1

Mediation and Civil Wars Involving Terrorism

Paulina Pospieszna1 University of Konstanz

Department of Politics and Management Paulina.pospieszna@uni-konstanz.de

Karl DeRouen Jr.

The University of Alabama, USA Department of Political Science

Paper prepared for presentation at the ECPR Joint Sessions 2011 Political Institutions and Political Violence Workshop

the University of St. Gallen, Switzerland 12-17 April, 2011

**Preliminary Draft, please do not cite or quote without permission from the authors**

Abstract The relationship between mediation and terrorism is likely to be non-recursive. Mediators are more likely to be involved in costly, intractable civil wars such as the type involving terror against civilians. In turn, mediation can impact the use of terror. Mediation might decrease the rate of terrorism if it successfully offers alternatives to waging war. However, mediation could lead to more terror if the process is exploited by the type of spoilers identified by Stedman. Thus terror attracts mediation which in turn leads to more terror. This two-way relationship is explored using a special modeling technique developed by Maddala and programmed by Keshk that can accommodate a dichotomous endogenous variable. Our model demonstrates that mediation and terrorism are related—terror attracts mediation but mediation increases the number of fatalities during the rebel terror act. Introduction

The prevalent form of armed conflict in the world is civil war. The vast civil war literature has mostly focused on conflict onset (see Fearon and Laitin 2003; Collier et al. 2003), outcome (see DeRouen and Sobek 2004), duration (see Fearon 2004), peace building (see Doyle and Sambanis 2000), and level of violence (see Kalyvas 2006; Krueger and Laitin 2008; Lacina 2006). However, the analysis of terrorism until recently was largely detached from the empirical study of civil war.

According to Krueger and Laitin (2008), the majority of countries with civil wars experience

terrorist incidents. Terrorism can be either a precondition or a cause of civil war, or both (Boulden 2009; Sambanis 2008). According to Sambanis (2008), if we consider terrorism as a strategy in civil war then the links between civil war and terrorism are obvious: civil wars create opportune environments for terror and terrorists.

1 Pospieszna would also like to thank the Fritz Thyssen Foundation for the support.

2

Given the importance of mediation as a means of managing civil violence and the threats that

internal conflicts pose to regional and global security, mediation of civil wars is receiving increasing attention from practitioners and scholars. However, while mediation in internal conflicts has been studied extensively, few scholars have examined mediation as a possible factor reducing the use of terror in civil war, thus leaving us with underdeveloped answers to important research questions about the impact of mediation on conflict intensity. This paper aims to fill this gap and by investigating the link between mediation and terrorism during the internal conflict.

The relationship between mediation and terrorism is likely to be non-recursive. Mediators are

more likely to be involved in costly, intractable civil wars (Bercovitch and Gartner 2006) such as the type involving terror against civilians. In turn, mediation can impact the use of terror. It might decrease the rate of terrorism if it successfully offers alternatives to waging war. However, mediation could lead to more terror if the process is exploited by the type of spoilers identified by Stedman (1997). Thus terror attracts mediation which in turn leads to more terror. This two-way relationship is explored using a special modeling technique that can accommodate a dichotomous endogenous variable developed by Keshk (2003).

This study proceeds as follows. First, we report findings from studies on terrorism in internal

conflict and review the literatures on mediation in civil wars. Second, we present descriptive statistics and data from two case studies. This is followed by research design, findings and conclusion.

Mediation and Terror within the Context of Civil War

Civil wars and terrorism are tightly linked and terrorism. Terrorism within the context of civil war is defined as a subset of violent strategies, or threat of use of violence, that can be used for political ends—e.g. to achieve national independence, influencing government policy, or overthrow the government (Lake 2002; Sandler 2003, 780). A recent study of Findley and Young (2011) shows a high degree of overlap between terrorism and civil war: the most incidents of terrorism take place in the geographic regions where civil war is occurring and during the ongoing war.

Terror can be used by both governments and rebels. Governments can use terror as a substitute for fighting if traditional combat is not possible. Since rebels rely on the population for support, the government can use terror to weaken the insurgency by killing civilians (Azam and Hoeffler 2002). However, terror is more likely to be used by the insurgents—rebels tend to be more violent on the whole, while governments commit relatively little violence except in those few years which see mass killings (Eck and Hultman 2007, 233).

There are several sometimes contradictory explanations of the use of terror by rebels. On one hand, rebel groups will be less likely to target civilians because they are dependent on the civilian population for resources, logistical support, or funding (Zahar 2001, 112). On the other hand, rebels will be more likely to kill civilians, because by punishing the broad population, extremists seek to provoke a response, to secure compliance, to demonstrate how determined they are (Kalyvas 1999; Kalyvas 2004; Kalyvas 2006), and to mobilize additional recruits and financial support (Humphreys and Weinstein 2006; Lake 2002). Wood (2010), however, argues that insurgents use violence against civilians to encourage support—weak insurgent groups have an incentive to target civilians because they lack the capacity to provide sufficient benefits to entice loyalty. Among other explanations, scholars suggest that

3

terrorism is used as a strategy to counter government strategies and to improve rebels’ bargaining position with the government (see Hultman 2007; Lake 2002). Rebel groups adopt extreme methods especially under time pressure or when are losing battles in order to compensate for their political weakness, and to impose extra political and military costs on the government.

Another argument is that violence allows rebels to fulfill a variety of functions related to the war. Kirk (1983), for example, offers a rent-seeking model of terrorism where violence becomes a means of generating wealth. Pape (2003) provides another model. He reports that suicide terrorism during civil wars follows a strategic logic, i.e. it is designed to achieve specific political purposes. That is to say rebels have used terror to compel an enemy to withdraw, e.g. to force Israel to abandon the West Bank and Gaza in 1994 and 1995, or to coerce modern democracies to make significant concessions, especially to eject foreign forces (e.g. Al Qaeda pressuring the United States to withdraw from the Saudi Arabian Peninsula).

Recent studies also explore terrorism in civil war in greater detail. Some scholars, like Borooah (2009) explores and compares the severity of terrorist incidents initiated by different rebel groups, others explore the possible consequences of domestic terrorism. For example, Lozano-Gracia et al. (2010) and Ibáñez and Velásquez (2009) find that violent conflicts in Colombia can play a role in increasing intractable conditions, such as the flows of households to other regions or migration, which poses new challenges to policymakers. Scholars also find that domestic terror can undermine stability not only in a country but in a region and lead to the internationalization of the war (see Bapat 2007).

Given the negative consequences of terrorism one may expect that the likelihood of mediation occurrence can be greater in countries where terror is severe—mediators may intervene in order to prevent the spread of violence. However, we should also bear in mind the impact of terrorism on mediation. Before we go any further in making any assumptions between mediation and terrorism in civil wars it is instructive to present the research on mediation and reasons for mediation incidences.

