Abstract
Armed groups have come to play an important role in contemporary international relations. Efforts to engage these groups in talks to reduce violence have been extensively studied, analyzed and criticized over the last couple of years. The same is the case for the
characteristics of armed groups that are used to better predict their behavior and formulate an effective policy response. A call from Bercovitch in 1991 to combine these two fields of study has for a large part been ignored however. This thesis aims to combine the study of armed groups with mediation scholarship by looking at how several characteristics of armed groups influence the mediation process. A literature study combined with expert interviews finds that the command and control structure of an armed group and the motivation of the group exert the most influence on the mediation process.
3
Table of Contents
1. Introduction: Armed Groups and Mediation p. 4
2. Theoretical Framework p. 5
2.1 Contingency Model 2.2 Nature of the Parties 2.3 Armed Groups
2.4 Characteristics of Armed Groups
3. Command and Control p. 13
3.1 Organizational Structure 3.2 Strategy
4. Motivation p. 18
4.1 Motivation of the Group 4.2 Motivation of the Individual
5. Support p. 24 5.1 Financial Support 5.2 Political Support 6. Conclusions p. 29 7. References p. 31 8. Appendices
4
1. Introduction: Armed Groups and Mediation
Mediation by third parties has been used as a conflict resolution method for centuries, but it only became a topic for scientific research fifty years ago (Kleiboer, 1996). There is a lot of catching up to do regarding our understanding of the dynamics of conflict mediation in comparison with its practice, but scholars have made considerable progress over the last couple of decades. This progress is definitely important at a time where conflicts with (non-state) armed groups seem to have replaced interstate wars, who require a different approach than conventional wars (Blin, 2011).
Even though today we know more about the mediation process than we did fifty years ago, some aspects still remain underexposed. The aspects that deserve more attention are
highlighted in the Contingency Model created by Bercovitch, Anagnoson and Wille (1991). Because the majority of the mediation scholars focuses on the characteristics of the mediator and on the timing of the mediation attempt, little attention is paid to aspects of the mediation process like the nature of the conflicting parties or the conflict environment. Bercovitch and Foulkes concluded that little has changed in the focus of mediation scholars since Bercovitch et al. first introduced their Contingency Model (2012). This thesis aims to fill the gap that still exists by focusing on one aspect of the Contingency Model: the nature of parties. Focusing on the nature of the conflicting parties can guide the mediator in the mediation process. Who should be included so the talks can be effective? Which incentives should be brought to the table? Which issues should be addressed? The globalization process of the last couple of years has increased the possibilities for armed groups to arm and fund themselves, which caused them to grow in number and size (Vinci, 2006b: 50). As an important actor in (international) violent conflicts I find them a good starting point to focus on while trying to fill the niche in our research about the effect of the conflicting parties on the mediation process.
In the chapters below, armed group literature will be combined with mediation scholarship to see which characteristics of armed groups can be used by mediators to improve the mediation process. After explaining Bercovitch’s Contingency Model in the theoretical framework in chapter 2, chapters 3-5 will analyze several characteristics of armed groups to see how they can influence the mediation process. This is analysis, based on existing literature, will be verified and complemented with experienced mediators Alvaro de Soto, Shadia Marhaban and Francesc Vendrell. In chapter 6 conclusions will be presented as to how characteristics of armed groups can influence the mediation process.
5
2. Theoretical framework
In the following paragraphs Bercovitch, Anagnoson and Wille’s Contingency Model will be explained (1991). The model shows why it is important to take a closer look at the conflicting parties to improve mediation efforts. Bercovitch, Anagnoson and Wille argue that besides the timing, the impartiality and the characteristics of the mediator, which is the focus of most scholars in the mediation literature, an in-depth knowledge of the conflicting parties is also important to improve the effectiveness of mediation (1991). Besides outlining the
Contingency Model, this chapter will also explain the focus on the nature of the parties within the Contingency Model and within that framework the specific focus on armed groups. 2.1 Contingency Model
Since the 1970s our knowledge about the mediation process and factors determining mediation success have significantly improved (Kleiboer, 1996). However, Bercovitch, Anagnoson and Wille noticed that the majority of this literature focused on the behavior and characteristics of the mediator or third-party (1991). They argued that to further improve the mediation process and the successfulness of mediation efforts, a broader scope of research was needed (1991: 9). This is why they created a Contingency Model, a model that gives an overview of all aspects of the mediation process. It not only looks at the outcome of the mediation effort but also at several context and process variables influencing the mediation outcome. Bercovitch et al. argue that in order to better prepare for a mediation attempt one must look at the identity and rank of the mediator, the environment or setting of the conflict, the different issues at stake and the nature of the conflicting parties. Furthermore they argue that to improve the mediation process one must focus on the timing and the setting of the mediation effort and also on the strategies and tactics that are used by the mediator (1991: 10-15). See figure 1 for a schematic overview of the model.
Despite the considerable effort Bercovitch and his colleagues have made in highlighting other aspects that contribute to the success of mediation efforts, the majority of mediation scholars keeps focusing on the mediator.1 Both Regan and Stam as well as Siniver have done
1 See for example: Bercovitch and Schneider (2000) Who Mediates? The Political Economy of International
Conflict Management; Bercovitch (1986) International Mediation. A Study of the Incidence, Strategies and Conditions of Successful Outcomes; Bercovitch (1991) International Mediation and Dispute Settlement. Evaluating the Conditions for Successful Mediation; Bercovitch (2006) Mediation Success or Failure. A Search for the Elusive Criteria; Bercovitch and Houston (2000) Why Do They Do It Like This. An Analysis of the Factors Influencing Mediation Behavior in International Conflicts; Bercovitch and Jackson (2001) Negotiation or
Mediation? An Exploration of Factors Affecting the Choice of Conflict Management in International Conflict and Bercovitch and Langley (1993) The Nature of the Dispute and the Effectiveness of International Mediation.
