“Violence is most persuasive and most successful when it is threatened and not used… The fact that a kidnap victim is returned unharmed, against receipt of ample ransom, does not make kidnapping a nonviolent enterprise.”
—Thomas Schelling, Arms & InfluenceIntroduction
On February 4, 2008, six million people took to the streets for the “Marcha de No Más” (“March of No More”) against the left-wing Fuerzas Armadas Revolucionarias de Colombia (Revolutionary Armed Forces of Colombia, or FARC). In the Colombian capital of Bogotá and another 192 cities worldwide, political, civil, and religious leaders came together with a singular focus: ending the FARC’s long-standing practice of kidnapping civilians (El Tiempo 2008). A “pantheon of actors” that united on social media forged a “social consensus that kidnapping was unjustifiable,” dealing a fatal blow to the once-popular rebels’ image (Fattal Reference Fattal2014).Footnote 1 Twelve years later, enmeshed in a truth and reconciliation process,Footnote 2 FARC leadership suddenly accepted responsibility for what they deemed “grievous mistakes.” They declared in September 2020,
From the bottom of our hearts, we ask public forgiveness of all of our kidnapping victims and their families… Kidnapping was a very serious mistake that we can only regret. It left a deep wound in the souls of those affected and mortally wounded our legitimacy and credibility. We took away the most precious thing: their freedom and dignity (Semana 2020).
Before this apology, the FARC had kidnapped more than 21,000 Colombians during the longest running insurgency in the Western Hemisphere (JEP 2021).
Ransom kidnapping—forceful abduction accompanied by monetary demands that condition the victim’s release—is a global, costly, and underexamined form of political violence. Though kidnapping represents only 7% of Global Terrorism Database (GTD) attacks, estimates suggest that 75–80% of worldwide kidnapping goes unreported (Forest Reference Forest2012a; Gallego Reference Gallego2019).Footnote 3 194 of the 231 armed groups included in both the Uppsala Conflict Data Program Actors Dataset and the GTD have kidnapped. Ransoms comprise an estimated 15% of global terrorist financing, representing what the U.S. Treasury Department has called “the most significant terrorist financing threat today” (Cohen Reference Cohen2012).
Rebel leaders have long touted the many concrete benefits that kidnapping can provide. In The Minimanual of the Urban Guerrilla, Carlos Marighella called kidnapping “a useful form of propaganda for the guerrillas,” and “form of protest against the penetration and domination of imperialism in our country” (Reference Marighella1969, 68). According to the leader of Al Qaeda in the Arabian Peninsula, “Kidnapping hostages is an easy spoil, which I may describe as a profitable trade and a precious treasure” (Callimachi Reference Callimachi2014). And yet, kidnapping presents significant costs to its perpetrators. It is a costly use of combatants’ time. During the days, weeks, or years it takes to plan, execute, and conclude a kidnapping, perpetrators’ attention is diverted from combat. Hostages must be kept alive and minimally healthy to secure an exchange, requiring food, medicine, and shelter. Armed groups may face increased counterinsurgency or policing while holding a hostage. As the Marcha de no Más suggests, kidnapping might damage a group’s reputation long term: advancing military attacks on the battlefield confers combatant legitimacy; abducting civilians is a war crime.Footnote 4 Given these obvious costs, why did the FARC kidnap? As a representative of Colombia’s Truth Commission put it, “Why attack whom you’re supposed to defend?”Footnote 5
This article focuses on Colombia to offer an explanation for why groups kidnap for ransom in civil war. I argue that rebel groups use kidnapping to enforce their taxation and protection rackets. Groups choose kidnapping as a lucrative way to punish those who refuse to pay the groups’ taxes, as well as a strategic means to compel future cooperation. Combining funding and force, kidnapping uses selective violence to punish and advertise the cost of shirking. However, as the case of the FARC suggests, this presents a temporal trade-off for kidnappers: although rebels may seek kidnapping’s short-term enforcement benefits in the midst of insurgency, they may downplay the long-term costs of attacking civilians. I test this argument using evidence from extensive interviews with and about actors from the Colombian civil war, including ex-combatants from the Revolutionary Armed Forces of Colombia (FARC) and National Liberation Army (ELN) guerrilla groups, as well as Colombian military, police, and private security personnel.
In doing so, this article makes several theoretical and empirical contributions. First, in showing how kidnapping is connected to rebel taxation, I illuminate important links between conflict funding, rebel governance, and violence against civilians. Second, this article features novel testimony from demobilized ex-combatants and hostage negotiators combating kidnapping in Colombia. Finally, this analysis contributes to the growing literature on patterns of political violence (Gutiérrez-Sanín and Wood Reference Gutiérrez-Sanín and Wood2017), building on and challenging existing explanations for civilian victimization. Specifically, my finding that FARC and ELN rebels kidnapped tens of thousands of ordinary Colombians for tax enforcement suggests that nonlethal violence, rather than killing, is an effective way to prevent shirking. Targeting satisfied both financial and ideological imperatives, and kidnapping violence depends on strong internal organizational control.
This article joins and builds on a small but growing literature that examines hostage taking in armed conflict (see Forest Reference Forest2012b; Gilbert Reference Gilbert2020a; Loertscher and Milton Reference Loertscher and Milton2018). In particular, it shares common ground with Shortland (Reference Shortland2019) in the centrality of protection rackets for explaining kidnapping behavior. Though Shortland focuses on the role of the private sector in providing extralegal governance of kidnapping violence, she also introduces the idea that kidnapping is related to criminal protection rackets. Specifically, Shortland suggests that kidnapping is disequilibrium behavior resulting from the information asymmetry of a population ignorant of their local “protector.” I build on this idea by developing a specific theory of rebel taxation and ransom kidnapping and rigorously testing it with evidence from the Colombian civil war.
Kidnapping and Explanations for Civilian Victimization
A growing body of scholarship explains why civilians are victimized in conflict. Scholars have examined lethal violence (Downes Reference Downes2008; Kalyvas Reference Kalyvas2006; Valentino Reference Valentino2014) and nonlethal forms of violence including displacement (Steele Reference Steele2017), forced recruitment (Blattman Reference Blattman2009), sexual violence (Cohen Reference Cohen2016; Wood Reference Wood, Kalyvas, Shapiro and Masoud2008), and torture (Sullivan Reference Sullivan2014). Taken together, these tactics comprise a repertoire of violence, “a set of practices that a group routinely engages in as it makes claims on other political or social actors” (Wood Reference Wood2009, 133). Despite this rich growth in research and simultaneous surge in kidnapping by groups from the Tamil Tigers to the Taliban, few studies have attempted to explain systematically why some but not all armed groups use this tactic.
Kidnapping is a form of hostage taking—the “seizing or detaining of an individual coupled with the threat to kill, injure, or continue to detain such individual in order to compel a third person or governmental organization to take some action” (Hostage Taking, 18 U.S.C. § 1203 [2020]). Two important features distinguish ransom kidnapping from similar forms of violence. First, ransom kidnappings are accompanied by monetary demands of third-party targets. The threat of further violence must be conditional—it is violence “avoidable by accommodation,” with assurances that satisfaction of the demands would prompt the hostage’s release (Schelling Reference Schelling1966, 2). The victim and target are separate persons: the victim is the hostage, whereas the target is the recipient of the demands, who can make concessions or not.Footnote 6 Second, ransom kidnappings are abductions—a subset of hostage taking in which the perpetrator moves victims to a new location rather than holding them in place as in hijackings or embassy sieges.
