Civil War and Terrorism: A Call for Further Theory Building
Summary and Keywords
Since the September 11 terrorist attacks on the United States, terrorism has gained increased prominence in both scholarship and the media. While international terrorist acts are quite visible and highly publicized, such attacks represent only one type of terrorism within the international system. In fact, a very large number of acts of terrorism take place within the context of civil wars. Given the great disparity in power in most civil wars, it is not surprising that terrorism might be seen as a tactic that is often used by insurgent groups, who may have few resources at their disposal to fight a much stronger opponent.
There is a clear linkage between the concepts of terrorism and civil war, yet until recently scholars have largely approached civil war and terrorism separately. Recent literature has attempted to specifically map the intersection of terrorism and civil war, recognizing the extent to which the two overlap. As expected, the findings suggest that civil war and terrorism are highly linked. Other scholars have endeavoured to explain why rebel groups in some civil wars use terrorism, while others do not. Further research focuses on how governments respond to terrorism during civil war or on how the decisions of external actors to intervene in civil wars are affected by the use of terrorism by insurgent groups.
These studies show that there is too little theorizing on the relationship between civil war and terrorism; while scholars are finally considering these concepts collectively, the full nature of their relationship remains unexplored. Additional research is needed to better understand the various ways that terrorism and civil war overlap, interact, and mutually affect other important international and domestic political processes.
For years the concepts of terrorism and civil war have largely been considered separately, in different sets of literatures and by different scholars. For example, a new volume, What Do We Know about Civil Wars? (2016), edited by T. David Mason and Sara McLaughlin Mitchell, presents a well-rounded discussion of the literature on civil wars. However, while the editors compile articles on the various causes of civil wars, on factors that contribute to ending wars and promoting peace, and on current trends in the study of civil war, not a single article considers the potential connections between civil war and terrorism. While terrorism is mentioned in a smattering of the chapters, the references are in passing, or concern largely the targeting of civilians within civil wars, often without actually mentioning terrorism. For example, a chapter on trends in civil war data by Cunningham, Gleditsch, and Salehyan (2016) mentions a few articles that connect terrorism and civil war, but the discussion of these articles is overly general and focuses only on the issue of whether or not rebel groups target civilians. The pieces they cite (such as Stanton, 2013 and Fortna, 2015) actually provide a much more nuanced discussion of the relationship between civil wars and terrorism than is present in the chapter.
Similarly, until recently, the literature focusing on the causes of terrorism often considered terrorism (both domestic and international terrorism) broadly, or focused more exclusively on transnational terrorism. While international terrorist acts have been on the rise, the fact remains that most terrorist events take place within the context of civil wars (Sambanis, 2008). A brief look at the Global Terrorism Database (GTD), one of the most comprehensive datasets of terrorist events, illustrates that many of the terrorist attacks included in the database occur within the context of civil war (National Consortium for the Study of Terrorism and Responses to Terrorism [START], 2016; Cunningham, Gleditsch, & Salehyan, 2016). In fact, Findley and Young (2012b) coded the geographic coordinates for most of the terrorist events in the GTD dataset and find that between 44% and 63% of these terrorist events took place within civil war zones (depending on the civil war measure used) and that 73% of civil wars experienced at least one terrorist event. These data demonstrate the clear connections and overlap between civil war and terrorism. Additionally, even international terrorist attacks are often connected to a civil conflict, demonstrated by the various international terrorist attacks carried out and claimed by Daesh (or the Islamic State) since 2015.
Furthermore, the context of civil war likely interacts with terrorism in many ways. On the one hand, the characteristics of a civil war and the country in which it is fought may influence the use of terrorism as a tactic by groups involved in the war; such characteristics might be linked to both domestic and international terrorism. On the other hand, terrorism may impact the civil war itself, how it is conducted, how long it lasts, how peace is constructed, and what the postwar environment may look like. Given the intricate connection between civil war and terrorism, it is problematic to consider these two concepts in isolation of one another. Terrorism and civil war overlap and influence one another in numerous ways. Despite the challenges, not only are there significant advantages to considering terrorism and civil war jointly, but given the strong connections between these concepts, it is difficult to fully analyze and understand these issues in isolation of one another. In particular, the causes of terrorism (and why some groups choose to use terrorism while others do not) cannot be completely evaluated, nor can the potential effects and consequences of the use of terrorism be adequately considered.
Defining and Measuring Terrorism and Civil War
Scholars may have been reluctant to consider both civil war and terrorism within the same research agendas because of the difficulties in defining these concepts, particularly terrorism. No scholarly consensus has been reached on the definition or measure of terrorism; in fact, there are nearly as many definitions for terrorism as there are articles on terrorism, and these definitions greatly vary in how closely they align with one another. Some have even argued that we know little about terrorism, and particularly less about terrorism than civil war because of the ambiguity of the concept of terrorism (Sánchez-Cuenca, 2014). While some of these definitions often share elements (such as violence for a political purpose), others are quite different. One important division in the literature is whether to approach terrorism from an “action” or “actor” perspective (Asal et al., 2012; Sánchez-Cuenca & de la Calle, 2009). Those researchers who approach terrorism as an action tend to focus on the elements of terrorism that make it different from other violent acts. Such definitions of terrorism may include elements that distinguish the target audience as separate from the targets, or they may focus on noncombatants or civilians as the target of such violence (examples include Butcher, 2016; Crenshaw, 1981, Findley & Young, 2012b; Kydd & Walter, 2006; and Stanton, 2013). On the other side, authors such as Sánchez-Cuenca and De La Calle (2009) argue that an actor-based approach is more appropriate. They suggest that action-based approaches are too broad, and based on such definitions too much violence within conflicts can be considered terrorism. They suggest an actor-based approach that pertains to groups and considers the asymmetry between these groups and the state. One key element of their definition of terrorist groups is that such groups do not control territory—this is what makes terrorist groups different from other insurgencies and guerrilla groups. Based on this definition, ETA (Euskadi Ta Askatasuna, Basque Homeland and Freedom) in the Basque areas of Spain and France would be a terrorist group, but the FARC (Revolutionary Armed Forces of Colombia) in Colombia and the LTTE (Liberation Tigers of Tamil Eelam) in Sri Lanka would be considered guerrillas. One consequence of this type of approach is that guerrilla actors and terrorist actors would be different by definition, and thus these groups could not overlap (Findley & Young, 2012b). Furthermore, authors have pointed out that such a definition fails a simple face validity test, as many groups that are “known” to commit terrorist acts (like the LTTE, the Shining Path) are not included as terrorist groups (see Asal et al., 2012). Clearly, depending on how terrorism is defined, the implications for its overlap and interaction with civil war would be dramatically different.