Mediation is a form of third-party intervention in civil war. A mediator can be an individual, group , state, or an organization that uses different strategies to facilitate peace between warring parties. However, compared to other forms of third-parties involvement, like military intervention, mediation is non-coercive engagement that does not resort to physical force in order to resolve differences between disputants (Bercovitch, Anagnoson and Willie 1991, 8). The occurrence of mediation incident requires both mediators’ willingness to intervene and disputant acceptance of mediation. Strategies are more likely to include communication and facilitation tactics like channeling information to the parties or bringing both parties together with some control over the conflict management process (e.g. timing or agenda-setting) and logistics (Bercovitch and DeRouen 2004).

In studying the characteristics of mediators that may make them more likely to intervene in the conflict, scholars also take into account political (alliance) and economic (trade) similarities between potential mediators and disputing states (Crescenzi et al. 2008; Werner and Lemke 1997). Scholars find that third party conflict management occurs more frequently when the potential mediator is democratic, powerful state and when countries of mediators and disputants share memberships in international organizations, when there is a short distance, close trade relationships, and alliance ties between potential mediators and conflicting parties.

Some studies categorize the driving motives behind mediation (Bercovitch and Schneider 2000; Mitchell 1988; Touval and Zartman 2001). Scholars stress that mediators derive many benefits from

4

mediation, such as international prestige, a boost in public opinion, influence over the disputing states, and the stability of economic and security ties, which gives potential mediators a strong incentive to position themselves as neutral and impartial. Svensson (2007), however, believes that a mediator may never be truly unbiased, thus he focuses in his research on biased mediators toward government and rebels, and nonbiased mediators who are not necessarily neutral in relation to the issue at stake, which is a requirement for unbiased mediators. Critics of the Norwegian mediation efforts in Sri Lanka, for example, have argued that Norway was biased towards the LTTE (Weerasinghe 2005) and that the Tamil diaspora in Norway had undue influence (Rovik 2005).

The warring parties are more likely to accept or even request an outside assistance if mediators can offer mutually beneficial solution, an alternative that is more attractive than a continuation of the conflict (Zartman 2000). However, non-state groups involved in conflict can also seek to involve mediators, because the process may grant them international recognition and legitimacy (Zartman 1995).

The community of civil war researchers investigates whether mediation is a promising avenue of dealing with or terminating civil wars. Defining mediation success based on whether mediation leads to the conclusion of an agreement or not, scholars find that mediation can be a useful conflict management mechanism ( see Beardsley 2008; Bercovitch, Anagnoson and Wille 1991; Bercovitch and DeRouen 2004; Bercovitch and Diehl 1997; Bercovitch and Langley 1993; Bercovitch and Schneider 2000; Gartner and Bercovitch 2006; Kleiboer 1996; Ott 1972; Regan and Stam 2000; Walter 2002; Werner and Yuen 2005; Wilkenfeld et al. 2005). The signing of political agreements represent consent on difficult to negotiate items and is considered to be a breakthrough in the conflict management, because often leads to complete resolution of conflict. However despite strong evidence that mediation helps end civil war, the use of terror makes a negotiated settlement more difficult (see Cronin 2009).

Scholars argue that terrorism is used to “spoil the peace”—to destroy ongoing peace negotiations (Cronin 2010; Kydd and Walter 2002; Kydd and Walter 2006; Stedman 1996). Stedman (1996, 369-71) argues that some leaders and parties may believe that peace can threaten their power and interests and thus use violence to undermine attempts to achieve it. Peace creates spoilers because it is rare in civil wars for all leaders and factions to see peace as beneficial. Spoilers use violence to attack the peace process and as an example may serve the Committee for the Defense of the Revolution (CDR) in Rwanda that used violence to prevent the implementation of the Arusha Accords.

Kydd and Walter (2002) find that if terrorist attacks occurred during negotiations talks it significantly reduced that number of treaties that was put into effect. Cronin (2010) discusses the conditions under which mediation may be more successful in ending the violence. Negotiation talks facilitated by the third parties can be promising when inter alia further violence in counterproductive, the popular support is waning, terror groups are losing ground in the conflict, there are strong leaders on both sides, and when negotiation talks are unmarred by spoilers.

Taking into account the presented research on terrorism and mediation in civil wars we may argue that mediation is more likely to occur in violent wars in order to protect civilians or to prevent the spread of terrorism. We may also expect that third party intervention will reduce terrorism, but mediation might also lead to greater terror because of spoilers who uses terror to destroy peace negotiations. Our research aims to establish the link between mediation and terrorism in civil wars, that has not been studied yet. An answer to these questions will enhance our knowledge about the impact of mediation, and will also send an important message to practitioners.

5

Empirical Patterns in Mediation and Terrorism

The study explores the link between mediation and terror acts in civil wars using three datasets: the UCDP Conflict Termination dataset, Civil War Mediation dataset, and the University of Maryland's Global Terrorism Database (GTD). To our knowledge this is the first study of its kind.

The UCDP Conflict Termination dataset (Version 2.1., 1946-2007) defines civil war as a contested

incompatibility that concerns government or territory where the use of armed force between two parties, of which at least one is the government of a state, results in at least 25 battle-related deaths.

Since insurgents are more likely to use terror and because they are more violent on the whole,

the study will focus on the terror incidents initiated by rebels. Data on terror acts come from the University of Maryland's Global Terrorism Database (GTD), 1970-2008. GTD offers the ability to generate data that correlates domestic terrorism with civil war, but not used by scholars for the analysis. There are other databases, such as the Memorial Institute for the Prevention of Terrorism (MIPT) Terrorism Knowledge Base and the International Terrorism Attributes of Terrorist Events (ITERATE) database, but they deal with international terrorism. More recently, there has been created UCDP data on one-sided violence from the Uppsala Conflict Data Program which provides yearly counts of the number of unarmed victims killed by government and rebel troops (Eck and Hultman 2007; Kreutz 2008) but the dataset lists terror acts only from 1989, and it provides less detailed information about the violent incidence as compared to GTD. GTD data includes specific dates of terror acts which allowed for precise count of direct and intentional killing of civilians during the civil war. The dataset also identifies the attack type, weapon used, target type, and number of fatalities, in addition to the location and the name of the group.

It should be noted that GTD dataset contains all terror events including those not related to civil

war. Also, some terrorists events within a country engaged in a civil war may not be related to the war with the particular rebel group. Therefore, we carefully selected terror acts that are related to a given civil war episode with a specific rebel group.

Finally, in order to determine linkages between civil war, terrorism and mediation this study also

relies on the Civil War Mediation (CWM) dataset (DeRouen, Bercovitch, and Pospieszna 2011), which provides detailed information about mediation efforts between 1946 and 2004 for each civil war episode identified by the Uppsala Armed Conflict Termination data. Because datasets provide different time ranges, we have created we dropped all wars that began and ended before 1970. As a result the dataset contains 203 civil war episodes that occurred between 1970 and 2004.

