Abstract
Although also victims, there is a growing appreciation that civilians are actors in civil wars. Scholarship on civilian agency shows how their decisions impact well-being and shape conflict dynamics. Civilians are increasingly framed as impartial forces resisting armed groups and cultivating peace. However, civilians are as likely to support armed groups and deepen violence. This paper seeks to better understand civilian partisanship. I frame forms of civilian support that enhance the coercive capacities of armed groups as ‘indirect violence’. Some forms of support, such as refusing to provide information or protesting enemy abuses, may reduce violence, while providing food or medicine is more neutral. Other actions, such as contributing funds, intelligence, recruits, and weapons, enable armed groups to carry out violence. That civilians contribute indirect violence does not mean they are not victims, but may call their innocence into question, providing a sober account of civilians in civil war.
In the wake of Aceh’s separatist conflict in 2008, I arranged what turned out to be a long interview with Free Aceh Movement (gam) Commander ‘Kowboy’. After recounting some tall tales over coffee, we turned to interactions with civilians, the topic of my dissertation. The Commander and his entourage explained that some civilians schemed against gam, persons he wrote off as opportunists. “Most Acehnese”, explained Kowboy, “helped us fight…Some told the army we went one way, when really, we were close. People told us where the army was. One old woman ran crying, saying gam killed her son, catch them, leading the army into our ambush.” When I asked if this is dangerous, he explained that “They are gam civilians. They are not fighters, but for Indonesia, they are not innocent” (interviews in Bireuen, February 2008).
At the time, I was interested in civilians opposing armed groups, subconsciously overlooking how they fuel war. While civilians also fled to find better lives and resisted combatants to demand peace, many supported state or rebel forces. Although many civilians were victims of this long-standing civil war, they were also participants whose actions increased the coercive capacities of armed groups. Indeed, many of the most committed supporters are motivated by being victimized by the other side, meaning that victims may simultaneously contribute to the victimization of others. Although civilians, neither taking part in active violence nor representing direct mortal threats, such partisans were hardly anti-violence nor necessarily innocent.
How do we understand civilian support for armed groups? How do civilians contribute to violence, strengthening the ability of combatants to fight? The ways that unarmed civilians enable violence by armed groups remain undertheorized, chafing with understandings of civilians as passive victims or peace advocates. I argue that many civilians contribute indirect violence through forms of support that enhance the coercive power of armed groups.
The first section of this paper reviews the literature on civilians in war, including a recent shift towards civilian agency, especially resisting armed groups and promoting peace. The second section reviews studies of civilian collaboration with and support for armed groups. In part three, I analyze how civilian support fuels violence. Some forms of civilian partisanship may be seen as decreasing violence or as neutral. We can refer to indirect violence when civilian support contributes to an armed group’s ability and likelihood to commit violent acts—contributing resources, providing intelligence, recruiting fighters, and procuring weapons. How these unfold and link to violence are illustrated with examples drawn from first-hand fieldwork in Southeast Asia as well as secondary sources. This paper concludes with some implications; highlighting indirect violence by civilians does not mean they are not victims, but may call their innocence into question. Of course, much depends on the degree to which support is coerced. However, the forms of support that most directly translate into violence require some trust and sustained commitment, and are thus those provided more voluntarily. Analyzing how civilians may help to fuel violence may mean they are not innocent, as they may knowingly contribute to violence, however this sober account of civilians in war should not be seen as undermining their state as civilians deserving protection.
Civilians in War: Victimhood & Agency
Although the civilian/combatant distinction is blurred by civilian contributions to violence, we must not dismiss this principle of distinction (Kinsella 2015). The Geneva Protocols (icrc 2021) define civilians as “any person not belonging to the armed forces”, which seems statist (excluding rebels) and focuses on affiliation rather than action. Although often conflated, ‘civilian’ differs slightly from ‘non-combatant’. If civilian is about status, non-combatant is about behaviour (see Table 1). A non-combatant is someone not actively taking part in hostilities, thus not representing a direct mortal threat. The Fourth Geneva Convention states that, in civil wars, non-combatants are especially deserving of protection, framed as “Persons taking no active part in the hostilities, including members of armed forces who have laid down their arms” (Geneva Convention 1949, 3:1). Meanwhile, a combatant may be a state soldier, rebel, militia member, or other armed persons affiliated with armed groups. By possessing weapons and some training, combatants represent direct mortal threats. Although speaking of civilians as shorthand, this paper focuses on civilian non-combatants, those inhabiting the bottom right of the table. Valentino, Huth, and Balch-Lindsay (2004, 378-379) define a non-combatant as “any unarmed person who is not a member of a professional or guerrilla military group and who does not actively participate in hostilities by intending to cause physical harm to enemy personnel or property.” This definition specifies civilians (non-members) and non-combatants (not participating in harm). Civilian non-combatants may include various societal groups, from farmers and villagers to students, religious leaders, activists, and officials. However, they may be members of organizations affiliated with armed groups or may play a role in training combatants, contributing forms of indirect violence.
Civilians, Non-Combatants, and Civilian Non-Combatants
| Combatants | Non-Combatants | |
|---|---|---|
| Armed Group Members | Soldiers | Medics, Engineers, pow s |
| Civilians | Rebels, Self-Defense Forces | Civilian Non-Combatants |
The study of war long reflected a statist bias, concerned with Great Powers and state security, while civilians were afforded limited attention. Following the Cold War, an emphasis on civil wars led to an expanded concern for protecting civilians (Bellamy 2009; Hultman 2013). Humanitarian concern generated a considerable scholarly literature theorizing the causes of civilian victimization, including armed group capacity (Wood 2010), conflict dynamics (Wood, Kathman, and Gent 2012), and rebel resources and organization (Weinstein 2007), to name a few. Titles include The Lessons of Terror: A History of Warfare against Civilians (Carr 2003); “Handling and Manhandling Civilians in War” (Humphreys and Weinstein 2006); “One-Sided Violence against Civilians in War” (Eck and Hultman 2007); Killing Civilians: Method, Madness, and Morality in War (Slim 2010); Targeting Civilians in War (Downes 2011); as well as on bombing, purging, detaining, protecting, and defending civilians. Here, civilians are solely victims, treated as dependent variables. The victimization frame endures for good reason—civilians are victims of war and are in need of assistance.
