Hostname: page-component-78c5997874-m6dg7 Total loading time: 0 Render date: 2024-11-19T08:46:10.921Z Has data issue: false hasContentIssue false

Compellence by denial against armed groups: UN peacekeeping in Ituri, the Democratic Republic of the Congo

Published online by Cambridge University Press:  04 November 2024

Ken Ohnishi*
Affiliation:
Policy Studies Department, National Institute for Defense Studies, Tokyo, Japan
Rights & Permissions [Opens in a new window]

Abstract

Peace operations are often deployed to countries where conflicts are effectively ongoing, with mandates to protect civilians and provide security. With the authorisation to use force beyond self-defence, peacekeepers are expected to exert military pressure on armed groups to induce them to abandon violence and join peace processes. Such an approach falls in the domain of compellence. Given that peace operations struggle to adapt to this new expectation, it is imperative to find effective ways of compelling armed groups. While studies on compellence agree on the effectiveness of denial-type pressure, its specific forms are highly context-dependent. What is the most effective way to achieve denial against armed groups in the context of peace operations? There are four such approaches: attrition, stronghold neutralisation, decapitation, and counter-coercion negation. A case study of peacekeeping by the United Nation in Ituri reveals that stronghold neutralisation and counter-coercion negation were especially important for compelling rebels to disarm. As existing studies have not examined the effectiveness of specific forms of denial in peace operations, the finding contributes significantly to the literature on compellence and peace operations.

Type
Research Article
Copyright
© The Author(s), 2024. Published by Cambridge University Press on behalf of The British International Studies Association.

Introduction

The limitation of military capability is a widely acknowledged feature of peace operations, especially those conducted by the United Nations (UN). Nevertheless, the UN Security Council (SC) often deploys peace operations to countries where conflicts are effectively ongoing, with mandates to protect civilians and provide security. Most operations are authorised to use force beyond self-defence, under Chapter VII of the UN Charter. However, as can be seen from the fact that peacekeepers often fail to respond to violence and that insecurity continues after their deployment, peace operations struggle to adapt to this new expectation and have much room for improvement.Footnote 1

If peace operations are to remain relevant to contemporary conflict management, they need to learn how to deal with hostile spoilers who are willing to deploy violence to subvert peace processes. As demonstrated in many post–Cold War missions, mere talks are insufficient in countering spoilers. The importance of credible deterrence in contemporary peace operations is widely acknowledged in both literature and policy.Footnote 2 However, it is not sufficient, because deterrence is a strategy used to maintain the status quo. In a situation without peace to keep, peacekeepers’ tasks are instead to create peace and stability – that is, to change the status quo on the ground rather than to maintain it. With the authorisation to use force beyond self-defence, peacekeepers are expected to change the status quo through exertion of military pressure on spoilers to induce them to abandon violence and join peace processes. They need a different strategy for that purpose.

Such an approach falls in the domain of compellence. Compellence is a strategy which employs pressure, including the threat and/or actual use of force, to induce others to take actions as demanded. As a threat-based strategy, compellence is the sibling of deterrence but is more proactive, as it aims to change the status quo. Compellence is also different from simply defeating others and achieving one’s objectives forcibly. Compellers try to manipulate targets’ cost–benefit calculation and thereby make the targets choose, however grudgingly, to accept the compellers’ demands.Footnote 3

Despite the suitability of compellence for contemporary peace operations, the strategy’s utility within this context is underexplored. Some studies have applied the concept of compellence and insights from the field to peace operations. Of these, some are descriptive and do not engage with theoretical questions.Footnote 4 While others are explanatory, they are small in number and have not yet reached a consensus about when and how peacekeepers should employ compellence.Footnote 5 Therefore, additional studies are required to advance our knowledge in order to successfully employ compellence in peace operations.

One of the few consensuses in the literature of compellence is that compellence using a denial type of pressure is more likely to succeed than is that using a punishment type of pressure. The pressure used in compellence, similar to deterrence, can be largely divided into two categories: punishment and denial. Compellence by punishment threatens to increase the costs of rejecting the compeller’s demands. On the other hand, compellence by denial threatens to reduce the possibility of the target attaining their goals.Footnote 6 In existing studies, compellence by denial is understood as being more effective than compellence by punishment.

However, this insight is insufficient for guiding the application of compellence in peace operations; as denial seeks to obstruct opponents’ efforts to pursue their objectives, its specific forms are highly context-dependent.Footnote 7 Therefore, the following question remains: what is the most effective way to achieve denial against armed groups in the context of peace operations? Given the challenges peacekeepers are faced with, it is imperative to understand what kind of pressure is effective in compelling non-state armed groups to give up violence. This article addresses this gap.

To answer this question, this article recaps the utility of compellence in peace operations and explains four specific forms of denial that can be applied: attrition, stronghold neutralisation, decapitation, and counter-coercion negation. Then, to examine which of the four is effective, the article analyses a case in which each of them was applied – namely, the operation of the United Nations Organization Mission in the Democratic Republic of the Congo (MONUC) in Ituri. The case study reveals that stronghold neutralisation and counter-coercion negation were especially important for compelling Ituri rebels to disarm.

Although the findings’ external validity should be empirically examined in further studies, this article contributes to the fields of compellence and peace operations. Studies on compellence have largely focused on that between states, and its utility against non-state actors is underexplored, despite such actors’ prominence in the contemporary security environment.Footnote 8 The identification of specific forms of denial effective against non-state armed groups is, therefore, a valuable new insight for the field and promotes our understanding of the strategy’s utility.

This finding is also helpful in advancing research on peace operations. Through demonstrating how compellence by denial works, this study provides a possible causal mechanism for the field. Although many studies have examined whether peace operations are effective in keeping peace or reducing violence,Footnote 9 only a few adequately address causal mechanisms of peacekeeping, and their focus is on deterrence and non-coercive mechanisms.Footnote 10 However, as mentioned above, deterrence is inapplicable to situations in which countering ongoing violence is the focus, and many post–Cold War missions have demonstrated that mere talks or positive incentives are insufficient to counter spoilers. Hence, the exploration of how peacekeepers can achieve objectives through compellence – which is a suitable strategy for addressing spoilers’ ongoing violence – reveals an unacknowledged causal mechanism of contemporary peace operations.

In addition, this article sheds light on an unrecognised case of successful compellence in Ituri. Although MONUC is one of the peace operations that employed compellence most successfully, this fact has not been acknowledged in existing studies. This article illustrates how MONUC applied denial pressure on Ituri armed groups and compelled them to stop engaging in violence.

Finally, this study has practical implications as well. Given that peacekeepers have difficulty in dealing with spoilers, practitioners need insights about when and how peacekeepers can use proactive pressure successfully to shape spoilers’ behaviours. The identification of specific forms of denial that are effective against armed groups can be a help for practitioners in designing a strategy to cope with spoilers.

Compellence and peace operations

That compellence is a useful strategy for peace operations may be surprising if one’s perception of peace operations is that of traditional peacekeeping. Traditionally, the expected role of force for peacekeepers was passive: to deter violence and defend themselves when under attack. For example, Lise Morjé Howard emphasises that peacekeepers do not and cannot resort to compellence because of capacity limitation and the three principles of peacekeeping: consent of the parties, impartiality, and non-use of force except in self-defence and defence of the mandate.Footnote 11

However, both UN and non-UN peace operations have employed compellence. Even if the scope is limited to cases wherein peace operations demanded non-state armed groups to completely disarm or cease armed activities, compellence has been employed in countries such as Somalia, East Timor, Sierra Leone, the Democratic Republic of the Congo (DRC), Haiti, and Mali.Footnote 12 There are more cases if the scope is extended to compellence with limited demands.

Some of the above attempts ended in failure, indicative of compellence not being risk-free. If targets do not yield to initial threats and pressure, compellers need to choose between escalating pressure or giving up.Footnote 13 Moreover, compellence is a double-sided game; the targets also attempt to compel the compellers to withdraw the demands.Footnote 14 The compellers may end up using substantial force and suffering huge costs in an escalating military confrontation.

Nevertheless, it seems that UN peace operations cannot deny the need for a more robust approach. The experience of the 1990s demonstrates that going back to the traditional mode is not an option. It is agreed that a political solution is important; however, people suffering from violence need immediate physical protection on the ground.Footnote 15 Considering the limitation of UN operations, arguments for the division of labour with coalitions of the willing and regional organisations are persuasive.Footnote 16 However, if deployed concurrently, spoilers would not distinguish between types of interveners and may attack UN troops.Footnote 17 Moreover, non-UN actors tend to hand over their missions to the UN with haste due to the financial burden.Footnote 18 Security-related tasks given to UN missions have expanded, and there are precedents for robust actions taken by the UN peacekeepers. It is unlikely that those who are suffering will stop expecting the UN to launch forceful interventions. For example, African states have called for more robust missions to halt violence on their continent.Footnote 19

UN documents also acknowledge the necessity of proactiveness in the use of force. The Capstone Doctrine states, ‘By proactively using force in defense of their mandates, [robust] United Nations peacekeeping operations have succeeded in improving the security situation and creating an environment conducive to longer-term peacebuilding in the countries where they are deployed’.Footnote 20 According to the so-called HIPPO Report, which is a review report on peace operations, peacekeepers should be prepared to use force ‘ranging from containment via deterrence and coercion [i.e. compellence] to direct confrontation’ to defend their mandates, including civilian protection.Footnote 21 Another review report, the so-called Cruz Report, also urges peacekeepers to avoid ‘Chapter VI Syndrome’ and take proactive and even pre-emptive stances to realise better force protection in hostile operational environments.Footnote 22

Hence, it is necessary to improve our understanding of the role of compellence in peace operations so that peacekeepers can use it effectively when required. Of course, this is not the preferred option for peacekeepers, and not all peace operations employ compellence. However, as discussed above, it is a logically appropriate strategy when peacekeepers are expected to change the status quo, and there are precedents for doing so. Given the call for robustness, it can be said that at least there exists a necessity to explore how the proactive use of military pressure can be effective in peacekeeping. Insights from studies on compellence provide clues for this exploration.