6
Figure 1. The Contingency Model (Bercovitch and Foulkes, 2012: 28)
important work in explaining why timing is important for the success of a mediation effort (2000; 2006). Regan and Stam explain how the timing of mediation efforts influences the duration of armed conflicts. They find that the sooner third parties intervene in a conflict, the better the success rates of mediation efforts seem to be. After a while, mediation efforts only seem to prolong the conflict as conflicting parties have time to catch a breath and regroup during these talks (Regan and Stam, 2000: 256-258). However, when the conflict has reached a mutually hurting stalemate time has come again for third parties to get the conflicting parties to the negotiation table. Siniver comes to the same conclusion while studying successful mediation efforts in the Middle East (2006). He finds that “the timing of
intervention has proved crucial to the disputants’ inclination to accept the mediation effort in the first place” (p. 823). The best timing for mediation is when the parties are “militarily exhausted”, which seems to be comparable with what Zartman calls a “mutually hurting stalemate” (2000: 823; 1985). A mutually hurting stalemate is a phase in the conflict where both parties “find themselves locked in a conflict from which they cannot escalate to victory
7
and this deadlock is painful to both of them” (2000: 828). William Zartman made a big contribution to mediation scholarship with his ripeness theory, arguing that “[w]hile most studies on peaceful settlement of disputes see the substance of the proposals for a solution as the key to a successful resolution of conflict, a growing focus of attention shows that a second and equally necessary key lies in the timing of efforts for resolution” (Zartman, 2001: 8) Another big debate in mediation literature considers the impartiality of the mediator. Several scholars (Inbar, 1991; Favretto, 2009; Smith, 1994) argue that a powerful mediator can push conflicting parties in the direction of a solution while others (Beardsley, 2008; Kydd, 2003; Kleiboer, 1996) argue that when a mediator puts pressure on the parties to solve their conflict an agreement that is reached is not likely to last. If the incentives put forth by the mediator lose importance over time and the capabilities and influence of all parties involved change, the disputing parties could become dissatisfied with the settlement agreement and might take up their arms again (Beardsley, 2008: 737). Different conflict situations require different types of mediators (Whitfield, 2010). Sometimes a mediator from a small NGO might be able to gain the trust of one of the parties when the actor refuse to talk to governmental
organizations. In another situation, pressure from the UN or individual states might be the most effective method to stop the violence.
When examining the Contingency Model we can conclude that only two of the factors influencing the success and effectiveness of mediation, i.e. mediation timing and identity and rank of the mediator, have been extensively covered by mediation scholars. The remaining variables continue to be under theorized. Except for Bercovitch his work, little has been written about how the conflict environment and nature of issues in a conflict effect the mediation process. And even though Siniver argues that timing and impartiality are the most important explanatory variables for successful mediation efforts, the other variables that Bercovitch, Anagnoson and Wille set forth also contribute to the improvement of the mediation process and the increase of success rates of mediation efforts (2006). 2.2 Nature of the parties
The part of the Contingency Model that this thesis aims to develop is the effect of the conflicting parties on the mediation process. An in-depth study of these parties can tell us more about how they will react to mediation efforts and if they are open to mediation to begin with. It is also important to know for the mediator how the conflicting parties are organized and how coherent they are in order to decide who to invite to the negotiation table.
8
Manikkalingam and Policzer argue for example that when a party - they focus on al-Qaeda - is decentralized it might be better to target groups at the periphery (2007). Groups at the periphery might be easier to convince to stop using violence because their goals or
motivations are different from those of the leaders of the organization. Another factor that can be of importance to the mediator is to know if the disputants are driven by greed or by
grievance issues. Actors driven by greed might be easier to approach than actors that are pursuing a more ideological goal. On the other hand, if disputants are driven by greed, the warlike situation in which they find themselves might be the best environment to enrich themselves (Vinci, 2006c). Mediation with these actors will be difficult because they profit from the situation as it is. Military action instead of mediation might be a better solution to end the violence in this case. All these examples are based on single articles, a more extensive literature study will show if these thoughts are shared by other scholars and experts.
Before further elaborating on the influence of the nature of parties on the mediation process it is important for the coherence and the time and space limits of this thesis to narrow down the scope of this research a little further. First of all we need to develop an understanding of which parties can influence the mediation process besides the mediating party. All conflicts start with two or more parties disagreeing on a specific issue. This issue can be anything from historical feuds to a clash of arms because of a specific policy set out by a country’s
government. Whenever a conflict becomes more violent, more parties tend to get involved. In the globalized world we live in today conflicting parties can seek support for their cause abroad which causes bigger actors (like states) to get involved as well. Even though targeting these actors might slow down the conflict, mediation to reach a peace agreement needs to be done with Track One disputants (Wallensteen and Svensson, 2014).2 Therefore this thesis will focus only on parties directly involved in the violence and not on parties supporting one side or the other. The parties that are directly involved in the fighting can still be a broad range of actors however. To further narrow down the scope of this thesis I therefore choose to focus on one important sort of actor often involved in violent conflicts: armed groups.
2.3 Armed groups
Armed groups are probably the most important actors in violent conflicts (Hofmann, 2006: 396). Whether they are called insurgent groups, guerrilla fighters, terrorists or militia’s, they have in common that they oppose the legitimate government of a country and that they took
2 Track One disputants are representatives of the main conflicting actors and not informal representatives or
9
up arms to fight it, thereby challenging the monopoly of violence originally held by the state (Shultz, Farah and Lochard, 2004). The variety of motivations, characteristics and strategies used by armed groups makes it hard to properly define them, but even more important to study because it are these differences that can explain why they react differently to mediation attempts (Vinci, 2006c; Hofmann, 2006). Vinci defines armed groups as “non-state
organizations that have the capacity for systemic military action” (2006c: 27). This rather broad definition includes all groups this thesis plans to study, namely all groups that could possibly be engaged in mediation efforts.
2.4 Characteristics of armed groups
The model set forth by Bercovitch, Anagnoson and Wille regards the outcome of a mediation attempt, whether it was successful or not, as contingent upon the nature of the conflicting parties (1991: 9). Whitfield confirms Bercovitch’s focus on the nature of the conflicting parties by saying that “(f)or a mediator or facilitator decisions regarding whether, and how, to engage will be rooted in detailed analysis of the armed group” (2010: 18). In the literature about armed groups several articles can be found that elaborate on this ‘nature of parties’ as Bercovitch et al. call it. Anthony Vinci, a specialist as it comes to armed groups has focused his research on the motivation of armed groups. In several articles he discusses what drives groups to fight (2006a; 2006c; 2006b).3 William Reno and Abdulkader Sinno both focus on the organizational structures of armed groups (2007; 2011). Reno discusses how patronage politics in West-African countries influence the organizational structure of armed groups and how these structures can be used to form a governmental structure, while Sinno writes about the phase after the organizational structure is formed. He discusses how these structures influence the strategic choices of a group and how these choices influence the chances of eventually reaching a peace agreement (Sinno, 2011). The violent nature of certain armed groups, terrorist group in particular, has caused a debate that is closely related to the subject of mediation. This debate will be further elaborated upon in chapter 3.2.
Instead of Bercovitch’s choice for the word “nature”, the rest of the document will use the word “characteristics”. The preference for characteristics over nature is merely for stylistics and clarity reasons. The “nature of a party” refers to an identity of a person or thing in
3 One article compares armed groups in Somalia and argues that a good understanding of the dynamics of
these groups is important to form a effective counter policy (2006a). Another article discusses why the greed and grievances debate has become obsolete. Vinci argues that the ultimate goal of each group is to survive in some way or form (2006c). The last article discusses how problems of mobilization that each armed group has to face form a better system of analysis than other taxonomies of armed groups (Vinci, 2006b).
