nonviolent methods of resistance result in rebels making political concessions or restraining their use of violence against nonmilitary targets. Most existing works on the effectiveness of nonviolent resistance against rebels have either been descriptive or, as will be elaborated below, have focused on areas where conditions were conducive to success.Footnote 4
This article addresses this gap by asking what factors influence both success and failure of civilian resistance against rebel groups, including under adverse conditions. Drawing on the resistance literature, as well as on classical works on mass protest and revolution, I identify three main factors that are likely to bear on a rebel group’s response to nonviolent action: the nature of civilian demands, the power of civilian resisters, and the rebels’ own power. These factors, I argue, help to explain why and when armed groups choose accommodation over repression or inaction (i.e. ignoring). Put simply, weak civilians placing maximalist demands on strong rebels face a high risk of repression, whereas strong civilians backed by powerful local institutions pursuing more moderate objectives have a greater chance of wringing concessions from their rebel rulers.
To test these propositions, I analyse the dynamics of civilian resistance against the Fuerzas Armadas Revolucionarias de Colombia (Revolutionary Armed Forces of Colombia, FARC) in Colombia’s Caquetá region. This case selection is premised on the following logic: Colombia is by far the most researched context for contentious civilian–rebel interaction.Footnote 5 This might be explained by the fact that civilians have been particularly ingenious and, perhaps, desperate to stand up to the country’s many non-state armed groups that have controlled large swathes of land for more than half a century. As I will explain below, existing research on civilian resistance in Colombia is focused on areas where rebels vied for control with one or several armed groups. We know comparably little about the dynamics of civilian resistance against rebels that succeeded in imposing themselves as the sole contender of the state. For much of the second half of the twentieth century, Caquetá constituted a paradigmatic example of a rebel stronghold where the FARC successfully built a counter-state. From the early 2000s, however, the Colombian army progressively pushed the FARC out of provincial towns. The combination of largely unrivalled rebel control in rural Caquetá with declining power in more central areas provides a ‘hard case’ for civilian resistance to succeed. Studying the dynamics of civilian resistance under unfavourable conditions promises to complement existing research, yield new insights and, I hope, provide a more nuanced understanding of what civilian resisters can realistically achieve when challenging well-organised, cohesive armed groups in their geographic strongholds.
The discussion will proceed as follows. After defining the ambiguous notion of civilian resistance, I continue by explaining the theoretical foundations for the claimed link between rebel responses to civilian resistance and the aforementioned three factors – the nature of civilian demands, civilian power and rebel power. I then explain how different factor configurations have played out empirically in Colombia, using secondary evidence from existing research. The remainder of the article analyses empirical findings made in the Caquetá, including civilian participation in state-led elections, which is a form of civilian resistance that featured prominently in interviews but has not been systematically analysed in the literature.
The empirical sections draw on three complementary sources: the first and primary source is interviews conducted with 76 civilians in Caquetá over eight weeks between May and June 2017. Operating on the assumption that the dynamics of civilian resistance against the FARC vary with different degrees of military control and political influence, I selected interview sites based on a territorial logic. Interviews took place in four settings: areas of historic FARC influence (the communist enclave of El Pato; see Figure 1), rural areas penetrated by the group during the 1980s (middle and lower Caguán), rural areas with limited FARC presence (southern Caquetá, along the main road), and urban areas dominated by the government (cabeceras municipales – municipal centres – and villages situated along government-controlled road axes connecting towns). Additional interviews with guerrilla fighters and commanders during a two-week stay at a FARC camp in La Montañita in June 2017, ten days before the group handed over its arms to the United Nations, provided a second source of information. I was accommodated in a tent together with FARC militia members, many of whom operated clandestinely in areas with a strong government presence. This was an intense time of deep immersion, which allowed me to allude to cases of civilian resistance reported in interviews with civilians, cross-check information and solicit the FARC’s perspective on specific incidents. The third source of information came from archival research in the capital, Bogota, which yielded quantitative data on specific cases of resistance that had received local and national media coverage, particularly civilians defying FARC-declared electoral boycotts (see Supplementary Materials for additional detail).
Figure 1. Caquetá Department
Risk of Reprisal as a Defining Element of Civilian Resistance
Civilians who refuse to submit to rebel rulers have three basic options. First, they can flee rebel-held areas and relocate to places beyond the rebels’ reach. Second, they can stay and resist rebels without using violence. Third, they can fight the rebels, in which case they cease to qualify as non-combatants. The scope of this article is limited to the second option. Unlike civil resistance, which is generally associated with nonviolent action against states, I use the term civilian resistance to refer to different forms of nonviolent action by non-combatant populations that challenge the authority of rebel groups at the risk of reprisal.Footnote 6
The notion of risk – understood as anticipated exposure to reprisal – provides a safeguard against an inflationary use of the term ‘resistance’. It helps us differentiate civilian resistance from civilian non-cooperation during armed conflict, which includes a much broader panoply of possible actions. Non-cooperation becomes an act of nonviolent resistance only if it violates rebel directives at the risk of reprisal. Consider the following example: a rebel group holds a political meeting to rally civilians behind their cause. For as long as participation is voluntary, civilian abstention should be viewed as an act of non-cooperation. Yet, if rebels request participation from all adults, the decision to stay at home becomes an act of resistance, precisely and only because it entails a risk of punishment. Moreover, connecting the term ‘resistance’ with the notion of risk makes it possible to distinguish nonviolent action from political mobilisation at large. As Juan Masullo writes, some armed groups have been open to criticism and have even created channels that allow civilians to voice their demands.Footnote 7 If civilians raise complaints with armed groups without risking repression, they are not committing an act of resistance. In a similar vein, civilian mobilisation against rebels in areas beyond the rebels’ reach should not be considered as a form of nonviolent resistance. Think of anti-rebel street protests in towns firmly under government control. If rebels lack the capacity to punish or credibly threaten protesters, participation in street marches or other forms of protest in government-controlled areas fall under the much broader category of political action.
The proposed definition of civilian resistance differs from the way nonviolent action against states is conceptualised in the literature. Civil resistance scholars conceive of nonviolent action as an extra-institutional form of political struggle. As Jonathan Pinckney writes, ‘Political actions that are nonviolent but fall within the normal bounds of regular politics (e.g. elections, lawsuits, lobbying) are not nonviolent resistance.’Footnote 8 Yet, disrupting ‘regular politics’, to use Pinckney’s words, can be a core objective of rebel groups. Preventing governments from holding elections in rebel strongholds was a common means used by the FARC and other rebels to achieve this.Footnote 9 In such situations, voting for incumbent parties despite a rebel-declared electoral boycott is a powerful, though dangerous, way to resist rebels in a nonviolent manner. What makes voting a form of nonviolent resistance is the risk of a violent backlash from rebels. When looking at nonviolent action against rebels, it thus seems more appropriate to take the risk of reprisal as a defining element of civilian resistance, rather than its institutional or extra-institutional character.