Civilians are typically approached not only as victims, but as innocent. After all, helping those implicated in violence will be less compelling than persons who are peaceful, but find themselves caught in the middle. The Red Cross states that “the main victims of war have been civilians. The protection of civilians during armed conflict is therefore a cornerstone of international humanitarian law…[especially] particularly vulnerable civilian groups such as women, children and the displaced” (icrc 2010). For Slim (2003, 483), “the phrase ‘innocent civilians’ is repeated like a chant by those concerned to argue the humanitarian case of distinction and immunity in war.” Charli Carpenter (2005) challenges the presumed innocence of women and children. Carpenter argues that assuming innocence based on ascriptive traits rather than individual action is problematic, potentially excusing violence against innocent men and overlooking how women and youths participate in warfare. Similarly, Kinsella (2015, 8) analyzes the idea of innocent civilians, including its conflation with women and children, pointing out that the idea of “a ‘guilty civilian’ appears oxymoronic”.
Civilians are not simply victims, nor are they always entirely innocent. Kalyvas (2006) explains violence against civilians in terms of armed group capacity across zones of control, but also recognizes that civilian support is a prized commodity for armed groups (111) and that civilians may use armed groups for personal ends (330). Scholars have discovered agency even among the victims of war. Coulter (2009) criticizes the tendency to treat women as passive victims; “The notion of victim has become a socially constructed identity that often reifies women’s experiences of war” (10, italics in original). In Sierra Leone, some women sided with rebels as wives and girlfriends in pursuit of security as well as wealth, sometimes contributing quite directly to violence. Inspired by Hirschman’s classic study, Barter (2012, 2014) delineates civilian options in terms of flight, voice, and support. Similarly, Baines and Paddon (2012) analyze civilian self-protection strategies in northern Uganda, noting how civilian strategies of neutrality, avoidance, and accommodation are overlooked by humanitarian agencies. Moore and Shellman (2006, 599) emphasize how scholars and humanitarians frame displacement as forced, even though not everyone flees, many do so at different times, and they may choose different destinations; “When faced with extraordinary circumstances, people still make choices.”
Recent research on civilian agency has shown that civilians are not passive actors, and can sometimes mobilize for peace. Kaplan (2017) analyzes civilian resistance and self-protection against armed groups in Colombia. He shows how civilian action can reduce violence, with strong social organization allowing communities to manage disputes and keep combatants at bay. Kaplan studies peaceful civilian action on the grounds that this is more common than civilian violence (16), perhaps overlooking grey areas of civilians contributing indirectly to violence and therefore war.1 Avant et al (2019) reject civil wars as limited to contests between combatants, highlighting how unarmed groups with “a reluctance to resort to violence” can mobilize for civil action. Reviewing the civilian agency literature, Masullo (2021) emphasizes how civilians reject armed groups and resist violence. Chenoweth and Stephan (2011) suggest that nonviolent movements are more successful than violent ones, largely because they are easier for people to join; “nonviolent campaigns have a participation advantage over violent insurgencies…The moral, physical, informational, and commitment barriers to participation are much lower for nonviolent resistance than for violent insurgency” (10). This suggests that civilians must enlist as fighters, overlooking how they support or join violent movements without the deeper commitment involved in taking up arms.
In many studies of civilian agency, combatants are violent, while civilians, having chosen not to fight, are peaceful protagonists. Although this field of research challenges civilian passivity, their innocence often remains intact. In reality, many persons that choose or are not allowed to take up arms contribute a great deal to violence, even if they do not directly commit it. Civilians do promote peace and bravely resist combatants, but they also support and strengthen them, indirectly fueling violence against other civilians.
Civilian Partisan Support for Armed Groups
An old woman gave us shelter
Kept us hidden in the garret
Then the soldiers came
She died without a whisper…
leonard cohen, “The Partisan” (1969)
For those studying elections and party politics, ‘partisans’ are dedicated supporters of a particular side. In studies of civil war, ‘partisans’ evoke images of armed resistance groups in Nazi-occupied Europe. Leonard Cohen’s take on a French resistance ballad frames the partisan as a guerrilla fighter. But those that have not taken up arms and support those that have, such as the old woman, are as much partisans, actively supporting a particular side despite grave risks.
It is important to define what is meant by partisan civilian support. Although initially speaking of support (2006, 91-92), Kalyvas prefers “collaboration” (104), framing civilians as opportunistic. Others speak of civilian “loyalty”, which may suggest affect rather than action (Johnson 1962). Arjona (2016, 46) differentiates between obedience (compliance) and support, preferring “cooperation”, defined as actions “that directly benefit the armed groups, independent of the motivations that underlie them” (25). Petersen (2001) fluctuates between collaboration and support, speaking broadly of wartime civilian “roles”. Following Wickham-Crowley (1992), Wood (2003, 17), and Barter (2014, 28), I prefer civilian ‘support’ for armed groups. Collaboration carries a stigma of opportunism, while cooperation is a broader term that includes basic compliance. Support refers to a wide range of civilian actions intended to benefit an armed group. The emphasis in this paper is on actions rather than motivations, although I will return to less voluntary forms of support at the end of the paper.
Civilian support has long been considered central to armed groups. For Wickham-Crowley (1992, 8), “peasant support is a crucial contributor to revolution”. Combatants often battle for ‘hearts and minds’, promising reforms and providing public goods to cultivate civilian support, and deprive their opponents. This is well understood by insurgent and counter-insurgent forces. Mao famously saw the people as the sea, with rebels swimming through them; the masses may sustain insurgents by providing “blankets, communication materials, transport, and facilities for propaganda work” as well as “revolvers, pistols, rifles, spears, big swords, and land mines and mortars of local manufacture” (1937, n.p.). In his Bolivian Diaries (1967), Che Guevara wrote that victory requires civilians to provide supplies, transport, information, finances, and networks. This has also been recognized in counterinsurgency operations. A widely distributed US Army field manual states that “In coin operations, the population is vital—since whoever the population supports has the advantage” (US Army 2009, 1-7). It defines rebel auxiliaries as “active sympathizers who provide important logistical services but do not directly participate in combat operations”, including women and children. Civilian auxiliaries may support rebels through intelligence; delivering messages; developing, storing, and transporting weapons; providing money; recruiting rebels; maintaining equipment; sabotaging state forces; and more (2-4). Many counterinsurgency operations are designed to sever rebels from supporters, as in various ‘strategic hamlet’ programs. Valentino, Huth, and Balch-Lindsay (2004) argue that state violence against civilians is not due to ethnic hatred or poor discipline, but may represent a strategy to weaken guerrilla forces reliant on civilian support, using violence to ‘drain the sea’. In other words, civilians are often victims because they are partisan supporters working to strengthen one side.