Compellence by denial

Existing studies on compellence have focused on compellence between states, with only a small number of explanatory studies dealing with compellence in the context of peace operations. However, this does not render the literature irrelevant for exploring the successful use of compellence against armed groups. Armed groups in civil conflicts are strategic actors; they use violence to achieve political goals.Footnote 23 As long as they act strategically, their behaviour may be affected by manipulating their cost–benefit calculations. Although appropriate adjustment and empirical applicability checks are required, insights from the literature on compellence sheds light on its utility in this underexplored context.

Previous studies examine various factors which can affect the outcome of compellence, including the types of demands and pressure, sources of threats’ credibility, and available resources for protagonists.Footnote 24 Although a consensus on the conditions for the success of compellence has not been reached, the literature agrees on the importance of some factors.

Empirical studies comparing multiple cases have found that compellence by denial is more likely to succeed than that by punishment.Footnote 25 The two types of pressures work on different parts of the targets’ cost–benefit calculations. As Lawrence Freedman explains, punishment threatens to impose costs and, in its purest form, does not hinder the targets’ capability to pursue their objectives. Conversely, denial threatens to prevent targets from achieving their goals, and its execution diminishes the targets’ capability for pursuing the same.Footnote 26 The more denial pressure is applied, the fewer options are available to the targets, who are ultimately forcibly prevented from pursuing their intended course of action, although reaching this point means that the strategy has turned from compellence to brute force.Footnote 27

The type of pressure to be used is a central question in designing any compellence strategy, and this insight is useful for compellence in peace operations as well. Compellence in this context is likely to be compellence by denial. This is because non-state armed groups – the likely targets of compellence in peace operations – seldom have assets that can be targeted for punishment.Footnote 28 To apply the pressure of punishment, targets must possess valuable assets that are not used in war efforts, but the loss of which would be costly. For example, if a compeller wishes to induce one country to cease its ground invasion of another, threatening to destroy or destroying the target country’s civil infrastructure or naval vessels, which are not being used in the invasion, would constitute pressure by punishment. However, non-state armed groups usually do not possess such valuable assets. They tend to be guerrilla forces composed of foot soldiers, with little to no heavy equipment. While they may have communal bases that can be targeted, threatening the civil infrastructure of such communities would contradict the purpose of peace operations. Therefore, compellence by peacekeepers is highly likely to rely on the pressure of denial that directly obstructs targets’ activities.

How to realise denial in peace operations

The specific ways in which compellers can achieve denial differ according to context. As denial aims to thwart opponents’ efforts to achieve their goals, various forms can be effective, depending on the opponents’ objectives and strategies.Footnote 29 For example, Robert Pape focuses on the use of air power and raises close air support, operational interdiction, and strategic interdiction as specific forms of denial.Footnote 30 However, this categorisation is only applicable to the cases in which compellers use air power as a main tool of pressure. In a recent study, Samuel Zilincik and Tim Sweijs introduced another categorisation composed of capability elimination, operational paralysis, tactical degradation, and strategic effect reduction.Footnote 31 However, as they conceive denial broadly, going beyond compellence and deterrence, their categories are too general and abstract to be a specific guide for compellence in peace operations. Regarding non-state actors, some studies on counterterrorism explore how denial-type pressure can be applied against terrorists, though the studies often do not distinguish compellence from deterrence.Footnote 32 While these are valuable contributions to the field, no study has focused on the use of denial in peace operations.

Therefore, to increase the success of compellence in peace operations, the most effective way to achieve denial in this context needs to be explored. Existing studies have not dealt with this issue, but it is possible to devise the forms of denial employable in peace operations based on related insights from the literature. There exist four such possible forms of denial.Footnote 33

The first is attrition. Attrition aims to wear down the opponents’ fighting force by killing and wounding combatants. Since manpower is indispensable to any insurgency,Footnote 34 armed groups lose capabilities as their casualties mount, which may convince them that winning is impossible.Footnote 35 Existing studies on counter-insurgency disagree about the utility of attrition. For example, one study claims that attrition through governments’ intelligence-led operations steered insurgencies in Uruguay and Northern Ireland to their decay.Footnote 36 Meanwhile, another study cast doubts on the effectiveness of attrition, as Israel could not stop Hamas in the Second Intifada, although Israel had killed or arrested a substantial number of Hamas members.Footnote 37 Although peacekeepers are not expected to conduct attrition for its own sake, they can kill or wound a large number of combatants as a result of their operations. For example, UN peacekeepers in Sierra Leone are thought to have killed 34 rebels and wounded 150 during an operation to rescue besieged peacekeepers.Footnote 38 MONUC also killed over 150 rebels in an attempt to halt their offensive towards the regional capital of North Kivu, a province in the DRC, after the mission’s warning of a forceful response was ignored.Footnote 39 At the least, attrition can serve as an effective pressure mechanism against non-state armed groups and deserves examination.

The second form of denial is stronghold neutralisation, which seeks to capture or destroy armed groups’ strongholds. Studies on insurgency and coercion against non-state groups point out the importance of strongholds or safe havens for armed groups.Footnote 40 Armed groups equip their forces, train them, plan operations, and launch attacks from their strongholds. Territorial control brings revenue from exploiting natural resources or taxing residents.Footnote 41 Therefore, neutralising strongholds would deprive armed groups of their bases, shrink their spheres of influence, and degrade their operational capabilities. For example, Iraqi insurgency dwindled after 2007 as counter-insurgents, composed of reinforced US troops and local tribal militias, cleared out rebels’ strongholds.Footnote 42 This approach can be effective in peace operations as well.

The third form is decapitation. This refers to the elimination of armed groups’ leaders by killing or capturing them. The groups’ activities can be disorganised and obstructed by eliminating leaders and cadres in command.Footnote 43 Also, the new leaders may be more susceptible to pressure and easier to persuade of the futility of their armed struggles than their predecessors. An example of effective decapitation is the Shining Path insurgency in Peru. The arrest of its leader by the Peruvian government in 1992 was a critical blow to the group, which heavily depended on its leader, and its violence sharply declined thereafter.Footnote 44 While the literature is divided on the utility of decapitation in general,Footnote 45 it can also work in peace operations, and the possibility deserves examination. Peace operations may entail directly killing or capturing leading members of an armed group, as when UN peacekeepers in the Central African Republic captured multiple armed group leaders during their efforts to stabilise the situation,Footnote 46 or may involve exploiting opportunities when armed group leaders are killed or captured by other actors.

The last approach to denial is counter-coercion negation. Even when compellers have inflicted serious damage on armed groups, they may keep up the resistance if they still have countermeasures and envision striking back. Therefore, containing armed groups’ counter-coercive attempts can be an effective approach in depriving them of the hope of prevailing in the confrontation. Daniel Byman and Matthew Waxman identified three specific counter-coercion strategies that armed groups can use against peacekeepers.Footnote 47 The first is ‘civilian suffering-based strategies’, in which armed groups disturb peacekeepers’ compellence by putting civilians in danger. For example, the use of human shields or the launch of retaliatory attacks on civilians by spoilers may complicate the planning and execution of compellers’ proactive operations or undermine support for them. The second is ‘coalition-fracturing strategies’, in which armed groups exploit rifts among troop contributors. As a multinational operation, any peacekeeping mission is composed of troops from countries with different interests and priorities. Armed groups can exaggerate and exploit such differences by taking fake amicable stances or attacking only certain countries, thereby undermining the unity of the mission’s efforts. The third is ‘casualty-generating strategies’, in which armed groups inflict casualties on peacekeepers and thereby induce them to halt their compellence. As the compellers seldom have any vital interests in the conflicts in which they are intervening, suffering casualties may break their will to continue the intervention. Some studies acknowledge the importance of containing opponents’ counter-coercion for successful compellence.Footnote 48 This approach can also be important in the context of peace operations.

The four approaches outlined above are intended to degrade and disturb armed groups’ capability to pursue their objectives and thus can constitute denial-type pressure. While each is effective theoretically, their actual effectiveness should be empirically examined. Therefore, a case study is presented in this article.

Case selection

As this study aims to examine the mechanism by which denial pressure works in compellence during peace operations, a detailed analysis of a successful case is required. Successful compellence in peace operations results in target armed groups accepting and implementing demands, such as ceasing violence or disarming. Success can be outright, partial, or temporary depending on the degree of acceptance. Even when compellence ends in success, it does not necessarily mean that long-term stability has been restored. Although the containment of violence via compellence should play a crucial role in stabilisation, such stability also requires many other peace-building efforts. In addition, peacekeepers may employ compellence against only some armed groups operating in a country. In a large country such as the DRC, multiple compellent campaigns can be employed independently against different armed groups in different theatres with various outcomes. Therefore, there can be a successful case of compellence even in a country which remains unstable overall.

