Abstract
Discipline forms the backbone of all military organisations. While discipline is traditionally associated with draconian punishment, this association is increasingly only applied to non-Western contexts. African rebel movements and similar, weak organisations are represented especially often as lacking non-coercive means of instilling discipline. This article explores the utility of coercive discipline in one such context – the Second Liberian Civil War (1999–2003). I argue that Liberia's weak military organisations faced significant restrictions when it came to employing direct coercion. Executions, which are often equated with coercion in existing literature, threatened to rive the already frail organisations. Even other formal instruments of discipline, such as military hierarchies and rules and regulations, remained contested throughout the war. Consequently, more indirect means were adopted. Ultimately, the main users of coercion were not military organisations, but peers. This suggests that it is easier for strong organisations to coerce their members, and that the relationship between coercion and organisational strength may need to be reassessed. Furthermore, existing positive perceptions of camaraderie between brothers-in-arms requires re-evaluation.
Nyonbu Tailey was an elephant hunter and a kinsman of President Samuel Kanyon Doe, who had risen to power through a military coup in 1980. When the rebels were moving closer to Monrovia in 1990, the desperate Doe promoted Tailey, who was not a soldier by training, to the rank of captain. When the war reached the capital, Tailey protected the port with his fellow soldiers from the Armed Forces of Liberia (AFL). But when the fighting intensified he began shooting at his own men, ordering them to stay put. Tailey's forces shrank in two waves, first after he shot at the soldiers, and later when others deserted in self-preservation. Finally, Tailey's remaining forces were so weak and demoralised that they fled in the face of the larger enemy. Consequently, the AFL lost the port to rebels, who later captured Doe there. He was subsequently tortured and killed. Tailey fared no better. After the president's demise, he and his young “death squad” attempted to take over the AFL – if not the state – to execute the maxims “No Doe, no Monrovia” and “It's not the size, it's the tribe” against the civilian population that remained in the capital. In effect, he and his underage followers began to burn down the homes of those deemed suspect, on ethnic grounds alone. 1 Tailey's actions against both his own fighters and civilians were soon perceived as being too violent by soldiers and civilians alike. He met his end when one of his own fighters knocked off his hat – which was supposed to give him supernatural protection against bullets – after which Tailey was shot and killed by his fellow soldiers.
Tailey is also accused of having led the massacre of hundreds of displaced people at the Lutheran Church compound (although his name is misspelled) (Williams 2002: 103–104).
This well-known narrative from the beginning of the First Liberian Civil War is a typical story of African warfare: It includes child soldiers, patrimony, supernatural forces, tribalism, and brutal violence used against both civilians and fellow soldiers. When it comes to the latter group, a “No die, no rest” attitude to discipline – subjugating defiance – existed in Liberia. Drawing from a 1980s Nigerian highlife song of the same name, the expression was used during Liberia's civil wars (1989–1996 and 1999–2003) to refer to situations where combatants would have to keep on fighting until killed either by the enemy or by their own comrades. Struggling with the problem faced by all military organisations of establishing discipline, “No die, no rest” thus began to characterise coercive situations where fighters were controlled through threats, if not actual use of force. Some military operations were even called “No die, no rest”, and contributed to the prevailing idea in the scholarly literature that the years of conflict in Liberia were particularly violent and uncivil (Edgerton 2002: 156–162; Ellis 2007: 20–22).
Yet when it comes to coercive discipline, Tailey's story is in fact atypical of war in Liberia. As his fate shows, Tailey's use of violence was simply too radical for his comrades. Instead of producing discipline, extreme coercion led to its disintegration. While “No die, no rest” existed as a notion, military organisations struggled to implement it in practice. The main argument of this article is that direct coercion – especially executions, which the literature often takes as the only measure of coercion – was never the main method of instilling discipline during the Second Liberian Civil War. This goes against the expectation that weak organisations lack non-violent means to control their members. These kinds of organisations are often seen to consist of those coming from the dregs of society, who can be controlled only through “indiscriminate use of drugs, forced induction, and violence” (Abdullah 1998: 223). Mueller agrees, and adds that contemporary wars are characterised by lack of discipline and almost exclusively occur in poor countries (Mueller 2003). Ultimately, there is an assumption that weak military organisations frequently resort to violence and extreme coercion in order to uphold discipline (Herbst 2000: 279–280).
This article seeks to examine this assumption through an investigation of the use of coercive discipline in Liberia. The utility of extreme coercion was limited, because if used on a wide scale it could have undermined the already bristling cohesion of Liberian military organisations. Because of lack of formalisation and shared norms, harsh disciplinary action was experienced as unjustified and illegitimate. Even further and as exemplified by Tailey's fate, extreme coercion potentially endangered the life of whoever was applying it. Consequently, the Liberian military organisations had to do as their like around the world, and resort to more indirect measures to instil discipline.