Almost 35% of all civil wars in our dataset were mediated by third parties. Our data confirm that

rebel terror is frequent in civil wars, because 120 civil wars (59%) have experienced rebel terror. From all episodes with rebel terror, there are more non-mediated than mediated civil war episodes (53% non-mediated and 47% mediated civil wars with rebel terror), and this proportion translates into the number of terror acts—non-mediated civil wars experience more terror acts than mediated, but the difference is not that significant ( 10452 and 9733 terror acts respectively) . However, if we compare a number of episodes with and without rebel terror within mediated and non-mediated civil wars, we notice that almost 80% of mediated conflicts experience terror acts from rebels as figure below demonstrates, whereas in case of non-mediated civil wars there as much violent as non-violent non-mediated civil wars. This finding suggests that mediation and terrorism during civil wars appear to be correlated. However, we cannot rush to judgment in terms of the direction of causality. Is mediation more likely to

6

take place in countries where terror is frequently used or is terror more likely to occur when mediators intervene in the conflict? Figure 1 . Mediated and Non-mediated Civil Wars and the Use of Terrorism, 1970–2004

Source: CWM, UCDP Conflict Termination dataset 2.1.

Breaking down rebel terror acts by type of conflict gives other interesting results as table below

demonstrates. Whereas in case of governmental wars, there are more terror acts committed in non-mediated than mediated ones, in case of territorial wars the trend is reverse. There are more terror acts occurring in mediated territorial wars than in non-mediated ones.

Table 1. Breakdown of Rebel Terror Acts by Type of Conflict, 1970–2004 Mediated Non-mediated

Territorial conflict

4193 2677

Governmental conflict 5802 6978

Source: CWM, UCDP Conflict Termination dataset 2.1.

There are six rebel groups that exhibit the greatest number of terrorist acts and Table 2

summarizes data for those cases. Of all these rebel groups, only Sendero Luminoso from Peru, which was the most violent rebel group in terms of number of attacks used and fatalities causes, and the Kurds in Turkey, have not negotiated with the government with help of mediators.

48%

79%

52%

21%

Non-mediated

Mediated

Rebel Terror

No terror

7

Table 2. The Most Violent Rebel Groups, 1970-2004 Rebel Groups

Country Conflict Dates Conflict Type Mediation Terror Acts

Fatalities Major Attack Type

Sendero Luminoso

Peru 8/15/1981- 12/31/1999

Governmental NO 4402 11582 Bombing/Explosion

LTTE, EPRLF, TELO

Sri Lanka 8/11/1984- 12/24/2001

Territorial YES 1048 8695 Armed Assault

FMLN El Salvador 1/1/1979- 12/31/1991

Governmental YES 3326 8458 Armed Assault

Contras/FDN Nicaragua 12/1/1981- 10/15/1989

Governmental YES 900 7795 Armed Assault

PKK/Kadek/KONGRA-GEL

Turkey 8/15/1984- ongoing

Territorial NO 1001 3518 Armed Assault

PIRA United Kingdom

8/1/1971- 12/15/1991

Territorial YES 1958 1516 Bombing/Explosion

Source: CWM, GTD, UCDP Conflict Termination dataset 2.1. We decide to choose two examples of civil war episodes—in Sri Lanka and El Salvador in order

to demonstrate the relationship between the occurrence of mediation incidence and terrorism during the civil wars. Does mediation makes a difference in these violent wars??

The Liberation Tigers of Tamil Eelam (LTTE) rebels of Sri Lanka were the second violent after

Sendero Luminoso and the most violent of all rebel groups in territorial civil wars. Almost 36% of the deaths from terror in territorial civil wars were due to LTTE bombings/explosions. LTTE fought to compel the Sri Lankan government to accept an independent Tamil homeland in the north and the east of the island. Several third parties were able to attract LTTE rebels to participate in mediation. The mediators included leaders of Indian government (Prime Ministers Indira Gandhi and Rajiv Gandhi, Foreign Secretary Romesh Bhandari), Foreign Monitors from Canada, Netherlands, Norway, and finally leaders and representatives of the Norwegian government. 2

As early as February 2000, Norway was asked to mediate by both sides, and initial international

diplomatic moves began to find a negotiated settlement to the conflict.3 The conflict came to the end in December 2001, when the LTTE announced a temporary ceasefire and pledged to halt all attacks against government forces. Later, at the beginning of 2002 the two sides formalized a Memorandum of Understanding and signed a permanent ceasefire agreement. Norway (Deputy Foreign Minister Vidar Helgesen, Norwegian peace negotiator Erik Solheim, and Minister of Foreign Affairs Thorbjoern Jagland) played an important role in monitoring the ceasefire and bringing the two parties to the negotiating table also when a significant setback with resurgence of violence took place in 2003.4

2 See for example Keesing's Record of World Events pages: K32630; K33098; K34383, K34875 K36282 K43361

K44157 3 About Norwegian mediation see Moolakkattu (2005) and Bullion (2001) and newspaper releases, such as BBC

News 2000, BBC News 2001a, BBC News 2001b, Frontline 2001, Guardian (London) 2002, Reuters 2001, Tamil Guardian 2001, The Sunday Times (London) 2001. 4 Keesing's Record of World Events pages: K44614, K45194, K45244, K44983, K45292, K45093, K45144

8

Figure 2. Mediations and the Use of Terror by LTTE rebels in Sri Lanka

Source: CWM, GTD.

The graphs above show mediation and LTTE terror acts over time and we can observe that when mediation activities increase, the number of terror acts decreases and vice versa. This finding suggests that lack of neutrality in Norwegian mediation efforts in Sri Lanka in fact had a negative effect on the use of terror by LTTE—the total number of terror attacks after Norwegian intervention into peace negotiations significantly dropped as graphs below show.

The situation however, looks different in case of a civil war episode in El Salvador, which was a

war over the control of government. There was no difference in the use of terrorism by the Farabundo Marti National Liberation Front (FMLN) rebels between years when mediators did intervene (1661 FMLN terror acts) and in years when mediators did not intervene (1665 FMLN terror acts).5 The conflict, and rebel terror, eventually ended with singing of a preliminary cease-fire agreement in the end of 1991 and a comprehensive peace settlement (the Chapúltepec Accord) two weeks later. However, the question remains why FMLN rebels used terror against civilians even in 1990 and in 1991 when both parties jointly declared their desire to end the war and appealed to UN Secretary-General Javier Pérez de Cuéllar for help in mediating negotiations? 6 Why despite mediations the violence occurred?

Both case suggest that we can obtain slightly different findings regarding the relationship between mediation and terrorism. The Sri Lankan case shows that there is a small negative relationship between mediation and number of terror acts, but we still cannot be sure whether it is mediation that ameliorates the use of terror by rebels in this conflict or maybe terrorism that discourages mediation. In case of El Salvador, however, we cannot be sure whether mediation triggered violence or maybe mediation encouraged rebel terrorism. Because these cases were not helpful in establishing the relationship between mediation and terrorism we need a more detailed analysis that examines whether type of war matters and whether mediation reduces terror in civil wars. Moreover, problems with interpretation of results have demonstrated we have to treat mediation and terrorism simultaneously as two mutually dependent variables that affect each other and in our statistical analysis in the next section we take this into account.