As mentioned above, peace operations have demanded that armed groups stop violence or disarm and have employed compellence in Somalia, East Timor, Sierra Leone, the DRC, Haiti, and Mali. Among them, East Timor, Sierra Leone, and the DRC have witnessed successful compellence. Non-UN actors played large roles in compellence in East Timor and Sierra Leone, while the UN led most compellence attempts in the DRC. Both MONUC and the succeeding United Nations Organization Stabilization Mission in the Democratic Republic of the Congo (MONUSCO) employed compellence during more than two decades of UN involvement in the country. The outcomes vary from a success in Ituri to a partial success and failures in Kivus.Footnote 49

MONUC’s operation in the Ituri region of the DRC is one of the most successful cases of compellence employed by UN peace operations in which all four possible approaches to denial were applied. Moreover, the compellence was protracted and can be divided into three phases with different outcomes, as explained below. This makes it possible to trace how the application of each type of denial pressure changed during the course of events and how the changes affected the outcome of the case. Therefore, it provides an opportunity to examine the extent of each approach’s contribution to the success of compellence. Such within-case temporal comparison has an advantage in that it is possible to analyse the effect of changes in independent variables – that is, the use of each type of denial pressure – while keeping other contextual variables basically constant, which is difficult if one compares different cases.Footnote 50 Admittedly, the method has a disadvantage in that the findings are derived from just one context, and their external applicability requires additional empirical examination. However, as a first cut at the utility of specific forms of denial in compellence in peace operations, a within-case examination is a legitimate and valuable approach to use in this study.

This study relied on public UN documents, media reports, and published works for its data. It is ideal but impossible to have direct access to the perceptions of armed group leaders; therefore, the best possible approach is to use diverse information from the UN and third parties. The UN Secretary-General regularly submits reports to the SC on UN peacekeeping operations, which provide authoritative information on what those missions do on the ground. The author used Nexis and Factiva to seek relevant news reports for this study. He reviewed all English-language news reports containing the word ‘Ituri’ issued between 2003 and 2008 by Africa News, AllAfrica, the Associated Press, BBC Monitoring Africa, and Reuters News. These press organisations not only cover MONUC’s press conferences but also carry reports from local media and correspondents and thus are a valuable source of information. Published works to which the author referred include the memoir of a UN force commander and think-tank reports based on fieldwork. Information from these sources makes it possible to reconstruct what happened in Ituri and to examine the effectiveness of each denial approach in compelling armed groups.

MONUC

Background

When the second Congo war ended in 2002, an independent conflict continued in the Ituri region, which had been under Ugandan control since its intervention in the war. Violence between two ethnic groups, the Hema and Lendu, flared up over control of land in 1999, and multiple armed groups, each supported by external actors, fought. The main armed group of the Hema, the Union of Congolese Patriots (UPC), enjoyed Ugandan support initially and later turned to Rwanda. Two splinter groups from the UPC, the Party for Unity and Safeguarding of the Integrity of the Congo (PUSIC) and the Armed Forces of the Congolese People (FAPC) received support from Uganda. The Lendu formed the Ituri Patriotic Resistance Force (FRPI) supported by the DRC government, and the Front of Integrationist Nationalists (FNI), supported by Uganda.Footnote 51

Violence in Ituri escalated and developed into a humanitarian crisis by 2003. The estimated cumulative death toll reached over 60,000 by May 2003, along with 500,000 to 600,000 internally displaced people.Footnote 52 Uganda could not control the situation and withdrew due to international pressure.

In response, the international community moved to military interventions. MONUC had been deployed to the DRC since 1999, but the UN peacekeepers were not prepared to confront violent spoilers. Therefore, the UNSC deployed the Interim Emergency Multinational Force (IEMF) – a French-led European Union operation – to the regional capital Bunia in June 2003, which quickly restored the security of the city.Footnote 53

As the IEMF was an interim measure focusing on Bunia, the UNSC decided to expand and enhance MONUC to stabilise the entire Ituri region. A brigade-sized force was deployed to Ituri, with various security-related mandates and the authorisation to ‘use all necessary means’ under Chapter VII of the UN Charter.Footnote 54

MONUC’s compellence in Ituri

The enhanced MONUC demanded that Ituri armed groups disarm and exerted military pressure in cooperation with the DRC government. Following the extensive use of force and multiple rounds of disarmament, demobilisation, and reintegration (DDR) programmes, most of the Ituri armed groups chose to disarm. However, some rebels did not accept disarmament, and, therefore, MONUC’s compellence was not a complete success. Nevertheless, it can be evaluated as rather successful, as most armed groups did disarm, and the security situation in Ituri improved greatly.Footnote 55

The compellence in Ituri was protracted and can be divided into phases based on the three DDR programmes. Phase one covered the period between September 2003 and June 2005 – from the beginning of the enhanced MONUC operation until the end of the first DDR programme. MONUC first declared Bunia, and then the entire Ituri region, as weapon-free zonesFootnote 56 and repeatedly issued warnings that the rebels must either disarm voluntarily or face forcible disarmament.Footnote 57 It engaged with the rebels frequently and inflicted many tactical defeats. MONUC also employed an ultimatum. After suffering casualties in combat in February 2005, MONUC’s Special Representative of the Secretary-General (SRSG), William Swing, demanded that the rebels lay down arms by the end of March 2005.Footnote 58 Once the deadline passed, MONUC intensified proactive military operations. These pressures were accompanied by the carrot of the DDR programme. In May 2004, the DRC government and the rebels agreed on the latter’s disarmament and integration into the security forces.Footnote 59 Six rebel leaders were appointed as brigadier generals,Footnote 60 and combatants who chose to return to civilian life were provided with a kit and a $50 allowance.Footnote 61 The first DDR programme disarmed around 16,000 combatants;Footnote 62 however, hardliners declined to disarm while keeping a substantial part of Ituri under their control.

Phase two spanned from July 2005 to October 2006 following the first DDR programme until the end of the second DDR programme. The remaining rebels had regrouped as the Congolese Revolutionary Movement (MRC), while the FNI and the FRPI continued to operate independently. MONUC, in cooperation with the reorganised and retrained national army, the Armed Forces of the Democratic Republic of the Congo (FARDC), repeatedly launched cordon-and-search operations against the rebels. The compellers employed ultimatums in this phase as well and urged the rebels to either voluntarily disarm by the deadlines or face forcible disarmament.Footnote 63 The final ultimatum included an offer of amnesty, and those disarmed also received a kit and a $110 allowance. Additionally, the DRC government and the armed groups agreed to assign rebel leaders as colonels of the FARDC.Footnote 64 The second DDR programme is said as having disarmed approximately 6,700 combatants, but the figure appears inflated, and the three armed groups were not dissolved.Footnote 65

Phase three spanned from November 2006 to November 2007 from after the second DDR programme until the end of the compellence operation. Soon after the election of President Joseph Kabila in November 2006, the government and the three remaining armed groups concluded agreements, which reaffirmed the rebels’ disarmament, the offer of amnesty, and the recognition of the rebels’ ranks in the FARDC.Footnote 66 The government later reassured the rebel leaders that they would not be prosecuted.Footnote 67 However, despite the agreements, the rebels were reluctant to begin disarmament and tried to renegotiate the modus of their integration into the FARDC to maintain their influence in Ituri.Footnote 68 Meanwhile, responding to the clash between the FNI and the FARDC in late December 2006, the FARDC and MONUC conducted military operations against the FNI’s stronghold.Footnote 69 The compellers’ last offensive against the FNI demonstrated that the rebels had to accept the offer or face military consequences. Thus, the three remaining armed groups joined the third DDR programme, and approximately 1,800 rebels were disarmed. The leaders of the three remaining groups also surrendered and left Ituri to join the national army.Footnote 70 Not every rebel was disarmed, and part of the FNI and the FRPI continued their activities, while the MRC was dissolved. However, most of the rebels were disarmed, and violence in Ituri greatly decreased.

Summing up, MONUC’s compellence in Ituri was successful only after the repeated application of pressure to target rebels. Negative and positive inducements in the first phase compelled approximately 16,000 less motivated rebels to disarm but were not sufficient in compelling the hardliners, and substantial violence continued. Therefore, compellence at this phase can be considered unsuccessful, or partially successful. In the second phase, the compellers continued to apply pressure, combined with positive inducements, but the regrouped hardliners disarmed only partially. Thus, compellence remained unsuccessful at this stage as well. Finally, additional pressure and carrots in the third phase induced the leaders and a substantial number of the remaining rebels to disarm, with the compellence finally turning successful.

MONUC, in cooperation with the DRC government, employed all four types of denial pressure in compelling the Ituri rebels to disarm. Thus, the question is, which of the four pressures were effective in this effort?

Attrition

The first approach to denial is attrition. Available data indicate that peacekeepers inflicted rather large casualties on armed groups in the context of peace operations, but a detailed examination suggests that attrition was not effective in this case.

Attrition was not realised in the first phase of compellence. According to news reports, the compellers killed approximately 130 rebels in phase oneFootnote 71 and possibly 100 more UPC combatants in a battle in Nyamamba.Footnote 72 These casualty figures are not small in the context of peace operations, but the attrition rate was quite low, as there were far more rebels. The exact number of Ituri rebels is uncertain, and the examination must rely on estimation, but there existed five armed groups with a substantial number of combatants before the first DDR. They were: the UPC with an estimated 5,000 to 9,000; the FNI with around 4,000 to 8,700; the FRPI with 4,000 to 9,000; the FAPC with 2,000 to 7,000; and the PUSIC with some 2,000 to 5,600 fighters.Footnote 73 As the size of each armed group was approximately several thousand, the reduction of their fighting power by 200 or so could not be considered a serious damage to them.

The compellers inflicted heavy casualties on the targets in phase two. The sum of media-reported rebel casualties indicates that over 530 combatants were killed and 100 wounded,Footnote 74 though the reported figures may be inaccurate. The size of target armed groups in phase two was much smaller than the previous phase, because approximately 16,000 rebels accepted disarmament in the first DDR programme. The MRC, formed by the remaining rebels following the first DDR programme, seemed to have 1,000 fighters at first, which later increased to approximately 3,000.Footnote 75 The FNI and FRPI also remained active and had about 3,000 and 500 combatants, respectively.Footnote 76 Therefore, there were approximately 4,500 to 6,500 combatants in total. This means that the compellers possibly neutralised 10 per cent or more of armed group members in combat, and if so, attrition can be considered applied in this phase. Although phase three did not involve additional attrition, the effect of attrition in the second phase may have been continuous and could have induced the rebels to disarm later.