The article proceeds as follows: the following section identifies coercion as both the main traditional source of discipline in military organisations and the use of power. As a result, Lukes’ three-dimensional view of investigating and exercising power is adopted as the theoretical framework that will later structure the investigation of coercive discipline in Liberia. These three dimensions respectively conceive power as decision-making, agenda-setting and preference-shaping. This section also advocates the use of (European) experiences of discipline as a heuristic tool to understand discipline in Liberia and elsewhere. The third section describes the ethnographic methods used for this study. Long-term presence in the field was arguably necessary for the investigation of a controversial topic in a politicised environment. The fourth section focuses on the Liberian context, and especially the military organisations there, which were rather ad hoc and weak, and fought in ways that made it difficult to apply the old mechanistic view of draconian discipline. The fifth section investigates the use of coercive discipline in Liberia through Lukes’ three dimensions of power. It is argued that there is little evidence that extreme coercion in the form of executions was widely used to instil discipline among the warring organisations. The norms of the top level within military organisations were not widely shared by their members, and remained contested. These difficulties regarding direct coercion are illustrated through an investigation of attempts to set up hierarchies and to uphold rules and regulations. Instead, less direct means had to be employed. The indirect means focused on limiting mobility: ethnic polarisation made it difficult to stay neutral in the conflict, and checkpoints kept fighters close to frontlines, where the circumstances alone forced them to fight. Finally, compliance was secured through dissemination of norms. Many combatants ultimately assumed a new fighter identity. But this identity came with the expectation that group norms would be adhered to. Ultimately, discipline was mainly maintained by peers. Because they often resorted to coercion, the previously upheld understanding of a positive role of comrades in arms should be re-evaluated. The sixth and final section discusses the theoretical implications of this investigation. 2
The author would like to thank the participants of the seminars at the Nordic Africa Institute in April 2014 and the Swedish Defence University in June 2014, the anonymous reviewers and the editors of this journal, and especially Peter Haldén, Danny Hoffman, Mats Utas, and Jan Ångström for their constructive comments.
Coercion, Discipline, Organisation, and Power
The defining characteristic of all military organisations is the use of organised violence, whereas the majority of those who form these organisations instinctively seek to survive these ordeals. This creates a clear contradiction between the interests of these organisations and their members, which becomes a fundamental problem these organisations need to solve. The traditional solution to this problem is discipline, which forms the backbone of all military organisations. To discipline essentially means to subjugate defiance; discipline is utilised to help ensure that orders are obeyed despite their unpleasantness or dangerousness (Baynes 1967: 180–181; Du Picq 2013: 111; Westmoreland 1971). As the notion of “No die, no rest” underscores, the disciplinary methods used in the Liberian military organisations discussed here are ultimately coercive.
As Mills argues, “the ultimate form of power is coercion by violence” (Mills 1999: 172). Power and coercion are even present in Dahl's standard definition of power, according to which “A has power over B to the extent that he can get B to do something that B would not otherwise do” (Dahl 1957: 202–203). The usefulness of power on the frontlines is obvious. For Weber, power could come in the form of either authority or coercion, where the former corresponds to legitimate use of power, and the latter to illegitimate use of power. In the absence of authorities, any use of power relies on force alone (Weber 1978: 212–215). In other words, in the absence of authority, coercion is required. A more recent take comes from Lukes, who criticises Dahl's “one-dimensional” understanding of visible use of power as insufficient and argues for a “three-dimensional view” of power as a contested concept. These three dimensions of power can be applied not only to analyse the use of power, but also to exercise it: Lukes’ “second dimension of power” focuses on the more indiscernible limiting of some alternatives of action altogether, while his “third dimension of power” equals “the capacity to secure compliance through the shaping of beliefs and desires, by imposing internal constraints under historically changing circumstances” (Lukes 2005: 144). This third dimension thus focuses not on forcing someone to comply, but instead on securing willing compliance. Lukes thus sees power as something much more than Dahl's classic, yet both more visible and narrower, understanding. For Lukes, “power is at its most effective when least observable” (Lukes 2005: 1). Lukes’ three dimensions will be employed later on as the theoretical framework that structures the investigation of Liberian coercive discipline.
Discipline and coercion are not necessarily obvious in the West any longer, and tend to be taken as given. This in part has to do with the recent professionalisation of military organisations, where those who form them are expected to be motivated by non-coercive causes. Even further, the coercion used in the conscript-based mass armies was often deemed legitimate to the extent that it was not strongly questioned. Other factors, such as nationalism, offered non-violent motivations. The Liberian case can shed light on both coercive and non-coercive motivations often assumed by Western political science. In what follows, the European history of discipline is used as a heuristic tool to understand the use of discipline not only in Liberia, but also elsewhere – including the contemporary West. Two things need to be pointed out in regards to this approach. First, for a central concept, surprisingly little has been written about discipline in the explicit military context. This almost necessitates the use of historical examples, which tend to be European. Second, while the decision to treat Liberian military organisations and their use of coercive discipline as cases of a broader phenomenon can be criticised on several accounts, so can the alternative of only focusing on these organisations as somehow inherently African. The latter alternative comes with an uncomfortable suggestion that Africa – if not Africans – is exceptional, and that it bears little similarity to other places and thus offers few opportunities for scientific exchange between continents. As the rest of this paper will show, the cases share certain similarities and can offer some interesting lessons, thus justifying this choice.
