Understanding Government Behavior during Armed Conflict
Summary and Keywords
How and why do governments choose the strategies that they do during armed conflict? While there is a substantial body of research on the use of different tactics by governments and rebels during armed conflict, this work has rarely made an attempt to engage with scholars of different tactics in order to develop a broader understanding of how and why actors make the behavioral choices that they do and how these choices advance certain intended strategies. Furthermore, the work across tactics does not have unified findings. Understanding both the range of behaviors by conflict actors and the motivations for these behaviors is an important and necessary step for understanding the course of armed conflict more generally and for developing relevant policy aimed at changing these behaviors.
Within existing literature on belligerent tactics, important lessons about government behaviors and why these specific behaviors are selected can be distilled. Objectives, strategies, and tactics should be disaggregated in order to think through the implications of different government decisions for understanding or changing behavior. This disaggregation aids us in identifying the areas of research in which we have confirmed findings and those questions regarding government behavior that require additional investigation. Moving forward we could and should develop a systematic list of the types of factors that impact certain behavioral choices, across tactics, but this is most useful if we can then link these factors to an understanding of the broader objectives and strategies that a government is trying to pursue.
The civil war in Syria began in 2011 following a political uprising linked to the protests of the Arab Spring. The Syrian government responded to this domestic threat with a policy of harsh repression against protesters including, “mass arrests, torture of detainees, deprivation of medical treatment for wounded protesters as well as the use of snipers and shooting with live ammunition into crowds” (Uppsala Conflict Data Program [UCDP], “Syria”). This program of harsh state repression against unarmed civilians contributed to the defection of members of the Syrian army and the creation of several insurgent groups, including the Free Syrian Army, whose goal was to overthrow the political regime of President Bashar al-Assad.
From the beginning of the conflict it has been clear that President Assad’s primary objective is to maintain personal power and control of Syria. As such, the Syrian government has focused on strategies to demonstrate state strength and control territory. The government has employed a variety of tactics to advance these strategies. Commonly, the regime has relied on the indiscriminate shelling of urban centers, particularly those towns and neighborhoods suspected of harboring rebels. The government has also laid siege to certain areas, “hindering the inflow of goods and supplies (such as medicines), as well as barring movement of people in and out of the besieged areas” (UCDP, “Syria”). The use of siege and shelling tactics has caused a high number of civilian casualties and heavy destruction of the country.
How and why did the Syrian government chose these tactics during the civil war? While we have developed a substantial body of literature on the different tactics employed by governments and rebels during armed conflict, works that focus on a specific tactic (e.g., work on sexual violence) have rarely made an attempt to engage with other research across tactics (e.g., work on massacres) in order to form a broader discussion about how and why actors make the behavioral choices that they do. Furthermore, this work has often focused exclusively on the tactical choices actors make without engaging in a broader discussion about that actor’s intended strategy and the government or rebel group’s overarching objective. Understanding both the range of behaviors and the motivations for these behaviors during armed conflict is an important and necessary step for understanding the course of armed conflict. Greater knowledge on these issues is particularly important for structuring relevant policy interventions aimed at influencing or ending certain behaviors such as the targeting of civilians.
The focus here is on the behavior of governments during armed conflict, in part, because extant empirical work has primarily emphasized the behavior of rebel groups (e.g., Bueno de Mesquita, 2013), ignoring or overlooking important behaviors and behavioral changes across government actors. In addition, government behavior may be an area in which the international community can have the most impact in changing or altering that behavior because of the role that government actors play in the international community. In this way, a better understanding of government behavior can have important and necessary implications for policy work. The term behavior refers here to the physical manifestation of a government’s conflict strategy through a compilation of tactics. In other words, this term encompasses the actions that a government takes in its interaction with a violent challenger. While the focus is on how governments respond to violent challengers, many of these arguments may apply equally well to governments challenged by nonviolent organizations.
While there is a well-developed body of literature in conflict studies that focuses on a specific arsenal of tactics, there is comparatively less work on objectives and strategies, and few scholars make the attempt to link these works. Furthermore, the work across tactics does not have unified findings. The existing literature on belligerent tactics is reviewed in order to distill important lessons about government behaviors and why these specific behaviors are selected. In particular, it disaggregates between objectives, strategies, and tactics in order to think through the implications of different government decisions for understanding or changing behavior. This disaggregation aids us in identifying the areas of research in which we have confirmed findings and those questions regarding government behavior that require additional investigation. Moving forward we could and should develop a systematic list of the types of factors that impact certain behavioral choices, across tactics, but this is most useful if we can then link these factors to an understanding of the broader objectives and strategies that a government is trying to pursue.
