Issues

The Chapultepec Peace Accords ended an armed confrontation between the Salvadoran military government and the Farabundo Marti National Liberation Front (FMLN) guerrillas, which resulted in approximately 70,000 deaths and 1,000,000 displaced people (^{ICG, 2017}). However, that milestone marked the beginning of a transition to a new cycle of violence. By the end of the 1990s, the U.S. government aggressively deported thousands of ex-convicts to Central America’s northern triangle (Guatemala, Honduras, and El Salvador). Members of rival street gangs, Mara Salvatrucha (MS-13) and Barrio 18, returned to the country in a post-conflict scenario marked by institutional weakness, a broken social fabric, and a destroyed economy, conditions that aided their quick expansion.

Until the beginning of the 2000s, the Salvadoran government paid little attention to the gangs, allowing them to operate with a certain degree of freedom. However, a few months before the presidential elections, where a resounding defeat of the ruling Nationalist Republican Alliance (ARENA in Spanish) party was in sight, then President Francisco Flores (1999-2004) launched the Mano Dura policy, pointing to these criminal groups as allegedly responsible for an outbreak of crime and violence that was supposedly affecting the small Central American country (^{ICG, 2017}). However, official figures refitted that thesis: at the time, El Salvador was going through the calmest period since the end of the counterinsurgency war, and violent deaths attributable to gangs represented about 10% of the national total (^{Aguilar, 2019}, p. 24).

The first anti-gang strategy was designed with an electoral zeal typical of punitive populism:, as evidenced by the fact that its launch was accompanied by a narrative that misrepresented its power and influence, ignoring its socio-economic origin (^{Hernandez- Anzora, 2017}). The demonization of the maras responded to a dynamic between electoral supply and demand, conditioned by the legacy of authoritarianism left by the armed confrontation, and reflected in some elements of the Salvadoran political culture, such as the discourse on crime (^{Amaya, 2017}). Consequently, it is possible to say that former president Flores approached the gang phenomenon by reproducing negative perceptions associated with the imaginary of the internal enemy, seeking to generate public anxiety.

This way, the president initiated a repressive effort against gangs through police raids aimed at “massively and indiscriminately capturing all those who, judging by their appearance, were presumed to be gang members” (^{Aguilar, 2019}, p. 13). In addition, there was the application of controversial transitory legislation, later declared unconstitutional by the Supreme Court on the grounds that it violated the Convention on the Rights of the Child (^{Wolf, 2017}, p. 50). This approach allowed ARENA to win a presidential victory, which is why it continued its approach under the name Super Mano Dura, with a greater punitive emphasis but considering prevention and rehabilitation components: Mano Amiga and Mano Extendida, aimed at youth in vulnerable situations and incarcerated youth, respectively. However, the low investment, the delay in its implementation, and the reduced number of participants affected and reduced its scope (^{ICG, 2017}, p. 16).

This harassment induced a sophistication of gang activity in response to the crusade undertaken by the government. Paradoxically, the allocation of prisons by membership made it possible to centralize operations, strengthen group cohesion, and manage the violence perpetrated by their cells in the streets through economies of protection (^{Lessing, 2017}). The collection of extortion taxes was their main means of subsistence and an instrument of territorial control, enabling them to create a para-state in marginalized communities. In addition, this anti-gang policy accentuated the use of lethal violence in three directions: to reinforce internal discipline, attack rivals who contributed to their persecution, and intimidate the population that cooperated with the police (^{Reyna, 2017}, p. 21). Consequently, this punitive strategy dramatically transformed gang behavior.

Although the Mano Dura policies were electorally profitable for the ruling party in the beginning, the public security crisis that followed their implementation led Salvadoran authorities to discontinue them progressively (^{Aguilar, 2019}). During their enforcement, the homicide rate practically tripled, while the number of extortions increased 14 times, as seen in Figure 2. This, in addition to prison overcrowding caused by relentless criminal prosecution, as the number of prison inmates doubled (^{ICG, 2017}, p. 10). Consequently, the official statistics more than demonstrated the ineffectiveness of these policies regarding the original objective: to combat violence and criminality associated with gangs, seriously compromising ARENAs continuity in the presidency.