10
general, while “characteristics” refers more specifically to several qualities one can posses (Website Thesaurus). These qualities added together then form the nature of a person, or armed group in this case. The method used most often in the literature to analyze armed groups is to put each group in a certain category.4 The problem with this approach is that once a certain designation like ‘guerilla’, ‘militia’ or ‘terrorist’, is given to a group, a strategy to deal with these groups is often based on this designation (Vinci, 2006b). The danger is then that once the group changes its strategy or maybe not even used this strategy in the first place, a wrong policy strategy to fight this group is followed. This was the case in Uganda where the government called the LRA a ‘guerilla insurgency’ and tried to isolate the group from the local support that guerilla insurgencies, like the GAM in Aceh for example, usually receive. However, the LRA never had much local support and the only effect this measure had was alienating the locals further from the government in Kampala (Vinci, 2006b: 49).
Furthermore, the definitions of terms like insurgency, terrorist movement, guerilla group or militia, differ significantly amongst different scholars. To avoid mistakes like the one in Uganda and to overcome the definitional problems of armed groups, Anthony Vinci came up with a different approach. After explaining why it is so difficult to properly define what an armed group is, he continues arguing that despite this fact, all armed groups face the same ‘problems of mobilization’ (2006b: 50). Problems of mobilization are the problems that any armed group must face in order to mobilize its fighters. Vinci states that the international community can determine its own strategy towards armed groups by analyzing how groups handle these problems, instead of basing its strategy on the designation given to the group. Furthermore Vinci’s method of analysis takes into account the “evolving and adapting nature of contemporary armed groups” (Vinci, 2006b: 50). When something within a group changes an approach based on a group’s characteristics will pick this up much quicker than an
approach based on any other taxonomy. Therefore I decided to look at how certain specific characteristics of armed groups as set out by Anthony Vinci can influence the mediation process.
The characteristics this thesis looks at are largely based on Vinci’s article about the
mobilization problems of armed groups. He starts with discussing the different motivations of people to take up arms and start fighting for an armed group. Subsequently he addresses how they get access to weapons and how they finance themselves, he calls this the logistical problems a group needs to solve. Finally he investigates different ways that groups can
4
11
organize their command, control and communication systems. Based on the specific
characteristics of a group one can then formulate an effective policy approach. For example if a group supports itself with the exploitation of natural resources and is motivated to fight merely because of economic incentives, like the MEND (Movement for the Emancipation of the Niger Delta) in Nigeria, one can decide to take Charles Taylor’s NPFL and the way the international community responded as an example of how to respond to them (Vinci, 2006b). This thesis looks specifically at what information these characteristics give the mediator and how they can affect the mediation process. Therefore not all characteristics in the form in which they are outlined by Vinci are relevant to discuss in this thesis. I thus decided to make some alterations in the model of analysis that Vinci proposes. First of all, several mediation scholars, like Aghedo and Osumah, Bercovitch and Langley, Wennman, and Withfield, highlight that it is important to look at what causes these groups to take up their arms (2012; 1993; 2009, 2010). In other words, what are the motivations of the parties in a conflict? Vinci mainly discusses the ways in which individuals are motivated to fight. But in accordance to other mediation literature this thesis will also discuss the motivation put forth by the group as a whole.
Secondly, a feature that is not per se necessary for a group’s survival and therefore not covered by Vinci is the political support a group receives from other actors. Support for armed groups, especially political support, is extensively discussed in other armed group literature.5 In international politics states often refuse to engage into negotiations with armed groups because this might give them a platform for their cause and could legitimize their means. The fact that a group receives political support from a third party could be important information for the mediator and will therefore be treated as a characteristic of an armed group in the literature study below.
Another form of support for an armed group is covered by Vinci under the name logistical problems. This section covers how groups get access to weapons, a feature that distinguishes them from other opposition groups, equipment and other means of survival. I will summarize these forms of support under the name financial support. Financial support can materialize in several forms. A group can be supporting, as indicated by Vinci, through looting, self-manufacturing, and purchasing or it can be supported by a third party which provides the group with direct financial support or weapons. The best known form of support to armed
5
12
groups by other actors is the support the US and USSR gave to opposition groups in countries loyal to the other superpower during the Cold War (Vinci, 2006b: 56).
The last characteristics Vinci discusses are the command, control and communication structures of an armed group. Each group needs some sort of organizational structure to be able to direct its forces (Vinci, 2006b: 56). In the literature of armed groups command and control is often discussed separately from strategy, with communication systems being the result of a certain command and control structure. Therefore a group’s strategy will be treated as a separate characteristic, while the communication system will not be discussed in this thesis. This is because a communication system is often dependent upon a certain command and control structure and will therefore not provide any additional information for the mediator.
In the following chapters, armed group scholarship will be combined with mediation literature to determine how characteristics of armed groups can influence the mediator and the
mediation process. In summary, the following characteristics of armed groups will be
discussed: the organizational structure of a group, their strategies, the group’s motivation and the motivation of individual group members and finally the political and financial support a group encounters. After discussing the theory from the armed groups literature, articles discussing mediation will determine how these influence the mediation process. A better understanding of armed groups forms a basis for analyzing the challenges and risks specific to engagement with each individual group. (Whitfield, 2010: 10).
13
1. Command and Control
This chapter and the following chapters will focus on the different characteristics that can be attributed to armed groups. This chapter addresses the command and control structures present in armed groups. The command and control structure are all the structures and processes put in place by the leadership of a group to facilitate the accomplishment of its mission (Pigeau and McCann, 2002: 54). A command and control structure consists of two key aspects: its organizational structure (i.e. centralized or decentralized) and its strategies of combat (Vinci, 2006c). In the paragraphs below will be discussed which options there are within each category and subsequently how these could influence the mediation process. 3.1 Organizational Structure
Each armed group has a certain hierarchy with a central command giving the orders to a broader support base. Even though each group has its unique background and history, many groups are organized along similar lines, either as centralized or decentralized and with a leadership format that allows them to give orders along certain hierarchical lines (Vinci, 2006b: 57).
Centralized organizations versus decentralized organizations
In a centralized organization most of the decision making power lies with the central command. Questions relating to the distribution of resources, combat strategy, enforcing discipline and formulating common goals (and enemies) are all solved by the central command (Sinno, 2011: 316). The more power is invested in local cadres or second-level commanders, the more decentralized an organization tends to be. Decentralized organizations like al-Qaeda, Boko Haram and the Sudanese sponsored Janjaweed allow for more
independence in individual cells. The problem with this structure is the risk of factionalization of groups at the periphery.
It is this aspect of the command and control structure of al-Qaeda that Policzer and
Manikkalingam ought to be important for a mediator to take into account (2007). They argue that its decentralized structure may serve as an opportunity for a mediator to isolate al-Qaeda’s central command from the bottom up by targeting the groups at the periphery. This approach is supported by Alvaro de Soto, renowned international mediator and former UN Under-Secretary-General, who adds that it is vital to the success of the talks that all
14
grievances are heard, especially when those of the periphery differ from those of the central command (Interview Alvaro de Soto, 28-05-2015).