However, a more detailed analysis casts doubt on the attrition’s effectiveness. Available media reports indicates that each of the three remaining armed groups seem to have suffered a more or less similar number of casualties in phase two: the MRC’s casualties seem to have been approximately 145 combatants, the FNI’s, 210, and the FRPI’s, 140.Footnote 77 Further, some media reports do not specify which group the killed or wounded rebels belonged to, and such unspecified casualties stood at around 135. Therefore, the above figures would fluctuate if those unspecified casualties are added according to their true belongings, but still each group’s casualties would not differ greatly from one another.

This means that the FRPI, the smallest among the three groups, had the highest casualty ratio and, therefore, should have been most receptive to disarmament if attrition was effective. Conversely, it was the most reluctant to disarm. The FRPI continued to attack the FARDC during and after the second DDR programme and remained active even after the third DDR, as it ultimately only partially disarmed.Footnote 78

Moreover, the rebels could recruit new members to make up for their losses. The rebel leaders actively boosted rank-and-file membership, because the larger the group, the higher the rank offered to the leader during integration into the FARDC.Footnote 79 Ex-combatants who just disarmed were good targets for recruitment. Many ex-combatants returned to armed groups due to threats and also because support for disarmed combatants was too small and slow.Footnote 80

Therefore, attrition is unlikely to have been a major mechanism in the success of this case. Although the compellers possibly neutralised over 10 per cent of the rebels in the phase two combats, the armed group with the highest attrition rate was most reluctant to disarm, and the rebels could recruit new members. Thus, other approaches to denial must have been more effective in the Ituri case.

Stronghold neutralisation

The second approach to denial is stronghold neutralisation. This type of pressure was frequently used in Ituri and seems to have played a critical role in compelling the armed groups to disarm.

MONUC intensively pursued stronghold neutralisation during the first phase of compellence. MONUC destroyed or captured at least 15 UPC camps around Bunia, including its brigade headquarters, 13 FNI camps around and north of Bunia including its headquarters in Loga, three FAPC camps in Mahagi and Ariwara in northern Ituri, and an FRPI camp south of Bunia.Footnote 81

The compellers continued their offensives against the targets’ strongholds in phase two. The MONUC-supported FARDC captured Mongbwalu, the FNI’s main stronghold, in October 2005.Footnote 82 In November 2005, the compellers also increased pressure on the MRC through a combined operation in Irumu, where the MRC was concentrated. Some 300 MRC members surrendered following this operation.Footnote 83 Between March and May 2006, the compellers conducted similar operations against the FNI and the FRPI in Djugu and Fataki.Footnote 84 In May 2006, the compellers also took Tchei in Irumu from the FNI and the FRPI, but the FRPI recaptured the town the following month.Footnote 85 Nevertheless, the compellers’ offensive against Tchei enhanced the effect of sensitisation towards disarmament, and approximately 3,000 combatants chose to disarm in the aftermath of the operation.Footnote 86

In phase three, the compellers’ attempt at stronghold neutralisation played a critical role in inducing the rebels to disarm. Despite agreeing to disarm, the FNI kept rejecting its implementation and sporadically clashed with the FARDC from December 2006.Footnote 87 In response, the FARDC and supporting MONUC moved to surround and cut off the supply lines of the FNI. In the process, the FARDC engaged with the FNI and flushed them out of the villages. The operation seriously threatened FNI’s stronghold, which finally induced the FNI to disarm.Footnote 88

Thus, stronghold neutralisation seems to have been effective, although it remained partial, as the compellers’ offensives could not neutralise all rebel strongholds.Footnote 89 Despite this, DDR participants increased after the compellers’ attacks on the rebel strongholds, and pressure on the FNI’s stronghold in phase three was the final push that induced the group to ultimately disarm. Therefore, it can be said that stronghold neutralisation was effective and necessary for the success in Ituri.

Decapitation

The third approach to denial is decapitation. While used extensively by the compellers, it seems to have been ineffective in this case.

Decapitation was mainly employed in phase one. In a cordon-and-search operation in Bunia on 15 September 2003, MONUC arrested approximately 70 UPC members, including Floribert Kisembo, the UPC’s Chief of Staff, though he was soon released on the instruction of MONUC headquarters.Footnote 90 The UPC leader Thomas Lubanga and FNI leader Floribert Ndjabu went to the capital, Kinshasa, in August 2003 to negotiate with the DRC government, which kept them under house arrest. MONUC also arrested Mathieu Ngudjolo, an FNI senior commander, and transferred him to Kinshasa.Footnote 91 Furthermore, MONUC rearrested the UPC’s Kisembo in June 2004 along with the FNI’s acting president Pitchou Iribi.Footnote 92

The compellers intensified decapitation when the FNI killed nine peacekeepers in late February 2005. In response to the attack, MONUC and the DRC government arrested several rebel leaders in March and April 2005. Those arrested included Lubanga and Ndjabu; Yves Kahwa, the PUSIC leader; Germain Katanga, the FRPI leader; and other cadres of the UPC and the FNI.Footnote 93 Thereafter, the FARDC arrested a senior MRC member, Innocent Kaina, in a combat in phase two,Footnote 94 but no other major arrest was made.

These arrests of rebel leaders fragmented the armed groups but did not stop their activities.Footnote 95 New leaders took over from their arrested predecessors and sustained their organisations. Furthermore, an arrest did not mean the end of a rebel life; the FNI’s Ngudjolo returned to Ituri after being released from prison and became the leader of the MRC.Footnote 96 Therefore, decapitation does not seem to have been a pressure that compelled Ituri rebels to disarm.

Counter-coercion negation

The final approach to denial is counter-coercion negation. As introduced earlier, armed groups can use three specific counter-coercion strategies: ‘civilian suffering-based strategies’, ‘coalition-fracturing strategies’, and ‘casualty-generating strategies’. Among the three, the Ituri armed groups tried to counter compellence by risking civilians and inflicting casualties on international troops. However, the compellers could effectively contain the countering efforts, and this seems to have been necessary for the success of compellence.

Ituri rebels often used human shields,Footnote 97 but the compellers could largely cope with this tactic. MONUC was cautious not to cause collateral damage, such as its rule that gunships fire warning shots before launching an attack.Footnote 98 Although it was impossible to completely avoid some accidental civilian casualties,Footnote 99 human shields did not hinder MONUC from conducting proactive military operations.

The rebels also attacked MONUC troops on numerous occasions, as the peacekeepers were an obstacle to rebel activity. The rebels had incentive to keep deploying violence as leverage to realise their greed. As rebel leaders in the second Congo war were incorporated into the DRC government under a power-sharing agreement, the Ituri rebel leaders perceived violence and territorial control as bargaining powers, which would lead to their entry into the circle of national power holders.Footnote 100 The Ituri rebel leaders also sought to maintain their influence in Ituri by placing conditions on the disarmament. Specifically, they demanded the establishment of the Ituri military region under their leadership, so that the rebel forces would be incorporated into the national army without dissolving their command structure. In pursuance of these objectives, the Ituri armed groups were willing to continue violence, while buying time by signing agreements and reneging them repeatedly.Footnote 101

MONUC understood that the rebels were attempting to force peacekeepers to retreat by inflicting casualties.Footnote 102 However, MONUC suffered few casualties, given the intensity of the operation. Summing up the media-reported casualties, MONUC in Ituri suffered 12 killed, 31 wounded, and 1 detained in phase one; 1 killed, 8 wounded, and 7 detained in phase two; and no casualties in phase three.Footnote 103 One reason for this low number was MONUC’s good force protection, facilitated by armoured vehicles and helicopters.Footnote 104 In particular, MONUC fully utilised helicopters for medevacs and air cover when troops were attacked or conducted offensive operations.Footnote 105 The other reason was reliance on the FARDC in the dangerous close combat operations. After the deployment of the FARDC, MONUC shifted its role to the provision of air, fire, and rear support for FARDC operations, thereby reducing the risk of casualties.Footnote 106

However, there was an instance in which even low casualties hindered MONUC’s compellence. In May 2006, the FNI attacked MONUC’s Nepalese troops conducting a cordon-and-search operation and captured seven peacekeepers, killing one and wounding three.Footnote 107 The hostage situation forced MONUC to halt offensives against armed groups while negotiating the release of the peacekeepers.Footnote 108 Although the hostages were released by July,Footnote 109 MONUC did not resume proactive operations as negotiations for peace between the DRC government and armed groups developed.

Considering the above, counter-coercion negation was necessary for the success of compellence. As the rebels intended to sell themselves at a higher price, they had incentive to continue the violence as long as possible, to secure a better deal. Hence, the rebels tried to drive out the peace-enforcing international troops; but MONUC was able to contain the rebels’ countermeasures. However, as seen in the fact that the FNI’s hostage taking of peacekeepers forced MONUC to cease offensives, the situation was precarious. If MONUC had suffered more casualties or caused more collateral damage, drawing criticism, it might have abandoned compellence altogether, as the rebels envisaged. Therefore, it seems that counter-coercion negation was necessary for maintaining compellent pressure and denying the Ituri rebels’ attempts to continue violence for bargaining power.

Impacts of other factors

While this study focuses on denial pressure, compellence is a complex phenomenon and other factors may well affect its outcome. As stated earlier, the use of within-case temporal comparison helps to ensure that many of these factors are held constant. However, even constant or only slightly varying factors might have played important roles in inducing rebels to disarm. Two such factors deserve examination in assessing the impact of denial pressure on the success of compellence in Ituri.