While several studies speak of popular support for rebel groups, only a handful provide in-depth conceptual discussion. Elizabeth Wood’s study of insurgency in El Salvador provides rich illustrations of civilian support as the lifeblood of insurgency. Observing that many civilians actively supported communist insurgents despite a lack of direct material benefits or security, she explains support in terms of a will to resist injustice and the pleasure in doing so. State violence drove many villagers to become active rebel supporters, maintaining silence to state forces, as well as providing information, food, and supplies to rebels (2003, 115). Wood also observes forms of support that enhance violence, including providing intelligence on army movements, transporting weapons, marking sites for bombs, and more (124, 173). Wickham-Crowley (1992, 54-55) lays out a seven-point scale of civilian support, including non-reporting (1); providing food (2); various errands (3); providing shelter (4); organizational cooperation (5); joining a militia (6); and enlistment (7). Petersen (2001) provides a similar model, placing inaction at 0 and +/- 3 as enlistment (+/- refers to either side). This leaves the realm of civilian support as +/- 1 (unorganized and sporadic support), while +/- 2 represents more sustained commitment, including membership in support organizations. Petersen is interested primarily in how people moved from one point to the other across fluctuating occupations in Lithuania. His work yields important insights for understanding civilian support as indirect violence, which is largely +/- 2 support. However, Petersen’s scale centres on depth of commitment, not contributions to violence. Many forms of support in the ‘2’ category, such as serving as a “partisan liaison…collecting food from other farmsteads to give to the regularly visiting partisans”, do not generate violence in the same ways as even occasionally “gathering guns and rifles” or recruiting fighters (199).
In addition to individual acts, civilian support is often channeled and amplified through affiliate organizations. Many communist rebel groups develop cooperatives, women’s groups, student groups, labour unions, and ngo networks. Wood (2003) shows this through political parties and farmer’s cooperatives aligned with the insurgents. Pye’s study of Malaysian communist insurgents emphasizes rebel ties with civilians, commenting on their “conviction that the rural population or the peasantry is a revolutionary element that can be relied upon in the struggle” (1950, 32). Pye documents how insurgents established civilian organizations and ‘United Fronts’, from women and youth groups to farmer cooperatives, schools, newspapers, and informant networks to sustain the rebellion. Affiliate organizations allow civilian partisans to join armed movements without joining armed groups, remaining civilian non-combatants that may fight through other means.
In sum, although we may see civilians as inert victims or as active peacebuilders, some studies have focused on civilian support, in which unarmed groups strengthen armed ones. This challenges views of civilians as passive or innocent victims. Despite differing on terms and motivations, scholars are clear that civilian support matters in war. However, forms of support are often grouped together, perhaps differentiated in terms of commitment, but not in terms of the effects on violence.
Civilian Support as Indirect Violence
It is important to disaggregate varied forms of civilian support and their relationship to violence. This does not necessarily track with the degree of commitment or depth of support, as a rebel medic may be part of a formal organizational structure but be responsible for less violence than a civilian making a single denunciation. Some forms of support for an armed group may reduce violence, with others being neutral or contributing only minimally. At the other extreme, persons may shed their civilian status when taking up rudimentary arms. This paper is concerned with the middle ground, illustrated in Table 2, with indirect violence shaded. This echoes Petersen’s spectrum, namely the +/- 2 level of support, but only partly. Petersen’s scale is based on degrees of involvement, not violence. It is conceivable that attending a rally, seen by Petersen as +/- 1, might entail criticizing the other side for human rights abuses or calling to punish collaborators. Table 2 also draws from Wickham-Crowley’s schema, even though he does not consciously refer to scales of violence. For instance, his level 5 (organizational cooperation), includes participation in rebel schools alongside making weaponry (1992, 54). Table 2 presents a spectrum of the relationship between support and violence, identifying forms of support that amplify violence to various degrees.
Civilian Support through a Spectrum of Violence
| Nonviolence | Neutral | Indirect Violence | Violence | |
|---|---|---|---|---|
 |
||||
| Inaction | Providing Aid: | Information | Information | Village Guards |
| Non-Reporting | Medical, | Funds | Spying | |
| Attending | Food, | Transport | Denouncing | → Enlisting |
| Rallies | Shelter | Symbolic | Others | |
| Information | Labour | Support | Recruiting | |
| Welfare | Organizations | Procuring | ||
| Programs | Governance | Weapons | ||
Support as Nonviolence
Not all forms of civilian support for armed groups should be seen as contributing to violence. On the left of the spectrum are forms of civilian support that may reduce violence and save lives. Examples may include inaction, where civilians avoid or refuse to support a specific side out of loyalty to another. Non-reporting, where civilians refuse to provide information, can be seen as support for the other side, but can reduce violence by slowing military operations. Civilians protesting violence by one side may also be seen as a form of partisan support, but by promoting nonviolence and human rights, opposes violence. This is not the case if protests overlook or justify violence by one side or demand violence against others. Finally, civilians may work with armed groups on projects that provide welfare for their society, enhancing the legitimacy and popularity of armed groups while improving overall welfare.