Government Behavior during Armed Conflict
Belligerent behavior during conflict is most often studied through the lens of the particular tactics that a group chooses and pursues during that conflict. These tactics can be used by the government or challenger group. They can be violent or nonviolent. They can be selectively, indiscriminately, or in some cases, collectively targeted at individuals or groups (Sullivan, 2012). Often a given actor uses a variety of tactics across time and space to different ends. For example, the use of sexual violence may be prevalent only in a certain area of the country and be committed only at a particular period in time, as in the case of Sierra Leone where the majority of reported incidents of wartime rape were committed by the Revolutionary United Front (RUF), despite the fact that the RUF made up roughly 34% of the combatants active during the war (Cohen, 2013).
This research on tactics has tended to concentrate on some key behaviors, such as the use of massacres and mass killing (Sullivan, 2012; Valentino, Huth, & Balch-Lindsay, 2004), torture (Conrad & Moore, 2010; Rejali, 2007), rape (Cohen, 2013; Cohen & Nordås, 2015; Leiby, 2009), and terrorism (Asal, Gill, Rethemeyer, & Horgan, 2015; Pape, 2005). The majority of these tactics of interest are violent; however, notable work has focused on nonviolent strategies such as the use of nonviolent resistance (Stephan & Chenoweth, 2008), covert state repression (Davenport, 2005), or the restriction of civil liberties (Davenport, 1996).
Despite extensive research on the use of different tactical behaviors on the part of government and rebels, there are two main areas of empirical and theoretical evaluation that are currently lacking. First, we are missing a systematic and comprehensive evaluation of the types of unifying factors that impact an actor’s decision to use all types of tactics and to switch between tactics. Research that theorizes a particular tactic (e.g., the use of civilian massacres) is often conducted in isolation from broader research that discusses other forms of civilian targeting such as sexual violence or enforced disappearances.1 This isolation is justified on the grounds that a particular tactic has a unique motivation from other forms of behavior; however, a unified assessment is needed in order to credibly make that claim. By failing to engage with a broader discussion of behavioral motivations we can say very little about why a particular tactic is chosen over another tactic, why that tactic was chosen for that location at a given point in time, or how different tactics are related via substitution.
Second, studies of individual tactics or broader patterns of actor behavior (such as civilian targeting) often lack a link to a more systematic understanding of the actor’s broader conflict strategy or objectives (Sullivan, 2012). For example, the finding that sexual violence is a tool for increasing group cohesion (Cohen, 2013) has the underlying assumption that cohesive groups are the most effective fighting structure for the broader goal of winning the war but does not explicitly lay out the trade-offs for press-ganging and the conditions under which governments and rebel groups benefit from cohesive units and when group cohesion is less necessary. In other words, we learn about the strong relationship between rape and group cohesion, without a broader understanding of the types of groups that would be more likely to need group cohesion in order to accomplish their particular conflict objectives. Understanding a conflict actor’s strategy and objectives for the war is a necessary step in predicting and ultimately preventing that behavior rather than encouraging substitution tactics. Furthermore, an understanding of the overall objectives of a conflict actor is essential for mediating or negotiating an end to ongoing conflict.
To address these limitations in the current theoretical and empirical literature, the universe of government behaviors is outlined via a discussion of the theoretical and empirical distinctions between an actor’s objective, strategy, and tactic.
In understanding government actions and activity during armed conflict, there are three overarching categories through which to view this behavior. First, a government2 has a principal objective that it is trying to achieve. In the rational choice literature, we assume objectives such as the desire to win the war or the desire for a particular individual or political party to stay in power (Fearon, 1995). While these objectives are often assumed and not tested, they are an important starting point for our future analysis, and certainly getting them wrong could have dire implications for our theories. In the Syrian example, President Bashar al-Assad has been clear that he is focused on maintaining office at all costs, costs that include the death of countless civilians. While it is theoretically possible for objectives to switch over time, we would expect this category of behavior to be comparatively static.