Centralized organizations are usually more disciplined and technologically better equipped, in order to stay in touch with geographically remote cells, compared to decentralized
organizations (Vinci, 2006b: 57). A downside of a centralized organization is that their structure makes them vulnerable for a power play by a foreign sponsor. Given that the central command sets out the strategy and goals for the entire organization, influence of a external actor, often a financial backer of the group, trickles down the entire organization. This is for example the case with Hezbollah which receives large amounts of money, equipment and knowledge from Iran. In exchange Iran expects Hezbollah to support the Assad regime in Syria, even though Hezbollah experienced a lot of resistance for its actions domestically (Stevenson, 2013).
A mediator can use this to his advantage by persuading the external actor to act differently. But it might just as well complicate things if a powerful external actor has interest in the conflict continuing. A case in point: China was pressured by the international community to talk to the Sudanese government for its behavior in the Darfur region. China’s vetoing of all UN missions in Sudan was discussed in the Security council after which China eventually decided to take action herself by pressuring the Khartoum regime to stop the violence (Taylor, 2008).
Centralized organizations usually have a relatively clear command and control structure, which makes them easier to approach for a mediator (Whitfield, 2010). Reaching out to decentralized organizations is often difficult because individual cells have more power. This complicates decisions-making processes within the group and makes it harder for the
mediator to get access to the group on a regular basis. An understanding of the command and control structure of an armed group, especially the decision making processes within the group, will furthermore help the mediator decide if his or her interlocutor is a legitimate representative of the group’s leadership (Whithfield, 2010: 19). In some cases this is very clear, when the group has a recognized leader like Gerry Adams of Sinn Fein and John
Garang of the SPLA/M, but frequently this will not be the case. The 2006-2008 peace talks in Juba between the government of Uganda and the LRA are an example of a situation where the representatives of the armed group, in this case the LRA, were never completely validated (Whitfield, 2010). The recognized leaders of the LRA, amongst them Joseph Kony, refused to
15
attend the talks lead by South Sudan’s vice-president Riek Machar, because of the ICC threats to arrest them.
Shadia Marhaban, former GAM activist and international mediator, argues that “without a strict or a definite command and control, you usually do not get the right answers” (Interview Shadia Marhaban, 05-06-2015). She compared the BRN group in Southern Thailand with the GAM in Aceh. The BRN group in Thailand is a leaderless movement she said. “One person will throw a bomb and claim to represent the group and so will others” (Interview Shadia Marhaban, 05-06-2015). This complicates the formulation of an agreement and coming to terms about the negotiations because when the mediator is addressing one cell, another cell might decide to carry out an attack. Francesc Vendrell, Special Representative of the
European Union for Afghanistan, Personal Representative of the UN Secretary-General and head of the UN Special Mission for Afghanistan, added that when a group does not seem to have a clear leader, there is less reason to bring them into the negotiations in the first place (Interview Fransesc Vendrell, 08-06-2015). With the GAM in Aceh, the command and control was very strong. “You talk to the leader and everything trickles down” (Interview Shadia Marhaban, 05-06-2015).
An interesting issue raised by Shadia Marhaban which could not be retrieved from the literature, is that in many cases groups do not send their leader right away. They usually send someone a little further down the hierarchy and then when something goes wrong, they will pull out. In these cases she said, the best thing to do for a mediator is to “be patient and keep hoping that one day they will send their leaders” (Interview Shadia Marhaban, 05-06-2015). In reference to that it is important to note that it also depends on the mediating party how cooperative an armed groups is willing to be (Whitfield, 2010). If a mediator is a UN special envoy, like Alvaro de Soto was in El Salvador, it comes a great way to a groups legitimacy if they are included in the talks and therefore groups will be easier persuaded to join the
negotiations than if a mediator is representing an NGO or a smaller human rights
organization. On this note, Nicholas Haysom, as quoted by Whitfield, says that “different kinds of mediation suggest different kinds of institutions; some require resources and leverage that only governments or multilateral institutions can bring to bear, while in others powerless can be an asset, a basis for confidence and trust building” (2010: 16).
Leadership format
16
group is a group’s leadership format. Just like the centralized or decentralized structure of a group, a certain leadership format can be an indication for a mediator of who exactly to talk to. Anthony Vinci addresses several forms of authority in his 2006 article (2006b). Within the central command of an armed group different kind of leadership formats are possible:
leadership based on charisma, patrimonial ties or bureaucratic authority (Vinci, 2006b). Some groups make use of a role model or hero that is the face and spokesperson of the organization. An example of such a leader is Che Guevara, who was a revolutionary, guerilla leader and who spread Marxists ideas in Cuba, the DRC and Bolivia. In this format, it is very clear who the go-to person is. Another leadership style is more patrimonial, often based on a monetary exchange but sometimes also based on loyalty (Vinci, 2006b). When the central government of Somalia collapsed, several clans tried to take over power of the Somali state. Clans appointed a war-leader who lead a clan-based faction militia in their strive against other clans. Even though this war-leader forms the central command and control of the armed group, the group’s loyalty remains with the clan-elders who appointed the war-leader. For mediators this means they should approach the clan-elders and not the central command of the armed group. Other militiamen are the so-called mooryaan, who fight for whoever pays the most (Vinci, 2006a). They are often hired by local businessmen or warlords for gain of territory or business protection. Appeals to clan loyalty is not likely to work for either the mooryaan nor their employers, financial incentives might be more successful (Vinci, 2006a). That leadership format is important is confirmed by Alvaro de Soto. He explained how Schafik Handal was an important spokesperson for the FMLN in El Salvador, which is a partnership between five different groups with the same goal, overthrowing the Salvadorian government, of different sizes and ideologies. Handal, even though he was the leader of the relatively small Communist Party, was highly regarded by the other leaders and by the international community and therefore seen as the leader of the FMLN negotiation team. 3.2 Strategy
Once a group decides to take up arms to fight for what they stand for it has to decide how it is going to fight the opposing party. Some armed groups owe their name to their combat
strategy. Guerrilla’s are groups that try to avoid a direct confrontation with the enemy. Terrorist execute acts of violence in order to create fear and do not shun hurting non-combatants (i.e. the civilian population). Strategy often determines the amount of (civilian) casualties and can therefore be a reason for a mediator or the international community as a whole, to not engage in talks with a group. The beheading of several hundreds of people by
17
the Islamic State makes it unlikely that mediation efforts will be employed in the near future. Francesc Vendrell confirms this. He argues that in trying to reach peace he would probably not include groups that “violate basic humanitarian and human rights norms”.
The arguments with which groups justify the violence they use are an important measure for mediators to decide whether they are willing to engage in negotiations with armed groups. The FMLN for example consisted of parties that no longer believed in changing the political situation in El Salvador through elections or peaceful protests and therefore decided to take up arms (Stanley and Holiday, 1997). Even though all violence should be eschewed, there are degrees as to which one can say that the violence used is reasonable. This is a very sensitive subject however, for which this thesis is not the right place.