There is a bit less variation in definitions of civil war, and much of the differences in how to approach civil war relates to the thresholds of battle-deaths used to distinguish civil war from other low-level conflict in a country. Civil war is usually defined as armed conflict that occurs between the government of a state and domestic groups within the state that are able to effectively wage a campaign against the state, with the threshold of violence within the definition varying depending on the source of the data (Sambanis, 2008). The Correlates of War (COW) Project (Sarkees & Wayman, 2010) and the Uppsala Conflict Data Program/Peace Research Institute Oslo (UCDP/PRIO) Armed Conflict Dataset (Melander, Pettersson, & Themnér, 2016; Gleditsch et al., 2002) provide the mostly widely utilized approaches to measuring and defining civil war. The COW Project includes civil wars as one type of intrastate war. In order for conflicts to be included as intrastate wars, they must result in a minimum of 1000 battle-related combatant fatalities within a one-year period. Civil wars differ from other types of intrastate war in that they involve the government of a state fighting against a nonstate entity. The UCDP/PRIO Armed Conflict Dataset includes a lower threshold for violence, with armed conflict being defined as occurring if at least 25 battle-related deaths occur within a calendar year. Civil wars, or intrastate conflicts, would have the government as one actor and one or more substate groups as the other actor.
Similar to how they treat the concept of civil war, many authors tend to utilize definitions of terrorism based on the datasets they use. The International Terrorism: Attributes of Terrorist Events (ITERATE) database focuses only on transnational terrorism (Mickolus et al., 2007). The RAND Database of Worldwide Terrorism Incidents (RDWTI) and the Global Terrorism Database (GTD) include both domestic and international terrorist events (RAND, 2016; START, 2016). In addition, the RDWTI and the GTD include the nature of the target within their datasets, which allows researchers to look more specifically at civilian targets. The scope of the GTD is much broader, as the RDWTI ends in 2009 and the GTD currently has data to 2015. Furthermore, the GTD is more comprehensive and includes more than 150,000 terrorist attacks, compared to a little more than 40,000 in the RDWTI. Both the RDWTI and the GTD have specific definitions of terrorism, and authors tend to utilize these definitions when using the corresponding dataset.
The problems associated with defining and measuring terrorism and civil war create both methodological and conceptual issues because in order to consider both concepts and how they overlap and interact, scholars must be able to define and measure these terms in a way that makes them conceptually and empirically distinct. Unfortunately, many definitions and measures of these concepts contain some potential overlap. This is particularly problematic in how terrorism is defined, as terrorism needs to be defined in such a way that the definition will not necessarily include all forms of violence that might take place in civil war. For example, Kalyvas (2004) approaches terrorism as the use of violence within a civil war, distinguishing two types of violence—discriminate and indiscriminate. Such an approach does not acknowledge any conceptual distinction between terrorist violence and other forms of violence. Rather, it makes distinctions based on how the targets are chosen. While the distinction between discriminate and indiscriminate violence is useful, it is not clear that lumping all such violence under the label of “terrorism” is extremely helpful in understanding terrorism more broadly and its connection to civil war more specifically.
While an actor-based approach has been argued to provide a clearer distinction between terrorism and other concepts (Sánchez-Cuenca & de la Calle, 2009), such an approach lacks nuance and makes it difficult to see the choices of tactics made by individual groups at different points in time. It assumes that groups are “terrorist” rather than that groups choose terrorism as a tactic among many different options, and that they may do so, or not, at different points in time. Such a limited definition does not allow researchers to consider terrorist violence within a civil war over time, nor really hone in on the multitude of potential motives and deterrents for using terrorism as a tactic within civil war. As such, an action-based approach provides more information from which to fully consider the overlap and interaction of terrorism and civil war and is more useful for research that aims to consider these two concepts in tandem.
However, not all action-based definitions of terrorism are completely distinct from civil war. For example, the GDT defines terrorism as an “intentional act of violence or threat of violence by a non-state actor” that also includes two of the following criteria: it (1) “is aimed at attaining a political, economic, religious, or social goal”; (2) “includes evidence of an intention to coerce, intimidate, or convey some other message to a larger audience (or audiences) other than the immediate victims”; and (3) “was outside the precepts of International Humanitarian Law” (START, 2016). This definition is very broad, however, and using it may make distinguishing terrorism from other violence very difficult. One of the key elements of this definition that helps distinguish terrorism from other types of violence that might take place in civil war is the second criterion, which includes the idea that the victims of the act are different than the audience. However, the definition of terrorism from the GDT could still be met without this criterion (since terrorism, as defined, must include only two of the three criteria), and the database includes attacks against the military, government, and police. Similarly, the RDWTI defines terrorism as an act that includes several key elements, such as violence or the threat of violence; is calculated to create fear; is intended to coerce; is motivated by a political objective; is generally directed against civilian targets; and can be perpetuated by a group or individual (RAND, 2016). While the definition states that terrorism “generally” targets civilians, the database includes military and government targets as well, so it is not entirely clear how this definition is used to determine terrorist incidents.
Given the difficulties of finding consensus on what terrorism is and defining it in a way that avoids too much overlap with the concept and definition of civil war, it is not surprising that some scholars have avoided using terrorism as a concept when discussing civil war and instead have often focused more narrowly on the idea of civilian casualties or civilian targets (Salehyan, Siroky, & Wood, 2014). While this does help alleviate some of the problems related to defining terrorism, it does so by merely avoiding the issue rather than by delving into a deeper discussion of what terrorism is and how it may be affected by and influence civil war more broadly. Rather than avoiding the problem, researchers should tackle these conceptual challenges. Only then can we begin to understand the complex ways that terrorism and civil war might interact.
It appears then that an action-based approach to terrorism would best allow researchers to consider the various ways that terrorism and civil war may overlap and/or interact. However, this approach should be coupled with a specific focus on the targeting of civilians, so that terrorism can be distinguished between other types of violence that would be present during a civil war.
Potential Theoretical Links between Terrorism and Civil War
Given the potential conceptual and methodological problems of considering the connections between terrorism and civil war, one may be inclined to wonder if it is worth the effort to disentangle these concepts. However, there are clear theoretical linkages between civil war and terrorism, and as such it is problematic to only consider these concepts in isolation of one another. For example, literature on the causes of terrorism and the strategies that terrorist groups use has increasingly moved toward the group level of analysis. Instead of focusing on the psychological reasons for individuals to use terrorism or to join groups that use terrorism, more and more scholars are considering the reasons groups use terrorism and the motives behind various terrorist strategies. These groups are often formed in resistance to governments and often oppose these governments in civil wars. In such instances, analysis of the causes of terrorism cannot be fully separated from the context in which this terrorism is taking place—in this instance, civil wars. The very nature of the civil war context itself frames the decisions of groups when determining tactics. Without considering these two concepts in tandem, we cannot fully understand the causes of terrorism in these cases.