Source: Created by author from ^{PNC, 2020}; ^{USAID/PNUD, 2021}.

Figure 2. Evolution of the homicide and extortion rates in El Salvador (2003-2014) 

After the FMLN came to power with Mauricio Funes (2009-2014), anti-gang policy took an erratic course. Initially, recognizing the need for a comprehensive approach, the new president launched a National Security and Coexistence Policy focused on crime prevention, control, and repression; social reintegration; attention to victims; and institutional reform (^{Wolf, 2017}, p. 230). However, the idea did not prosper because of 1) the lack of bureaucratic experience necessary to implement the planned changes; 2) the lack of inter-agency coordination to prioritize the lines of action; and 3) the collapse of police capacities in the face of a deterioration in the security situation reflected in an extreme homicide rate associated with gang activity (^{Van Der Bough & Savenije, 2013}).

Naturally, the level of alarm among the Salvadoran population grew, as did questions about the inability of the new authorities to confront the public security crisis (^{Aguilar, 2019}). This situation prompted the militarization¹ of the fight against gangs, an option essential to demonstrate control and strength. Because 73.9% of the public opinion supported this option (^{IUDOP, 2009}, p. 86), the government significantly increased the number of soldiers in the fight against crime, giving them greater functional autonomy, and expanding their powers and competencies (^{Amaya, 2012}, p. 76). Therefore, military participation in public security went from being an exceptional resource to becoming a regular procedure. Between 2009 and 2014, the military personnel assigned to this sector increased 28%. During this period the Army participated in a wide range of plans (^{Aguilar, 2019}, pp. 39-40)², in contravention of the reforms achieved after the Chapultepec Accords.

A Law for the Prohibition of Gangs was also enacted, condemning their legalization, financing, and third-party collaboration. This action constituted more of a symbolic reaction to the public commotion, as instruments for their criminal prosecution already existed (^{Van Der Bough & Savenije, 2015}, p. 166). Seeking to repeal the law and initiate a dialogue, the maras forced a two-day national public transportation strike that, “according to official figures, mobilized 60% of public transport, causing a 40% commercial activity drop in the main cities and a loss of around 24 million dollars” (^{Martinez, 2019}). With this powerful gesture, they believed they had enough relevance to be heard as any other political actor (^{Cruz, 2013}). Due to this, Defense Minister General David Munguía Payés requested that the maras were to be considered a serious security threat: “The gangs want to scare the population, show their strength. A democratic government like ours cannot negotiate with criminal organizations” (^{Martinez, 2019}).

This discourse permeated the vision of the former Salvadoran president, causing a shift in his anti-gang policy: to intensify the repressive effort. As a result, in November 2011, Munguia Payes took over the Ministry of Public Security, blaming the maras exclusively for 90% of the murders, dismissing the impact of other criminal actors on the lethality rate (^{Vasquez & Marroquin, 2014}). Based on this premise, the new minister promised to reduce homicides by 30% in one year, telling the press that to achieve this goal, he was preparing a massive repression scheme (^{Aguilar, 2019}). Nonetheless, violence continued to rise, as shown in Figure 2, demonstrating the ineffectiveness of militarization in combating gangs and forcing the Funes administration to modify its strategy, opting for the path of agreements. This is also confirmed by the trends expressed in Figure 3.

Source: Created by the author from ^{PNC, 2020}; ^{USAID/PNUD, 2021}.