Before moving on with the analysis, it is necessary to acknowledge conceptual and empirical ambiguities. Civilians can not only combine institutional with extra-institutional forms of nonviolent resistance; they can also rely simultaneously or successively on nonviolent and violent tactics.Footnote 10 The rondas campesinas in Peru are a case in point. Created in the 1980s, these vigilante groups played a role in mobilising nonviolent protest against the Shining Path, a leftist guerrilla group. But they also defended their villages from guerrilla intrusions with weapons supplied by the Peruvian army.Footnote 11 When enlisting with local defence forces, civilians thus transgress the conceptual boundary of nonviolent resistance. Another example that defies easy classification is civilians collaborating with government forces by providing intelligence (see next section). Although the line between violent and nonviolent resistance can be blurred, the distinction between these two types of action is nonetheless useful, primarily because they pose different problems to armed groups: those who take up arms against them or collaborate with their military rivals are, by definition, enemies of the rebels. A violent response to their actions and demands is more easily justified from both a military and an ideological perspective. As I will argue below, nonviolent forms of resistance are often mounted by the rebels’ own constituents. A harsh response to nonviolent resisters is thus much more likely to generate a backlash and breed alienation, including within the rebels’ civilian support base, and thereby diminish the rebels’ power.
Contentious Civilian–Rebel Interaction
As Michael Walzer noted in Just and Unjust Wars, a civilian ‘knows the most important military secret; he knows who the guerrillas are. If he doesn’t keep this information to himself, the guerrillas are lost.’Footnote 12 The ability to withhold or disclose information about rebel fighters gives civilians power over armed groups – though not the kind of power invoked by scholars of civil resistance against states. According to Gene Sharp, a leading theorist of nonviolent action, the power of rulers in any political system depends on the consent of the ruled. By withdrawing that consent, the ruled can control and ultimately destroy their masters’ power.Footnote 13 If civilians disclose information on armed groups, they also display agency. In Michael Rubin’s words, ‘Because even well-financed rebel organizations rely on civilian support and population concealment to move freely through contested territory and avoid counterinsurgents, they are not immune to civilians’ exercising of agency.’Footnote 14 Unlike withdrawal of consent, however, civilian betrayal of armed groups poses a much more immediate threat. When civilians provide information on the whereabouts of rebel fighters to government forces, they can indirectly inflict serious, physical harm. Government forces can use their military superiority, particularly their air supremacy, to weaken rebel groups, by killing their leaders with air strikes.
Theoretical Implications
The idea that civilian discontent poses a serious risk to armed groups – particularly where they are outnumbered, outgunned and surrounded by government forces – has intriguing theoretical implications. On the one hand, it suggests that armed groups will be particularly hostile towards civilian resisters and respond with repression. This is not to say they will kill anyone who dares to raise their voice against them. A less violent way of dealing with dissent might be to expel individuals or groups who stand up to them. On the other hand, armed groups also have an incentive to accommodate civilian demands as quickly as possible and thereby remove the cause of civilian discontent, instead of expelling or eliminating resisters.
Beyond these two basic courses of action – that is, repression and accommodation – rebels may sometimes have no choice but to ignore civilian resisters because their organisations are entirely absorbed by fighting military rivals or because they lack the capacity to identify and punish civilian resisters effectively (or both). For weak rebels who are unable to establish a monopoly of violence in areas they claim to control, ignoring civilian demands is a third possible response. Ignoring civilian resisters, however, is a risky course of (in-)action and one chosen due to a lack of alternatives.
Factors Bearing on the Success or Failure of Civilian Resistance
If an armed group chooses to accommodate civilian demands, resistance can be considered a civilian success. Failure occurs if the armed group responds with repression or ignores civilian demands. Three main factors are likely to bear on the choice of one response over another. The first and most obvious is the nature of the demands civilians place on their rebel rulers. It is logical that any political authority – whether a rebel group or a state – will find it easier to accommodate demands that require moderate policy or behavioural adjustments than maximalist demands that challenge its claim to power. In the latter case, as Timothy Wickham-Crowley pointed out, the most likely response is repression: ‘If terror is the response of a government to a decline of – and an open challenge to – its authority in the body politic, then we would expect to find guerrilla terror against the peasantry in similar circumstances.’Footnote 15 The contention that certain types of demands are more likely to elicit a violent counteraction is supported by classical works on mass protest and revolution. William Gamson, for instance, argued that contending groups seeking to replace an incumbent face a higher risk of repression than those pursuing more moderate objectives.Footnote 16 Likewise, Charles Tilly also supported the proposition that those in power will react differently to resistance, depending on how acceptable they find contenders’ demands.Footnote 17
Civilian demands are likely to be more acceptable to rebels under two related conditions: first, while clashing with low-ranking commanders, civilian demands may coincide with the interests of an armed group’s leadership. One such example, discussed in the empirical section, is civilians calling for the removal of particularly abusive local commanders. For an armed group that is eager to restrain the use of violence against non-military targets, civilians complaining about local excesses to the leadership may be literally knocking on open doors, while risking reprisal on the part of field commanders. Such situations are examples of what Oliver Kaplan refers to as ‘joint-ness of interests’.Footnote 18 Second, civilian demands are likely more acceptable to the extent that they require only moderate adjustments on the part of rebels in specific policy areas, such as taxation, education and healthcare, or the movement of goods and people across military lines of control. Unarmed actions aiming to change rebels’ conduct of warfare (e.g. attacks on enemy forces in populated areas) or the use of specific weapons (e.g. anti-personnel mines) also qualify as moderate, as long as they do not challenge rebels’ use of violence for political ends altogether. In contrast, maximalist demands involve a direct and overt rejection of an armed group’s authority over civilian populations and claim to power, at a much higher risk of reprisal. This is why, as James Scott wrote, subordinate groups tend to start with ‘low-grade forms of resistance … at comparably little risk’ before resorting to more contentious actions that carry a high risk of reprisal.Footnote 19
Tilly identified a second factor that influences an incumbent’s response to dissent: the power of the contending group. The more powerful a contending group, the less repression it is likely to experience.Footnote 20 More recent literature on contentious civilian–rebel interaction contains variations of this argument. Ana Arjona theorises the conditions under which civilians are most likely to challenge their rebel rulers.Footnote 21 She identifies the quality of pre-existing institutions as a key explanatory factor in whether civilian resistance occurs. According to Arjona, an institution’s quality is a function of both organisational capabilities and less tangible properties, such as its legitimacy in the eyes of both civilians and rebels. Although Arjona does not use the notion of civilian power, the distinction she draws between organisational capabilities and legitimacy is also made in classical works on power. Drawing on Max Weber’s broad definition, power is usually understood as the ability of a social actor to use material resources (e.g. organisational capabilities) and social resources (e.g. legitimacy, interpersonal ties) to get others to do what they otherwise would not.Footnote 22 Traditionally, the main concern in social sciences has been to understand how dominant social groups gain and sustain power over others. By looking at how subordinate actors can limit the exercise of power by dominant actors, resistance studies provide a necessary corrective to this one-directional focus on power.
