Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS (oxfordre.com/politics). (c) Oxford University Press USA, 2018. All Rights Reserved. Personal use only; commercial use is strictly prohibited. Please see applicable Privacy Policy and Legal Notice (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 18 December 2018

Bargaining Theory, Civil War Outcomes, and War Recurrence: Assessing the Results of Empirical Tests of the Theory

Summary and Keywords

Once ended, a significant number of civil wars recur. One influential empirical international relations theory on which scholars have drawn in an effort to provide an explanation for this phenomenon is the bargaining model of war. Devised initially for the study of interstate war, the theory posits that bargaining problems may prevent belligerents from reaching a deal that enables them to avoid a costly war. Bargaining problems also have been identified as contributing to the recurrence of armed intrastate conflict. Working within the framework of bargaining theory, a number of scholars have claimed that the most effective way to inhibit a return to civil war is to end the conflict via military victory as such an outcome is thought to help solve key bargaining problems. However, a growing number of empirical tests cast doubt on this proposition. An analysis of the results of these tests as well as new scholarship on civil war termination highlight some of the limitations inherent in employing a theory devised for the study of interstate war to analyze questions related to civil wars.

Keywords: civil war recurrence, bargaining model of war, civil war outcomes, bargaining problems, empirical tests, empirical international relations theory

Introduction

Once ended, a significant number of civil wars recur. One influential empirical international relations theory on which scholars have drawn in an effort to provide an explanation for this phenomenon is the bargaining model of war. Devised initially for the study of interstate war, the theory posits that bargaining problems may prevent belligerents from reaching a deal that enables them to avoid a costly war. Bargaining problems also have been identified as contributing to the recurrence of armed intrastate conflict. Working within the framework of bargaining theory, a number of scholars have claimed that the most effective way to inhibit a return to civil war is to end the conflict via military victory as such an outcome is thought to help solve key bargaining problems. As this article makes clear, a growing number of empirical tests cast doubt on this proposition. An analysis of the results of these tests as well as new scholarship on civil war termination highlight some of the limitations inherent in employing a theory devised for the study of interstate war to analyze questions related to civil wars.

Scholars’ use of bargaining theory to account for civil war termination and the recurrence of civil wars is outlined. The results of efforts empirically to test the central claim that one type of war outcome—military victory—is better at stabilizing the peace following civil war than is another—negotiated agreements—are examined. Efforts are made to assess the mixed results that these tests have yielded. Implications for our understanding of civil war recurrence stemming from emerging scholarship that relaxes some of the assumptions of bargaining theory are reviewed. Conclusions are then drawn regarding the merits and limitations of borrowing from a theory created to understand one kind of behavior—interstate war—to explain another—civil war.

The Bargaining Model of War, Civil War Termination, and Recurrence

According to the bargaining model of war, armed conflict is one of the policy tools available to rational actors who seek to resolve a dispute. Parties can be expected to resort to bargaining in an effort to resolve disagreements over the allocation of goods, resources, or policies (Reiter, 2003). Because fighting is costly, there is assumed to be some agreement that is preferable to war (Fearon, 1995; Powell, 2006). The failure by belligerents to arrive at such a deal without resorting to the use of armed conflict has been attributed to bargaining problems. These bargaining problems consist of asymmetric information regarding the distribution of capabilities or resolve, which parties may have an incentive to misrepresent; an inability on the part of actors credibly to commit to abide by the terms of an agreement; and the existence of stakes whose lack of divisibility makes it difficult to strike a mutually beneficial bargain.1

Scholars working within the bargaining model framework note that fighting should help parties overcome the information problem. If war occurs because actors are unable to agree on which party should make concessions and what the nature of those concessions should be, fighting should provide the actors with information that subsequently allows them to agree on the terms of a bargain. By revealing adversaries’ relative capabilities, fighting helps generate convergent beliefs on the adversaries’ parts regarding the consequences of continued combat. With uncertainty reduced as a result of fighting, warring opponents should be able to agree on a set of concessions or terms regarding a division of benefits or resources that all sides prefer to that of continuing the war (Reiter, 2003; Werner & Yuen, 2005). Having reached such a bargain, peace should persist as long as the groups remain committed to the new bargain, an outcome that should prevail as long as the groups in question have convergent expectations regarding the outcome of a future hypothetical war. However, if those expectations should change, war is predicted to become more likely (Wagner, 1993; Werner, 1999; Werner & Yuen, 2005).