Numerous examples attest to forms of civilian support that may diminish violence. In southern Thailand, silence has represented a dominant response to military sweeps. Given uncertainty, even if civilians might know where rebels are or who is helping them, they are unlikely to tell Thai authorities (interviews in Patani, Summer 2008). In Aceh, villagers explained that they played dumb, pretending not to speak Indonesian when soldiers arrived. This corresponds to Wickham-Crowley’s non-reporting (1992, 55) and Petersen’s 0-point of neutrality (2001, 8). Civilians forced by the state to serve as night guards communicated to rebels that they are ‘off duty’, eventually creating systems where hanging their sarongs signaled inaction (interviews in Aceh Besar, November 2007). These conflicts also saw protests criticizing government leaders and army human rights abuses, events sympathetic to rebel groups. In Mindanao, civil society organizations often have ideological alignments (Rood 2005, 37), with pro-rebel protests calling out army violence. Here, emphasizing human rights abuses strengthens the narrative of reducing violence, which may also constrain rebel groups. In Aceh, separatist groups initially rejected human rights language, but came to ally with civil society critics of Indonesian abuses; doing so undermined Indonesia, but also changed the rebels, slowly incorporating human rights language (Aspinall 2002, 19-20). In Mindanao, ‘Zones of Peace’, where residents declare their communities to be off-limits to combatants, came to be partly coopted by the state, making them a form of support as landlords and Christian villagers restricted rebel operations. This said, we can still see them as reducing violence, since they maintained a nonviolent, albeit conservative status quo.
Information provided by civilian supporters often generates violence, discussed below, but can also decrease it. In my fieldwork, I was told of many instances in which civilians informed rebels of approaching troops, allowing them to escape and avoid battles. In some instances, religious leaders sounded prayer drums, church bells, or the call to prayer as a signal to rebels that troops are present or had left (interviews in Aceh, February 2008 and Mindanao, June 2008). Such forms of support enable combatants to commit violence in the future, but here, the civilian act helps to avoid the loss of life. Civilians can also serve important roles in encouraging desertion from the other side. For armed groups, desertion represents a direct threat, although one not often publicized (McLauchlin 2015). In many conflicts, civilian messengers and officials aligned with one side are known to help rival soldiers demobilize with some guarantees of safety. In Mindanao, this role was played by Islamic leaders (ustadz), while in Aceh, some village chiefs worked with Indonesian officials to allow rebels to demobilize. This is an important form of civilian support, aiding one side by peacefully reducing the number of combatants available to the other side.
Finally, many civilian partisans join auxiliary organizations that play largely peaceful roles, such as promoting welfare. In Aceh, rebels provided and encouraged donations for mosques, widows, and army victims, work typically carried out by Islamic and university students (interviews in Bireuen, November 2008). In Sri Lanka, the Tamil Tigers briefly established a proto-state, which while largely concerned with security, also provided welfare and health services in cooperation with affiliated civilian organizations, improving people’s well-being. The Tamil Rehabilitation Organization was run by civilian partisans, channeling international assistance to provide services to internally displaced persons (idp s) and other conflict victims (Stokke 2006, 1029). Similarly, armed groups might establish a sort of ombudsperson role in which civilians can share complaints about combatant behaviour. Such systems might help armed groups to police their ranks and improve their reputation, with texts and social media messages informing civilian officials of violence, leading to internal investigations and reforms. In Aceh, such roles were sometimes played by rebels, but also by pro-rebel student activists (Aspinall 2009, 138). The above forms of support strengthened a particular side, but in ways that conceivably reduced violence.
Support as Neutral
The next cluster of civilian support can be seen as largely neutral in terms of violence. This refers mostly to forms of support that enhance the well-being of individual combatants, such as providing medical assistance, food, shelter, or other provisions such as clothing, but do not directly enhance offensive capabilities. These actions thus contribute to violence very indirectly, encouraging healthier combatants and in aggregate strengthening a fighting force, but also saving lives and improving well-being.
Many civilians take up sustained roles as nurses, cooks, or billets to assist armed groups. That civilians provide medicine, food, and shelter to armed groups is a core point in counterinsurgency strategies. In the Malayan Emergency, severing civilians from rebels was intended to ‘starve out’ guerrillas (Pye 1950). Such support roles are especially common for women, whose gendered peacetime labour transfers to wartime support. Wickham-Crowley (1992, 22) finds that, even though leftist guerrillas promoted women’s equality, in practice “women in guerrilla movements apparently were relegated to typical ‘support’ roles rather than active combat”. Horton (1998, 207) focuses on peasant support for Contra fighters in Nicaragua, where women and youths joined networks to supply food and other supplies. In Aceh, Aspinall (2009, 93) observes that “women were assigned support roles, such as provision of logistics, medical assistance, or hiding and smuggling male fighters”, almost never participating in combat. Such roles were risky, especially for nurses on the front lines; “It was very dangerous. We helped wounded gam soldiers, protecting our brothers” (interviews in Bireuen November 2007). Youths were also tasked with collecting and delivering food to rebels. Such actions helped to sustain armed groups, making healthy soldiers that could continue fighting, but their immediate effects were improved well-being, with distant causal links to violence.
Civilians may also serve as labourers and engineers, building encampments or bridges, improving roads, enhancing irrigation, or providing other forms of labour. Such work may enhance an armed group’s popularity or capacity, but may also serve the entire community as public goods. While building clinics improves lives, building roads can help the community and enables armed group deployment and resource extraction. Such public goods serve multiple ends, but minimally are less directly linked to violence than other forms of support. Many such actions may be a result of coercion, provided involuntarily. Shelter, food, labour, and donations may be provided through intimidation and coercion, support that is important to armed groups, but with indirect links to violence.
Additionally, religious figures such as chaplains may excuse or justify violence. In Aceh, the rebels benefitted from the gradual support of traditional Islamic leaders, which rebel commanders recognized as improving morale and public legitimacy (interviews in Bireuen, February 2008). While some Islamic leaders encouraged violence in their sermons and councils, others spoke of justice and nonviolent ways to achieve their goals. Such religious officials supported and even joined armed groups, but with little effect on overall violence.
Support as Indirect Violence
Indirect violence begins where partisanship increases the potential of armed groups to commit violence while not decreasing violence or suffering. Table 2 features two columns, representing varied distances between civilian action and combatant violence.