Hostage-taking violence comes in many forms, and its targeting, technique, and frequency vary (Gutiérrez-Sanín and Wood Reference Gutiérrez-Sanín and Wood2017); perpetrators may demand prisoner exchanges, policy change, and publicity (Gilbert Reference Gilbert2020b). However, this investigation focuses on ransom kidnappings for two central reasons. First, ransom kidnappings appear to represent the vast majority of kidnappings worldwide.Footnote 7 Second, ransom kidnapping is a priority for policy makers. Beyond the massive sums paid to transnational groups, ransoms are a pervasive form of local extortion. Armed actors’ ability to extract payments from local communities presents a persistent governance challenge, as groups coerce cooperation in their spread of violence.
Unlike other forms of civilian victimization, kidnapping explicitly links funding and force. Yet, the monetary component should not obscure that kidnapping is indeed violence inflicted to further political objectives. In addition to the “take,” when victims are forcibly abducted at knife or gunpoint, hostages face varying levels of violence during captivity. For example, among the 39,058 verified kidnap victims in Colombia, 45% reported intimidation with weapons and 17% reported being tortured in captivity (CNMH 2013a). Hostages were forced to march through the jungle for weeks on end and required to defecate in front of armed guerrillas. They were isolated, beaten, raped, and frequently put in chains (JEP 2021). They also endured psychological abuse. As a former FARC hostage recalled, “I cried—but not out of sadness. Out of ire, out of impotence. When you’re in captivity like this, you know neither when, nor how, nor if you’ll ever be rescued.”Footnote 8
While there are no existing explanations for ransom kidnapping that have been formally tested in the political violence literature,Footnote 9 there are some common intuitions. For instance, one might assume that rebels kidnap when they need money. If this were sufficient, we could expect poor groups to kidnap wealthy individuals. However, most rebellions need money, and not all kidnap, whereas some of the wealthiest rebels kidnap often. Given kidnapping’s high costs, it is especially puzzling that rebels with alternative funding sources in their “resource portfolios” kidnap for ransom (Rettberg and Ortiz-Riomalo Reference Rettberg and Ortiz-Riomalo2016). Indeed, the most prolific kidnappers increase hostage taking while profiting from resource-related extortion.Footnote 10
Kidnapping might instead be used for publicity. Hostage taking has long attracted outsized media attention, and Colombia has been no exception (Fattal Reference Fattal2014; Reference Fattal2018). In the hijackings and embassy sieges that peaked in the 1970s and 80s, perpetrators staged spectacular, media-seeking violence, explicitly demanding media coverage or holding high-profile hostages. Publicity usefully explains kidnappings of politicians and journalists, but its role in ransom kidnappings is less clear, as most are never publicized (Gilbert Reference Gilbert2020a).
Because these accounts do not seem to explain variation in ransom kidnapping, I apply other theories of civilian victimization to deduce explanations for kidnapping. One primary explanation for civilian victimization describes violence as a solution to challenges of territorial control. For instance, proxying lethal violence for violence in general, Kalyvas argues that rebels use violence to punish defection when they have near, but incomplete, control of a territory (Reference Kalyvas2006). Steele argues that armed groups use information about civilians’ loyalties to displace rivals in their bid to conquer territory (Reference Steele2017). These arguments would expect kidnappings to occur in areas of near but incomplete territorial control and that rebels would kidnap to target disloyalty. Other scholars see violence stemming from principal-agent problems, a negative externality of weak leadership, centralization, or hierarchy. Poor internal control may derive from a deliberate trade-off between order and secrecy (Shapiro Reference Shapiro2013) or from recruiting greedy, undisciplined “opportunists” (Weinstein Reference Weinstein2007). Asal, Rethemeyer, and Schoon explain kidnapping thus—a form of violence correlated with poor internal control, which does “not require institutional infrastructure or specific skills” (Reference Asal, Rethemeyer and Schoon2019, 401). Several implications follow: without explicit orders, kidnappings would be crimes of opportunity, captivity would be short, and hostages should have a high likelihood of being killed, as combatants would not be able to devote much time to resolving them successfully.
Other explanations related to organizational imperatives may explain kidnapping’s pattern of political violence. For instance, some scholars have shown that civilian victimization contributes to bargaining problems, improving rebels’ negotiating position by imposing costs on the government for continuing to fight (Hultman Reference Lisa2007; Walter Reference Walter2009). Rebels might improve bargaining leverage by capturing state agents or other prominent hostages to increase bargaining leverage. IdeologyFootnote 11 can play several different roles to shape and constrain armed group behavior (Thaler Reference Thaler2012). It can frame organizational strategy and structure (Kalyvas and Balcells Reference Kalyvas and Balcells2010), provide normative commitments and constraints (Gutiérrez-Sanín and Wood Reference Gutiérrez-Sanín and Wood2014), or legitimate target selection (Drake Reference Drake1998). Among left-wing groups, communist organizations commit less rape (Hoover Green Reference Hoover Green2018), whereas Maoist groups institute more effective forms of governance (Mampilly Reference Mampilly2011). We might expect to see kidnapping by groups for whom it maps onto ideological commitments, such as left-wing rebels that kidnap the wealthy elite or Islamist insurgents targeting proponents of Western mores.
These explanations fall short in explaining variation in ransom kidnapping. The argument that violence follows territory is dissatisfying, as kidnapping proliferates within and outside areas of rebel control, including in major cities. Principal-agent problems cannot explain kidnapping violence. As with recent findings regarding torture or sexual violence as organizational policy (Richardot Reference Sophie2014; Wood Reference Wood2018), systematic kidnapping depends on deliberate organizational adaptations; it is a “precious treasure” explicitly ordered by leaders. Captivity is long-lasting, and most hostages survive.Footnote 12 Bargaining explanations also provide an incomplete account of kidnapping violence. Rebels kidnap for leverage in negotiations, though they often capture state agents for prisoner exchange, not civilians for ransom. Ideology also seems insufficient to explain rebel kidnapping, as only a quarter of leftist groups kidnap frequentlyFootnote 13 and neither leftist nor Islamist ideology predict kidnapping (Forest Reference Forest2012b). Still, ideology may feature in designating appropriate targets, as explained below. What then explains ransom kidnapping? The answer, I argue, is related to armed group funding—but not in the way we might expect.
Enforcing the Protection Racket
Consider that a group of people has decided to take up arms in a political fight against the state. To pay recruits, and buy weapons and other materials, they will need to raise a significant amount of money. They may get lucky: some armed groups have unfettered start-up funding for their fight, such as voluntary support from local patrons, international organizations, diaspora communities, or foreign governments (Adamson Reference Adamson2005; Byman Reference Byman2005). Others rely on natural resources—cultivating and selling drugs or expropriating oil fields or mines (Collier and Hoeffler Reference Collier and Hoeffler2004; Felbab-Brown Reference Felbab-Brown2009; Weinstein Reference Weinstein2007). But lacking any of these sources, groups need to find substantial funding quickly. To do so, they may turn to predatory funding—nonvoluntary resource collection including theft, extortion, and taxation enforced by violence (Levi Reference Levi1988; Sabates-Wheeler and Verwimp Reference Sabates-Wheeler and Verwimp2014). This could include payoffs from local businesses, ordered taxes, or fees at transportation checkpoints (Mampilly Reference Mampilly2011; Revkin Reference Revkin2020).