It is useful to begin the investigation of discipline from Frederick the Great and the mid-eighteenth century. Half of Frederick's troops consisted of foreign mercenaries “whom no bond ties to the state, [and who] seek to run away at every opportunity” (Frederick the Great 1753). Underlining how pressing the problem of discipline was, he devoted the very first section of his
Yet, it was already recognised that while soldiers were isolated from society and as a result could be treated harshly, officers were different. This began to change as Enlightenment ideas gained ground, and soldiers were increasingly recognised not only as individuals with honour but, because of growing nationalist sentiments, equally as citizens. The military, in turn, came to be portrayed as a school of the nation. New tactical innovations were less fixed and required more autonomy from soldiers, and as a result discipline became looser. Consequently, draconian discipline began to give way to more humane treatment and patriotic education even among the rank and file. By the time Du Picq died in 1870, he had already noted this important change. According to him, “discipline is for the purpose of dominating […] horror by a still greater horror, that of punishment or disgrace” (Du Picq 2013: 94). Only a century before, when it was still up for debate whether the rank and file had honour to begin with, disgrace was perceived to have limited utility. This had left corporal punishment as the predominant disciplinary method.
Whereas previous wars had been fought with smaller and more homogeneous forces, in the First World War whole nations were mobilised in an industrial manner and on a grand scale to fight a total war with an almost industrial quality of death and destruction. While this could be taken as a return to machinelike discipline, the war shows that, in fact, the influx of civilians into the military also led to inevitable norm conflicts between professionals and citizen-soldiers. For instance, Sheffield notes that in the British volunteer forces, where personal relationships between commanders and subordinates remained crucial for discipline, “an attempt to impose Regular-style discipline would have led to men leaving the Auxiliary forces” (Sheffield 2000: 17). It was only with the First World War that self-discipline began to play a greater role than imposed formal discipline (Kellett 1982: 9; see also Baynes 1967). Yet this is not to say that coercive discipline had ceased to play a role in military organisations. In the aftermath of the Second World War, an extensive questionnaire-based research programme conducted for the United States War Department targeted American servicemen. When asked about punishment, only 23 per cent of the officers and 20 per cent of enlisted men agreed to the claim that it was the “best way” to ensure good behaviour. However, 46 per cent of officers and 67 per cent of enlisted men agreed that “the main reason most soldiers obey rules and regulations is because they are afraid of being punished” (Stouffer 1949: 417).
Coercive discipline is, however, rarely discussed when it comes to contemporary Western military organisations, which have little desire for machinelike soldiers, as personal autonomy and initiative have become prized and sought-after attributes. Coercive discipline would also not help with recruitment and retention of voluntary forces (Strachan 2006: 226). The gradual shift away from coercive discipline has, however, been very different in other contexts. For instance, different rules applied when it came to what was described as exceptional but necessary colonial disciplining of “uncivilised” “other bodies” (Rao and Pierce 2006). This perception arguably continues in a sense, as coercive discipline has mostly been omitted in literature that investigates contemporary Western military organisations. As a result, it remains a practice associated with others.
This association to some extent arises from the perceived connection of discipline and forms of organisation. In fact, Du Picq went so far as to say that it is only organisations that can produce discipline (Du Picq 2013: 122; see also van Creveld 1982: 163). Some organisations are, however, unmistakably designed for warfighting and possess highly developed organisational structures that offer a wide variety of instruments to create and maintain discipline. Others, like the ones in Liberia, are comparatively ad hoc and are seen to possess more limited means of fostering discipline, which are furthermore allocated to more immediate incentives. A dichotomy of stronger and weaker military organisations thus exists, where the former are seen to mainly resort to non-coercive means.
Finally, while coercive discipline has changed profoundly over time in the West, existing literature still typically equates coercion with executions (King 2013: 214-215). While executions are seemingly easily observable and quantifiable, they do not say much about the less visible (and more everyday) forms of coercion. In addition, because of the sensitive nature of the topic of execution, there is reason to be sceptical of some of the existing source material (Ylikangas 2009: 89–156). Both of these issues raise questions about the usefulness of executions as a measure of coercion.
Ethnographic Fieldwork with Former Combatants in Liberia
This article builds mainly on data gathered from ethnographic fieldwork. I spent a total of ten months during two field trips (in 2012 and 2013) investigating Liberian military organisations, focusing on former combatants from all major factions of the civil wars in seven of the fifteen counties of Liberia. Semi-structured interviews with former combatants were the main method employed. The length of interviews varied from brief, unplanned encounters to living for weeks with informants. Fieldwork concentrated on approximately 30 key informants of varying ethnicity, faction, geographic location, and military rank. I spent anywhere from a few dozen to several hundred hours with each. Some of the findings received from these key informants were verified in interviews with other former combatants, who number in the few hundreds. Semistructured group interviews became an especially effective way to accomplish this.
A long process of building trust was deemed necessary to undertake this research, as many aspects of the wars remain sensitive to this day. An element of informants’ potential mistrust of researchers is the lingering rumours concerning a potential war crimes court that could especially target those who held higher positions during the war. All sources have been anonymised in order to prevent any negative consequences for them.
Several independent groups of informants were interviewed based on the snowball-sampling recruitment method (Russell 2006: 192–194). Additionally, my long-term presence and open intentions (which were deemed necessary to avoid feeding into suspicions) led to additional spontaneous encounters with former combatants. Most of these key informants were revisited after a year, when I revisited my main fieldwork sites. This made it possible to return to certain topics, thereby improving the reliability of the data.
The analysis also benefitted from the work of Anders Themnér, who permitted me to utilise a further 238 semi-structured interviews of former combatants conducted for another research project. This material contained information about disciplinary measures, and was collected in eight counties between 2010 and 2013 by Themnér and myself. It has been used to further control some of the findings.