One is the use of ultimatums. Existing accounts of the Ituri conflict emphasise the role of the ultimatum issued by SRSG Swing in March 2005 in inducing the rebels to disarm.Footnote 110 The disarmament proceeded apace after Swing’s ultimatum, and some rebels mentioned fear of the consequence of ignoring the deadline as the reason they joined the DDR programme.Footnote 111 Moreover, ultimatums were also used in phase two, and the FNI and the MRC made deals with the government around the deadline.Footnote 112 These indicate that the use of ultimatums facilitated disarmament in Ituri.

Nevertheless, the effect of ultimatums was limited. Swing’s ultimatum could not convince the hardliners to disarm, and the armed groups, including the newly formed MRC, continued their activities for more than two years thereafter. Also, the armed groups’ agreements to disarm after the ultimatum in phase two seem to have been delaying tactics, as they tried to renegotiate the terms repeatedly. As a result, the compellers had to keep increasing denial pressure until the rebels committed to disarmament.

The other factor is the use of positive inducements. Past studies of compellence largely regard the use of carrots as one of the factors favouring its success,Footnote 113 and this was effective in Ituri as well. As mentioned earlier, the compellers provided the rebels with amnesty, higher-ranking posts in the FARDC, and reintegration support in inducing them to disarm. In particular, the positive inducements offered in phase three were critical for the success. While the inducements were not new, their repeated promise was important, as the rebels were sceptical as to whether the promised carrots would be provided after the presidential election.Footnote 114

Although carrots were effective, the sticks played a vital role. As discussed earlier, the Ituri rebels were greedy, trying to sell themselves as dearly as possible, which led to the repetitive making and breaking of deals with the government. If the compellers had solely relied on carrots, the rebels would have raised their conditions for disarmament endlessly, including the establishment of the Ituri military region led by the rebel leaders, which was repeatedly rejected by the government. Military pressure in combination with positive inducements was necessary for inducing the rebels to disarm.

Therefore, though other factors facilitated disarmament, the application of denial pressure was indispensable to the success of compellence in Ituri. The use of ultimatums or positive inducements was insufficient to induce the rebels to disarm. Only with military pressure applied could MONUC compel the rebels to disarm.

Conclusion

Although not widely acknowledged, compellence is an option for contemporary peace operations. There are precedents, and its use will continue as long as peacekeepers are expected to protect civilians and stabilise violence-ridden countries. Therefore, as it is a risky strategy, accumulating insights on how to employ the strategy successfully in the context of peace operations is important.

An important factor to consider in designing a compellence strategy is what type of pressure to use. It is agreed that compellence by denial is more effective than that by punishment, but how to realise denial differs according to context, and, thus, a context-specific analysis is required.

Therefore, this study has attempted to answer the following question: what is the most effective way to achieve denial against armed groups in peace operations? Based on insights from existing studies, four specific approaches to denial, applicable to this context can be devised: attrition, stronghold neutralisation, decapitation, and counter-coercion negation. The effectiveness of each approach has been examined through MONUC’s operation in Ituri, in which all of the four approaches were employed.

The case study indicates that two of the approaches to denial were effective in Ituri. The first was stronghold neutralisation. MONUC, in cooperation with the FARDC, extensively cleared or captured rebel strongholds. The proactive operations led to the increase of rebel participants in the DDR programmes, and pressure on the FNI’s stronghold induced the recalcitrant group to disarm. The second was counter-coercion negation. MONUC could contain the rebel’s attempts to exploit civilian casualties and inflict casualties on peacekeepers. Failure to do so could have led to the abandonment of compellence, with the rebels endlessly increasing conditions for accepting disarmament and continuing to use violence as a bargaining chip.

However, attrition and decapitation were not effective in this case. The compellers possibly killed or wounded over 10 per cent of rebel combatants, but the effect is in doubt, as the FRPI, which suffered the highest attrition rate, was the most reluctant to disarm. The compellers also arrested numerous rebel leaders and cadres, but this did not stop the rebels, as other members just assumed the positions of those arrested. While the direct effects of these two approaches are in doubt, there remains a logical possibility that they facilitated stronghold neutralisation and counter-coercion negation and thus indirectly contributed to the success of compellence. However, the course of events casts doubts on such a possibility as well. As MONUC had been able to apply and realise stronghold neutralisation and counter-coercion negation from phase one, attrition, which MONUC only realised in phase two, was not necessary in employing the two effective approaches. Decapitation was extensively used in phase one but not in the following phases, with the exception of the arrest of a senior MRC member. This means that decapitation was hardly employed against the MRC, which was formed in phase two. Nevertheless, MONUC could apply stronghold neutralisation and counter-coercion negation against the MRC; thus, decapitation was also unnecessary in employing these two effective approaches.

The finding that stronghold neutralisation and counter-coercion negation are effective against armed groups in the context of peace operations adds new insight to the literature. Existing studies of compellence have largely focused on compellence between states, and how compellers can apply the denial type of pressure against non-state armed groups is underexplored. Given the prominence of non-state armed groups in the current strategic environment and the context-dependent nature of denial, the identification of effective approaches to denial against such actors by a context-specific empirical study is a meaningful contribution to the research field of compellence. Turning to peace operations, this field lacks insight about how peacekeepers can achieve challenging objectives of creating stability in a no-peace-to-keep situation. The causal mechanisms of peacekeeping are especially understudied, and few existing studies solely focus on deterrence and non-coercive mechanisms that are inappropriate for countering ongoing violence. This study demonstrates how peacekeepers can effectively apply compellence by denial in inducing rebels to give up violence, using an as-yet-unrecognised successful case of Ituri as evidence. This sheds new light on one of peacekeeping’s causal mechanisms, which is often ignored but is suitable for security-related objectives requiring a change in the status quo. Therefore, the above finding addresses remaining gaps in the two fields of study and advances our understanding of important issues in contemporary security affairs.

Moreover, the finding has practical implications. Today’s peacekeepers struggle to deal with spoilers, and the findings provide valuable clues about the effective use of force during peace operations. If peacekeepers are required to compel spoilers to give up violence, the findings suggest that peacekeepers should neutralise armed groups’ strongholds while containing their counter-efforts. Stronghold neutralisation requires that peacekeepers have the capability to launch expeditionary military operations to clear and capture territories. Meanwhile, counter-coercion negation requires the careful use of force to avoid civilian casualties that can be exploited by spoilers, as well as good force protection to avoid peacekeeper casualties. Although self-preservation is a natural requirement and its importance for peacekeepers is obvious, this research indicates that it is also important for pursuing more proactive purposes.

While this is a single case study, the prolonged compellence in Ituri can be divided into three phases with different outcomes, and within-case analysis made it possible to discern the utility of different approaches to denial. This study’s findings may well apply to other cases of compellence in peace operations. Regarding attrition, it seems difficult for peacekeepers to inflict large enough casualties on armed groups. Peacekeepers’ rules of engagement tend to be restrictive, as MONUC’s attack helicopters had to fire warning shots before actual attacks, allowing the rebels to escape or take cover. Instead of killing and wounding rebels, proactive operations by peacekeepers focus on clearing and holding areas, and thus stronghold neutralisation would be an option better suited to them. The effect of decapitation may vary depending on how much an armed group relies on its leader. However, there are resilient groups that can survive decapitations, as in the case of the Ituri rebel groups, so it may not be effective in other cases either. Finally, it is easy to imagine the importance of counter-coercion negation when we consider famous cases of its absence, such as Somalia. Its importance may well be confirmed in other cases too.

Of course, the finding’s external validity should be empirically examined. Given the difficulty peacekeepers face, the utility of specific forms of denial deserves further exploration. Further empirical studies are required to promote our understanding of this issue.

Acknowledgements

I am grateful to Yasuhito Fukushima, Akira Jingushi, Hiroto Sawada, and two anonymous reviewers for their helpful comments on earlier drafts. Any remaining errors or mistakes are my own.

Competing interests

The author declares none.

Ken Ohnishi is Senior Fellow at the National Institute for Defense Studies (NIDS), Ministry of Defense, Japan. His main fields of research are compellence/coercive diplomacy and peace operations. Since joining NIDS in 2010, he has authored articles analysing the use of force by peacekeepers and taught classes on peace operations. He received his PhD in Politics and International Studies from the University of Leeds.

References

1 Alex Bellamy and Charles T. Hunt, ‘Using force to protect civilians in UN peacekeeping’, Survival, 63:3 (2021), pp. 143–70.

2 E.g. Sarah-Myriam Martin-Brûlé, ‘Tackling the anarchy within: The role of deterrence and great power intervention in peace operations’, in Stephen M. Saideman and Marie-Joëlle Zahar (eds), Intra-state Conflict, Governments and Security: Dilemmas of Deterrence and Assurance (Abingdon: Routledge, 2008), pp. 182–204; UN Department of Peacekeeping Operations and Department of Field Support, ‘United Nations peacekeeping operations: Principles and guidelines’ (2008); Andrea Ruggeri, Theodora-Ismene Gizelis, and Han Dorussen, ‘Managing mistrust: An analysis of cooperation with UN peacekeeping in Africa’, Journal of Conflict Resolution, 57:3 (2012), pp. 387–409; Vanessa F. Newby, ‘Offering the carrot and hiding the stick? Conceptualizing credibility in UN peacekeeping’, Global Governance, 28:3 (2022), pp. 303–29.