Scholars have long investigated the decisive role of predatory funding in building political organizations. Nascent European nation-states behaved like criminal “protection rackets,” extracting from the population to wage war and offer protection (Tilly Reference Tilly, Evans, Rueschemeyer and Skocpol1985). This criminal model can explain the behavior of state and nonstate actors alike: extracting money from the population helps the racketeer improve its own internal capacity while raising funds for the fight. Extortionists may extract unpredictably, but regularizing the transaction benefits both population and racketeer. Populations prefer “stationary bandits,” predatory actors who repeatedly tax locals at a less than maximum rate, to “roving bandits,” who unreliably extract heftier sums (Olson Reference Olson1993; Sabates-Wheeler and Verwimp Reference Sabates-Wheeler and Verwimp2014; Shortland Reference Shortland2019). The racketeer prefers to encourage production rather than capital flight (Rodriguez-Franco Reference Rodriguez-Franco2016) and generate predictable, ongoing revenue.
When extraction is regularized in this latter way, it manifests as a tax. Taxation is the imposition of costs “administered according to publicly known rules and procedures,” as opposed to in an arbitrary and unpredictable manner (Revkin Reference Revkin2020). Like the state, rebels can systematically and predictably collect revenue. Rebels also decide what, precisely, they wish to tax—whether to capture wealth or to regulate and charge for certain activities and behaviors. Whereas the former might look like property or income taxes, the latter might look like tolls, customs, or business licensing fees. Predatory funding can thus be classified on two dimensions—the degree to which the process is regularized or ad hoc, and what is taxed. I divide this latter dimension into “possession” (what one has) and “activity” (what one does—to include travel, production, and shipping). This yields a typology of predatory funding, in which rebels might rely on theft, extortion, commodity taxes, or usage fees (Table 1).
Typology of Predatory Funding
Armed Group Kidnapping in Colombia, Perpetrator Presumed (1970–2010)
Source: Data from the Centro Nacional de Memoria Histórica (CNMH 2013a).
Note: Several rebel groups, including the Guevarista Revolutionary Army (ERG), Quentin Lame Armed Movement (MAQL), Revolutionary Workers Party (PRT), and the Socialist Renewal Current (CRS) are not recorded as the perpetrators of any kidnappings.
Regularized taxation has a wide range of benefits, including and beyond revenue generation. Taxation can help bolster group ideology, as it does for Islamist and communist groups (Breslawski and Tucker Reference Breslawski and Tucker2021; Revkin Reference Revkin2020). It can be used to impose social control (Arjona Reference Arjona2016) or reinforce state-building processes (Mampilly Reference Mampilly, Di Cosmo, Fassin and Pinaud2021; Rodriguez-Franco Reference Rodriguez-Franco2016). For example, India’s NCSN-IM rebels collect a “house tax,” which doubles as a census (Mampilly and Thakur Reference Mampilly and Thakur2021). As a “technology of governance,” taxation can play a central role in rebel governance, adopting the revenue-collecting functions of states to mirror the wide range of services that some rebel groups provide (Arjona Reference Arjona2016; Mampilly Reference Mampilly2011; Stewart Reference Stewart2018).
Nevertheless, there are organizational and political costs to imposing taxes. Regularized taxation requires surveillance and collection, which taxpayers may not like. Thus, even with its revenue and state-building benefits, taxation is often avoided by resource-rich political actors, from rentier states to rebel groups (Breslawski and Tucker Reference Breslawski and Tucker2021; Ross Reference Ross1999). The central problem with regularized taxation, however, is that it requires enforcement. The “taxman” faces a choice when someone refuses to pay. He may let it slide, especially if monitoring costs are high and tax revenue is low. But if he values the revenue or governance functions of taxation and wants them to continue, he should offer a conditional bargain through “quasi-voluntary compliance” (Levi Reference Levi1988): pay, and you will be protected; refuse, and you will be punished. Moreover, because cash-strapped rebel groups rely on local populations, they should strive to reduce overall levels of violence against their tax base while issuing a threat so credible that majorities are compelled to comply; the taxman wishes to maximize compliance while hurting only a few. The ideal enforcement will demonstrate his presence, incentivize compliance, and—if he’s lucky—recover money owed.Footnote 14
I argue that rebel groups kidnap to enforce regular tax payments. By kidnapping those who shirk, rebels have the chance to recoup through ransom any tax money lost. From the group’s perspective, therefore, ransom kidnapping is a better way to punish tax evasion than killing or otherwise punishing the accused, because it is violence that can pay for itself.Footnote 15 In this way, kidnapping is retrospective—it is an efficient, lucrative way for the group to compel tax compliance. But kidnapping is also prospective in its ability to deter future shirking. Groups use selective violence to communicate to their tax base: compliance protects you; shirking puts a target on your back. This mirrors Kalyvas’s contention that selective killing operates on two levels: “coercive violence tends to be both retrospective in its intention to punish an action that has already taken place and prospective in its goal to deter a similar future action by someone else” (Reference Kalyvas2006, 27). When racketeers demonstrate their ability and willingness to punish defection, they deter shirking and compel compliance (Shortland Reference Shortland2019).Footnote 16
Inherent in this argument is the need for kidnappers to make their violence sufficiently public that the broader population can glean its threat. Yet there is reason to believe kidnappers can communicate a credible threat while remaining out of the public eye. Scholars have repeatedly found evidence for a “collapse of compassion”—greater attention to violence with fewer victims (Gilbert Reference Gilbert2020a; Schelling Reference Schelling and Chase1968). This suggests that attacks on individuals—like kidnappings—can effectively capture the attention of a broader public and make everyone feel at risk. Kidnappers may thus be able to spread their message widely with limited, admonitory violence. They can project their ability and intention to enforce their rules in an efficient, less deadly way.
There are four observable implications of my theory, regarding the perpetrators, victims, and dynamics of violence. First, rebels that rely on regularized taxation will be more likely to kidnap than those that do not. This taxation can come in various forms including property taxes, income taxes, border taxes, and licensing fees, as well as production and shipping fees for illicit commodities. This suggests a conditional relationship between natural resources and civilian victimization: although groups that tap, extract, and sell natural resources themselves would have no need to kidnap for enforcement, those that tax resource production would be likely to do so.
Second, rebels should target for ransom kidnapping those who might not otherwise pay taxes. This can apply to whomever the group has designated as their tax base: the wealthy, certain economic sectors, or the whole population. However, a broad tax base can present a problem because rebels want to limit violence against their present and future constituency. Rebels should thus be expected to mitigate legitimacy costs through ideologically aligned target selection (Drake Reference Drake1998; Gutiérrez-Sanín and Wood Reference Gutiérrez-Sanín and Wood2014). Kidnappers act as “discriminating monopolists,” choosing targets not only for tactical, but also for symbolic reasons (North Reference North1985).