Providing information could fit multiple categories depending on its content as well as frequency. If civilians provide information that helps combatants avoid an ambush, we can see this as nonviolent, even it eventually enables later attacks. This is different than intelligence helping an armed group conduct an ambush of its own, a direct link to violence discussed below. In between, we may see civilians provide information by delivering messages for armed groups to find supplies or report intra-organizational intelligence. For example, couriers may tell combatants where to pick up funds, maps, or other supplies that later enable them to launch attacks. US counterinsurgency experts refer to civilians providing “passive intelligence collection” and performing “courier operations” (US Army 2009, 2-4). In Aceh, youths and Islamic leaders were tasked with delivering messages across districts and into state strongholds. According to one report (Child Soldiers 2008), 14–18-year-olds were pressured “to run errands, look out for police and purchase supplies”. The report frames these youths as child soldiers, misrepresenting their combatant status, intended to criticize rebels but unintentionally inviting harm against civilians. According to one rebel commander in Aceh, “young men and girls would provide information, like where we could pick up supplies, sometimes uniforms or weapons. They were proud to do this, our scouts, sometimes our sons” (interviews in Bireuen, 2008).
Donations, tolls, taxes, resource extraction, and other forms wealth may be used for various ends, including purchasing weapons. Funds may derive from legal or illicit sources, from religious and diasporic donations to oil revenue, gems, black markets, ransom, and narcotics. Rebel and also state military revenue may be generated by vast networks of civilian supporters. Studies of civilian victimization tend to paint a picture of resource rich combatants not relying on civilians, whereas resource poor rebels rely on civilian support and thus treat them better (Weinstein 2007; Fortna, Lotto, and Rubin 2018). Things are less clear in practice. Petroleum extraction demands expertise and equipment, while gemstone extraction requires miners and transporters, and drug cultivation demands a labour force of pickers and refiners (Lujala 2009, 54). In regions lacking such resources, donations represent key sources of revenue for procuring weapons and other supplies. Money may be provided voluntarily or coerced through tolls, robbery, and taxes, with entrepreneurs and large businesses pressured for rents. In Aceh, Islamic leaders used a local tradition of roadside donations for mosques to help fund rebels, while village chiefs faced demands from both sides for scarce village funds (Barter 2014, 75, 82). Aspinall (2009, 155) describes networks of gam civilians serving as tax collectors and donors. This allowed the rebels to provide stipends and public works, but also to pay rebel soldiers and purchase weapons.
Similarly, providing transport for combatants may enable combat operations. Transport may involve simply taking a combatant home, delivering food, or helping combatants escape attack. But transport or serving as local guides may also enable armed groups to position themselves for offensives or maintain economic activities. In Colombia, civilian transportation was crucial for illicit economies, bringing narcotics from remote rural regions to towns, ports, and international markets (Arjona 2016, 184). Civilians may donate vehicles or have them commandeered. More voluntary support is found in terms of maintaining equipment and vehicles. In Mindanao, rebels employed mechanics, typically persons operating private businesses known to serve rebel vehicles (interviews in Cotabato June 2008). This echoes what Zhukov (2016, 399) dubs “the logistics of violence”, the technical, engineering side of war carried out by non-combatants; “One cannot kill without means to reach a target. Without transport and open lines of communication, combatants cannot easily deploy their forces, reload their weapons, refuel their vehicles, repair their equipment, feed their troops, evacuate their wounded, or send detainees to camps.” Myriad smaller support roles played by civilians allow armed groups to carry out attacks. This includes forging documents that allow armed groups to arrange shipments, prepare for attacks, or cross borders. Civilians may also manufacture equipment, include clothing, footwear, tents, flares, glasses, radios, mobile phones, printers, dredgers, and boats, equipment used to organize attacks.
Civilians may encourage violence through symbolic support such as propaganda, art, and graffiti, as well as attending rallies. Above, I noted that such expressions may diminish violence by criticizing one side’s human rights record. Such acts may also increase violence when favouring one side, especially when justifying or encouraging violence. In Aceh, civilians helped the rebels post notices warning ethnic minorities to leave or else face attacks (interviews in Nagan Raya, December 2006). Supporters may write, print, and spread rebel propaganda that valorizes victories, justifies attacks, or downplays violence. Throughout Southeast Asian separatist conflicts, national media outlets employed journalists that promoted the government line that rebels are criminals, focusing on rebel violence and justifying military attacks (McCargo 2006). Keels and Kinney (2019) show how political wings of armed groups deflect media attention from violent attacks, highlighting precipitants by the other side. Such propaganda wings feature distinctive operational structures, expertise, and perhaps above-ground civilian organizations.
Membership in affiliate organizations represents an important form of support, buttressing the above methods and allowing membership without taking up arms. Affiliate organizations, including farmer cooperatives, study groups, women’s organizations, political parties, and artist collectives can elevate an armed group into a social movement. Wood (2003) notes that, in El Salvador, communist organizations encouraged land occupations by peasants, promoting attacks on landlords to decrease the suffering faced by the poor. Some armed groups enjoy linkages with political parties, i.e., the Irish Republic Army and Sinn Féin. This is also a core lesson in the literature on rebel governance, as armed groups create formal organizations to carry out governance functions and cultivate wider support. State forces may also create ethnic organizations, veterans’ groups, and more as a means to enhance counterinsurgency operations. In East Timor, many violent militias were rooted in welfare and vocational training organizations, with unemployed youths gaining job skills while serving as reserves able to target dissidents (Barter 2013, 82). Support organizations are primarily civilian, but membership enables civilians to provide material benefits to armed groups and join violent movements.
An important example of rebel organizations, and a foundational component of rebel governance, is courts (Arjona 2016). Although sometimes overseen by fighters, justice systems are often operated by elders, activists, chiefs, and religious figures. Sometimes providing order and reducing violence, courts typically involve coercion, perhaps sentencing collaborators or others to capital punish or banishment. In Aceh, many Islamic leaders served as judges in rebel courts. For smaller offenses, villagers might face fines, or exile, but state informants (cuak) faced violence. In Mindanao, rebels explained that civilians told them about a rebel that had committed sexual violence against a young woman. This led to the individual’s capture, trial, beating, and eventual disappearance (interviews in Cotabato, 2007). Armed group justice is often violent, attacking perceived enemies while increasing the control of armed actors.