3 Thomas C. Schelling, Arms and Influence (New Haven, CT: Yale University Press, 1966), pp. 69–73, 79–80.

4 E.g. Donald C. F. Daniel, Bradd C. Hayes, and Chantal de Jonge Oudraat, Coercive Inducement and the Containment of International Crises (Washington, DC: United States Institute of Peace Press, 1999); David E. Johnson, Karl P. Mueller, and William H. Taft V, Conventional Coercion across the Spectrum of Operations: The Utility of U.S. Military Forces in the Emerging Security Environment (Santa Monica, CA: RAND, 2002); Wallace J. Thies, ‘Compellence failure or coercive success? The case of NATO and Yugoslavia’, Comparative Strategy, 22:3 (2003), pp. 243–67; Kersti Larsdotter, ‘Military strategy and peacekeeping: An unholy alliance?’, Journal of Strategic Studies, 42:2 (2019), pp. 191–211.

5 E.g. Fred Tanner, ‘Weapons control in semi-permissive environments: A case for compellence’, International Peacekeeping, 3:4 (1996), pp. 126–45; Peter Viggo Jakobsen, ‘Reinterpreting Western use of coercion in Bosnia-Herzegovina: Assurances and carrots were crucial’, Journal of Strategic Studies, 23:2 (2000), pp. 1–22; David Carment and Frank Harvey, Using Force to Prevent Ethnic Violence: An Evaluation of Theory and Evidence (Westport, CT: Praeger, 2001); Taylor B. Seybolt, Humanitarian Military Intervention: The Conditions for Success and Failure (Oxford: Oxford University Press, 2008).

6 Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, NY: Cornell University Press, 1996), pp. 18–19.

7 Karl Mueller, ‘Strategies of coercion: Denial, punishment, and the future of air power’, Security Studies, 7:3 (1998), pp. 182–228(pp. 214-15).

8 E.g. Alexander L. George and William E. Simons (eds), The Limits of Coercive Diplomacy, 2nd ed. (Boulder, CO: Westview Press, 1994); Peter Viggo Jakobsen, Western Use of Coercive Diplomacy after the Cold War: A Challenge for Theory and Practice (Basingstoke: Macmillan Press, 1998); Barry M. Blechman and Tamara Cofman Wittes, ‘Defining moment: The threat and use of force in American foreign policy’, Political Science Quarterly, 114:1 (1999), pp. 1–30; Robert J. Art and Patrick M. Cronin (eds), The United States and Coercive Diplomacy (Washington, DC: United States Institute of Peace Press, 2003); Todd S. Sechser, ‘Militarized compellent threats, 1918–2001’, Conflict Management and Peace Science, 28:4 (2011), pp. 377–401; Phil Haun, Coercion, Survival, and War: Why Weak States Resist the United States (Stanford, CA: Stanford University Press, 2015); Dianne Pfundstein Chamberlain, Cheap Threats: Why the United States Struggles to Coerce Weak States (Washington, DC: Georgetown University Press, 2016); Robin Markwica, Emotional Choices: How the Logic of Affect Shapes Coercive Diplomacy (Oxford: Oxford University Press, 2018).

9 E.g. Kofi Nsia-Pepra, UN Robust Peacekeeping: Civilian Protection in Violent Civil Wars (New York: Palgrave Macmillan, 2014); Lisa Hultman, Jacob D. Kathman, and Megan Shannon, Peacekeeping in the Midst of War (Oxford: Oxford University Press, 2019); Allison Carnegie and Christoph Mikulaschek, ‘The promise of peacekeeping: Protecting civilians in civil wars’, International Organization, 74:4 (2020), pp. 810–32; Anup Phayal and Brandon C. Prins, ‘Deploying to protect: The effect of military peacekeeping deployments on violence against civilians’, International Peacekeeping, 27:2 (2020), pp. 311–36; Barbara F. Walter, Lise Morje Howard, and V. Page Fortna, ‘The extraordinary relationship between peacekeeping and peace’, British Journal of Political Science, 51:4 (2021), pp. 1705–22.

10 Virginia Page Fortna, Does Peacekeeping Work? Shaping Belligerents’ Choices after Civil War (Princeton, NJ: Princeton University Press, 2008); Lise Morjé Howard, Power in Peacekeeping (Cambridge: Cambridge University Press, 2019).

11 Howard, Power in Peacekeeping, pp. 129–30, 199. For more details on the three principles, see UN Department of Peacekeeping Operations and Department of Field Support, ‘Principles and guidelines’, pp. 31–5.

12 Ken Ohnishi, ‘Analysing the use of compellence during peace operations’, PhD thesis, University of Leeds, 2020, p. 109.

13 Alexander L. George, ‘Coercive diplomacy: Definition and characteristics’, in Alexander L. George and William E. Simons (eds), The Limits of Coercive Diplomacy, 2nd ed. (Boulder, CO: Westview Press, 1994), pp. 7–11(p. 9).

14 Lawrence Freedman, ‘Strategic coercion’, in Lawrence Freedman (ed.), Strategic Coercion: Concepts and Cases (New York: Oxford University Press, 1998), pp. 15–36 (p. 30); Daniel Byman and Matthew Waxman, The Dynamics of Coercion: American Foreign Policy and the Limits of Military Might (Cambridge: Cambridge University Press, 2002), pp. 37–8.

15 Peter Nadin, Patrick Cammaert, and Vesselin Popovski, Spoiler Groups and UN Peacekeeping (Abingdon: Routledge, 2015), p. 109.

16 E.g. James Sloan, The Militarisation of Peacekeeping in the Twenty-First Century (Oxford: Hart Publishing, 2011), pp. 294–5; Howard, Power in Peacekeeping, pp. 194–6.

17 Charles T. Hunt, ‘All necessary means to what ends? The unintended consequences of the “robust turn” in UN peace operations’, International Peacekeeping, 24:1 (2017), pp. 108–31(p. 114).

18 Chiyuki Aoi and Cedric de Coning, ‘Conclusion: Towards a United Nations stabilization doctrine: Stabilization as an emerging UN practice’, in Cedric de Coning, Chiyuki Aoi, and John Karlsrud (eds), UN Peacekeeping Doctrine in a New Era: Adapting to Stabilisation, Protection and New Threats (Abingdon: Routledge, 2017), pp. 288–310(p. 302).

19 Seun Abiola, Cedric de Coning, Eduarda Hamann, and Chander Prakash, ‘The large contributors and UN peacekeeping doctrine’, in Cedric de Coning, Chiyuki Aoi, and John Karlsrud (eds), UN Peacekeeping Doctrine in a New Era: Adapting to Stabilisation, Protection and New Threats (Abingdon: Routledge, 2017), pp. 152–85 (pp. 159–60, 167–8); John Karlsrud, The UN at War: Peace Operations in a New Era (Cham: Palgrave Macmillan, 2018), pp. 123–4, 129–30.

20 UN Department of Peacekeeping Operations and Department of Field Support, ‘Principles and guidelines’, p. 34.

21 UN, ‘Report of the High-Level Independent Panel on Peace Operations on uniting our strength for peace: Politics, partnership and people’, A/70/95-S/2015/446, para. 128.

22 Carlos Alberto dos Santos Cruz, William R. Phillips, and Salvator Cusimano, ‘Improving security of United Nations peacekeepers: We need to change the way we are doing business’, UN, 2017, available at: {https://peacekeeping.un.org/sites/default/files/improving_security_of_united_nations_peacekeepers_report.pdf}, accessed 12 June 2024.

23 E.g. Kelly M. Greenhill and Solomon Major, ‘The perils of profiling: Civil war spoilers and the collapse of intrastate peace accords’, International Security, 31:3 (2006/7), pp. 7–40.

24 E.g. George and Simons (eds), Limits of Coercive Diplomacy; Jakobsen, Western Use of Coercive Diplomacy; Blechman and Wittes, ‘Defining moment’; Art and Cronin (eds), United States and Coercive Diplomacy; Haun, Coercion, Survival, and War; Chamberlain, Cheap Threats; Kelly M. Greenhill and Peter Krause (eds), Coercion: The Power to Hurt in International Politics (New York: Oxford University Press, 2018); Melanie W. Sisson, James A. Siebens, and Barry M. Blechman (eds), Military Coercion and US Foreign Policy: The Use of Force Short of War (Abingdon: Routledge, 2020).

25 Pape, Bombing to Win; Art and Cronin (eds), United States and Coercive Diplomacy; Rob de Wijk, The Art of Military Coercion: Why the West’s Military Superiority Scarcely Matters, 2nd ed. (Amsterdam: Amsterdam University Press, 2014).

26 Freedman, ‘Strategic coercion’, pp. 29–30.

27 Byman and Waxman, Dynamics of Coercion, pp. 4–5, 78.

28 Barry R. Posen, ‘Military responses to refugee disasters’, International Security, 21:1 (1996), pp. 72–111.

29 Mueller, ‘Strategies of coercion’, pp. 214–15.

30 Pape, Bombing to Win, pp. 69–79.

31 Samuel Zilincik and Tim Sweijs, ‘Beyond deterrence: Reconceptualizing denial strategies and rethinking their emotional effects’, Contemporary Security Policy, 44:2 (2023), pp. 248–75.

32 E.g. Wyn Q. Bowen, ‘Deterrence and asymmetry: Non-state actors and mass casualty terrorism’, Contemporary Security Policy, 25:1 (2004), pp. 54–70 (pp. 61–2); James H. Lebovic, Deterring International Terrorism and Rogue States: US National Security Policy after 9/11 (Abingdon: Routledge, 2007), pp. 162–7; James M. Smith and Brent J. Talbot, ‘Terrorism and deterrence by denial’, in Paul R. Viotti, Michael A. Opheim, and Nicholas Bowen (eds), Terrorism and Homeland Security: Thinking Strategically about Policy (Boca Raton, FL: CRC Press, 2008), pp. 53–68 (pp. 56–9); Alex S. Wilner, ‘Deterring the undeterrable: Coercion, denial, and delegitimization in counterterrorism’, Journal of Strategic Studies, 34:1 (2011), pp. 3–37 (pp. 22–4); Alex S. Wilner, Deterring Rational Fanatics (Philadelphia, PA: University of Pennsylvania Press, 2015), pp. 61–6.