Third, rebel groups should develop an enforcement infrastructure, showcasing their ability to monitor the population and punish at scale. For kidnapping to help tax enforcement, rebels must demonstrate that their threats are credible—that they are both able (via combatant capacity and skills) and willing (via occasional acts of violence) to enforce their threats. Therefore, we should see investment in an armed group’s enforcement infrastructure—special roles for gathering intelligence and executing kidnappings. Together with the implication of regular taxation, we should expect to see kidnapping from groups with high organizational capacity and connections to the local population.
Last, as kidnapping is established as the punishment for shirking, the threat should affect the behavior of their target population over time. Some locals might opt to leave the territory (Rodriguez-Franco Reference Rodriguez-Franco2016) or find violent ways to resist the protection racket (Moncada Reference Moncada2019). But if kidnapping “works” to deter shirking, then the perpetrators should see an increase in tax compliance, as the credible threat of violence convinces would-be shirkers that it is wiser to pay. When targets believe that the cost and probability of kidnapping outweigh the cost of compliance, they should comply. This reflects Shortland’s contention that kidnapping is rare, “off the equilibrium path” behavior (Reference Moncada2019). In reality, this equilibrium often breaks down, as targets opt to exit, resist, or seek alternative protection, rather than comply. Moreover, due to competition and counterinsurgency, rebel territory is not static, so groups’ protection rackets are in constant flux, making perfect information unlikely. Under these conditions, we might instead see a shift from regularized taxation to ad hoc extraction, as rebels are forced to cast a wider net. Although I save this possibility for future work, it can help explain indiscriminate kidnapping by once-discriminating rebels. Table 2 summarizes my argument and the alternative explanations considered, including observable implications of each.
Theoretical Predictions: Explanations for Ransom Kidnapping
Note: Theories labeled “territory,” “poor internal control,” “bargaining problems,” and “ideology” do not purport to explain kidnapping violence explicitly; I have applied their broader expectations deductively.
Research Design
This article explains ransom kidnapping by rebel groups in civil war. It does not explain kidnappings unaccompanied by ransom demands or kidnappings by other perpetrators, including criminal organizations, “lone wolf” individuals, or government actors; it focuses on organizations’ violent, extralegal actions to coerce behavior change for political ends.Footnote 17 Though any armed band could capture the occasional hostage, I limit this study’s scope to formal rebel organizations with the potential and capacity to do so repeatedly and systematically.
I explore this phenomenon in the Colombian civil war for three central reasons. First, rebel kidnapping may be more prevalent in Colombia than anywhere else in the world. For more than five decades, Colombia’s multiactor conflict pitted the state, left-wing guerrillas, and right-wing paramilitaries against each other, while the population suffered tremendously. Beyond the estimated 200,000 killed and seven million displaced, there were 39,058 verified individual kidnappings, 84% of which came with a ransom demand (CNMH 2013a). Colombia has the second highest kidnapping toll of any country in the GTD. Any theory of ransom kidnapping that cannot explain Colombia’s experience would be an incomplete explanation.
Second, there is empirical variation in Colombia’s pattern of kidnapping. Among the conflict’s dozen left-wing rebel groups, the largest and most enduring were the FARC, which began demobilizing in 2017, and the ELN, active at the time of writing. Although the Marxist-Leninist rebels and right-wing paramilitaries all perpetrated mass violence against civilians, groups differed dramatically in their tactics, varying in their share of assassinations, massacres, and abductions. Paramilitaries had an outsized role in the deaths, displacements, and disappearances during the conflict (Daly Reference Daly2016; Steele Reference Steele2017), but guerrillas were far more likely to kidnap. Specifically, of the 39,058 individual kidnappings between 1970 and 2010 documented by the Centro Nacional de Memoria Histórica Impreso en Colombia (CNMH), 73% of all attributed kidnappings were committed by guerrillas. Only 377, less than 1% of total attacks, were attributed to paramilitaries. Moreover, there is significant variation among the left-wing groups’ kidnapping behavior. Whereas the FARC and ELN are presumed responsible for more than half of all kidnappings,Footnote 18 other major guerrilla groups, like the April 19th Movement (M-19), Popular Liberation Army (EPL), and Popular Revolutionary Army (ERP), kidnapped only dozens of times. Several guerrilla groups never kidnapped at all.Footnote 19
Last, the Colombian civil war serves as a hard case for my argument about kidnapping violence. Colombia is resource rich, with some of the world’s largest depositories of emeralds, gold, and oil and an ever-booming drug trade. The guerrilla and paramilitary organizations were notoriously connected to the country’s drug cartels. If armed groups derive their riches from coca production and sales, it seems especially counterintuitive that they would need money from ransom payments. Instead, my theory suggests that the mere presence of resources is less important than understanding how those resources are used.
To test my theory, I focus on the two most prolific kidnappers in Colombia: the FARC and ELN.Footnote 20 I examine why and how rebels kidnapped through direct testimony from perpetrators. Process tracingFootnote 21 serves to show that the mechanisms posited by the theory operate as I suggest (Bennett and George Reference Bennett and George1997). Over three trips to Colombia in 2017, 2018, and 2019, I used snowball sampling to conduct 78 semistructured interviews with and about these groups in research sites across the Colombian departments of Atlántico, Bolívar, Cundinamarca, Santander, and Tolima.Footnote 22 This included 36 interviews with demobilized combatants from the FARC and ELN and 42 interviews with subject matter experts including commanders from the police and military GAULA antikidnapping and extortion units;Footnote 23 senior officials from the Justice Department and independent truth commission; NGO leadership; and hostage negotiators, insurance agents, and former Colombian intelligence officers, currently serving as the chief security officers of multinational corporations including Pfizer, Phillip Morris, and Sony. I met these interview subjects through journalists, academics, kidnap and ransom insurance executives, and Colombian military officers.