Bargaining problems, particularly those related to information and commitment issues, have been identified as posing more severe obstacles to efforts to end civil wars than to the termination of interstate conflicts (Walter, 2009). In contrast to disputes between countries, cases in which governments are likely to know something about the size of an opposing state’s military and its previous performance in war, governments in a civil war context may have little concrete information about the size and fighting ability of rebel forces, nor the degree of support they have from internal and external actors. Non-state actors may also lack a sense of their fighting capabilities prior to engaging in battle and may well be reluctant to share that information lest government forces use it against them. Additionally, unless governments have previously been involved in a civil war with other rebel forces, considerable uncertainty is likely to exist regarding their willingness to fight or to settle a conflict (Walter, 2006).

Commitment problems are also deemed to constitute a serious obstacle to the striking of a deal between governments and non-state actors. Within the context of the state, a bargain is likely to require non-state actors to lay down their arms, thus exposing them to risk of attack by the government and leaving them with little means to enforce the terms of an agreement. Aware of the potential that such a situation has for their exploitation once a war has ended, non-state actors will be extremely reluctant to enter into an agreement. Governments may also be wary of agreeing to a deal if they believe that non-state actors have the potential to gain significant influence or power over time as the latter groups would then have an incentive to demand a new and more favorable deal, an outcome a relatively weaker government might not be able to impede.

Despite the obstacles posed by information and commitment problems, governments and rebel groups have agreed to bargains that end civil wars short of a military victory by one side. Such bargains, arrived at via negotiated settlements and negotiated truces, are considered to be inherently unstable by many adherents of the bargaining school and thus to have a heightened risk of recurrence. One source of this instability has been identified as the fact that such bargains allow non-state actors with the organizational capacity to reinitiate conflict to continue to exist (Wagner, 1993; Werner, 1999). The potential for the balance of power between the state and these groups to evolve over time also has been noted as being potentially destabilizing (Walter, 2009). Yet another source of instability is the failure of “unnatural” settlements reached via third-party pressure to reflect adversaries’ expectations regarding the distribution of benefits they could or should have received from the conflict (Werner & Yuen, 2005). These factors, scholars claim, undermine bargains reached via negotiation by leaving open the possibility that some group could secure better terms for itself through a return to war. Finally, the failure by third parties fully to live up to the security guarantees they offer warring groups in order to convince them to reach a negotiated end to a civil war and for at least some of the terms of terms of negotiated agreements to be implemented also have been identified as reasons to expect negotiated deals to break down and civil war to recur (Jarstad & Nilsson, 2008; Ottmann & Vüllers, 2015).

In light of the perceived instability of the bargains associated with negotiated settlements and negotiated truces, several scholars have posited that the best way to ensure that a civil war does not recur is to end it via a military victory. Military victories are assumed to resolve information problems by clearly communicating the distribution of capabilities and resolve. This should stabilize expectations, and thus the peace, by providing the losing side(s) of the conflict with unambiguous information regarding its (their) likelihood of winning a future hypothetical war. Additionally, military victories are assumed to result in the destruction of the organizational structures of all the parties to the conflict but the one that emerges as the winner. This means that military victories effectively “solve otherwise difficult commitment problems . . . [because the] victor simply implements its favored policies” (Walter, 2009, p. 257).

Tests of the Theory: The Influence of War Outcomes on Civil War Recurrence

Claims regarding the superiority of military outcomes of civil wars as a means of minimizing the likelihood of conflict recurrence began to appear in the scholarly literature on civil wars nearly a quarter century ago (Licklider, 1995; Luttwak, 1999; Wagner, 1993).2 In one of the earliest empirical tests of this proposition, Licklider (1995) found that civil wars that ended in military victories were less likely to see the parties return to armed conflict, an outcome he attributed to the winning side’s ability to subjugate its adversary. Subsequent studies by DeRouen and Bercovitch (2008) and Rustad and Binningsbø (2012) have also provided support for the stabilizing effect of military victories.

Arguing that the identity of the military victor in a civil war has an impact on the stability of the peace, a number of scholars have distinguished among conflicts that end in a government victory, a rebel victory, and a negotiated agreement. Tests of the peace-extending effects of victory by rebels vs. victory by government forces have yielded mixed results. Finding that civil wars that end via negotiated agreement were twice as likely to recur as those terminating in military victory, Duffy Toft also finds that, among military victories, conflicts won by rebel groups are the least likely to recur (2010). Positing that a rebel victory is the best means by which to dismantle the condition of dual sovereignty characteristic of civil wars states, tests by Quinn, Mason, and Gurses (2007) and Mason, Gurses, Brandt, and Quinn (2011) also provide support for the stabilizing effects of rebel victory. On the other hand, Kreutz (2010) and Zeigler (2016) find evidence indicating that government victories are more effective at lowering the likelihood that civil war will recur.

The results of yet other tests of the relationship between civil war outcomes and conflict recurrence have failed to provide support for the theoretically generated expectation that military outcomes have a stabilizing effect on the peace. Pearson, Lounsebery, Walker, and Mann (2006) find no evidence that military victory yields a more durable peace. A study by Walter (2004) indicates that neither military victory nor negotiated agreements have a significant effect on civil war recurrence. Fortna (2008) and Hartzell (2009), on the other hand, find that both types of war outcomes can produce a durable peace. Doyle and Sambanis (2006) conclude that negotiated agreements lower the likelihood of war recurrence while military victories have no discernible impact on the peace.

Assessing the Results of Empirical Tests of the Theory

The inconsistent nature of the results of tests of the effect that civil war outcomes have on conflict recurrence raises several questions. One is how to account for the mixed nature of the evidence regarding the relationship between outcomes and recurrence. Is it an artifact of the tests themselves? Or are the results inconsistent because theoretical testing of the argument has been fragmented and not always cumulative? Second, in light of their mixed nature, how do we evaluate the findings? Do they provide support for or disconfirm the expectations regarding civil war recurrence that have been derived from the bargaining model of war?

Part of the explanation for the mixed nature of the evidence regarding the relationship between civil war outcomes and conflict recurrence is likely to lie in differences among the tests that scholars have employed to test the proposition that civil wars ending in military victories are less likely to recur. One way in which these studies differ is in the varying definitions of intrastate conflict they employ, with some tests focusing on conflicts that produce at least 1,000 deaths during the course of the conflict, some including conflicts with lower death thresholds, and some employing annual casualty counts. Differences in the rules for coding what counts as an intrastate conflict have yielded differences in the numbers of conflicts that are the subject of empirical tests as well as, arguably, in the types of conflicts that are being analyzed.

In conjunction with the point made above, differences in the methodologies employed by various scholars are also likely to be factors contributing to the mixed nature of the results produced by tests of the effects civil war outcomes have on war recurrence. Several models, for, example, code civil war recurrence dichotomously, employing logistical regression to analyze the effect war outcomes have on the dependent variable. Most of those tests have focused on a five-year period of time following the end of the fighting to code whether conflict recurs or not, although other studies have employed time frames ranging from 1 to 2 to 10 years in length, with obvious implications for the number of cases available for analysis in each instance. More recent studies have employed event history analysis, thus shifting the dependent variable from a focus on whether or not conflict recurs to how long peace endures following the end of a conflict.

Theoretical testing of the core argument regarding the relationship between civil war outcomes and peace duration also has been fragmented, a factor that may have contributed to the inconsistent nature of the results as well as to a lack of cumulation. For example, a number of scholars have sought to disaggregate negotiated agreements, moving beyond coding whether or not a civil war ended via negotiation to examining the contents of those agreements. The contents of agreements have been defined on the basis of whether or not they call for power sharing (e.g., Hartzell & Hoddie, 2007), cost-increasing and fear-reducing provisions (Mattes & Savun, 2009), judicial reforms and amnesty provisions (Bell & O’Rourke, 2009), and gender provisions and more (Badran, 2014; Joshi & Darby, 2013). Although most of the aforementioned studies have sought to disaggregate only the agreements of wars that ended via negotiation, others have attempted to determine whether power-sharing provisions, for example, characterize the settlements of wars ending via military victory, negotiated agreements, or truces (Findley, 2012; Hartzell, 2009; Mukherjee, 2006). While these efforts have served to highlight the fact that negotiated agreements differ in significant ways, some of which may have implications for the likelihood that conflict will recur, they also have made it more difficult to sort out how the results of studies relate to the core argument as well as to one another.

As this overview of empirical tests focused on civil war recurrence makes apparent, there is a lack of clear support for the theoretical proposition, derived from the bargaining model of war, that military victory is the best means for overcoming bargaining problems that lead to the renewed outbreak of civil war. Although this may suggest that there are limitations inherent in the use of bargaining theory to understand civil war recurrence, the results should not be interpreted as definitive evidence that the model is of little utility for studying this issue. That is because, arguably, tests of the theoretical proposition have been flawed, or at least incomplete, in nature. There are three reasons to make this claim. First, bargaining theory emphasizes that wars end in bargains, settlements that “specify who gets what and when” (Werner & Yuen, 2005, p. 262). However, rather than looking at the terms of those bargains, scholars have opted to focus on the outcomes of civil wars. On the face of it, such a focus makes sense since outcomes serve to “define the range of available settlements” (Quackenbush & Venteicher, 2008, p. 729). But settlements and outcomes are not identical concepts. Settlements themselves have the potential to be structured in such a manner as to help overcome bargaining problems. Studies that concentrate on outcomes to the exclusion of settlements thus fail to assess the independent effect that settlements may have on civil war recurrence.

Second, a number of tests have used war outcomes as a proxy for the causal mechanisms that have been identified as reducing or solving bargaining problems, rather than directly testing the effects of the mechanisms themselves. For example, scholars long have assumed that military victories result in the destruction of the losing factions’ organizational structures, thus precluding any commitment problem, while negotiated agreements provide protections for defeated groups, thus raising the possibility that they might return to civil war. However, a test of the hypothesized mechanism found that a significant number of wars ending in military victory saw more than one faction preserve its organizational identity while a number of wars ending in a negotiated agreement saw the dismantling of the organizational structures of all but the winning faction. Once the fates of the rival factions were taken into account, both military victory and negotiated settlements were found to lower the likelihood that civil war would recur (Hartzell, 2009).

Third, tests that have focused on war outcomes in isolation from the characteristics of conflicts have failed to identify ways in which “[c]ontextual conditions surrounding the termination of wars shape postconflict possibilities” (Zeigler, 2016, p. 35). Interacting war outcomes with conflict attributes could help to reveal the extent to which different war outcomes are associated with the mitigation of the various forms of commitment problems. In keeping with such an approach, tests should also endeavor to examine the joint influence that settlements and outcomes may have on the stability of the peace.

Emerging Scholarship: Potential Challenges to the Bargaining Model of War?

More thorough testing of the central claims and associated mechanisms derived from the bargaining model of war, as well as of environmental factors associated with civil war termination, is necessary in order to judge the utility of bargaining theory for helping to explain the conditions under which civil wars are likely to recur. At this point in time it is impossible to know whether or not tests of the nature described above will provide support for one of the central propositions derived from bargaining theory, the claim that civil war outcomes have an impact on civil war recurrence. In the interim, new scholarship on civil war termination is generating a new set of potential challenges to theoretically generated expectations derived from bargaining theory.

One of the central sources of these challenges is studies that relax key assumptions associated with bargaining theory. Central among these are assumptions that a fixed set of actors participates in civil war and that actors are unitary. Speaking to the first of these, Driscoll (2012) observes “. . . civil wars since 1945 have rarely featured professional armies with coherent and unified lines of command and control” . . . [thus making] “. . . [a]nalogies to two-player firm or interstate bargaining problems . . . strained” (p. 122). Driscoll, like Rudloff and Findley (2016) and Zeigler (2016), also focuses on the splintering of rebel groups during civil war, a process that all three studies suggest generates significant commitment problems that have implications for the duration of the peace.

Building on these two insights, Driscoll describes civil war termination as a “coalition formation game” in which peace is achieved via “a constantly renewable process of contracting and bargaining between violence entrepreneurs” (2012, p. 122) with winners and losers in the state consolidation process not necessarily constituting the victors or losers of the civil war. While drawing on insights from the bargaining model of war, the picture of peace that Driscoll paints differs from that of the traditional two-player bargaining model in key ways. First, in Driscoll’s work, peace is a product of multiple private bargains struck between the leader of a state and individual warlords. Second, peace is possible, and enforceable, because rebel groups remain armed, thereby mitigating the commitment problem. Finally, power sharing allows a number of warlords to “reinvent and redefine themselves as the state” (p. 122), an outcome that further challenges bargaining theory assumptions regarding the invariable identities of the parties (i.e., states and rebel groups) to a conflict.

Further challenges to the bargaining model’s conceptualization of civil war as involving two opposed sets of unitary actors (the government and a challenger) stem from the emerging literature on side switching in civil wars. As a growing number of scholars have noted, many civil wars involve not only multiple rebel groups but pro-government militias, non-state armed actors aligned with the government (see, for example, Carey, Mitchell, & Lowe, 2013; Jentzsch, Kalyvas, & Schubiger, 2015). These sets of non-state actors do not always maintain a fixed relationship (i.e., supporter or adversary) with the government. As Otto (2017) makes clear, side switching, or an intentional change of alignment with the central government during the course of an armed conflict, takes place in a significant proportion of multiparty civil wars. Although scholars have not yet explored the implications that side switching may have for the bargaining model of war, two in particular come to mind. First, it is difficult to know what kind of information battles convey regarding the strength of the various actors involved in a civil war in which side switching takes place. Is an actor that won a number of battles as a result of a group aligning itself with it during the course of a conflict likely to be perceived clearly by others as victorious? Could evaluations of that nature be subject to revision following defection by a group? And what types of effects is all of this likely to have on the terms of the settlements that end the war? Second, it may well be the case that commitment problems are exacerbated in the case of conflicts in which some groups have a history of switching sides. Not only are parties less likely to trust actors with a history of side switching but the possibility that a former side-switching party to a settlement might opt at some point to defect from one group and align itself with another could serve to undermine the peace by prompting other actors to take preemptive moves.

Staniland (2012) seeks to address a weakness of the bargaining model of war, the failure of studies that focus on civil war outcomes to encapsulate political relationships between state and non-state actors, by conceiving of civil war as a process of competitive state building. Relationships between states and insurgents, he observes, “should be determined not solely by military violence but also, like the evolution of states, by political relationships and bargains” (p. 244). Based on this insight, Staniland relaxes the bargaining model’s assumption that states and insurgents seek to monopolize power and resources. Rather, he assumes that they seek to optimize authority, a goal that can lead them to establish wartime political orders, sustained relationships ranging from “collusion and shared sovereignty to spheres of influence and tacit coexistence to clashing monopolies and guerrilla disorder” (p. 244). These orders matter, Staniland posits, because they can produce types of stability: outcomes that cannot be coded as cases of victory, defeat, or formal peace agreements.

Conclusion

Scholars of civil war have debated the “best” way to end a civil war in order to ensure that fighting does not recur. Drawing on the bargaining model of war, a number of scholars have argued that intrastate conflicts that terminate in a military victory are the most likely to be followed by prolonged periods of peace because military victories can help to resolve bargaining problems that lead to war recurrence. Empirical tests of the proposition have failed to provide clear-cut support for this claim. The lack of support may be a product of the failure of the tests fully to engage with the theory undergirding the bargaining model of war. This suggests that it would be premature, at least based on the results of the tests, to conclude that the bargaining model of war is of limited value for analyzing questions relating to civil war recurrence.

There are other reasons, however, to question the utility of using a model derived for the study of interstate war to analyze civil war. One is the different nature of the belligerents involved in each of the types of conflicts. Competitive rebel coalitions, warlords, and other actors involved in civil wars may pose information and commitment problems different from those at the heart of the bargaining model of war and thus not solvable via combat. If civil war adversaries seek not to achieve some optimal allocation of goods but to maximize authority (Staniland, 2012), or if “incumbent and insurgent military commanders are potentially interchangeable candidates for postwar government spoils” (Driscoll, 2012, p. 119), it is difficult to see how military victories address these issues in a manner that helps to stabilize the peace.

Relatedly, it is also not clear whether fighting has the same ability to reveal information in civil war contexts as it does in interstate wars. If a civil war involves multiple (and potentially shifting) factions, for example, fighting may not generate clear-cut signals regarding actors’ capabilities and resolve. It is also not clear what kind of information military victories convey if they do not consistently result in the dismantling of the organizational structures of defeated groups. Fighting that takes the form of hit and run tactics, or symmetric nonconventional conflicts that match adversaries with low levels of military technology, types of civil war combat that became more common following the end of the Cold War (Balcells & Kalyvas, 2014), may also fail to yield information that helps to clarify the terms of a bargain.

Finally, more consideration should be given to the inherent differences between interstate war-ending settlements and intrastate war-terminating settlements. The bargaining model of war defines settlements in both instances as a set of terms regarding a division of benefits or resources that all sides prefer to that of continuing the war. Because the parties to a civil war must continue to exist within the confines of a state, civil war settlements almost invariably contain terms relating to the means by which such interactions are to take place. The possibility that these terms, which often take the form of institutions, have the potential to alter rivals’ preferences in a manner that makes for a stable peace, is one factor that distinguishes civil war settlements from interstate war settlements. Indeed, the international community appears to have seized on this notion in recent years, helping conflict actors design settlements that provide adversaries with a stake in the peace. Although such settlements have not always proved stable, there is some evidence that learning is taking place as a growing number of measures with the potential to help stabilize the peace are identified (Badran, 2014).

Civil war outcomes have been accorded a central explanatory role in studies of civil war recurrence for the past 25 years. The time is ripe for thinking critically about the role outcomes, and the bargaining model of war more broadly, should play in theories of civil war recurrence. The most significant contribution of the bargaining model of war where the question of civil war recurrence is concerned lies in the framework it provides scholars for thinking systematically about the types of problems—information asymmetries and commitment problems—that may contribute to civil war recurrence. Noting that these types of problems may be more difficult to solve in the case of civil wars given the severe nature of the security dilemma and the often confusing information conveyed by combat, a number of scholars have concluded that ending a war via military victory is the best means that exists for securing the peace. Recognizing that civil wars differ from interstate wars along those dimensions, and adapting the analytical power of the bargaining model of war accordingly, has helped to advance knowledge regarding the challenges inherent in stabilizing the peace after civil war. But as new scholarship on civil war makes clear, civil wars differ from interstate wars in other key ways that scholars interested in questions of war recurrence would do well to take into account. In some instances, doing so may involve modifications of the bargaining model of war. In others, it may involve developing new models to understand civil war recurrence.

References

Badran, R. (2014). Intrastate peace agreements and the durability of the peace. Conflict Management and Peace Science, 31(2), 193–217.Find this resource:

    Balcells, L., & Kalyvas, S. N. (2014). Does warfare matter? Severity, duration, and outcomes of civil wars. Journal of Conflict Resolution, 58(8), 1390–1418.Find this resource:

      Bell, C., & O’Rourke, C. (2009). Counting peace agreements: The transitional justice peace agreements database. Transitional Justice Institute Research Paper No. 09-12. Transitional Justice Institute, Ulster University.Find this resource:

        Carey, S., Mitchell, N., & Lowe, W. (2013). States, the security sector, and the monopoly of violence: A new database on pro-government militias. Journal of Peace Research, 50(2), 249258.Find this resource:

          Cunningham, D. E. (2006). Veto players and civil war duration. American Journal of Political Science, 50(4), 875–892.Find this resource:

            DeRouen, K., Jr., & Bercovitch, J. (2008). Enduring internal rivalries: A new framework for the study of civil war. Journal of Peace Research, 45(1), 43–62.Find this resource:

              Doyle, M. W., & Sambanis, N. (2006). Making war and building peace: United Nations peace operations. Princeton, NJ: Princeton University Press.Find this resource:

                Driscoll, J. (2012). Commitment problems or bidding wars? Rebel fragmentation as peace building. Journal of Conflict Resolution, 56(1), 118–149.Find this resource:

                  Fearon, J. D. (1995). Rationalist explanations for war. International Organization, 49(3), 379–414.Find this resource:

                    Findley, M. G. (2012). Bargaining and the interdependent stages of civil war resolution. Journal of Conflict Resolution, 57(5), 905–932.Find this resource:

                      Fortna, V. P. (2008). Does peacekeeping work? Shaping belligerents’ choices after civil war. Princeton, NJ: Princeton University Press.Find this resource:

                        Hartzell, C. A. (1999). Explaining the stability of negotiated settlements to intrastate wars. Journal of Conflict Resolution, 43(1), 3–22.Find this resource:

                          Hartzell, C. A. (2009). Settling civil wars: Armed opponents’ fates and the duration of the peace. Conflict Management and Peace Science, 26(4), 347–365.Find this resource:

                            Hartzell, C. A. (2016). Civil war termination. Oxford Research Encyclopedia of Politics.Find this resource:

                              Hartzell, C. A., & Hoddie, M. (2007). Crafting peace: Power-sharing institutions and the negotiated settlement of civil wars. University Park: Pennsylvania State University Press.Find this resource:

                                Jarstad, A. K., & Nilsson, D. (2008). From words to deeds: The implementation of power-sharing pacts in peace accords. Conflict Management and Peace Science, 25(3), 206–223.Find this resource:

                                  Jentzsch, C., Kalyvas, S. N., & Schubiger, L. I. (2015). Militias in civil wars. Journal of Conflict Resolution, 59(5), 755–769.Find this resource:

                                    Joshi, M., & Darby, J. (2013). Introducing the peace accords matrix (PAN): A database of comprehensive peace agreements and their implementation, 1989–2007. Peacebuilding, 1(2), 256–274.Find this resource:

                                      Kreutz, J. (2010). How and when armed conflicts end: Introducing the UCDP conflict termination dataset. Journal of Peace Research, 47(2), 243–250.Find this resource:

                                        Licklider, R. (1995). The consequences of negotiated settlements in civil wars, 1945–1993. American Political Science Review, 89(3), 681–690.Find this resource:

                                          Luttwak, E. N. (1999). Give war a chance. Foreign Affairs, 78(4), 36–44.Find this resource:

                                            Mason, T. D., Gurses, M., Brandt, P. T., & Quinn, J. M. (2011). When civil wars recur: Conditions for durable peace after civil wars. International Studies Perspectives, 12(2), 171–189.Find this resource:

                                              Mattes, M., & Savun, B. (2009). Fostering peace after civil war: Commitment problems and agreement design. International Studies Quarterly, 53(3), 737–759.Find this resource:

                                                Mukherjee, B. (2006). Why political power-sharing agreements lead to enduring peaceful resolution of some civil wars, but not others? International Studies Quarterly, 50(2), 479–504.Find this resource:

                                                  Ottmann, M., & Vüllers, J. (2015). The power-sharing event dataset (PSED): A new dataset on the processes and practices of power-sharing in post-conflict countries. Conflict Management and Peace Science, 32(3), 327–350.Find this resource:

                                                    Otto, S. (2017). The grass is always greener? Armed group side switching in civil wars. Journal of Conflict Resolution. Advance online publication.Find this resource:

                                                      Pearson, F. S., Lounsebery, M. O., Walker, S., & Mann, S. (2006). Rethinking models of civil war settlement. International Interactions, 32(2), 109–128.Find this resource:

                                                        Powell, R. (2006). War as a commitment problem. International Organization, 60(1), 169–203.Find this resource:

                                                          Quackenbush, S. L., & Venteicher, J. F. (2008). Settlements, outcomes, and the recurrence of conflict. Journal of Peace Research, 45(6), 723–742.Find this resource:

                                                            Quinn, J. M., Mason, T. D., & Gurses, M. (2007). Sustaining the peace: Determinants of civil war recurrence. International Interactions, 33(2), 167–193.Find this resource:

                                                              Reiter, D. (2003). Exploring the bargaining model of war. Perspectives on Politics, 1(1), 27–43.Find this resource:

                                                                Rudloff, P., & Findley, M. G. (2016). The downstream effects of combatant fragmentation on civil war recurrence. Journal of Peace Research, 53(1), 19–32.Find this resource:

                                                                  Rustad, S. A., & Binningsbø, H. M. (2012). A price worth fighting for? Natural resources and conflict recurrence. Journal of Peace Research, 49(4), 531–546.Find this resource:

                                                                    Staniland, P. (2012). States, insurgents, and wartime political orders. Perspectives on Politics, 10(2), 243–264.Find this resource:

                                                                      Toft, M. D. (2010). Securing the peace: The durable settlement of civil wars. Princeton, NJ: Princeton University Press.Find this resource:

                                                                        Wagner, R. H. (1993). The causes of peace. In R. Licklider (Ed.), Stopping the killing: How civil wars end (pp. 235–268). New York: New York University Press.Find this resource:

                                                                          Walter, B. (1997). The critical barrier to civil war settlement. International Organization, 51(3), 335–364.Find this resource:

                                                                            Walter, B. (2004). Does conflict beget conflict? Explaining recurring civil war. Journal of Peace Research, 41(3), 371–388.Find this resource:

                                                                              Walter, B. (2006). Building reputation: Why governments fight some separatists but not others. American Journal of Political Science, 2(50), 313–330.Find this resource:

                                                                                Walter, B. (2009). Bargaining failures and civil war. Annual Review of Political Science, 12, 243–261.Find this resource:

                                                                                  Werner, S. (1999). The precarious nature of peace: Resolving the issues, enforcing the settlement, and renegotiating the terms. American Journal of Political Science, 43(3), 912–934.Find this resource:

                                                                                    Werner, S., & Yuen, A. (2005). Making and keeping peace. International Organization, 59(2), 262–293.Find this resource:

                                                                                      Zeigler, S. M. (2016). Competitive alliances and civil war recurrence. International Studies Quarterly, 60(1), 24–37.Find this resource:

                                                                                        Notes:

                                                                                        (1.) I focus on information and commitment problems in what follows, as it is not clear that issue indivisibility poses a serious problem for civil war recurrence.

                                                                                        (2.) A number of the empirical tests discussed in this section were previously reviewed in Hartzell (2016).