5 The Farabundo Marti National Liberation Front was a coalition of five leftists and communist revolutionary

groups 6 Other mediators involved in peace talks were Álvaro de Soto acting as de Cuéllar's Special Envoy, representative

of the Roman Catholic Church— Arturo Rivera y Damas, Archbishop of San Salvador, and the representatives of the American government (Dunkerely 1994; UPI 1990a and 1990b).

0

1

2

3

4

5

19

80

19

82

19

84

19

86

19

88

19

90

19

92

19

94

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

Me

dia

tio

n in

Sri

Lan

ka

0

20

40

60

80

100

120

140

160

19

80

19

82

19

84

19

86

19

88

19

90

19

92

19

94

19

96

19

98

20

00

20

02

20

04

20

06

20

08

20

10

LTTE

Te

rro

r

9

Research Design

Variables

In line with the most recent studies in both literatures on mediation and violence in civil wars we will use several control variables, such as war duration, internationalization of war, regime type, incompatibility type, and Cold War. Since we are not trying to explain the onset of the conflict, we will not incorporate possible causes for outbreak of conflict, such as ethnicity, religion, or economic situation. Lacina’s (2006) study on why some civil wars are much deadlier than others, demonstrates that the determinants of conflict severity are quite different from those for conflict onset. Thus, following recent research on one-side violence and mediation we find it appropriate to investigate the relationship between mediation and the use of terror acts by rebels, together with the following independent variables:

Regime Type. We use the average score of ‘polity2’ variable from Polity IV dataset as an indicator of the country’s regime type. The "Polity Score" classifies a regime using a 21-point scale ranging from -10 (hereditary monarchy) to +10 (consolidated democracy).It is commonly known that regime type influences the onset of conflict, and conflict severity (Collier and Hoeffler 2000; Fearon and Laitin 2003; Gurr 2000; Harff 2003; Muller 1985; Ron 2001; Bohara, Mitchell, and Nepal 2006). The democratic peace literature states that democracies are not only more peaceful in their international relations but also in their domestic politics than non-democracies (Brown, Lynn-Jones, and Miller 1996; Doyle 1983; Oneal and Russett 1999; Ray 1995; Russett and Oneal 2001). Scholars find that democracies not only do not go to war with one another, but also are better able to resolve their bargains peacefully (conflict is resolved through negotiations and compromise) at home and thus are less subject to experience civil wars and other domestic disturbances including revolutions, guerilla warfare, killing, torture, imprisonment and the like, than authoritarian states (Diamond 1995, 6-7; Rummel 1997, 85). Democratic governments are unwilling to use the strict measures against rebels or to cause great loss of life among civilians because of development of a democratic culture and norms, and restraints on decision makers through the mechanism of competitive elections (Rummel 1995, 4).

However, regime type influences not only the selection of violence by government but also by opposition in a democratic state. It is believed that the use of violence by the opposition is less severe in democracies because of high political participation and better channels of communication between the government and opposition. According to social movement scholars high participation builds social capital and narrows the policy window for political violence and thus the insurgents’ use violence (Tarrow 1994; McAdam 1982). Studying one-side violence Lacina (2006) finds that democracy is most strongly correlated with fewer deaths in a civil war. On the other hand we should bear in mind research that show that democracy’s effect on terror incidents can be unclear because many democracies are targets of terror (see Pape 2003). As an example may serve the use of terror acts by PIRA in the United Kingdom, as Table 2 shows.

In case of the impact of regime type on mediation, the effect is not entirely clear-cut. Scholars

argue that democracy should increase chances of mediation because democracies are more amenable to peaceful resolution of disputes (see DeRouen and Goldfinch 2004; Dixon 1994; Greig 2005). Wallensteen et al. (2009) report that some level of democracy is required before substantive mediation incident can occur. However, Savun (2009) find that democracy had a negative impact on the probability of mediation in interstate war.

10

War Duration. Using the episode start and end date from the Uppsala Armed Conflict Termination dataset we have calculated the number of days the war lasted. For ongoing wars we calculated duration of conflicts based on end of dataset, i.e. 12/31/2004. One of the most stable findings is that the duration of conflict affects the severity of conflict. A long-standing conflict manifests itself in cyclical patterns with frequent bursts of violence (Coleman, 2000; Putnam and Wondolleck 2003; Regan 2002). Such conflicts perceive to be destructive, they bear devastating financial costs and cause moral and physical harms and can motivate terrorist activity.7 Taking into account these findings we find it appropriate to link duration of civil wars to the number of terror acts in civil wars.

Incompatibility. Coded as 1 if the war concerns autonomy or secessionist claims and 0 otherwise. In defining the type of incompatibility we rely on the data provided by the UCDP Conflict Termination dataset 2.1. The literature on civil wars makes an important distinction between territorial and governmental types of incompatibility. Governmental conflicts concern the type of political system, the replacement of the central government, or the change of its composition, while territorial conflicts are secessionist or autonomy-seeking wars. Some of the most highly visible civil wars, such as Sri Lanka, Northern Ireland, Israel, and Spain concern the status of a specified territory, often regarding demands for autonomy or secession. In general, scholars find that territorial wars last longer, are intractable and are more likely to recur than wars over control of government (DeRouen and Bercovitch 2008).

Moreover, territorial wars end much less conclusively, i.e., they are less likely to end with peace agreements or military victory than government wars. Fearon (2004) reports on the intractability of territorial wars. However, at the same time territorial wars are also more likely to be mediated than governmental wars (DeRouen, Bercovitch, and Pospieszna 2011). Eck and Hultman (2007, 240) find that for rebel actors half of the fatalities are in governmental conflicts, with only slightly fewer in territorial conflicts; however for government actors, 99% of fatalities are in governmental conflicts. Their data also show that rebels kill almost six times more civilians than governments in territorial conflicts (Eck and Hultman 2007, 241). There is a good reason to expect that territorial conflicts, which are often underlain by ethnic divisions, are more violent, i.e. that they have more terrorism causalities than governmental ones. However, as descriptive statistics part already demonstrated, the most violent non-state groups exist in internal conflicts over the control of government. It also is reasonable to expect that territorial wars are more likely to be mediated because of their intractable nature.

Internationalized war. This variable is coded as 1 when the UCDP Conflict Termination dataset identifies a conflict as a civil war with intervention from other states. On one hand, the civil internationalized with the presence of foreign troops may be more violent. As Pape (2003) reports suicide terrorist campaigns have been used to compel foreign military forces to abandon a country, thus we may also expect that any foreign intervention may increase terror attacks directly against the foreign military forces in the country. On the other hand, it is reasonable to expect that these wars should be less violent, because the third party military involvement may derail the use of terror by rebels. Mediation may be expected to be less likely in internationalized internal conflicts as external actors see that the presence of armed foreign allies takes their role. On the other hand we may also expect that mediation is more likely to accompany the third party military involvement as this intervention increases the changes of successful mediation.

