Local non-violent strategies amid Guatemala’s post-accord violence: understanding the potential and limitations in poor urban neighbourhoods

ABSTRACT

The impact residents of violence-affected communities can have on addressing insecurity is underexplored, particularly amid criminal violence. In non-conflict contexts, can non-violent actions by committed individuals transform their violent environments given criminal groups’ social control and capacity to exert violence? Investigating two urban neighbourhoods in post-accord Guatemala, the article evidences how residents engage in violence disruption, breaking generalised self-protection strategies – silence, avoidance and displacement – to proactively address local violence. In doing so, residents can attain tangible security improvements and foster cohesion, social capital and informal leadership – facilitating further collective action. The article draws on 47 interviews, triangulated with police data and documentary evidence. The article contributes novel empirical evidence on the underexplored phenomenon of non-violent engagement with criminal groups. Conceptually, the article advances the emerging concept of violence disruption as a distinct form of agency and a useful framework to neither overlook nor overstate the impact of grassroots strategies.

KEYWORDS:

• Non-violence
• local
• agency
• post-accord
• Guatemala

Introduction

The causes and consequences of violence in Latin America have received extensive academic attention. However, residents of violence-affected communities have been understood narrowly, as being victims or perpetrators. The impact that residents may have, through non-violent means, in shaping how and whether violence happens in their communities remains underexplored. In non-conflict contexts, can non-violent actions by committed individuals transform their violent environments given criminal groups’ social control and capacity to exert violence?

The article investigates local non-violent strategies in post-accord Guatemala. After ambitious peace accords in 1996,¹ the country experienced decades of mass violence. Despite a tendency to explain post-accord violence as a by-product of macro-level factors (socio-economic marginalisation, inequality, low state capacity, conflict legacies), the country showcases vast subnational variations in homicide rates and manifestations of violence.² Moreover, while vast inequalities, a co-opted and/or absent state, and widespread presence of criminal groups persist, homicide rates have decreased sustainably between 2009 and 2021.³ These factors indicate the importance of better understanding causal mechanisms that are local in scope, including different forms of local non-violent action. Academic literature is progressing in this direction, documenting how civilians self-protect and resist violence, but so far focused on armed conflicts.⁴

The article presents novel evidence from urban Guatemalan case studies, expanding knowledge on grassroots agency amid post-accord criminal violence and advancing the concept of violence disruption. The article argues that residents of violence-affected communities are not restricted to perpetrator or victim roles; they engage in peaceful contention that impacts the dynamics of post-accord criminal violence. Conceptually, these strategies are best conceived as violence disruption. Grassroots actors disrupt prevailing self-protection strategies and can foster further collective action by restoring cohesion, strengthening social capital and fostering informal leadership. Residents also disrupt violent cycles, attaining tangible security improvements locally. However, the impact of violence disruption is contingent on the power of violent actors and its scale appears insufficient to transform post-accord violence’s complex causes.

The article unfolds in five sections. First, I review overlapping literature on civilian agency amid violence. I follow this review with a conceptual exploration of how the strategies evidenced are distinct forms of agency: violence disruption. I then present the article’s methods and the context of the case studies. The fourth and fifth sections present the empirical material, respectively, evidencing the impact of dispute mediation by informal leaders and of local negotiations with gangs. I conclude by discussing the novelty of the empirical material, new avenues for research and the relevance of the conceptual contribution of the article.

Civilians amid violence: contexts, types and drivers of local strategies

Our understanding of how residents in violent contexts protect themselves or prevent escalations is growing, nourished by cross-disciplinary perspectives. For instance, literature on civil wars has shifted from war’s causes and consequences to trying to understand patterns of violence itself. The importance of civilians, their strategic choices and how they interact with armed group behaviour has become a key focus.⁵ Further developments in International Relations have foregrounded civilian agency amid violence. The local and resilience turns in peacebuilding research have shifted analytical lenses towards local capacity to resist in, adapt to and transform violent contexts.⁶ Moreover, researchers have increasingly documented how civilian rescue behaviour emerges even in the worst contexts of genocide.⁷ Critiquing a ‘salvation paradigm’⁸ undergirding traditional understandings of protection, studies have evidenced how protection during conflict is frequently the result of local strategies in contexts as varied as Colombia, Iraq, Uganda or Liberia.⁹

Compared with evidence from armed conflicts, knowledge is sparse on how populations cope with and seek to address non-conflict violence. To illustrate the state of knowledge, it is helpful to consider Jose and Medie’s conceptualisation of civilian self-protection strategies as falling into three broad categories: avoiding engagement with armed actors, non-violent engagement and violent engagement.¹⁰ Studies on Latin America have documented plural avoidance strategies. For instance, studies have documented links between criminal violence and forced displacement,¹¹ gang dominance and decreased political mobilisation,¹² and coping strategies such as avoiding neighbours or changing mobility patterns to avoid victimisation.¹³ Violent engagement is similarly well explored. For example, Moncada investigated the conditions under which business associations resist extortion through vigilantism rather than avoidance.¹⁴ Vigilantism and lynching have received extensive attention altogether.¹⁵

In contrast, evidence on non-violent engagement is tentative. There have been calls to explore the causal link between local capacity for organisation and Central America’s variable homicide levels.¹⁶ Similarly, Berg and Carranza evidenced differences in collective organisation among Honduran neighbourhoods with high and low homicides.¹⁷ However, there is scant work focusing on the concrete strategies residents of violent communities deploy. As a noteworthy exception, Arias documented sporadic protests and more complex strategies to constrain (via civil society coalitions) and to engage with gangs and police in Jamaica, Brazil and Colombia.¹⁸

Consequently, evidence on what factors drive non-violent engagement largely comes from armed conflicts. A crucial consensus is that civilian strategies are driven by endogenous characteristics of the population but contingent on violent dynamics and types of armed groups. Social cohesion, understood as having effective platforms for community organisations (development committees or religious congregations), has been proposed to explain the emergence and scope of anti-violence strategies.¹⁹ The importance of charismatic leaders, capable of galvanising community-wide action or preventing polarisation, has also been identified amid sectarian violence.²⁰ Endogenous characteristics interact with local environments, with prior work suggesting criminal groups to be least amenable to engagement by residents.²¹ Alternatively, intervention by NGOs may foster organisation, increase capacities or open space for intervention.²² In the context of criminal violence, it is necessary to re-assess if similar patterns of non-violent contention exist, what factors drive them, and how to conceptualise their impact.

Violence disruption as a distinct form of agency

I propose understanding the strategies researched in urban Guatemala as forms of violence disruption. The concept carves out a conceptual space for forms of agency distinct from survival,²³ everyday resistance,²⁴ coping,²⁵ or friction²⁶ amid violent actors. Such literature studies rightly highlight civilian agency amid violence. Yet keeping to self-protection conceptualisations risks underestimating the potential that residents hold to affect violence. Similarly, in resilience frameworks, concepts to situate local agency precisely are lacking. Resilience practitioners and researchers conceptualise grassroots agency as showcasing absorptive, adaptive or transformative capacities.²⁷ These capacities allude, respectively, to the capability to survive a shock, adapt to changing contexts or transform said environments. Residents who remain in extremely violent Guatemalan neighbourhoods exhibit absorptive and adaptive capacities, as they survive and self-protect from local violence. Yet residents also engage in more ambitious strategies, breaking away from self-protection behaviours and seeking proactive solutions to violence. However, transformative capacities are often conceived as changing governance patterns, institutions and culture – a degree of impact not evident from the actions of grassroots groups researched. The concept of violence disruption bridges this gap, situated within a continuum of forms of agency between adaptation and transformation.

The notion of disruption proves especially helpful to situate the type of agency on display: strategies that are disruptive with regard to their origin and their impact. For Mac Ginty, who coined the term conflict disruption, it is the result of ‘individuals and groups who disrupt the main dynamic, logic, or narrative of a conflict’.²⁸ I adapt the concept to violence disruption, alluding to key contextual differences. In post-accord Guatemala, violence does not respond to one overarching conflict dynamic, logic or narrative, but rather to plural manifestations, drivers and causes. Therefore, violence disruption encompasses non-violent actions by individuals or community groups aiming to proactively prevent, mitigate or affect the plural dynamics through which different manifestations of violence ensue. Residents who participate in violence disruption are already disruptive, regardless of their impact. The emergence of violence disruption breaks with patterns of behaviour that have been internalised because of mass violence, and which stymie collective action. Engaging in violence disruption breaks deep-seeded self-protection dynamics that have become the norm amid post-accord violence (e.g. silence, avoiding participation or displacement). Therefore, violence disruption represents a rupture with local dynamics that inhibit collective action and can disrupt local violent cycles attaining tangible security improvements, as the empirical section shows.

Highlighting limits in the impact of grassroots agency also justifies using the term disruption. In post-accord contexts, disruption does not appear to transform the conditions from which violence originates. Mac Ginty argued that conflict disruption can be a ‘precursor’, which ‘can create space in which other forms of peace may take root’,²⁹ a stance similarly taken in the literature on civil action amid conflict.³⁰ I sustain that violence disruption creates space for further collective action at a similar level of incidence. However, residents face unequal power relations with violent groups and violence driven by plural micro-meso-macro factors, and so the transformational capacity over the complex drivers of post-accord violence should not be overstated.

Methods, cases and their context

Data were collected between November 2019 and November 2020. I conducted 47 semi-structured interviews with security experts (civil society and government) and case study informants – residents, journalists and staff from educational centres (Annex 1). Interviews ranged from 22 to 150 minutes (median: 64) and were conducted, transcribed and analysed in Spanish. Participants were asked about local violent dynamics (actors, changes in intensity/manifestations, role of state institutions, local responses and coping strategies). My affiliation with a British university likely favoured access to respondents generally, except with local residents, who were reticent to be interviewed unless an acquaintance introduced me (mostly people collaborating with Asociación Grupo Ceiba – AGC). Being a native Spanish speaker, living for extensive periods in Guatemala and having family locally enabled the use of accessible language in interviews and facilitated understanding idioms and references. As a foreigner, I was an outsider to the experience of coexisting with fear of violence and its impact. However, participants regularly positioned me as an outsider in need of such learning, and were frequently compelled to offer detailed explanations of assumed knowledge. Following COVID-19 restrictions, local respondents were interviewed via telephone or videoconference (from March 2020 onwards). This medium may have affected reading non-verbal cues, but aided participants’ ease given the potential danger of being labelled by gangs as an informant for speaking to an outsider. Interview data were triangulated with the literature, police data obtained through official requests, and a newly-created database (210 local news items from newspapers Prensa Libre and El Sol de San Pedro Ayampuc). The University of Bristol approved ethical protocols, which mandate that names and information that could identify interviewees are withheld.

Case studies were selected in the framework of a project on post-accord violence, broader than this article. Guatemala’s armed conflict came to a negotiated end in 1996,³¹ after decades of insurgency, state terror and genocide.³² Since then, Guatemala exemplifies a paradox seen across Latin America: mass violence followed democratisation and peace accords.³³ Post-accord violence follows complex patterns, with networks of drug trafficking organisations, gangs, oligarchic elites and remnants of counter-insurgent state security pursuing distinct illicit, economic, or political goals, yet colluding frequently.³⁴ Guatemala exemplifies another regional phenomenon: immense subnational variation in manifestations of violence and homicide rates.³⁵

The neighbourhoods of El Limón and Brisas de San Pedro were selected because they present most-similar cases. They present a similar vulnerability to violence, constituting similar-size poor urban areas in proximity; having limited state presence, except for bouts of mass police-military deployment; not having been the epicentre of combat nor mass militarisation during the armed conflict. Despite similarities, El Limón has seen a decrease in homicides, while in Brisas they have increased. The divergent trend is evident in National Civilian Police data (where the smallest geographic disaggregation is an area larger than neighbourhood). Zone 18 (El Limón), saw a peak of 321 homicides in 2009, decreasing to 151 in 2011 and stabilising thereon, with a low of 77 in 2015. San Pedro Ayampuc municipality (Brisas) saw 21 killings in 2011, then a sustained increase, tripling by 2018. Interview accounts corroborate police data: Brisas’ interviewees regret worsening lethal violence, while El Limón’s describe fewer homicides.

Both cases are in Guatemala Department, where the capital is situated. El Limón was created after 1976 on land allocated to re-house victims of an earthquake, in which people fleeing the armed conflict and seeking economic opportunities also settled. The neighbourhood is reputed for its insecurity and is one of the first where gang presence is reported.³⁶ Brisas is more recently populated, and grew exponentially with people searching for affordable housing in the early 2000s. A large landowner with dubious claims to legal tenancy sold small plots to most families in Brisas, and residents have struggled for decades to have their ownership recognised. Thereon, rapid growth has continued and also fed by arrivals of families displaced because of intimidation and violence by criminal groups.

Interviews and local media showed that variations in homicides in the cases stemmed from contrasting situations of competition or control by violent groups. El Limon’s lethal violence peaked during a war between gangs −2006 to 2010. As a group affiliated with Barrio 18 gained unchallenged control, homicides decreased dramatically, but coercive social control became severe, as did quasi-universal extortion (house-to-house and local businesses). In contrast, ongoing competition between gangs explains increasing lethal violence in Brisas. Coercion against residents is less prevalent – not inexistent–, extortion has not reached the extent seen in El Limón and most violent events stem from clashes between rivals. Consequently, the cases offer the chance to investigate non-violent strategies amid comparable yet distinct contexts.