33 Ohnishi, ‘Analysing the use of compellence’, pp. 89–92.

34 Steven Metz and Raymond Millen, Insurgency and Counterinsurgency in the 21st Century: Reconceptualizing Threat and Response (Carlisle: Strategic Studies Institute, 2004), pp. 7–8.

35 J. Boone Bartholomees, Jr, ‘The issue of attrition’, Parameters, 40:1 (2010), pp. 5–19 (pp. 9–10, 16).

36 Cameron I. Crouch, Managing Terrorism and Insurgency: Regeneration, Recruitment and Attrition (Abingdon: Routledge, 2010).

37 Avi Kober, Israel’s Wars of Attrition: Attrition Challenges to Democratic States (Abingdon: Routledge, 2009), p. 163.

38 V. K. Jetley, ‘“Op Khukri”: The United Nations operation fought in Sierra Leone part-II’, USI Journal, 137:568 (2007).

39 International Crisis Group (hereafter ICG), ‘Congo: Bringing peace to North Kivu’, Africa Report 133, 2007, p. 8; Jason Stearns, From CNDP to M23: The Evolution of an Armed Movement in Eastern Congo (London: Rift Valley Institute, 2012), p. 29; Nadin, Cammaert, and Popovski, Spoiler Groups and UN Peacekeeping, p. 75.

40 Keith B. Payne, Shmuel Bar, Patrick Garrity et al., Deterrence and Coercion of Non-state Actors: Analysis of Case Studies (Fairfax, VA: National Institute for Public Policy, 2008); Ben Connable and Martin C. Libicki, How Insurgencies End (Santa Monica, CA: RAND, 2010), pp. 34–49; Jeffrey Treistman, ‘Home away from home: Dynamics of counterinsurgency warfare’, Comparative Strategy, 31:3 (2012), pp. 235–52.

41 Bard E. O’Neill, Insurgency and Terrorism: Inside Modern Revolutionary Warfare (Dulles, VA: Brassey’s, 1990), pp. 55–6; Luis de la Calle and Ignacio Sánchez-Cuenca, ‘How armed groups fight: Territorial control and violent tactics’, Studies in Conflict and Terrorism, 38:10 (2015), pp. 795–813.

42 ICG, ‘Iraq after the surge I: The new Nunni landscape’, Middle East Report 74, 2008; Connable and Libicki, How Insurgencies End, pp. 41–3; Stephen Biddle, Jefferey A. Friedman, and Jacob N. Shapiro, ‘Testing the surge: Why did violence decline in Iraq in 2007?’, International Security, 37:1 (2012), pp. 7–40; Lawrence E. Cline, ‘The two surges: Iraq and Afghanistan in comparison’, Small Wars and Insurgencies, 33:7 (2022), pp. 1152–76.

43 Stephen T. Hosmer, Operations against Enemy Leaders (Santa Monica, CA: RAND, 2001), pp. 3–4; Wilner, ‘Deterring the undeterrable’, p. 20.

44 Audrey Kurth Cronin, How Terrorism Ends: Understanding the Decline and Demise of Terrorist Campaigns (Princeton, NJ: Princeton University Press, 2009), pp. 18–20; Stewart J. D’Alessio, Lisa Stolzenberg, and Dustin Dariano, ‘Does targeted capture reduce terrorism?’, Studies in Conflict and Terrorism, 37:10 (2014), pp. 881–94; David Scott Palmer, ‘Revolutionary leadership as necessary element in people’s war: Shining Path of Peru’, Small Wars and Insurgencies, 28:3 (2017), pp. 426–50.

45 E.g. Cronin, How Terrorism Ends; Jenna Jordan, ‘When heads roll: Assessing the effectiveness of leadership decapitation’, Security Studies, 18:4 (2009), pp. 719–55; Patrick B. Johnston, ‘Does decapitation work? Assessing the effectiveness of leadership targeting in counterinsurgency campaigns’, International Security, 36:4 (2012), pp. 47–79; Alex S. Wilner, Deterring Rational Fanatics (Philadelphia, PA: University of Pennsylvania Press, 2015); Yasutaka Tominaga, ‘Organizational context matters: Explaining different responses to militant leadership targeting’, Conflict Management and Peace Science, 38:3 (2021), pp. 270–91.

46 E.g. UN, ‘Report of the Secretary-General on the Central African Republic’, S/2017/473, para. 11; UN, ‘Central African Republic: Report of the Secretary-General’, S/2020/545, para. 31.

47 Daniel Byman and Matthew Waxman, ‘Defeating US coercion’, Survival, 41:2 (1999), pp. 107–20(p. 111).

48 Byman and Waxman, ‘Defeating US coercion’; Willem Martijn Dekker, ‘What the bombing of Hanoi tells us about compellence theory’, EUI Working Paper 2011/28, European University Institute, 2011; Frank Harvey and Alex Wilner, ‘Counter-coercion, the power of failure, and the practical limits of deterring terrorism’, in Andreas Wenger and Alex Wilner (eds), Deterring Terrorism: Theory and Practice (Stanford, CA: Stanford University Press, 2012), pp. 95–114.

49 UN peace operations in cooperation with the DRC government employed compellence on at least three occasions in Kivus. MONUC tried to compel the Democratic Forces for the Liberation of Rwanda (FDLR) to disarm but failed. MONUC also tried to compel the National Congress for the Defence of the People (CNDP) to end violence and join the national army. The compellence was a partial success, because an agreement was reached with a major compromise on the compeller’s side. MONUSCO tried to compel the M23 to end rebellion, but the compellence was nearly a failure, as the rebel group was almost defeated and the objective was almost forcibly achieved. Ohnishi, ‘Analysing the use of compellence’, pp. 105–7.

50 Arend Lijphart, ‘Comparative politics and the comparative method’, American Political Science Review, 65:3 (1971), pp. 682–93 (p. 689); Stephen Van Evera, Guide to Methods for Students of Political Science (Ithaca, NY: Cornell University Press, 1997), pp. 82–3; Alexander L. George and Andrew Bennett, Case Studies and Theory Development in the Social Sciences (Cambridge, MA: MIT Press, 2005), pp. 166–7.

51 ICG, ‘Congo crisis: Military intervention in Ituri’, Africa Report 64, 2003, pp. 7–10.

52 UN, ‘Second special report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2003/566, para. 10.

53 Peacekeeping Best Practice Unit, ‘Operation Artemis: The lessons of the Interim Emergency Multinational Force’, Peacekeeping Best Practice Unit, UN, 2004.

54 UNSC, ‘Resolution 1493 (2003)’, S/RES/1493.

55 Jim Terrie, ‘The use of force in UN peacekeeping: The experience of MONUC’, African Security Review, 18:1 (2009), pp. 21–34 (pp. 23–4, 30–1); Alan Doss, ‘In the footsteps of Dr Bunche: The Congo, UN peacekeeping and the use of force’, Journal of Strategic Studies, 37:5 (2014), pp. 703–35(pp. 716–17).

56 UN, ‘Fourteenth report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2003/1098, para. 8; Jan-Gunnar Isberg and Lotta Victor Tillberg, By All Necessary Means: Brigadier General Jan-Gunnar Isberg’s Experiences from Service in the Congo 2003–2005 (Stockholm: Swedish National Defence College, 2012), pp. 49–50, 58, 78.

57 E.g. Eddy Isango, ‘UN promises final crackdown on militias in Congo’s northeast’, Associated Press International (hereafter API) (21 January 2004); Integrated Regional Information Networks (hereafter IRIN), ‘Fighting between UN troops, militias leaves 50 dead’, Africa News (2 March 2005).

58 IRIN, ‘UN envoy gives militiamen ultimatum to disarm’, Africa News (14 March 2005).

59 ICG, ‘Maintaining momentum in the Congo: The Ituri problem’, Africa Report 84, 2004, pp. 9–10.

60 UN, ‘Seventeenth report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2005/167, para. 13; ICG, ‘Congo: Four priorities for sustainable peace in Ituri’, Africa Report 140, 2008, p. 30.

61 Tsjeard Bouta, ‘Assessment of the Ituri disarmament and community reinsertion program (DCR)’, Netherland Institute of International Relations, 2005, p. 14.

62 UN News Service (hereafter UNNS), ‘Disarmament in Ituri progresses, but other steps needed – UN official’, Africa News (8 November 2007).

63 Radio Okapi, ‘UN commander warns militias in northeastern region’, BBC Monitoring Africa (1 April 2006); IRIN, ‘New disarmament deadline, amnesty offer for militiamen’, Africa News (15 June 2006); IRIN, ‘Deadline for militias to disarm extended to 15 July’, Africa News (6 July 2006).

64 IRIN, ‘New disarmament deadline’; IRIN, ‘Militiamen disarm ahead of deadline’, Africa News (29 June 2006); IRIN, ‘Two militia leaders appointed army colonels’, Africa News (11 October 2006).

65 ICG, ‘Four priorities’, p. 34.

66 Crispin Nlanda, ‘Close to 6,000 militiamen to surrender their weapons in Ituri’, AllAfrica (1 December 2006).

67 ICG, ‘Four priorities’, p. 3.

68 Radio Okapi, ‘Calm restored in northeast DRCongo village after army, militia clash’, BBC Monitoring Africa (2 February 2007); ICG, ‘Four priorities’, p. 3.