Stratum. Within each dyad we counted stratum of conflict, to reflect if the termination occurred during the first war, second war, third war, etc . Scholars find that successive wars are not independent and the previous events could shape the outcome (DeRouen, Bercovitch, and Wei 2009). DeRouen, Bercovitch,

7 For a more in-depth discussion on the definition and meaning of the term intractable conflict, see Kriesberg,

et al., 1989; Coleman 2000; Crocker, Hampson and Aall, 2004; Putnam & Wondolleck, 2003.

11

and Pospieszna (2011) find negative effect of previous war on the likelihood of mediation which may suggest that mediators may be more reluctant to intervene in a civil war that recurred. We will control for this variable in order estimate its effect on mediation only.

Cold War. Coded as 1 for 1991 and beyond and 0 otherwise. This variable is used to test its impact of mediation. We may expect that because of super powers’ backing during the Cold War the civil wars were more likely to be mediated. On the other hand, the end of the Cold War could create more opportunities for mediation offered by different entities.

Method As we mentioned earlier we expect that mediation and terror affect each other—the number of

people killed by terror can influence the chance of mediation and the presence of mediation should reduce terror incidents but may not due to spoilers. Since we have continuous (log killed) and dichotomous (mediation) endogenous variables and we hypothesize that they simultaneously determine each other, we employ two-stage probit least squares estimation method described by Maddala (1983) and programmed by Keshk (2003; see also Keshk, Pollins, and Reuveny 2004). Violating the recursiveness (causality in only one direction from the independent variable to the dependent variable) assumption can lead to flawed inferences. We use log of fatalities because of extreme outliers in the measure.

Findings

The results are reported in Table1. We find below that terror attracts mediation but that mediation increases the number of fatalities during the rebel terror act. In terms of control variables, we find that war duration has a positive effect on terrorism causalities. If the war is internationalized there are fewer killed, which may indicate that terror is less likely in internationalized war though examples of wars with third party military presence, like Afghanistan, may seem to contradict this finding. Contrary to our expectations we find that democracies do not experience less violence from rebels’ side during the civil war than authoritarian states, but this this finding is not statistically significant. Also, we find that terrorism in territorial wars is not deadlier than in governmental wars. The negative coefficient indicates that there are less terror acts in territorial wars. We created a dummy for Africa, because this is where most terror acts are and most territorial wars occur there. However, surprisingly we obtained a negative coefficient on log killed, which suggests that African countries are less likely to experience terrorism causalities in civil wars.

In the second equation, the signs of all control variable coefficients generally match those

reported in the mediation literature. Mediation is more likely if it’s a territorial war, post-Cold War, and internationalized and these findings are statistically significant. However, the regime type of the country experiencing the war does not seem to play a consistent role in mediation. Also, a negative coefficient of stratum indicates that mediation becomes less likely with successive wars. The more rounds of violence a conflict exhibits, the more reluctant are mediators to intervene. This seems to contradict the expectation that the more complicated and intractable civil wars are, the more likely they are to be mediated (see Bercovitch and Diehl 1997; Bercovitch and Gartner 2006; Bercovitch and Jackson 2001). However, this finding relates to recent studies by DeRouen, Bercovitch, and Pospieszna (2011).

12

Table 3. Two-stage Probit Least Squares Model: Mediation and Terrorism

-------------------------------------

Logkilled

-------------------------------------

Mediation 1.568**

(.791)

Territorial war -.886

(.602)

Internationalized -2.309*

(1.208)

Africa -.796*

(.467)

Polity2IVavg .007

(.020)

War Duration .0004**

(.0002)

Constant 3.378**

(.967)

-------------------------------------

Mediation

-------------------------------------

Logkilled .228**

(.067)

Territorial war .436**

(.223)

Internationalized 1.203**

(.400)

Polity2IVavg -.005

(.010)

Cold War .395

(.250)

Strata -.067

(.151)

Constant -1.431**

(.312)

------------------------------------

Main entries are coefficients and numbers

in parentheses are corrected standard

errors; * sig. at .10, ** sig. at .05, two-

tailed tests

As expected, we find that the more violent the war becomes the more likely it is to attract

mediators, but at the same time a mediation incidence increases causalities in rebel terror acts. There can be different interpretation of the second finding—we may explain it by using one of the reason why mediators are being invited: non-state group seek to involve mediators, simply because the process can grant them international recognition, and regardless mediation efforts these groups continue or even may increase the use of terror acts. However, we also could tie our finding into the spoiler theory—spoilers use violence to undermine attempts to achieve peace because any peace attempts (e.g. peace agreement) threatens their power or interests (Stedman 1996; 1997; ). In line with this theory we may argue that the probable reason why mediation increases rebel terrorism is that mediation facilitates peace talks which in turn create spoilers who want to undermine peace because they do not see this peace process as beneficial to them. Undoubtedly, it will require more tests and data to test the theory, but at least in regards to terror, our results may show the first empirical proof of spoilers theory by Stedman.

13

Conclusion

This paper has broken ground on an important yet under-studied relationship between terror, civil war and mediation. The descriptive statistics part has showed that as others before us such as Krueger and Laitin (2008) - what may actually be considered quite obvious: cruelty is the steady companion of civil wars. We also showed that mediation and terrorism in civil wars are tightly linked. But by bringing in case studies of Sri Lanka and El Salvador we demonstrated the difficulty with the interpretation of the results. Therefore, we conducted a statistical analysis in order to detect the direction of relationship between mediation and terrorism and in order to test their impact on each other in the presence of other control variables. Using simultaneous equations techniques, the study has demonstrated a couple of statistically significant findings: Terrorism attracts mediation. The more violent the conflict gets the more it receives attention from mediators. It could be explained by the fact that the third parties may be interested in mediation activities to prevent the killings of civilians, or to prevent the spread of terrorism beyond the borders. Long wars cause more terrorism. This result relates to earlier studies that the duration of conflict affects the severity of conflict. One reason could be that in wars that are long eventually the rebels turn to terror (e.g., Colombia). Terror is less likely in internationalized wars. It can be explained by the fact that the third party military involvement may derail the use of terror by rebels. Terrorism is less severe in territorial wars. Both descriptive and statistical analysis demonstrated that the most violent non-state groups are in conflict fought over the control of government. Mediation is more likely if it’s a territorial war, post-Cold War, and internationalized. These results regarding the impact of control variables on mediation are consistent with the great majority of studies in mediation literature. Mediation appears to increase the use of terror by rebels in civil wars. We look for the explanation of this phenomenon in the theory of spoilers presented by Stedman (1996 and 1997). The mediation usually leads to the conclusion of peace agreements, this process however may create spoilers who perceive the provisions included in an agreement as unattractive for or as excluding spoilers’ demands. Resort to violence amidst some signs of commitment to peace may demonstrate rebels’ dissatisfaction with the peace process and what it gives them.