Figure 3. Monthly evolution of homicide and extortion rates during the gang truce 

Although it is estimated that the agreement would have saved 5,501 lives (^{Katz & Amaya, 2015}, p. 43), its detractors argue that it would have allowed rival gangs, MS-13 and Barrio 18, to expand geographically, dividing up the disputed areas without the need to resort to aggression or assassinations, generating an undesirable result: consolidating a pax mafiosa in Salvadoran territory (^{Felbab-Brown, 2020}, p. 33). This problematic scenario also hindered the execution of the sanctuary municipalities project because these criminal groups never relinquished control over their domains. This is evidenced by the fact that extortions varied only 5% during the term of the pact, as shown in Figure 3.

Without a source of funding or alternative means of subsistence, it was difficult to think that the gangs would stop taxing communities under their control. Moreover, there was a strategic dilemma that prevented them from stopping extortion, as Adam Blackwell explained in an interview with the digital newspaper El Faro: “[It was impossible to demand that these groups give up] their main source of power if the government remained on a war footing, because it would be illusory to ask them to surrender and then be executed by the enemy” (^{Sanz, 2012}). His statement suggests that the maras never had any real intention of making commitments to their criminal agenda, such as abandoning illicit activities (^{Cockayne et al. 2017}).

For this reason, it is possible to say that the negotiated pacification process with the gangs never achieved social legitimacy and that there was a high degree of skepticism among the Salvadoran population regarding its true impact or supposed positive externalities in daily life (^{FUSADES, 2013}, p. 123). In fact, this situation was clearly reflected in public opinion polls conducted during the gang truce, which also did not indicate significant improvements in victimization rates. Thus, in December 2012, 19.9% of those surveyed said they had been victims of a crime in the past year. While in the same period of 2013, this figure reached 20.5%, an increase of 0.6% (^{USAID/PNUD, 2021}).

The discovery of 97 clandestine graves by the Institute of Legal Medicine raised doubts on the veracity of the drop in homicides (^{HIIK, 2014}, p. 82). ^{Cruz & Duran- Martinez (2016}) suggest that such a situation would indicate a hypothetical capacity of the maras to strategically manipulate the visibility of the violence³, looking to make a convincing simulation of compliance with the agreement. Thus, the gangs would have concealed the killings to avoid attracting repressive state efforts and implying a deceptive reduction of homicidal violence. In this sense, it is worth noting that during the truce, the rate of missing people rose from 39.3 to 54.5 per 100,000 inhabitants, an increase of 38.7% (^{USAID/PNUD, 2021}). Therefore, the negotiated solution was a damage control strategy, as it did not seek to modify the scale of predatory crimes committed by the gangs but rather to control their behavior: low profile, without a confrontational stance, and exhibiting moderate levels of lethality.

Behavior

The Funes administration proved incapable of structuring a congruent narrative about the gang phenomenon -beyond its criminal nature- that would allow for the validation of the pacification process as a viable option to solve the public security crisis and to reach common minimums among the population to guarantee its sustainability. Under this logic, it is worth mentioning that the lack of transparency about its role in the dialogue exposed the Salvadoran government to ethical questions associated with an alleged corruption of the bureaucratic apparatus as, by deciding to negotiate with the maras, it was breaking its anti-gang legislation, setting a bad precedent on the rule of law in the small Central American country.

However, it should be noted that the intelligence services informed former President Funes and Munguía Payés about possible dialogue risks, thus increasing transparency (^{Felbab-Brown, 2020}, p. 29). From this angle, it is worth remembering that the exploratory contacts with both gangs (MS-13 and Barrio 18) were developed stealthily by the unofficial mediators to facilitate consensus building at an early stage, as well as to prevent government authorities from being compromised or singled out in the event of failure. Fabio Colindres and Raul Mijango led the team in charge of doing good offices and reporting the progress of the negotiations, whose profiles inspired confidence in the counterpart: the moral authority of the former military chaplain and the honorability of the recognized former guerrilla commander.

Had the negotiations been known in their early stages, they would surely never have prospered due to the strong public repudiation of the very idea of making a pact with the community’s main public enemy: the maras, a perception created by the Salvadoran establishment and reinforced by negative media coverage. Therefore, a communications strategy was needed to tie up loose ends, close the flanks of attack, or mitigate the collateral damage caused by the publication of the report in the independent newspaper El Faro, which detailed the negotiation process and subsequent transfer of some thirty members of the largest groups from the maximum-security prison Zacatecoluca to prisons with fewer restrictions (^{Felbab-Brown, 2020}, p. 29). Even so, the dialogue process continued.

