Incentivizing peace how international organizations can help prevent civil wars in member countries

LIST OF FIGURES3.1 Count of highly structured IGOs in the international system over time 3.2 Comparison of the membership size of all highly structured IGOs 3.3 Membership size of highly

Incentivizing Peace

Oxford University Press is a department of the University of Oxford It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide Oxford is a registered trade mark of Oxford University Press in the UK and in

certain other countries.

Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.

© Oxford University Press 2018

All rights reserved No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by license, or under terms agreed with the appropriate reproduction rights organization Inquiries concerning reproduction outside the scope of the above should be

sent to the Rights Department, Oxford University Press, at the address above.

You must not circulate this work in any other form and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data Names: Tir, Jaroslav, 1972- author | Karreth, Johannes, author.

Title: Incentivizing peace : how international organizations can help prevent civil wars

in member countries / Jaroslav Tir and Johannes Karreth.

Description: New York, NY : Oxford University Press, [2018] | Includes bibliographical references.

Identifiers: LCCN 2017026625 (print) | LCCN 2017052236 (ebook) | ISBN 9780190699543 (epub) | ISBN 9780190699536 (updf) | ISBN 9780190699512 (hardcover) | ISBN 9780190699529 (pbk.)

Subjects: LCSH: International agencies–Membership | Peace-building–International cooperation | Conflict management–International cooperation | Civil war–Prevention.

Classification: LCC JZ4850 (ebook) | LCC JZ4850.T57 2018 (print) | DDC 303.6/4–dc23

LC record available at https://lccn.loc.gov/2017026625

1.1 THE PROBLEMS OF CIVIL WAR AND CIVIL WAR MANAGEMENT

1.2 A NOVEL PERSPECTIVE ON CIVIL WAR MANAGEMENT

1.3 CONTRIBUTIONS AND IMPLICATIONS

1.4 OUTLINE OF THE BOOK

2 Managing Civil Wars from the Perspective of Their Development

2.1 INTRODUCTION

2.2 CAUSES, CONSEQUENCES, AND MANAGEMENT OF CIVIL WARS

2.3 CIVIL WAR DEVELOPMENT AND ESCALATION

2.4 WHAT DOES IT TAKE TO PREVENT LOW-LEVEL ARMED CONFLICT FROM ESCALATING TO CIVIL WAR? 2.5 THE (IN)ADEQUACY OF COMMON THIRD-PARTY CONFLICT MANAGING POLICIES IN PREVENTING

CONFLICT ESCALATION

2.6 OUR THEORETICAL FRAMEWORK FOR ESCALATION PREVENTION

3 The Interplay Between Civil War Development and Highly Structured Intergovernmental

Organizations

3.1 THE DEFINITION, EVOLUTION, AND DISTRIBUTION OF HIGHLY STRUCTURED IGOS

3.2 THE IMPORTANCE OF IGOS’ INDEPENDENCE FROM AND LEVERAGE OVER MEMBER STATES

3.3 THE ROLE OF HIGHLY STRUCTURED IGOS IN PREVENTING CIVIL WAR

3.4 HOW HIGHLY STRUCTURED IGOS EXERCISE INFLUENCE

3.5 THE PRIMARY HYPOTHESIS

4 The Empirical Record of Highly Structured Intergovernmental Organizations and Armed ConflictEscalation

4.1 A SYSTEMATIC EXAMINATION OF CONFLICT ESCALATION

4.2 THE UNIVERSE OF CASES

4.3 ARMED CONFLICT ESCALATION

4.4 MEASURING THE INFLUENCE OF HIGHLY STRUCTURED IGOS

4.5 OTHER POTENTIAL DETERMINANTS OF ESCALATION

4.6 HIGHLY STRUCTURED IGOS REDUCE THE RISK OF ARMED CONFLICT ESCALATION

4.7 CONCLUSION

5 The Logic of Institutional Influence: Conceptual and Methodological Implications

5.1 COSTS AND BENEFITS

5.2 PEACEFUL SETTLEMENTS

5.3 MEDIATION AND INTERVENTION

5.4 POTENTIAL MEMBERSHIP SCREENING BY HIGHLY STRUCTURED IGOS

5.5 OTHER POTENTIAL INFLUENCES ON CONFLICT ESCALATION

5.6 THE ADDED VALUE OF HIGHLY STRUCTURED IGOS INEXPLAINING ESCALATION

7.1 BRIEF SUMMARY OF FINDINGS

7.2 LESSON 1: THE INTERNATIONAL ENVIRONMENT CAN CONTRIBUTE TO CIVILWAR PREVENTION 7.3 LESSON 2: ECONOMIC INFLUENCE IS KEY TO CIVILWAR PREVENTION

7.4 LESSON 3: COORDINATION INCREASES THE IMPACT

7.5 LESSON 4: FURTHER INVESTIGATE THE IMPACT OF HIGHLY STRUCTURED IGOS ON REBELS 7.6 LESSON 5: PEACE PAYS OFF (FOR REBELS, TOO)