If one accepts the idea that the ruled can sometimes restrain the power of their rulers, it follows that subordinate groups also have power of their own. Power, in other words, is a two-way relation,Footnote 23 whereby dominant and dominated actors mobilise different kinds of material and social resources to their advantage. While power is a ‘sociologically amorphous’ concept that is not easily measured,Footnote 24 material resources, such as organisational capabilities, can be approximated empirically. Organisational capabilities, as Arjona suggests, vary with the extent to which civilians can draw direct and indirect support from existing institutions, such as faith-based entities. Kaplan also identifies pre-existing community-based organisations as a crucial social resource enabling civilians to ‘retain autonomy’ under rebel rule.Footnote 25 Varying levels of organisational capacity are also likely to influence the success of civilian resistance (not only its occurrence). However, my empirical research in Colombia’s Caquetá suggests that organisational capabilities were neither a necessary nor sufficient condition for success. Legitimacy was the more decisive social resource that seemed to weigh most heavily on the FARC’s preference for accommodation over repression. Like power, legitimacy is another concept that is not easily measured. A useful proxy measure for a resister’s legitimacy is their personal connections with individuals and local institutions close to the FARC. In Colombia’s Caquetá region, civilian resisters were often members of local governance institutions that were respected and actively supported by the FARC, former guerrilla members (i.e. veterans), or people with close personal ties to active FARC members. Violent repression of their demands would have carried a high price for the FARC, risking a reduction in their support base, which, as argued below, is a key determinant of rebel power.
This brings us to the third explanatory factor in the success or failure of civilian resistance, which is not explicitly identified in the literature on contentious rebel–civilian interaction: the power of the rebels themselves. In any armed conflict, a rebel group’s power will vary depending on its ability to establish a monopoly of violence in at least some territories. But this ability is in itself contingent upon both social and material resources. A key material resource, besides weapons, is physical terrain that gives weaker rebels a military advantage over typically stronger state rivals, as well as access to economic resources that can be exploited under wartime conditions.
Whereas physical terrain is a static (material) resource for rebel power, gathering popular support is a political process that relies largely, though not exclusively, on a transactional logic. ‘In a relationship of power’, Sherry Ortner wrote, ‘the dominant often has something to offer, and sometimes a great deal (though always of course at the price of continuing in power)’.Footnote 26 As noted in the literature, rebel-sponsored illicit economies may benefit civilians economically, thereby increasing popular support.Footnote 27 Beyond direct, financial benefits, many rebels may seek support by providing public goods that the state is unable or unwilling to supply, including basic security and alternative justice (e.g. people’s courts). Lastly, armed groups can also invest in political education or – in the case of faith-based groups – religious teaching. As revealed during interviews, FARC cadres disseminated the group’s ideology at the grassroots level prior to establishing a military presence. Transcending a purely transactional approach, such efforts aimed to cultivate a populace supportive of the guerrillas’ objectives and actions.
In places where social and material resources of power converge, rebels tend to portray themselves as alternative governments. In such rebel strongholds, which include most rural parts of Colombia’s Caquetá, we can expect rebels to respond to maximalist demands that pose a direct challenge to their rule with repression. Wickham-Crowley captured this idea particularly well: ‘When the guerrilla movements claim that they constitute legitimate counter-states, new governments in microcosm, they are sharply challenged, usually through the populace’s unwillingness to cooperate with the “legitimate directives” of the guerrillas.’Footnote 28 Looking at Taliban strongholds in rural Afghanistan prior to the group’s accession to state power, Jori Breslawski finds that ‘attempts to challenge the Taliban directly, both in terms of establishing community autonomy as well as armed confrontation, have for the most part been unsuccessful in lessening violence against civilians’.Footnote 29
The existing literature is rich in theoretical insights that help us develop hypotheses for how strong rebels may react to unarmed challenges. Yet theorising about the likely response of weak rebels to civilian resistance is a much more speculative endeavour. On the one hand, weak rebels who are unable to establish a monopoly of violence over a given area may plausibly allocate limited resources to fighting armed contenders and therefore ignore unarmed resisters. On the other, in areas of contested control, rebels also compete with their rivals for popular allegiance. Accommodating both moderate and maximalist demands may be to their advantage. Still, in areas marked by uncertain and frequently changing lines of military control, rebels are particularly vulnerable to civilian betrayal due to the geographic proximity of enemy forces that can react swiftly and more effectively to intelligence than in well-protected rebel strongholds. One might thus argue that a likely response of weak rebels to civilian resistance is repression. Since none of these courses of action can be excluded outright, caution is warranted when theorising how rebels respond to civilian resistance from a position of weakness or declining power. All we can say is that the response of weak rebels will be more uncertain than those of strong rebels, and is likely to fluctuate between repression, accommodation, and inaction (i.e. ignore).
Table 1 summarises hypothetical outcomes for different configurations of the three factors of interest: civilian power, the nature of civilian demands, and rebel power.