Liberian Military Organisations
All Liberian military organisations have been influenced by the Armed Forces of Liberia (AFL). The AFL has its roots in the Liberian Frontier Force (LFF), which was established in 1908 to pacify Liberia's interior in order to keep encroaching colonialist neighbours at bay. Organised by British officers following European models used in neighbouring countries but soon overtaken by Americans, the LFF acted with impunity and often paid little heed to its political masters in the capital. There were too many officers, who furthermore were selected due to patronage rather than competence. Paid a meagre salary that was often several months in arrears, LFF soldiers often resorted to looting and robbery (Akingbade 1976: 136–166). Many of these problems formed a legacy that was handed down first to the AFL, and subsequently to the military organisations formed during the two civil wars, as former soldiers from the AFL manned central positions in all of them.
These two wars were fought between 1989 and 2003. The First Civil War began on Christmas Eve in 1989 and continued until 1996, with presidential elections the year after. The Second Civil War began only a few years after these elections, following a short period of calm brought about as a result of the electoral victory of the rebel leader Charles Taylor. As Taylor proved unable to bring stability, many began to see regime change as the only way to bring peace. By 1999, Liberian dissidents operating from neighbouring Guinea, who came to be known as Liberians United for Reconciliation and Democracy (LURD), attacked northern Liberia. But as Taylor's Government of Liberia (GoL) proved difficult to dislodge, fighting dragged on. In early 2003, southeastern Liberia was invaded by another group of rebels, who assumed the name Movement for Democracy in Liberia (MODEL). Following the departure of Taylor in August 2003, the government forces, LURD and MODEL combined to create a transitional government, thus ending the conflict that had begun fourteen years earlier (Käihkö 2015).
In order to link Liberia to the more general discussion of discipline above, the nature of the latter war in Liberia warrants some discussion. This was a small war, with limited combat. Fighting forces were typically dispersed into small formations, which ensured that casualties between them remained low. Much of the fighting was done in densely forested areas and required significant independence and initiative, which further made upholding strict discipline difficult. There were long periods that saw little fighting on several fronts. Violence, therefore, does not always feature prominently in fighters’ narratives of war. Second, while many of these fighters had inherently political motives, there was no overarching ideology such as nationalism that they could draw from. Not only was the war a civil war, but those who fought in it had very mixed origins: in addition to Liberians from different ethnic groups, the factions also included foreign nationals, such as Guinean rebels, Ivoirian militiamen, and Sierra Leoneans from the Revolutionary United Front (RUF) and the Civil Defence Forces. To some extent, ethnicity played a similar, yet lesser role, but it was never backed up with formal structures comparable to nation-states, for instance. As argued by Malešević, this lack of “institutional and ideological devices” has profound consequences for discipline. According to him, it is “social organisation, an external mechanism of social control, [which] is a backbone of military might” (Malešević 2010: 113). Curiously, the GoL forces were arguably
Lukes’ First Dimension of Power: Never Quite “No die, no rest”
As noted, “No die, no rest” was used in Liberia to express the idea that the only alternative to fighting was death – either at the hands of the enemy, or through execution by one's own side. Yet it was difficult to put this idea into practice, as exemplified by the case of Captain Tailey. He remains a pertinent example often used by informants with military backgrounds (who, because of their positions, were the people who would have employed coercion) to describe the problematic utility of executions and other direct coercive techniques to establish discipline during the Liberian civil wars. This should, however, not be altogether surprising, considering that even an extreme case like the Lord's Resistance Army – which almost completely relies on forced recruitment –could not exclusively rely on coercion (Blattman and Annan 2010; Titeca 2010). Such coercive measures fit under Lukes’ first dimension of power, which focuses on the visible use of power that coerces someone to do something they would not otherwise do.
Why did coercion fail to achieve the desired effect in Liberia? Ultimately, the disintegration of discipline among Tailey's forces was the result of the AFL's weak organisation. For Weber, discipline is rationalised, bureaucratised, and above all impersonal (Weber 1946: 253–254). This kind of formalisation did not take hold in Liberia, where the weak organisations never succeeded in convincing their members of the necessity of drastic measures. Following Weber, enforcing punishments not based on shared norms is illegitimate. As Westmoreland observes, military justice should promote discipline – “an unfair or unjust correction never promotes the development of discipline” (Westmoreland 1971). Fairness can only be achieved through shared understanding, and unfair punishments do not easily lead to discipline. This is also evident in the way peers often begin to control each other's behaviour. As one American study noted, “where the formal controls […] are not supported by the informal social pressures of one's fellows, not to mention internalisation which operates even in the absence of one's fellows, there is almost certain to be widespread violation of the rules” (Stouffer 1949: 411). While collective punishments can be used for similar effect, as long as they are illegitimate they remain problematic. From the point of view of an organisation, self-discipline is naturally preferable because it spares resources. As will be discussed later, signs of this were visible during the Second Liberian Civil War.
In the case of Liberia, the weak organisation and the consequent lack of shared norms contributed to a limited use of executions during the Second Civil War. For instance, only two executions took place in Voinjama when it served as the headquarters of LURD from 2000 to 2003. Both resulted from quarrels between members of the rebel movement. 3 At least one of these executions was carried out by the provost marshal, and the accused was shot on the spot. Several informants state that no executions took place among the MODEL fighters. 4 As one MODEL fighter noted, “rebel war” had “no court trial system.” 5
High-ranking LURD commander, 11 September 2012, Monrovia
MODEL task force commander, 11 August 2012, Zwedru; MODEL commander, 24 August 2012, Zwedru.