Martha Crenshaw’s foundational work on terrorism (1981) provides a theoretical basis for beginning to see the connections between terrorism and civil war, and her theories are often invoked by those researchers who have begun to tie terrorism and civil war together more explicitly. Crenshaw considers the reasons groups use terrorism, and sees terrorism as a rational political choice made by groups. Thus, terrorism is a logical means to desired ends—groups use this tactic because they believe it will help them attain their objectives. While not all groups have the same ultimate goals, Crenshaw argues that common patterns of more short-term objectives drive terrorist strategy. One clear goal of terrorism which Crenshow identifies is to gain recognition or attention and to advertise their cause and gain publicity. Another goal is to disrupt and weaken the government, making it more difficult for the government to function properly and to potentially weaken the government’s control over its territory. Crenshaw further suggests that terrorism may be aimed at provoking a reaction (or overreaction) from the government in an attempt to delegitimize the government even further by decreasing public support for government actions.
In terms of the public, Crenshaw argues that terrorism has both positive and negative effects—it can create sympathy among the population, or it may engender fear in the parts of the population that are targeted. In addition, terrorism can be used to maintain control, discipline, and morale within a terrorist group, or even as an instrument of rivalry between competing factions. In this way, various groups can use terrorism to compete with one another for public and international recognition (a process others term “outbidding”).
Crenshaw suggests that terrorism is a strategic and logical choice for groups when they have these types of goals and that there is a large power imbalance between the group and the government. Government repression can weaken organizations, emphasizing this power differential. However, groups may be weak for other reasons as well, including timing constraints that make groups impatient for using legal methods of redress that might be more time consuming. Alternatively, the group may just be unpopular and unwilling or unable to mobilize greater support. While Crenshaw does not specifically tie her discussion of the causes of terrorism to civil war, her discussion is largely about groups using terrorism to challenge the state’s authority and is generally applicable to a civil war context. Her argument, however, might be further strengthened if she more explicitly tied the context of civil war to the various causes of terrorism she discusses.
Like Crenshaw (1981), Kydd and Walter’s (2006) influential work focuses on the strategies of terrorism but again fails to fully incorporate a discussion of the civil war context. The authors focus on four different questions: the types of goals terrorists seek to achieve; the strategies used to pursue those goals; why the strategies work in only some cases; and how government can best respond to and prevent terrorism. With regard to the first three questions, there are clear connections between terrorism and civil war, though Kydd and Walter do not explicitly make these connections. They define terrorism “as the use of violence against civilians by nonstate actors to attain political goals” (Kydd & Walter, 2006, p. 52). While this definition does not necessarily mean that such acts would take place within civil wars, it is clear that many such acts would be part of a broader civil war.
Kydd and Walter point to five goals of terrorists: regime change, territorial change, policy change, social control, and status quo maintenance. There are clear links between some of these goals and civil war. For example, many of the cases of terrorism whose goals are regime change and overthrow of the government would be expected to be part of a larger civil war campaign (the authors give the example of the Shining Path in Peru). Similarly, groups that have territorial change (and secession or joining with another state) as their goal are also likely to be embroiled in a civil war with the state (the examples of the Tamil Tigers in Sri Lanka and Lashkar-e Yayyiba in Indian Kashmir would fit here, as discussed by the authors). Thus, while all of these goals might not be equally likely to also be part of a civil war campaign, clearly there is substantial overlap between some of these goals and the goals of civil war more generally. Without fully parsing out these connections, we are unable to see how the pursuit of these goals may vary within and outside a civil war context and we cannot evaluate the success of reaching these goals inside and outside civil wars. Thus, the presence of a civil war can impact the choice of goal and the success at achieving these goals, but without incorporating such discussions, theoretical understanding and empirical testing of these potential impacts are lacking.
In terms of strategies, again some potential connections to civil war are evident. Kydd and Walter focus on five specific strategies, two that are directed at the government that a group hopes to influence (attrition and spoiling) and three that are directed at the population that might support the group (intimidation, outbidding, and provocation). Attrition is generally a battle of wills between the state and the group, with the group attempting to inflict great costs on the state in an effort to gain concessions. This is similar to Crenshaw’s (1981) discussion of the goal of weakening the state. Spoiling occurs when terrorist attacks are used to mar the peace at the end of a conflict. Both of these tactics point to close connections to civil war, and in fact, the act of spoiling would likely take place during the peace process following a civil conflict. Intimidation is similar to deterrence in that it works by groups demonstrating their ability and power to punish and showing that the government is unable to stop them. The strategy of intimidation is similar to some of Crenshaw’s ideas, specifically in relation to weakening the government, demonstrating its inability to control its territory, and instilling fear in the population. According to Kydd and Walter, intimidation is most likely to be used by groups attempting to overthrow a government or to gain social control of the population within the country. As mentioned earlier, the act of attempting to overthrow the government would likely also be part of a civil war campaign, so these cannot easily be separated. Outbidding, a process whereby multiple groups are competing for the public’s support, is also a process that is likely to take place within civil wars (in fact, Kydd and Walter give the example of Hamas and Fatah). This process is similar to the process of competition discussed by Crenshaw. Finally, provocation—the use of terrorism in an attempt to provoke an overreaction by the government—is often used in pursuit of regime change, according to the authors (it is also a strategy discussed by Crenshaw).
With many of the strategies discussed by Kydd and Walter (2006), as well as Crenshaw (1981), there are clear connections to civil war. The exclusion of civil war from such discussions is particularly problematic because we would expect different outcomes based on variations in the civil war context itself. As such, considering more explicitly how terrorism and civil war overlap and impact each other is extremely important. Given the strong potential for these theoretical discussions to relate to civil war, it is not surprising that the majority of recent research connecting terrorism and civil war utilizes the strategies and theories presented by Crenshaw (1981) and Kydd and Walter (2006). Both of these works take a rationalist approach to terrorism, considering the cost-benefit calculations of various actors. Nearly all of the literature on the relationship between civil war and terrorism takes a similar rational-type approach.