At the beginning of the peace process, Mauricio Funes found himself in a difficult position due to the erratic course of his anti-gang policy, reflected in the failed attempt to install a comprehensive approach and the counterproductive militarization of its combat. In addition, the weakness, corruption, and inefficiency of the Salvadoran security forces hindered the implementation of his approach. Meanwhile, the maras already exercised tight territorial and socioeconomic control, replacing the state in providing basic services: their members and the dominated communities were totally merged (^{Felbab-Brown, 2020}, p. 12). Moreover, while the gangs allowed their cells or clicas to act with a certain degree of autonomy, the leaders possessed a high capacity for command and cohesion that allowed them to better negotiate.

This cohesion was demonstrated by the abrupt drop in murders observed as soon as the truce began to be in force and the capacity to maintain it relatively stable for 15 months, thus contributing to improving public security conditions during its duration, which was the gangs’ main negotiating tool with the Salvadoran government. However, the loss of control of the gang leadership in prison over the actions of the clicas on the outside, after the blocking of the communicating vessels, prevented them from continuing to regulate homicidal violence. Even so, the gangs were in a position of strength due to their size, geographic reach, and social capital, having enough capacity to seal or implement agreements. Moreover, the involvement of the two main maras (MS-13 and Barrio 18) neutralized the risk of possible sabotage by adversarial groups.

Unlike the Salvadoran government, which left the negotiations in the hands of third parties, taking away their official character, the two main gangs were negotiating from a position of leadership. However, it should be emphasized that both negotiating teams were legitimate and representative and had a direct channel of communication with their counterpart. For this reason, it was to be expected that the agreements reached would be sufficiently forceful to be accepted by both sides. Although the gang leaders’ demands were reasonable, including improved prison conditions and implementation of social intervention projects in their communities, certain concessions made were difficult to assimilate, such as the omission of an anti-gang legal framework or the withdrawal of the military from public security functions (^{Felbab-Brown, 2020}, p. 18).

There are two points worth noting regarding the previous. On the one hand, crucial issues were not discussed during the dialogue, including the possible repeal or reform of anti-gang legislation, the granting of judicial benefits (amnesties or reduced sentences to gang members), and the implementation of mechanisms for reparations to their victims. The exclusion of these issues from the negotiation agenda was a normative and strategic error because “if you don’t hold the gangs accountable for their acts, you lose credibility with them” (^{Felbab-Brown, 2020}, p. 33). In addition, it is necessary to mention that the agreement did not contemplate the dismanding of the maras, only the conditional cessation of their illicit activities within the framework of the sanctuary municipalities project, significantly reducing their potential for transformation. Yet again, we see that the pact was not conceived as a strategy for the sustained reduction of lethality but rather as a form of control and change of behavior.

On the other hand, it should be noted that, during the ceasefire, the Funes administration proved incapable of adequately managing the criminal agendas of the gangs, which is why the dialogue process led to the strengthening of these criminal groups as privileged interlocutors with the Salvadoran government. This was reflected not only in the tenor of the concessions made by Munguía Payés but also in the fact that they flouted the conditions imposed by the government: they continued extorting communities under their control and committing murders underhandedly. Undoubtedly, the verification of the transgressions to what was agreed also influenced the change of strategy adopted by the next minister of public security, Ricardo Perdomo, thus truncating the process of pacification negotiated with the two main maras in El Salvador.