While the above forms of civilian support for armed groups contribute to violence, the following forms feature even more direct linkages to violent acts. Civilian violence remains indirect, but here, violence would likely not occur without these acts of support. Civilians may not pull the trigger, but they create contexts where others can, perhaps supplying the weapon or the person using it.
To complete the gamut of information provision, many civilians supply high-level intelligence to armed groups that enables violence. The narrative that began this paper provides one example, where a civilian provided false information leading to an ambush. Armed groups maintain networks of civilian spies and informants that enable them to attack their enemies. For example, rebels in Aceh’s west coast describe how a woman, angry that her sister was involved with an Indonesian soldier, told the rebels when he was likely to visit, leading to his capture (interviews in South Aceh, April 2009). Information also enables violence against other civilians. Rebel informants enabled gam to purge hundreds of state informants across Aceh. This included the families of informants and security officers, with civilian intelligence enabling rebels to target their opponents’ friends and families. Balcells (2010, 297) describes how “civilians can either facilitate or constrain” armed groups arresting, processing, and executing suspects by denouncing their neighbours; here, “civilian agency is relevant for the perpetuation of direct violence”. Denunciations are enabled by what Kalyvas (2006, 332) calls the intimacy of violence, where community ties produce knowledge used against neighbours, the “dark face of social capital.” Denunciations occur not despite an aversion to violence, but because of it, as civilians are able to indirectly utilize coercion for personal gain without taking action themselves.
Another way that civilian support drives violence is through recruitment, expanding the number of combatants and thus the scale of war. Civilians may play roles in screening potential recruits, investigating their backgrounds and providing indoctrination to those willing to take up arms.2 Recruitment is a role especially open to those not expected to fight, but with contact and influence over those that are. In Southeast Asia, some mosques and boarding schools aided in rebel recruitment, as trusted Islamic leaders encouraged their students to join. Women and elders were core recruiters, encouraging their sons and other men to enlist. In Mindanao, rebels explained that former fighters know war and are credible, directing young men to enlist; “it is hard to refuse someone who has risked his own life” (interviews in Buluan, June 2008). For one veteran of an earlier Islamic rebellion in Aceh, “I am still a soldier…but now I serve as a gam civilian. I provided old rifles and trained young people to use them” (interviews in Aceh Besar, 30 October 2007). By providing more combatants and training them to fight, civilians contribute directly to violent wars.
Just as civilian support may put more combatants in the field, it can also put weapons in their hands. Civilians may maintain, store, transport, donate, buy, and sell weapons to armed groups. Horton (1998, 207) describes popular support for Nicaraguan Contras, with peasants leading Sandinistas into ambushes as well as hiding or transporting weapons. In El Salvador, trusted civilians were tasked with purchasing materials with rebel funds and “transporting heavy weapons” (Wood 2003, 173). In Eastern Europe, Petersen (2001, 7) describes bunkers that were maintained by civilians, stored with provisions as well as weapons, including machine guns. Weapons may be procured from everyday people, such as farmers or veterans donating rifles. Youths may transport guns, bullets, and explosives for armed groups, or hide weapons before or after attacks. In Aceh, civilians report hiding weapons during military sweeps, enabling rebel soldiers to blend in and shed incriminating evidence. State soldiers, as well as civilians pressed to join state-sponsored militias, sold weapons to rebel forces; such leakages are key methods through which rebel groups acquire weapons, and often take place through civilian intermediaries. Civilians may be tasked with manufacturing and assembling weapons; in El Salvador, rebel bases included civilian-run clinics, schools, and “workshops to build crude weapons” (Wood 2003, 122). This may include developing improvised explosive devices (ied s). In Iraq, US military reports emphasize the importance of bombs, but also the bombmakers, noting how civilians “provide active support when they transport, emplace, and detonate bombs” (McFate 2005, 40). The role of civilians in weapons provision can sometimes lead experts to “qualify them as combatants and therefore as legitimate targets” (Lichtenberg 1994, 347).
Support as Direct Violence
At the edge of the spectrum are forms of support that represent direct violence, at which point civilian status dissolves and a person becomes a combatant, a status defined by representing a direct mortal threat. There are, however, some grey areas demanding discussion. Civilians may support armed groups through armed defense patrols, support challenging not only civilian innocence, but also civilian status. When armed with clubs or rudimentary weapons, such persons represent minimal mortal threats to others. This can be complicated, since self-defense forces might begin as lightly armed and loosely organized, but then be coopted over time to become pro-state militia groups, as Blocq (2014, 710) finds in South Sudan and Starn (1999) finds in Peru. We might see a self-defense group armed with clubs as civilians, perhaps supporting a particular side, but those armed with guns as combatants. As civilians participate more directly in violence, we may no longer see them as civilians.
Conclusions: Implications of Civilian Partisans
Clearly, civilians are victims of wartime violence. One reason they are targeted is that civilians are not only victims, but also actors providing forms of support for armed groups that may sustain or scuttle rebellion. Civilians are often victims in part because they are actors. When not seen as anonymous and inert, civilians are often framed as innocent and pro-peace. This may be true, but civilians are not always innocent, nor peaceful. This paper has laid out forms of support where civilians help sustain violence, increasing the coercive capacities of armed groups through forms of support that can be understood as indirect violence.
Understanding how civilians supply indirect violence brings important implications and concerns. This study focused on civil wars, with greater weight attached to rebel forces that depend on civilian support to survive. However, nothing here should exclude state actors. Be they medics or cooks, engineers, recruiters, or others, civilians play important roles in maintaining state security forces. Those in combat areas may help state forces in ways similar to how others help rebels, and those further away may contribute to the fighting capacity of armed forces in numerous ways. Youth scouts and veteran’s organizations are often sources of recruitment and legitimacy for state forces. And the military industrial complex, the companies and employees responsible for designing, testing, manufacturing, and transporting weapons, are contributing to indirect violence. A factory worker at Lockheed Martin is not committing acts violence, but their support very much makes violence possible, thus sharing some responsibility for war.