Research Connecting Civil War and Terrorism
Scholars have more recently begun to consider the various ways that terrorism and civil war overlap and influence one another. Initial works attempted to generally discuss the ways that terrorism and civil war are conceptually and empirically related and/or overlap. From these studies, two broad categories of research have emerged. The first category of research focuses on why some groups use terrorism within civil wars and others do not. Studies in this category might focus on characteristics of the state or characteristics of the group, or even characteristics of the civil war itself, to help explain why groups choose terrorism within the context of civil war. These studies demonstrate the difficulty of fully explaining the causes of terrorism without also considering the civil war context. The second category includes studies that focus on the potential effects of terrorism in civil wars. The potential impact of terrorism (both domestically and internationally) is an important avenue for research, and again, one that cannot be fully pursued without a consideration of both terrorism and the context of civil war. In both of these categories, researchers rely largely on rationalist explanations and theories to explain outcomes, whether the outcome is terrorism or the effects of terrorism, and specifically utilize the theoretical discussions of Crenshaw (1981) and Kydd and Walter (2006).
First Insights: Conceptual and Empirical Connections Between Terrorism and Civil War
Increasingly, researchers are attempting to discover the links between terrorism and civil war. Boulden (2009) begins to chart the connections between terrorism and civil war and argues that both terrorism and civil war are forms of political violence and that both are tactics or tools used by nonstate actors to pursue their goals. Thus, she argues that terrorism and civil war are two possible options on the spectrum of political violence that groups might choose from (Boulden, 2009). Boulden’s article is one of the first to attempt to more explicitly link terrorism and civil war conceptually, but unfortunately her article remains largely at the conceptual level with no empirical analysis of these various connections. In her conclusion, she emphasizes the importance of moving beyond state-centric analyses and also focusing on the causes of political violence. However, such focus “requires a far more nuanced sense of the contexts at both the national and international levels that generate political violence than we have achieved to date” (p. 20). Analyzing the various ways that civil war and terrorism overlap and influence each other is one step in that direction. This analysis requires focusing on group-level dynamics in addition to state-level factors, and while Boulden’s article is useful for outlining the potential conceptual overlap between terrorism and civil war, it falls short of providing solid ground for empirical testing of these various connections.
Sambanis (2008) discusses the conceptual differences and similarities between civil war and terrorism and suggests that “we must conceptualize terrorism as a distinct form of violence and not just as a strategy that can take place in the context of any other form of violence” (p. 176). Sambanis proposes a typology of political violence that seeks to help conceptualize the distinctions between civil war and terrorism more concretely. His typology includes a consideration of the key actors, their degree of organization and public support, the balance of power among various actors, and the level, target, and purpose of violence. Sambanis is working to find some conceptual and empirical grounds to help distinguish terrorism from civil war.
Sambanis argues that certain elements of this typology are more likely to help distinguish terrorism from civil war. The degree of organization of actors can be particularly useful for helping distinguish terrorism and civil wars, as terrorist groups, Sambanis argues, are not generally as structured and organized as rebel armies. This argument makes groups, rather than a particular action, the central feature of terrorism. Thus, a distinction is made here between rebel groups and terrorist groups. Similarly, Sambanis argues that a key difference between civil war and terrorism is that terrorist groups have less public support than rebel armies. This suggests that rebel armies fight in civil wars and that terrorism is separate from civil war. While this is a good start in attempting to separate civil war and terrorism, Sambanis’s actor-based approach creates a dichotomy between civil war and terrorism rather than acknowledging that terrorist violence might take place within civil war or that rebels within civil wars might sometimes utilize terrorism as a tactic.
Building on the rationality-based arguments of Kydd and Walter (2006), Findley and Young (2012b) empirically examine the extent to which terrorism and civil war overlap. They consider the link between terrorism and civil war broadly and include temporal and spatial patterns of overlap, analyzing how terrorism may be used prior to war, during war, and after war. As might be expected, Findley and Young find substantial overlap between civil war and terrorism and conclude that the majority of incidents of terrorism take place during ongoing civil wars and in regions where civil wars are occurring. They find that smaller proportions of terrorist incidents occur prior to war, after war, or in zones that are not experiencing war.
Similar to the approach of Kydd and Walter, they suggest that rationalist approaches to conflict are often used when discussing various conflict processes, including civil war and terrorism. They argue that “different varieties of violence may actually fit within a similar strategic approach” (p. 287). Regardless of what these actors are called, and regardless of the fact that they are often considered separately (i.e., dissidents, rebels, insurgents, terrorists, revolutionaries, etc.), Findley and Young suggest that “they may follow a similar strategic logic in their choices to use violence” (p. 287).
Findley and Young (2012b) rely on the five strategies discussed by Kydd and Walter (provocation, attrition, intimidation, outbidding, and spoiling) and discuss how these strategies might correspond to different periods of a civil war (before, during, or after). Provocation is an attempt to provoke a government into using a disproportionate response against a group, therefore causing other members of the population to radicalize and support the group using terrorism. Findley and Young connect this strategy to the beginning of violent conflict, or before a civil war begins, as a strategy to provoke the state and gain support and capacity. Attrition is the need for a group to demonstrate its strength and resolve to fight a conflict and thus to inflict high costs on their adversary. In terms of civil war, the authors argue that this strategy might be used to instigate conflict or to maintain it, but is unlikely to occur after the end of a conflict. Intimidation of the population of a state is another strategy of terrorism. Findley and Young suggest this tactic would be most likely to be used during a civil war, as both sides are fighting to maintain their dominance and thus groups may need to employ it for control. Outbidding, another tactic, is one that Bloom also discusses (2005). When multiple groups are all competing for public support, the use of terrorism, especially suicide terrorism, may increase in an attempt to set a group apart from the rest. As such, this strategic use of terrorism would be used during a civil war, but the authors argue that it is unclear whether it would occur prior to or after a conflict. Finally, the authors consider the strategy of spoiling—where groups use terrorism to attempt to spoil a peace agreement. This strategy would be likely to be employed after a civil war or during lulls or periods of negotiation during the war. Based on the discussion of these strategic approaches, Findley and Young argue that terrorism and civil war overlap quite a bit, with four of the strategies being likely during an ongoing civil war.
Findley and Young, unable to test these specific strategies, instead use them more as a theoretical basis for why terrorism and civil war might be expected to overlap. They proceed to map out the overlap of terrorist incidents with civil war. They find strong overlap between civil war and terrorism, with terrorism being particularly prevalent during civil wars broadly. They also find a high amount of terrorism prior to civil war, but only in Latin America. Finally, they find more terrorism since the end of the Cold War and suggest this is because groups cannot rely on material support from sympathetic superpowers. Overall, their findings suggest that terrorism and civil war are empirically linked. Such an empirical link demonstrates not only a theoretical reason to consider terrorism and civil war together, but also a methodological one. Given that these concepts relate closely to one another, it is necessary to model these interactions in a way to account for their impact within empirical models.