Attitudes

In 1992, the FMLN sealed a peace agreement with the military regime seeking to end a bloody civil war. Ironically, 20 years later, it was the government of the former Salvadoran guerrilla the first to negotiate a nationwide ceasefire with the largest gangs, given the meager results of the punitive policies implemented. However, the Funes administrations attitude became ambivalent, making it difficult to determine the real support of the bureaucratic apparatus for the pacification process. While Munguía Payés’ support for the negotiating team was vital to its progress, the fierce opposition of Attorney General Luis Martinez reflected a high degree of political dissidence. Moreover, when the talks aired, generating strong popular rejection, Mauricio Funes blamed the head of public security and the mediators.

This hesitant attitude prevented the Funes administration from transforming the agreement into a pacification strategy that would lead to a sustained reduction in criminal violence. Moreover, because the truce had mainly been motivated by the need to obtain benefits for the presidential elections, the government hesitated to assume it as its own, given its unpopularity, or to abandon it when Munguía Payés was removed as minister of public security. Consequently, the social polarization around the maras and their instrumentalization as an electoral tool influenced the decision to agree on a truce to reduce the high rates of homicidal violence.

Due to the widespread rejection caused by the pact, the next president, Salvador Sánchez Cerén, not only ruled out the possibility of continuing it under his administration but also redoubled repressive efforts against the gangs, despite having secretly negotiated the electoral support of these criminal groups in the run-up to the ballot. This was how his government sought to frustrate the incitement to new peace negotiations, criminalizing both the calls for dialogue and its promoters and mediators (^{Tabory, 2016}). In fact, both Mijango and Munguía Payés would later be arrested and prosecuted on charges of unlawful association in the context of the truce agreed upon during Funes’ five-year term (^{Carlson, 2020}).

During the first stage of the dialogue with the Salvadoran government, the gangs had a committed attitude because, although they used the negotiation process to further their criminal agendas, they maintained a high degree of self-criticism regarding their responsibility for the public security crisis. “The leaders in prison had a positive attitude and wanted to collaborate; they didn’t want to see their children make the same mistakes” (^{Felbab-Brown, 2020}, p. 25). However, their attitude became distrustful when the Funes administration failed to comply with the conditions agreed upon, such as the abandonment of the sanctuary municipalities project, the obstruction of the negotiating teams efforts, and the military’s offensive launch after the abrupt departure of the truces architect.

Similarly, the truce led to a political awareness of the gangs based on the public ostentation of their monopoly of violence in relation to the legitimate government, using the homicide rate as a tool for negotiation, or, rather, for pressure (^{Hernandez-Anzora, 2017}). Undoubtedly, another instrument used by these criminal groups was the territorial control accumulated during the implementation of sanctuary municipalities to the detriment of the sovereignty of the State in such areas, allowing them to tip the balance towards a specific candidacy and openly intervene in the results of an electoral process, as happened with the FMLN during the 2014 presidential runoff. From this angle, it should be noted that the exchange of bullets for votes in favor of political elites sponsoring their criminal agendas validated a corrupt symbiosis between the two, posing a serious risk to the legitimacy of the democratic system.

Nevertheless, it was possible to observe a certain political awareness when agreeing to the truce, as the gangs acted strategically, leaving aside their traditional enmity to strengthen their group cohesion based on a collective identity reflected in the ability to unify criteria on the contingency, speak with one voice through joint communiqués, or negotiate agreements that expressed common demands (^{Cantor, 2016}). It is possible to affirm that the controversial pacification process with the government of Mauricio Funes allowed the largest gangs to dialogue and build mutual trust, demonstrated by their coordinated response to the re-launching of the military offensive by Perdomo.

On the other hand, the OAS showed a cooperative attitude, being the main international support for the negotiation process: it not only encouraged the promotion of dialogue between the MS-13 and Barrio 18 so that they would learn to communicate in a civilized manner, instead of resorting to assassinations or attacks, but they also sought to provide its expertise on conflict mediation in the technical committee created to address the root causes of the gang phenomenon. The will of the OAS to aid the process contrasted with the aggressive attitude of the U.S. government regarding the truce due to the size of the gang diaspora in its territory and the multiple economic and political-diplomatic efforts committed to combating them.