Abstract
The article examines the links between paramilitary groups and the Colombian state within a context of pervasive violence. Colombia represents a particularly interesting case as high-intensity violence is accompanied by the preservation of a relatively strong institutional framework. Most interpretations of this relationship consider it to be either a sign of state weakness or a centralized strategy to outsource violence. Taking a different stance, the paper argues that the existence of paramilitary groups compels us to analyze government through practices vis-à-vis the treatment of violence. A policy linking private security and counterinsurgency, crafted in the early 1990s and known as Convivir, provides an illustration of this approach.
Introduction
In September 2015 the Medellín High Court of Justice ruled against several former paramilitary members for the crimes committed by the Cacique Nutibara Bloc. 1 The decision attracted a great deal of media attention, not necessarily because of the crux of the ruling but on account of a provision defended by the head of the tribunal. Judge Rubén Darío Pinilla called for further inquiries into the role of the former president, Álvaro Uribe, in creating and bolstering paramilitary groups. One of the points made by Judge Pinilla concerned the security policies implemented by Uribe as governor of Antioquia province (departamento) between 1995 and 1997. Pivotal to these policies was the creation of “special vigilance and private security services” (servicios de vigilancia y seguridad privada) – namely, vigilante groups, which were referred to as Convivir (“living together”). These groups were created by decree in 1994 and were supposed to organize security for communities affected by insurgent activity. Uribe, who as head of a provincial executive only had limited statutory prerogatives within this policy, wholeheartedly endorsed it and actively encouraged citizens to arm themselves and collaborate with the military in counterinsurgent endeavors.
I wish to thank Laurent Gayer, Gilles Favarel-Garrigues, and the other members of the Group for Research and Analysis on Vigilantism (GRAV), as well as the two anonymous reviewers for their comments on previous versions of this paper.
Chief Justice Pinilla's opinion echoed a number of critiques that had been made against the Convivir policy since its inception. Convivir has been accused of aggravating the fragmentation of violence, which is a common trait in contemporary Colombian history, and – more importantly – of supporting paramilitary militias, who gained control of large parts of the Colombian territory between the late 1990s and the early part of the first decade of the twenty-first century. The controversy surrounding the Convivir program persists today, mainly due to the involvement of political figures who subsequently became major protagonists. Uribe – who has never denied his firm support of Convivir, and who has been accused on many occasions of being lenient on or even actively supportive of paramilitary activities – is consequently at the center of this politico-legal controversy.
Yet a closer look at the Convivir case compels us to adopt a historical and political focus, which shall not be limited to criminal and political responsibilities. Understanding the role of Convivir in the history of paramilitarism provides us with a finer picture of the complex, multifarious relationship between private violence and the Colombian state. This issue has led to a number of competing interpretations. For Gustavo Duncan (2005), paramilitary groups were “warlords,” whose power was external to the political game and to statutory institutions, and whose violent rise should be interpreted as an instance of (the now hackneyed concept of) “state failure” (for a critical analysis of the state failure literature, see Hagmann and Hoehne 2009; Grimm, Lemay-Hébert, and Nay 2014). This interpretation echoes the current interest in the issue of “warlordism” in comparative politics (Marten 2012), which links this phenomenon to a larger context of state fragility and weak governance. Other scholars have integrated the use of violence into their analyses of the political game but have considered it as a sign of “state capture” (Garay Salamanca 2008; López 2010; Romero 2011). According to this stance, statutory institutions have been diverted from undertaking their official purposes and are instead put at the service of criminal designs. More radical critiques contend that state officials deliberately subcontracted violence to paramilitary militias in an effort to deploy high levels of violent repression while preserving international respectability (Avilés 2007; Zelik 2015).
These accounts of state and paramilitary violence in Colombia reflect normative biases, defining political power and institutions not as they actually are but as scholars think they should be. Concepts such as “warlords” or “state capture” fail to account for the entanglement of violence and state formation, a constant feature in the history of states (Tilly 1985; González, Bolívar, and Vázquez 2003). While Avilés's denouncement (2007) of the links between state and criminal actors corresponds to empirically verified situations – for instance, in violent land-grabbing cases (Ballvé 2012; Grajales 2013) – his account reifies the state and conceals more than it reveals as it screens the relations of power inside the state, whether between institutions or individuals.