While this study provides empirical support for the link between civil war and terrorism, it suffers from some key shortcomings. Although the study distinguishes between terrorism that takes place within a civil war versus terrorism outside of a civil war context, it does not fully elaborate on the issue of different types of violence within civil war, and it does not help us fully understand exactly why terrorism is used in some civil wars and not others. In fact, the authors define terrorism very broadly as “the threatened or actual use of illegal force and violence by a non-state actor to attain a political, economic, religious or social goal through fear, coercion or intimidation” (p. 289). This is a conventional and increasingly accepted definition of terrorism and comes from the GTD. However, it is not entirely clear when such “illegal force and violence by a non-state actor” within the context of civil war would NOT be considered terrorism. Certainly, not all acts of violence from the nonstate actor side are terrorism, but this study provides no way to fully distinguish these types of violence, and thus no way to consider when terrorism is used and when it is not used within a civil war context. Such nuances are important in any exploration of the relationship between terrorism and civil war.
Some of the factors included in Findley and Young’s paper can help explain when groups might use terrorism. For example, attrition is theorized to be most effective against democracies, so it may be used in those cases more frequently. Outbidding is a tactic that would be used when there is splintering within rebel groups or when there are multiple nonstate actors in the same conflict. The authors point to issues, such as spoiling, that might affect the outcome of civil war, particularly the resolution of these conflicts. Thus, the authors, using much of the theoretical underpinnings provided by Kydd and Walter (2006), suggest many specific ways that terrorism and civil war may overlap and provide empirical support for a connection between these concepts.
Causes of Terrorism within the Context of Civil War
Studies focused on attempting to explain the use of terrorism within civil wars typically take an action-based approach to terrorism. In this way, these studies consider terrorism as one type of violence that might occur within civil wars and attempt to explain the conditions under which such violence emerges. Most of these works also use rationalist arguments, based on strategic calculations, to explain the use of terrorism within civil wars. As might be expected given the difficulties involved, some of these studies approach either terrorism or civil war more tangentially rather than fully engaging in a discussion of these concepts and their overlap.
For example, Salehyan, Siroky, and Wood (2014) look at the question of why some rebel groups target civilians while others do not. They argue that rebel organizations that receive state funding are more likely to target civilians because they have less need of support from the population, but that this effect will depend on the characteristics of the external sponsor(s). Using a principal-agent approach, they consider how the characteristics of the external government (principal) affect the likelihood that the rebel organization (agent) will target civilians. They suggest that while sponsorship may generally increase the likelihood of targeting civilians, support by democratic governments, particularly with high numbers of human rights organizations, will make groups less likely to target civilians. Salehyan and colleagues use data on one-sided violence against noncombatants from the UCDP, which measures intentional and direct use of violence against civilians. Their results support their hypotheses. Again, the authors do not make any connection to terrorism as a concept; rather, they rely solely on the discussion of targeting civilians.
Instead of focusing on the characteristics of external countries, Stanton (2013) more explicitly examines why some rebel groups use terrorism within civil war while others do not, and she incorporates characteristics of the civil war state’s government into her analysis. She argues that there are a variety of strategic reasons for targeting civilians, and focuses specifically on violence targeting civilians that is aimed at coercing the opponent/government. Stanton’s definition of terrorism is important; she uses an action-based definition and excludes from her definition violence against government and military personnel not actively engaged in military operations; her sole focus is on violence against civilian targets. This is significant because it helps create a greater conceptual distinction between civil war and terrorism and allows the relationship between the two concepts to be considered more fully.
Similar to Crenshaw and others, Stanton uses a rationalist type of approach, discussing the strategic calculations made by groups. She points to two missing components in previous literature focusing on the strategic calculations of rebel groups. First, in considering and weighting the benefits of violence, she argues that rebel groups consider characteristics of their government and observe particularly whether the government is likely to make concessions in response to the violence. Some types of governments, like democracies, are likely to be more sensitive to civilian losses and therefore are more likely to make concessions. Second, rebel groups will consider the costs of using terrorism. Rebel groups have different political objectives, but those that have objectives requiring that they appeal to a broad civil population will likely be more adverse to the costs associated with targeting civilians. While a variety of scholars have discussed the link between regime type and terrorism, these scholars have either focused on transnational terrorism or have failed to fully consider the context of civil war.1 Stanton specifically makes the connection between regime type and terrorism within the context of civil war.
Stanton finds that rebel groups are more likely to use terrorism (targeting civilians) when the government is democratic. Further, she finds that while groups that have more inclusive political objectives are just as likely to use terrorism, they are less likely to use terrorism with high levels of casualties. Thus, she finds support for the idea that rebel groups are more likely to see the benefit of using terrorism against democracies, but also that groups that rely on greater public support are more sensitive to the high costs of using terrorism.
Political Instability and Weak States as a Cause of Terrorism
Another government characteristic that may affect the use of terrorism (both domestic and international) is political instability and state strength. Given that civil war is quite common in weak states and is a clear sign of political instability, it would make sense that civil war would be fully incorporated into studies on the link between political instability and terrorism. In fact, Crenshaw (1981) discusses weakening government and weakening the government’s control of territory as goals of terrorism. Unfortunately, most of these works consider political instability broadly, utilizing civil war as one potential measure of this instability, and provide scant discussion of the theoretical connections between political instability, state strength, civil war, and terrorism.
Looking only at international terrorism, Lai (2007) argues that civil war impacts a government’s ability to control its territory, and will lower the costs of terrorism by reducing the fear of government reprisals. He finds that civil war, among other factors that weaken the strength of a state, increases the likelihood of international terrorism. Campos and Gassebner (2013), consider political instability more generally, analyzing whether civil war might increase the likelihood of international terrorism. The authors specifically separate terrorist acts from civil wars, guerrilla warfare, and riots, arguing that because terrorist attacks target civilians and not military targets they are different from these other types of political violence. The authors use various types of political instability, including civil war and guerrilla warfare, as indicators of political instability and test whether these factors lead to international terrorism. Campos and Gassebner find that civil war and guerrilla warfare, as well as riots, increase the likelihood of international terrorism. While the authors do a good job of separating civil war from terrorism, their study is theoretically limited. The theoretical connections are broadly related to political instability and terrorism, and there is limited discussion about how different types of instability and violence might have different effects. Another shortcoming is the measure and discussion of terrorism. The authors purposely focus only on international terrorist acts, but it would be useful to compare domestic and international terrorism, particularly in relation to how such acts might be impacted by civil war.