LIST OF FIGURES

3.1 Count of highly structured IGOs in the international system over time

3.2 Comparison of the membership size of all highly structured IGOs

3.3 Membership size of highly structured IGOs over time

3.4 States’ participation in highly structured IGOs over time

3.5 The geography of highly structured IGOs

4.1 Map of all low-level armed conflicts, 1946–2000

4.2 Map of escalated conflicts, 1946–2000

4.3 Highly structured IGOs and estimated escalation risk

4.4 Changes in escalation risk

4.5 Political institutions and escalation risk

4.6 Changes in the risks of conflict onset and escalation

5.1 HSIGOs commanding substantial resources and changes in escalation risk

5.2 HSIGOs commanding substantial resources and conflict escalation

5.3 Changes in the probability of low-level armed conflict settlements

5.4 Highly structured IGOs and low-level armed conflict settlement

5.5 Distribution of mediation and intervention efforts

5.6 The role of highly structured IGOs intervention, and mediation

5.7 Mediation, intervention, highly structured IGOs, and conflict escalation

5.8 Low-level armed conflict escalation patterns over time

5.9 Highly structured IGOs and escalation risk, using instrumental variable estimates5.10 HSIGOs, conflict escalation, and natural resources

5.11 HSIGOs, rebel strength, and escalation risk

5.12 HSIGOs, space, time, and changes in escalation risk

5.13 HSIGOs, conflict diffusion, and changes in escalation risk

5.14 Variable inclusion in BMA

5.15 Posterior probabilities of coefficients in BMA

LIST OF TABLES

2.1 An Assessment of Common Third-Party Conflict Management Policies

3.1 List of Highly Structured IGOs

3.2 An Assessment of Third Parties’ Roles in Pre–Civil War Bargaining

4.1 Low-Level Armed Conflicts That Escalated to Civil Wars, 1946–2000

4.2 Descriptive Statistics for All Variables in Model 2 in Table 4.3

4.3 Probit Estimates of the Escalation of Low-Level Armed Conflict to Civil War

4.4 Heckman Probit Estimates of Escalation, Accounting for Selection of Low-Level Armed

Conflict

5.1 Descriptive Statistics for All Analyses

5.2 The Influence of Highly Structured IGOs Commanding Substantial Resources

5.3 Highly Structured IGOs and Conflict Settlement

5.4 Mediation, Intervention, Highly Structured IGOs, and Escalation

5.5 Instrumental Variable Estimates

5.6 Natural Resources and Escalation

5.7 Natural Resources and a Possible Conditional Impact of Highly Structured IGOs

5.8 Rebel Strength and Escalation

5.9 Spatial and Temporal Trends of Escalation

5.10 Conflicts in Proximity and Escalation

5.11 Territorial Conflicts and Escalation

6.1 Summary of Key Expectations and Anticipated Evidence

6.2 Summary of Evidence Pertaining to the Three Cases

6.3 Expectations and Observed Evidence

This book has grown out of a long-term collaborative project between the authors Along the way,several colleagues and friends have generously provided feedback and offered helpful thoughts onour work For their comments on the manuscript along its different stages, we thank Kyle Beardsley,Ken Bickers, Margit Bussmann, Courtenay Conrad, Paul Diehl, Bryan Early, Page Fortna, MichaelGreig, Håvard Hegre, Ann Karreth, Carmela Lutmar, Helen Milner, Sara Mitchell, Glenn Palmer,Burcu Savun, Carolyn Tir, and Geoffrey Wallace John Vasquez, Patrick James, Paul Diehl, andMatthew Ingram provided advice and guidance in developing and placing the book manuscript

At Oxford University Press, we are grateful to David McBride for his editorial guidance Thefeedback from two anonymous reviewers greatly improved the manuscript We also thank EmilyMackenzie, Ed Robinson, Claire Sibley, and Kathleen Weaver for their assistance during theproduction of this book and Susan McClung for copy-editing our manuscript

In facilitating this research, we have benefited from institutional support from the Departments ofPolitical Science at the University of Colorado Boulder and the University at Albany, the Department

of Politics and International Relations at Ursinus College, and the Kroc Institute for InternationalPeace Studies at the University of Notre Dame For research assistance, we thank Steven Beard, CoriCichowicz, Timothy Passmore, Elisa Elvove, and Charmaine Willis We also owe a debt of gratitude

to the scholars and researchers who collected and made publicly available the data we use throughoutthis book

The authors contributed equally and interactively to this book This book is one of several outputs

of a joint project, and the order of their names follows a principle of rotation

Incentivizing Peace

Introduction

1.1 The Problems of Civil War and Civil War Management

Civil wars are one of the most pressing problems facing the world today Beyond causing millions ofcasualties and immeasurable human suffering in the post–World War II era,1 such wars havedevastating social and economic consequences Public health institutions and outcomes in countriesexperiencing civil war systematically suffer.2 The mass migration of people fleeing violence anddestruction can destabilize neighboring countries and regions, creating new challenges.3 Thedestruction of physical and human capital during civil wars creates serious and lasting damage toeconomic prosperity and international trade.4 Experiences of violence lead to substantial increases inpolitical and social intolerance of ethnic, religious, and other out-groups.5