Table 1. Hypothetical Outcomes for Different Factor Configurations
Secondary Evidence
Let us now turn to the existing literature to analyse how different configurations of factors can play out empirically. The increasingly rich scholarship on civilian resistance contains numerous examples of unarmed action against armed groups. Documented cases range from public dissemination of instances of rebel violence against civilians and refusal to comply with Islamic State educational policies in Syria,Footnote 30 evading forced conscription imposed by Kurdish rebels,Footnote 31 and attempts at preventing rebels from governing certain areas, generally referred to as ‘peace zones’ (see below). Barring a few exceptions, however, existing scholarship is not explicit on whether or not documented acts of civilian resistance have succeeded in pressuring armed groups to accommodate their demands. One such exception is Till Förster’s research on Côte d’Ivoire, where civilians managed to wring concessions from armed groups without challenging an armed group’s authority. In response to widespread road racketeering and arbitrary taxation at rebel checkpoints, ‘influential traders’ called on rebel leaders to regulate tolls along the main transport axes. Förster finds that rebel leaders accommodated these demands and agreed to establish a formal system of fees and tax collection.Footnote 32
Many other examples of successful resistance have been documented in other contexts, particularly Colombia. Drawing on interviews with civilians in Colombia’s Nariño region, Annette Idler and her colleagues find that local communities managed to push the Ejército de Liberación Nacional (National Liberation Army, ELN), one of many left-wing guerrilla groups active in Colombia over the past century, to inform civilians about the presence of landmines.Footnote 33 Kaplan offers another systematic attempt to measure the impact of civilian efforts to change harmful policies. His monograph includes a chapter on community-led resistance in Colombia’s Santander department.Footnote 34 In this part of the country, where various armed actors vied for territorial control, the execution of alleged enemy spies was particularly widespread. To stop such killings, civilian community leaders negotiated an arrangement with different armed groups to allow individuals accused of collaboration to take flight and relocate to another area.
In these examples, the factor configuration was favourable from the resisters’ perspective. Rather than challenging the authority of armed groups as a whole, civilians demanded that rebels change existing practices and policies on landmine use and the treatment of enemy collaborators. Driven by strong leaders and backed by legitimate, pre-existing institutions, civilian resistance occurred in areas where territorial control was disputed by multiple armed actors. In line with the propositions above, relatively weak rebels responded with accommodation to moderate demands by strong civilians.
The literature also includes examples of civilians placing more radical demands on rebels. For instance, Arjona relates the case of an Indigenous mayor of a rural town in Colombia’s Cauca region. The man refused to step down despite the FARC ordering all mayors in the area to quit. He was arrested but released following protest by a large crowd of Indigenous people. As Arjona suggests, the political price for killing unarmed Indigenous protesters would have been too high.Footnote 35 This is indeed plausible, given that the FARC always presented itself as an anti-imperialist champion of the poor,Footnote 36 and Latin America’s Indigenous populations certainly suffered extensively at the hands of European imperialism. This is one reason why the FARC leadership publicly apologised for violence against Indigenous leaders as early as 1987 and also recognised Indigenous people’s right to establish their own governance institutions.Footnote 37
Another prominent example of civilian resistance in Colombia, which qualifies as a maximalist demand, is the so-called ‘peace zones’.Footnote 38 Masullo provides a particularly comprehensive analysis of a peace zone in a small town called San José, located in the northern region of Urabá. Beyond describing the community’s unarmed struggle, Masullo explains how different armed actors responded to the declaration of territorial autonomy.Footnote 39 As in Arjona’s case study, local mobilisation in San José was driven by strong Indigenous communities, supported by the Catholic Church. Yet whereas the mayor’s release in Cauca was a clear success, residents of the San José Peace Community were not so lucky. While armed groups could not stop civilians from defending their claim to neutrality, the San José Peace Community suffered violent incursions by both left-wing guerrilla groups and right-wing paramilitaries.Footnote 40 This speaks to the proposition above that weak armed groups with limited and fluctuating degrees of territorial control respond to maximalist demands more erratically, in ways that oscillate between accommodation, inaction and repression.
In none of these examples did armed groups respond consistently with repression. This might be linked to the fact that none of the above-cited scholars analysed civilian resistance in areas where rebels had established themselves as the sole contender for state power.Footnote 41 As I have argued, strong rebels are more likely to repress maximalist demands under such conditions than when their power is more limited. This gap in the literature provided the main justification for selecting Colombia’s Caquetá region as a subnational case for empirical research. The objective is to complement existing research focused primarily on civilian resistance against weak rebels by assessing the prospects and limits of civilian resistance in one of the FARC’s historic strongholds.
Caquetá: Resistance under Unfavourable Conditions
There are several reasons why Caquetá constitutes a hard case for civilian resistance to succeed. First, the department lacks powerful Indigenous communities, which could have posed a serious challenge to the FARC’s dominance. Unlike other parts of Colombia with sizeable Indigenous populations, such as the Cauca department, only 1.6 per cent of the total population in Caquetá are Indigenous people. But the absence of powerful Indigenous institutions in not merely a matter of demographics. Caquetá has a number of designated Indigenous reservation areas. As Teófilo Vásquez Delgado agues, these ‘do not possess the same cohesion and strength as in other parts of the country’.Footnote 42 Other local governance institutions exist in the department, including in rural areas where FARC influence was strong. As I will show, however, these entities cannot be conceived as autonomous institutions; they rather formed part of FARC’s local governance structure.
Second, Caquetá is part of the region the Colombian army referred to as the guerrilla group’s ‘strategic rear-guard’, at the intersection of the eastern Andes and the Amazon plain.Footnote 43 In early 1999, Caquetá’s second-largest town, San Vicente del Caguán, became the site of an unsuccessful round of peace talks between the FARC and the Colombian state. The government agreed to remove its armed forces from an area the size of Switzerland. This provided the FARC with a unique opportunity to showcase its ability to govern, both to domestic and to international audiences.
Third, unlike in other parts of Colombia, right-wing paramilitary groups failed to establish themselves as a countervailing power. Towards the end of the 1990s, some parts of south-western Caquetá came under the control of paramilitary groups.Footnote 44 However, according to FARC commanders, these paramilitaries never developed the capacity to operate beyond areas already under government control.Footnote 45 While such statements should be interpreted with caution, the commander of one of the main paramilitary groups publicly acknowledged that he had only 500 men deployed in southern Caquetá – an insufficient number to seriously challenge the much stronger FARC.Footnote 46 In fact, there is no evidence to suggest that the paramilitaries drove the FARC out of any of their rural strongholds.Footnote 47
Fourth, after peace talks broke down in early 2002, Caquetá became the epicentre of a massive counterinsurgency campaign known as ‘Plan Patriota’.Footnote 48 During this government offensive, which received financial backing from the United States under Plan Colombia,Footnote 49 the FARC lost some of its traditional areas of control to the government. The group’s military wing abandoned more central areas around municipal centres and retreated to remote parts of the Amazon plain and the Andes mountains. In the following years, until a new round of peace talks officially resumed in 2012, maintaining political influence over areas it had lost to Colombia’s armed forces was vitally important to the FARC. One way of achieving this objective was to declare electoral boycotts, which prevented the government from holding elections in Caquetá. The FARC also imposed travel bans on the department’s main roads, attempting to paralyse traffic. Enforcing these orders became part of the group’s survival strategy, leaving little room for civilian defiance.