On the subject of the justified use of violence Alvaro de Soto said he would only talk to an organization as Boko Haram, who often rape and abuse the girls and women they capture in their attacks, if “there is a lot of evidence that they have a justification for what they are doing, because it does not appear to me that there is one” (Interview Alvaro de Soto, 28-05-2015). Francesc Vendrell said about Boko Haram and like-minded groups that there is a difference in involving them in negotiations with the government and just establishing contacts with such groups. At this time, he would not include Boko Haram or al-Qaeda related groups in Syria, like al-Nusra, in peace negotiations (Interview Francesc Vendrell, 08-06-2015). A group’s combat strategy thus influences the mediation process in such a way that it can determine whether mediators or the international community want to talk to them in the first place.
18
1. 4. Motivation
This chapter will focus on the questions: what drives these people to pick up arms and what motivates individuals to join an armed group? When focusing on the motivations of the armed group in general a distinction can be made between an individual goal and a group goal, each comes with different issues that need to be addressed and both need a different approach (Bercovitch and Kadayifci, 2009). As has been discussed in the previous chapter it is important for the mediator to decide whether the motivation of an armed group is a
justification for the violence used. Or to speak with Mr. de Soto “they want to take us back to the 7th century with 21st century means [about IS] (Interview Alvaro de Soto, 28-05-2015). 4.1 Motivation of groups
When asked what he would want to know about the groups he is asked to engage with, the first thing Francesc Vendrell said was that he would want to know what their objectives are. Why do groups fight? This is possibly the most important question that armed group scholars try to answer. What causes a group to give up their peaceful protests and take up their arms? Schafik Handal, leader of the Communist Party in El Salvador said that after two deceitful elections he no longer believed that the regime could be defeated through elections (Interview Alvaro de Soto, 28-05-2015). That is when he decided to join the armed opposition forces united in the FMLN.
Political reforms or feelings of injustice towards the political environment in a country
together with the more abstract goal of a religious ideology are the most common motivations for groups to use violence (Piazza, 2009). Islamic terrorist organization like Qaeda, al-Shabab, Boko Haram and IS have drawn media attention to Islamic religious violence in recent years, but religious violence occurred and still occurs in almost all religions. For a mediator, the goal a groups has, which is the most basic form of information one can have, can also indicate their willingness to engage into talks. Groups with political goals often evolve into a violent organization because of the injustice of, or their dissatisfaction with, the current political climate. They are usually separatist or secessionist movements active in post-colonial countries.6 In their hopes for political change they are more willing to talk and negotiate than groups with a more abstract ideological goal (Piazza, 2009). The violence that politically driven groups use is often aimed at their direct opponent, trying to defeat them or
6
19
persuade them to act differently or get them to negotiate, but using too much violence might alienate their rank and file and rule out any future talks with the government. This is different for ideologically driven groups whose support base is much more vague. In the case of al-Qaeda it is the complete Muslim community who are the beneficiaries of their actions. The tactical objective of violent attacks for them is usually more communicative (Hoffman and McCormick, 2004). The high amount of casualties with their attacks is a way to draw media attention, spread their message and show commitment to their strife. Groups with a political goal usually have a clearly defined constituency on whose behalf they execute attacks. They rely on these constituencies for support both financially and politically and using too much violence might alienate their support. This was ultimately the case with the LTTE whose methods at the end of the war alienated them from their local supporters (Asia Report nr. 186, 2010).
For a mediator it is therefore easier to approach and negotiate with armed groups that have a political objective than with groups that have an ideological goal. Also, negotiations with politically driven groups tend to be more fruitful because they rely on their constituencies much more, who can eventually tire from all the violence and settle for less than ideologically driven groups would (Piazza, 2009).
Talking to clan-elders is an important measure for mediators when dealing with clan or ethnicity based violence (Vinci, 2006c). They thereby often surpass the military leader of the group. In other cases, talking to the government of the country where the group is active might suffice. If the motivation of the group is based on injustice or unequal treatment compared to other (ethnic) groups, as was the case with the LTTE in Sri Lanka and currently with the Nagas in Northern India, persuading or pressuring the government to adjust its laws or change its behavior towards these groups might suffice instead of the entering the long process of gaining trust and engaging with these armed groups themselves. Shadia Marhaban notes however that armed groups do want to engage in talks, as a breather and to take time to regroup, but the government refuses to talk and rejects mediation efforts because it does not like the meddling of third parties in its internal business (Interview Shadia Marhaban, 05-06-2015).
The type of goal a group has can also help determine the organizational structure of a group (Piazza, 2009). Groups with a concrete political goal seem to be more centralized with a clear command and control structure formed by their non-violent past as a political opposition
20
party, while ideologically driven groups usually follow a more decentralized structure. Al-Qaeda for example, follows a rather decentralized structure with many affiliated groups all over the world. Their ambitious and highly ideological goal is to unify the Islamic world through an Islamic jihad and to rid this world of non-Muslim’s cultural and political influence (Piazza, 2009: 66). As can be concluded from the previous chapter, centralized organizations are easier to approach and to communicate with for mediators, but the decentralized structure of some organizations might also bring possibilities for the mediator as has been indicated by Manikkalingam and Policzer (2007).
Another popular categorization for the motivations of armed groups is the greed versus grievances classification. Even though this categorization has been subject to some criticism over the last couple of years, it is still frequently used by mediators in the field.7 Using violence out of greed or grievances indicates which incentives mediators should bring to the table. However the discussion in scientific literature about which kind of complaint is easier to handle for a mediator does not have a clear winner. Some say that groups acting out of greed can be easier persuaded to stop the violence if they are offered financial alternatives. An example is the MEND in the Niger Delta who wanted to be compensated for the oil that was obtained from their region (Aghedo and Osumah, 2012: 854). This argument is refuted by Collier and Hoefler who argue that frequently warfare is such a lucrative business that it is harder to persuade these greed-driven groups to stop (2004). For groups with grievances on the other hand, like the political grievances discussed above, the same problems seem to occur for the mediator. Some groups are easier to approach because they have a clear, concrete grievance with which a mediator can go to the other parties in the conflict to discuss it. But in other cases grievances are unrealistic and do not offer the mediator anything to go with, which is the case with IS.
Because research proved that the greed and grievances debate was not very useful to
formulate an effective counter policy, Vinci decided to reassess the greed versus grievances debate and come up with an alternative approach. He uses a realist approach and argues that the ultimate goal of each armed group is to survive and maintain power (Vinci, 2006c). This does not necessarily have to be violent. They may for example be incorporated into the national military. This means a mediator should look for a, often political, way in which a group can continue to exist. Alvaro de Soto said about his experiences in the El Salvador
7 See Vinci’s 2006 article “Greed and Grievances Reconsidered: The Role of Power and Survival in the
21
peace process that while one or two leaders of the FMLN might have had personal interest they would like to see fulfilled through the negotiations, the majority of the leaders needed no reminder as to what they were fighting for (i.e. political reforms) and all started out as
peaceful opposition movements thus had no trouble going back to that format (Interview Alvaro de Soto, 28-05-2015).