Examining such a sensitive topic presents logistical and ethical challenges, as well as questions of bias. Kidnapping is illegal—in Colombia and around the world—and admitting to illegal behavior could put interview subjects at risk.Footnote 24 Scholars have considered the challenges of retrospective research with former combatants (Fujii Reference Fujii2017; Parkinson and Wood Reference Parkinson and Wood2015). Time and memory inevitably obscure important details (Cohen Reference Cohen2016; Daly Reference Daly2016); subjects may also intentionally misrepresent information, diminishing or aggrandizing former crimes. Consequently, I considered it critical to speak to ex-combatants with a wide range of perspectives, across organizations, regions, rank, gender, decade of participation, and experience demobilizing (Oppenheim et al. Reference Oppenheim, Steele, Vargas and Weintraub2015). I sought out subjects that I expected would offer competing perspectives. For example, I spoke to FARC ex-combatants living in an official demobilization zone, ELN ex-combatants at a transitional safe house, retired guerrillas in their 70s and 80s, formerly imprisoned ex-combatants living in major cities, and former commanders turned politicians. As subjects from different backgrounds and sides of the conflict corroborate the same story, it adds internal validity to the explanation presented. However, there remain important limitations to my approach. I explain one form of kidnapping in a single country context; I only explore one type of armed group. Interview subjects were not selected randomly across fronts, some of which kidnapped much more than others. Future work should address these limitations, expand these parameters, and explore kidnapping elsewhere.Footnote 25
Evidence from the Colombian Civil War
During the Colombian conflict, various actors used hostage taking to attract attention and make demands (Fattal Reference Fattal2018). From the 1980s onward, the FARC and ELN kidnapped two sets of victims with concomitant demands: “retenciones económicas” (“economic retentions,” holding civilians hostage for ransom) and “prisoneros de guerra” (“prisoners of war,” holding soldiers, police, and government officials hostage).Footnote 26 These were separate forms of violence operating under different logics, with different targets, demands, and outcomes.Footnote 27 Ransom kidnapping comprised the vast majority of the groups’ kidnapping—over 96% of kidnappings by the FARC and ELN came with a ransom demand (CNMH 2013a). These victims represented a broad swath of the Colombian populace and socioeconomic scale. According to the CNMH, kidnapping was “universal”—the FARC and ELN “victimized not only the wealthy but also the poor and common citizens,” resulting in “a collective feeling of vulnerability” (CNMH 2013b).Footnote 28
In this section, I draw on interview evidence to demonstrate that rebels use ransom kidnapping to enforce taxation. If my argument is correct, I must demonstrate, first, that kidnapping groups imposed a system of taxation, second, that they kidnapped those who would otherwise not pay, third, that kidnapping rebels have an apparent enforcement infrastructure, and fourth, that kidnapping affected future noncompliance. I trace this logic through the cases of the FARC and the ELN. I then provide evidence to challenge alternative explanations for kidnapping violence.
The Revolutionary Armed Forces of Colombia (FARC)
When the FARC began kidnapping in the 1980s, the group had been active for decades but was not the strong, wealthy insurgency it would become. Unlike other Latin American insurgencies, Colombian guerrilla groups were not funded by their ideological role models. One ex-combatant explained, “No one gave us help. Not Russia, not Venezuela. We were a movement born without money. Unlike the state, we didn’t have tax money, and we needed it for shoes and food.”Footnote 29 Conflict experts agree that the Colombian insurgency was “self-financed”Footnote 30—rebels received scant support from communist parties in Colombia, Cuba, or the Eastern Bloc (Otis Reference Otis2014). Thus, the inchoate insurgency had to look elsewhere for funding. Resources were needed to “sustain the organization—politically, militarily, and [provide] our alimentation.”Footnote 31 As a “regular army” of “thousands,” the FARC needed resources to support their health and infrastructure.Footnote 32
The FARC turned to several forms of predatory funding. Throughout the 1980s and 1990s, they expanded rapidly by diversifying to three funding sources: “agriculture, voluntarily from the people, and what our enemies call ‘extortion.’”Footnote 33 While taxation is not always violent, it contains an underlying threat: compliance is “voluntary because taxpayers choose to pay. It is quasi-voluntary because the noncompliant are subject to coercion if they are caught” (Levi Reference Levi1988, 52). Despite the guerrillas’ distinction between “voluntary” and “extortion” models of funding, the two were inextricably linked over the barrel of a gun. In this vein, Colombian guerrillas refer to their protection payments as the “vacuna” (vaccine)—a payment that protects those who comply from violence of the group’s own making.
These payments were formalized in 1982 when the FARC established Law 001, defining the commodity and usage taxes to fund their movement. An ex-combatant explained their logic and targeting: “People who made more than 1,000,000,000 pesos per yearFootnote 34 had to contribute, to pay for peace.”Footnote 35 Another said, “It was a law of tribute: the capitalist oligarchs were responsible for the conflict, so we taxed them. They could pay, and they were notified—pressured—to pay.”Footnote 36 Another justified the charge: “Guerrillas are a belligerent group. They don’t have fees and taxes, but they need money to maintain their capacity. [Law 001] helped us pay for food and resources—exactly like a tax.”Footnote 37 Simultaneously, the FARC formalized its taxation of the Colombian drug trade: they began levying a 10% per kilogram production tax on coca base and taxing marijuana farmers, poppy farmers, and every drug-related flight leaving their territory (Otis Reference Otis2014). From 1996 onward, written policies outlined the FARC’s regularized taxes from merchants,Footnote 38 narcotraffickers, landowners, and ranchers on their possession and activity (Richani Reference Nazih2013).
In 2000, the FARC established Law 002, which justified what they call “economic retentions”—kidnapping any millionaire who has not paid sufficient war tax. After outlining grievances against the Colombian state, American “imperialists,” and the “plundering” of Colombia’s resources, Law 002 affirms that the FARC’s effort “demands sufficient money” to pursue its objectives (FARC 2000). Current fundraising was falling short: “the resources coming from the voluntary contributions that many compatriots send us and those originated by our own investments are insufficient to cover the needs demanded by the struggle of the FARC-People’s Army” (FARC 2000). The FARC’s communiqué then imposes a tax on all “natural or juridical persons” worth more than $1 million USD. The law concludes with a formal justification for kidnapping: “those who do not comply with this requirement will be retained [kidnapped]. Your release will depend on the payment that is determined [ransom]” (FARC 2000). Law 002 thus provides official FARC rationale for kidnapping as tax enforcement.
According to security personnel, this message of conditional violence was abundantly clear, supporting my argument that tax compliance is enforced through kidnapping threats: “Extortion means: pay me now, or I’ll kidnap your family later.”Footnote 39 Put simply, this is “extortion, backed up by a physical threat: pay us or we’ll come get you again.”Footnote 40 Kidnapping skyrocketed during this period—reaching an all-new height in 2001 (CNMH 2013a). Table 3 shows government estimates of one year of FARC funding, which shows that over 80% of the organization’s revenue derived from predatory funding. This emphasis on taxation and extortion was likewise central to my interviews (Table 4)—92% of FARC ex-combatants and 85% of subject matter experts interviewed said the FARC relied on at least one form of predatory funding; few claimed the population offered voluntary support.
FARC Finances, 1998–1999, in Millions of Dollars
Source: Data from Consejaria de Seguridad de la Presidencia, estimates of the National Police and the Military, as presented in Richani (Reference Nazih2013, 62).
Note: One year of FARC finances illustrates the group’s main funding sources.
Percentage of Interview Subjects’ Mentioning Funding Sources: FARC
Note: “Subject matter experts” in Tables 4 and 6 include government, private sector security, and civil society actors. Estimates are conservative for two reasons: First, I did not explicitly ask about every funding source. Second, I have coded “Drugs” for any mention of drugs/narcotrafficking in the interview; this may refer to a tax on traffickers or to rebel involvement in sales.