Understanding an actor’s objective gives us greater insight into why particular strategies are chosen during conflict. For example, Barnes (2005) distinguishes between two objectives, or what she calls “categories of intent” for the government use of genocide; namely, she investigates the differences between genocide as “predominantly instrumental tactics” versus a government who sees genocide as “an integral goal” (Barnes, 2005, p. 309). Barnes argues that the distinction between these two strategies is driven by the political, economic, and strategic factors which impact a government’s conflict objective. In this way, a government’s strategy results directly from its chosen objective(s). As an illustration, a government may choose a strategy of group destruction (or genocide) based on its goal to win a conflict through the elimination of a perceived threat (Valentino, 2005). A rational choice approach to understanding government behavior during conflict focuses on the cost-benefit calculus of certain actions. In this framework, governments are ultimately concerned with achieving the greatest possible benefit from the civil war (Fortna, 2004), though this calculus is often considered with an eye toward future challengers and outcomes (Walter, 2006). As there are a variety of strategies through which to achieve a given aim, governments select strategies weighing the costs and benefits of each option.
In his work on government use of massacres, Sullivan (2012) identifies two main conflict strategies: the elimination of a particular threat and the projection of state control. He calls these the “strategic settings” in which tactical decisions are made (Sullivan, 2012). Loyle (2016a) has written about the government use of judicial processes during armed conflict as linked to a strategy of demonstrating strength or addressing rebel grievances. Called during-conflict justice, these activities can be used in parallel with violent tactics to advance certain government strategies. Similarly, Wood (2010) argues that actors can switch between two strategies in their interaction with civilians, namely either a “carrot” or a “stick” approach. With a “carrot” strategy, the government or rebel group focuses on the provision of security and public goods. With a “stick” strategy, a group uses violent attacks or other forms of intimidation to advance their conflict objectives (Wood, 2010).
An alternative conception is to think about government strategies in terms of macro and micro policies (Gates, 2002; Loyle, 2016a). Governments can pursue macro strategies aimed at defeating a rebel group through undermining the group’s ability to rebel. Macro strategies involve tactics such as directly attacking the group militarily through conventional warfare, targeting the group’s supply lines or external funding, or assassinating rebel leaders. Governments could alternatively pursue micro strategies targeted at individual rebels and their decision to rebel. Micro strategies include tactics that address the motivations and opportunities that individuals have to join rebellions (Humphreys & Weinstein, 2008; Gates, 2002; Collier & Hoeffler, 2004; Regan & Norton, 2005), for example tactics that increase the likelihood of defection or reduce individual material or personal support for the rebel cause. Civil wars and rebellion are ultimately about the individuals who choose to fight in those conflicts. As Walter (2004) argues in regards to conflict recurrence, it is necessary to understand the micro motivations of individual farmers, shopkeepers, and workers to understand the perpetuation of conflict (p. 372). Micro strategies, therefore, directly target rebel resolve and decrease popular support for rebel groups, increasing defections and reducing the potential pool of rebel recruits and supporters.
Conflict strategies are advanced through a variety of tactics designed to increase the government’s chances of achieving its conflict objective. Once a government has chosen a particular strategy, tactics are selected to advance those aims. Tactics are by far the most researched and theorized component of government conflict behavior. This is in large part because this behavior is the easiest to identify, measure, and track over time. While intentions and motivations are important to theorize about, they have been historically difficult concepts for social scientists to gauge.
Governments challenged by violent groups have a variety of tactics at their disposal during their fight. In addition to direct combat, they can employ tactics such as sexual violence (Cohen & Nordås, 2015), torture (Conrad & Moore, 2010), terrorism and civilian targeting (Hultman, 2007; Pape, 2005), and enforced disappearances (Humphreys & Weinstein, 2006; Loyle, 2016b). Furthermore, in addition to violent tactics governments use nonviolent tactics aimed at influencing the course of the conflict and its outcome. Governments may engage in coercive nonviolent strategies such as surveillance (Davenport, 2005) or they may choose to employ conciliatory nonviolent actions, for example by engaging in peace negotiations (Fortna, 2004; Slantchev, 2003), granting concessions (Walter, 2006), offering public goods provisions (Wickman-Crowley, 1991), or adopting amnesty or reparation processes (Loyle & Binningsbø, 2017). In other words, different tactics are used in different contexts to potentially different ends.
Government tactics can be broadly characterized into those that selectively, indiscriminately, or collectively target challengers. Selective tactics are those that focus on the punishment of individuals who are directly in confrontation with the government. For example, targeted arrests or killings of rebels or individuals assisting the rebel cause. Here individuals are targeted based on personal characteristics, such as their participation in certain challenging behaviors. Indiscriminate tactics are those that target all civilians with equal probability, such as the indiscriminate shelling in Chechnya described by Lyall (2009) or the shelling of urban centers in Syria. In this case the government may be unable or unwilling to identify those individuals directly engaging in challenging behavior (Lyall, 2009), or the government may be trying to signal its strength more widely (Sullivan, 2012). Finally, Sullivan (2012) discuses a third category of tactics—collective targeting. Collective targeting refers to tactics, such as a state massacre, where the government targets victims based on group characteristics but not their direct involvement in the conflict.
Figure 1 maps the distinction and relationship between the three concepts of objective, strategy, and tactic and provides some examples of the behaviors across these three categories. It is important to note that the distinction between these categories is not rigid. For example, certain behaviors can be classified as either a tactic or a strategy, such as genocide, which is often argued to function as both (Barnes, 2005).
The theories and empirical findings that have been generated about government behavior can be broadly divided into those factors impacted by government characteristics, challenger characteristics, and those characteristics unique to the conflict itself. Across these categories, the relationships between these factors and government decisions regarding conflict objective, strategy, and tactic selection are examined.
Government Characteristics and Behavior during Armed Conflict
Existing work on the impact that the characteristics of a given government has on its behavior during armed conflict has primarily focused on two structural variables, namely regime type and government capacity.
Regarding regime type, it is argued that democratic governments pursue different objectives, and employ different strategies and tactics, from those of autocratic states (Colaresi & Carey, 2008). The presence of institutions to peacefully resolve conflict (Rummel, 1994) as well as the low-cost avenues through which citizens can sanction leaders who pursue personal or private objectives (Bueno de Mesquita, Smith, Siverson, & Morrow, 2003) impact the ways in which governments behave during armed conflict. Specifically, unstable political circumstances are likely to be handled differently in a democracy than in an authoritarian regime, helping to explain the conditions under which certain tactics, such as genocide, are more likely (Colaresi & Carey, 2008). For these reasons, democratic governments are likely to employ different objectives, strategies, and tactics from their nondemocratic counterparts.
The institutional characteristics of democratic governments also impact the tactical choices governments make (e.g., Davenport, 2007b). It is what Davenport refers to as the “voice” and “veto” channels within a democracy that determine how likely that state is to employ repressive strategies and tactics against its own citizens (Davenport, 2007b). Voice refers to the “pacifying influence of elections and the representation of diverse political orientations” (Davenport, 2007b, p. 13). Veto is made up of the “pacifying influence of checks and balances, executive constraints, and veto points/players” (Davenport, 2007b, p. 13). According to Davenport, democratic institutions increase the costs of using repression, democracies accept values that deter repression, and democracies provide alternative mechanisms for making challenges to the state less likely in the first place (Davenport, 2007b). Importantly, however, these institutions are found to be most effective in constraining government behavior when a democratic state is not facing a domestic threat (Davenport, Moore, & Armstrong, 2007). In countries where governments are less concerned or less constrained we should expect to see an increase in strategies and tactics that involve state repression and/or civilian targeting.
Similarly, in their work on interstate wars, Reiter and Stam (2002) find that democratic governments have a more conservative calculus in choosing to engage in certain violent behavior due to domestic audience costs. This finding has theoretical applications for domestic conflict as well. Furthermore, a free press and civilian sanctioning in democratic governments make this regime type more susceptible to naming and shaming campaigns by domestic and international civil society groups. While there is scant evidence on the ability of naming and shaming to permanently arrest certain conflict strategies or tactics, DeMerritt, Conrad, and Fariss (2017) find that naming and shaming causes governments to move from one tactic to another, in particular to another tactic that can be useful for hiding the named abuse. The impact of democratic institutions does not apply to all types of conflict behavior. For example, Eck and Hultman (2007) do not find an effect of democracy in the decision to commit one-sided violence against civilians.
Government capacity is another important indicator of government behavior during armed conflict. The overall level of development for a particular country and the resources available to a given government will determine the strategies and tactics at its disposal. For example, government capacity plays a role in how likely a government is to be violently challenged in the first place. Buhaug (2006) finds that wealthier countries are less likely to face armed opposition, but they are particularly unlikely to be the site of “large-scale revolutionary movements” (p. 703).
In regards to tactics, Sullivan (2012) finds that massacres in Guatemala were much less likely to occur in areas where the state had strong control. Cohen (2013) finds that sexual violence is more common among armed groups and government armies that have low social cohesion, in particular among groups which use press-ganging. Colaresi and Carey (2008), however, find that military capabilities are an inconsistent predictor of state-sponsored mass killings because strong militaries can be used to protect or repress civilians; therefore, it is necessary to understand the underlying strategies and ultimately the objectives of a government. Finally, Ahram (2013) argues that weak states are often able to delegate atrocity behavior to nonstate actors or pro-government militias who are able to carry out this violence and, therefore, state capacity is not a good proxy for the ability of governments to employ certain tactics.
Arguments about government capacity can also relate to the degree of command control the government holds over its security forces. For example, variation across fighting units predicts violence against civilians during the civil war in Sierra Leone (Humphreys & Weinstein, 2006). “Fighting units composed of individuals motivated by private goals, with high levels of ethnic diversity, and weak mechanisms to maintain internal discipline commit the highest levels of abuse” (Humphreys & Weinstein, 2006, p. 444). This suggests that lower capacity governments may be more prone to civilian abuse strategies based on their inability to choose policies requiring more discipline from their troops.
Challenger Characteristics and Government Behavior during Armed Conflict
In addition to the characteristics of the government, characteristics of the challenger that the government is fighting impact government behavior during armed conflict. In particular, scholars have noted the impact of the level of threat of the challenger (or the balance of power between the government and the rebel group), goals of the rebel group (e.g., the conflict incompatibility), as well as how the challenger group is behaving during the conflict.
Balance of Power
The objective, strategy, and tactics a government chooses to pursue has been found to be a product of the interaction between the government and rebels (Kalyvas & Balcells, 2010); in particular, government behavior is related to the level of threat a challenger poses to the government. The belligerents’ relative capabilities and the marginal utility for fighting are important indicators of combatant behavior on both sides of the conflict (Goemans, 2000; Cunningham, Gleditsch, & Salehyan, 2009; Buhaug, Gates, & Lujala, 2009) and the subsequent behavior of the government.
Cunningham et al. (2009) analyze the balance of power between government and rebels and find that strong rebels are able to evoke concessionary strategies from the government. Valentino, Huth, and Balch-Lindsay (2004) find that the intentional killing of civilians during civil war is often a calculated military strategy designed to combat powerful guerrilla insurgencies (Valentino et al., 2004, p. 375). The authors further argue that mass killing of civilians in guerrilla warfare often emerges out of frustration with conventional tactics in an effort to stave off defeat (Valentino et al., 2004, p. 377). In other words, governments have little incentive to resort to mass killing if they believe a less violent strategy would be effective (Valentino et al., 2004, p. 386).3
Performance on the battlefield has also been found to impact the selection of targets for specific tactics. In her study of rebel group behavior, Hultman (2007) argues that rebel groups use civilian casualties to increase the costs imposed on the government when the rebel group is losing the war. Clayton (2013) finds that relatively strong rebels are more likely to overcome the strategic bargaining problems that prevent the end of conflict. In this case, weaker governments (vis-à-vis the rebels) may be more willing to enter into negotiations or employ other types of concessionary strategies. These works suggest that the relative capacity of a government impacts the strategy, military or otherwise, that the government is likely to pursue.
In addition to the relative strength of a government vis-à-vis its challengers, the challenger group’s goals and the number of challengers that a government faces have been found to impact government strategy, particularly in regard to the likelihood of adopting concessionary behavior. Walter (2006) argues that governments may choose particularly hardline strategies, such as a lack of willingness to negotiate or make concessions, when there is a chance that in the future a government will be faced with other groups with similar demands. In regards to the study of self-determination movements, however, Cunningham (2014) argues that governments may chose concessionary strategies to reveal information about rebel groups—increasing the chances of a favorable position.
Finally, the behavior of the challenger group, for example how they fight and how they mobilize, has been determined to have an impact on government behavior during armed conflict. In their study of the Peruvian Civil War, Fielding and Shortland (2012) find that civilian targeting by one side of the conflict was associated with subsequent civilian targeting by the other. Work on one-sided violence against civilians, for example, finds that targeting civilians is related to current conflict-intensity rather than previous levels of violence in the country (Eck & Hultman, 2007). In his work on Chechnya, Lyall (2009) finds that insurgent tactics are related, under certain conditions, to a government’s decision to use indiscriminant violence.
How a challenger group mobilizes has also been found to impact government behavior, for example, the level of support that group has from the population or the ethnic composition of the group. Kalyvas (1999) argues that civilian massacres can be a rational tool for maximizing support for the government under a given set of circumstances. Valentino et al. (2004) find this to be particularly true when governments face guerrilla insurgences. Because guerrilla forces are difficult to defeat directly, governments have strong incentives to target the guerrillas’ civilian base of support and often rely on the strategy of mass targeting of civilians, employing the tactic of civilian massacres (Valentino et al., 2004, p. 375).
Conflict Characteristics and Government Behavior during Armed Conflict
Finally, it is likely that characteristics unique to the conflict itself impact a government’s behavior during that conflict. In particular, the literature has focused on the geography of an area and natural resource availability.
The geography of a particular area can make it more difficult to get information about civilian loyalties and the conflict itself impacting the types of strategies and tactics a government is likely to choose (Wood, 2010). Furthermore, difficult geographies can increase the principal-agent problem (Gates, 2002). Buhaug, Gates, and Lujala (2009) find that geography significantly affects the duration of armed conflict by impacting the relative capabilities of the belligerent parties, in particular, by impacting the dynamics of a particular conflict (p. 566). Mason and Fett (1996) investigate the conditions under which a government would be most likely to negotiate and find that conflict characteristics are an important determinant of that behavior.
Natural resources impact the strategies and tactics that a government employs and also the overarching objectives of the conflict, as maintaining control of particular areas may require different strategic concerns than simply maintaining power more broadly. Ross (2004) finds evidence to suggest that the presence of natural resources can impact the intensity of the conflict as it contributes to “resource battles” over access to natural resources. Alternatively, Ross (2004) found partial evidence for “cooperative plunder,” in which the presence of resources made it more likely for the government and challenger group to cooperate over the extraction of certain resources, creating a kind of “commercial equilibrium” (p. 56).
Moving Forward in the Study of Government Behavior during Armed Conflict
Understanding government behavior during armed conflict has significant implications for the interventions that we use to impact or modify that behavior. Despite the importance of understanding the tactics and overall strategies that governments use to pursue certain war objectives, research has rarely engaged with a unified understanding of how these concepts are related and the potential impact that these meta-concepts of behavior have for our understanding of the more micro-tactical decisions that governments make during armed conflict. Linking our understanding of how different tactics relate to certain strategies and how these strategies are part of broader government objectives can have important implications for research. In particular, understanding the often assumed meta-concepts of behavior can help us to address such issues as the substitution of tactics and how the use of certain tactics are interlinked and potentially complementary.
While it is potentially useful to understand the larger motivations for government behavior, such as government objectives, we have comparatively little research on these concepts. Some of the research on tactics can be applied to our understanding of strategies and objectives, but there has been little systematic attempt to link these broader patterns of motivations with behavioral outcomes. However, it is clear that a systematic understanding of government behavior during armed conflict must engage with these broader objectives in order to evaluate the underlying objectives behind different government strategies.
Table 1 charts some of the findings from the literature highlighted, noting the concepts in which there has been little theoretical or empirical work or inconsistences across the findings.
Table 1. Mechanisms that impact government objectives, strategy, and tactics during armed conflict
Democratic norms (Davenport, 2007a)
Indivisible goods (Cunningham, 2014)
Losing with conventional tactics (Valentino, Huth, & Balch-Lindsay, 2004)
Military resources available for combat (Ross, 2004)
Impact of the strength of “voice” (Ross, 2004)
Table 1 demonstrates that while we have focused on some factors that have impacts across various types of government decisions, we are lacking in a systematic understanding of other influences. For example, work on regime type, specifically the behavior of democratic governments during armed conflict, has investigated the impact of democratic institutions, audience costs, and democratic norms on the objectives, norms, and tactics that governments choose. However, we have been less thorough in our understanding of how challenger and conflict characteristics condition these behaviors. While we know that the level of threat against the government softens the impact of democratic institutions, we have been less systematic in laying out the nuances of that threat, bringing in literature on challenger characteristics, for example, to discuss the domestic democratic peace. Moving forward, work on government behavior should address the distinct and overlapping causal logic across the selection of given tactics as well as the ways in which these causal logics impact broader decisions regarding government objectives and strategies.
Ahram, A. I. (2013). The role of state-sponsored militias in genocide. Terrorism and Political Violence, 26(3), 488–503.Find this resource:
Asal, V., Gill, P., Rethemeyer, R. K., & Horgan, J. (2015). Killing range: Explaining lethality variance within a terrorist organization. Journal of Conflict Resolution, 59(3), 401–427.Find this resource:
Barnes, C. (2005). The functional utility of genocide. Journal of Genocide Research, 7(3), 309–330.Find this resource:
Bueno de Mesquita, B., Smith, A., Siverson, R., & Morrow, J. (2003). The logic of political survival. Cambridge, MA: MIT Press.Find this resource:
Bueno de Mesquita, E. (2013). Rebel Tactics. Journal of Political Economy, 121(2), 323–357.Find this resource:
Buhaug, H. (2006). Relative capability and rebel objective in civil war. Journal of Peace Research, 43(6), 691–708.Find this resource:
Buhaug, H., Gates, S., & Lujala, P. (2009). Geography, rebel capability, and the duration of civil conflict. Journal of Conflict Resolution, 53(4), 544–569.Find this resource:
Clayton, G. (2013). Relative rebel strength and the onset and outcome of civil war mediation. Journal of Peace Research, 50(5), 609–622.Find this resource:
Cohen, D. K. (2013). Explaining rape during civil war: Cross-national evidence (1980–2009). American Political Science Review, 107(3), 461–477.Find this resource:
Cohen, D. K., & Nordås, R. (2015). Do states delegate shameful violence to militias? Patterns of sexual violence in recent armed conflicts. Journal of Conflict Resolution, 59(5), 877–898.Find this resource:
Colaresi, M., & Carey, S. C. (2008). To kill or to protect. Journal of Conflict Resolution, 52(1), 39–67.Find this resource:
Collier, P., & Hoeffler, A. (2004). Greed and grievance in civil war. Oxford Economic Papers, 56, 563–595.Find this resource:
Conrad, C. R., & Moore, W. H. (2010). What stops the torture? American Journal of Political Science, 54(2), 459–476.Find this resource:
Cunningham, D. E., Gleditsch, K. S., & Salehyan, I. (2009). It takes two: A dyadic analysis of civil war duration and outcome. Journal of Conflict Resolution, 53(4), 570–597.Find this resource:
Cunningham, K. G. (2014). Inside the politics of self-determination. Oxford: Oxford University Press.Find this resource:
Davenport, C. A. (1996). “Constitutional promises” and repressive reality: A cross-national time-series investigation of why political and civil liberties are suppressed. Journal of Politics, 58(3), 627–654.Find this resource:
Davenport, C. (2005). Understanding covert repressive action: The case of the U.S. government against the Republic of New Africa. Journal of Conflict Resolution, 49(1), 120–140.Find this resource:
Davenport, C. (2007a). State repression and political order. Annual Review of Political Science, 10, 1–23.Find this resource:
Davenport, C. (2007b). State repression and the domestic democratic peace. Cambridge, U.K.: Cambridge University Press.Find this resource:
Davenport, C., Moore, W. H., & Armstrong, D. (2007). The puzzle of Abu Ghraib: Are democratic institutions a palliative or panacea?.
DeMerritt, J., Conrad, C., & Fariss, C. (2017). Hiding human rights violations in response to UN naming and shaming. Working Paper.Find this resource:
Eck, K., & Hultman, L. (2007). One-sided violence against civilians in war: Insights from new fatality data. Journal of Peace Research, 44(2), 233–246.Find this resource:
Fearon, J. D. (1995). Rationalist explanations for war. International Organization, 49(3), 379–414.Find this resource:
Fielding, D., & Shortland, A. (2012). The dynamics of terror during the Peruvian Civil War. Journal of Peace Research, 49(6), 847–862.Find this resource:
Fortna, V. P. (2004). Does peacekeeping keep peace? International intervention and the duration of peace after civil war. International Studies Quarterly, 48(2), 269–292.Find this resource:
Gates, S. (2002). Recruitment and allegiance: The microfoundations of rebellion. Journal of Conflict Resolution, 46(1), 111–130.Find this resource:
Gibney, M., Cornett, L., Wood, R., Haschke, P., & Arnon, D. (2016). The political terror scale 1976–2015.