Vinci also underlines that it is important to understand a group’s motivation and possible differences in motivation between members of the group to formulate an effective policy response (2006b: 25). A group of which the cells have diverse motivations is Boko Haram (Aghedo and Osumah, 2012). Boko Haram is a militant organization driven by a
fundamentalist Islamic ideology. It pursues the Islamisation of Nigeria and it repudiates democracy and Western education (Aghedo and Osumah, 2012: 858). These issues have been controversial however, because some factions of the group tend to focus on local grievances and joined the group to enrich themselves, given that an important part of Boko Haram’s finances is derived from looting (2012: 859). The other faction, who preaches the true teachings of Islam, is said to receive support from al-Qaeda and get military training in the Maghreb dessert in Mauritania by AQIM-fighters (al-Qaeda in the Islamic Maghreb). This difference in motivations means that a mediator should come up with two completely different strategies in a possible mediation effort.
Besides the things mentioned above, Francesc Vendrell adds that a group’s motivation is also an indication “whether they are willing to compromise at all towards other minorities or other groups” (Interview Francesc Vendrell, 08-06-2015). In summary, the motivation of an armed group can be used by the mediator as an indication for which issues should be addressed during the negotiations and how easy or hard it might be to approach the group and persuade them to join peace talks.
4.2 Motivation of individuals
Individuals join armed groups for several reasons. The ones that are mentioned most often in the literature are ethnic ties to the group, clan-loyalty, greed, religious ties and ideological ties (Vinci, 2006c; Reno, 2007; Hofmann, 2006; Watkin, 2010). Francesc Vendrell said that “it matters for how you approach a solution if most of the militants are fighting because of their ideological commitment, or whether they are fighting purely for financial reasons. Financially meaning purely monetary gain.” (Interview Francesc Vendrell, 08-06-2015). “Clearly you
22
still have to deal with them, whichever their motivation is. But you may have to approach the solution in a different way.” (Interview Francesc Vendrell, 08-06-2015).
How does a group recruit people that want to fight for them? In his 2003 article Kalyvas explains that the goals of the leaders of a civil war can differentiate from the goals of local supporters of the war. He explains that choosing one side of the master cleavage of a civil war or the other by groups at the periphery does not necessarily depend on what is right or wrong according to these groups, it may just as well depend on a former enemy choosing the other side of the conflict. With the support of the elite, local factions may try to solve decades long feuds with other local actors (Kalyvas, 2003). This happened for example in the civil war in Syria where several groups chose the side of Assad because his regime tolerated them, while their future under a new regime was unknown.
A mediator needs to find out if the ideological goal of the elite is the same as that of the fighters at the bottom of the hierarchical ladder. If an agreement is reached where the elite groups decide to stop the violence because they resolved their issues, this may not be satisfactory to groups at the bottom who can than either decide to continue fighting on their own or align with another group who can support them in their battle, with the mediation effort than missing its purpose. Also, talking to a group leader might not always result in an agreement while individual fighters might be converted by offering them an education as was done with the child soldiers of the Cobra faction in South Sudan.8
When talking about the Taliban, Francesc Vendrell said that the reasons to join the Taliban might not be the same for every fighter. Some may be fighting for local, tribal or ethnic reasons, others may be fighting for ideological reasons and yet another might take up arms because the government is corrupt. At the same time you cannot start dealing with various groups and individual motivations. The key question is, he said, does the Taliban leadership appear to have command and control. He thereby hints that for him as a mediator, the
command and control structure of a group provides him with more important information than does motivation of the group. When there is no solid command and control, engaging with the
8
The Cobra Faction is an armed group in South Sudan led by David Yau Yau who took up arms in 2010 to defend the rights of the marginalized Murle minority in a country dominated by Dinka and Nuer tribes (Hofman, 2015). This year all 3.000 child soldiers will be released after the Cobra Faction signed a peace agreement with the South-Sudanese government in July 2014. In the Piborregion, where the Murle live, sending a child to fight for an armed group is seen as an opportunity for a better life because parents often cannot offer their children enough. Offering their children an education might persuade parents to not let their children join these groups, so the child abuse and casualties decline (Hofman, 2015).
23
group might not be effective and it may be better to look for other ways to resolve the conflict (Interview Francesc Vendrell, 08-06-2015).
Shadia Marhaban indicates that the separatist and secessionist movements she works with in Southeast Asia often have a much stronger foundation compared to groups like Boko Haram because of their feelings of injustice toward their government (Interview Shadia Marhaban, 05-06-2015). Motivations of group member are therefore likely to be the same as the motivations of the group’s leadership.
24
5. Support
The remaining characteristics of armed groups that need elaborating can be summarized under the term ‘support’. Armed groups receive support in two ways: financially and politically by a regional or international support base who identify themselves with the ideas, ethnicity or beliefs of the group or with whom they share a common enemy.
5.1 Financial Support
Like any other organization armed groups need to find a way to support themselves. Because they break the monopoly on violence held by the state armed groups are often forced to support themselves through illegal means. Kidnapping for ransom, narcotics trade and looting are important sources of income for armed groups. Most of the time however, armed groups can also find financial support in the form of a large diaspora, another armed group or a sovereign state. The Tamil Tigers of Sri Lanka were originally sponsored by their diaspora in India, while al-Qaeda supposedly supports groups like AQIM, Boko Haram, Jabhat al-Nusra and al-Shabaab in Somalia (Sengupta and Ganguly, 2013; Piazza, 2009). Al-Qaeda in turn, is said to receive support from Saudi Arabia, and Iran supports Hezbollah with arms, technology and money (Salehyan, 2007).
Below will be discussed what these different types of support mean for the mediation process. Natural resources: Some groups might make use of their surroundings and extract timber, diamonds, oil and other natural resources from their habitat. This is how the RUF in Sierra Leone supported itself (Vinci, 2006b: 56). According to smugglers and Iraqi officials, IS exports oil from its territory in Iraq out to Turkey, Jordan and Iran (Hawramy, Mohammed and Harding, 2014). IS was able to use existing smuggling routes from Iraq to the surrounding areas that were established years ago and received millions of dollars in exchange for their oil supplies.
Extortion: Groups that are relatively stationary and intend to stay in a certain area for a longer period of time might decide to extort money from inhabitants of that area by raising taxes in exchange for protection (Vinci, 2006b: 55). The SPLA/M is an example of an organization that uses this method. Boko Haram extorts money from Nigerian governors, senators and other officials. Some experts question however if all the money is extorted or if some of it might be voluntary given by members of (former) president Jonathan’s opposition.
25
Kidnapping: In theory most countries agree that it is counterproductive to pay ransom to armed groups because this finances their battle. In practice however the New York Times claims that Belgium, Denmark, France, Germany, Italy, Spain, Sweden and Switzerland all paid ransom to IS in order to free some of their citizens (Taylor, 2014).9 Kidnapping for ransom is therefore still an important source of revenue for armed groups. Boko Haram received nearly seventeen million dollar in ransom for their kidnapping activities from the French government in 2013 alone (Website Stratfor Global Intelligence).
Looting, narcotics trade and other criminal activities: Because armed groups are often forbidden or repressed by the national government they are forced to support themselves through illegal means as was mentioned in the introduction of this chapter. Weapons they use in combat can also be used for activities like looting and the lucrative narcotics trade.