What did this mean for targeting decisions? If my argument is correct, it must hold at the micro level: guided by Law 002, targets should be chosen because they are perceived to have evaded taxes. A former commander told me, “If someone doesn’t comply with Law 002, we take them, in order to fund our war. This wasn’t kidnapping—it was an economic retention.”Footnote 41 Another echoed, “After [the year] 2000, we had Law 002, and it was very clever: if you don’t pay us, we can retain you. Going forward, you owed us 10% of all your liquid assets. The state collected taxes to make war against us, so we had to place a tax to make peace.”Footnote 42 Ex-combatants emphasized that hostage takings were not attacks on enemies but a means of guaranteeing finances: “It was to fund the conflict, but it wasn’t part of the conflict.”Footnote 43
This resulted in kidnapping “businessmen, ranchers, merchants, companies from abroad, all industry … and their kids, wives, and moms.”Footnote 44 As Danton writes, the FARC’s progressive taxation scheme “almost universally exempted the smallholding peasantry from paying any significant contribution to the ‘cause,’ whereas companies, landlords, cattle-ranchers, and, to a lesser degree, better-off social layers such as shopkeepers and sometimes even outsider teachers, had to contribute” (2018, 31). Every former FARC combatant I spoke to listed the same priority targets: “merchants and ranchers,”Footnote 45 “foreigners involved in business and people who [extracted] natural resources” and “someone with money: working in infrastructure, gasoline, or contraband.”Footnote 46 These professions are not merely Colombia’s elite but reflected the wealth and economic sectors of the FARC’s tax base. Sometimes, the FARC overestimated targets’ wealth, kidnapping Colombians who were neither wealthy nor elite. An executive emphasized, “They take a man with one cow because they think he has two cows.”Footnote 47 Targets deemed “class enemies” and “oligarchs” were ideologically sanctioned, but the FARC simultaneously targeted poor and middle-class farmers through taxes on production (JEP 2021, 90).
A third implication of my theory is that kidnapping rebels will develop an enforcement infrastructure, and indeed the FARC created a “special assignment to take care of [hostages].”Footnote 48 Following commanders’ prerogatives, kidnapping required specialized roles and systematic monitoring of the local population. A former combatant said, “I would spend three months with a group of hostages, like a rancher. I brought them their food, and if something went wrong, I’d bring them medical attention.”Footnote 49 An ex-combatant recounted the extensive, hierarchical work required to pull off each attack:
We were responsible for acquiring provisions, logistics, identifying people to kidnap, and gathering intelligence on the police. Once we got the information about our target to the commanders, they make decisions about everything, and the process followed accordingly. Every single person had their separate role, which was almost always the same: there’s someone good at driving, someone good at talking on the phone, someone to help with wounds. My specialty was logistics, and how to move the person. It was my job to guard the ‘take.’ Everyone has their function.Footnote 50
Security personnel called kidnapping “sophisticated,”Footnote 51 “a lot of work,”Footnote 52 and evidence of “impressive organizational capability from the guerrilla groups.”Footnote 53 Every step was delegated to those with specific training or skills, befitting intentional violence.
Last, if kidnapping works to enforce taxes, we should see targets respond accordingly. Recall that there are two primary benefits groups seek in kidnapping: the tactical benefits of punishing noncompliance and the strategic benefits of communicating a message. In addition to extracting payment from those who would otherwise not pay, the group used kidnapping to deter others from nonpayment: “If we take money this way, it sends a message. And if you try to move, there are [FARC combatants] elsewhere, so we’ll [kidnap] you there.”Footnote 54 Developing a reputation for kidnapping is crucial to its strategic success.
The FARC’s tax base responded in three ways to kidnapping threats. Some simply left dangerous territory. As security personnel recalled, once kidnapping became a formidable risk, wealthy Colombians stopped visiting their countryside fincas or left the country altogether. However, there was not as much emigration as in other conflicts in Latin America, suggesting that the FARC calibrated their violence against elites (Wood et al. Reference Wood, Gibson, Ribeiro and Hamsho-Diaz2010). Other Colombians resisted taxation at high personal cost. Some adopted the mentality of “wishful thinking and armored cars,”Footnote 55 increasing vigilance in their daily routines. One executive explained, “I don’t open my windows, walk on the street, or go to any unfamiliar places. I travel with a driver, and I alternate my route to work.”Footnote 56 Targets turned to alternative protectors to resist FARC extortion—whether the state, paramilitaries, or private security providers. In the 1980s and 1990s, Colombia’s security sector exploded as former federal agents launched companies to mitigate executives’ kidnapping risk: “We studied the roads, the city, and made specific protection plans for everyone. Companies spend a lot of money to be in an armed bubble.”Footnote 57
Still, many targets were effectively deterred from noncompliance, allowing the FARC to increase revenue extraction while decreasing kidnapping in their territory. One former kidnapping specialist explained, “We delivered a card to you—you can pay or be kidnapped, and some paid.”Footnote 58 One year, he collected 100 million pesos [$32,000] without kidnapping anyone; targets complied once he issued his threat. A corporate security officer noted that this became widespread policy: “What did the companies do? They were sandwiched between the paramilitaries and the guerrillas. They’d pay for protection. Many American companies paid. The oil companies paid. The FARC and ELN set up checkpoints, and you had to pay the ‘vaccine’ in areas of their control.”Footnote 59 Perhaps the most prominent example of the threat’s coercing compliance is that of U.S.-based banana conglomerate, Chiquita, which paid extensively documented protection payments to the FARC, ELN, and right-wing paramilitaries for decades. What Chiquita lawyers and executives described as “the cost of doing business in Colombia” amounted to millions of dollars, directly contravening U.S. law, to protect their employees from risk of kidnapping or death (Navarrete, Evans, and Restrepo Reference Navarrete, Evans and Restrepo2017).Footnote 60
While focusing on the short-term benefits of tax enforcement, the FARC downplayed the long-term legitimacy costs they incurred. One former commander told me that kidnapping “did not put our relationship with the locals at risk. [They] understood what we’re doing here; our targets weren’t the pueblo.”Footnote 61 However, lower-level ex-combatants were more candid about targeting mistakes. Did kidnapping risk harming the FARC’s relationship with the local population? “Claro!” several exclaimed. “We kidnapped from the farms. It was a disaster.”Footnote 62 Another confessed, “the FARC committed errors, and they will have to pay.”Footnote 63
The National Liberation Army (ELN)
Kidnapping by the National Liberation Army (ELN) likewise demonstrates the tax enforcement logic. While the FARC demonstrates how taxation was used to enforce taxes on both wealth and production, ELN taxation—and thus kidnapping—stressed the latter. A central contention of this article suggests that groups with resource wealth should not be expected to take hostages: if they have access to riches from the ground, they would not need to extract from the population. However, if groups rely not on looting resources directly but, rather, indirectly—taxing those who profited off of the land—they would be expected to kidnap. In other words, extractive industries may comprise an ideologically appropriate tax base. As with the FARC, I offer evidence for the observable implications of my theory: the ELN levied taxes, kidnapped tax shirkers, invested in enforcement infrastructure, and affected compliance.
Colombia has enormous stores of resources and primary commodities, including oil and the largest coal reserves in Latin America (Rettberg et al. Reference Rettberg, Nasi, Leiteritz and Sanabria2019). It is the world’s top emerald exporter and was formerly the region’s largest gold producer. These vast mineral stores have attracted global investors, contributing to Colombia’s prosperous mining industry. The country’s temperate climate also provides ideal growing conditions for its most profitable crops: coffee, cacao, and the coca leaf.