More fundamentally, the fact of considering the “state” to have subcontracted the exercise of violence hinders any understanding of (1) conflicts between the executive and the judiciary, (2) the controversies regarding the scope of judicial control over military activity, and (3) the balance of power inside the government. It also ignores the fact that violence in Colombia has long been a “public problem” that has generated concurrent definitions and courses of action that correspond to a diversity of institutional settings, values, and interests. More broadly, not only does it homogenize the state – which might best be analyzed as a “field” (Jensen 2001; Wacquant 2009) – it also defines it as “a freestanding object or actor,” disregarding the processes of differentiation that produce and maintain “the distinction between state and society,” an operation which is “itself a mechanism that generates resources of power” (Mitchell 1991: 90).
The Convivir policy is one example of what I call “government in the midst of violence.” This notion, developed elsewhere (Grajales 2016a), covers the twofold dimension of state action during periods of widespread and pervasive violence and proposes a more encompassing view of the link between paramilitary groups and the state. I argue that state activity is characterized both by practices of governing through violence and by a government of violence. This means that the violence of nonstatutory armed groups is, on the one hand, a resource mobilized by state agents to repress opponents and social movements and, on the other hand, a public problem that is endorsed by various public institutions, such as security agencies, peacebuilding organizations, and the judiciary. Of course, many civil wars or other situations of pervasive violence can be analyzed through this lens. Yet, Colombia provides us with a “borderline case” (for more on selecting cases in qualitative social science, see Lund 2014; Flyvbjerg 2006), as high levels of violence coexist with statutory institutions that retain political power. Moreover, the coexistence of extensive collusion between state actors and private armed groups does not preclude the possibility of violence becoming the target of public intervention, as the regulation of violence has been widely embraced by officials in various institutions since at least the 1980s.
This dual nature of nonstatutory violence helps us understand the singular position of Colombian paramilitary groups. Examining the Convivir program enables us to untangle the complex interaction between violence and the institutional legal field. Despite remaining allies with the military and key supporters of local elites for years, paramilitary groups were also categorized as criminal actors – a status that placed constraints on the collusive relations they held with other actors. The Convivir policy is intimately linked to a situation of political and legal vulnerability, as one of its key effects was to provide a legal facade that facilitated the collaboration between paramilitaries and the military. At the same time, the Convivir program was championed by its proponents as an initiative designed to tackle violence fragmentation. In addition to establishing a legal framework for the creation of regular security firms, the Convivir policy was also supposed to prevent the formation of illegal “selfdefense” groups.
Consequently, my analysis will follow two complementary approaches. First, Convivir will be assessed as an example of the government of violence, formulated at the crossroads of counterinsurgent endeavors and regulatory preoccupations. Second, it will be analyzed as a practice of government through violence, as it created a common ground for collaboration between the military, the local elites, and private-violence professionals.
The data used in this contribution were gathered through fieldwork and archival research undertaken during a period of 12 months in different regions of Colombia. They were collected as a part of my doctoral thesis on the links between paramilitary groups and the state, which was recently published in French (Grajales 2016a) and Spanish (Grajales 2017). This article's conclusions are more specifically drawn from the interviews conducted with former security officials and on the analysis of several judicial procedures against former Convivir proponents.
The inception of Convivir must be linked to the evolution of the security agenda. The treatment of violence in Colombia since the mid-1980s has been extremely equivocal. Given the proliferation of organized armed groups, security policies did not seek to eliminate or to prevent violence, but sought to limit its consequences. This is a fundamental trait of the government of violence. Here, it will not be seen as a planned and centralized design led by a rational unified actor but rather as various forms of intervention in the field of organized violence, which are based on diagnoses and hierarchies of threats that remain fluid and contingent. One of the outcomes of this political activity is the convergence of private security and counterinsurgency, which provided a favorable context for the creation of Convivir.
The Convergence of Policies
The 1980s were a time of erratic negotiations between the government and the Revolutionary Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia, FARC). The fragile truce established in 1983 was finally broken in 1990. This marked the transformation of the FARC into the main focus of security concern. Between 1991 and 1992, representatives of the government and the guerrilla met in Mexico and Venezuela in an unsuccessful attempt to resume negotiations. The government then announced a “total war,” which was intended to bend the guerrillas’ will. In response, the FARC reorganized its military structure at its 1993 conference in order to prepare for a direct confrontation with the military (Pécaut 2008; Villamizar 2017).
At the same time, the government redefined its security policies to target the guerrillas, which were now considered to be criminal organizations financed by drug trafficking and kidnapping (Leal Buitrago 2006). This was the first time that political violence came to be understood as part and parcel of the fight against crime. This public policy framework constituted the basis for the convergence of private security and counter-insurgency.