The connection with spoiler theory in this study undoubtedly requires more study. We call upon the scholar and practitioner communities to look at these interrelated relationships in closer detail. Perhaps the study should take into account variables that explain why mediation is less likely to work. One focus is mediation strategy—maybe there is a certain form of mediation that will be more successful in civil wars involving terror.

The future research should also account for the difference between types of wars in order to

understand better the link between mediation and terrorism. Results based on examples of Sri Lanka and El Salvador for territorial and governmental civil wars may only be country-specific but at the same time may indicate the type of civil war may matter in the relationship between mediation and terrorism. Mediation is more likely to take place in territorial wars and as other studies show mediation seems

14

more likely to work in territorial wars (Wallensteen et al. 2009). Because of a special link between territorial wars and mediation, we may expect that once mediators intervene in these wars, the terrorism from rebels decreases, as the case of Sri Lanka demonstrates in this paper. On contrary, governmental wars are less likely to be mediated, and mediation efforts bring worse results in these wars, thus regardless of mediation efforts rebels use terror tactics to influence a government. It could also be that the rebels are simply trying to inflict costs on the government so that more concessions can be made at the negotiating table. References Azam, Jean-Paul, and Anke Hoeffler. 2002. “Violence Against Civilians in Civil Wars: Looting or Terror?”

Journal of Peace Research 39 (4): 461-485.

Bapat, Navin A. 2007. “The Internationalization of Terrorist Campaigns.” Conflict Management and Peace Science 24 (4): 265-280.

BBC News. 2000. Norway role in Sri Lanka peace plan. February 1, by Susannah Price http://news.bbc.co.uk/2/hi/south_asia/627281.stm

BBC News. 2001a. Sri Lanka rebels announce truce. December 19.

BBC News. 2001b. "Sri Lanka enters truce with rebels. December 21. http://news.bbc.co.uk/1/hi/world/south_asia/1723419.stm

Beardsley, Kyle. 2008. “Agreement without Peace? International Mediation and Time Inconsistency Problems." American Journal of Political Science 52 (4): 723-740.

Bercovitch, Jacob and Allison Houston. 1993. “Influence of Mediator Characteristics and Behavior on the Success of Mediation in International Relations.” International Journal of Conflict Management 4(3): 297–321.

Bercovitch, Jacob and Gerald Schneider. 2000. “Who Mediates? The Political Economy of International Conflict Management.” Journal of Peace Research 37(2): 145-165.

Bercovitch, Jacob and Jeffrey Langley. 1993. “The Nature of the Dispute and the Effectiveness of International Mediation.” The Journal of Conflict Resolution 37 (4): 670-691

Bercovitch, Jacob and Karl DeRouen. 2004. “Mediation in Internationalized Ethnic Conflicts: Assessing the Determinants of a Successful Process.” Armed Forces and Society 30 (2): 147-170.

Bercovitch, Jacob and Paul F. Diehl. 1997. “Conflict Management of Enduring Rivalries: The Frequency, Timing and Short-Term Impact of Mediation.” International Interactions 22 (4): 299-320.

Bercovitch, Jacob and Richard Jackson, 2001. “Negotiation or Mediation? An Exploration of Factors Affecting the Choice of Conflict Management in International Conflict.” Negotiation Journal 17(1): 59-77.

Bercovitch, Jacob and Richard Jackson, 2001. “Negotiation or Mediation? An Exploration of Factors Affecting the Choice of Conflict Management in International Conflict.” Negotiation Journal 17(1): 59-77.

Bercovitch, Jacob and Scott Sigmund Gartner. 2006. “Is there a Method in the Madness of Mediation? Some Lessons for Mediators from Quantitative Studies of Mediation.” International Interactions 32: 329-354.

15

Bercovitch, Jacob, J.T. Anagnoson and D.J. Wille. 1991. “Some Conceptual Issues and Empirical Trends in the Study of Successful Mediation in International Relations.” Journal of Peace Research 28: 7-17.

Bohara Alok K., and Neil J. Mitchell. 2006. “Opportunity, Democracy, and the Exchange of Political Violence A Subnational Analysis of Conflict in Nepal.” Journal of Conflict Resolution 50 (1): 108-128.

Borooah, Vani K. 2009. ”Terrorist Incidents in India, 1998-2004: A Quantitative Analysis of Fatality Rates.” Terrorism and Political Violence 21 (3): 476 – 498

Boulden, Jane. 2009. “Terrorism and Civil Wars.” Civil Wars 11 (1): 5-21.

Brown, Michael E., Sean M. Lynn-Jones, and Steven E. Miller. 1996. Debating the Democratic Peace. Cambridge, Mass.: MIT Press.

Bullion, Alan.2001. “Norway and the Peace Process in Sri Lanka.” Civil Wars 4 (3): 70-93.

Coleman, P. T. 2000. Intractable Conflict. In Deutsch, M. and Peter T. Coleman, (eds.). The Handbook of Conflict Resolution: Theory and Practice. Francisco: Jossey-Bass Publishers.

Collier, P., and A. Hoeffler. 2000. “Greed and Grievance in Civil War.” World Bank Policy Research Working Paper 2355.

Collier, Paul, Lani Elliot, Havard Hegre, Anke Hoeffler, Marta Reynal-Querol and Nicholas Sambanis. 2003. Breaking the Conflict Trap: Civil wars and Development Policy. Washington, D.C. and New York: World Bank and Oxford University Press.

Crescenzi, Mark, Kelly M. Kadera, Sara Mitchell, & Clayton L. Thyne. 2008. “A Supply Side Theory of Mediation.”Available at Iowa Research Online, http://ir.uiowa.edu/polisci pubs/6

Crocker, C. A., Hampson, F. O. and Aall, P. 2004. Taming intractable conflicts: Mediation in the hardest cases. Washington, D.C.: United States Institute of Peace Press.

Cronin, Audrey K. 2010. “When should we talk to Terrorists?” USIP Special Report number 240.

Cronin, Audrey Kurth. 2010. When Should We Talk to Terrorists? United States Institute of Peace Special Report 240, May, available at http://www.usip.org/files/resources/SR240Cronin.pdf

DeRouen Jr., Karl, Jacob Bercovitch and Paulina Pospieszna. 2011. “Introducing the Civil Wars Mediation (CWM) Dataset.” Journal of Peace Research. Forthcoming.

DeRouen, Karl, and David Sobek. 2004. “The Dynamics of Civil War Duration and Outcome.” Journal of Peace Research 41(3): 303-320

DeRouen, Karl, and Shaun Goldfinch. 2005. “Putting the Numbers to Work: Implications for Conflict Prevention.” Journal of Peace Research 42(1): 27-45.