Fahey and LaFree (2015) consider whether social disorganization and state instability (including revolutionary and ethnic war, adverse regime change, and genocide) increase terrorist attacks but also fail to fully theorize and specify the relationship between civil war (and various types of war) and terrorism. Similarly, Coggins (2015) considers the links between civil war and terrorism, though this is a secondary focus of her analysis. However, while she is primarily looking at the relationship between state failure and terrorism, there are important theoretical linkages to civil war. She argues that some state failings are more likely to generate conditions that are conducive to terrorism. Most importantly, she argues that “war fighting’s relationship to terrorism has not received sufficient theoretical attention” (p. 457). As she notes, “[c]ontemporary civil and international wars routinely feature an insurgent strategy utilizing terrorism as a tactic” (p. 457). In this way, terrorism is linked to civil war (and international wars) as a tactic used during such conflicts.
Coggins includes the concept of “war” as a factor in the political collapse of a state. In her discussion of different types of state failure, she includes civil war as an example of political collapse. Specifically, she identifies civil war as a control variable in her study, and not as a primary variable of interest. As such, there is inadequate theorizing about the full nature of the relationship between civil war and terrorism, though, not surprisingly, she finds that civil war has a significant positive effect on the likelihood of terrorism. Furthermore, Coggins uses the RAND Database of Worldwide Terrorism Incidents (RDWTI), which includes all acts of terrorism and does not distinguish the identity of the perpetrators or the cause. While the definition of terrorism includes civilians as targets, the perpetuator could be a group or individual and is not limited to acts that originated from domestic groups (and thus includes acts of transnational terrorism and domestic terrorism). Including and discussing both domestic and transnational acts of terrorism could be useful, but without separating them out, it is difficult to fully assess how civil war might affect terrorism, particularly since the context of civil war might have different effects for different types of terrorism and terrorist motives. Thus, while Coggins notes the importance of uniting the scholarly studies of civil war and terrorism, her connection of these concepts lacks specificity and needs additional theorizing and empirical testing.
A clear critique of all of this research focusing on political stability and state strength is that civil war is treated more tangentially, and often as a control variable, rather than being fully incorporated into the theoretical discussion and empirical model. Because civil war may impact both political instability and terrorism, further research needs to better parse out these theoretical and empirical connections.
Outbidding as a Cause of Terrorism in Civil Wars
Group-level factors are also considered an important element in determining the use of terrorism in civil wars. One cause of terrorism that has received a lot of attention is the idea of outbidding or competition between various groups. While much of the literature focusing on outbidding and competition does not specifically discuss civil war as a context, the theoretical linkages are clear, and increasingly scholars are connecting these concepts.
Bloom (2005) suggests that insurgent groups often splinter into multiple factions, and this division creates outbidding between rival groups that may increase the likelihood of domestic terrorism, particularly suicide terrorism. Bloom’s argument focuses on a specific type of terrorism, suicide terrorism, and moves beyond some other works, such as Kydd and Walter (2002, 2006), in that she looks at the importance of public opinion in affecting the use of terrorism by groups. Bloom argues that groups are playing to both domestic and international audiences in using suicide terrorism and that groups must compete for this support. Not all domestic political contexts will make outbidding likely. Bloom approaches the choice to use suicide terrorism as a cost-benefit analysis on the part of groups, and the costs will be affected by how the government responds and by the government’s policies, as well as by the social acceptability of these tactics. While Bloom is not necessarily only discussing terrorism within the context of civil war, several of the cases she discusses, such as Palestine, Sri Lanka, and Turkey, take place within civil wars and, thus, her theory has clear applicability to the civil war context but does not fully address the elements of civil war and how that might affect the various calculations of the actors.
Findley and Young (2012a) consider the theory of outbidding more specifically within countries engaged in armed conflict and move beyond case studies. Utilizing data from the GTD, they analyze whether having a greater number of opposition groups within an armed conflict will make suicide terrorism, or terrorism more generally, more likely. Furthermore, Findley and Young build on Bloom’s idea that certain factors, such as social acceptability, will provide conditions where outbidding will be most likely. One such factor is identity. Bloom considers the idea that it is easier to attack the other side when its members are from a different ethnic group, religion, or race. Findley and Young also include ethnic and cultural differences to see if outbidding is more likely under these circumstances. They find little support for the idea that more opposition groups (and thus, more groups competing for the public’s support) affect the use of suicide terrorism or terrorism more generally. In terms of identity, they find little statistical significance that outbidding is more likely when there are more ethnic and cultural groups. They do find, however, that in cases that have the highest number of minority groups, there is actually a lower likelihood of terrorism. The authors also find some support that outbidding might be important within the Israel case.
Nemeth (2014) advances a more nuanced analysis of the issue of outbidding, looking at the specific instances when this type of process might occur. He agrees with Bloom (2005) that outbidding will be most likely take place in cases where a favorable government policy and social acceptability of violence are present. In cases where these conditions are not present, Nemeth predicts a lower likelihood of terrorism. Though not limited to civil wars, Nemeth uses civil war as a control variable, expecting civil war to have a positive effect on the number of domestic terrorist attacks because governments will be less able to control their territory and will have reduced ability to combat terrorism; the presence of civil war will allow groups to use more of their resources on terrorism. Using the GTD, Nemeth finds support for his theory regarding the outbidding process and for civil war as a predictor of terrorism. While this work is interesting and provides a more nuanced look at the process of outbidding, its treatment of civil war is underwhelming. Utilizing civil war as a control variable is a start, but it doesn’t allow for parsing out how outbidding might work differently within and outside of this civil war context.
Effects of Using Terrorism in Civil Wars
A general debate in the terrorism literature concerns the effectiveness of terrorism. Much of this literature does not necessarily focus on terrorism within the context of civil war, or does it do so only tangentially. For example, Pape (2003, 2005) suggests that suicide terrorism has generally been successful in gaining concessions and points to several cases where groups utilizing suicide terrorism were able to gain concessions against what they might see as “foreign occupiers.” Some of these cases involve external occupations; others are more domestic in nature and would fall within the realm of civil wars (Sri Lanka, Turkey, Israel, and India are examples). Pape further argues that suicide terrorism is particularly successful against democracies. Similarly, Kydd and Walter (2006) contend that terrorists are generally successful, though again they do not consider civil wars specifically. Others have found little support for Pape’s findings (Abrahms, 2006, 2012; Abrahms & Gottfried, 2016), but have not considered terrorism within civil wars.