In August 1993 the president obtained a special mandate from Congress in order to regulate the creation and operation of private security companies as well as the right to bear firearms. At that time, the private security sector accounted for more than 700 firms and was expanding rapidly without any real administrative control (Arias 2009). Moreover, since the late 1980s, the Ministry of Defense had been warning the government about the increasing porosity between private security firms and drug trafficking enterprises and had stressed the urgent need to establish a regulatory framework (Krauthausen and Sarmiento 1991: 91).
The convergence of counterinsurgency and private security was made official by the Gaviria administration's (1990–1994) security white book entitled Seguridad para la gente (Security for the people). The document acknowledged the necessity for collaboration between the public and private sectors and that the actions of private groups should complement those of the state (Presidencia de la República de Colombia 1993). This convergence was strongly endorsed by the military lobby. At the time, most of the private security companies belonged to former army men. The development of the private security sector provided profitable career prospects to former military, whose specific savoir faire and social networks made them valuable assets. Following the introduction of new regulations on firearms licenses in 1993, their importance to the private security sector was reinforced by their ability to facilitate the purchase of firearms. Institutional support by the military for private security was further determined by the fact that the army had a monopoly on the production, import, and sale of firearms and supplies.
This regulatory paradigm resulted in the creation of the Superintendence of Surveillance and Private Security (Superintendencia de Vigilancia y Seguridad Privada, SVSP) – a watchdog agency that was placed under the authority of the Ministry of Defense. More critically, Decree No. 356 (of 11 February 1994) made it possible for a single individual or corporation to set up a “special security service” for their own protection but not for offensive operations. These organizations were permitted to operate in high-risk zones and to use heavy weaponry. They were presented as an answer to the difficulties faced by regular private security firms operating in guerrilla-struck zones.
Regulating Violence
The new provisions were immediately mobilized by political actors who advocated a broader remit for special security services, which they thought should serve as a basis for the development of vigilante groups with more offensive orientations. The Colombian Federation of Stockbreeders (Federación Colombiana de Ganaderos, FEDEGAN) was a key actor in this lobby and called for strong mobilization against racketeers and kidnappers. In late 1994 FEDEGAN representatives met with President Ernesto Samper, a few weeks after he took office, and pressed for stronger action against the guerrillas. They denounced the kidnapping of more than 400 ranchers in a few months and pleaded for less restrictive regulation on private security so that they could bolster militarized legal vigilantes.
These demands were favorably received by the new defense minister, Fernando Botero Zea, and by the armed forces commander, General Harold Bedoya. In December 1994 Botero was invited to FEDEGAN's annual meeting, where he announced a new government scheme (i.e. Convivir) that would not only ensure the security of individuals and corporations but also actively participate in the counterinsurgent effort. He stated that a fixed perimeter of intervention would be defined, but that Convivir groups would operate as military auxiliaries. Botero's announcement surprised his fellow cabinet members – notably the interior minister, Horacio Serpa, who harshly criticized the project. The justice minister and the peace commissioner also expressed their opposition, arguing that the measure would be counterproductive and would aggravate violence. The Senate Peace Committee believed that Convivir would pose an obstacle to peace talks with the guerrillas and requested that the government dismiss the project. In addition, a number of governors, including those from the most violence-struck regions, opposed Convivir, fearing that it would foster the development of paramilitary groups.
Botero was a very influential member of the cabinet. A Harvard University graduate, he kept close relationships with government officials in Washington and was considered by the US administration as a major asset in Colombia's engagement in the war against drugs. 2 He was also the son of the renowned artist Fernando Botero Angulo and a member of Bogotá's social elite. Botero also served as the president's campaign manager and had earned the support of the richest Colombian Liberal Party donors.
This is an ironic situation as Botero was found guilty in 1996 of participating in the illegal financing of the Samper campaign with money from the Cali cartel.
As he threw his weight behind Convivir, Botero claimed his intellectual affiliation with the Peruvian Rondas Campesinas, a network of legal paramilitary organizations controlled by the army. Shortly after he took office as defense minister, he invited Nicolás Hermoza, the commander of the Peruvian armed forces, and Hernán Garrido, a close adviser of President Alberto Fujimori, to Bogotá. Both men were pivotal in the implementation of the Rondas Campesinas and participated in working groups to design the Colombian vigilantes. For Botero, the privatization of violence was a strategic imperative. In early 1995 he declared that “no nation in history has ever been able to overcome the problem of rural criminality with the sole effort of the armed forces” and that counterinsurgency required “the support of civil society organizations” (Semana 1995a).