Diamond, Larry.1995. Promoting Democracy in the 1990w: Actors and Instruments, Issues and Imperatives. Report to the Carnegie Commission on Preventing Deadly Violence. New York: Carnegie Corporation.

Dixon, William J. 1996. "Third-Party Techniques for Preventing Conflict Escalation and Promoting Peaceful Settlement." International Organization 50(4): 653-81.

Dixon, William. 1994. "Democracy and the Peaceful Settlement of International Conflict." American Political Science Review 88(1): 14-32.

16

Doyle, Michael & Nicholas Sambanis. 2000. “International Peacebuilding: A Theoretical and Quantitative Analysis” American Political Science Review 94 (4): 779-801.

Doyle, Michael W. 1983. "Kant, Liberal Legacies, and Foreign Affairs". Philosophy and Public Affairs 12 (3): 205–235.

Dunkerley, James. 1994. The Pacification of Central America: Political Change in the Isthmus, 1987-1993. London and New York: Verso.

Eck, Kristine and Lisa Hultman. 2007. “One-Sided Violence Against Civilians in War : Insights from New Fatality Data.” Journal of Peace Research 44 (2): 233-246.

Fearon, James D., and David D. Laitin. 2003. “Ethnicity, Insurgency, and Civil War.” American Political Science Review 97: 75-90.

Fearon, James. 2004. “Why do some Civil Wars Last So Much Longer than Others?” Journal of Peace Research 41(3): 275-301.

Findley, Michael G, and Joseph K. Young. 2011. “Terrorism and Civil War: A Spatial and Temporal Approach to a Conceptual Problem.” Paper presented at the 52st Annual Convention of the International Studies Association, Montreal, Quebec, Canada, 16-19 March, 2011

Frazier, Derrick V, and William J. Dixon. 2006. "Third-Party Intermediaries and Negotiated Settlements, 1946-2000." International Interactions 32: 385-408.

Frontline. 2001. The Faciliator Fracas. July 6, by D. Jeyaraj, pp.44-5

Gartner, Scott Sigmund, and Jacob Bercovitch. 2006. "Overcoming Obstacles to Peace: The Contributions of Mediation to Short-Lived Conflict Settlements." International Studies Quarterly 50 (4): 319-28.

Greig, J. Michael. 2005. “Stepping into the Fray: When Do Mediators Mediate?” American Journal of Political Science 49(2): 249-266.

Guardian (London). 2002. Ceasefire raises Sri Lankan peace hopes. February 22, available at http://www.guardian.co.uk/world/2002/feb/22/srilanka1

Gurr, Ted Robert. 2000. Peoples versus states. Washington, D.C.: U.S. Institute for Peace.

Harff, Barbara. 2003.”Nolessons learned from the Holocaust? Assessing risks of genocide and political mass murder since 1955.” American Political Science Review 97 (February): 57-73.

Hensel, Paul. 2001. “Contentious Issues and World Politics: The Management of Territorial Claims in the Americas, 1816-1992." International Studies Quarterly 45(1):81-109.

Hultman, Lisa, 2007. “Battle Losses and Rebel Violence: Raising the Costs for Fighting.” Terrorism and Political Violence 19 (2): 205 – 222.

Humphreys, Macartan & JeremyWeinstein. 2006. “Handling and Manhandling in Civil War.” American Political Science Review 100(3):429–447.

Ibáñez, Ana María, and Andrea Velásquez. 2009. “Identifying Victims of Civil Conflicts: An Evaluation of Forced Displaced Households in Colombia.” Journal of Peace Research 46 (3): 431-451.

Kalyvas, Stathis N. 2006. The Logic of Violence in Civil Wars. Cambridge: Cambridge University Press.

Kalyvas, Stathis. 1999. “Wanton and Senseless: The Logic of Massacres in Algeria.” Rationality and Society 11(3):243–285.

Kalyvas, Stathis. 2004. “The Paradox of Terrorism in CivilWar.” The Journal of Ethics 8:97–138.

17

Keshk, Omar M. G. 2003. “CDSIMEQ: A program to implement two-stage probit least squares.” The Stata Journal 3 (2): 1–11.

Keshk, Omar M.G.; Brian M. Pollins, and Rafael Reuveny. 2004. “Trade Still Follows the Flag: The Primacy of Politics in a Simultaneous Model of Interdependence and Armed Conflict.” Journal of Politics 66 (4): 1155-1179.

Kirk, Richard M. 1983. “Political Terrorism and the Size of Government: A Positive Institutional Analysis of Violence Political Activity.” Public Choice 40: 41-52.

Kleiboer, Marieke. 1996. "Understanding Success and Failure of International Mediation." Journal of Conflict Resolution 40(2): 360-89.

Kreisberg, L. 2005. Nature, dynamics, and phases of intractability. In Crocker, C. A., Hampson, F. O. & Aall, P. (eds.), Grasping the nettle: Analyzing cases of intractable conflict. Washington, D.C.: United States Institute of Peace Press.

Kreutz, Joakim. 2008. UCDP One-Sided Violence Codebook, v.1.3.

Krueger, Alan, and David D. Laitin. 2008. “Kto Kogo?: A Cross-Country Study of the Origins and Targets of Terrorism” In Terrorism, Economic Development, and Political Openness, eds. Philip Keefer and Norman Loayza. New York: Cambridge University Press.

Kydd, Andew and Barbara Walter. 2006. “The Strategies of Terrorism.” International Security 31(1):49–80.

Kydd, Andrew and Barbara Walter. 2002. “Sabotaging the Peace: The Politics of Extremist Violence.” International Organization 56:263–296.

Lacina, Bethany. 2006. “Explaining the Severity of Civil Wars.” Journal of Conflict Resolution 50 (2) 276-289.

Lake, David A. 2002. “Rational Extremism: Understanding Terrorism in the Twenty-first Century.” Dialog-IO, Spring, 15–29.

Lozano-Gracia, Nancy et al. 2010. “The Journey to Safety: Conflict-Driven Migration Flows in Colombia.” International Regional Science Review 33 (2): 157-180.

Maddala, G.S. 1983. Limited-dependent and qualitative variables in econometrics. Cambridge University Press, Cambridge.

McAdam, Doug. 1982. The political process and the development of black insurgency. Chicago: University of Chicago Press.

Mitchell, Sara McLaughlin. 2002. “A Kantian System? Democracy and Third-Party Conflict Resolution." American Journal of Political Science 46(4):749-759.

Moolakkattu, John Stephen. 2005. “Peace Facilitation by Small States: Norway in Sri Lanka Cooperation and Conflict.” Cooperation and Conflict 40 (4): 385-402.

Muller, Edward N. 1985. “Income inequality, regime repressiveness, and political violence.” American Sociological Review 50:47-61.

Oneal, John R., and Bruce Russett. 1999. "The Kantian Peace: The Pacific Benefits of Democracy, Interdependence, and International Organizations.” World Politics 52 (1): 1-37.