Context offers key insights into the type of agency in display. In spite of differences between cases, pervasive insecurity elicits similar self-protection strategies that hinder cohesion, social capital and grassroots leadership. This fact contradicts the notion that these factors predict the scale of anti-violence strategies and evidences how violence disruption strategies represent a rupture with behaviour that inhibits collective action (rather than emerge as a consequence of favourable local conditions). First, forced displacement undermines social capital and leadership in both neighbourhoods. Displacement is especially common in response to extortion in both neighbourhoods.³⁷ For instance, El Limón featured in a 2014 police report as Guatemala’s neighbourhood with most abandoned houses due to violence.³⁸ Coercive violence also triggers displacement. In Brisas, for example, several neighbours requested that the military establish a presence and subsequently became forcibly displaced after gang threats.³⁹ In El Limón, after local leaders were sought by the Ministry of Interior as local liaisons for a military outpost in 2012, several were killed or fled after threats.⁴⁰ Displacement results in lost social capital for remaining residents, as known neighbours depart. Abandoned homes, too dangerous to sell or rent, are often occupied by gang members.⁴¹ Moreover, since displacement especially affects proactive local leaders, key human capital is lost.

Insecurity also leads many to self-protect by not engaging with neighbours⁴² or avoiding the community altogether.⁴³ By doing so, residents minimise the risk of interacting with gang members or unwittingly signalling affiliation to any group, which could elicit reprisals.⁴⁴ Silence entails more than compliance and, deployed strategically, it can protect some modest forms of collective action that do not threaten violent actors’ interests. For example, in Brisas mobilisation to demand land titles is robust, yet active residents have internalised that intervening in security issues would endanger them.⁴⁵ In El Limón, despite historically strong local organisation, residents increasingly avoid participation because Barrio 18 controls collective initiatives, such as food distribution to poorer neighbours.⁴⁶ Overall, the prevalence of silence to self-protect hinders social capital and cohesion to pursue collective goals related to insecurity.

Relatedly, insecurity undermines platforms for potential mobilisation, such as development committees or religious congregations. Local development committees actively avoid dealing with local violence, even when they are active on issues like local utilities in both neighbourhoods.⁴⁷ Moreover, in El Limón and Brisas, committees were widely mistrusted, amid accusations of co-optation by clientelist networks, or of lacking internal democracy.⁴⁸ Further potential catalysts for collective action, such as churches, were described as actors curtailed by the imperative to not antagonise gangs.⁴⁹ In Brisas, there is scant evidence of work by religious networks to address violence. In El Limón, the Catholic Church was a key actor historically,⁵⁰ but progressively reneged from involvement in insecurity, to avoid antagonising gangs as the environment became riskier.⁵¹

Despite the risks, violence does not fully subdue community activism. In 1989, activists from El Limón, initially supported by the Catholic parish, created a group to further community development and prevent violence by and against youth.⁵² In 1995, local activists broke ties with the church and became legally recognised as Asociación Grupo Ceiba (AGC).⁵³ AGC’s work expanded to other neighbourhoods, including Brisas. The approach of strengthening local networks of committed citizens who could devise and implement actions to address violence grew into a method named Urban Peace. Combined with educational initiatives, the approach continues to connect neighbours to foster action by small groups of leaders to address violence and mediate disputes. The approach ultimately expanded to include work to create similar groups for dispute mediation inside prison.

The networks established came to be defined as micro-peace platforms (MPPs): local groups for political participation and action. They seek to break a state of ‘chronic terror’,⁵⁴ working at the scale of the street or housing block. At this level, leadership falls on charismatic individuals who are not traditional powerholders (i.e. not religious leaders or members of development committees). The persons who set up MPPs and mobilise peers are nicknamed ‘subsoil’ leaders, for their informal and less obvious legitimacy. MPPs seek to build an alternative form of power, one built on solidarity, dialogue and non-violence,⁵⁵ reminiscent of Mac Ginty’s ‘everyday peace power’.⁵⁶ MPPs organise meetings to analyse community dynamics and organise micro-actions in response to perceived needs. Each group makes independent decisions on their course of action. Examples of activities are as varied as mediating in disputes between neighbours, organising food distributions, or petitioning municipal government for better services. MPPs collaborate with local institutions when the latter operate locally (e.g. public services or NGOs). However, members often portray themselves as being the only ones acting in response to violence: perceiving that other institutions do not work in their communities or abandon them, that state-led initiatives for violence prevention are absent, and mistrusting security forces for their authoritarian approaches. MPP members are all residents, supported by members of AGC, who advise on how to broach local conflicts. AGC staff have also engaged in direct negotiations with gangs.

The emergence of anti-violence strategies given the context provides initial insights on violence disruption. As insecurity weakened cohesion, social capital and leadership, local action was sustained over time. Consequently, the case studies suggest that the waning or strengthening of these factors does not explain more or less ambitious strategies in post-accord contexts. Otherwise stated, rather than being a product of favourable endogenous conditions, violence disruption represents rupture with prevailing self-protection strategies, which stymie collective action. Moreover, leadership, cohesion and social capital are fluid: they are affected by insecurity, but can also be restored as by-products of violence disruption strategies, as shown subsequently.

Grassroots dispute mediation and the disruption of violent cycles

Subsoil leaders are charismatic individuals, active locally prior to any MPP involvement. They may be bakers, cleaners, or housewives who, by virtue of their local knowledge and legitimacy, can motivate others to carry out micro-actions alleviating violent dynamics, mainly through dispute mediation, breaking away from prevailing self-protection strategies.⁵⁷ MPPs help create opportunities for action by creating safe spaces for dialogue, pooling collective knowledge, and networks, thereby restoring social capital and cohesion. The examples below demonstrate how MPPs can intervene in various violent dynamics: in the sphere of the family, in interpersonal disputes and, in rare instances, in conflicts involving gangs. The examples demonstrate how the notion of violence disruption helps grant due importance to such forms of grassroots agency, and how they attain security gains.

A crucial contribution of MPPs to local dynamics is disrupting patterns of self-protection in favour of collective action. MPP members report a greater willingness to act since they started participating, greater confidence to initiate actions to solve community problems and greater awareness of channels to advocate for community rights.⁵⁸ Interviewees widely reported improved social cohesion – articulated as greater ‘unity’, ‘trust’, or ‘organisation’.⁵⁹ Overall, grassroots leaders understand that participating in MPPs increases their capacity to identify, analyse and intervene in conflict situations, through mediation. This increased capacity rests on two major aspects: the support of peers with whom to analyse the nature, scale or actors involved in local conflicts,⁶⁰ and improving mediation skills.⁶¹ Increased involvement, in turn, deepens the legitimacy of subsoil leaders among other neighbours, who seek leaders out when conflicts arise.⁶² Consequently, we should conceive the establishment and work of MPPs as disrupting patterns of self-isolation and silence, fostering leadership, rebuilding social capital and restoring cohesion, which all favour further violence disruption.

Moreover, the work of MPPs disrupts violent cycles, attaining tangible security gains. Many interviewees pointed to participation in MPPs as crucial to changing their own attitudes towards violence and in the way they approach disputes with neighbours and family.⁶³ The impact of individual rejections of violence is heightened by that of emboldening leaders to intervene in wider community conflicts. Local leaders have the greatest confidence identifying and mediating in situations of violence against women or children.⁶⁴ Interventions aimed at solving domestic or family disputes, an important and often silenced phenomenon, have a significant impact when considering the wider context. For instance, identifying and disrupting violence at home against children can constitute a preventive intervention for gang recruitment. There is a well-accepted connection between the experience of abuse at home and joining a gang.⁶⁵ Accession into gangs is progressive and most frequently non-coerced, and children begin with menial tasks.⁶⁶ Prior to becoming full members, they enter an interim stage, called chequeo, where they must prove their worth, often with a violent action. Given this incremental process, mediating in domestic conflicts, and identifying early signs of involvement in gangs (which may itself be driving domestic tension) is a way MPP leaders understand they are able to disrupt recruitment.⁶⁷ Consequently, the impact of intervening in the domestic sphere can be better ascertained when conceived as a form of disruption of violent cycles (e.g. gang recruitment).

Leaders also play important roles in interpersonal disputes. Interviewees agreed that work by MPPs in mediation and promoting dispute resolution eased conflict between neighbours.⁶⁸ In both neighbourhoods, violence is used frequently to deal with mundane tensions – where gangs are not the trigger–, such as conflicts over parking or waste disposal. Respondents described multiple pathways for mundane conflicts to escalate, given the access to weapons and persons willing to use them.⁶⁹ The most obvious escalation is through aggressions between parties involved. Gang presence also makes contract killings a prevalent phenomenon, including for revenge against small-time criminals or neighbours who have committed a perceived transgression.⁷⁰ As this interviewee narrated: ‘If I know so-and-so, I can go call him, and for 100 quetzals he’ll do me the favour of planting a bomb in that guy’s house’.⁷¹

Escalation also ensues via kinship relations, linking back to gangs. For example, in Brisas, an entire school suspended activities after receiving bomb threats. The dispute originated when a teacher suspended a child, whose father was a gang leader. The issue was resolved only when the teacher apologised to the gang leader, who accepted the apology but reiterating that he would kill those who committed any other slight.⁷² Given the plural pathways for mundane disputes to escalate, by contributing to a rejection of violence and mediating in interpersonal disputes, MPPs contribute tangibly to local security: disrupting escalations of violent cycles.

Grassroots leaders also affect more complex patterns of collective violence, involving gangs. These security gains are rarer, but significant. Given the threat posed by gangs, intervening in many violent phenomena was widely seen to be off-bounds, for instance episodes of domestic violence where a gang member is involved or any criminal activities conducted by gangs.⁷³ However, MPPs constantly re-draw these boundaries and newly-gained cohesion, social capital and capacity to analyse threats collectively make risky interventions possible. For example, in El Limón, when a neighbour faced extortion threats, a group of local leaders intervened on his behalf. The group was able to identify where the threat originated, analysing affiliations of the perpetrator, determining him to be an ‘imitator’ (not a gang member capable of following through), and finally convincing him to desist peacefully. As this interviewee describes:

I remember that we were able to engage in finding a solution to the problem, because it was becoming more complicated that it really was. But it was for the same reason, that we were able to detect it as a group.⁷⁴

The quote illustrates how MPPs become fora for participation, but also where to pool social knowledge.

In this case, collective analysis enabled the identification of the problem and its collaborative evaluation. The MPP provided a platform that was crucial to dissect the issue, exchange local knowledge and identify an unlikely opportunity to intervene in an episode of extortion (widely understood to be too risky). The same respondent explained how similar actions were previously unthinkable: ‘we realised that it was a situation where we weren’t putting ourselves at risk. If it had been a risky situation, we wouldn’t have done it’.⁷⁵ Through increased collective analysis at the MPP, the problem was identified, drawing on collective knowledge of the evolving threat environment, MPP members ascertained that the problem originated in an imitator, and therefore intervention became feasible.

The episode exemplifies the potential of grassroots agency and the usefulness of the concept of violence disruption. In a pessimistic reading, wider extortion operations remained untouched, as did the conditions of impunity which enable them. Yet as a counterfactual, without grassroots intervention, the problem would likely have never been identified. At best, the threatened neighbour would have become a long-term victim of extortion. At worst, this episode could have led them to become forcibly displaced. Plausibly, given the extraction of payments with impunity, the imitator might have been emboldened to threaten other residents. Grassroots non-violent action disrupted this cycle of fear, impunity and displacement. In a more expansive view, non-violent intervention initiated a virtuous cycle: disrupting patterns of silence and avoidance, channelling cohesion towards action and, more speculatively, likely encouraging those involved to face similar threats in the same way.

Grassroots strategies are not a panacea, but notions like disruption are helpful to grant them due importance. Subsoil leaders can disrupt local cycles of violence, contributing for instance to alleviate gang recruitment, lethal violence originating in neighbourly disputes and – in rarer instances – complex phenomena where gangs are directly involved, such as extortion. They do so building on a form of leadership that differs from that seen to be relevant amid, for example, sectarian violence:⁷⁶ rather than galvanise entire communities to action or enforce norms therein, subsoil leaders amid post-accord criminal violence work at a more micro scale, yet attaining important security gains. MPPs’ impact, however, appears contingent on power imbalances with violent actors, which often render action too risky. Therefore, there is scant evidence of their strategies becoming precursors of wider transformation of the causes that enable post-accord violence at large, contrary to research emerging from armed conflicts.⁷⁷ Nonetheless, establishing MPPs can restore cohesion, regenerate social capital, and foster alternative modes of leadership, underscoring how factors often seen as drivers of grassroots strategies can instead emerge as products of violence disruption, favouring further collective action at a similar scale.

Assessing the potential of localised negotiations with gangs

AGC members have engaged directly with gangs in issue-specific negotiations – i.e. dissolution or disarmament was never on the table. Negotiations’ impact may seem modest in terms of transforming the conditions that enable violence, but immediate security gains have been attained. Negotiation can also effectively increase access for NGOs or public services in violence-affected communities. However, the success of negotiations is contingent on local dynamics of violence, in particular the balance of power between local violent groups.

When a single gang consolidated control in El Limón, residents became subject to a virtually universal ‘taxation’ system, with house-to-house extortion based on the appearance of wealth.⁷⁸ Drawing on communication channels to gang leaders inside prisons, AGC advocated for an end to the practice, appealing to gang leaders’ sense of neighbourhood belonging.⁷⁹ Gang leadership committed to ending house-to-house extortion but not extortion aimed at businesses.⁸⁰ Two reasons, impossible to confirm without gang leader accounts, might explain the concession (beyond the mere incentive to improve Barrio 18’s image). When the agreement was concluded, in 2017, many better-off families and other victims of extortion threats had abandoned El Limón anyway. Additionally, excluding businesses likely meant that major sources of income were unaffected. Nevertheless, given the impact of extortion – victimising those who pay, triggering displacement among those who cannot and provoking repercussions against those who will not–, any reduction matters.

Negotiation also facilitated better access to services for residents, particularly education. In El Limón, when Barrio 18 consolidated control and extortion grew, transport operators, large and small businesses, and NGOs were targeted. Among those affected by extortion were schools and their staff.⁸¹ In response, several schools were on the brink of terminating local work. Building on its community credibility and contacts with Barrio 18 leadership, AGC similarly led negotiations and secured an exoneration of schools from extortion.⁸² The agreement persisted at the time of data collection and survived changes in gang leadership.⁸³ Brisas similarly benefitted from better access to services. It was a local common occurrence that youth heading to or returning from school would be threatened while crossing territory controlled by rival groups. For instance, in 2015, two students were attacked when returning from AGC’s school.⁸⁴ Again, through direct negotiation with imprisoned gang leadership, AGC was able to advocate for pupils not to be targeted on their way to or back from school.⁸⁵

Access to wider services also improved through pacts. In El Limón, a violence-free zone known as ‘Peace street’ was established so social, cultural and commercial activities could happen without requiring Barrio 18 authorisation. Negotiation included the gang’s leadership, subsoil leaders and local police.⁸⁶ As this interviewee describes: ‘in this street any activity can take place, whether it’s cultural or sports, without anyone coming and asking: “what are you doing here?”, or without any attacks or anything’.⁸⁷ Crucial to securing the gang leadership agreement was that police similarly committed to refraining from raids and repressive action in that street.⁸⁸ The neutral zone enabled the market to happen unhindered, protected school pupils and church-goers on that street, and enabled communal activities, such as street cinema or outreach by social services.⁸⁹

Several elements stand out from the two examples of negotiation. Negotiators have ‘insider’ legitimacy, but rely on being ‘outsiders’ to attempt mediation safely. As described, AGC is a grassroots initiative, born in El Limón, and the association has been a unique advocate for the rights and advancement of local youth. Consequently, for instance, while other NGOs and schools have suffered extortion, the organisation has not.⁹⁰ The insider position grants its recognition that extends to gang leadership: many of whom or their relatives attended AGC schools or grew up knowing the association’s activities.⁹¹ As this respondent articulates: ‘[gang members] recognise Ceiba as legitimate, because it gave them studies, it supported them, it gave them affection, so they appreciate Ceiba. This permitted Ceiba to gain that recognition until now, to avoid conflict with them’.⁹² Secondly, AGC’s ability to scale up its work – professionalising and expanding geographically – enabled the association to play a useful outsider role. While subsoil leaders widely considered it too unsafe to advocate openly for victims of extortion or against gang activities, AGC members – some of whom are not local residents – can navigate these boundaries safely as outsiders not living directly under gang control.

Moreover, instances of negotiation have in common that they took place with imprisoned gang leaders.⁹³ The insider network and trust that AGC has generated aid dialogue with gang leadership.⁹⁴ However, it is because of the connections made inside the prison system – through social work funded by international cooperation – that sustained channels of communication exist.⁹⁵ Those connections were made by the more professionalised branch of the association, working inside the prison system, who are not from the communities where the organisation works and are able to broach negotiations and bypass safety concerns.⁹⁶ In sum, the dual position as legitimate insider but more distant outsider, facilitates negotiation and new opportunities for violence disruption. This finding supports prior work concluding that external actors can be crucial to foster capacities for local agency,⁹⁷ yet indicates that grassroots strategies can be fostered in more nuanced arrangements than as interactions between clearly demarcated external or local actors.

A further point on negotiation concerns its modest capacity to reduce competitive violence. There was no evidence of impact on violence between rivals in the cases, but some emerged outside the two neighbourhoods. In Guatemala City’s El Esfuerzo neighbourhood, AGC facilitated a truce between local cliques of Barrio 18 and Mara Salvatrucha, including a halt to criminal activity throughout Christmas 2010.⁹⁸ However remarkable, the truce’s impact was short-lived: the day it expired, a young gang member was killed and violence resumed.⁹⁹ Months later, the gang leaders who participated in the truce were assassinated – it is unclear if as a consequence of participating in the truce.¹⁰⁰

Negotiations, therefore, appear severely affected by changing local violence. The successes in El Limón versus a more modest track record in Brisas point to a paradox: while more control by a single gang made extortion and coercive social control more severe, it facilitated negotiating around these very phenomena. In El Limón, having a single gang facilitated communication and the credibility of pacts reached, whereas having multiple gangs would entail getting rivals to communicate and trust one another. As this interviewee explains:

[In Brisas] between youth gangs it has been a bit difficult –not because there haven’t been attempts, because there have been dialogues with their members – but they are somewhat unwilling to talk between them because of their rivalry. The advantage in El Limón is that, as I mentioned, there is only one youth gang.¹⁰¹

In El Limón, when the gang war was at its height, even beginning a process of dialogue was seen as risky.¹⁰² In contrast, the subsequent control by Barrio 18 over the neighbourhood (and its own members) meant that agreements reached were upheld.¹⁰³ Relatedly, this fact signals the impact that internal gang fragmentation could have on negotiated gains, which rest largely on gang capacity to enforce agreements internally.

Complementing evidence on state-led negotiations with criminal groups,¹⁰⁴ the findings highlight the importance of civil society-led negotiations. Through negotiation, access to public services improved, including establishing violence-free zones. Negotiations have attained security gains, disrupting patterns of victimisation such as extortion, but falling short of transforming the violent context alone. The ability to extract concessions in scenarios of competition diminishes, as the modest gains in Brisas indicate. This insight nuances those by Arias:¹⁰⁵ while he found negotiated engagement to be frequent across different configurations of armed group control or competition, Brisas and El Limón suggest the effectiveness of negotiating increases when a single criminal group exerts control. Rare negotiated truces between rivals may present opportunities for disrupting competitive violence between gangs. Yet such gains appear to be temporary, and not viable pathways to address underlying causes behind the gang phenomenon, to transform drivers of gang enmity or to scale up truces to a national or even regional level.

Dialogue with gangs also elicits difficult moral questions, which hinder the transformative potential of negotiations. Negotiation with criminal groups is unpopular, regardless of effectiveness.¹⁰⁶ Reservations are held simultaneously by state authorities and community members. Both see pacts as morally questionable, and concessions by gangs as unreliable,¹⁰⁷ even if security gains can be attained and state-led negotiations can potentially result in disarmament and demobilisation, as Ecuador shows.¹⁰⁸ As an interviewee lamented: ‘some say “once you negotiate with pandilleros you become part of them”. But no, it is about gains and saving lives, not about negotiating’.¹⁰⁹ In this context, modest gains may continue to be attained locally, but there appears to be scant potential for negotiation to be a precursor to more transformative action, leading to wider demobilisation or disarmament.

Conclusion

The article assessed the impact of non-violent strategies in urban neighbourhoods in post-accord Guatemala with consolidated gang presence. Such forms of local agency to address criminal violence are risky and exceptions, but constitute an important and often-overlooked dimension in analyses of non-conflict contexts. Far from a binary of victims and perpetrators, the article showed residents engage in distinct forms of agency that should be conceived as violence disruption.

The findings that grassroots efforts impact violent dynamics underscore how research on Latin America’s violence should evolve towards a greater focus on local causal mechanisms. The article demonstrates that non-violent approaches, such as dispute mediation and negotiation with gangs, alter micro-dynamics of violence and attain tangible security gains (e.g. restricting extortion’s impact, preventing displacement or safeguarding access to services). In presenting this argument, the article contributed novel empirical material given its focus on a post-accord non-conflict context, of great relevance to the region. The article’s evidence demonstrates that exploring local causal mechanisms, among which is grassroots agency, is an important avenue to understand better the variations in Latin America’s homicide rates and experiences of (in)security.

The notion of violence disruption advanced in the article can help future research to neither overlook nor overstate grassroots agency. The strategies evidenced in urban Guatemala can disrupt violent cycles in subtle localised ways (e.g. those leading to gang recruitment or lethal escalations of neighbourly disputes). This impact is significant for the neighbourhoods under study, but differs from that observed in armed conflicts, where community groups were able to remain autonomous from rival armed groups or fully restrain violent escalations. Given these differences in impact, and plural manifestations of violence in Guatemala (and the region), an important question to explore is how non-violent strategies impact similar post-accord contexts but with different violent dynamics (e.g. with mass state repression, consolidated vigilante groups, or private security companies defending extractive projects).

The case studies also contribute important insights on drivers of local agency amid violence. Violence disruption approaches emerged in neighbourhoods where insecurity hindered cohesion and social capital, pushed community groups to avoid engaging in local affairs and prominent leaders were forcibly displaced. Violence disruption strategies documented represent a rupture from self-protection strategies (silence, avoidance and displacement) that inhibit collective action, rather than a product of favourable conditions. Moreover, through violence disruption strategies, social capital can be regenerated, alternative forms of leadership fostered, and social cohesion strengthened, facilitating further collective action.

By highlighting how cohesion, social capital and leadership can be a by-product of violence disruption, the article tackled the crucial issue of whether grassroots efforts can aggregate to transform the conditions that enable post-accord violence. Violence disruption approaches identified show scant potential to transform the micro-meso-macro challenges that enable the reproduction of violence in Latin America. In urban Guatemala, residents face grave risks, including social control and the threat of violence by gangs, and they have limited leverage to shape gang behaviour at large. Therefore, the persistence of local violence should never be understood as a failure to act. Grassroots violence disruption is a demonstration of courage, with tangible impact, but not a panacea for transforming communities affected by mass criminal violence.

Acknowledgments

The author would like to thank Roddy Brett, Ana Juncos, Patience Schell, Hannah Gracher and two anonymous reviewers for their helpful feedback on earlier versions of the draft, but all errors and omissions remain mine. The author would also like to thank Asociación Grupo Ceiba for the important work they do, and especially for their collaboration and logistical support to conduct the research that led to this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by an Economic and Social Research Council doctoral grant, under Project Reference [1930852].

Notes on contributors

Daniel S. Herrera Kelly

Daniel S. Herrera Kelly is an academic and practitioner focused on violence, peacebuilding, and civil society’s role therein, specialised in Central America. He has worked for over a decade in NGOs, in humanitarian and peacebuilding initiatives in Latin America, Africa and the Middle East. He obtained his PhD from the University of Bristol, supported by an ESRC doctoral grant. He also holds master’s degrees from the University of St Andrews and the University of Granada. He is currently an Honorary Research Associate at the School of Sociology, Politics and International Studies, at the University of Bristol.

Notes

¹ Susanne Jonas, Of Centaurs and Doves: Guatemala’s Peace Process (Boulder: Westview, 2000).

² Caroline Moser and Cathy McIlwaine, Encounters with Violence in Latin America: Urban Poor Perceptions from Colombia and Guatemala (New York: Routledge, 2004); and Deborah Yashar, Homicidal Ecologies: Illicit Economies and Complicit States in Latin America, Cambridge Studies in Comparative Politics (Cambridge: Cambridge University Press, 2018).

³ Guillermo Trejo and Camilo Nieto-Matiz, ‘Containing Large-Scale Criminal Violence Through Internationalized Prosecution: How the Collaboration Between the CICIG and Guatemala’s Law Enforcement Contributed to a Sustained Reduction in the Murder Rate’, Comparative Political Studies (2022): 1–37.

⁴ Betcy Jose and Peace Medie, ‘Understanding Why and How Civilians Resort to Self-Protection in Armed Conflict’, International Studies Review 17 (2015): 515–35.

⁵ Stathis Kalyvas, The Logic of Violence in Civil War (Cambridge: Cambridge University Press, 2006); and Ana Arjona, ‘Civilian Cooperation and Non-Cooperation with Non-State Armed Groups: The Centrality of Obedience and Resistance’, Small Wars & Insurgencies 28, no. 4–5 (2017): 755–78.

⁶ Roger Mac Ginty and Oliver Richmond, ‘The Local Turn in Peace Building: A Critical Agenda for Peace’, Third World Quarterly 34, no. 5 (2013): 763–83; Philippe Bourbeau, ‘Resilience and International Politics: Premises, Debates, Agenda’, International Studies Review 17, no. 3 (2015): 374–95; David Chandler, Peacebuilding: The Twenty Years’ Crisis, 1997–2017 (Cham: Springer, 2017); and Ana Juncos and Jonathan Joseph, ‘Resilient Peace: Exploring the Theory and Practice of Resilience in Peacebuilding Interventions’, Journal of Intervention and Statebuilding 14, no. 3 (2020): 289–302.

⁷ J. Semelin, C. Andrieu, and S. Gensburger, eds., Resisting Genocide: The Multiple Forms of Rescue (Oxford University Press, 2014).

⁸ Frédéric Mégret, ‘Beyond the “Salvation” Paradigm: Responsibility To Protect (Others) vs the Power of Protecting Oneself’, Security Dialogue 40, no. 6 (2009): 575–95.

⁹ Daniel Levine, ‘Some Considerations for Civilian – Peacekeeper Protection Alliances’, Ethics & Global Politics 6, no. 1 (2013): 1–23; Paul Williams, ‘Protection, Resilience and Empowerment: United Nations Peacekeeping and Violence against Civilians in Contemporary War Zones’, Politics 33, no. 4 (2013): 287–98; and Jose and Medie, ‘Understanding’.

¹⁰ ‘Understanding’.

¹¹ Sindy Hernández Bonilla and Adela Morales Orellana, ‘Un Éxodo Que No Se Nombra. Aproximaciones al Desplazamiento Forzado Interno Por Violencia En Guatemala (2010–2019)’ (Guatemala City: Universidad Rafael Landívar and Procuraduría de los Derechos Humanos, 2020), https://www.pdh.org.gt/documentos/seccion-de-investigacion/investigacion/informes-tematicos/6265-documento-completo-un-exodo-que-no-se-nombra/file.html.

¹² Abby Córdova, ‘Living in Gang-Controlled Neighborhoods: Impacts on Electoral and Nonelectoral Participation in El Salvador’, Latin American Research Review 54, no. 1 (2019): 201–21.

¹³ Moser and McIlwaine, Encounters.

¹⁴ Eduardo Moncada, Resisting Extortion: Victims, Criminals, and States in Latin America (Cambridge: Cambridge University Press, 2021).

¹⁵ Angelina Snodgrass-Godoy, ‘When “Justice” Is Criminal: Lynchings in Contemporary Latin America’, Theory and Society 33 (2004): 621–51; and Regina Bateson, ‘The Socialization of Civilians and Militia Members: Evidence from Guatemala’, Journal of Peace Research 54, no. 5 (2017): 634–47.

¹⁶ Jenny Pearce, ‘Central America: From War to Violence’, in The Palgrave Handbook of Disciplinary and Regional Approaches to Peace, eds. O. Richmond, S. Pogodda, and J. Ramović (Basingstoke: Springer, 2016), 450–62.

¹⁷ Louis-Alexandre Berg and Marlon Carranza, ‘Organized Criminal Violence and Territorial Control: Evidence from Northern Honduras’, Journal of Peace Research 55, no. 5 (2018): 566–81.

¹⁸ Enrique Desmond Arias, ‘Social Responses to Criminal Governance in Rio de Janeiro, Belo Horizonte, Kingston, and Medellín’, Latin American Research Review 54, no. 1 (2019): 165–80.

¹⁹ Mohammad Suleman and Sue Williams, ‘Strategies and Structures in Preventing Conflict and Resisting Pressure: A Study of Jaghori District, Afghanistan, under Taliban Control’ (Cambridge, MA: CDA Collaborative, 2003), https://www.cdacollaborative.org/wp-content/uploads/2016/01/Strategies-and-Structures-in-Preventing-Conflict-and-Resisting-Pressure.pdf; Oliver Kaplan, Resisting War: How Communities Protect Themselves (Cambridge: Cambridge University Press, 2017); and Berg and Carranza, ‘Organized Criminal Violence’.

²⁰ Ami Carpenter, ‘Havens in a Firestorm: Perspectives from Baghdad on Resilience to Sectarian Violence’, Civil Wars 14, no. 2 (2012): 182–204; and Jana Krause, Resilient Communities: Non-Violence and Civilian Agency in Communal War (Cambridge: Cambridge University Press, 2018).

²¹ Kaplan, Resisting War; and Moncada, Resisting Extortion.

²² Ellen Furnari, Huibert Oldenhuis, and Rachel Julian, ‘Securing Space for Local Peacebuilding: The Role of International and National Civilian Peacekeepers’, Peacebuilding 3, no. 3 (2015): 297–313; and Oliver Kaplan, ‘The International Committee of the Red Cross and Support for Civilian Self-Protection in Colombia’, International Interactions 47, no. 5 (2021): 898–927.

²³ Jose and Medie, ‘Understanding’.

²⁴ James Scott, Weapons of the Weak: Everyday Forms of Peasant Resistance (New Haven: Yale University Press, 1985).

²⁵ Moser and McIlwaine, Encounters.

²⁶ Annika Björkdahl and Kristine Höglund, ‘Precarious Peacebuilding: Friction in Global – Local Encounters’, Peacebuilding 1, no. 3 (2013): 289–99.

²⁷ Carpenter, ‘Havens’; Alistair Cook, ‘Civilian Protection, Resilience, and Insecurity in Myanmar’, in: Civilian Protection in the Twenty-First Century: Governance and Responsibility in a Fragmented World, eds. C. Jacob and A. Cook (New Delhi: Oxford University Press, 2016), 127–46; Mercy Corps, ‘Mercy Corps Resilience Framework’ (Portland: Mercy Corps, 2019), https://www.mercycorps.org/sites/default/files/2019–11/Resilience_Approach.pdf; and Thomas Tanner, Aditya Bahadur, and Marcus Moench, ‘Challenges for resilience policy and practice’ (London: Overseas Development Institute, 2017), https://cdn.odi.org/media/documents/11733.pdf.

²⁸ Roger Mac Ginty, Everyday Peace: How So-Called Ordinary People Can Disrupt Violent Conflict (Oxford: Oxford University Press, 2021), 190.

²⁹ Everyday Peace, 18.

³⁰ D. Avant, M. Berry, E. Chenoweth, R. Epstein, C. Hendrix, O. Kaplan, and T. Sisk, eds., Civil Action and the Dynamics of Violence (New York: Oxford University Press, 2019).

³¹ Jonas, Of Centaurs.

³² Roddy Brett, The Origins and Dynamics of Genocide: Political Violence in Guatemala (London: Palgrave Macmillan, 2016).

³³ K. Koonings and D. Kruijt, eds., Societies of Fear: The Legacy of Civil War, Violence and Terror in Latin America (London: Zed Books, 1999).

³⁴ Susan Peacock and Adriana Beltrán, Hidden Powers in Post-Conflict Guatemala: Illegal Armed Groups and the Forces behind Them (Washington, DC: WOLA, 2003).

³⁵ Yashar, Homicidal Ecologies.

³⁶ Juan Manuel Merino, ‘Las maras en Guatemala’, in Maras y Pandillas en Centroamérica, eds. ERIC, IDESO, IDIES, IUDOP (Managua: Nicaragua, 2001), 109–218.

³⁷ Interviews 27, 29, 31, 33, 40 and 43–44.

³⁸ Jerson Ramos, ‘Vecinos abandonan casas por violencia’, Prensa Libre, November 26, 2014, https://www.prensalibre.com/guatemala/justicia/nacionales-vecinos-abandonan-casas-por-violencia-extorsiones-pandillas-0–1255674421/.

³⁹ Interview 29.

⁴⁰ Interviews 27 and 45.

⁴¹ Interviews 30 and 47.

⁴² Interviews 33, 38–39, 44 and 46.

⁴³ Interviews 38, 40–41 and 46.

⁴⁴ Interviews 32, 40, 42, 44 and 46.

⁴⁵ Interviews 29, 35–37 and 40.

⁴⁶ Interviews 27, 31–33, 35, 38 and 45.

⁴⁷ Interviews 27, 29, 38 and 44–45.

⁴⁸ Interviews 26–41.

⁴⁹ Interviews 32–34, 36–37, 41, 43–44 and 46–47.

⁵⁰ Interviews 27, 33 and 47.

⁵¹ Interviews 32–33 and 47.

⁵² Claudia Dary, Cristianos en un país violento. Respuestas de las iglesias frente a la violencia en dos colonias del área metropolitana de Guatemala (Guatemala City: USAC, 2016), 39–48.

⁵³ AGC, ‘Prevención del fenómeno droga y mara en áreas marginales urbano y rurales’ (CEPAL, 2005), https://dds.cepal.org/innovacionsocial/e/proyectos/gt/drogaymara/prevenciondrogafinal2.pdf.

⁵⁴ Marco Castillo, ‘Implementación de Microplataformas de Paz: Camino Integral Para La Construcción de Redes Comunitarias de Paz En Contextos Urbanos y Rurales En Violencia Crónica y Alta Marginalidad’, in Experiencias Urbanas, eds. Á. Mora and L. Herrera Robles (Juárez, Mexico: IMIP, 2020), 18–52.

⁵⁵ Ibid.

⁵⁶ Mac Ginty, Everyday Peace.

⁵⁷ Interviews 32–34, 36–41 and 46.

⁵⁸ Interviews 29, 33–34, 36–37, 41 and 46.

⁵⁹ Interviews 32–34, 36–37, 39–42 and 45–46.

⁶⁰ Interviews 32, 34, 37 and 39.

⁶¹ Interviews 37 and 41.

⁶² Interviews 33, 37, 39 and 46.

⁶³ Interviews 32–34, 39–40 and 42.

⁶⁴ Interviews 32, 34, 37, 39 and 41.

⁶⁵ ERIC et al., Maras y pandilla.

⁶⁶ Interviews 27, 32, 35, 43 and 46–47.

⁶⁷ Interviews 27, 32 and 37.

⁶⁸ Interviews 32, 34, 36–39, 41, 42 and 44.

⁶⁹ Interviews 26, 27, 30, 39, 41 and 47.

⁷⁰ Interviews 30, 43, 44 and 47.

⁷¹ Interview 30.

⁷² Interview 44.

⁷³ Interviews 27, 32, 38 and 42.

⁷⁴ Interview 32, emphasis mine.

⁷⁵ Interview 32.

⁷⁶ Carpenter, ‘Havens’; and Krause, Resilient Communities.

⁷⁷ Mac Ginty, Everyday Peace; and Avant et al., Civil Action.

⁷⁸ Interviews 27, 33, 35 and 47.

⁷⁹ Interview 47.

⁸⁰ Ibid.

⁸¹ Interviews 33, 35 and 47.

⁸² Interviews 2, 35 and 47.

⁸³ See note 79 above.

⁸⁴ Omar Sandoval, ‘Estudiantes Asesinados’, El Sol de San Pedro Ayampuc, 10 July 2015, https://www.calameo.com/read/0008484577fd6f96f670b.

⁸⁵ See note 79 above.

⁸⁶ Interviews 27 and 30.

⁸⁷ Interview 27.

⁸⁸ Ibid.

⁸⁹ See note 86 above.

⁹⁰ Interviews 35 and 47.

⁹¹ Interviews 27, 29–30, 35 and 47.

⁹² Interview 35.

⁹³ Interviews 27, 30, 35 and 47.

⁹⁴ Interviews 27, 30, and 35.

⁹⁵ Interviews 2 and 47.

⁹⁶ Interviews 26, 28 and 35.

⁹⁷ Alther, ‘Colombian Peace Communities’; Furnari, Oldenhuis, and Julian, ‘Securing Space’; and Kaplan, ‘The International Committee’.

⁹⁸ Interviews 30 and 35; also Prensa Libre, ‘Pandillas rivales pactan una tregua por Navidad’, Prensa Libre, December 20, 2010, https://www.prensalibre.com/guatemala/justicia/pandillas-rivales-pactan-tregua-navidad-0–393560674/.

⁹⁹ See note 71 above.

¹⁰⁰ See note 92 above.

¹⁰¹ See note 87 above.

¹⁰² See note 71 above.

¹⁰³ See note 79 above.

¹⁰⁴ Achim Wennmann, ‘Negotiated Exits from Organized Crime? Building Peace in Conflict and Crime‐affected Contexts’, Negotiation Journal 30, no. 3 (2014): 255–73; and José Miguel Cruz and Angélica Durán-Martínez, ‘Hiding Violence to Deal with the State: Criminal Pacts in El Salvador and Medellin’, Journal of Peace Research 53, no. 2 (2016): 197–210.

¹⁰⁵ Arias, ‘Social Responses’.

¹⁰⁶ Silvia Roque, ‘Between New Terrains and Old Dichotomies: Peacebuilding and the Gangs’ Truce in El Salvador’, Contexto Internacional 39, no. 3 (2017): 499–520; and Rachel Kleinfeld, A Savage Order: How the World’s Deadliest Countries Can Forge a Path to Security (New York, 2018).

¹⁰⁷ See note 95 above.

¹⁰⁸ David Brotherton and Rafael Gude, ‘Inclusión Social Desde Abajo’ (Washington, DC: Inter-American Development Bank, 2018), https://publications.iadb.org/publications/spanish/document/Inclusion-social-desde-abajo-Las-pandillas-callejeras-y-sus-posibles-efectos-en-la-reduccion-de-la-tasa-de-homicidios-en-el-Ecuador.pdf.

¹⁰⁹ See note 79 above.

Annex I:

Interview list and interviewee profiles

Table

Interview Number	Respondent profile	Interview place (or medium, if remote)	Date
1	Security analyst, female	Guatemala City	November 2019
2	Practitioner, youth violence prevention, male	Guatemala City	November 2019
3	Security analyst, male	Guatemala City	December 2019
4	Practitioner, youth in conflict with law, male	Guatemala City	December 2019
5	Practitioner, violence prevention, Ministry of Interior, male	Guatemala City	December 2019
6	Practitioner, youth violence prevention, male	Guatemala City	January 2020
7	Practitioner, Human Rights organisation, male	Guatemala City	January 2020
7 (2^nd interview)	Practitioner, Human Rights organisation, male	Online	May 2020
8	Practitioner, violence prevention, female	Guatemala City	January 2020
9	Practitioner, police reform, male	Guatemala City	February 2020
10	Practitioner, violence prevention, female	Guatemala City	February 2020
11	Practitioner, Human Rights organisation, female	Guatemala City	February 2020
12	Practitioner, police reform, male	Guatemala City	February 2020
13	Practitioner, police reform, male	Guatemala City	February 2020
14	Practitioner, police reform, female	Guatemala City	February 2020
15	Government official, Ministry of Interior, male	Guatemala City	February 2020
16	Security analyst, male	Guatemala City	February 2020
17	Practitioner, gender-based violence prevention, female	Guatemala City	February 2020
18	Practitioner, violence prevention, male	Guatemala City	February 2020
19	Practitioner, violence prevention and former Ministry of Interior official, female	Guatemala City	February 2020
20	Practitioner, Human Rights organisation, male	Guatemala City	February 2020
21	Security analyst, female	Guatemala City	March 2020
22	Practitioner, Human Rights organisation, female	Guatemala City	March 2020
23	Security analyst, female	Online	May 2020
24	Practitioner, violence prevention, male	Online	May 2020
25	Practitioner, Human Rights organisation, male	Online	June 2020
26	Case study NGO practitioner, male	Online	July 2020
27	Case study NGO practitioner, male	Online	July 2020
28	Case study NGO practitioner, female	Online	July 2020
29	Case study NGO practitioner, female	Phone	July 2020
30	Case study NGO practitioner, male	Online	July 2020
31	Case study NGO practitioner, female	Online	July 2020
32	Case study resident, male	Phone	August 2020
33	Case study resident, male	Phone	August 2020
34	Case study resident, female	Phone	August 2020
35	Case study NGO practitioner, male	Online	August 2020
36	Case study resident, female	Phone	August 2020
37	Case study resident, female	Phone	August 2020
38	Case study resident, male	Phone	August 2020
39	Case study resident, female	Phone	August 2020
40	Case study resident, female	Phone	August 2020
41	Case study resident, female	Phone	August 2020
42	Case study resident, male	Phone	September 2020
43	Case study resident, male	Online	September 2020
44	Case study education practitioner, male	Phone	October 2020
45	Case study resident, male	Online	October 2020
46	Case study resident, female	Phone	October 2020
47	Case study NGO practitioner, male	Online	October 2020