Despite the opposition met by Botero's project, the president approved it. Yet, the creation of Convivir did not involve a new law of decree but was merely an SVSP administrative decision that extended the scope of the special security services. Enact private security regulation using the 1994 decree, which had been issued by the preceding government, could have undermined the legal status of Convivir, as not only can administrative decisions be easily dismissed, such an approach could have been interpreted as reflecting weak support for the minister's initiative and the president's limited latitude of action.
The Convivir policy transformed the institutional role of the SVSP by linking its area of responsibility to counterinsurgency, meaning it was no longer limited to the independent task of regulating the private security sector. Thus, evidence of loose regulation should not be seen as an administrative malfunction but rather as the consequence of the political mandate given to the agency. In that context the anti-Convivir critique that the initiative encouraged the development of paramilitary groups was easily dismissed; the issue was simply that Convivir opponents and Convivir promoters had different visions of the paramilitary problem.
The justifications for the Convivir scheme are a good illustration of the project of governing violence that characterized an element of the Colombian establishment at that time. For instance, Botero believed that private armed militias would be an inevitable consequence of the armed conflict. For him, violence privatization could not be prevented, only regulated:
The dilemma for the country is not to choose to have rural security cooperatives or not. The real choice is between allowing cooperatives supervised by the state or having the uncontrolled development of self-defense and paramilitary groups created outside the law. (Semana 1995a)
As a consequence, the Convivir project was presented as a method to bring the paramilitaries back into the scope of the state and subject them to public regulation. Herman Arias Carrizosa, the head of the SVSP, expressed a similar idea:
The paramilitaries told those gentlemen who owned large estates in the Bogotá plateau, “We can give you security […] Do you want security? […] We will give it to you. Give us the money and do not ask any questions.” What response can the state provide? The possibilities are twofold. On the one hand, you can deny that reality, saying that it does not exist, that everything is false. On the other hand, you can accept that reality and try to incorporate those people into the state, next to the law enforcement agencies, such as the army, the police or the marines, with them, under their surveillance, under the wardship of the state, within a legal framework. That is the importance of Convivir. (Cien Días 1997)
Arias Carrizosa was a member of the same social class that was radically opposed to the guerrillas. His father, José Manuel Arias Carrizosa, served as justice minister and president of the Colombian Association of Banana Growers (Asociación de Bananeros de Colombia, Augura), and his brother was in the armored vehicle industry. As head of the SVSP, Arias Carrizosa never fully acknowledged the role of Convivir as something other than a self-defense association. For instance, he claimed that Convivir only possessed self-defense weapons and that most of their equipment only included communications systems. However, in 1996 the SVSP authorized the use of 445 machine guns, 373 9 mm pistols, 217 shotguns, 70 assault rifles, and 41 “restricted” assault weapons – a category that included mortars, grenades, and grenade launchers. The role of the SVSP clearly exceeded its control functions, as illustrated by 1997 press articles on conflicts of interests. According to the journalists, not only did the SVSP actively promote the formation of Convivir, but members of the board of directors were also employed as consultants for the security companies they were supposed to control (Alternativa 1997).
This brief account shows how the Convivir policy emerged at the convergence of two distinct sectors of the security field: counterinsurgency and private security. Convivir, however, was not the first of its kind to legally facilitate the collaboration between armed civilians and the state. Until 1989, Cold War legislation provided a legal basis for the creation of self-defense committees (juntas de autodefensa), which were organized groups in rural areas that acted as military auxiliaries. These paramilitary groups were later banned by the government and the Supreme Court after it became clear that they were intimately linked to drug trafficking. Promoters of Convivir were thus accused of trying to revive self-defense committees. However, Convivir was not merely an attempt to revamp a former strategy, but its reformulation under the terms of the 1990s neoliberal state. The issue of private security and state-led counter-insurgency was redefined through the engagement of corporate actors in security policy. This was fully compatible with the “private–public approach” that had been praised in the 1993 white book on security. Rather, Convivir was seen as a response to security policy objectives and to the “uncontrolled” development of paramilitarism. Moreover, the inability to eliminate organized violence highlighted the need to develop provisions that could help to govern violence, merging its dangerous and beneficial manifestations. In order to better understand the implementation of this policy, we shall now examine the practices of collaboration that it rendered possible.
Home to more than six million people, Antioquia is the most populous province in Colombia. 3 Its capital city, Medellín, has over two million inhabitants and is the country's second industrial and financial hub. But other areas of Antioquia are politically and economically marginalized. In these zones political power is exercised simultaneously by precarious public institutions, local strongmen, and armed groups.
This case study relies heavily on documents retrieved from the trial archives of several Convivir promoters. Thanks to the support of the NGO Corporación Jurídica Libertad, I obtained access to the archives of the cases brought against Alberto Osorio, Arnulfo Peñuela Marín, and Rafael Emilio García – all three of whom were accused of promoting paramilitary groups. The three trials were conducted by the Special Prosecutor's Office 29, Medellín (Fiscalía 29 Especializada de Medellín).
Because of their economic and geographic characteristics, several regions of Antioquia were at the core of the offensive launched by the FARC from 1993. At the same time, Antioquia was also one of the provinces where Convivir groups enjoyed their strongest support from the regional government and landed elites. Situated in northern Antioquia and bordering the province of Córdoba on the shores of the Caribbean Sea, Urabá – one of the most violent spots in the province and the cradle of paramilitary militias – illustrates the link between Convivir, the dynamics of the armed conflict, and the political economy of agribusiness.
Urabá exemplifies the notion of government through violence. In this region, as elsewhere in Colombia, Convivir created new opportunities for promoters of paramilitarism. Paramount to their strategy was the capacity to establish strong links with statutory institutions, especially the military. While a part of these networks were kept secret, Convivir provided the possibility to stabilize these collusive relations within a legal framework.
By the time Convivir was created, Antioquia was experiencing high levels of violence. In the late 1980s the conflict between the FARC and the paramilitaries in Urabá had subsided. The demobilization of the Popular Liberation Army (Ejército Popular de Liberación, EPL) in 1991 did not help to calm the situation, as former rebels were considered traitors by the FARC. Thus, in order to protect their lives, some of these former rebels created a new armed group called the People's Commandos (Comandos Populares), which ended up being absorbed by paramilitary groups (Martin 1997; Ortíz Sarmiento 2007; Suárez 2007). The situation in Urabá was gradually becoming a national problem, particularly because of the constant flow of forcibly displaced people (desplazados), many of whom sought refuge in Medellín.
The FARC targeted all the symbols of the state's presence, police stations in particular. Moreover, in order to finance their deployment, they intensified their racketeering and kidnapping activities, severely affecting rural entrepreneurs. The election of Álvaro Uribe as governor of Antioquia in 1994 coincided with this critical situation. Uribe, who belonged to a family of landowners, was a controversial individual: both he and his brother had been accused of participating in the training of paramilitary groups. Although the investigation was never fully conclusive, these accusations were often used by his political challengers. Despite his suspected collusion with paramilitaries, Uribe firmly committed himself to creating Convivir groups in Antioquia. His closest collaborator, Pedro Juan Moreno, who was the governor's security advisor, was a divisive character who had been repeatedly targeted by censure motions in the Provincial Assembly and accused of supporting paramilitary groups. These accusations did not stop him from actively promoting the Convivir policy throughout the province.
In late 1995 there were 48 Convivir groups in the province. Praising their effectiveness, Uribe advocated a legal reform that would allow vigilantes to use more powerful weapons and would transform some of these groups into civilian commandos acting in close collaboration with armed forces (Semana 1995b). The governor also promoted “active neutrality,” what he defined as the obligation for all citizens to actively collaborate in law enforcement by becoming informants of security agencies. These developments were severely criticized by other public officials, such as the local representative of the People's Ombudsman (Defensor del Pueblo), and both the conservative El Colombiano (the most prominent newspaper in Medellín) and the left-wing press. At the time, the left-wing Alternativa wrote:
In Antioquia, practically anyone who has 80 million pesos [approximately USD 100,000 at the time] – whether it be a landlord, a businessman, a paramilitary, a drug trafficker […] – can go to a provincial government office, where an army colonel will explain how to create one's own private army. (Alternativa 1996)
How Did Convivir Work?
During the first half of the 1990s, Urabá became a paramilitary stronghold. In 1994 Carlos Castaño, who had been an active drug trafficker and paramilitary promoter in eastern Antioquia and neighboring Córdoba province, created the Córdoba and Urabá Peasant Self-Defense Forces (Autodefensas Campesinas de Córdoba y Urabá, ACCU). Castaño had taken over the counterinsurgent endeavor from his brother, Fidel, who died in mysterious circumstances. The three Castaño brothers – Fidel, Carlos, and Vicente – played a key role in the history of paramilitarism in Colombia. Their efforts embody the convergence of counterinsurgency, organized crime, and the response of the landed elites – all of which were pivotal to the formation and development of paramilitary groups (Romero 1995, 2003; Grajales 2016a). Urabá became the center of a nationwide paramilitary network that had been structured by Castaño: the United Self-Defense Forces of Colombia (Autodefensas Unidas de Colombia, AUC). This loose coalition consisted of a number of independent violence entrepreneurs and remained in Urabá and Córdoba at least until the mid-1990s. Most of the paramilitary chiefs that marked Colombia's recent violent history had some kind of link to the region. While a sociological history of paramilitarism in Urabá exceeds the scope of this contribution, a finer understanding of the role of Convivir in the development of paramilitary groups would be conducive to better understanding security policies and the involvement of armed groups in state-led counterinsurgency strategies (Torres Bustamante 2004).
Urabá was not initially supposed to be included in the Convivir policy due to the opposition of several mayors. However, Raúl Hasbún – a wealthy banana entrepreneur from Medellín and an influential member of Augura – played a key role in generating support for paramilitarism among plantation owners. The first vigilante groups were created as early as 1996; by 1997 there were 12 Convivir groups. Although administratively distinct, these groups functioned as a single organization and were headquartered next to the 17th Brigade military base. They maintained permanent relations with the local military, primarily collaborating on intelligence sharing. Convivir groups functioned as brokers between paramilitary groups and the military, allowing the latter to transmit information to the former. In this sense Convivir was less a form of violence privatization than a readjustment of the relations between public and private actors, built upon a more forceful institutionalization of intelligence flows, that aimed at securing the positions of all parties, especially that of the military.
With the support of the plantation managers, Convivir vigilante groups deployed a communication network that covered a large portion of the Urabá territory. According to Hasbún, up to a thousand radio sets were distributed among the plantations, linking them to Convivir patrols, the army, and the police. Antennas situated within the perimeter of two military bases provided coverage for the whole region.
By passing information from the army to the paramilitaries, Convivir groups facilitated and indirectly controlled extra-legal violence, keeping it within a counterinsurgent strategy. In other words, Convivir played a part in the regulation of paramilitary violence, steering it towards objectives which were perceived as legitimate by the military. As Raúl Hasbún acknowledges:
We would have information from the army or the police […] they arrested someone but were not able to carry on with an indictment because information was lacking. Then, if we had the chance, we immediately executed or killed [sic!] the person, because we knew that he was from the guerrillas, but the justice system did not have enough evidence to act. (Raúl Hasbún, statement given to the Attorney General's Office [Fiscalía General de la Nación], 20 August 2008) 4
All legal documents are part of the archives mentioned above.
Uraba Convivir groups are the most prominent example of the instrumental character of Convivir in the development of paramilitary militias. In the mid-1990s leading characters of what would become the AUC set up a number of Convivir vigilante groups. One of them was Salvatore Mancuso, who is today serving a 15-year sentence in a US penitentiary for drug trafficking. In 1996 Mancuso created three Convivir groups that operated in the Córdoba province and secured licenses for the purchase of submachine guns, assault rifles, and other firearms. The same link between Convivir and paramilitarism in other parts of the country has been confirmed by former paramilitaries. Whenever the necessity to create a legal facade for fundraising or to share intelligence would arise, Convivir vigilantes were recruited by paramilitary chiefs themselves or by allies or frontmen, who were often members of the landed elite.
A sociological analysis of the composition of Convivir groups reveals the importance of multipositioned actors. Convivir members and paramilitaries did not represent two distinct social groups; on the contrary, most vigilantes had previously acted within paramilitary groups. However, Convivir constituted a vector of professionalization and bureaucratization of the paramilitary milieu. For instance, the transformation of paramilitaries into legal vigilantes enabled paramilitary commanders to offer their rank-and-file members employment contracts and social benefits. Plantation managers also operated as multipositioned actors.
For instance, Arnulfo Peñuela was a former army sergeant who later became a storekeeper and businessman in Urabá and was elected mayor of Carepa in 2008, home to Convivir headquarters. A few months after his election, he was arrested and convicted for his links to paramilitary groups. At his trial, it was revealed that he had been a broker between the army, Convivir groups, and paramilitaries. There is also the case of Antonio Arboleda, a former EPL guerrilla who joined the paramilitaries after his demobilization and also became the manager of six banana plantations. Officially, he was a Convivir group member and in charge of one of the group's “operative sectors.”
Peñuela's and Arboleda's dual roles as Convivir actors and plantation managers allowed them to keep strong relations with military officers. The social proximity between members of the army and representatives of the agribusiness elite structured their relations. In the words of a former military officer stationed in Urabá:
The relations between military officers and local elites are considered to be a fundamental aspect of our job by the military institution. It is said that military officers must develop strong relations
[with the elites]. That means having frequent contact with all those that matter locally: the large landowner [hacendado], the mayor, the businessman, the guys from the corporate associations [gremios]. (Interview, Bogotá, March 2011)
Consequently, people with elite profiles were in charge of maintaining Convivir links to security services and the military. Alberto Osorio, a key member of the Urabá Convivir, embodied the social characteristics required for this type of position: not only was he a successful businessman in the banana sector, he was closely acquainted with the military, having been awarded the Ayacucho medal for his “professionalism, complete devotion to the Colombian infantry and true sense of collaboration and solidarity” by the commander of the 17th Brigade. 5 Osorio was also invited by the 17th Brigade to be a panelist for a talk about the partnership between Convivir groups and the army. 6 This all clearly establishes Osorio's role of broker. Brokers are instrumental in the functioning of Convivir groups as “screens” – statutory institutions that transmit information from one side of the legal gap to the other.
Letter from Brigadier General Rito Alejo del Rio to Alberto Osorio, 7 August 1997.
Letter from Brigadier General Rito Alejo del Rio to Alberto Osorio, 2 February 1998.
As such, individuals like Osorio were in charge of maintaining paramilitary networks, which contribute to the reproduction of criminal enterprises across time and space. Rocco Sciarrone's (2000) analysis of the Italian mafia corresponds on every point to the organizational dynamics of paramilitary militias:
One of the mafia's strong points is its capacity to obtain the cooperation of other social sectors which are situated outside its organizational core – in other words, its capacity to entertain relations of collusion and complicity with different spheres of the civil society and the institutions. (Sciarrone 2000: 36)
The same analysis can be applied to the relations between Convivir groups and the agribusiness sector. I have established the importance of multipositioned individuals and the links between the Convivir intelligence networks and plantation managers. Yet, the relation between agribusiness and Convivir groups was not only a matter of interpersonal relations, it was also institutionalized within the Convivir organization. The practical functioning of Convivir illustrates this argument.
In 2001 Urabá Convivir vigilantes became a single administrative entity. First called the Papagayo Security Association and later renamed Urabá Special Services for Security and Surveillance (SEVSP), this structure officially became a nonprofit organization. In 2006, at the SEVSP's general assembly at the Medellín Country Club, 55 agribusiness corporations (owners of several hundred plantations in Urabá) came together to discuss various financial and organizational issues. The associates validated the financial report, which would then be presented to the Antioquia provincial government. The organization's administrative management was entrusted with electing assembly members, who were typically senior managers of agribusiness corporations (minutes of the general assembly, 26 April 2006).
The networks supporting Convivir groups were not only composed of military officers and agribusiness actors from Medellín. The vigilantes also kept relations with foreign companies active in the region. The example of the US agribusiness corporation Chiquita finely illustrates these relations. In 2002 Chiquita's board of directors ordered an independent audit of the payments made by its Colombian subsidiary Banadex to paramilitary groups. These payments, while apparently legal, fell under the scope of US antiterrorist legislation, as paramilitary groups were considered terrorist organizations following an administrative decision issued on 10 September 2001. Chiquita's lawyers advised the board to conclude a plea agreement with the US Justice Department. The terms of the deal included a USD 25 million fine and the delivery of several thousand documents evidencing the relations between Chiquita executives and paramilitary groups. A large share of those documents were released under the Freedom of Information Act following the intervention of the National Security Archive. 7
This is an initiative hosted by George Washington University. Thousands of unclassified documents can be retrieved at <http://nsarchive.gwu.edu> or <http://search.proquest.com>.
This episode gives us a measure of the breadth of the collusion between transnational companies and paramilitary groups in Colombia (for another example, analyzed in great detail, see Gill 2007). The Chiquita documents recorded payments to paramilitary groups as far back as 1993, the year that marked the beginning of paramilitary activities in Urabá. At that time, Chiquita was also paying the FARC and EPL guerrillas. While the company argued that it was a victim of racketeering, internal documents suggest that the FARC provided the company with security services, such as protecting plantations from common criminals and even preventing strikes. An analysis of this issue would require further data, but the evidence presents a picture of unequivocal relations between racketeers and racketed companies.
The records show that in 1996 Chiquita paid USD 21,763 to Convivir groups and over USD 200,000 to the guerrillas. However, from 1997 the paramilitaries became the primary beneficiaries of Chiquita's payments, with the rebels only receiving USD 80,000 and Convivir groups receiving USD 120,000 that year. At the end of the decade, the paramilitaries monopolized the funds and put in place a centralized system of collection. For each exported banana crate, Chiquita had to pay a “tax” of three cents. This system was very profitable, earning the paramilitaries USD 1.7 million from 1998 to 2002. The payments were made to Convivir groups and recorded in the company's financial books as “security services.” Because the money was managed by Convivir groups, Chiquita and other companies could argue that they were ignorant of the links between Convivir vigilantes (legal) and paramilitaries (illegal). According to demobilized paramilitaries, other agribusiness multinationals such as Dole and Del Monte were also among the first contributors to Convivir groups. As a matter of fact, the three-cent tax applied to all exports, irrespective of whether they were managed by Colombian or foreign companies.
The links between these companies and Convivir vigilantes were not limited to the financial level. The companies’ facilities were also used by the paramilitaries. This kind of collaboration meant that companies were sometimes involved in clear-cut criminal activities. Chiquita's subsidiary Banadex, for instance, allowed for its port to be used to unload and store 23 containers filled with arms and ammunition. An inquiry by the Organization of American States (OAS) revealed that the 3,000 AK-47 assault rifles and 2.5 million rounds of ammunition had been purchased from the Nicaraguan army by an Israeli arms dealer, Shimon Yelinek, claiming to represent the Panamanian police. However, after leaving Nicaragua, the shipment was diverted to Colombia and eventually delivered to paramilitaries, who were the actual purchasers of the arms and ammunition (OAS Secretary General 2003).
Even though a 1997 decision by the Constitutional Court later restricted the legal purchase of firearms by Convivir groups, these groups continued to serve as screen organizations for paramilitaries. By that time, paramilitary groups had already attained a more sophisticated level of national coordination through the creation of the AUC. To assess the true extent of Convivir's impact on the expansion of paramilitarism would require more information and further analysis; nevertheless, the institutionalization of collaboration networks that Convivir groups facilitated undoubtedly contributed to the extension of paramilitarism.
In a context where there are numerous violence entrepreneurs with strong links to the drug economy, the project of governing violence was seen by its promoters as a matter of security and efficiency in counterinsurgency. Although official discourse considered that such an approach should prevent the development of illegal armed groups, the evidence shows that Convivir bolstered the expansion of paramilitary groups, especially by facilitating the collaboration between paramilitaries and the military. As a matter of fact, the efficiency of Convivir groups results from their use of multipositioned actors – who simultaneously operate within paramilitary groups, the legal economy, and organized crime – to maintain and develop paramilitary groups’ social networks.
Overall, Convivir's strategic role was intrinsically linked to the fact that paramilitaries, despite being supported by sectors of the national and provincial establishments, were illegal. For Zelik (2015: 175), this was a minor aspect of the issue, as illegality would have merely been a “secondary consequence of a generalized crisis.” Yet, notwithstanding the egregiously violent context, the law was never dismissed as a political repertoire or a social field in the Colombian conflict (Lemaitre Ripoll 2009). Studying Convivir offers a good angle to approach and analyze the link between violence and criminal characterizations. As a matter of fact, scholarly literature on vigilantism (a sociological category seldom used in the Colombian case) places this relation at the core of its concerns. The development of vigilantes in very different contexts is inseparable from the labeling process that includes or excludes them from the legitimate political order (Pratten and Sen 2007; Fourchard 2008). Convivir made it possible for paramilitaries to obtain resources denied them due to their illegal status. Social networks, as well as the financial contribution of domestic and foreign companies, passed through the hands of Convivir groups. In addition, the opportunity to legally and overtly collaborate with the military and/or the business establishment brought about transformations inside the paramilitary milieu, such as the division of labor (some actors were responsible for public relations; others, financial management) and the bureaucratization of social relations and hierarchies.
Therefore, a closer look at the Convivir case advocates for a more complex analysis of the links between paramilitaries and the state. As argued in this contribution, posing the problem using the antagonistic categories of autonomy and integration (for two contrary views of this matter, see Rangel 2005; Zelik 2015) is a theoretical contradiction. An in-depth inquiry into this issue would enable us to break down the state into areas of collusion and areas of denunciation (see also Grajales 2016b), seriously taking into account the existence of legal categories and the fact that the law and criminal violence interact in various ways (Briquet and Favarel-Garrigues 2010). The most obvious reason for this socio-legal approach is a methodological one. As a matter of fact, most of the scholarly literature that has been produced since 2007 – including this paper – relies heavily on legal sources. If the questioning of the social conditions of production of this legal discourse is a basic methodological rule in social sciences (Bourdieu 1986), one should seriously consider the – apparently paradoxical – force of the law in the Colombian conflict.