Fortna (2015) addresses this shortcoming by considering the effectiveness of terrorism within a civil war context. Specifically, she compares the outcomes of civil wars to see if groups that use terrorism are more likely to achieve their objectives. Fortna uses a rationalist approach and considers the various advantages and disadvantages of using terrorism, and some of the ways in which various scholars have suggested that it works, such as attrition, provocation, outbidding, and spoiling (see Kydd & Walter, 2006). Overall, she argues that the advantages of terrorism are more likely to help rebel groups survive rather than to win. She finds that while civil wars that involve terrorism do, in fact, last longer than those that do not involve terrorism, rebel groups that utilize terrorism as a tactic are generally less likely to achieve their larger political objective than groups that avoid the use of terrorism as a tactic.
Thomas (2014) moves beyond the achievement of objectives to consider whether groups that use terrorism in civil wars are more likely to be offered greater concessions and to be included in negotiations. Using monthly data on African civil conflicts from 1989 to 2010 (from the UCPD), Thomas finds that groups employing a greater number of terror attacks are more likely to participate in negotiations and more likely to gain more concessions on their aims.
Terrorism might be “effective” in these cases, Thomas argues, because the use of terrorism by groups within a civil war undermines the states’ ability to win the conflict, more specifically by providing rebels with an asymmetric “power to hurt.” The idea is that terrorism can hurt governments in a way that they may be unable to reciprocate. The use of terrorism as a tactic can demonstrate that rebels are willing to impose great costs on those that support the state, and that the state is unable to protect these individuals. This may even lead civilians to seek protection through rebel groups, causing the state to lose civilian support. Such a defection may also occur if the government uses excessive force in response to terrorism. In such a situation, where the state has lost popular support and is experiencing greater costs from rebel groups owing to their use of terrorism as a tactic, the government may be more inclined to cooperate with rebel groups, bringing them in to negotiations and bargaining. As in most of the other articles discussed in this article, Thomas utilizes rationality-based arguments, looking at the potential costs and benefits associated with using terrorism. More specifically, his theory is similar to the idea of attrition presented by Kydd and Walter (2006) and the idea of weakening the government as discussed by Crenshaw (1981). Thomas (2014) finds that groups that employ more terror attacks are more likely to negotiate with their governments and that strong rebels are more likely to be selected for negotiations than weaker rebels. Further, such rebels are also likely to gain more concessions. By incorporating the civil war context, Thomas provides a stronger theoretical argument relating to the effectiveness of terrorism that is not limited to international campaigns.
While Thomas (2014) focuses on how terrorism might affect whether governments grant concessions or negotiate with groups, Findley and Young (2015) consider the effects of terrorism on war resolution. They point to the theory of spoiling and analyze whether terrorism might help to derail peace efforts (see Kydd & Walter, 2002, 2006). Again, this work focuses more on rationalist explanations to explain how terrorism might spoil the peace process, as they argue that violent behavior effectively undermines peace because it creates distrust among parties, may provoke harsh responses from the government (provocation), and generally decreases the ability of the actors to make credible commitments to peace. As such, they expect more frequent terrorism to lead to longer wars and shorter postwar peace periods. Utilizing the GTD, the authors find that increased terrorism does increase the duration of civil wars and that as terrorism increases, the duration of time until war reoccurs also decreases. In other words, terrorism is found to have a spoiling effect on peace, both preventing peace from occurring in the first place and undermining peace agreements.
Pospieszna and DeRouen (2017) also analyze whether terrorism, and specifically violence against civilians, affects mediation efforts and peace. These authors build on the ideas found in Thomas (2014), as well as the theoretical foundations provided by Crenshaw (1981) and Kydd and Walter (2006). Pospieszna and DeRouen argue that the use of terrorism increases the likelihood of mediation, for terrorism can demonstrate the state’s illegitimacy and inability to provide security, while also potentially provoking a harsh response from the government that may encourage greater civilian support of groups utilizing terrorism. In such circumstances, the state may be willing to work with these groups for a peaceful solution to the conflict. However, Pospieszna and DeRouen also argue that terrorism increases the stakes of a conflict and may make mediation more difficult; they say that there may be incentives for groups (or splinter groups) to attempt to spoil this process, particularly if the mediation process is not working in their advantage. As such, Pospieszna and DeRouen argue that mediation may also increase the likelihood of rebel violence targeting civilians. Thus, terrorism may lead to mediation, but it may also be the result of mediation. The authors find support for both of their claims.
Butcher (2016) considers the broader effects of using terrorism within civil wars, specifically the effects of such tactics on foreign intervention into civil wars. As is true of in many of the other works previously discussed, Butcher focuses on rationalist arguments, and the costs associated with supporting either rebel groups or governments when terrorism is prevalent within a civil war. Butcher’s findings suggest that while rebel groups may use terrorism as a way to gain external support within civil wars, they are not successful. Further, in civil wars with large numbers of terrorist attacks, there is a greater likelihood that external actors will intervene on behalf of the government. Yet, suicide terrorism is found to potentially decrease the likelihood of external interventions on behalf of the government and on external diplomatic interventions. Thus, the potential costs associated with intervening in a case with greater levels of suicide terrorism may serve as a deterrent for external actors to aid the government or to intervene to diplomatically resolve the dispute. Butcher’s work is limited, however, to 1970–1999 and is unable to fully assess the potential impact of terrorism on intervention in a post–September 11 international climate.
Researchers are increasingly considering the ways that civil war and terrorism are connected, but they have only scratched the surface. There are many advances still to be made in relation to building and evaluating empirical theory. First, the research connecting terrorism and civil war has largely relied on a rationalist approach, but these studies are still very few in number. While these works have made some important contributions to the theoretical applications of theories such as attrition, spoiling, intimidation, outbidding, and provocation, many gaps remain and there are unanswered questions in terms of the applicability of these theories within the civil war context.
Furthermore, researchers have relied almost exclusively on these rationalist approaches and have not fully considered other theories that might help explain terrorism within a civil war context. Some recent research has been published attempting to broaden the theoretical scope of these connections. Zech and Gabbay (2016) utilize social network analysis to investigate how militants within a civil war interact with one another. While explanations such as outbidding and spoiling begin to get at these intergroup dynamics, Zech and Gabbay focus more extensively on social network analysis to examine the various patterns of relations and the social structure connecting groups within the boundaries of a civil war. Such newer approaches can help move beyond rationalist explanations and to fill in the gaps where these explanations fall short.
Finally, greater theorizing is needed regarding the relationship between civil war and terrorism and the way they interact with and influence one another. Such a discussion is imperative not only to fully evaluate the causes of terrorism (and why some groups use terrorism and others do not), but also to understand the effects and consequences of terrorism. Current research tends to adopt various theories related to the strategies of terrorism, but it does not fully consider how the civil war context itself might affect these theories. This also pertains to a larger discussion of the interactions between international and domestic terrorism, both of which can be affected by or take place within a civil war context. In this broader theoretical discussion, it will be important to continue to consider how terrorism and civil war might be conceptualized, defined, and measured in ways that will make considering their relationship increasingly possible. Overall, although there is increasing research on the connection between terrorism and civil war, there are still many gaps in this young research agenda that provide fruitful areas for scholars of terrorism and civil war.