MODEL fighter, 12 September 2012, Monrovia.
Military norms failed to completely replace civilian ones in Liberia. Furthermore, lack of formalisation in organisations highlighted the importance of personal relationships (Brabazon 2003: 7). Several commanders were afraid of using too much coercion: as Tailey's example shows, executions could have easily turned fighters against their commanders. “Execute somebody – they execute you, too,” was how one of them put it.
6
Even less severe punishments could be seen as unreasonable, as illustrated by the case of MODEL's commanding general, who repeatedly beat up his fighters for failing to follow orders. His perceived violent nature caused ill feelings towards him, and even made his
LURD commander, 2 November 2013, Zwedru.
MODEL fighter and
LURD official, 16 November 2013, Monrovia.
Punishing relatives was impossible in practice, which is why commanders could send their family members to other units 9 and choose to serve away from their home communities. 10 Finally, as all the factions ultimately struggled with manpower issues, in most cases they could ill afford to execute their own fighters. This imperative to preserve limited force also contributed to the limited use of executions. Several fighters who knew that they could face consequences for misbehaviour found safety in the frontlines.
LIMA fighter, 3 October 2012, Monrovia.
Low-level MODEL commander, 18 July 2012, Grand Gedeh.
All in all, interviews do not support the hypothesis that the popular idea of “No die, no rest” was a widely implemented notion, or that executions were extensively used as a disciplinary measure during the Second Liberian Civil War. This may sound puzzling, as commanders sometimes claim that they could execute their own men for mistreating civilians. Even ex-combatants often refer to other commanders who were extremely violent, and who used indiscriminate violence against their own men. Yet, curiously, no combatants serving under commanders who claim to have exercised harsh discipline have confirmed these accounts.
This discrepancy between the commanders’ and fighters’ narratives may be explained by the fact that few interviewees wanted to admit to participating in brutality during the wars. Similarly, ethical considerations make it difficult to pose direct questions about what often amount to war crimes. Accounts of specific events, though, were often brought up by interviewees themselves. The few commanders who claim to have executed those who were brutalising civilians may be attempting to justify some violence, or at least to paint an image of a force more disciplined than it was. The latter tendency was especially obvious during interviews with political leaders associated with the military organisations. Yet if executions were employed as a deterrent, they should have been advertised within organisations as a warning to others, which was not the case. Similarly, fighters who presented themselves as victims might have been expected to emphasise the presence of extreme coercion as a reason for their participation in the war (Utas 2003: 22–24); however, only a few did so – mostly government fighters and those recruited by LURD in the final stages of the war during its march towards the capital. As a result, there is limited evidence that executions constituted a common instrument of coercive discipline among the rebel forces of the Second Liberian Civil War.
Naturally, Liberian military organisations sought ways to instil discipline among their ranks. Two methods stand out: imposing hierarchies, and establishing rules and regulations. These will be investigated in the following sections.
Hierarchy
Military organisations are almost invariably hierarchical. Hierarchies –and their outward manifestations, such as rank and insignia – demand obedience, and thus help to generate authority and discipline. In Liberia, the lack of formalisation had a major negative effect on attempts to develop hierarchies. As a Library of Congress report dating from 1985 describes, the 1980 coup had seriously affected discipline in the military, and resulted in a situation where “the reluctance of most officers to impose discipline had combined with the unwillingness of more than a few enlisted men to accept it.” Even further, “since the coup, the AFL had operated almost exclusively on the basis of directives, rather than written regulations that codify standard methods of operation” (Ehrenreich 1985: 273). In addition, when many officers were retired after the coup, their replacements were often selected with criteria other than merit in mind (Ehrenreich 1985: 263). Tailey's promotion shows that this practice still existed a decade later. This lack of professionalism among the officer corps continued to be a source of disciplinary problems until the beginning of the war in 1989, when some soldiers refused to take orders from their illiterate and poorly trained officers. 11 While the AFL obviously struggled to impose a hierarchy, it must still be seen as the most formalised of all the Liberian military organisations, as it drew on some shared military norms. These were not easily accepted by the citizen-soldiers fighting in the three organisations of the Second War, as most of their members lacked previous experience with the AFL (Käihkö forthcoming).
MODEL task force commander, 11 August 2012, Zwedru.
While military hierarchies existed in all factions, they were partly based on a different logic than the strict bureaucratic one, with the result being that formal hierarchies became disputed. The best example is the commanding general of MODEL mentioned above. Because he spent most of his time in Côte d'Ivoire, other top commanders were able to establish themselves in Liberia. When one such commander struck the commanding general in the face after accusing him of looting too much, no disciplinary action followed. Despite such insolence, there was little the commanding general could do. 12 Clearly, discipline could not be automatically imposed through hierarchies.
MODEL task force commander, 19 August 2012, Zwedru.
As already discussed, non-military dynamics also affected military hierarchies. It was often difficult to know for certain who ranked higher, and why. In one MODEL case, each of two commanders claimed that he, himself, ranked higher. 13 One of them, though, recognised the other as an “elder brother” (a slightly older man from the same area, to whom he probably had a distant blood relation), which made the asymmetry in military rank irrelevant. “Tradition [was] more important than rank” 14 because the younger man needed to be mindful of the consequences he would face after the war if he broke the prevailing social norms. Similarly, because of the fear of upsetting social relations (for a comparison, see Sheffield 2000: 17), some promotions were based on seniority rather than merit. Political meddling also served to confuse the situation: because hierarchies were ultimately based on control of supplies, politicians could bypass chains of command and directly empower their trusted favorites. This kind of favouritism led other, less favoured commanders to question hierarchies altogether. 15
MODEL commander, MODEL company commander, 19 August 2012, Grand Gedeh.
LURD commander, 21 October 2013, 2 November 2013, Zwedru.
MODEL commander, 19 July 2012, Zwedru.
Ultimately, many fighters – with or without experience in the AFL – drew a distinction between “city rank” (of formal organisations such as the AFL) and informal “bush rank” (used in the rebel organisations and militias), and noted that only the latter really played a role, as it was based on actual performance. 16 Some aspects of meritocracy thus existed within all Liberian military organisations: the commanders that could both grant special privileges to their subordinates and keep them alive were understandably favoured by fighters. By controlling many fighters, such commanders could in turn claim recognition of rank. As a result, at least one ambitious LURD commander left a safe position behind the lines for frontline duty simply because this was the only way he could ascend in the organisation. 17 While hierarchies could be used to uphold discipline to some extent, they remained contested. Relationships between commanders and subordinates remained more important than formal hierarchies for discipline.
GoL commander, 14 April 2012, Monrovia.
LURD frontline commander, 23 November 2013, Monrovia.
Rules and Regulations
As Ellis notes, Liberia came to epitomise the anarchical and uncivil wars of Africa (Ellis 2007: 17–22). This description would suggest that even the military organisations involved with these wars would be anarchic and uncontrolled. Liberian fighters’ narratives, however, do not necessarily support this view. In the Second Civil War, all factions aimed to observe standard operating procedures (SOPs), which regulated fighters’ conduct towards both each other and civilians. In addition, factions had military police or so-called “task forces,” which were described by interviewees to have focused on disciplinary issues. LURD even had a provost marshal. Finally, since many leadership positions within factions were filled by former AFL soldiers, they brought with them AFL regulations, and could even refer to the Uniform Code of Military Justice (UCMJ) and the Geneva Conventions. The existence of these mechanisms counters the idea of the Liberian military organisations as lacking discipline.
Despite these accounts suggesting that disciplinary mechanisms were in place in all factions, the reality was more complicated. Such mechanisms were probably not well known or well used. While those familiar with the AFL regulations could refer to a suitable regulation to justify disciplinary action, such regulations meant little to the vast majority of fighters. Virtually no former AFL soldier could explain the contents of the UCMJ, except to mention that the last officer responsible for enforcing it in the AFL had been killed by President Taylor after verbally offending him. 18 The same went for the Geneva Conventions, with the exception of policies concerning prisoners of war. In LURD, the provost marshal worked mostly on logistics behind the lines, and was only involved in the two executions discussed earlier. Military police and task forces did exist, but were few and only occasionally concentrated on upholding discipline due to lack of manpower. SOPs certainly existed, and most fighters can recall the transgressions identified in them. For instance, the MODEL SOP was “No looting, no raping, no loving to your friend['s] wife or girlfriend, don't kill friend or innocent people.” 19 Yet there is little agreement on how such transgressions should be punished. For instance, many fighters claimed that rape would be punishable by death, and harassing civilians by flogging. Most fighters, though, struggle to name a single person punished, let alone executed, for any violation of an SOP. The SOP prohibiting looting was contradicted by the fact that government and NGO property were sometimes declared to be fair game as it “belonged to the people.” Similarly, “abandoned” civilian property (in many cases possessions of those who had fled the fighting) could be taken. Looting was thus used as a positive incentive to encourage fighters, but at the same time commanders sought to control looting for tactical as well as selfish reasons. The part about relationships with “reserved women” was important for minimising conflicts between fighters.
AFL soldier, 1 October 2012, Monrovia.
MODEL fighter, 20 September 2013, Monrovia.
Rules and regulations were thus characterised by lack of formalisation, and the resulting negotiability. This is made evident in the varying understandings regarding the procedures of the SOPs, which were likely never documented. If they were written down, they simply did not reach down the chain of command to the frontlines. This lack of formalisation resulted in flexible rules that were difficult to enforce since they were not clearly understood. Typically, the local commander was left with the arbitrary power to enforce SOPs. While some commanders were stricter, this arrangement did not usually result in high discipline, or control of forces. Many commanders understood that they needed to choose between military efficiency and limiting immediate civilian suffering, and they chose the former. 20 The fact that perhaps the most important GoL military police commander was mainly responsible for disarming fighters coming from frontlines to the capital also suggests that the regime valued its own security over that of its citizenry. 21 To sum up, SOPs were more guidelines than actual rules, and were selectively enforced.
Mitton raises an important question when he asks whether the punishments meted out in the RUF were meant to reduce abuse, or rather to uphold commanders’ authority and control of resources; Mitton 2015: 89–90. I, too, find this question difficult to answer.
GoL Anti-Terrorism Unit military police commander, 23 April 2012, Monrovia.
Ultimately, direct coercion fared poorly in Liberia. As Marks notes in regard to the RUF, “harsh punishments were unsustainable,” as they “would have eliminated some of the fiercest fighters upon whom the RUF relied for strength in battle, including key commanders” (Marks 2013: 161). As a result, executions were not extensively used. Organisations also failed to formalise themselves in ways that would have promoted much discipline – for instance, by legitimising the use of extreme coercion. Hoffman's description of the organisation of the Civil Defense Forces in neighbouring Sierra Leone thus seems fitting even in Liberia: “hierarchies were rooted in a less bureaucratic and more personalistic set of understandings about authority and responsibility” (Hoffman 2011b: 132). As a result, more indirect and subtle means were required.
Lukes’ Second Dimension of Power: Limiting Mobility
Arguably, the one-dimensional power in Liberia was overshadowed by Lukes’ second dimension: limiting alternatives of action. As Kellett notes, discipline “limits the range of possible behaviour open to the soldier and thus increases the likelihood of his compliance” (Kellett 1982: 325). Even further, war itself becomes a force that limits the alternatives of those caught up in it. In Liberia, this offered the military organisations some ways to recruit and control fighters. The first method was the creation of in-groups and out-groups, and the second the more concrete limitation of mobility through the erection of checkpoints. From an organisational point of view, this indirect strategy was superior to direct coercion because it required substantially fewer resources. Even further, it spared organisations from much of the conflict and grievances that outright coercion would have created.
The Creation of In-Groups and Out-Groups
Ethnic polarisation has historical roots in Liberia. When the rebels first entered the country in 1989, they targeted two ethnic groups they viewed as supporters of the Doe regime: the Krahn and the Mandingo. Doe had persecuted the Gio and the Mano following unsuccessful coup attempts. In Lofa, Loma fought Mandingo for local dominance, and atrocities were committed by both sides during the early 1990s. Particular ethnic groups associating with particular military organisations strengthened the creation of in-groups and out-groups, making it difficult for individuals to stay neutral in the conflict.
These ethnic grievances had not vanished by the beginning of the Second Civil War (and still have not more than a decade after its end). While LURD in particular sought to present itself as a movement that fought only against Charles Taylor and welcomed all ethnic groups to join it, it was easy for all the factions to exploit existing ethnic grievances. Whereas the GoL could claim to be protecting the state against rebels, many LURD and MODEL fighters saw fighting as the only way to return home from exile. All three movements could present themselves as protectors of one or more ethnic group, which helped mobilisation against certain out-groups. Ethnic groups identifying themselves and being identified by others with certain factions enforced polarisation. As Maček argues, “war has a propensity to enforce both an antagonistic division between groups and homogeneity within groups” (Maček 2009: 208). Both divisions and homogeneity were enforced by tactical action. For instance, one side's systematic execution of prisoners belonging to certain ethnic groups likely led to the adoption of similar tactics by the opponent. This kind of indirect use of power contributed to making fighting a “choiceless decision” (Aretxaga 1997). As a result, it was easier for military organisations to control those who lacked viable alternatives for not falling in line.
Checkpoints
The second indirect way to instil discipline and to reduce desertion was to limit the mobility of fighters and thus keep them on the frontlines, where the conditions alone would force them to fight. Similar logic has been applied around the world. For instance, military barracks are typically located in remote places in order to remove soldiers from society –and thus make it easier to retain them. Another example is the Lord's Resistance Army, which “sought to limit escape opportunities by quickly moving the abductee as far as possible from home” (Blattman and Annan 2010: 141). Travelling through the bush in Liberia is difficult, which meant that a string of checkpoints – often called gates – erected on roads behind the frontlines had a significant effect on stopping desertion. As one LURD commander explained, “gate controls frontline.” 22 While it was possible to pass through one or two checkpoints by negotiating or by exploiting personal relationships, at some point a pass or a permit became necessary. Checkpoints, enemy attacks, hunger, and terrain all contributed to limiting the alternatives to not fighting. For instance, many LURD fighters in northern Lofa County claim that they were “captured” by LURD, but never forced to fight. One sample of 26 fighters from the same unit that I interviewed in May 2012 included only six who claimed to have joined willingly. None, however, stated that they were ever forced to fight. As even these respondents brought up, struggling with a difficult situation, many ultimately had little choice whether to join – or fight – or not. While the grim situation was at least partially the responsibility of LURD, it was difficult to hold the capturing commander personally responsible for the prevailing circumstances. In fact, associating with a military organisation often provided significant benefits for the recruits. Not only did it provide security, but it also increased the possibility of getting supplies such as food and medicine, and improved some recruits’ social status and future prospects. Refusing to follow orders would have negated these benefits and possibly worsened the situation even further. In other words, active coercion was not always required to recruit and retain fighters.
LURD task force commander, 19 November 2013, Bomi.
Coercion was focused on the controlled areas close to frontlines, whereas the coercive means available to military organisations behind these areas – in towns, villages, and refugee camps – were much more limited. The weak formalisation of military organisations and the relative calm on many fronts is emphasised by the fact that many combatants spent considerable amounts of time away from their units, behind the frontlines (for a Sierra Leonean parallel, see Peters 2011: 66). Some interviewees (mainly but not exclusively those with rank) simply chose to stay away from the front without facing any consequences. Runaways often faced the choice of being disarmed or being returned to the frontlines when caught.
A different kind of dichotomy existed between civilians and fighters despite their ethnicity. This kind of polarisation of available alternatives is evident in statements made about civilians such as “You're either with us or against us” and “One way or the other you'll participate.” 23 This kind of acceptance of military norms that differ significantly from civilian norms is best discussed separately.
LURD official, 16 November 2013, Monrovia.
Lukes’ Third Dimension of Power: Securing Compliance
Lukes’ third dimension is the most potent of his three categories, but is also typically the most difficult to achieve. In a similar manner as with the second category, the context of war arguably helped. As noted, associating oneself with a military organisation was often thought of as the sensible thing to do in difficult circumstances. Yet, closer association came with certain expectations, which in many cases led to dissemination of norms and adoption of a new identity – that of a fighter.
Disseminating Norms
The final means of coercion concerned the dissemination of norms throughout organisations. In established military organisations, norms are disseminated through training (Kellett 1982), a process enhanced by isolating individuals from competing norms (Goffman 1987). While training officers existed in all Liberian organisations and while the government forces established dedicated training bases, norms were arguably disseminated more by unit commanders and peers on the frontlines than by centralised training regimens. While many fighters possessed previous military training, wartime formal training in all three warring factions typically consisted of little more than the basics of handling a weapon, whereas the rest was learned on the front. Ideological training was especially lacking.
While it has been noted above that the AFL possessed the strongest norms in Liberian civil wars, even the identities of ordinary fighters with no previous military experience were affected by years of war. Many began to identify with their comrades to the point that they became an in-group. Furthermore, other factions, military units, and ethnic groups were commonly perceived to be less brave and less capable than one's own, which points to the formation of group identities and
LURD fighter, 12 November 2013, Voinjama.
Examples of negative consequences for breaking established norms come from the frontlines. Once in the front, it was considered a breach of solidarity for anyone to abandon comrades. Such attempts were typically met with hostility or even aggression. For example, purposefully injuring oneself in order to leave the front (typically, shooting oneself in the palm or leg) might be left untreated by one's fellow fighters. Some went even further and shot the other palm or leg. At worst, attempts to desert raised the question of whether the fighter attempting to flee had betrayed their comrades to the enemy through “conniving.” As elsewhere, such cases of potential “different intention” were typically met with even more hostility by suspicious comrades.
The existence of at least some shared identification and norms is also evident in the lack of condemnation concerning the treatment of those who refused to fight: they were often forced to carry heavy loads or exposed to long periods of direct sunlight. Some interviewees expressed sympathy towards commanders who used violence against disobedient subordinates. These officers were described as having been carried away by anger after repeated insubordination; the blame was thus put on the victims instead of the perpetrators. Such views were more common among former AFL soldiers, but were also held by other fighters. This kind of socialisation clearly included a different idea of what was just. Such shaping of beliefs and desires corresponds to Lukes’ third dimension of power, in that norms had been changed to the extent that discipline was upheld by peers, who saw in-group cohesion as a way to increase their chances of survival in a violent environment.
This normative change and the resulting discipline upheld by peers was arguably the most important instrument for upholding discipline in Liberia. Whereas the organisations struggled to instil discipline through direct means, peer pressure was clearly more successful – although not necessarily in ways that guaranteed that force could be utilised strategically, as widespread preying on civilians and disregarding orders illustrates. At times, such tactical behaviour went against strategic interests, especially to the extent that it increased ethnic polarisation.
This upholding of norms by peers was also not necessarily as brotherly or buddy-like as is typically portrayed in the existing literature. For instance, Marshall notes that a soldier risks his life because of “friendship, loyalty to responsibility, and the knowledge that he is a repository of the faith and confidence of others” (Marshall 1947: 161; see also Baynes 1967:183; Little 1964; Shils and Janowitz 1948: 284). This may be true to some extent in Liberia, but the prospect of being punished by peers after failing to meet expectations played a role as well. A to some extent comparable account comes from Eisenhart's personal experiences from the United States Marine Corps boot camp in the 1970s, where violence was used against norm-breakers (Eisenhart 1975). This finding calls for a re-evaluation of the positive role played by peers in military organisations.
Conclusions and Theoretical Implications
This article has investigated the use of coercive discipline epitomised in the notion of “No die, no rest” during the Second Liberian Civil War. It argues that Liberian military organisations faced concrete restrictions when they sought to rely on extreme coercion and especially executions to instil discipline. Much rested on the very weakness of these organisations: they failed to formalise themselves, and consequently to disseminate and uphold norms effectively. As a result, the Liberian military organisations could be compared to citizen-armies with recruits with little previous idea – let alone acceptance – of military norms. These norms thus remained contested and extreme coercion illegitimate. Executions in particular were recognised as potentially hazardous for those who sought to resort to them. Individual commanders and in particular peers became the source of most discipline, ostensibly in order to increase their own chances of survival. Because they often resorted to violence against norm breakers, the positive view of camaraderie of brothers-in-arms held in previous literature should be re-evaluated.
The fact that Liberian military organisations struggled to employ coercive discipline and as a result had to resort to more indirect means challenges the prevailing idea that weak organisations must invariably resort to force in order to instil discipline among their ranks. In fact, the investigation suggests that weak organisations can face collapse if they attempt to rely on extreme coercion. Employing coercion remains much easier for strong organisations, because they can disseminate and uphold norms much more effectively. Without shared norms, coercion remains illegitimate. Similarly, the Liberian military organisations had very limited power to use coercion in areas they did not directly control. In comparison, Finnish deserters during the Second World War could never truly escape the total war: they were often arrested if they returned home. Some were executed (Ylikangas 2009). Similar coercive apparatuses did not exist in Liberia. Further comparative in-depth studies would be helpful in developing the idea of the legitimacy and utility of coercive discipline in both weak and strong military organisations.