Thus Caquetá qualifies as a paradigmatic example of a rebel stronghold. Congruent with the theoretical proposition formulated above, we can expect an armed group’s tolerance of civilian defiance and willingness to accommodate maximalist demands to be particularly limited in such areas.
Settlers and Guerrillas: A Symbiotic but Conflictual Relationship
Roughly the size of Portugal, Caquetá contains huge swathes of sparsely populated and partly uncharted territory. The majority of its current 465,000 inhabitants live in a relatively small stretch of land between two natural borders: the Andes mountains in the north-west and the Amazon forest in the south-east (see Figure 1). This area was cleared for farming during successive waves of colonisation by non-Indigenous populations. The first settler communities arrived in the late nineteenth century, but by the mid-twentieth century Caquetá had a population of barely 50,000.Footnote 50 The pace of colonisation accelerated in the second half of the twentieth century, with the arrival of people fleeing poverty and violence in more central regions during a period of Colombia’s history known as ‘La Violencia’ – a civil war that escalated in 1948, when liberals and communists took up arms against conservatives.Footnote 51 This particular combination of favourable physical terrain – mountains and dense forests – with marginalised populations receptive to left-wing ideas facilitated the FARC’s subsequent expansion and consolidation of power. In line with the theoretical precepts outlined above, in places where social and material resources of power converge, rebels are likely to respond to unarmed resisters with repression, particularly if civilian demands pose a direct challenge to their strategic interests and ability to inflict military and political harm upon their state rivals.
1964–78: Communist Enclaves
During La Violencia, armed peasant communities settled in an area called El Pato, which includes the north-eastern corner of Caquetá and parts of the neighbouring Meta department. El Pato was one of several ‘liberated zones’ established on the margins of the Colombian state, with Communist Party support.Footnote 52 Since their creation in the 1950s, these communist enclaves had been subject to military incursions by the Colombian army. In the early 1960s, state repression intensified due to domestic and international political dynamics. Alarmed by the Cuban revolution, the United States stepped up its military engagement in Latin America as part of a larger campaign to contain the spread of communism. In 1962, the Colombian army began a major US-backed military offensive aimed at bringing the ‘independent republics’ back under state control.
In May 1964, Colombia’s Communist Party established the Resistance Secretariat to coordinate the military response to the government assault on its strongholds. During a conference held later that year, the secretariat created the Bloque Sur (Southern Bloc) as an umbrella organisation, putting different peasant defence forces under a centralised command structure. At a second guerrilla conference in 1966,Footnote 53 the Bloque Sur changed its name to Fuerzas Armadas Revolucionarias de Colombia (FARC). At that time, the FARC was still a small organisation with roughly 370 fighters.Footnote 54 The group grew to only about 1,000 fighters over the next 12 years. During this period of ‘organic growth’ (‘crecimiento vegetativo’),Footnote 55 FARC activity in Colombia remained limited to the Pato region and other rural areas in neighbouring departments with a strong communist influence. This changed only after the 6th FARC Conference in 1978, during which the FARC doubled the number of its military fronts.
The 1980s: A Bloodless Expansion
Between 1980 and 1981, the FARC’s 13th, 14th and 15th Fronts left the Pato region for other areas of rural Caquetá. This geographic expansion was not a military campaign of territorial conquest, but a relatively bloodless undertaking that unfolded ‘in the silence of the jungles’.Footnote 56 As one guerrilla fighter explained,Footnote 57 before entering new areas front commanders would send small teams, including at least one woman, to carry out what the FARC called ‘trabajo de masas’ – political education at the grassroots level.Footnote 58 Dressed in plain clothes, these advance parties prepared the ground for the deployment of FARC military units. Meetings were held with settler communities to gather information on the terrain and its population, to disseminate FARC ideology, and to inform villagers of the guerrilla fighters’ impending arrival. This gradual expansion into rural areas lasted for about a decade. Towards the end of the 1980s, the FARC had eight fronts across all of Caquetá’s 16 municipalities.
By the time the guerrillas moved into Caquetá’s southern Amazon plain, the local economy had already begun to revolve around the production and processing of coca leaves.Footnote 59 Since the late 1970s, the influx of people attracted by the prospects of rapid fortune engendered a host of societal problems, which traditional settler communities were ill equipped to deal with: high levels of inter-personal violence, theft, drug addiction, gambling, prostitution, and other forms of economic exploitation. The FARC quickly organised communities, which established basic rules of conduct for everyday life. Over the years, these rules evolved into fairly complex regulatory frameworks, sometimes codified into what are known locally as ‘manuales de convivencia’ (manuals of cohabitation). These local codes were not FARC documents per se.Footnote 60 Rather, they were compiled and disseminated by community boards, called ‘juntas de acción comunal’. Officially created in the 1950s as local parastatal governance entities,Footnote 61 community boards existed throughout Colombia. Interestingly, rather than replacing these entities with its own parallel structures, the FARC encouraged their creation in its strongholds and infiltrated existing ones by co-opting local leaders.Footnote 62
One key function of the community boards was to provide basic public goods, such as health, education and transportation. The ‘Code of Conduct for the Middle and Lower Caguán’, obtained during field research,Footnote 63 states that every community member over 14 and under 55 was required to participate in communal work brigades.Footnote 64 Failure to comply was punishable by up to 30 days of labour on the roads, which involved repairing existing trails or cutting new ones through the rainforest.Footnote 65 To be effective, different rules and related punishments set by community boards required the presence of an armed actor willing to enforce compliance. The literature singles out precisely this role – guarantor of the common good – as a reason why people supported the FARC.Footnote 66 Qualitative evidence gathered in interviews partly supports the view that investments in local governance provided the FARC with a source of popular support. But research in Caquetá also reveals a more complex and often contentious relationship between local community boards and the FARC. While the guerrilla group used community boards to convey information to the population, civilians relied on them to express their discontent. To address abuse and misbehaviour by local commanders, which was a frequent cause of civilian discontent, community board members had to deploy personal connections with the FARC to escalate their demands. As interviewees explained, the act of bypassing local commanders – not actual discussion with the FARC’s leadership – carried a risk of reprisal.
Local Resistance: A Partial Success
Civilians in Caquetá placed two distinct types of demands on the FARC. The first included collective petitions to remove abusive local commanders. Consider the following account by Don Héctor, a man in his 70s.Footnote 67 In the late 1990s, he moved to the Amazon plain and settled in a village called Campo Hermoso. After the peace talks broke down in 2002, a new guerrilla detachment arrived in this area. This group, Don Héctor explained, differed from other FARC entities, mainly because its members were not drawn from the region. The new group was much more violent than any other FARC entity he had seen. Within weeks several civilians had been killed, most accused of being government spies. According to Don Héctor, who was openly sympathetic to the FARC, these accusations were baseless: ‘They killed the wrong people.’
Following unsuccessful attempts to discuss these incidents with the local commander in charge, a commission made up of several villagers, including Don Héctor and the president of the local community board, left Campo Hermoso to travel to the adjacent Macarena region, where they met with the FARC’s central command to complain about the new guerrilla detachment. According to Don Héctor, the entire group was then removed from Campo Hermoso and never seen again. Subsequent interviews with both the president of the Campo Hermoso community board and the FARC confirmed this.Footnote 68
The second type of civilian demand comprised collective and individual calls for greater flexibility in applying rules and regulations issuing from the local codes of conduct. Restrictions imposed on freedom of movement were a key area of contention. As stipulated in the ‘Code of Conduct of the Middle and Lower Caguán’, everyone aged over 14 had to register with their respective community board and carry a membership pass, which functioned as a de facto identity card in FARC territories. People who wished to move from one community board’s geographic jurisdiction to another’s needed an invitation letter from a local resident who had lived in the area for at least five years. Permanent resettlement required an ‘entry permit’, signed by every member of the community board’s directorate. Moreover, newcomers were prohibited from leaving their new place of residence for two years.
Full compliance with these measures was impossible for government employees. Because the Ministry of Education assigned positions centrally, newly appointed schoolteachers could not obtain recommendation letters from local residents, nor stay at their workplace for two years. Teachers had to travel to the municipal centre or to the departmental capital for training and other administrative business. To ensure that local schools could function, community boards negotiated travel restriction exemptions for teachers and other public service providers. However, low-ranking FARC cadres who belonged to the group’s militia structure did not have the authority to grant exemptions. This required intervention from above – from area or front commanders. As several interviewees confirmed, the FARC leadership did accommodate requests for exemptions. The main contention was often local militia members, who were over-zealous in applying the existing rules or refused to refer matters to their organisational hierarchy.
A teacher who was also president of the local community board’s education committee in south-western Caquetá related another instance involving a local militia member, who opposed the construction of a student residence. Given that the rural inhabitants of Caquetá were scattered across wide geographic areas with poor road infrastructure, it was impossible for many children to commute daily to the nearest school. Parents had to send their children to student hostels during the week. As the teacher explained, the International Committee of the Red Cross (ICRC) offered to build a hostel in her village, but the local militia member forbade any ICRC staff from entering the area, using the risk of infiltration as justification. Frustrated with the militia’s intransigence, the teacher took the matter directly to the FARC front commander for the area:
I went to the other side [the FARC]. Locally they did not want to listen. I went to a higher level to get the permits. So I sat down and talked to them and told them the things as they are: these are children who will benefit from the facility [the hostel]. They are children from here. They are not the children of the Red Cross. They are your children.Footnote 69
She also brought pictures to show the FARC commander of children sleeping on concrete floors in classrooms. Following her intervention, the FARC front granted the ICRC access to the area to build a student hostel, which opened eight months later.
In the cases related above, the FARC chose accommodation over repression of civilian demands. Successful examples are characterised by the absence of a conflict of interest, which confirms that the nature of civilian demands is relevant to a rebel group’s response to civilian resistance. What links these different cases of successful resistance is the fact that civilian demands did not clash with the FARC’s politico-military strategy. Take the example of education: unlike other armed groups – such as the Afghan Taliban, which deliberately targeted government schools – the FARC did not oppose public education.Footnote 70 Preventing the construction of student hostels was incongruent with the group’s ideology and interests. Likewise, the FARC had nothing to gain from commanders killing suspected spies at their own discretion. In fact, according to a former FARC commander, increasing numbers of executions of suspected spies at the hands of local commanders prompted the FARC to issue an instruction in the mid-1990s that required front commanders to seek approval for capital punishment from the FARC’s regional command structures.Footnote 71
Another commonality between the above examples relates to the legitimacy of the people who stood up to their rebel rulers. Typically, they were members of local institutions that were respected and actively supported by FARC, the so-called community boards. Many also had personal ties to FARC commanders, either because they had been part of the FARC’s fighting force and militia groups in their youth or because they had friends and relatives still enrolled with the guerrillas. Their demands could not be repressed easily. Responding with violence would have turned the FARC against its own constituents. Moreover, personal connections with the FARC enabled civilians to bypass lower-ranking guerrilla cadres. Whereas local commanders or militia members ignored civilian demands, the FARC leadership responded with accommodation. This confirms the importance of civilian power as a factor influencing an armed group’s response to nonviolent resistance and suggests that interpersonal ties between civilian resisters and rebel rulers are an essential social resource of power and a key determinant of civilian legitimacy. This finding also resonates with observations made in other contexts, including Sri Lanka. Looking at civilian resistance against Tamil Tigers in the north and east of the island nation, Nimmi Gowrinathan and Zachariah Mampilly show that only rebel constituents could challenge armed groups in their strongholds without facing repression.Footnote 72
In summary, as long as their demands did not clash with the FARC’s strategic interests, civilians in Caquetá managed to wring numerous important concessions from the rebel group: travel restriction exemptions for teachers and health workers, access permits for humanitarian agencies, and the removal of abusive local commanders. While these achievements are remarkable and largely unexpected, such civilian demands do not qualify as maximalist as they only entailed adjustments in specific policy areas, without challenging the FARC’s rule altogether. The underlying constellation of factors in the cases above was similar: strong rebels facing moderate demands from strong individuals or groups who could not be easily repressed without facing a backlash. Congruent with the theoretical framework presented above (see also Table 1), the FARC responded with accommodation. However, its tolerance of dissent had clear limits. As soon as civilians openly defied the FARC’s authority by interfering with the group’s strategic interests, the customary response was outright violence, threats and other coercive measures – in other words, repression.
The 1996 Peasant Marches: A Turning Point in Popular Attitudes
In August 1996, tens of thousands of peasants marched from Caquetá’s rural areas to the town of Florencia to protest against the government’s aerial spraying of coca fields. In FARC-held areas, where coca cultivation was the main source of income, this spraying hit people particularly hard. Although civil society organisations officially called for the march, the FARC supported the peasant mobilisation. The march served to pile political pressure on the government at a time when Bogota was already under strain due to several recent military defeats to the FARC. But the guerrilla backing was not merely political. The group also played a key role in mobilising peasants in its rural strongholds. Protest participation, however, was not achieved solely through pressure. People in rural areas were truly upset with the government for destroying their livelihoods without offering any economic alternative to coca cultivation. Juan Guillermo Ferro and Graciela Uribe describe the situation as follows: ‘Even though without the initiative of the FARC the mobilisation of [coca] growers would not have occurred, it is evident that all the participants rejected the State’s heavy-handed policies.’Footnote 73 The following statement – from a peasant from the neighbouring Putumayo department, where coca eradication campaigns provoked concomitant protests – poignantly captures the ambivalent civilian–rebel relationship: ‘We were voluntarily obliged to take to the streets.’Footnote 74 Regardless of whether and to what extent the FARC’s political strategy played to popular anger, those who would have preferred to stay at home seemingly had no choice but to participate.
According to the interviewees, everyone was obliged to go on the march, including children and the elderly. In families with livestock farms that could not be abandoned, one adult was allowed to stay behind. Failure to comply was punishable by expropriation. A middle-aged man who arrived in Caquetá as a coca harvester in the early 1980s recalled that he was walking on crutches at the time.Footnote 75 He was still obliged to participate, although the protest route included a three-day march on foot. Another rural resident, Don Alfredo, related a similar case: in 1985, he had led a previous protest march against the government in the Pato region, during which two of his children were killed in army custody.Footnote 76 He thought this was reason enough for exemption this time, but when he did not board the ferry the FARC had chartered to transport people to the municipal centre, the president of the local community board sought him out and delivered a message: ‘Don Alfredo, the order is for everyone to go on the march, dead or alive.’
Don Alfredo’s initial refusal to go on the march provides a good example of what Scott famously describes as ‘everyday forms of resistance’: petty acts of foot-dragging and evasion on the part of rural folks who lack the leadership and discipline to launch a more organised campaign.Footnote 77 Some people may have deceived the FARC and hidden from the local community boards. Yet there was no organised resistance campaign, as the fact that 75,000 people participated in the peasant march clearly shows.Footnote 78 For a department with fewer than half a million inhabitants, this was an exceptionally high rate of protest participation.
Forced mobilisation during peasant marches was frequently mentioned as a criticism of the FARC. Even civilians who otherwise valued the FARC’s role in organising communities considered forcing an entire population to abandon their farms for several weeks an abuse of power. One way civilians expressed their growing opposition to the FARC was by defying the group’s authority at the ballot box. When asked about instances of nonviolent resistance against the FARC, interviewees often mentioned cases of people voting against pro-FARC candidates in local elections. This is an interesting finding, given that the literature has not regarded political participation in state-led elections as a form of nonviolent resistance against rebel power.Footnote 79
Electoral Participation as Civilian Resistance
In 2006, when the military campaign against the FARC was in full swing, Colombia held congressional elections. In Caquetá, 90,000 electors were called on to choose their regional representatives for the national congress. As previously, the FARC sought to influence the outcome – this time by promoting their favourite candidates. A resident of Remolinos del Caguán (see Figure 1) remembered the local community board calling a meeting a few days before the election.Footnote 80 The objective of this gathering, which the FARC attended, was to pre-select a candidate. According to the resident, the FARC nominee was endorsed by the attendees, who ‘clapped their hands loudly’. Given that Remolinos was situated in an area with a strong FARC presence, the ballot box was not located in the village itself, but at the municipal centre of Cartagena del Chairá. On polling day, the FARC arranged for the villagers to be transported to Cartagena del Chairá. When the electoral results were announced, less than a quarter of all votes in the Remolinos ballot box had been cast for the FARC candidate. The majority of votes went to a candidate who openly opposed the FARC.
This is an example of electoral resistance. The implicit demand directed at the FARC was for the group to stop (ab)using its power to influence elections. It was a maximalist demand that challenged the FARC’s claim to power and ability to place its supporters in local political institutions. The example qualifies as a form of nonviolent resistance because, by voting for an anti-FARC candidate, civilians incurred a risk of reprisal. In a place like Remolinos, situated close to the FARC’s military camps in the Amazon forest, the FARC clearly had the capacity for carrying out collective punishment. Luckily this did not occur. According to residents, the FARC acted as if nothing had happened. The Remolinos case is interesting as it contradicts several theoretical propositions outlined at beginning of this article. First, the voting example does not neatly match the supposed sources of civilian power. Casting a vote, unlike campaigning for political office (see next paragraph), is an individual form of resistance that requires no mobilisation of specific material or social resources. Second, the example suggests that even strong groups may sometime have no choice but to ignore civilian resisters, for a lack of alternatives. There was nothing the FARC could do to selectively punish those residents who had frustrated the group’s attempt to promote its favoured candidate. Given the secrecy of the vote, it was impossible to identify individuals who had voted for the anti-FARC candidate. Collective punishment would have amounted to a public acknowledgment of the group’s lack of popular support in its own stronghold. Moreover, killing or expelling all civilians would have played into the government’s hands by draining the proverbial sea, which, as Mao Zedong famously noted, is the guerrilla’s most crucial advantage and – one might add – most precious social resource of power.Footnote 81
However, where identifying defiant individuals was possible, the FARC’s response was swift and violent. The following incident, which occurred in the same municipality a year later, illustrates how the FARC dealt with individuals who ran for office despite a FARC-declared electoral boycott. Unlike voters, who benefitted from the secrecy of the ballot, the identity of individuals campaigning for political office is, by definition, public. In October 2007, Liliana Polanía and Grataliano Murcía left the municipal centre in Cartagena del Chairá to campaign for election as mayor and member of the regional chamber of representatives, respectively. Both were affiliated with Convergencia Ciudadana, a political party that was part of the governing coalition in Bogota and whose founding members were later condemned for links with right-wing paramilitary groups.Footnote 82 One of their campaign managers explained that the local police chief had tried to convince the two candidates to stay within the urban perimeter of the government-controlled town of Cartagena del Chairá.Footnote 83 Nevertheless, the two ventured into rural areas with a strong FARC presence. According to the campaign manager, FARC fighters stopped their car less than 30 minutes’ drive from the municipal centre. Both candidates were shot dead on the spot. The remaining members of the campaign team were allowed to return with their bodies.
The decision to campaign in areas under the FARC’s influence despite its boycott of the election was an act of resistance with a fatal outcome. Quantitative data on political homicides in Caquetá show the killings were not an isolated incident, but part of a systematic campaign of targeted violence against local politicians:Footnote 84 as demonstrated by Figure 2 – which compares the number of FARC-attributed political homicides in Caquetá between 1990 and 2010 with the number of civilian killings (excluding politicians) during the same period – roughly one in four civilian fatalities in Caquetá was a politician.
Figure 2. Political Homicides and Other Civilian Killings Attributed to the FARC, Caquetá: 1990–2010
Between 1990 and 2010, the FARC killed 36 political officeholders or electoral candidates. A further six civilians lost their lives during these attacks, bringing the total number of civilian victims of politically motivated attacks to 42. The political homicides fall into two categories. The first includes politicians who ran for office despite a FARC-declared electoral boycott. They were killed either during their electoral campaigns or shortly after taking office. The peaks in political homicides in 1997, 2000 and 2007 represent this category. The second category of victims comprises politicians who refused to resign from office despite FARC directives to do so. Such orders were often based on individual accusations of corruption,Footnote 85 or issued as collective threats: in April 2005, for instance, the local radio station Caquetá Stereo received a flyer ordering all council members from the Puerto Rico municipality to resign or face death.Footnote 86 The councillors refused to step down. A month later, six of the 13 municipal counsellors died in the single most deadly FARC attack perpetrated against local officeholders in Caquetá. Prior to this event, the FARC had already killed three politicians from the same municipality. The 2005 peak of the black line in Figure 2 represents these councillors.
Defying FARC-declared electoral boycotts or ignoring rebel orders to step down represented implicit calls on the rebels to respect civilian rights to vote, campaign and exercise political office. These were maximalist demands that could not be accommodated easily through policy adjustments or amendments of specific rules and regulations. They directly challenged the FARC’s political authority, particularly in the more developed parts of Caquetá. Most of the political killings occurred in municipal centres, which had fallen back under government control in the course of a massive counter-insurgency campaign.Footnote 87 Although the FARC had lost its monopoly of violence in these areas and therefore no longer qualified unambiguously as a strong rebel group, it still retained the capacity for hit-and-run attacks from nearby bases. Killing local politicians in urban areas, who lacked personal ties with the FARC, was less costly in terms of reputational damage than killing members of local community boards, who operated with the group’s explicit approval. In other words, nonviolent action in the above examples of electoral resistance occurred under highly adverse conditions: weak civilians placing maximalist demands on weakened rebels who retained the capacity to punish resisters, including in areas they had lost to the government. The FARC consistently responded with repression.
Did political killings help the FARC achieve its politico-military objectives? Placing these targeted killings into the broader context of the armed confrontation helps to address this question. As seen in Figure 2, the first peak in political killings occurred in 1997, about 18 months before the beginning of unsuccessful peace negotiations. The FARC’s decision to boycott municipal elections in 1997 was a show of power ahead of the peace talks, signalling the group’s capacity to hit the government both militarily and politically. It also coincided with the establishment of paramilitary groups in some municipalities under government control.
After a period of relative calm between 2001 and 2004, political killings re-escalated in 2005. Even as the total number of civilians killed began to drop from 2002 onwards, political homicides continued to rise. More than half of all the civilians (23 out of 42) killed in the FARC’s targeted attacks on local politicians were assassinated over a five-year period: from 2005 to 2009. The 2005 re-escalation occurred a year after the Colombian government launched the so-called ‘Plan Patriota’. Caquetá and the adjacent regions of Meta and Guaviare were the epicentre of this operation, which involved the largest mobilisation of combat troops in the history of the conflict.
Plan Patriota was part of a wider strategy known as ‘La Seguridad Democrática’ (‘Democratic Security’).Footnote 88 One of the centrepieces of President Álvaro Uribe’s administration (2002–10), the objective was to recover areas that had been under FARC control for decades. Disrupting local elections and punishing local officeholders accused of corruption proved an effective way of discrediting the government’s counter-guerrilla strategy. It is indeed difficult to imagine a better way of exhibiting the ineffectiveness of a strategy called ‘Democratic Security’ than by making democratic elections insecure. The intended message was clear: you can push the FARC into the jungle, but you cannot prevent the guerrilla from striking back – not even in heavily fortified towns you claim to control. Eventually, the resulting stalemate helped to create the momentum for a renewed commitment to negotiations with the FARC. In 2010, Colombia voted for a new president, Juan Manuel Santos, who built his campaign on the promise of finding a political solution to over 60 years of armed conflict.
Conclusion
Field research in Caquetá showed that local resistance is unlikely to trigger change when directed against policies or tactics that form part of an armed group’s larger organisational (survival) strategy. The FARC chose accommodation over repression only in situations when civilian demands did not compromise its capacity to inflict military and political harm on the government. When civilian resistance clashed with the guerrilla group’s strategic objectives, as during the 1996 peasant marches and subsequent electoral boycotts, the FARC consistently responded with repression. We lack evidence for civilians in Colombia or elsewhere managing to push armed groups to make far-reaching, strategic concessions, let alone defeat rebels via nonviolent action. Thus, civilian resistance against rebels seemingly failed to replicate the claimed success of civil resistance against states.
Still, many armed groups suffer defeat. However, victory over rebels rarely results from a single course of action. Rather, it comes from a combination of violent and nonviolent forms of struggle. The example of the FARC clearly illustrates this point. What pushed the guerrilla group to the negotiation table was a massive counter-insurgency campaign launched in 2002. In the following years, the Colombian army severely weakened the FARC, yet without achieving outright military victory. The FARC’s decisive defeat was accomplished not on the battlefield but at the ballot box. The anecdotal electoral rebuff the FARC sustained in the village of Remolinos, when civilians voted for an anti-FARC candidate, was only a small foretaste of what awaited the guerrilla group after it disarmed. As part of the 2016 peace agreement, the FARC transformed into a political party. During the first post-agreement election in early 2018, the FARC party earned a meagre 0.3 per cent, relegating the group to the sidelines and leaving it unable to challenge the Colombian government against which it had fought for more than half a century.
It is plausible that Colombians used the 2018 post-agreement election to punish ex-rebels for their excessive use of violence, including against unarmed civilians who had stood up to the FARC. The violent repression of civilian resistance, however, failed to generate a massive wartime backlash against the FARC – at least within the group’s historic strongholds. Ironically, it was the FARC’s decision to leave the path of war and commit to peaceful forms of political struggle that triggered a decisive backlash at the ballot box. In sum, trying to assess the effects of non-violent resistance in isolation from other forms of power contestation – be it institutionalised electoral campaigns or military counterinsurgency campaigns – is problematic from both a practical and a theoretical perspective. Rather than viewing nonviolent resistance as a ‘distinct sui generis phenomenon’Footnote 89 with clearly attributable outcomes, it is more promising and relevant for future research to analyse cumulative effects of different form of past and present power struggles – including violent and non-violent forms of resistance.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022216X25100692.
Open access