With a sudden boom in Colombia’s oil and mineral industries in the 1980s, the ELN could tap large, local stores of wealth, imposing a vacuna and other usage fees. Rather than engage in natural resource extraction or drug cultivation themselves, the guerrillas moved to “collect all of their fees” from those extracting emeralds, oil, and gold.Footnote 64 A former combatant put it simply: “We raised money through taxes. We taxed use of the river, just like a mayor would. The state has people pay monthly or yearly, so we did the same.”Footnote 65 They expected money from executives from multinationals and agricultural firms in their territory, which “can afford to pay us.”Footnote 66
While the FARC focused on businessmen and ranchers, the ELN focused on extractive industries. As an ex-combatant put it, “I don’t like the rich who rob our country. Carbon mines, natural gas … how much money does the state make from that?”Footnote 67 Thus, they sought funding from “officials from petrol companies and foreigners and their partners, technicians. It was about extortion from the companies.”Footnote 68 According to the former director of the CNMH, “The ELN combined extorsive kidnappings with their imposition of taxes. They put quotas on all of the petrol companies, but it’s difficult to extort multinationals without a threat.”Footnote 69 Kidnapping, in other words, served as tax enforcement and an effective deterrent from shirking. As Table 5 shows, the ELN derived nearly half of its funding from predation—taxation, kidnapping, and extortion—a focus emphasized in nearly all of my interviews (Table 6). As with the FARC, few interviews suggested that the ELN’s funding was voluntary.
ELN Finances, 1998–1999, in Millions of Dollars
Source: Consejaria de Seguridad de la Presidencia, estimates of the National Police and the Military, as presented in Richani (Reference Nazih2013, 62).
Percentage of Interview Subjects’ Mentioning Funding Sources: ELN
Security personnel echoed that whereas the FARC taxed ranchers and businesspeople, the ELN taxed extractive industry workers, with a “more sophisticated” justification: “They had a Robin Hood story. They would take from rich families explicitly because of exploitation of their country and their people, so they had the right to ask them for money.”Footnote 70 Still, the ELN’s logic of enforcement mirrored the FARC’s. As one ELN ex-combatant put it, “Who gets kidnapped? Someone who owes us money. And if he doesn’t pay, you take him. Once they pay, they’re released.”Footnote 71 Another former combatant echoed, “We tell them, ‘We’re the guerrillas here in this department. We’re fighting for equality of the people, so please support us voluntarily.’ But if they didn’t collaborate, we would use the means of kidnapping.”Footnote 72 As a scholar of the ELN emphasized, “It’s a systematic practice—if you don’t pay taxes, tributes, they’ll retain you. Just like the state.”Footnote 73
Like the FARC, the ELN created an enforcement infrastructure to facilitate systematic kidnapping. Some highlighted their well-known monitoring capacity: “In the mountain, someone is always giving you some intelligence.”Footnote 74 Others highlighted the division into specialized roles. One former ELN combatant broke it down: “There are three or four working in intelligence to get the person, three or four in the car to get them, one person to call the family, and one person to stay with the hostage, to talk and entertain them. Everyone is part of it.”Footnote 75
Finally, the ELN’s reputation for kidnapping effectively deterred future noncompliance. A former commander put it plainly: “Kidnapping was a threat to cover our extortions. If you pay us voluntarily, we’ll leave you alone. So many paid.”Footnote 76 As the former director of the CNMH put it, “Why does the ELN keep kidnapping? Because it’s difficult to extort multinationals without a threat.”Footnote 77 Chiquita, for example, paid the ELN an estimated $236,000–$314,000 for “security” under threat of violence (Navarrete, Evans, and Restrepo Reference Navarrete, Evans and Restrepo2017). A peace advocate emphasized that kidnapping was crucial to enforce the broader funding scheme: “Kidnapping is not a big part of the ELN’s funding; the vacuna is major.”Footnote 78
Alternatives
The interviews provided substantial evidence that kidnapping is used to enforce rebel tax collection. They also demonstrate the need to move beyond the assumption that kidnapping is a straightforward money-making tool. If kidnapping were just about profits, we might expect groups to expend labor to kidnap exclusively high-value targets and cease kidnapping as they developed other money-making techniques in their “resource portfolio” (Rettberg and Ortiz-Riomalo Reference Rettberg and Ortiz-Riomalo2016). Instead, FARC and ELN kidnapping peaked as the wealthy rebels became increasingly involved in Colombia’s drug trade. This reflects two dynamics: first, taxation contributed to smooth functioning of illicit markets, including coca production. Second, because fronts were responsible for their own finances, Colombia’s geographic variation could result in regional variation in funding (Rettberg et al. Reference Rettberg, Nasi, Leiteritz and Sanabria2019). One front might finance itself from involvement in drugs, whereas others relied on predatory funding.
Publicity, though central to much hostage-taking violence, played a minor role in Colombia’s ransom kidnappings. More than 23,000 kidnappings recorded by the CNMH were never reported to the press, and the recent investigation by the Jurisdicción Especial para la Paz (“Special Jurisdiction for Peace,” or JEP) found around 12,000 unreported kidnappings by the FARC alone (JEP 2021). Although ex-combatants repeatedly stressed the need to keep ransom kidnappings secret, the relationship to publicity was complex. For years, the “Voces del Secuestro” (Voices of Kidnapping) radio program on Noticias Caracol, a national Colombian news program, invited hostages’ families to leave anonymous messages for their loved ones on air. Although neither designed to bring attention to specific cases nor aid the FARC, as families kept names hidden for fear of retribution, the messages were broadcast nationally, conveying the scope and scale of the guerrillas’ kidnapping threat.Footnote 79
Similarly, the interviews suggest that prevailing explanations for civilian victimization are insufficient to explain ransom kidnapping. For instance, territory was not directly linked to kidnapping numbers (see Appendix 3.5). The FARC and ELN kidnapped in areas of full state control, urban centers including Bogotá, Cali, and Medellín. These kidnappings, I contend, helped support the prospective function of kidnapping—communicating an effective deterrent against future noncompliance. Urban kidnappings demonstrate to mobile targets that they cannot protect themselves by moving elsewhere, as monitoring extended beyond areas of rebel control. Thus, while the guerrillas targeted locals, they also pursued a sustained practice of abducting people from cities or highways, only to transfer them to remote areas.Footnote 80 Moreover, although competitive state building explains displacement in Colombia (Steele Reference Steele2017), with limited exceptions, ransom kidnapping was not used a tactic to target the politically disloyal within rebel territory.Footnote 81 The FARC and ELN targeted not only ideologically sanctioned enemies but also the agricultural workers of their base.
As the evidence about kidnapping’s enforcement infrastructure suggests, kidnapping was not a consequence of poor internal control. One ex-combatant put it bluntly, “There was discipline. You have to kidnap. If you don’t participate, [the leaders] kill you.” She continued, “We were all given a special role, but there was only one person making the decisions.Footnote 82 Ex-combatants stressed that orders came through the hierarchy and indiscipline was severely punished. “Orders always came from the top. If you committed violence voluntarily, or collected taxes or kidnapped without instruction, that would be outside the line.”Footnote 83
Bargaining-related explanations, which would posit that rebels kidnap civilians to impose costs on the state prior to negotiations, are also insufficient to explain ransom kidnapping—though interviews suggested both the FARC and ELN kidnapped soldiers for leverage in peace talks.Footnote 84 The bargaining logic assumes that rebels cannot impose sufficient battlefield costs on the state and its soldiers and thus resort to attacking civilians to raise the cost of fighting. In fact, as kidnapping reached its peak, the FARC was formidable and feared by the state. As former President Pastrana told me, while the state was pursuing negotiations with the FARC, the Colombian military would simply surrender when they faced the FARC in battle: “That was a huge problem for morale—when your military would prefer to be kidnapped than fight.”Footnote 85 Conversely, the Colombian government has made relinquishing kidnapping a precondition of the ELN for peace talks. The group has thus far refused to do so, but neither have they increased kidnapping in recent years.