Goemans, H. E. (2000). War and punishment: The causes of war termination and the First World War. Princeton, NJ: Princeton University Press.Find this resource:
Hultman, L. (2007). Battle losses and rebel violence: Raising the costs for fighting. Terrorism and Political Violence, 19(2), 205–222.Find this resource:
Humphreys, M., & Weinstein, J. M. (2006). Handling and manhandling civilians in civil war. American Political Science Review, 100(3), 429–447.Find this resource:
Humphreys, M., & Weinstein, J. M. (2008). Who fights? The determinants of participation in civil war. American Journal of Political Science, 52(2), 436–455.Find this resource:
Kalyvas, S. N. (1999). Wanton and senseless? The logic of massacres in Algeria. Rationality and Society, 11(3), 243–285.Find this resource:
Kalyvas, S. N., & Balcells, L. (2010). International system and technologies of rebellion: How the end of the Cold War shaped internal conflict. American Political Science Review, 104(3), 415–429.Find this resource:
Leiby, M. L. (2009). Wartime sexual violence in Guatemala and Peru. International Studies Quarterly, 53(2), 445–468.Find this resource:
Loyle, C. E. (2016a). Justice during armed conflict: Addressing grievance and opportunities for rebellion. Working Paper. American Political Science Association Annual Conference, September 1–4, 2016.Find this resource:
Loyle, C. E. (2016b). Without a trace: Enforced disappearance as a strategy during armed conflict. Working Paper. Peace Science Society International Annual Conference, October 21–22, 2016.Find this resource:
Loyle, C. E., & Binningsbø, H. M. (2017). Justice during armed conflict: A new dataset on government and rebel strategies. Journal of Conflict Resolution.Find this resource:
Lyall, J. (2009). Does indiscriminate violence incite insurgent attacks? Evidence from Chechnya. Journal of Conflict Resolution, 53(3), 331–362.Find this resource:
Mason, T. D., & Fett, P. J. (1996). How civil wars end a rational choice approach. Journal of Conflict Resolution, 40(4), 546–568.Find this resource:
Pape, R. (2005). Dying to win: The strategic logic of suicide terrorism. New York: Random House.Find this resource:
Regan, P., & Norton, D. (2005). Greed, grievances, and mobilization in civil wars. Journal of Conflict Resolution, 49(3), 319–336.Find this resource:
Reiter, D., & Stam, A. C. (2002). Democracies at war. Princeton, NJ: Princeton University Press.Find this resource:
Rejali, D. M. (2007). Torture and democracy. Princeton, NJ: Princeton University Press.Find this resource:
Ross, M. L. (2004). How do natural resources influence civil war? Evidence from thirteen cases. International Organization, 58(1), 35–67.Find this resource:
Rummel, R. J. (1994). Death by government. New Brunswick, NJ: Transaction Publishing.Find this resource:
Slantchev, B. (2003). The principle of convergence in wartime negotiations. American Political Science Review, 97(4), 621–632.Find this resource:
Stephan, M. J., & Chenoweth, E. (2008). Why civil resistance works: The strategic logic of nonviolent conflict. International Security, 33(1), 7–44.Find this resource:
Sullivan, C. M. (2012). Blood in the village: A local-level investigation of state massacres. Conflict Management and Peace Science, 29(4), 373–396.Find this resource:
Uppsala Conflict Data Program. Government of Syria—Syrian insurgents.
Valentino, B. A. (2005). Final solutions: Mass killings and genocide in the 20th century. Ithaca, NY: Cornell University Press.Find this resource:
Valentino, B., Huth, P., & Balch-Lindsay, D. (2004). “Draining the sea”: Mass killing and guerrilla warfare. International Organization, 58(20), 375–401.Find this resource:
Walter, B. F. (2004). Does conflict beget conflict? Explaining recurring civil war. Journal of Peace Research, 41(3), 371–388.Find this resource:
Walter, B. F. (2006). Building reputation: Why governments fight some separatists but not others. American Journal of Political Science, 50(2), 313–330.Find this resource:
Wickman-Crowley, T. (1991). Exploring revolution: Essays on Latin American insurgency and revolutionary theory. Armonk, NY: M. E. Sharpe.Find this resource:
Wood, R. M. (2010). Rebel capability and strategic violence against civilians. Journal of Peace Research, 47(5), 601–614.Find this resource:
(1.) A notable exception to this trend is the literature on state repression, which has generally categorized all physical integrity violations into a single measure of repressive behavior (e.g., Gibney, Cornett, Wood, Haschke, & Arnon, 2016). This trend has its own limitations, as we are not able to distinguish between the strategic logic of behaviors such as enforced disappearances, torture, or extrajudicial killings.
(2.) For the purposes of this discussion I am writing about the government as a unitary actor; however, important work on conflict behavior has stressed principal-agent problems (e.g., Gates, 2002; Conrad & Moore, 2010) as well as the diverse nature of actors involved in decision-making within the government.
(3.) While this article makes a substitution argument, it is not tested.