Abrahms, M. (2006). Why terrorism does not work. International Security, 31(2), 42–78.Find this resource:
Abrahms, M. (2012). The political effectiveness of terrorism revisited. Comparative Political Studies, 45(3), 366–393.Find this resource:
Abrahms, M., & Gottfried, M. S. (2016). Does terrorism pay? An empirical analysis. Terrorism and Political Violence, 28(1), 77–89.Find this resource:
Asal, V., de la Calle, L., Findley, M., & Young, J. (2012). Killing civilians or holding territory? How to think about terrorism. International Studies Review, 14(3), 475–497.Find this resource:
Bloom, M. (2005). Dying to kill: The allure of suicide terrorism. New York: Columbia University Press.Find this resource:
Boulden, J. (2009). Terrorism and civil wars. Civil Wars, 11(1), 5–21.Find this resource:
Butcher, C. (2016). Terrorism and external audiences: Influencing foreign intervention into civil wars. Terrorism and Political Violence, 28(4), 774–794.Find this resource:
Campos, N. F., & Gassebner, M. (2013). International terrorism, domestic political instability, and the escalation effect. Economics and Politics, 25(1), 27–47.Find this resource:
Chenoweth, E. (2013). Terrorism and democracy. Annual Review of Political Science, 16, 355–378.Find this resource:
Coggins, B. (2015). Does state failure cause terrorism? An empirical analysis (1999–2008). Journal of Conflict Resolution, 59(3), 455–483.Find this resource:
Crenshaw, M. (1981). The causes of terrorism. Comparative Politics, 13(4), 379–399.Find this resource:
Cunningham, D. E., Gleditsch, K. S., & Salehyan, I. (2016). Trends in civil war data: Geography, organizations, and events. In T. D. Mason & S. M. Mitchell (Eds.), What do we know about civil wars? (pp. 247–260). Lanham, MD: Rowman & Littlefield.Find this resource:
Fahey, S., & LaFree, G. (2015). Does country-level social disorganization increase terrorist attacks? Terrorism and Political Violence, 27(1), 81–111.Find this resource:
Findley, M. G., & Young, J. K. (2012a). More combatant groups, more terror?: Empirical tests of an outbidding logic. Terrorism and Political Violence, 24, 706–721.Find this resource:
Findley, M. G., & Young, J. K. (2012b). Terrorism and civil war: A spatial and temporal approach to a conceptual problem. Perspectives on Politics, 10(2), 285–305.Find this resource:
Findley, M. G., & Young, J. K. (2015). Terrorism, spoiling, and the resolution of civil wars. Journal of Politics, 77(4), 1115–1128.Find this resource:
Fortna, V. P. (2015). Do terrorists win? Rebels’ use of terrorism and civil war outcomes. International Organization, 69, 519–556.Find this resource:
Gleditsch, N. P., Wallensteen, P., Eriksson, M., Sollenberg, M., & Strand, H. (2002). Armed conflict 1946–2001: A new dataset. Journal of Peace Research, 39(5), 615–637.Find this resource:
Kalyvas, S. N. (2004). The paradox of terrorism in civil war. Journal of Ethics, 8(1), 97–138.Find this resource:
Kydd, A., & Walter, B. F. (2002). Sabotaging the peace: The politics of extremist violence. International Organization, 56(2), 263–296.Find this resource:
Kydd, A. H., & Walter, B. F. (2006). The strategies of terrorism. International Security, 31(1), 49–80.Find this resource:
Lai, B. (2007). “Draining the swamp”: An empirical examination of the production of international terrorism, 1968–1998. Conflict Management and Peace Science, 24, 297–310.Find this resource:
Mason, T. D., & Mitchell, S. M. (Eds.). (2016). What do we know about civil wars? Lanham, MD: Rowman & Littlefield.Find this resource:
Melander, E., Pettersson, T., & Themnér, L. (2016). Organized violence, 1989–2015. Journal of Peace Research, 53(5), 727–742.Find this resource:
Mickolus, E. F., et al. (2007, November 26). International terrorism: Attributes of terrorist events. Dunn Loring, VA: Vinyard Software.Find this resource:
National Consortium for the Study of Terrorism and Responses to Terrorism (START). (2016). Global Terrorism Database [Data file]. Retrieved from https://www.start.umd.edu/gtd.
Nemeth, S. (2014). The effect of competition on terrorist group operations. Journal of Conflict Resolution, 58(2), 336–362.Find this resource:
Pape, R. (2003). The strategic logic of suicide terrorism. American Political Science Review, 97(3), 343–361.Find this resource:
Pape, R. (2005). Dying to win: The strategic logic of suicide terrorism. New York: Random House.Find this resource:
Pospieszna, P., & DeRouen, K., Jr. (2017). Civil war mediation and rebel use of violence against civilians. Armed Forces and Society, 43(3), 500–522.Find this resource:
RAND Database of Worldwide Terrorism Incidents (RDWTI). (2016). [Data file]. Retrieved from http://smapp.rand.org/rwtid.
Salehyan, I., Siroky, D., & Wood, R. M. (2014). External rebel sponsorship and civilian abuse: A rebel principal-agent analysis of wartime atrocities. International Organization, 68, 633–661.Find this resource:
Sambanis, N. (2008). Terrorism and civil war. In P. Keefer & N. Loayza (Eds.), Terrorism, economic development, and political openness (pp. 174–206). New York: Cambridge University Press.Find this resource:
Sanchez-Cuenca, I. (2014). Why do we know so little about terrorism? International Interactions, 40, 590–601.Find this resource:
Sanchez-Cuenca, I., & de la Calle, L. (2009). Domestic terrorism: The hidden side of political violence. Annual Review of Political Science, 12, 31–49.Find this resource:
Sarkees, M. R., & Wayman, F. (2010). Resort to war: 1816–2007. Washington, DC: CQ Press.Find this resource:
Stanton, J. A. (2013). Terrorism in the context of civil war. Journal of Politics, 74(4), 1009–1022.Find this resource:
Thomas, J. (2014). Rewarding bad behavior: How governments respond to terrorism in civil war. American Journal of Political Science, 58(4), 804–818.Find this resource:
Zech, S., & Gabbay, M. (2016). Social network analysis in the study of terrorism and insurgency: From organizations to politics. International Studies Review, 19(2), 214–243.Find this resource: