Ideational Factors and Civilian Contention in Civil War: How normative commitments and political ideas shape civilian collective responses to armed groups

Juan Masullo J.
Leiden University
Institute of Political Science

This is an extended and annotated (earlier) version of:

Find the PDF version of this paper here.

Paper prepared for the ”Annotation for Transparency Inquiry (ATI) Workshop.” New York, November 29 – 30, 2018. I thank Sebastian Bitar, Nelson Kasfir, Donatella della Porta, Pam Oliver, Annette Idler, Sidney Tarrow, Libby Wood and participants in the EISA 5th European Workshops in International Studies (June 2018, University of Groningen), the Young Scholars in Social Movements Conference (April 2018, University of Notre Dame), the ISA 59th Annual Convention (April 2018, San Francisco), the Workshop on Civilian Activism in Civil Wars (Feb 2018, University of Konstanz), the Annual Meeting of the Conflict Research Society (Sep 2017, University of Oxford) and the Workshop on Ideology and Armed Groups (June 2017, University of Montreal) for useful comments on previous versions of this paper.

Abstract

For long civil war scholarship largely ignored, or at best downplayed, the role of ideology in shaping both macro processes and micro dynamics of civil strive. While in recent years we have seen a renewed interest in exploring whether and how ideology matters, most of this work, naturally, has focused on how ideology shapes armed groups’ behavior. In this paper, I shift the focus away from armed organizations and argue that ideational factors are also fundamental to understand the ways in which organized civilians respond to armed groups. I focus on one particular response, civilian noncooperation — i.e., the refusal to cooperate with each and every armed group present in their territory – and argue that the form that noncooperation takes is – at least partly – shaped by normative and ideological commitments. Ideational forces have an impact on civilians’ repertoire of action and contentious performances: they push civilians towards nonviolent forms of noncooperation (as opposed to armed resistance) and incentivize them to engage in more confrontational forms of noncooperation. Relying on original micro-level data collected in warzones in Colombia, and using the strategy of paired comparisons, I provide detailed evidence for this argument and cast doubt on some alternative explanations. I trace the effect of ideational factors on the repertoire of action in three campaigns that engaged in distinguishable forms of noncooperation and show that the effect is independent and comes from exogenous sources.

Introduction

Despite of a background recognition that wars are fundamentally a political enterprise and that ideas are at the core of the struggle, when explaining civil war dynamics and processes, both at the micro and macro levels, civil war scholarship has privileged structural conditions, organizational characteristics and/or situational factors. Ideational forces have been left out of the picture or, at best, have remained largely marginal. In recent years, however, we have seen a renewed interest in understanding how ideology matters, especially as it relates to armed groups’ behavior.

From this work we have learnt, for example, that ideology can play a central role in armed groups’ decisions to use violence against civilians. Tactical escalation in violence by Sendero Luminoso in the Peruvian Civil War has been explained as resulting from an interaction between the ideology of the organization and the broader political environment (Ron 2001). Likewise, changes in patterns of selective and indiscriminate targeting overtime by insurgents in Mozambique and Angola has been explained as shaped by the salience of Marxist-Leninist ideals (Thaler 2012).¹ Ideology can also shape rebel governance (Kalyvas 2015). During the Greek Civil War, communist insurgents established systems of governance that differed significantly from those introduced by their more traditional, non-communist counterparts.

At the macro-level, ideology also seems to matter: civil wars where the main rebel group embraces a socialist or marxist ideology tend to be fought as irregular wars, last longer, have more fatalities and are more likely to lead to rebel defeat (Balcells and Kalyvas 2015). To be sure, these insights provide a more complete (and sophisticated) understanding of how armed organizations behave and how some war dynamics unfold. They extend, refine and sometimes challenge existing theories of civil war (see Thaler (2012)).

This paper joins this work by stressing the ideational factors also matter in the unfolding of civil war social processes; that is, the transformation of social actors, structures, norms, and practices in the context of war Wood (2008). However, rather than exploring how ideology shapes the behavior of armed organizations, I focus on how ideology shapes the behavior of civilians. Concretely, it explores the interplay between ideational factors and civilian noncooperation with armed groups. In doing so, it bridges two areas of inquiry that are currently gaining increasing attention in civil war research: ideology and civilian agency. Moreover, by understanding civilian cooperation as a form of contentious collective action and exploring issues that have been at the core of social movements studies – such as the choice of repertoires and performances (see, e.g., Tilly 2008), it strengthens a dialogue between civil war studies and contentious politics which, even if called for more than a decade ago (see, Tarrow 2007), is only recently capturing the attention of a larger number of scholars.²

Civilian noncooperation designates a set of behaviors by which civilians refuse to collaborate with each and every armed group present in their territory, including state and non-state armed forces. It can be individual or collective, can take a violent or nonviolent form, and can rest on acts of omission, commission or combinations of both. (Masullo 2017b). What is common to these different expressions of noncooperation is that it entails civilian behaviors that negatively affect armed groups and that can even harm them (Arjona 2017; Masullo 2017b).

While civilian support has long been central in civil war studies (Johnson 1962; Wood 2003; Kalyvas 2006), noncooperation has received scant attention in the literature. This omission is problematic, as recent work has shown that forms of noncooperation have the power to shape civil war trajectories in consequential ways, affecting crucial processes such as armed groups’ violence against civilians (Kaplan 2013a, 2013b), the systems of governance that armed factions establish in the areas they control (Arjona 2016), and the capacities of communities to engage in peace-building and post-conflict reconstruction work in the aftermath of war (Masullo 2018). While cooperation and displacement are the dominant strategies followed by civilians in warzones, civilian noncooperation is more prevalent that what available theories of civil war and collective action would predict and what observers and analysts would expect. Ordinary civilians organizing to refuse to cooperate with heavily armed groups ready to use violence against them seems a rather obtuse choice. However, I have identified instances of noncooperation in at least 18 different countries that have experienced or are experiencing civil war and, in Colombia alone, I have identified more than 50 experiences of nonviolent noncooperation in at least 19 of the country’s 32 departments (Masullo 2017b, 2017a).³

While in previous work I have theorized the emergence of noncooperation (Masullo 2017a), in this paper I take on the task of explaining why it takes different forms when in emerges. Civilian noncooperation campaigns can be violent or nonviolent, and may involve different degrees of confrontation vis á vis armed groups, ranging from oblique manifestations of disobedience to public, unilateral declarations of entire areas off limits to armed groups. I propose that ideology is a central factor – even if not the only one – in explaining this observed variation. Following the strategy of paired comparisons (Tarrow 2010), I explore three campaigns of civilian non- cooperation in the Colombian civil war, representing three different types that exhibit varying levels of confrontation. The cases are located in three different regions of the country that were particularly affected by armed conflict. The basic composition of the communities that engaged in noncooperation and the situational war factors under which each campaign emerged were fairly similar: all three communities are peasant communities which engaged in noncooperation campaigns when territorial control was in flux, when the levels of violence against them was sharply increasing and when civilians came to perceive targeting as unavoidable. Nonetheless, the form that noncooperation took differed importantly from campaign to campaign.

The aim of this paper is two-fold: (a) provide an accurate account of the ideational and normative commitments and constraints underlying the three contentious campaigns under study; and (b) make the case that ideational factors have an independent effect on the form noncooperation takes when it emerges. In doing so, I heed Gutierrez Sanín and Wood (2014) call to bring the role of ideas back to the analysis of civil war. A few recent studies spelled out how (some) armed groups make efforts to inculcate communities with norms and ideas to promote their own ideologies (Wood 2009; Mampilly 2011; Gutierrez Sanín and Wood 2014) through processes such as socialization into violence (See contributions to a recent special in the Journal of Peace Research Checkel 2017). Beside this, the way ideational factors shape civilian choices and behavior in war has been largely ignored. In this paper I aim to show that the study of ideational factors in the context of civil war is also pertinent to understand civilian agency, in particular when civilian responses to armed groups takes a collective, organized form. I hope to open a new complementary avenue of research in the nascent program on ideology and civil war that looks beyond armed organizations and that recognizes civilians as agents, not only as mere victims or resources to be plunder.

The Argument

While in the context of civil war survival considerations are indeed a central driver in civilian decision making, civilian choice does not only follow a strategic reasoning. As Steele (2017, Chap. 2) has shown in her study of wartime migration, civilian choices are also a function of political preferences and loyalties. The preference for noncooperation, and the capacity to act upon this preference, is formed by a from a mix of situational factors stemming from localized war dynamics and factors associated with community structure (Masullo 2017a).⁴ However, when it comes to the choice of repertoire of action – that is, choosing whether noncooperation is violent or nonviolent and more or less confrontational –, I argue that ideational forces play a central role, even if they are note the sole factor shaping these contentious choices.

I follow Gutierrez and Wood’s (2014, 214) in their contention that a set of more or less systematic ideas can identify a constituency and define the objectives that a given collectivity pursue and a (perhaps vague) program of action, including strategies and institutions for the realization of those objectives. Rather than focusing on the way a given ideology defines a constituency (e.g., civilians), the challenges this constituency faces (e.g., violence from armed groups), and the objectives it pursues (e.g., self-protection from violence), my argument emphasizes on the latter, the ways in which a set of more or less systematic ideas can prescribe the strategies chosen for the realization of group goals. I stress the role ideational factors play in the mobilization phase (rather than in shaping decisions of an already formed group, something that is more common in the literature), when some of the fundamental features of a campaign of collective action are decided (Ruggeri and Costalli 2017). Concretely, I look into how normative commitments and political ideas define and constrain the strategies and tactics of noncooperation at two different levels:

1. civilians’ choice for nonviolent methods over violent ones and,

2. civilians’ choice of the level of confrontation involved in the campaign.

In exploring these two levels, my argument takes the study of ideational factors beyond an exclusively instrumental approach, adopting instead what Gutierrez Sanín and Wood (2014) call a “strong program”. However, I stress – and substantiate – that the role ideological andmoral commitments does not make civilian choices less strategic (For this debate in the context of nonviolent actions, see McAdam and Tarrow 2000).

By examining in detail three instances of noncooperation from the Colombian civil war comparatively, I show that normative commitments to nonviolence, stemming largely from religious beliefs, played an important role in civilians’ decision to opt for a nonviolent repertoire of action. In addition, I show that left-wing oppositional ideologies, linked to the workings of pre-existing political parties, trade unions, peasant associations and even armed groups, shaped the level of confrontation displayed in each campaign under analysis.

Theories of political choice commonly recognize an effect of ideas on actors’ choices. Choices generally flow from cognitions and cognitions, in the forms of beliefs, are commonly informed by ideas (Elster 2007). The argument proposed here shares this basic understanding, but goes beyond. A key requirement for a deeper ideational account that can make the case that ideas have some independent effect on choice is that ideas are not endogenous to the material and situational features of the choice situation which is presently being explained (Bennett and Checkel 2014, 44, fn.2).

I embrace this intuition and show that the sources of this ideational input were exogenous to the conditions that gave birth to the choice for noncooperation. In tracing the effects of ideational factors on noncooperation, I account for the “exogenous condition” by expanding the empirical scope of the analysis beyond both the concrete moment in which the choice for noncooperation was made and the main dyad of actors in the decision making process (that is, civilians–armed groups). I specify the main channels of moral commitments and political ideas into noncooperation campaigns – community leadership and external networks of support – and show that in the processes of working towards preference convergence (Schelling 1960) and securing mobilization resources, these actors found enough room to imprint a particular ideational outlook to the campaign.

I show empirically that the embracement of this ideological content pre-dated the choice for noncooperation and the process of mounting the campaign. The actors who pressed for particular normative commitments and political ideas came to the decision making process with a set of ideas they had already developed or embraced, rather than developing them at the moment of defining the program of action for noncooperation. External actors, such as the Church or national NGOs, brought with them an existing set of normative commitments to the process that was a defining feature of their organizational identity. Community leaders brought to the process particular political ideas to which they had been exposed and had embraced long time before through their participation in previous experiences of contentious collective action within local organizations and associations. It is in this light that I argue that the effect of these ideational factors on the repertoire of action was largely independent.

Data and Structure of the Comparison

The evidence supporting this argument constitute a rich body of original micro-level data collected through immersive field research in multiple warzones in Colombia in 2014 and 2015. I gathered fine-grained data on the histories of the localities where each of the three campaigns emerged, covering both pre- and during-war experiences. I collected information on the social structure of the communities inhabiting these localities and traced the presence of and influence of local organizations, participation in social and political movements, and the existence and role of community leaders. Moreover, I gathered detailed data on localized war dynamics, including patterns of violence and territorial control and the ways in which local residents interacted with different armed organizations.

I made multiple visits to each locality, involving long-term, engaged presence. As the proponents of the contentious politics approach have noted, fieldwork is the method perhaps best suited to the study of contention (McAdam, Tarrow, and Tilly 2008, 317; Tarrow 2007). The chief data collection technique that I used was interviewing. I conducted over 150 semistructured interviews and plenty less formal and more conversational interviews. The core questions and main probes included in the interview instruments were the same for all three campaigns to ensure the collection of comparable data. I spoke to a wide variety of stakeholders, including civilians, (ex) members of state and non-state armed groups, local politicians, and staff from different organizations that played a role in the process of mounting noncooperation. While I spent a considerable amount of time interacting with leaders, which allowed me to build “leadership profiles” for each of the campaigns understudy, I also interviewed “rank-and-file” participants and non-participants who lived in the localities, always trying to keep a gender balance and getting access to people who were present during the process leading to the organization of non-cooperation. This purposive sampling strategy, allowed me to capture a wide range of vantage points and build a multi-perspectival orientation (Snow and Trom 2002, 154; Fujii 2011, 30), which would have been virtually impossible had I followed a random strategy to select informants. As I visited each locality several times during the course of the project, I interviewed some key informants multiple times and held follow-ups with several interviewees (some times upon their own request). This proved vital to cover sensitive issues and to go beyond what Tilly (1999) once called “troubled standard stories” – of which I got a lot in my first visit to each community.

Beyond interviewing, I conducted memory workshops (Grupo de Memoria Historica 2011; Arjona 2016) and map-drawing exercises (Wood 2003; Petersen 2001). These two techniques were particularly effective in helping participants recall past events and ease them back into specific time periods. With this techniques I built timelines and mapped spatial dynamics. Overall, these interaction-based techniques were especially well suited to access the rich and nuanced data needed for understanding the complexities related to strategic choice and mobilization (Jasper 2004, 11), in particular in the context of civil war (see, e.g., Wood 2003; Shesterinina 2016; Jentzsch 2014). Finally, while in the field, I conducted archival research at local public and private archives, collecting and analyzing different types of documents, from press releases to documents created by the communities and leaflets distributed by armed groups as events unfolded. This information allowed to triangulate and cross-check data coming from testimonies, countering some of the potential issues that can arise from problems associated with memory or the strategic reconstruction of events given the realities at the moment in which testimonies were collected.

I engaged deeply with the communities living in the localities I studied. I lived in their villages and spent most of my time in activities other than (explicitly) gathering data (working in cacao or banana fields, reconstructing rural trails, participating in community potlucks, studying with kids in local schools, playing dominoes with elders in the evenings, and practicing sports with youngsters, among many others). This not only allowed me to built a rapport with the communities, but also to learn a big deal about the context and subjects that I was studying through detailed observation. This enabled an understanding of a very complex social process – civilian noncooperation with armed groups – that I could have not reached with a less engaged and immersive approach to the field, let alone through other data collection techniques such as surveys or event data analyses alone.

The structure and logic of comparison

I investigate three different campaigns of noncooperation in the Colombian civil war, covering the range of variation that noncooperation can take according to a typology that I offered elsewhere (Masullo 2017b) – each of the types are briefly described in the next section when I introduce each of the cases under study. The first campaign is the Youth’s Project of Peace (Joppaz), which emerged in 2000 in the municipality of San Carlos, in the Eastern Antioquia region in northwestern Colombia. This is a case of what I call oblique noncooperation, the less confrontational type in the typology. The second one is the Peace Community of San José de Apartadó (PCSJA), which emerged in 1997 in the municipality of Apartadó, in the Urabá region in northwestern Colombia. This constitutes an instance of unilateral noncooperation, the most confrontational type analyzed in this paper. Finally, the third case is the Peasant Worker Association of the Carare River (ATCC), which emerged in 1987 in the Magdalena Medio Region in central eastern Colombia. This experience is an expression of pacted noncooperation, a type that is more confrontational that oblique noncooperation but less than unilateral.

I analyze them comparatively at the two levels proposed in the argument, through two sets of paired comparisons (Tarrow 2010): first, Joppaz and the PCSJA, and then the PCSJA and the ATCC. As for Level (i), both paired comparisons (all three cases) show that normative commitments brought into the process by the Church and other faith-based organizations shaped civilians choice for nonviolent forms of action. The first pair compares two cases in which both the Church was present and civilians opted for nonviolent forms. Therefore, the question of whether the story would have been different in the absence of the Church is left open. The ideal control case to test this proposition would have that one of two forms: either one case in which the Church (or any other equivalent “nonviolent entrepreneur”) was not present to explore whether in its absence villagers would have engaged (or at least tried to) in armed resistance; or one in which civilians did engage in violent forms of action to explore whether the Church or any other “nonviolent entrepreneur” was indeed absent. Unfortunately, I did not find such a control case. However, the second paired comparison provides a second-best plausibility test for my argument. By including the ATCC, I analyze one case in which community leaders explicitly considered violent forms (and perceived they had the means to engage in it), but dropped idea when a “nonviolent entrepreneur” entered the process and explicitly made nonviolence a condition for its participation. Within-case evidence from this case, allowing to compare a before and after and think counterfactually, provides more confidence on the argument.

As for Level (ii), the two paired comparisons together include the three different degrees of confrontation that noncooperation can take according to my typology and each pair yields additional weight to the argument in different ways. The first pair compares the two cases in the extremes of the continuum; that is, a case with very low levels of confrontation (Joppaz – oblique) to one with very high levels (PCSJA – unilateral). However, as much as these two cases differ in terms of their level of confrontation, they also differ significantly in their organizational capacity, with the more confrontational case, the PCSJA, exhibiting considerable higher levels of organizational capacity. This co-variation raises the question of whether ideational factors play any role or if capacity is doing all the work in explaining variation in form. To address this issue, ideally I should compare two cases that are fairly comparable in terms of organizational capacity and where villagers engaged in similarly demanding (from a mobilization and organizational resources point of view) campaigns of noncooperation. In other words, a pair of cases that allow me to keep capacity constant and explore the role of ideology more closely. In this case, I was fortunate to find these cases. The second paired comparison examines two campaigns with very similar levels of capacity for collective action that engaged in two campaigns of noncooperation that where highly demanding in terms of mobilization and organizational resources: the PCSJA and the ATCC. Despite of both having high levels of organizational capacity, the campaigns still differed in their level of confrontation: the former engaged in a unilateral campaign, while the latter in a pacted one.

Empirics

Paired Comparison I: Joppaz & PCSJA

The Joppaz and PCSJA represent respectively the least and the most confrontational forms of noncooperation. A comparison between these two campaigns shows how normative commitments shaped the form noncooperation both in terms of villagers’ decision to restrict their repertoire to nonviolent methods and in the level of confrontation involved.

In oblique forms of noncooperation civilians refuse to cooperate with armed groups in an indirect, disguised way. Disguised not in the sense of concealing, as civilians do engage in overt and visible actions, but in the sense of advancing it through activities that are not openly related to war dynamics and that do not imply a direct expression of defiance. Civilians practice noncooperation without publicly declaring it, and many times remain carefully circumspect and institutionally invisible. In contrast, in unilateral campaigns civilians refuse to collaborate with armed groups in an overt and open way. Noncooperation is unequivocally signaled to armed groups and civilians often make public declarations of their decision to not cooperate with armed groups. Moreover, the activities they engage in to advance the campaign, and the norms of behavior they design and opt to abide to, are explicitly and directly related to the dynamics of war.

In the municipality of San Carlos, in the late 1990s, the control of the municipal center the Revolutionary Armed Forces of Colombia (FARC) – the largest left-wing rebel group in the Colombian civil war – had was fiercely challenged by one paramilitary block, the Bloque Metro (BM). Responding to the disorder and uncertainty that competition commonly induces into civilian life (Arjona 2016), a group of high school students gave birth to Joppaz. With the support of the Pastoral Social of the Catholic Church and a body of volunteer firefighters, Joppaz organized street social activities in the evenings, convening people to play board games and share chocolatadas (i.e., making hot chocolate in huge pots for everybody to drink). They did so precisely when, due to gunfire, firecrackers and bombings, people had opted to lock themselves in their houses after dusk. Many residents welcomed the campaign. From a group of 8 or 10 that showed up in the first evenings, participants went up to 50. By doing so, they refused to cooperate with both factions. They defied an implicit dusk-to-dawn curfew that armed groups had placed on the town, avoided recruitment (especially youth recruitment), and countered the distrust and isolation that was feeding deadly cycles of denunciations and counter-denunciations. Joppaz was an innovative and courageous experiment of civilian self-protection in the midst of war.

The experience of Joppaz is an illustrative example of what I term oblique noncooperation. Jimena, a young resident of San Carlos who took part in the activities, characterized the campaign as “visible and invisible”: while people were in the streets, they were not in the central square; while the campaign was about opposing armed groups, there were playing Parcheesi and eating together.⁵ While residents of the municipal center refused to cooperate with armedgroups by disobeying their implicit and explicit norms, they did so without any overt outward manifestation of dissent.⁶ Naturally, as noncooperation is ’masked’ behind other activities, direct interaction with armed groups is minimal and the level of confrontation between civilians and armed factions is very low. Rather than substantially building a new order, as the one we will see in the cases of the PCSJA and ATCC, San Carlos’ residents sought to “work the order to their minimum disadvantage”, as in Hobsbawm (1973). To do so, they opted to minimize confrontation. This is not to say, however, that participants were not aware of the oppositional nature of what they were doing. This is especially the case for organizers. In fact, Camila, an active organizer of Joppaz, described these activities as “civil disobedience” with a “dissimulation mechanism”, and added that it was done that way to avoid being “questioned or sanctioned by them [armed groups].”⁷

The form noncooperation took in the municipality of Apartadó is quite different. When most of the residents of the village of San José had fled the area due to violence coming from multiple armed groups, around 1200 peasants decided to stay put, declared themselves neutral and established a “peace community”, the PCSJA. The war context was very similar to that of San Carlos: territorial control was in flux and violence against civilians was peaking following the military challenge that FARC was facing from the Peasant Self-Defense Forces of Córdoba and Urabá (ACCU), an arriving paramilitary army. By establishing the peace community, villagers pledged not to participate in any possible way in the war and disavow any form of cooperation with all armed groups present in the village, including the national Army and the Police. With flags, symbols, billboards and fences, they explicitly designated physical areas where community members would live and where armed groups, without distinction, were not allow to circulate or pernoctate. On March 23 1997, with representatives of national and international NGOs, the Catholic Church, and the local government in attendance, villagers publicly signed and presented a Declaration stating their commitment to noncooperation and informing armed groups operating in the area about their choice. The declaration contains the basic principles, norms of behavior and organizational structure that since 1997 and until today, have governed villagers’ daily lives.

No previous consultation, let alone negotiations took place between armed groups and civilians. According to demobilized rank-and-file combatants who operated in the area, civilians decision to engage in noncooperation came largely as a surprise.⁸ The case of the PCSJA is an illustrative example of unilateral noncooperation. By delimiting a physical space off-limits to armed groups in an area of strategic importance, this campaign involved high levels of confrontation. In these areas armed groups are particularly unlikely to give in to civilian pressures (Arjona 2016). Therefore, noncooperation gave rise to a series of contentious encounters between armed groups and villagers that still persist today.⁹ In many occasions villagers have had to ask armed groups to leave their lands and even to physically push them away. These encounters, in particular during the initial months of the campaign, have turned violent on the side of armed groups, with many civilians loosing their lives.

Level (i): The choice for nonviolent methods

Both campaigns, despite the different levels of confrontation involved, are resolutely nonviolent. To be sure, the decision to stick to nonviolent methods of action was influenced by the fact that, after so many years of violence, villagers were, as a volunteer firefighter that participated in the Joppaz campaign put it, “sick of violence”.¹⁰ However, in this section I argue that external actors that supported the emergence of these campaigns, by providing material and non-material mobilizing resources, shaped civilians’ choice for nonviolent methods. In both Joppaz and the PCSJA, the Catholic Church and the normative commitments it upholds, played a role in restricting the campaigns’ repertoire of action exclusively to nonviolent methods of action.

The Pastoral Social of the Catholic Church, in particular the local Parish, supported and advised the group of high school students that launched Joppaz in how to bring their idea into practice. Jaider, one of the campaigns funding fathers, recalls well the day when one of his classmates broke the silence and asked him and some of his friends whether they wanted to do something to counter the war that was being waged in their town. By that time, young males were constantly being approached by the different armed actors (mostly the paramilitaries) to join their ranks. School was a prime recruitment pool. In fact, by the time Jaider’s friend broke the silence, several classmates had already joined armed factions, did intelligence work at school and patrolled the town’s streets at night.¹¹

After holding a clandestine meeting to discuss their options, Jaider and five of his classmates addressed the Pastoral Social for support and express their disagreement with what armed groups were doing in town. “We want to carry out some actions, we want to mobilize people”, they ventured.¹² In secrecy and with the support of the Parish, they explored different forms to manifest their dissent, from symbolic “light marches” (walking around central square with candles) to overt protest against armed groups. After long deliberations, they came with the idea of creating Joppaz and organize street social activities in the evenings with the aim of shaking off people’s apathy and reactivating social interaction.

Jaider, today a promising social and political leader of San Carlos, by the late 1990s was only a high school student. As the other organizers of Joppaz, he was short of a community leader and he was not involved in politics. However, he (as well as other classmates) had been active in many activities organized by the Pastoral Social. This gave them some recognition within the community and allowed them to socially appropriate (McAdam, Tarrow, and Tilly 2001) the Church for launching an organized civilian response to violence. Doing so proved to be crucial for the success of the campaign. The Parish and the Pastoral Social provided material and non-material mobilizing resources that were critical for creating Joppaz and mobilizing the population. The Church was a relatively safe physical space in a town constantly exposed to military presence and incursions from different armed factions. By the time Joppaz was created, gathering in public spaces was very dangerous, as armed groups were highly suspicious about these meetings and people were exposed to denounced to either faction by distrustful fellow residents. Under these circumstances, counting on a relatively safer physical space to meet after school hours was a vital organizational resource. The Church was safe both in terms of being somehow shielded from military attacks and of not being subject to infiltration and control. As in many other civil war contexts, the Church served as a sort of sanctuary (Hancock and Mitchell 2007).

In this way, the Pastoral social was key in solving coordination problems in a context in which interaction was avoided and people had enduring incentives to falsify their private preferences (Kuran 1995). In addition, the Church provided encouragement and validation. Several interviewees recalled the role of one particular priest, Father Hector Blandón, who got deeply involve with this initiative. Although not all of the organizers and participants of Joppaz were devout Catholics, organizers highlighted the importance of feeling that an institution that has been so central in the social and community life of San Carlos, and a well-known and well-respected personality in town, such as Father Blandón, backed them.

The supportive role of the Church was not, however, limited to facilitating collective action. With their own moral commitments, their pacifist approach to war, and their tradition of organizing activities in rejection to violence in San Carlos (and the Eastern Antioquia region at large), the Church had an effect on the form the campaign took. It promoted values of non-violence and invited the youth to embrace a discourse of peace and reconciliation, rather than one of protest and opposition.¹³ In line with the tradition of the Church in the municipality, Joppaz committed to the exclusive use of nonviolent methods. The centrality of “peace” in the whole framing of the campaign, present even in the name they gave to the group, was directly influenced by the Church. Organizers were unequivocal in stressing that they never considered any armed response to armed groups. However, some noted that, as the idea was coming from high school students that were constantly being approached by armed groups, their original ideas were much more along the lines of “resisting recruitment” than of “building peace”. It was the Pastoral the one that provided the “peace” frame, which easily resonated with the organizers as both things, fighting recruitment and building peace, were perfectly consistent.¹⁴

Beyond this framing, as the campaign unfolded the evening activities began to involve performances closer to the repertoire of acton that the Church had exhibit in San Carlos in the past and that reflected more their moral, pacifist commitments. For example, while the activities were mostly about playing board games and eating something together, participants told me that in the middle of the evenings, they would stop and pray the Rosary. It is hard to tell if these would have been otherwise if the Church would have not supported Joppaz, as the campaign evolved within the context of a highly Catholic community. However, organizers note that they did not originally conceived this as part of the activities. When I asked one of them about where their repertoire of action came from, she underscored that beyond the games and the chocolatadas, many things were taken from the few “acts of resistance” that they Church had organized in the town before. Some examples, beyond praying, included wearing white shirts, marching around the park and lighting up candles.¹⁵ All these, to be sure, are methods of nonviolent action (Sharp 1973).

The Catholic Church also played an important role in helping the PCSJA mobilize and, in doing so, they manage to transmit key ideas and norms of nonviolence. As in San Carlos, the Church also impacted the framing and performances of the campaign. However, in this experience its role was more explicit and went far beyond. The Church served as a key broker for mobilization, a role that has been found repeatedly as a central enabling force behind multiple expressions of collective contentious action (see, among others, McAdam, Tarrow, and Tilly 2001). Since the early days of this campaign, the Diocese of Apartadó supported villagers and, to facilitate organization and mobilization, it brokered with different external actors that came to support the few residents of San José that had not yet displaced.

Similar to the way the process unfolded in San Carlos, after discussing how to respond to intolerable levels of violence for several days (even months according to some accounts), San José villagers decided to reach out the Church for support. Violence against civilians was becoming more frequent and the paramilitaries that were breaking into the area had already given civilians explicit orders to leave their lands unless they wanted to be massacred. Leonidas Moreno, a priest from the Diocese of Apartadó, was the person who responded the peasants’ call:

I remember well when peasants from La Unión [a rural hamlet where residents were ordered to flee] arrived to my office to ask me what do to, where to go. They told me that the Army had gone to them, said that although they had no issues with them [...] they had to leave within 15 days, otherwise the “beheaders” [los mochacabezas] will come and get rid of everybody.¹⁶

When asked about the decision to reach the Church for support, a pivotal leader in the process recognized that, despite of them being resolute about staying put and refusing to cooperate with the parties in contest, they were not clear about how to do it and thought the Church could help in this regard.¹⁷ They were right. Some people within the Church, such as Monsignors Tulio Duque and Isaías Duarte Cancino, had already thought about possible ways to deal to the emergency situation in the region. Therefore, they had some ideas of how to go about it and eventually proposed to set up a “peace community”.

In fact, as in the Joppaz campaign, the mere concept (and name) of “peace community” came from the Church. Already in 1995, in the neighboring municipality of Turbo, Monsignor Tulio Duque had publicly promoted the creation of peace communities as a strategy to protect civilians in the midst of war (Hernandez Delgado 1999, 72). While none of the members of the PCSJA I spoke with reported being aware of this, almost all of them recalled Monsignor Isaías Duarte Cancino, Bishop of Apartadó, when the initial steps towards mobilization took place, and some recognized him as one of the initial instigators of the idea of a peace community. Apparently influenced by the ideology of the Liberation Theology and the experience of the Basic Ecclesial Communities in Brazil (Aparicio 2009, 107), Duarte Cancino had proposed the formation of a neutral zone in San José de Apartadó (SJA) already in 1996. With the idea of “peace community” channeled by the Church, responses that would include violent methods of action or siding with one faction – such as the idea of “active neutrality” proposed by the then Governor of Antioquia, A´lvaro Uribe Vélez, which implied noncooperation with illegal armed groups and cooperation with the forces of the state – were ruled out. With this, the Church had a strong impact on the form the noncooperation campaign took in San José, limiting too its repertoire of action to nonviolent action. As noted by a representative of Fellowship of Reconciliation (FOR), an interfaith non-profit organization based in the US that has supported the PCSJA for many years now, the “Peace Community” label sends a clear signal, both to the inside and the outside, that whatever villagers autonomously decide to do in order to advance and sustained their campaign, that has to be nonviolent and in the name of peace.

However, the ideational input of the Church went beyond this. Through the Diocese of Apartadó, and in particular with the help of the then mayor of Apartadó, Gloria Cuartas, the Church brokered with other actors that came to support the process. During the still incipient initial stage of mobilization, a group of nuns came to San José and, working along with the Red Cross, provided villagers with humanitarian assistance. This help proved to be vital, as they managed to bring foods and medical supplies to the place where those who stayed were confided due to a road block established by the paramilitaries at the beginning of the access road to the village. Every and each of my respondents highlighted that they could neither leave the village, nor could not bring supplies from the Municipal capital without risking their lives. Indeed, many lost their lives or were disappeared in a place called Tierra Amarilla, where the paramilitaries established a permanent base for over 9 months.

While these nuns played a central role and many villagers remember them – in particularMother Clara – as central to the process of formation of the Peace Community, the external actors that had a stronger impact on the Peace Community both organizationally and ideationally were the Center for Investigation and Popular Education (CINEP) and the Inter-congregational Commission of Peace and Justice (CIJP), two faith-based organizations and well-known national NGOs based in Bogotá, the country’s capital. Within these organizations, two individuals stand out in devotedly supporting and shaping the PCSJA in definitive ways: Javier Giraldo, a Jesuit priest, and Eduar Lancheros, a human rights defender. As original documents from these organizations reveal, both CINEP and CIJP were already working in alliance with the Church in other conflict-affected regions of the country to defend and protect civilians, especially displaced populations. Therefore, when they arrived to San José, brokered by the Church and Cuartas, they had already experience in dealing with similar situations and had in mind a set of ideas of how to do so.

Cuartas refers to the arrival of these organizations to San José, as a “great moment of encounter”.¹⁸ In fact, they arrived precisely when San José villagers needed them the most, as the community had lost most of its leaders – some had been killed and many others had left – and thus lacked the mobilizing muscle to set up a campaign of noncooperation. These two organizations provided training in leadership and Father Giraldo and, in particular, Eduar, served as vivid examples of what a leader should be. All the members of the first Internal Council [the governing board of the PCSJA] received formal and informal training and guidance from these two organizations. Moreover, they acted as leaders themselves. Giraldo was a member of the Internal Council for several years and still today, 20 years after its creation, accompanies and advices the PCSJA in crucial matters. Eduar eventually moved to San José and became a full member of the community, accompanying and helping lead the process until 2012 when he died. During the initial stages of mobilization, villagers relied heavily on them — and in particular on Eduar, who was regarded by most peasants as “highly knowledgeable” and “trustworthy” — to create an encouraging environment and give momentum to the process.¹⁹

As in San Carlos, these actors brought a normative commitment to nonviolence into the process. As it has been identified in multiple instances of collective action involving the support of external actors (Benford and Snow 2000), these actors implicitly (and sometimes explicitly) condition their support on basic levels of resonance. In the case of the PCSJA, a pivotal aspect in this resonance was a strong and clear commitment to nonviolent methods. Although not part of the standard narrative of the process of the PCSJA, some few informants told me that some villagers individually considered arming themselves to counter armed groups violence. However, when it came to moving from individual preferences to collective action, they noted that this idea was not pursued, not even openly discussed. When I asked why, they stressed that it was because of the centrality the Church, and in particular, of Javier Giraldo. One noted explicitly that with arms in the process, they would have lost their main ally [referring concretely to the CIJP].²⁰

None of the funding leaders reported having considered any form of violent resistance. They even noted that those who wanted to use violence had joined one of the factions in competition. In other words, recruitment by armed groups served as a sorting device selecting those who rejected violence into the noncooperation campaign. Members of the Church, the then mayor Gloria Cuartas and of other organizations that later came to support the process (such as FOR), explicitly stressed that a commitment to nonviolence was a precondition for them to back the campaign. This was largely an implicit condition. However, on occasions it became more explicit. As an interviewee who was directly involved in the humanitarian assistance provided by the Catholic nuns and the Red Cross told me, aid was conditional on embracing the moral commitments of the Church.²¹

Level (ii): The degree of confrontation

Both Urabá and Eastern Antioquia, the regions where the PCSJA and Joppaz emerged respectively, were deeply influenced by left-wing political movements throughout the 1960s, 1970s and 1980s. Both regions, in fact, are well known for having hosted strong trade unions, social and civil movements and left-wing political parties. However, while the PCSJA was heavily informed and influenced by this ideological history, the Joppaz campaign was not. I argue in this section that this difference had an impact on the form noncooperation took.

Joppaz emerged from a group of motivated high school students who were not bearers of any particular political ideology. Moreover, given past experiences of heavy repression, it did so in a municipality in which most people were disenchanted with left-wing mobilization “The more confrontational, the worst”, was one of the lessons than two decades of violent repression silencing social protest left in the municipality.²² In the late 1970s and during the 1980s, residents fiercely reacted against mega-infrastructural projects that were being advanced in the regions. They organized into what came to be known as the Civic Movement. Within this dynamic regional context of contentious collective action, the population of San Carlos was particularly strong and determined. For example, during an iconic regional civic strike organized in 1984, when all the other municipalities in the region decided to step back, San Carlos’ inhabitants went on striking for several more days, organizing street protests almost on a daily basis.

The experience of these years is one of successful mobilization as much as one of heavy repression. Already in 1978, during the first strike that was organized regionally, protesters were heavily repressed by the state forces. While in the initial encounters with repression no killings were involved, as political contention progressed and deepened over the years, disappearance and assassination of leaders became more and more frequent. Social protest was heavily criminalized, with activists being increasingly associated with the left-wing rebel groups that were gaining presence and salience in the region. Media reports of the time from Antioquia’s main news outlet (El Colombiano) reflect well this stigmatization: activists were commonly referred to as “subversives”, “rebels”, “insurgents”, and “anarchists”, the same labels used to make reference to the emerging guerrilla groups.²³ With this framing, the interests of the traditional political class in power aligned with those of the paramilitary groups that, even if not yet stationed in the

region at that time, had began to make sporadic incursions. The Death to Kidnappers (MAS), a paramilitary organization closely link to drug cartels created in the early 1980s, severely targeted activists, threatening and killing leaders and visible figures of the Civic Movement. Data collected by CINEP – a faith-based nongovernmental organization that keeps systematic track of human right violations in the country – shows that between January 1988 and October 1991 alone, 66 social and political activists were killed in the region. According to both my interviews and secondary sources (e.g., Olaya 2012), by the end of the 1980s and beginning of the 1990s, most leaders in the region had been killed or had fled.

Olaya (2012, 90), a historian native from San Carlos, notes that the experience of this cycle of protest constitutes “an unforgettable experience...recorded in the minds [of the inhabitants of San Carlos]”. While the testimonies I collected back this claim, they help specify it further. While those who participated in or witnessed this first wave of contention unequivocally recalled this experience and stressed its importance, when asked to elaborate on its legacies several noted that it taught them how costly and risky voicing discontent was. Without probing, some interviewees explicitly stated that this event constituted a serious disincentive to further engage in collective action (see, also, Olaya 2012, 133). Even people who were known among the community for having been active, and thus could have incentives to aggrandize this past experience, recognized that the lesson was tough and had negative consequences in term of subsequent collective action efforts. It left people “without energy and without soul”.²⁴

To be sure, San Carlos does have a very rich history of contentious political action, with very well organized movements (such as the Civic Movement), deeply influenced by left-wing political ideologies, and that managed to mount highly confrontational forms of contention. However, these experiences were also harshly represses and left among the population the feeling that collective action, at least when confrontational, was more risky than useful. Therefore, Joppaz organizers, cognizant of this and not having had any direct experience in these movements, to be successful in their campaign consciously tried their best to not evoke concrete political discourses and avoid involving political forces. As an organized explicitly told me, “[Joppaz] was not about politics, it was about stopping violence, building peace, taking spaces and kids out of the hands of war.”²⁵

Besides this subjective legacy of previous experiences of collective action, the role the Church played in the process of giving Joppaz shape further tempered the level of confrontation involved in the campaign. A member of the Church recounted that when Father Blandón was approached by Jaider and his classmates, he insisted that whatever they would do in response to the situation had to unite the community rather than further divide it. He noted that if they exacerbated existing divisions between residents and armed groups, they were going to create obstacles for other youths to defect armed groups and participate in the activities they would organize, compromising one of the main objectives of the whole campaing – fighting youth recruitment. Organizing activities that were not overly related to the dynamics of war and that were not confrontation vis á vis armed groups, such as playing board games or preparing community potlucks, would allow more people to jump on board, including youth at risk of being recruited and even those that had already joined the ranks.²⁶ Reflecting this position of the Church, as Jaider likes to put it, Joppaz ultimately was meant “to bring people together, to unite the community”.²⁷

In stark contrast, in San José, many of the villagers that set up the Peace Community were active militants of left-wing social and political movements and had been deeply socialized into a particular militant ideology. Beyond opting out of war, as residents of San Carlos aimed to do, the campaign in San José involved the creation of local institutions, such a local school and a system of fair trade for their produce. To put it in Gandhian terminology, unlike Joppaz, the PCSJA campaign embraced a “constructive program”, coupling noncooperation with community self-improvement and self-sufficiency, building structures that are alternatives to oppression and reflect the different values participants uphold (Gandhi 2001). Even if some respondents had embraced an outward discourse of being “apolitical” (most likely to avoid being stigmatized), their local institutions and form of organizing reflects well a left-wing political ideology and a sense of communal governance that is in line with the political ideals and modus operandi of the local organizations and political parties that influenced leaders and residents for several years prior to the creation of Peace Community.

More than three decades of multiple forms of social organization and mobilization in the village of San José left important legacies that were tapped into when the time to set up the Peace Community came. These legacies, that can be traced back to the process of colonization of the area in the late 1950s and early 1960s, favored the emergence of the PCSJA by providing leadership and other organizational resources Masullo 2017a. Additionally, it shaped the form the campaign took by providing ideological guidelines and content.

While in the field I was able to reconstruct the profile of some of the original settlers of San José, some of which played an important role in the process leading up to the creation of the PCSJA. From this exercise, I learned that most of them came from Liberal municipalities in the Department of Antioquia and many were in fact fleeing from Conservative violence.²⁸ Upon arrival, they had to organize and engage in demanding collective action, sometimes highly contentious, to colonized a largely inhabited and wild area. During the 1960s and 1970s, colonization advanced through complex land invasions. These were the first experiences of “working together” that the older residents of San José recall, experiences they claimed were the basis of further collective action in the village and that several villagers identify as the seeds of “resistance”.²⁹ As in El Salvador, where land occupations shaped subsequent insurgent collective action (Wood 2003, Chap. 1), in San José they sow the seeds of a highly organized and confrontational experience of civilian noncooperation.

Interviews with elders revealed that peasants were not alone in these collective efforts. Local organizations, with clearly defined political ideas, supported the process. Older villagers, those who took part in the process of setting the village of San José up, recall the Colombian Com- munist Party (PCC) and the National Association of Campesino Users (ANUC) as the most salient. While a radical faction within the latter promoted land invasions as a strategy to force agrarian reform, the former actively supported and guided the settlement of peasants and help them organizing in different forms, following the “communist” general guidelines.

Moreover, the accelerated pace of population growth that followed the expansion of the banana agro-industry (San José is located in the country’s “Banana Belt”) put new issues on the agenda and made peasant mobilization go beyond land invasions (Uribe de Hincapié 1992, 163). The peasantry began to demand more favorable working conditions and the provision of public services from both the State and the companies in the banana industry. Many residents of San José, even if not all working in the banana industry, got involved in workers’ marches, demonstrations and strikes. Moreover, as it was to be expected, some of the strongest trade unions in the country emerged in the region, some with clear links to the Communist Party. Unions became a vital catalyst of organization and mobilization, promoting a more-urban inspired set of left-wing ideas and goals, and stimulating a stronger oppositional attitude toward the state.³⁰

In the late 1970s and throughout the 1980s, social mobilization and community organization in San José entered into a whole new phase. Rebel groups, in particular the Popular Liberation Army (EPL) and the FARC, arrived in San José and, as they did pretty much in the entire Urabá region, permeated and radicalized most of the already existing social and political movements, including parties, trade unions, and the Local Junta Council (JAC) (Carroll 2011; Bejarano 1988; Ramírez Tobón 1993; García 1996). They capitalized on the existing local organizations to mobilize supporters and further promoted peasant organization in direct and indirect wars. In doing so, they managed to advance their insurgent ideals and agenda.³¹

Referring largely to the FARC, many elder residents of San José noted that, even if the penetration of the rebels into their organizational processes justified violent policing from the state and later on from the paramilitaries, it undoubtedly strengthened many of the existing organizations. In advancing their insurgent agenda, rebels not only lectured peasants on revolutionary ideals, but also exposed them to means of action that were not part of their repertoire of action. They ultimately promoted a much more oppositional attitude, both vis á vis the private sector (banana companies) and the state (Romero 2003, 170; Hernandez Delgado and Salazar Posada 1999, 32). This reading differs sharply from the recollections of older residents of San Carlos, who tend to blame rebel groups for spurring harsher repression from the state and paramilitary groups and for weakening civic mobilization in the region.

During this period, strong local communal organizations, such as the Cooperativa Balsamar, and political parties, such as the Patiotic Union (UP), were formed in the region and had strong influence in the social and political life of San José peasants. The UP, a national political party formed by the FARC within a larger coalition of left-wing movements and parties, became one of the main political forces in Urabá and, particularly in San José, harnessed wide popular support.³² Pacho, who was tasked by the UP to gear and monitor support for the party in San José, recalls that almost every leader in every hamlet – including many of those who ended up organizing the PCSJA – worked with the UP. It was hard to find a “single soul” in the village that was not voting for the UP or who was not participating in at least some the meetings the party used to organize, he highlighted.³³ Notorious community leaders, such as Bartolomé Catan˜o – the founder of the village of San José –, moved from leading the local JACs to (successfully) running for council posts within UP lists. Balsamar was an even more local and rural expression of communal organization. It emerged in the village of San José as an experiment to empower peasants and stimulate civic engagement around the production and management of cacao. While not openly oppositional as the UP, ideologically speaking it was clearly left-wing, upholding some of the organizational forms the Communist had exhibit in the region and embracing a strong sense of peasant autonomy.

Finding evidence for the influence of these experiences on the creation of the Peace Community is no easy task. Apart from direct testimonies, participation – especially “rank-and-file” in these movements leaves little trace and, fearing stigmatization and association with FARC villagers have reasons to avoid being linked to this contentious past. However, to my surprise, some leaders not only recognized their past in the PCC and/or the UP, but also linked it explicitly to their efforts to set up the experiment of Peace Community. For example, Consuelo, a recognized local leader of the PCSJA, noted:

I believe that the experience of having been a communist and part of the UP was a privilege...I was involved in everything: parent associations, the JAC, peasant associations, I helped organize trade unions or banana workers... Look, since I was in the Communist Party I developed the skills to organize people, to help those who suffer...[in the context of creating the PCSJA] we had to lead a whole new process and we did not know well what were we doing, but we had the experience in the PCC and the UP to build on.³⁴

While these organizations had disappeared or were not necessarily active when the community organized into noncooperation, many villagers – in particular elders and community leaders identified them as the building blocks of the Peace Community. In the words of Cuartas, the mayor of Apartadó when the PCSJA was created, villagers’ experience with the UP and Balsamar constitutes “a milestone for understanding the organization, the sense of defending the territory and of doing community work, and the practice of joint-decision making [of the PCSJA].”³⁵

All in all, unlike Joppaz where organizers were “new actors” and residents who were active in left-wing social movements had left of were disenchanted, in San José some of the same villagers that for many years were active in these movements and parties were directly involved in mounting noncooperation to respond to civil war violence. Not only the had the “know how” to organize the community and work together, but were also quite used to highly confrontational and high-risk forms of collective action.

Unlike the situation I found in San Carlos, where the legacies of a strong tradition of collective action pushed the youth to take a stand away from politics and to avoid giving any political content to their campaign of noncooperation, in San José the legacy of over 30 years of social and political mobilization in the form of peasant associations, unions, clandestine and legal political parties (and even rebel groups), left a legacy that was effectively tapped into by villagers to mount noncooperation. Even when many community leaders were killed or had left the village, people that actively participated in movements and parties in the 1970s and 1980s put their experience at the service of the Peace Community. Beyond acquired skills, these previous experiences of confrontational contentious action left shared beliefs about the importance and rewards of expressing dissent collectively to redress grievances. If residents in San Carlos lacked a sense of “collective efficacy” (Klandermans 2013), San José had it all. While respondent after respondent in San Carlos highlighted the costs of contentious collective action when describing past experiences, villagers in San José underscored how the memories of collective action incentivized them to find a collective response to the changing dynamics of war.

These legacies came along with particular political ideologies and normative commitments that shaped the attitudes and behaviors of the Peace Community. Compared to Joppaz, participants – in particular leaders – opted for a more overt and oppositional form of noncooperation. Rather than disguising their activities, the declared noncooperation publicly. While Joppaz activities were intended to be as inclusive as possible (even including youth that were joining armed factions) to avoid creating and strengthening cleavages, the Peace Community demarcated an area off limits for armed groups and broke radically their interactions with them. The different ideological input that informed these two campaigns also shaped the way both communities relate with the institutions of the state. While both campaigns refused to cooperate with the armed forces of the state, including both the Police and the Army, Joppaz remained more open and even willing (according to the accounts of their organizers) to coordinate activities with local government officials as long as the army was not involved. In contrast, PCSJA villagers saw the state with distrust and, on occasions, even as an opponent, just as they did in previous experiences of collective action. This position went as far as to, in early 2005, a full and radical brake with every institution of the state, even with agencies intended to support victims of the conflict.

Moreover, building on a strong tradition of communal governance, were community work and autonomy from the state was highly valued, the PCSJA established days devoted to community work (Thursdays at the time of this research) and developed their own institutions to regulate social life. They engaged in a campaign of noncooperation that, involving a ’constructive program’, gave them greater autonomy and reflected the political and social ideas they had been socialized into. In contrast, Joppaz restricted their action to activities aimed strictly at reactivating social interaction and recovering the community’s social fabric to protect themselves from further violence or, at least, counter its negative effects. In this sense, as Gutierrez Sanín and Wood (2014, 214) note, a particular set of more or less systematic ideas prescribed not only the form the strategies these two communities opted for when mounting noncooperation, but in the case of the PCSJA also provided blueprints for the institutions they set up for the realization of their objectives.

The table below summarizes this first paired comparison:

	JOPPAZ	PCSJA
Level i	Local Pastoral Social	Regional Diocese
Level ii	—	Left-wing mov.; parties & rural unions
Outcome	Nonviolent – Oblique	Nonviolent – Unilateral

Table 1: Paired Comparison 1

This comparison provides evidence for an effect of ideational factors on the form that non- cooperation takes. Additionally, it further shows that leaders, acting as political entrepreneurs, were the ones doing the work of translating ideas into practice, just as in other experiences of collective action studied by scholars interested in the role of ideas and emotions (Costalli and Ruggeri 2015; Ruggeri and Costalli 2017). Having compared one case in which there is some level of continuity in terms of leadership between previous experiences of collective action (in which people were socialized into left-wing oppositional ideologies) and noncooperation, to one in which organizers were new actors that had not been directly exposed to these ideological content, allows to see more clearly the role of ideational factors. In this sense, Joppaz is used as a “control” case for the argument.

Nevertheless, both cases vary considerably in terms of capacity for collective action. Given the way past experiences of collective action mapped into the process of the PCSJA and not in Joppaz, San José villagers enjoyed noticeable higher levels of organizational and mobilization capacity than residents of San Carlos. Therefore, it can be reasonably argued that villagers from San José went for a more confrontational (and more comprehensive) type of campaign not because the oppositional ideational content that informed the formation of the campaigns, but just because they had the needed capacity to do so. To explore this plausible alternative explanation, in the next section I compare the Peace Community with a campaign with comparable levels of capacity for collective action: the ATCC. This second paired comparison strengthens the confidence on the proposed ideational argument by showing that villagers from the ATCC, despite of having the mobilizing resources to mount a highly confrontational campaign such that of the PCSJA, they instead went for a less confrontational campaign of pacted noncooperation.

Paired Comparison II: PCSJA & ATCC

The experience of the ATCC lays in between oblique and unilateral types of noncooperation and represents yet another distinct type: pacted noncooperation. Unlike participants of Joppaz, ATCC villagers did engage with armed groups openly and directly; unlike the experience of the PCSJA, this engagement was based on dialogue and negotiation, rather than unilaterally declared.

In pacted forms of noncooperation dialogue and negotiation with armed groups are the distinctive feature in both the process by which these campaigns emerge and are sustained, defining the terms of interaction between civilians and combatants. While in unilateral forms the rules of the game are defined by civilians alone and imply a radical break in civilian–combatant interactions, in brokered forms civilians establish and maintain an open and constant channel of communication with armed groups. Action here is direct, but comes after some form of tacit or explicit agreement between the different actors involved or, at a minimum, a previous discussion of civilian intentions. The concrete mechanisms and institutions to advance noncooperation commonly result from bargaining processes between the parties involved and many times are jointly designed and implemented. Notwithstanding, these campaigns are premised on the non- negotiable choice of refusing to cooperate with any side. However, the fact that there is some sort of negotiation and that dialogue is constant, tempers the degree of direct confrontation between combatants and noncombatants. To maintain these dialogues going and ensure that armed groups stick to the agreements, villagers deliberately refrain from engaging in practices that could be seen as too confrontational.

The ATCC is an illustrative example of pacted noncooperation. In early 1987, militaries and paramilitaries together convened peasants of the village of La India – in El Carare region, north central Colombia – and gave them an unequivocal message. Villagers had about 15 days to choose between joining the paramilitaries (the Peasant Self-defense Groups of Magdalena Medio (ACMM)), siding with the rebels (FARC), leaving the region or being killed. Villagers recall this as “the ultimatum”. Despite of the fear this credible threat generated, a group of about 12 community leaders held an almost permanent ad hoc assembly for about three days to analyze each of these options. After carefully going through them, they came to the realization that there was another course of action: staying put without cooperating with any side. To do so, they created the ATCC. In the words of one of the founding fathers of the Association, in that assembly they concluded that “Neither everybody will die, nor people will go with an armed group. We are from here, we are colonos [colonizers, settlers], so they [armed groups] will respect us. We will be neutral...”³⁶

Instead of making a unilateral declaration of this choice, as the PCSJA did ten years later in San José, villagers from La India opted to address armed groups to inform them about their choice and try to get them to buy into the process. Peasants plainly expressed their determination to stay in their lands while not taking sides in the conflict. Their refusal to cooperate with armed groups was unequivocal. In the words of an ATCC peasant, “whoever calls himself ATCC has to declare himself neutral in regards to armed groups”.³⁷ However, the concrete from this took was negotiated with each of the different factions present in their territory. After several meetings, armed factions accepted civilians decision and agreed on a series of behaviors to keep civilians outside of the war. This was how noncooperation emerged in La India on May 21, 1987. The date corresponds to the first dialogue peasants held with FARC. Still today, 30 years after its creation, the ATCC governs the life of hundreds of peasants living in La India and surrounding hamlets.

Level (i): The choice for nonviolent methods

Introducing this second paired comparison provides additional leverage to the argument that normative commitments have an effect on the choice for nonviolent methods of action. Both the PCSJA and the ATCC rely exclusively on nonviolent methods of action. Nevertheless, the process by this outcome came about in the ATCC campaigns, as opposed to the PCSJA, allows to provide stronger evidence for the role of ideational factors. Analyzing processes, rather than focusing exclusively on variation in the outcomes, is in fact one of the strengths of paired comparisons (see, e.g., McAdam, Tarrow, and Tilly 2001)

Through the first comparison I showed that normative commitments, due to the role the Catholic Church played, were present in both campaigns and provided evidence suggesting that the Church ideational content shaped villagers choices. However, that first comparison did not allow me to go much further in that none of the cases serve as good counterfactual to explore what would have happened (in terms of form) had the Church (or nonviolent ideational forces) were absent. Organizers of the Joppaz campaign apparently never considered violence as an option. In the PCSJA, I only found scattered evidence suggesting that a couple of villagers privately did consider this option, but none of these villagers had leadership roles within the community and they themselves noted that the option was never discussed collectively.

The second comparison corrects for this. In contrast, in the case of the ATCC I did found clear evidence that some of their main leaders seriously contemplated arming themselves, but during the early mobilization stage they dropped this idea. Why did they do so? If my argument is right, I should find evidence showing that normative commitments (in the form, for example, of ideals of nonviolence) are part of the answer. While the ideal design would have included a negative case in which villagers opted for an armed form of noncooperation and normative commitments were clearly absent in the process, within-case variation in the the process by which the ATCC opted for nonviolent forms on action provides a second best way to explore whether the normative factors play any role in shaping the form of noncooperation.

After villagers of La India received the ultimatum, deciding that noncooperation was the way to go about was not an easy process. Not only because this was a particularly risky option given the credible threat from the paramilitaries, but also because leaders had diverging views of how the process should look like. One of the areas in which they differed was precisely that of the methods of action. Whether the campaign was going to be armed or unarmed was an issue of intense discussion. In what follows, I argue that normative commitments coming from the Church – this time the Adventist Church – were critical in overcoming these differences and in pushing some influential leaders to drop the idea of armed resistance and stick exclusively to nonviolent methods of action.

The Adventist church settled in La India in the early 1960s, during the early years of colonization of the village. Since then, it has been a focal point for community congregation and an engine for communal work. Julio was one of the first settlers of what eventually became La India and the person who established the Adventist Church on those lands. “[T]he Adventist Church began in the middle of the jungle”, he noted. Given the multiple challenges of settling in the wild, the Church ended up deeply involved in the process of organizing the population to effectively colonize the area. What the PCC did in San José was to a large extent taken

over by the Adventist Church in La India. The colonization process forged some of the community leadership that most of my respondents reported as indispensable for the emergence of the ATCC back in the late 1980s. Julio, for example, became a natural community leader by working towards the construction of the only road connecting La India and Cimitarra – the closest municipal capital –, as well as by being active in the struggle to bring electricity to the village. This gave him and the Adventist Church a privileged position within the emerging community. Not only he ended up acting as the president of the first JAC in La India, but also it were the Adventists who accounted for the majority of the Board members.³⁸

Adventists and non-Adventists alike recognize that the church became a central (if not, the central) associational space that allowed for preference convergence and that encouraged people to mobilize in the process of mounting noncooperation. Without any probing, Julio made clear that he deliberately used the Church to shape people’s preferences and beliefs regarding participation in the war and, in particular, in relation to the use of violence.³⁹ He, as well as some of his closest followers, noted that while addressing the population, they deliberately linked concrete teachings of the Bible to the situation the village was going through. Almost every resident I spoke to recalls that the Adventist message has been always one of peace. They noted that the Adventists have always stressed, directly and indirectly, that by getting involved with armed factions one ends up promoting violence in one way or another. This had been one of their key message even years before the creation of the ATCC.⁴⁰

Julio remembers well the day he discussed the issue of methods of action with Josué, the most salient leader of the community, founding father of the ATCC and its first president. They were in the hamlet of Santa Rosa (few kilometers South of la India). Josué told Julio “man, we need to do something because this people will finish us. Let’s arm ourselves as other people are doing in Puerto Boyacá [a neighboring Department infamous for the emergence of various self-defense groups many of which later evolved as paramilitary armies].” Julio was not surprised, as he knew that Josué and others had entertained that idea for some time. His reply was unequivocal: “we [the Church] are here to collaborate, but not with arms. The Adventist people don’t kill’.”⁴¹

This was confirmed by David, another historical community leader and founding father of the Association. He confirmed that arming themselves was an idea that Josué and others in the village had considered for a long time and noted that they were just waiting for the right moment to propose it. In their view, the ultimatum presented them with an opportunity to do so. They did not consider joining any of the existing factions, but creating a self-defense group of their own was discussed as a possible course of action. David himself recognized that he was not fully against this option and noted that was ready to put at the service of it his previous experience with liberal guerrillas in the region he was originally from, before fleeing to La India. This is not necessarily the part of the processes that members stress the most when reconstructing the process by which the ATCC was created. In fact, the first time I spoke with David none of this was mentioned.⁴²

However, David, Josué and pretty much all the other community leaders seemed to be well aware that to be successful in whatever they would organize as a response to the ultimatum, armed or unarmed, they needed almost universal participation from the community. For this, given the social structure of the community of La India, the full support of the Adventist Church was indispensable. As a consequence, those who entertained the idea of forming a self-defense group, including Josué, eventually bent their arms and dropped the idea. They were cognizant that, as recent theories of nonviolent action have evidenced (see Chenoweth and Stephan 2011), armed opposition would set higher barriers (some of them moral) for participation and that it would likely preclude the participation of the Adventists. Given the centrality of the Church as a center for social aggregation and its tradition in community organizing and communal work in the village, they exerted influence not only among the formal members of the Church (a substantial portion of the residents of the village, in any case), but over residents at large.

It is hard to tell whether in the absence of the Adventist Church we would have observed armed noncooperation instead. However, compared to the case of the PCSJA (and even more, the Joppaz), there are good reasons to believe that this could have a possible outcome. First of all, those who contemplated the idea (as opposed to what seems to have been the case in San José) were historical community leaders and were among those who took an active role in mounting noncooperation in the village. Second, evidence indicated that some had some sort of previous experience that could have been put at the service of armed resistance. Moreover, according to David for example, securing access to arms would have not been that hard. In other words, not only armed resistance was cognitively available (see Kalyvas and Sanchez-Cuenca 2005), but also the material conditions to do so (at least as perceived by some villagers) were in place. This helps to cast some doubt on two of the more plausible alternative explanations of why violence did not emerge.

Finally, providing direct support to the proposed ideational argument, evidence suggests not only that villagers in La India had embraced normative commitments against violence over the years before the creation, but that the Adventists – in particular Pastor Julio, the founder of the Church and a pivotal community leader, deliberately used the leverage of the Church to shape the type of collective response the community of La India was crafting.

Level (ii): The degree of confrontation

In the first paired comparison I showed that, unlike the case of Joppaz were campaign organizers had not been socialized into oppositional ideologies and political ideas were deliberately kept at bay, in San José oppositional ideologies had long influenced villagers – including some of those who later helped organized the Peace Community – and provided guiding principles in previous experiences of collective action. Given this different trajectories, oppositional ideologies informed and shaped the form that noncooperation took in San José, but not in San Carlos. In this second paired comparison, I bring the case of the ATCC, which emerged in a very similar ideational landscape to that of the PCSJA, with particularly strong regional left-wing movements and parties in the years preceding the emergence of noncooperation.

From the 1960s, and in particular in the first half of the 1970s, left-wing political parties and oppositional movements were very popular in the Carare region, especially in the municipality of Cimitarra, where the Association has a big portion of its area of influence. As in San José, the PCC was a central political actor and different left-wing organizations, such as the Liberal Revolutionary Movement (MRL), the National Popular Alliance (ANAPO) and the National Opposition Union (UNO), enjoyed wide support among the population. As an indication of the strength of this movements, in 1976 the UNO, a coalition of the Communist Party and other left- wing political movements, got over 60% of the votes in Cimitarra’s municipal elections, winning six of the ten seats of the Municipal Council (Equipo Nizkor 2001). This took place precisely when the left was beginning to loose ground in the region and traditional political parties were recovering ground.

Throughout the 1960s and part of the 1970s, these left-wing organizations, some of which embraced strongly oppositional ideologies, worked towards organizing the population and stimulating communal work. As in San José, these initial experiences of collective action left important legacies that enhanced importantly the population’s capacity to mobilize. Moreover, the work advanced by these movements, along with the the political work done by FARC in the 1970s and part of the 1980s, was central in promoting socialist ideals and in stimulating a contentious spirit among the population of the region (Grupo de Memoria Histórica 2011, 81). If villagers in both San José and La India were exposed to similarly strong left-wing movements and parties that embraced and promoted similarly (some times the same) oppositional ideologies in the decades before the emergence of noncooperation, why do the PCSJA and the ATCC differ in terms of how confrontational their noncooperation campaigns were?

In this section I argue that in La India these ideologies did not inform the creation of the ATCC to the same extent they did in the PCSJA. I found evidence of forces both blocking and tempering this “contentious spirit”. Concretely, repression blocked the advancement of strong left-wing oppositional ideologies, and the general approach advocated by the Adventist Church, besides pushing the campaign towards nonviolent methods of action, tempered the level of confrontation involved. The fact that both communities enjoyed comparable levels of organizational capacity and that both opted for deeply comprehensive campaigns, casts doubt on the argument that variation is driven by capacity. Something the previous paired comparison could not do.

The electoral success of the UNO in the more urban areas of Cimitarra, and the increasing control of FARC in the rural areas, led to a highly repressive response. First, by the state, in the mid/late 1970s; and later, during the 1980s, by paramilitary groups coming from neighboring areas. As a result, some years before the creation of the ATCC, the political left had almost fully disappeared from the political landscape. Even Cimitarra, where the strength of these movements persisted over a longer period, witnessed what residents still today call the “extermination of the left”.⁴³ But it was not repression per se what prevented these ideologies to permeate the ATCC. For that matter, left-wing political movements in San José were also violently repressed and, as in the case of the UP, entire forces were exterminated. What blocked the advancement of oppositional ideologies and prevented them to find their way into the ATCC was the subjective effects of repression, operating both at the level of leadership and the population at large. In this regard, the experience of the ATCC is in fact closer to that of Joppaz.

During the initial months of harsh repression, leaders of the most salient political movements were selectively targeted. Some were killed and many others were forced to leave the region. While this did not leave La India leaderless – available leadership in fact played a central role in the emergence of the ATCC (Masullo 2017a) –, it did sort out and transform the leadership that remained. As David – a highly politicized person himself – noted, the strongest left-wing influences had fled the area before villagers began to organize the Association and the leaders who remained were largely those who had been somehow less involved in politics (such as those more closely related to the Adventist Church). Those who were more “into into political thing” (metidos en la cosa política) and were “more rebellious”“ (revoltosos)”, either left or, like himself, were left with “no enthusiasm [sin gana] to get into politics”. He recognizes that he still feels deeply identified with left-wing ideologies, like Marxism and Leninism and reads about these ideas, he does not “bring that into meetings and things of the ATCC”.⁴⁴

When agents of repression were not longer able to further select victims in light of their visible position in opposition groups, they (in particular the paramilitaries) gradually move towards collective forms of targeting, using (quite loose) group identifiers as proxies for political loyalty.⁴⁵ Basically those who lived in areas where FARC had strong influence were taken as part of the insurgent base of support and, consequently, residents of entire hamlets became targets of violence. Through this form of targeting many peasants learnt that being associated with FARC, or any organized expression of the left, was indeed a risky enterprise. As a peasant who is today a central leader of the ATCC told me:

“...it was better to be seen as apolitical, this is why we created a peasant association not a political movement. We don’t follow any color. We are not part of any political or ideological group. This allows for dialogue [with armed groups] and does not lead us to take radical positions.⁴⁶”

In fact, when explicitly asked about the differences between the ATCC and the PCSJA, this same leader, without any probing, noted that while they were apolitical, the Peace Community was “very radical, very left-wing”. Then he added, “[t]hey are more radical than us, they are different because they have an ideological color that we do not have.”⁴⁷

The wave of repression that peasants of La India experienced in the late 1970s and early 1980s was very successful in achieving what Anderson (1988), in the context of the authoritarian regimes of the Southern Cone of the 1070s, termed “subjective traumatization”. It made left oppositional ideologies a taboo, something that people prefer not to appeal to, let alone embrace. It is hard to tell why harsh repression did not have the same effect in San José. My evidence does not provide a definitive answer to this puzzle, but points to some plausible explanations. The life histories of members of both campaigns, and the profiles of leaders that I was able to reconstruct in the field, clearly suggest that while various leading figures of the PCSJA were active participants in several of the left-wing oppositional movements that were active in San José, many of those who pushed the ATCC process forward did not militate or were not as actively involved in the movements that emerged in their region. This could have been the case for several reasons. In El Carare, unlike Urabá, some of the most influential oppositional movements were mostly based and active in urban areas – for example, the municipal council of Cimitarra. In addition, while the ATCC area was inhabited largely by “pure campesinos” (in the sense of people that work and live of the land), many of San José residents, while still campesinos, had closer links with the banana agro-industry. This exposed directly to the forms of organizing, protest repertoires, and left-wing ideologies that were promoted by some of Colombia’s strongest and, at the time, most oppositional trade unions.

Besides repression, the Adventist Church also played its part in making the ATCC less confrontational. Since the very beginning of the process, Pastor Julio pushed towards a non confrontational take towards armed actors. For example, he stressed the importance of not portraying armed factions as their enemies and, instead, pushed for recognizing them as valid interlocutors. He emphasized the importance of relying on constant and transparent dialogues with both the guerrillas and the paramilitaries. These ideas permeated so strongly into the creation of the Association, that when I explicitly tried to discus the issue of ideology with members, more than one replied – without any probing – that ATCC’s ideology is dialogue.⁴⁸

A very illustrative example of how this approach was evidently present in origins of the ATCC, comes from what many consider the foundational moment of the Association. That is, the first meeting villagers held with FARC’s Central High Command. On that occasion, Josué opened the conversation with the following words:

Today we want to reach a sincere and unanimous agreement with the FARC and then we will seek the same with the military and the paramilitary groups ... We simply seek to live in peace and work, we are not your enemies. This is not a meeting against the guerrilla, but against all those who violate our rights.⁴⁹

These words are particularly telling, given that Josué was perhaps the most confrontation leader in the entire community and on who, in fact, considered armed resistance as a response. The fact that even a person like Josué, who was not a member of the Church, embraced these ideas Gives a good indication of the strength of these ideational forces.

Besides words, the modus operandi of the ATCC reflects well the approach promoted by the Adventist Church. Most notably, villagers have never stoped interacting with armed groups. On the contrary, to facilitate the advancement of the pacted process, they opted to tap into existing ties between civilians and combatants, taking advantage of the embeddedness of armed groups (especially of FARC) within the community. In the initial years, to get the dialogues flowing, the Association even identified some members with this type of ties with armed groups and named them mediators. As a peasant put it, “we need to have a relationship with the armed actor...To dialogue one needs to take advantage of many things, [such as] of this neighbor having a good relationship [with an armed group], of this commander being the uncle of this or that person...”⁵⁰

This less confrontational approach, largely promoted by the Church, sets the ATCC in stark contrast to the PCSJA. Not only it led to a pacted process – as opposed to a unilateral declaration, but it also had practical implications in the way both campaigns have dealt with similar issues.

To start with, while in San José armed groups are banned to enter the hamlets where members of the Community live, in the ATCC’s area of influence armed groups have the right to transit and even stay over as long as they do not use members’ houses as shelter, enter spaces designated for the exclusive use of the ATCC, and/or take advantage of services provided by and for the Association. This included, for example, a communal shop located in La India and long-tail boats used to transport people from hamlet to hamlet through the Carare river.

Similarly, while since its creation the PCSJA has been engaged in a restless and constant effort to publicly denounce violations and military activity, blaming and shaming armed groups both nationally and internationally, the ATCC has tried to avoid doing so – at least in an overt, explicit manner. For the Peace Community, “fighting against” impunity is one of the central duties of each of its members and is even stated in the banners that designate the Community’s area of influence (which are largely addressing armed groups). In stark contrast, the ATCC agreed to, before making public denunciations, address first the high commanders of each faction to clarify events or, at most, address the judicial branches of the state (rather than the national and international media).

The following testimony is illustrative of how the ATCC feels about this issue: “... I am convinced that denouncing does not build peace. By shouting one does not fix the issues of this country. When you shout, you provoke people.”⁵¹ Even when three of the main leaders of the Association (including Josué) were massacred in 1990 – the toughest assault on the campaign in its 28 years of existence –, the noted that “Instead of making public accusations of the killings of our leaders, we need to intensify our efforts to get closer to those who declare to be our enemies in order to demonstrate them that, in practice, for us no one is an enemy.”⁵² This take on denunciation is clearly in line with the ideas promoted by the Adventist Church and explicitly preached by Pastor Julio.

Finally, what is perhaps the most important institution created by the ATCC, which recent evidence shows that has saved a considerable number of lives (Kaplan 2013b), rests on this continued interaction and dialogue. When the ATCC was created, civilians and armed groups pacted that when there were allegations of a member of the Association collaborating with an armed faction, before killing that person, commanders will reach the Association and give it the opportunity to investigate whether the allegations were true. If the Association would find that the person was not implicated (and, therefore, was respecting the norms of behavior of the ATCC), the ATCC board would pass the evidence to the commanders and the life of that person would be respect. On the contrary, if the person was found guilt by the Association, he or she would have to either reform his/her behavior (if armed actors accepted this solution) or leave the community. Without a less confrontational approach, that allows for constant interaction and fluid dialogue, such an institution would be impossible.

In sum, while both noncooperation campaigns emerged in areas were left-wing organizations (both legal and illegal, unarmed and armed) had historical presence and gained noticeable strength, oppositional ideologies did not inform the ATCC campaign to the extent they did in the PCSJA. In stark contrast to the experience of San José, some of the leaders who were vital for the emergence of the ATCC were not as strongly permeated by the available oppositional ideologies, and the existing associational spaces on which villagers of La India drew to set up the ATCC (such as the Adventist Church) were not as tightly linked to left-wing oppositional movements as they were in San José (such as the Balsamar cooperative). Moreover, the concrete ideational content of the Adventist Church, which was at the core of the emergence of the ATCC, emphasized constant and transparent dialogues and ruled out unilateral actions precisely to avoid confrontation. As an outcome, despite of having comparable levels of collective action capacity and, therefore, having the mobilizing and organizational resources to mount complex and sophisticated campaigns – the ATCC engaged in a pacted campaign of noncooperation that was noticeably less confrontational than the unilateral campaign of the PCSJA.

The table below summarizes the second paired comparison:

	PCSJA	ATCC
Level i	Regional Diocese	Adventist Church
Level ii	Left-wing mov.; parties & rural unions	Urban Left-wing mov.
Outcome	Nonviolent – Unilateral	Nonviolent – Brokered

Table 2: Paired Comparison 2

Conclusion

This paper underscores that ideational factors are central in shaping the behavior of civilians living in warzones, just as they are for armed groups. By focusing on a collective and risky enterprise such as noncooperation, I provide evidence that civilians indeed “use ideas when taking literally life and death decisions” (Gutierrez Sanín and Wood 2014, 213). While dealing with an actor that has received comparably less attention in the emerging literature on ideology in civil war, and covering an under-explored topic in the civil war literature – civilian noncooperation – , this paper complements and expands existing findings on how ideational factors play a role in the context of war and links this literature with the growing research on civilian agency.

Beyond civil war scholarship, this paper makes a contribution to other fields that, while dealing with similar issues, have advanced with little dialogue. The arguments advanced here are in line, for example, with recent work in civil resistance that shows that ideological variation shaped the form of protest in the Middle East (Asal et al. 2013). Similarly, it yields additional support to social movement scholarship that has identified the role of the Catholic Church and its normative commitments in shaping contentious collective action in various settings, from left- wing mobilization in Latin America (Peterson 1996), to struggles for civil rights in the United States (Morris 1986) and resistance against communism in Hungary (Wittenberg 2006). In this sense, this paper shows how a unified framework for the study of contentious politics can richly advance our knowledge of similar phenomena taking place is very different settings (McAdam, Tarrow, and Tilly 2001; Tilly and Tarrow 2015).

Finally, methodological speaking, this paper supports the claim that, with a close examination of concrete processes and a purposively designed micro-comparative structure – based on data collected via immersive fieldwork – we can in fact trace the effect of ideas (Jacobs 2014; Bennett and Checkel 2014). This approach allowed me not only to observe “ideas in action”, but also opened the door to explore whether ideational factors have an effect that is independent from material or situational forces. In trying to do so, this paper shows the benefits of expanding the temporal and spatial scope of analysis, looking beyond the concrete moment in which crucial mobilization choices were made and the main actors involved in the interaction, that is, the “usual suspects” of mobilization (in this case, the civilian–armed groups dyad). If there is any evidence for an exogenous role of ideas on a given campaign of collective action, this is likely to be found in the past history of communities that pre-dates the concrete choices that we study (i.e., the pre-mobilization stage. See, e.g. Staniland (2014), O’Connor and Oikonomakis (2015), and Ruggeri and Costalli (2017)) and/or by looking at the larger set of actors that constitute the strategic field of contention (Fligstein and McAdam 2015; Jasper 2004). Echoing the most recent advancements in process tracing (Lyall 2014; Jacobs 2014; Bennett and Checkel 2014), further work on the roles of ideas and emotions in collective action in civil war and other expression of contentious politics is likely to benefit from these methodological insights.

References

Anderson, Perry. 1988. “Dictadura y Democracia en America Latina”. In Democracia y social- ismo: La lucha democrática desde una perspectiva socialista, 43–66. Buenos Aires: Editorial Tierra del Fuego.

Aparicio, Juan Ricardo. 2009. “La ’Mejor esquina de Suramérica’: Aproximaciones etnográficas a la protección de la vida en Urabá”. Antípoda 8 (Enero-Junio): 87–115.

Arjona, Ana. 2017. “Civilian Cooperation and Non-Cooperation with Non-State Armed Groups: The Centrality of Obedience and Resistance”. Small Wars & Insurgencies 28, no. 4 (): 755– 778.

———. 2015. “Civilian Resistance to Rebel Governance”. In Rebel Governance, ed. by Ana Arjona, Nelson Kasfir, and Zachariah Cherian Mampilly, 180–202. Cambridge: Cambridge University Press.

———. 2016. Rebelocracy: Social Order in the Colombian Civil War. Cambridge: Cambridge University Press.

Asal, Victor, and R. Karl Rethemeyer. 2008. “Dilettantes, Ideologues, and the Weak: Terrorists Who Don’t Kill”. Conflict Management and Peace Science 25, no. 3 (): 244–263.

Asal, Victor, et al. 2013. “Gender ideologies and forms of contentious mobilization in the Middle East”. Journal of Peace Research 50, no. 3 (): 305–318.

Balcells, Laia. 2017. Rivalry and Revenge: The Politics of Violence during Civil War. Cambridge: Cambridge University Press.

Balcells, Laia, and Stathis N. Kalyvas. 2015. “Revolutionary Rebels and the Marxist Paradox”.

Duke University & Yale University.

Bejarano, Ana María. 1988. “La violencia regional y sus protagonistas: el caso de Urabá”. Análisis Político 4:41–68.

Benford, Robert D., and David A. Snow. 2000. “Framing Processes and Social Movements: An Overview and Assessment”. Annual Review of Sociology 26 (1): 611–639.

Bennett, Andrew, and Jeffrey T. Checkel, eds. 2014. Process Tracing: From Metaphor to Analytic Tool. Cambridge: Cambridge University Press.

Carroll, Leah Anne. 2011. Violent Democratization: Social Movements, Elites, and Politics in Colombia’s Rural War Zones, 1984-2008. South Bend: University of Notre Dame Press.

Checkel, Jeffrey T. 2017. “Socialization and violence: Introduction and framework”. Journal of Peace Research 54, no. 5 (): 592–605.

Chenoweth, Erica, and Maria J. Stephan. 2011. Why civil resistance works? The strategic logic of nonviolent conflict. New York: Columbia University Press.

Costalli, Stefano, and Andrea Ruggeri. 2015. “Forging political entrepreneurs: Civil war effects on post-conflict politics in Italy”. Political Geography 44 (): 40–49.

Drake, C. J. M. 1998. “The role of ideology in terrorists’ target selection”. Terrorism and Political Violence 10, no. 2 (): 53–85.

Elster, Jon. 2007. Explaining social behavior: more nuts and bolts for the social sciences. Cambridge: Cambridge University Press.

Equipo Nizkor. 2001. Colombia Nunca Mas. Crímenes de Lesa Humanidad 1966 - 1998 [Zona XIV. Tomo 1. Capítulo II: Cimitarra]. Equipo Nizkor.

Fligstein, Neil, and Doug McAdam. 2015. A Theory of Fields. Oxford University Press.

Fujii, Lee Ann. 2011. Killing Neighbors: Webs of Violence in Rwanda. Ithaca: Cornell University Press.

Gandhi, Mohandas Karamchand. 2001. Non-Violent Resistance. Reprint edition. Mineola, N.Y: Dover Publications.

Garcia, Alejandro. 1996. Hijos de la violencia: campesinos de Colombia sobreviven a ”golpes” de paz. Madrid: La Catarata.

García, Clara Inés. 1996. Urabá. Región, actores y conflicto. Bogotá: Gente Nueva Editorial.

Goodwin, Jeff. 2007. “”The Struggle Made Me a Nonracialist”: Why There was so Little Terrorism in the Antiapartheid Struggle”. Mobilization: An International Quarterly 12, no. 2 (): 193–203.

Grupo de Memoria Historica. 2011. San Carlos. Memorias del éxodo en la guerra. Bogotá: Comision Nacional de Reparacion y Reconciliacion CNRR.

Grupo de Memoria Histórica. 2011. El orden desarmado. La resistencia de la asociación de trabajadores campesinos del Carare (ATCC). Bogotá: Comisión Nacional de Reparación y Reconciliación (CNRR).

Gutierrez Sanín, Francisco, and Elisabeth J. Wood. 2014. “Ideology in civil war Instrumental adoption and beyond”. Journal of Peace Research 51, no. 2 (): 213–226.

———. 2017. “What Should We Mean by “Pattern of Political Violence”? Repertoire, Targeting, Frequency, and Technique”. Perspectives on Politics 15, no. 1 (): 20–41.

Hallward, Maia, Juan Masullo, and Cécile Mouly. 2017. “Civil Resistance in Armed Conflict: Leveraging Nonviolent Action to Navigate War, Oppose Violence and Confront Oppression”. Journal of Peacebuilding & Development 12, no. 3 (): 1–9.

Hancock, Landon E., and Christopher R. Mitchell, eds. 2007. Zones of peace. Bloomfield, CT: Kumarian Press.

Hernandez Delgado, Esperanza. 1999. “Comunidad de Paz de San José de Apartadó”. In Con la esperanza intacta. Experiences Comunitarias de Resistencia Civil No Violenta, ed. by Esperanza Hernandez Delgado and Marcela Salazar Posada, 48–92. Bogotá: Oxfam-GB.

———. 2004. Resistencia civil artesana de paz: Experiencias indigeneas, afrodescendientes y campesinas.

Bogotá: Editorial Pontificia Universidad Javeriana.

Hernandez Delgado, Esperanza, and Marcela Salazar Posada, eds. 1999. Con la esperanza intacta. Experiences Comunitarias de Resistencia Civil No Violenta. Bogotá: Oxfam-GB.

Hobsbawm, Eric. 1973. “Peasants and politics”. Journal of Peasant Studies 1, no. 1 (): 3–22.

Jacobs, Alan M. 2014. “Process Tracing the Effects of Ideas”. In Process Tracing: From Metaphor to Analytic Tool, ed. by Andrew Bennett and Jeffrey T. Checkel, 41–73. Cambridge: Cambridge University Press.

Jasper, James M. 2004. “A Strategic Approach to Collective Action: Looking for Agency in Social Movement Choices”. Mobilization 9 (1): 1–16.

Jentzsch, Corinna. 2014. “Militias and the Dynamics of Civil War”. PhD thesis, Yale University.

Johnson, Chalmers A. 1962. “Civilian Loyalties and Guerrilla Conflict”. World Politics 14 (4): 646–661.

Kalyvas, Stathis N. 2015. “Rebel Governance during the Greek Civil War, 1942-1949”. In Rebel Governance, ed. by Ana Arjona, Nelson Kasfir, and Zachariah Cherian Mampilly, 179–209. Cambridge University Press.

———. 2006. The Logic of Violence in Civil War. Cambridge: Cambridge University Press.

Kalyvas, Stathis N., and Ignacio Sanchez-Cuenca. 2005. “Killing Without Dying: The Absence of Suicide Missions”. In Making sense of suicide missions, ed. by Diego Gambetta, 208–232. Oxford: Oxford University Press.

Kaplan, Oliver. 2013a. “Nudging Armed Groups: How Civilians Transmit Norms of Protection”.

Stability: International Journal of Security and Development 2, no. 3 ().

———. 2013b. “Protecting civilians in civil war The institution of the ATCC in Colombia”. Journal of Peace Research 50, no. 3 (): 351–367.

———. 2017. Resist War. How Communities Protect Themselves. Cambridge: Cambridge University Press.

Klandermans, Bert. 2013. “Collective Efficacy”. In The Wiley-Blackwell Encyclopedia of Social and Political Movements. Blackwell Publishing Ltd.

Kuran, Timur. 1995. Private truths, public lies: the social consequences of preference falsification.

Cambridge: Harvard University Press.

Lyall, Jason. 2014. “Process Tracing, Causal Inference, and Civil War”. In Process Tracing: From Metaphor to Analytic Tool, ed. by Andrew Bennett and Jeffrey T. Checkel, 186–208. Cambridge: Cambridge University Press.

Mampilly, Zachariah Cherian. 2011. Rebel rulers : insurgent governance and civilian life during war. Ithaca: Cornell University Press.

Masullo, Juan. 2017a. “A Theory of Civilian Noncooperation with Armed Groups. Civilian Agency and Self-protection in the Colombian Civil War”. Doctoral Dissertation, European University Institute.

———. 2018. “Civilian Noncooperation as a Source of Legitimacy. Innovative Youth Reactions in the Face of Local Violence”. In Local Peace Building and Legitimacy, ed. by Christopher R. Mitchell and Landon E. Hancock. London: Routledge.

———. 2017b. “Refusing to cooperate with armed groups. Civilian Agency, Noncooperation and selfprotection in civil war”. In OCV Seminar, Yale University. New Haven, CT.

Masullo, Juan, and Francis O’connor. 2017. “PKK Violence against Civilians Beyond the Individual, Understanding Collective Targeting”. Terrorism and Political Violence.

McAdam, Doug, and Sidney G. Tarrow. 2000. “Nonviolence as Contentious Interaction”. PS: Political Science and Politics 33, no. 2 (): 149–154.

McAdam, Doug., Sidney G. Tarrow, and Charles. Tilly. 2001. Dynamics of contention. Cambridge: Cambridge University Press.

McAdam, Doug, Sidney Tarrow, and Charles Tilly. 2008. “Methods for Measuring Mechanisms of Contention”. Qualitative Sociology 31, no. 4 (): 307–331.

Morris, Aldon D. 1986. The Origins of the Civil Rights Movement: Black Communities Organizing for Change. New York: Free Press.

O’Connor, Francis Patrick, and Leonidas Oikonomakis. 2015. “Preconflict Mobilization Strategies and Urban-Rural Transition: The Cases of the PKK and the FLN/EZLN”. Mobilization: An International Quarterly 20, no. 3 (): 379–399.

Olaya, Carlos. 2012. Nunca más contra nadie. Medellín: Cuervo Editores.

Petersen, Roger Dale. 2001. Resistance and Rebellion. Lessons from Eastern Europe. Cambridge: Cambridge University Press.

Peterson, Anna L. 1996. Martyrdom and the Politics of Religion: Progressive Catholicism in El Salvador’s Civil War. Albany: State University of New York Press.

Ramírez Tobón, William. 1993. “Estado y crisis regional: el caso de Urabá”. Análisis Político

20:23–38.

Reiniciar. 2006. Historia de un Genocidio. El exterminio de la Unión Patriótica en Urabá. El Plan Retorno. Bogotá: Gente Nueva Editorial.

Restrepo, Gloria Inés. 2005. “Dináicas e interrelaciones en los procesos de resistencia civil. Estudio de caso comparado de los procesos de resistencia civil organizada de la Asociación de Trabajadores Campesinos del Carare y la Comunidad de Paz de San José de Apartadó”. PhD thesis, Universidad Nacional.

———. 2006. “Dinámicas e interacciones en los procesos de resistencia civil”. Revista Colombiana de Sociología 27:169–202.

Roldán, Mary. 2002. Blood and Fire: La Violencia in Antioquia, Colombia, 1946-1953. Durham: Duke University Press Books.

Romero, Mauricio. 2003. Paramilitares y Autodefensas, 1982-2003. Bogotá: Instituto de Estudios Políticos y Relaciones Internacionales.

Ron, James. 2001. “Ideology in Context: Explaining Sendero Luminoso’s Tactical Escalation”.

Journal of Peace Research 38, no. 5 (): 569–592.

Ruggeri, Andrea, and Stefano Costalli. 2017. “Emotions, Ideologies and Armed Mobilization”. Baltimore.

Schelling, Thomas C. 1960. The Strategy of Conflict. Reprint edition. Cambridge, Mass: Harvard University Press.

Sharp, Gene. 1973. The Politics of Nonviolent Action (Part 2) ”The Methods of Nonviolent Action”. Vol. 2. Boston: Porter-Sargent.

Shesterinina, Anastasia. 2016. “Collective Threat Framing and Mobilization in Civil War”. The American Political Science Review; Washington 110, no. 3 (): 411–427.

Snow, David A., and Danny Trom. 2002. “The Case Study and the Study of Social Movements”. In Methods of social movement research, ed. by Bert. Klandermans and Suzanne. Staggenborg, 146–172. Minneapolis: University of Minnesota Press.

Staniland, Paul. 2014. Networks of Rebellion: Explaining Insurgent Cohesion and Collapse.

Ithaca: Cornell University Press.

Steele, Abbey. 2017. Democracy and Displacement in Colombia’s Civil War. Itahca, NY: Cornell University Press.

Straus, Scott. 2015. Making and Unmaking Nations: The Origins and Dynamics of Genocide in Contemporary Africa. Itahca, NY: Cornell University Press.

Tarrow, Sidney. 2010. “The Strategy of Paired Comparison: Toward a Theory of Practice”. Comparative Political Studies 43, no. 2 (): 230–259.

Tarrow, Sidney G. 2007. “Inside Insurgencies: Politics and Violence in an Age of Civil War”. Perspectives on Politics 5 (3): 587–600.

Thaler, Kai M. 2012. “Ideology and Violence in Civil Wars: Theory and Evidence from Mozambique and Angola”. Civil Wars 14, no. 4 (): 546–567.

Tilly, Charles. 2008. Contentious performances. Cambridge: Cambridge University Press.

Tilly, Charles. 1999. “The trouble with stories”. In The social worlds of higher education: Handbook for teaching in a new century, ed. by Bernice A. Pescosolido and Ronald J. Aminzade, 256–270. Thousand Oaks: SAGE Publications.

Tilly, Charles, and Sidney G. Tarrow. 2015. Contentious Politics. (2nd Edition). New York: Oxford University Press.

Uribe de Hincapié, María Teresa. 1992. Urabá, ¿regio´n o territorio? Un análisis en el contexto de la política, la historia y la etnicidad. Medellín: Iner/Corpourabá.

Wittenberg, Jason. 2006. Crucibles of Political Loyalty: Church Institutions and Electoral Continuity in Hungary. New York: Cambridge University Press.

Wood, Elisabeth J. 2009. “Armed Groups and Sexual Violence: When Is Wartime Rape Rare?”

Politics & Society 37, no. 1 (): 131–161.

———. 2003. Insurgent Collective Action and Civil War in El Salvador. Cambridge: Cambridge University Press.

———. 2008. “The Social Processes of Civil War: The Wartime Transformation of Social Networks”.

Annual Review of Political Science 1 (1): 539–561.