Narcotics trade also provides them with a smuggling network through which all other sorts of goods, i.e. weapons, can be transported. Groups known for their large share in narcotic trade are the FARC in Colombia and Hezbollah in Lebanon (Berry, Curtis, Hudson and Kollars, 2002). Looting, robbing banks and other criminal activities brought Boko Haram more than three million dollar in 2011 (Website Stratfor Global Intelligence).
External support: Other groups might receive support from outside the territory they operate in. This can be in the form of a diaspora supporting them, which was the case for a long time with the LTTE and their diaspora in India and for the GAM in Aceh (Sanguly; Interview Shadia Marhaban). Sometimes groups are supported by other groups with similar beliefs. Boko Haram, for example, allegedly received seventy million dollars from other Islamist organizations like AQIM and the al-Muntada al-Islami Trust. Militant Islamist organizations like AQIM and al-Shabaab also provide Boko Haram with military training and the
knowledge behind the fabrication of explosives. In some cases state governments support armed groups in other countries. Even though most states officially deny the accusations, Eritrea, Iran, Saudi-Arabia, Syria, Qatar and Yemen all have been accused of providing funds for al-Shabaab (Website CFR). And Sudan sponsored the Lord’s Resistance Army in its strife against the Ugandan government in retaliation of the Ugandan support for the SPLA (Vinci, 2006b: 56).
9 Until recently only the US and the UK refused to pay ransom for their kidnapped citizens. However, on June
24th 2015 Obama announced that the US will adjust its policy towards the hostage situations of citizens. It will not change its policy of not paying ransom to armed groups, but it will engage in talks with the hostage takers. Furthermore it will help the families that want to pay the ransom themselves by facilitating negotiation talks instead of threatening them with criminal charges as is the case now (Davis, 2015).
26
What do these forms of support mean for the mediator? The ancient rule that money is power also goes for armed groups. In order to get an insight into the command and control structure of an armed group, which is very important according to Francesc Vendrell, a mediator can look at the flow of money to gather intelligence about the structure of the group if they are not willing to provide that information themselves. Figuring out where the financial support is coming from is often an indication of who is in charge. For example, groups that receive great amounts of funding from their constituency are usually more decentralized organizations (Duverger, 1959). Diasporas do not belong to the constituency of a group in this case, because their money goes directly to the leadership.
In cases where another state funds the armed group, a mediator can decide to include
representatives of that state in the mediation talks because chances are that these states have a great say in the functioning of the armed group. When trying to talk to Hezbollah, it is
important also to include Iran, because a lot of the orders of Hezbollah come from Iran (Stevenson, 2013). This is often difficult, because as has been said before, states usually do not admit that they are sponsoring armed groups. Furthermore they might have an interest in the continuation of the conflict which might even complicate negotiations. Shadia Marhaban said that what makes it difficult for the mediator to use funding in its approach to the group is that sometimes the government that the armed group opposes also profits through corrupt officers for example. This chain of interdependence is a huge challenge to break for the mediator. And in general, support in the form of money or weapons for an armed group is often not public information or information that third parties have access to.
In summary, the influence of the financial support an armed group receives on the mediation process is twofold. First it highlights if the mediator should address more parties than just the conflicting parties. And secondly, it indirectly gives the mediator information about the organizational structure of an armed group. In situations where the organizational structure of an armed group is not clear, as is often the case, following the money flow within the group might give a good indication of who is in charge (Whitfield, 2010).
5.2 Political Support
Political Support sometimes goes hand in hand with financial support. But very often, political support is a much scarcer good to find for an armed group than financial support. Political support can be derived from two sources: from actors in the international community
27
or actors in the constituency of a group. The most important debate in the scholarly literature about political support for armed groups is the legitimacy question. Legitimacy is an
important reason for states not to engage in talks with armed groups, because it can
undermine their authority and threaten their territorial integrity and give these group leverage in negotiations with international organizations (Steinhoff, 2010). When states do engage into talks with armed groups for whichever reason, i.e. the release of prisoners, persuading them to put down their arms or involving them in negotiations, it might give the group an impression that their cause is valid and so are their means (Toros, 2008). It can also be an incentive for other groups to start using violence to gain the attention of the international community. Steinhoff notes that besides these legitimate reasons not to engage with armed groups, states might also have alternative agenda’s not to engage with them (2010). Fear of losing power and a change of the status quo can be the real reason for states not to engage with armed groups within their borders (Steinhoff, 2010). For a mediator it is thus important to determine what the real reason is that a state refuses to engage with an armed group in order to formulate an effective mediation strategy.
There are different forms of engagement with armed groups however. Francesc Vendrell indicated that it is always useful to establish contacts with armed groups (Interview Francesc Vendrell, 08-06-2015) These contacts can be used either to gather information or get a better understanding of their cause. According to Toros, these ad hoc dialogues are unavoidable at times, while actual political negotiations can be counterproductive and dangerous (2008: 411). And while states claim that engaging with armed groups might undermine their authority, this rhetoric is undermined by the fact that they still communicate and work with organizations like the ICRC and other NGO’s which regularly negotiate with armed groups.
A mediator thus needs to determine what engagement with armed groups will do for the political support and legitimacy of the group and he or she needs to adjust the format of the engagement to these findings. Labeling a group as a terrorist organization can be a method from the international community to show that the actions of these groups are wrong to compensate for the fact that engagement is taking place (Toros, 2008).
The other form of political support a group can receive is from their constituency. The broader their support base, the bigger chances are that their cause is picked up by the
28
support. They can lobby in their countries of residence for political support for their cause. The large Tamil diaspora in India is a good example in this case. Afraid to lose the support of the large Tamil diaspora in Tamil Nadu and also out of fear for a secessionist movement within its own borders, the Indian government decided to support the Tamil insurgents in Sri Lanka (Sengupta and Ganguly, 2013: 112). For a mediator political support from a group’s rank and file, especially that of a diaspora, can indicate which other parties besides the conflicting parties can be approached to join the negotiations or pressure the conflicting parties in one way or another.
29
6. Conclusions
Based on the research above, several conclusion can be drawn about the influence certain characteristics of armed groups have on the mediation process. However the characteristics looked at in this thesis are not all encompassing and even though I chose them with care and consideration, in practice mediators might look at more characteristics than the ones that are discussed in the scholarly literature. The aim of this thesis was to start filling the gap in the existing research about mediation with armed groups by focusing on one aspects of
Bercovitch, Anagnoson and Wille’s Contingency Model (1991). A lot more research still needs to be done in order to completely understand how the nature of conflicting parties can influence the mediation process. One interesting question might be if the influence on the mediation process is different when the conflict consists of a state and a non-state actor versus two non-state actors. Below the conclusions from the previous chapter will be discussed. The last paragraph contains the central conclusion of this thesis.