69 Eoin Young, ‘Peter Karim surrenders 170 men’, AllAfrica (28 February 2007); UN, ‘Twenty-third report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2007/156, para. 19.

70 UNNS, ‘Disarmament in Ituri progresses’; UNNS, ‘Transfer of ex-combatants major step for peace in northeast DR Congo – UN’, Africa News (5 November 2007).

71 E.g. ‘UN troops kill 10 militiamen in eastern Congo’, Reuters News (8 May 2004); Bryan Mealer, ‘U.N. peacekeepers kill nearly 60 militia in battle in northeastern Congo’, API (2 March 2005); Radio Okapi, ‘Army kills four militiamen, arrests 30 in retaliation operation’, BBC Monitoring Africa (24 March 2005); David Lewis, ‘U.N. troops kill 38 militiamen in Congo raid’, Reuters News (3 April 2005).

72 Isberg and Tillberg, By All Necessary Means, pp. 91–103.

73 IRIN, ‘MONUC helps free sex slaves, civilian prisoners held by militiamen’, Africa News (4 December 2003); ICG, ‘Maintaining momentum’, p. 9; Bouta, ‘Assessment of the Ituri DCR’, p. 13; Kristof Titeca, ‘Access to resources and predictability in armed rebellion: The FAPC’s short-lived “Monaco” in eastern Congo’, Africa Spectrum, 46:2 (2011), pp. 43–70(p. 50); Isberg and Tillberg, By All Necessary Means, p. 21.

74 E.g. Anjan Sundaram, ‘Deadly military operation in East’, AP Worldstream (16 November 2005); David Lewis, ‘Congo army attacks militia to help 50,000 displaced’, Reuters News (23 November 2005); UNNS, ‘UN peacekeeper, DR of Congo troops and scores of Ugandan rebels killed in sweep’, Africa News (27 December 2005); Radio Okapi, ‘Thirty-eight killed in fighting in Ituri’, BBC Monitoring Africa (27 April 2006); Anjan Sundaram, ‘Hundreds of militia fighters raid army base in eastern Congo; 53 dead’, The Associated Press (14 May 2006); Radio France Internationale, ‘Eighty six said killed in recent offensive against rebels in northeast’, BBC Monitoring Africa (2 June 2006).

75 IRIN, ‘Aid arrives for displaced thousands but fears over safety persist’, Africa News (16 February 2006); IRIN, ‘Two militia leaders appointed army colonels’.

76 IRIN, ‘Voluntary militia disarmament ends’, Africa News (26 July 2006); Edward Harris, ‘Three militia groups in eastern Congo to lay down arms ahead of elections’, API (27 July 2006).

77 If a media report raises two names of armed groups that the killed or wounded rebels belonged to, the reported casualty figure is halved between the groups.

78 E.g. IRIN, ‘12 militias killed in clashes with army in Ituri’, Africa News (9 October 2006); Henning Tamm, FNI and FRPI: Local Resistance and Regional Alliances in North-Eastern Congo (London: Rift Valley Institute, 2013), pp. 38–42.

79 IRIN, ‘Recently demobilised militiamen re-arming in volatile Ituri district’, Africa News (17 September 2006).

80 Ibid.; ICG, ‘Four priorities’, p. 34; Alex Veit, Intervention as Indirect Rule: Civil War and Statebuilding in the Democratic Republic of Congo (Frankfurt: Campus Verlag, 2010), pp. 177–8.

81 E.g. David Lewis, ‘Militia, UN clash in north Congo as peace stalls’, Reuters News (7 December 2004); Radio Okapi, ‘Militia attack convoy of UN peacekeepers in northeastern district’, BBC Monitoring Africa (22 April 2005); UN, S/2005/167, para. 19; Isberg and Tillberg, By All Necessary Means, pp. 91–103, 150–2.

82 Veit, Intervention as Indirect Rule, pp. 171, 222; Tamm, FNI and FRPI, p. 31.

83 UN, ‘Eighteenth report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2005/506, para. 25; UN, ‘Twentieth report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2005/832, para. 22.

84 UN, ‘Twenty-first report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2006/390, para. 33; UN, ‘Twenty-second report of the Secretary-General on the United Nations Organization Mission in the Democratic Republic of the Congo’, S/2006/759, para. 42; ICG, ‘Four priorities’, p. 33.

85 UN, S/2006/390, para. 33; ICG, ‘Four priorities’, p. 34.

86 Patrick C. Cammaert, ‘Learning to use force on the hoof in peacekeeping: Reflections on the experience of MONUC’s eastern division’, Situation Report, Institute for Security Studies, 2007, p. 7.

87 IRIN, ‘Displaced civilians desperate for help’, Africa News (4 January 2007); Radio Okapi, ‘Army clashes with rebels in northwest’, BBC Monitoring Africa (1 February 2007).

88 Eoin Young, ‘Gen. Mayala – The only obstacle that remains is Peter Karim’, AllAfrica (24 February 2007); MONUC, ‘Monthly human rights assessment – February 2007ʹ, AllAfrica (20 March 2007); UNNS, ‘Notorious militia leader disarms, demands amnesty’, Africa News (28 February 2007); UN, S/2007/156, para. 19.

89 IRIN, ‘Voluntary militia disarmament ends’; IRIN, ‘Recently demobilised militiamen re-arming’.

90 Isberg and Tillberg, By All Necessary Means, pp. 51, 63.

91 Henning Tamm, UPC in Ituri: The External Militarization of Local Politics in North-Eastern Congo (London: Rift Valley Institute, 2013), p. 32.

92 IRIN, ‘UN arrests political leaders in Ituri’, Africa News (29 June 2004); Eddy Isango, ‘United Nations brokers Congo peace talks after fighting kills at least 19’, API (6 July 2004).

93 IRIN, ‘Another key Ituri leader arrested’, Africa News (22 March 2005); ‘Militia leader arrested in eastern Congo’, Reuters News (9 April 2005); Eddy Isango, ‘Peacekeepers and police arrest another warlord in eastern Congo’, API (15 April 2005); ICG, ‘Four priorities’, p. 31.

94 IRIN, ‘Army captures militia commander’, Africa News (17 May 2006); ‘Militia leader captured in eastern Congo – U.N.’, Reuters News (17 May 2006).

95 IRIN, ‘Ituri militias take war to civilians’, Africa News (23 March 2005).

96 Tamm, FNI and FRPI, p. 35.

97 E.g. Bryan Mealer, ‘Officials: U.N. peacekeepers in northeastern Congo clash with militia fighters’, AP Worldstream (27 June 2005); David Lewis, ‘Congo militia use civilians as human shields – UN’, Reuters News (1 March 2006); Isberg and Tillberg, By All Necessary Means, p. 58.

98 Cammaert, ‘Learning to use force’, p. 7; Rajesh Isser, Peacekeeping and Protection of Civilians: The Indian Air Force in the Congo (New Delhi: KW Publishers, 2012), pp. 50–2.

99 Isberg and Tillberg, By All Necessary Means, p. 181.

100 Veit, Intervention as Indirect Rule, pp. 150, 177.

101 Ibid., pp. 156, 161, 177; ICG, ‘Four priorities’, p. 3.

102 Radio Okapi, ‘UN blames east DRCongo armed groups for deaths of “blue helmets”, urges arrests’, BBC Monitoring Africa (26 February 2005); Isberg and Tillberg, By All Necessary Means, pp. 49, 57.

103 E.g. Radio Okapi, ‘UN blames east DRCongo armed groups’; IRIN, ‘Peacekeeper dies, six others wounded in ambush’, Africa News (13 May 2005); ‘U.N. peacekeeper killed in Congo attack’, Reuters News (3 June 2005); David Lewis, ‘Congo militia holds 7 Nepali UN soldiers – sources’, Reuters News (29 May 2006).

104 Isberg and Tillberg, By All Necessary Means, p. 189.

105 United States Department of State, ‘Aviation essential for Democratic Republic of Congo security’, Africa News (9 June 2006); Isser, Peacekeeping and Protection of Civilians, pp. 93, 104–11, 129.

106 Philipp Münch and Alex Veit, ‘Intermediaries of intervention: How local power brokers shape external peace- and state-building in Afghanistan and Congo’, International Peacekeeping, 25:2 (2018), pp. 266–92(pp. 282–4).

107 David Lewis, ‘Congo militia holds 7 Nepali UN soldiers’.

108 David Lewis, ‘Congo militia threaten to execute UN peacekeepers’, Reuters News (26 June 2006); IRIN, ‘Interview with Brig-Gen Mahboob, commander of Monuc’s Ituri brigade’, Africa News (11 July 2006).

109 API, ‘5 remaining U.N. peacekeepers taken hostage in Congo are freed’, API (8 July 2006); MONUC, ‘Captured UN peacekeepers released’, Africa News (8 July 2006).

110 E.g. Stephanie Wolters, ‘Is Ituri on the road to stability? An update on the current security situation in the district’, Situation Report, Institute for Security Studies, 2005; Veit, Intervention as Indirect Rule, p. 158; Emizet Fançois Kisangani, Civil Wars in the Democratic Republic of Congo, 1960–2010 (Boulder, CO: Lynne Rienner Publishers, 2012), p. 205.

111 Radio Okapi, ‘Militiamen reportedly swamp disarmament sites ahead of deadline’, BBC Monitoring Africa (31 March 2005).

112 ICG, ‘Four priorities’, p. 34.

113 E.g. George and Simons (eds), Limits of Coercive Diplomacy; Byman and Waxman, Dynamics of Coercion; Art and Cronin (eds), United States and Coercive Diplomacy.

114 ICG, ‘Four priorities’, p. 34; IRIN, ‘Militiamen still taxing civilians despite leader joining army’, Africa News (27 October 2006).