This paper has maintained a pacifist tone, largely critical of violent acts. This brings up an ethical point, that it may be understandable and even legitimate for civilians to provide indirect violence against an abusive adversary. It is probable that many people support a particular side because they wish to reduce long-term violence through victory. At times, there may be very good reasons for civilians to support armed groups, such as Ukrainians helping their army against Russian invasion. My goal is not to suggest that civilians are inherently wrong to provide indirect violence, but more to emphasize the importance and different forms of them doing so, and to recognize how this produces violence. For many civilians, this may very much be their goal, as they consciously hope that their support for an armed groups helps to defeat a hated enemy.
One potential concern is that this line of inquiry, of civilians fueling violence, may undermine civilian / combatant distinctions. This article shows that civilians may be closely aligned with armed groups, even members of affiliate organizations, and contribute indirectly to violence. Appreciating how civilians fuel violence does not mean that we should disregard civilian / combatant distinctions. If anything, this divide should now be clearer, identifying which forms of support make for indirect and direct violence. Civilian non-combatant supporters responsible for indirect forms of violence are still civilian non-combatants. This speaks to campaigns related to child soldiers as well as studies of female combatants (Wood 2019), which by confusing support roles and organizational membership with combatant status, may unintentionally lead to these groups being targeted. Support roles are especially likely among persons who are not allowed or expected to take up arms, the very women, youths, and elders that make for marketable victims (see Horton 1998, 206).
This study has focused more on civilian actions than on their motivations. It is notoriously difficult to clearly establish intent by any actor. In civil wars, few decisions are made freely, with some civilian support for armed group forced, and most civilian partisanship a result of some degree of pressure. Many forms of support analyzed are indeed products of duress, perhaps representing what Coulter (2009, 146) refers to as “choiceless decisions”. This is true for both civilians and combatants, as the actions of supporters and recruits are often a result of duress. If civilians are truly forced to provide forms of support that contribute to violence, then we can still see them as innocent, as they are not responsible for violence. This said, we must not write off civilian agency for the sake of clearer moral categories, maintaining the innocent civilian by characterizing indirect violence as involuntary. The above caveats about helping state forces and the potentially legitimate nature of indirect violence both suggest voluntary actions. Involuntary support can be costly for armed groups, requiring resources, undermining popularity, and inviting forms of resistance. The forms of support that most clearly drive violence, especially over time, are those that are more or less voluntary. Civilians that build or store weapons, or help with recruitment or training, are likely to be trusted by and committed to the armed group. In my fieldwork, many civilians report being proud of their contributions. This was true even for those in contested regions, where partisanship was costly and armed groups lacked the capacity for oversight (see also Wood 2003, 12). Many civilian partisans take action out of conviction, perhaps for revenge or belief in a given cause, as well as profit, social status, and various other reasons that may overlap or shift over time. When civilians are forced to provide indirect violence, we may still see them as innocent, although many forms of support that represent indirect violence are especially likely to be more voluntary in nature.
Crucially, many of the most dedicated supporters of armed groups are also victims of the other side, with civilians motivated by a need for justice or revenge. This means that many victims are simultaneously responsible for indirect violence that drives the victimization of others. This may complicate humanitarian work, since innocence is more likely to satisfy donors and compel action. But a more realistic approach may also make for more effective humanitarian aid. The chequered nature of civilian innocence should not make civilians legitimate targets in war—they remain unarmed and do not represent direct mortal threats, deserving civilian status and protection.
Acknowledgements
The author would like to thank panelists and audiences at 2022 wpsa, isa, and isa-West for critical feedback, as well as research assistants Sachi Nishida, Jerry Yang, Mukesh Bastola, and Mira Peregud. Thank you also to the Jeremy, Berit, and Felicity as guest editors and helpful comments from three reviewers.
Notes
Kaplan briefly discusses civilians allying with armed groups, for “protection from that actor as well as incursions by a nearby enemy army” (2017, 45). This seemingly frames support as rooted in necessity rather than conviction.
Civilians may recruit other supporters—identifying, training, and vouching for persons to provide forms of support discussed in this paper.
References
Arjona, A. (2016). Rebelocracy: Social Order in the Colombian Civil War. Cambridge University Press.
Aspinall, E. (2002). Sovereignty, the Successor State, and Universal Human Rights: History and the International Structuring of Acehnese Nationalism. Indonesia 73, pp. 1–14.
Aspinall, E. (2009). Islam and Nation: Separatist Rebellion in Aceh, Indonesia. Stanford University Press.
Avant, D., et al. (2019). Civilian Action and the Dynamics of Violence. Oxford University Press.
Baines, E., and Paddon, E. (2012). ‘This is How We Survived’: Civilian Agency and Humanitarian Protection. Security Dialogue 43 (3), pp. 231–47.
Balcells, L. (2010). Rivalry and Revenge: Violence against Civilians in Conventional Civil Wars. International Studies Quarterly 54 (2), pp. 291–313.
Barter, S.J. (2012). Unarmed Forces: Civilian Strategy in Violent Conflicts. Peace & Change 37 (4), pp. 544–71.
Barter, S.J. (2013). State Proxy or Security Dilemma? Understanding Anti-Rebel Militias in Civil War. Asian Security 9 (2), pp. 75–92.
Barter, S.J. (2014). Civilian Strategy in Civil War: Insights from Indonesia, Thailand, and the Philippines. Palgrave.
Bellamy, A.J. (2009). Responsibility to Protect. Polity Press.
Blocq, D.S. (2014). The Grassroots Nature of Counterinsurgent Tribal Militia Formation: The Case of the Fertit in Southern Sudan, 1985-89. Journal of Eastern African Studies 8 (4), pp. 710–24.
Carpenter, C. (2005). ‘Innocent Women and Children’: Gender, Strategic Frames, and the Protection of Women and Children. International Studies Quarterly 51 (1), pp. 99–120.
Carr, C. (2003). The Lessons of Terror: A History of Warfare against Civilians. Random House.
Chenoweth, E., and Stepan, M.J. (2011). Why Civil Resistance Works: The Strategic Logic of Nonviolent Conflict. Columbia University Press.
Child Soldiers. (2008). “Child Soldiers: Global Report 2008 Indonesia.” Coalition to Stop the Use of Child Soldiers.