Last, there is mixed evidence for the role of ideology in Colombia’s ransom kidnapping. That the most prolific perpetrators were left-wing groups, whereas that right-wing paramilitaries abstained supports the refrain that “the Left kidnaps; the Right displaces.”Footnote 86 But this neat division cannot explain why left-wing groups other than the FARC and ELN kidnapped seldom or never at all. The EPL rarely kidnapped, claiming that it violated their ideological commitments. The FARC initially forbade the practice for that reason.Footnote 87 Instead, differences in groups’ capacity for taxation and enforcement infrastructure can better explain the lack of kidnapping by Colombia’s other left-wing rebel groups.
Nevertheless, ideology played an important role in how the groups approached kidnapping, supporting both “normative commitments” (i.e., to economic redistribution) and “normative constraints” (i.e., do not attack the pueblo; Gutiérrez-Sanín and Wood Reference Gutiérrez-Sanín and Wood2014). As the Bogotá mission chief of the ICRC recounted, “The FARC and ELN talk about the social fabric of the country, ‘if you’re not with us, if you don’t distribute your wealth in line with the values of liberation theology, then you’re a potential target.’”Footnote 88 However, emphasizing the primacy of tax enforcement, both groups frequently took hostages ideologically proscribed.
Conclusion
This article develops a theory of why rebels kidnap for ransom. Why, given the long-term costs of kidnapping, did the FARC and ELN consider it central to their strategy? Because it paid handsome benefits in the short term, shoring up tax revenue on wealth and production, rebels downplayed long-term backlash. The FARC and ELN kidnapped to enforce taxation, enjoying kidnapping’s dual compellent and deterrent effects.
Kidnapping by armed groups in Colombia suggests there are underexplored avenues in the relationship between funding and force in civil war. Despite its centrality to foreign policy debates and adoption by some of the world’s largest and most violent armed groups, kidnapping has been largely absent from political violence literature. Yet, it addresses several fundamental questions of governance and security. First, kidnapping represents both a form of civilian victimization and armed group funding in which perpetrators turn a local population into a resource to be extracted. It illustrates a new way to consider the relationship between natural resources and civilian victimization, conditional on whether resources are taxed. This has implications for local governance, as armed groups tax and extort from civilian populations, as well as global terrorist financing, as groups use ransom payments to fund their violence. Despite a wealth of research on conflict funding, this pervasive form of finance has been neglected.
Second, due to its high human cost, this phenomenon has implications for postconflict reconciliation. Kidnapping comprised a central form of suffering in the Colombian civil war—so central that the JEP took up the FARC’s kidnapping as its first and most high-profile case. Whereas FARC leadership have recently expressed their regret, the ELN has refused to disavow the tactic, insisting on its legality and importance for their security. Understanding the role of kidnapping in civil war will provide insights for bringing groups to the negotiating table, facilitating the demobilization and reintegration of former combatants, and reconciling with the broader population.
Kidnapping demonstrates the myriad costs that violence can produce. While the FARC and ELN adapted to the short-term costs of kidnapping, it seems less clear whether—or how—they considered the legitimacy costs of adding kidnapping to their repertoire. Given the ample damage kidnapping wrought to these at-times popular guerrillas, it seems that they underestimated the long-term consequences of their violence. Legitimacy costs also represent an area where combatants’ experience mattered deeply (Oppenheim et al. Reference Oppenheim, Steele, Vargas and Weintraub2015). Those who had deserted the groups eagerly admitted that kidnapping had been a mistake, such as one former ELN commander who demobilized over kidnapping: “It was a disgrace. It was outside the norms of war; there was a river of blood, and it really delegitimized our opposition… I said we had to stop kidnapping. The most important thing is to morally justify the act.”Footnote 89 Conversely, FARC commanders who participated in the peace process continued justifying the practice long after capturing their last hostage. As the Bogotá ICRC chief said, “Kidnapping is worse than killing people… The most striking part of this phenomenon is the emotional impact of kidnapping on families and society. It has been underestimated by kidnappers, but it will harm them forever.”Footnote 90
The causes and consequences of nonlethal violence are crucial to understanding civil war outcomes. Beyond Colombia, pervasive kidnapping violence could dramatically alter rebels’ reputation with local populations. This in turn may affect rebel recruitment or postconflict outcomes, including support for reconciliation or revenge. Future research should explore the temporal trade-offs and legitimacy costs of kidnapping violence. It should also further explore kidnapping’s pattern of political violence beyond Colombia to understand when armed groups seek or eschew publicity, make nonmonetary demands, and target locals versus foreign nationals. More work exploring the intersection of nonlethal violence and rebel governance can help us understand the purposes, processes, and manifestations of understudied tactics. Such research can provide insights as to why—and when—armed groups exercise different aspects of their repertoire of violence and how doing so conditions insurgencies’ failure or success.
SUPPLEMENTARY MATERIALS
To view supplementary material for this article, please visit http://doi.org/10.1017/S0003055422000041.
DATA AVAILABILITY STATEMENT
Research documentation of this study is openly available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/50M5Z4. Limitations on data availability are discussed in the text and appendix.
ACKNOWLEDGMENTS
The views expressed in this article are those of the author and do not represent the United States Air Force Academy, Department of the Air Force, or Department of Defense. Approved for public release, PA#: USAFA-DF-2020-415. The author would like to thank Elizabeth Acorn, Nicholas Anderson, Erica de Bruin, César Caballero, Alex Downes, Eugene Finkel, Hein Goemans, Buddhika Jayamaha, Josh Kertzer, Kendrick Kuo, Janet Lewis, Yonatan Lupu, Cynthia McClintock, Jonathan Pike, Stephen Rangazas, Livia Schubiger, Abbey Steele, Michael Weintraub, Rachel Whitlark, Sherry Zaks, three anonymous reviewers, and the editors of the American Political Science Review for their invaluable comments. This article benefited from presentations at the Institute for Security and Conflict Studies at George Washington University, the Political Economy of Security Workshop at Boston University, the Triangle Institute for Security Studies “New Faces” Conference, the 2019 ISSS-IS Conference in Denver, Colorado, and the 2019 Peace Science Society in Manhattan, Kansas.
FUNDING STATEMENT
This research was funded by a Peace Scholar Award from the United States Institute of Peace and Minerva Research Initiative, a World Politics and Statecraft Fellowship from the Smith Richardson Foundation, and the George Washington University.
CONFLICT OF INTEREST
The author declares no ethical issues or conflicts of interest in this research.
ETHICAL STANDARDS
The author declares that the human subjects research in this article was reviewed and approved by the George Washington University and certificate numbers are provided in the appendix. The author affirms that this article adheres to APSA’s Principles and Guidance on Human Subject Research.
Comments
No Comments have been published for this article.