The command and control structure of an armed group turned out to be the most important characteristic for the mediation process. Centralized organizations are easier to approach for a mediator and are more vulnerable for the influence of an external sponsor. In negotiating with centralized organizations a mediator can be relatively sure that eventual agreements are affecting the entire organization. This will not always be the case for decentralized
organizations. In decentralized organizations it is harder to identify the leadership and during negotiations a mediator cannot be sure that he or she is talking to an official representative of the group. The only effect the combat strategy of an armed group has on the mediation process is that it can be decisive in whether or not mediators are prepared to engage with the group in question.
The motivations for groups to start using violent means to pursue their goals are very divers. They can be broadly categorized as being either political or ideologically driven or based or greed or grievances. Even though the greed and grievance explanations for the violence of armed groups are extensively discussed in the literature, for the mediation process they do not seem to make much of a difference. Differentiating between political and ideologically motivated groups provides the mediator with more information. Politically driven groups are usually more centralized and thus easier to approach for the mediator. These groups
furthermore rely on a constituency that might eventually tire from all the violence that is being used which makes these type of groups more flexible in negotiations. Ideologically
30
driven groups are harder to catch, they usually do not have a clearly defined constituency for whom they have to justify their means and are thus less willing to settle in negotiations. Individual motivations to join an armed group are only of interest for a mediator in the case of decentralized organizations. In these cases the central command is not always able to
represent all members and individual motivations that come to play a role in order for agreements to be effective. A solution for these types of organizations might be to address individual cells instead of the central command.
The way in which an armed group supports itself or is supported by an external source gives the mediator an indication of how the group is organized and who have the final say in the organization. Another effect the financial support a groups receives has on the mediation process is when the group is sponsored by an external actor. This actor possibly influences, positively or negatively, the actions of the group and could be an important actor to invite to the negotiation table. The political support a group receives is important for the legitimacy of the group and might be a factor to reckon with for the mediator when deciding to engage the group in the negotiations.
In summary the command and control structure of an armed group is the most important characteristic to look at for mediators in the mediation process. It gives them information about who to address and whether or not a possible agreement will be effective. The motivation of an armed group is important for mediators because it tell them which issues should be addressed and what incentives might persuade groups to put down their arms. The other characteristics generally only have an indirect influence on the mediation process.
31
7. References
- Aghedo, I and Osumah, O. (2012) The Boko Haram Uprising: how should Nigeria
respond? Third World Quarterly 33 (5): pp. 853-869.
- Alberts, D. (2007) Agility, Focus and Convergence: The Future of Command and
Control. The International C2 Journal 1(1): pp. 1-30.
- Alberts, D. and Hayes, R. (2006) Understanding Command and Control. Command
and Control Research Program (CCRP) Washington, DC.
- Asia Report nr 186 (2010) The Sri Lankan Tamil Diaspora after the LTTE.
International Crisis Group.
- Beardsley, K. (2008) Agreement Without Peace? International Mediation and Time
Inconsistency Problem. American Journal of Political Science 52 (4): pp. 723-740.
- Bercovitch, J. (1986) International Mediation: A Study of the Incidence, Strategies
and Conditions of Successful Outcomes. Cooperation and Conflict 10 (11): 155-168.
- Bercovitch, J. (1991) International Mediation and Dispute Settlement: Evaluating the
Conditions for Successful Mediation. Negotiation Journal 7 (1): pp. 17-30.
- Bercovitch, J. (2006) Mediation Success or Failure. A Search for the Elusive Criteria.
Cardozo Journal of Conflict Resolution 7: pp. 289-302.
- Bercovitch, J. and Foulkes, J. (2012) Cross-Cultural Effects in Conflict Management:
Examining the Nature and Relationship between Culture and International Mediation. International Journal of Cross Cultural Management 12 (1): pp. 25-47.
- Bercovitch, J. and Houston, A. (2000) Why Do They Do it Like This? An Analysis of
the Factors Influencing Mediation Behavior in International Conflict. Journal of Conflict Resolution 44 (2): pp. 170-202.
- Bercovitch, J. and Kadayifci, A. (2009) Religion and Mediation: The Role of
Faith-Based Actors in International Conflict Resolution. International Negotiation 14: pp. 175-204.
- Bercovitch, J. and Langley, J. (1993) The Nature of the Dispute and the Effectiveness
of International Mediation. The Journal of Conflict Resolution 37 (4): pp. 670-691.
- Bercovitch, J. and Schneider, G. (2000) Who Mediates? The Political Economy of
32
- Bercovitch, J., Anagnoson, J. and Wille, D. (1991) Some Conceptual Issues and
Empirical Trends in the Study of Successful Mediation in International Relations. Journal of Peace Research 28 (1): pp. 7-17.
- Berry, L. Curtis, G. Hudson, R. and Kollars, N. (2002) A Global Overview of
Narcotics-Funded Terrorist and Other Extremist Groups. Library of Congress – Federal Research Division
- Blin, A. (2011) Armed Group and Intra-State Conflicts: the Dawn of a New Era?
International Review of the Red Cross 93 (882): pp. 287-310.
- Chamberlin, P. (2012) When it Pays to Talk to Terrorists. New York Times,
September 3.
- Collier, P. and Hoefler, A. (2004) Greed and Grievances in Civil War. Oxford
Economic Papers 56: pp. 563-595.
- Cronin, A. (2006) How Al-Qaida Ends: The Decline and Demise of Terrorist Groups.
International Security 31 (1): pp. 7-48.
- Davis, J. (2015) In Hostage-Terrorist Policy Shift, Obama Admits Failures. The New
York Times 24 June. (http://www.nytimes.com/2015/06/25/world/middleeast/obama-softens-policy-on-hostage-negotiations-with-terrorist-groups.html?ref=us&_r=0)
- Dudouet, V. (2010) Mediating Peace with Proscribed Armed Groups. United States
Institute of Peace Special Report 239.
- Duverger, M. (1959) Political Parties. Wiley and Sons, New York: p. 58-59. - Favretto, K. (2009) Should Peacemakers Take Sides? Major Power Mediation,
Coercion, and Bias. The American Political Science Review 103 (2): pp. 248-263.
- Hawramy, F., Mohammed, S. and Harding, L. (2014) Inside Islamic State’s Oil
Empire: How Captured Oilfields Fuel Isis Insurgency. The Guardian, 19 November. Available from: http://www.theguardian.com/world/2014/nov/19/-sp-islamic-state-oil-empire-iraq-isis. [Last accessed on 23-06-2015].
- Henderson, D. (1996) Mediation Success: An Empirical Analysis. Ohio State Journal
on Dispute Resolution 11 (1): pp. 105-147
- Hoffman, B. and McCormick, G. (2004) Terrorism, Signaling and Suicide Attack.
Studies in Conflict and Terrorism 27: pp. 243-281.
- Hofman, L. (2015) Eindelijk zijn deze kindsoldaten vrij (alleen nog niet van hun
verleden). de Correspondent 16 April.
- Hofman, L. (2015) In Zuid-Soedan is Vrede, Vrijheid en Democratie vooralsnog