Ott, Marvin C. 1972. “Mediation as a Method of Conflict Resolution: Two cases.” International Organization 26: 595-618.

18

Pape, Robert A. 2003. “The Strategic Logic of Suicide Terrorism.” American Political Science Review 97 (3): 343-361.

Putnam, L. L. and J. M. Wondolleck. 2003. Intractability: Definitions, dimensions, distinctions. In Lewicki, R. J., Grey, B. & Elliott, M. (Eds.), Making sense of intractable environmental conflicts: Concepts and cases. Washington: Island Press.

Rauchhaus, Robert. 2006. "Asymmetric Information, Media tion and Conflict Management." World Politics 58(2): 207-41.

Ray, James L. 1995. Democracy and International Conflict. Columbia: University of South Carolina Press.

Regan, P. 2002. “Third-Party Intervention and the Duration of Intrastate Conflicts.” Journal of Conflict Resolution 46(1), 55–73.

Regan, P.M. 1996. Conditions of Successful Third-Party Intervention in Intrastate Conflicts. The Journal of Conflict Resolution 40 (2): 336-359.

Regan, Patrick M., and Allan C. Stam. 2000. "In the Nick of Time: Conflict Management, Mediation Timing, and the Duration of Interstate Disputes." International Studies Quarterly 44: 239-60.

Regan, Patrick, and Aysegul Aydin. 2006. "Diplomacy and Other Forms of Intervention in Civil Wars." Journal of Conflict Resolution 50(5): 736-56.

Reuters. 2001. Norwegian peace envoy arrives in Sri Lanka. April 30.

Reuveny, Rafael, and Quan Li. 2003.”The Joint Democracy–Dyadic Conflict Nexus: A Simultaneous Equations Model “. International Studies Quarterly 47 325–346

Ron, James. 2001. “Ideology in context: Explaining Sendero Luminoso’s tactical escalation.” Journal of Peace Research 38:569-92.

Rovik, Falk. 2005. “Norway: A terrorist safe haven?” In Peace in Sri Lanka: Obstacles and Opportunities. Nugegoda: World Alliance for Peace in Sri Lanka, Colombo.

Rummel, R. J. 1997. Power Kills: Democracy as a Method of Nonviolence. New Brunswick, N.J.: Transaction

Rummel, R.J. 1995. “Democracy, Power, Genocide, and Mass Murder.” Journal of Conflict Resolution 39 (1): 3-26.

Russett, Bruce and John R. Oneal. 2001. Triangulating Peace: Democracy, Interdependence, and International Organizations. W. W. Norton & Company.

Sambanis, Nicholas. 2008. “Short- and Long-Term Effects of United Nations Peace Operations.” World Bank Economic Review 22 (1): 9-32.

Sambanis, Nicholas. 2008. “Terrorism and Civil War.” In Terrorism, Economic Development, and Political Openness, eds. Philip Keefer and Norman Loayza. New York: Cambridge University Press.

Sandler, Todd. 2003. “Collective Action and Transnational Terrorism.” World Economy 26: 779-802.

Savun, Burcu. 2009. Mediator Type and Effectiveness of Information-Provision Strategies in the Resolution of International Conflict. In.: Jacob Bercovitch and Scott Gartner eds. International Conflict Mediation: New Approaches and Findings. London: Routledge.

19

Schneider, G. and Bercovitch J. 2006. Analyzing the Effectiveness of International Conflict Resolution: Structural and Personal Factors in Mediation. Conference Papers—International Studies Association.

Stedman, Stephen John. 1996. Negotiation and Mediation in Internal Conflicts, In Michael E. Brown, ed., The International Dimensions of Internal Conflict. Cambridge, Mass.: MIT Press/

Stedman, Stephen John. 1997. “Spoiler Problems in Peace Processes.” International Security 22(2): 5-53.

Svensson, Isak. 2007. "Bargaining, Bias and Peace Brokers: How Rebels Commit to Peace." Journal of Peace Research 44(2): 177-94.

Tamil Guardian. 2001. Offensive seriously undermines Norwegian initiative, May 2, p.1.

Tarrow, Sidney. 1994. Power in movement: Social movements, collective action and politics. New York: Cambridge University Press.

The Sunday Times (London). 2001. New bid to end Sri Lanka war. May 6, by P.Conradi, p.25

Touval, Saadia and I. William Zartman. 1985.” Introduction: Mediation in theory.” In: Saadia Touval & I. William Zartman eds. International Mediation in Theory and Practice. Boulder, CO: Westview, 7-18.

UPI (United Press International). 1990a. Fighting continues as El Salvador faces uncertain New Year., December 29, 1990, Saturday, by Daniel Alder.

UPI (United Press International). 1990b. Congress approves aid bill, October 27, 1990, Saturday, Washington News.

Wallensteen, Peter; Karl DeRouen, Jacob Bercovitch and Frida Möller. 2009. “Democracy and mediation in territorial civil wars in Southeast Asia and the South Pacific.” Asia Europe Journal 7 (2): 241-264.

Walter, Barbara F. 2002. Committing to Peace: The Successful Settlement of Civil Wars. Princeton, NJ: Princeton University Press.

Weerasinghe, Mahinda. 2005. “Kissinger, Nobel committee, Kissinginian mutants, or how Norway became a sanctuary for terrorists”. In Peace in Sri Lanka: Obstacles and Opportunities. Nugegoda: World Alliance for Peace in Sri Lanka.

Werner, Suzanne and Douglas Lemke. 1997. “Opposites Do Not Attract: The Impact of Domestic Institutions, Power, and Prior Commitments on Alignment Choices.” International Studies Quarterly 41 (3): 529-546.

Werner, Suzanne, and Amy Yuen. 2005. "Making and Keeping Peace." International Organization 59(2): 261-92.

Wilkenfeld, Jonathan, Kathleen J. Young, David M. Quinn, and Victor Asal. 2005. Mediating International Crises. New York: Routledge.

Wilkenfeld, Jonathan, Kathleen Young, Victor Asal, and David Quinn. 2003. "Mediating International Crises: Cross National and Experimental Perspectives." Journal of Conflict Resolution 47(3): 279-301.

Wood, Reed M. 2010. “Rebel capability and strategic violence against civilians.” Journal of Peace Research 47(5) 601–614.

20

Zahar, Marie-Joelle. 2001.‘‘Protégés, Clients, Cannon Fodder: Civilians in the Calculus of Militias,” In Adekeye Adebajo and Chandra Lekha Sriram, eds., Managing Armed Conflicts in the 21st Century. London: Frank Cass.

Zartman, I. William. 1995. Elusive peace: Negotiating an end to civil war. Washington, DC: Brookings Institute.

Zartman, I. William. 2000. Ripeness: The hurting stalemate and beyond. In P.C. Stern and D. Druckman (eds.) International conflict resolution after the cold war, Washington, DC: National Academy Press.