Coulter, C. (2009). Bush Wives and Girl Soldiers: Women’s Lives Through War and Peace in Sierra Leone. Cornell University Press.
Downes, A.B. (2011). Targeting Civilians in War. Cornell University Press.
Eck, K., and Hultman, L. (2007). One-Sided Violence against Civilians in War: Insights from New Fatality Data. Journal of Peace Research 44 (2), pp. 233–46.
Fortna, V.P., Lotto, N.J., and Rubin, M.J. (2018). Don’t Bite the Hand that Feeds: Rebel Funding Sources and the Use of Terrorism in Civil Wars. International Studies Quarterly 62 (4), pp. 782–94.
Geneva Convention Relative to the Protection of Civilian Persons in the Time of War (1949).
Guevara, C. (1967). The Bolivian Diary. Available at www.marxists.org/archive/guevara/1967/bolivian-diary/index.htm.
Horton, L. (1998). Peasants in Arms: War and Peace in the Mountains of Nicaragua, 1979-1994. Ohio University Center for International Studies.
Hultman, L. (2013). UN Peace Operations and Protection of Civilians: Cheap Talk or Norm Implementation? Journal of Peace Research 50 (1), pp. 59–73.
Humphreys, M., and Weinstein, J. (2006). Handling and Manhandling Civilians in War. American Political Science Review 100 (3), pp. 429–47.
icrc (2021). “ihl Rule 5: Definition of Civilians.” Available at https://ihl-databases.icrc.org.
icrc (2010). “Civilians Protected under International Humanitarian Law.” Available at www.icrc.org/en/doc/war-and-law/protected-persons/civilians/overview-civilians-protected.htm.
Johnson, C.A. (1962). Civilian Loyalties and Guerrilla Conflict. World Politics 14 (4), pp. 646–61.
Kalyvas, S. (2003). The Ontology of ‘Political Violence’: Action and Identity in Civil Wars. Perspectives on Politics 1 (3), pp. 475–94.
Kalyvas, S. (2006). The Logic of Violence in Civil War. Cambridge University Press.
Kaplan, O. (2017). Resisting War: How Communities Protect Themselves. Cambridge University Press.
Keels, E., and Kinney, J. (2019). ‘Any Press is Good Press?’ Rebel Political Wings, Media Freedom, and Terrorism in Civil Wars. International Interactions 45 (1), pp. 144–69.
Kinsella, H.M. (2015). The Image before the Weapon: A Critical History of the Distinction between Combatant and Civilian. Ithaca: Cornell University Press.
Lichtenberg, J. (1994). War, Innocence, and the Doctrine of Double Effect. Philosophical Studies 74 (3), pp. 347–68.
Lujala, P. (2009). Deadly Combat over Natural Resources Gems, Petroleum, Drugs, and the Severity of Armed Civil Conflict. Journal of Conflict Resolution 53 (1), pp. 50–71.
Mao, Z. (1937). On Guerrilla Warfare. Available at https://www.marxists.org/reference/archive/mao/works/1937/guerrilla-warfare/ch06.htm.
Masullo, J. (2021). Refusing to Cooperate with Armed Groups: Civilian Agency and Civilian Noncooperation in Armed Conflicts. International Studies Review 23 (3), pp. 887–913.
McCargo, D. (2006). Communicating Thailand’s Southern Conflict. The Journal of International Communication 12 (2), pp. 19–34.
McFate, M. (2005). Iraq: The Social Context of ied s. Military Review. Arlington: Office of Naval Research.
McLauchlin, T. (2015). Desertion and Collective Action in Civil Wars. International Studies Quarterly 59 (4), pp. 669–79.
Moore, W.H., and Shellman, S. (2006). Refugee or Internally Displaced Person: To Where Should One Flee? Comparative Political Studies 39 (5), pp. 599–622.
Petersen, R.D. (2001). Resistance and Rebellion: Lessons from Eastern Europe. Cambridge University Press.
Pye, L.W. (1950). Guerrilla Communism in Malaya: Its Social and Political Meaning. Princeton University Press.
Rood, S. (2005). Forging Sustainable Peace in Mindanao: The Role of Civil Society. East-West Center Policy Studies 17.
Slim, H. (2010). Killing Civilians: Method, Madness, and Morality in War. Oxford University Press.
Slim, H. (2003). Why Protect Civilians? Innocence, Immunity, and Enmity in War. International Affairs 79 (3), pp. 481–501.
Starn, O. (1999). Nightwatch: The Politics of Protest in the Andes. Duke University Press.
Stokke, K. (2006). Building the Tamil Eelam State: Emerging State Institutions and Forms of Governance in ltte-Controlled Areas in Sri Lanka. Third World Quarterly 27 (6), pp. 1021–40.
US Army. (2009). “Tactics in Counterinsurgency: Field Manual Interim.” Washington DC: Department of the Army.
Valentino, B., Huth, P., and Balch-Lindsay, D. (2004). “Draining the Sea”: Mass Killing and Guerrilla Warfare. International Organization 58 (2), pp. 375–407.
Weinstein, J. (2007). Inside Rebellion: The Politics of Insurgent Violence. Cambridge University Press.
Wickham-Crowley, T. (1992). Guerrillas & Revolutions in Latin America: A Comparative Study of Insurgents and Regimes since 1956. Princeton University Press.
Wood, E.J. (2003). Insurgent Collective Action and Civil War in El Salvador. Cambridge University Press.
Wood, R.M., Kathman, J.D., and Gent, S.E. (2012). Armed Intervention and Civilian Victimization in Intrastate Conflicts. Journal of Peace Research 49 (5), pp. 647–60.
Wood, R.M. (2010). Rebel Capability and Strategic Violence against Civilians. Journal of Peace Research 47 (5), pp. 601–14.
Wood, R.M. (2019). Female Fighters: Why Rebel Groups Recruit Women for War. Columbia University Press.
Zhukov, Y.M. (2016). On the Logistics of Violence: Evidence from Stalin’s Great Terror, Nazi-Occupied Belarus, and Modern African Civil Wars. In: C. Anderton and J. Brauer, eds., Economic Aspects of Genocides, Other Mass Atrocities, and their Prevention, Oxford University Press, pp. 399–424.