These and other consequences of civil wars are in part responsible for setting the stage for futurecivil wars Nearly 50 percent of civil wars return to violence within five years of the initial cessation

of hostilities.6 The danger of recurrence is so great that researchers have characterized manydomestic conflicts as enduring internal rivalries.7

Given the far-ranging and well-known consequences of civil wars, academics and policymakershave devoted much attention to finding solutions for managing these conflicts and their consequences.This attention has led to some positive outcomes Among a variety of approaches to civil warmanagement by third parties, the most common ones include mediation, peacekeeping, and militarizedinterventions These third-party responses have frequently helped They have contributed to thetransition of civil war–plagued countries toward normalcy and to the prevention of civil warrecurrence Some of the oft-heralded success stories of third-party conflict management includeCambodia, Sierra Leone, Angola, the Eastern Slavonia region of Croatia, Kosovo, and the DaytonAgreement in Bosnia More broadly, and focusing more on international wars, Joshua Goldsteinargues that the world has become more peaceful in recent times and gives at least some credit for thisdevelopment to concerted international conflict management efforts.8

Yet, despite some successes, the sad fact remains that civil wars continue to occur, and recur, tothis day with notable regularity.9 Indeed, critics note that the successful track record of conflictmanagement efforts is marred by a notable undersupply of third-party involvement or by outrightfailures Some high-profile cases where third parties failed to manage and halt conflicts in the last 25years include pre-Dayton Bosnia, Rwanda, Somalia, Sudan, and Syria In these and other cases,conflict management failed either because third parties took far too long to engage in robust, decisiveconflict management efforts or because they did not supply any meaningful conflict management at all

Multilateral peacekeeping missions in particular are deployed to only about one-third of civilwars.10 And in cases where international peacekeepers were involved, their engagement often leads

to speculation about whether the presence of peacekeepers actually helped local populations, andwhether they made substantial contributions to stabilization and peace.11 This mixed success record

of peacekeeping missions has also created a robust debate on how and why third parties choose when

to get involved in conflict management and when to stay out One criticism maintains thatpeacekeeping missions’ main agenda relates more to pursuing goals relevant to their sendingorganization or states These include accusations of selecting civil wars that are easier to resolve inorder to make organizations such as the United Nations look successful and helpful.12 Anothercritique notes that peacekeepers are often sent to help allied or economically important civil warstates, while ignoring others.13 Therefore, even if peacekeeping were a panacea to political violence,such selective deployment of peacekeeping efforts is one of the reasons why civil wars persist today

Questions of bias favoring the interests of sending states as opposed to civil-war states are evenmore prominent in the case of militarized interventions In these cases, intervening states generallyaim to protect or advance their economic or strategic interests and to make sure that their favored sidewins the war.14 And an unfortunate consequence of military interventions is that they can intensifypolitical violence and prolong armed conflicts,15 further amplifying the negative long-termconsequences of civil wars noted previously

International mediators are closest to being truly neutral actors who have the interests of the war state and its population at the forefront of their agenda Yet while mediation attempts receivemuch press for helping negotiate short-term ceasefires, it is not all that clear that they are particularlyeffective in actually ending civil wars.16 Additionally, mediators do not appear to go to all theconflict zones or do so particularly quickly For example, in the context of sub-Saharan Africa,Michael Greig reports that the chances that a conflict is mediated in its first three years are less than 1

civil-in 100.17

In short, despite many conflict management successes, scholars’ and practitioners’ understanding ofhow to manage civil wars can still be substantially improved With a better understanding of how touse existing international structures effectively to mitigate civil wars, countries facing high risks ofpolitical violence could improve their odds for domestic peace and stability Addressing this issueand highlighting a potential solution are the subjects of this book We preview this approach and ourargument next

1.2 A Novel Perspective on Civil War Management

As the previous discussion suggests, the international community has had only limited success inmanaging civil wars For that reason, we argue for the need to fundamentally rethink how third partiescan approach civil wars In this book, we propose that international, third-party conflict management

efforts will be particularly effective when they focus on civil war prevention, rather than on dealing

with already ongoing civil wars Civil wars do not occur overnight, but take time to develop.Acknowledging this fact allows us to investigate whether and how emerging conflicts can be stoppedbefore they escalate to full-scale civil wars Unfortunately, scholarly attention has thus far mostlyfocused on how typical third-party policy responses (peacekeeping, mediation, and intervention)affect later phases of a conflict Existing studies often ask whether mediation can help reach

ceasefires, or whether peacekeepers prolong the peace after civil war These outcomes are all

relevant once hostilities have already escalated to civil war The timing of the respective effortsillustrates this issue Mediators, intervention forces, and peacekeepers are typically deployed onlyafter a conflict has already become a full-blown civil war Across the post–World War II period that

we investigate, and according to the data that we use in this book, the median timing for mediation is

48 months into the conflict The typical intervention is deployed 23 months into the conflict.Peacekeepers are almost never sent before a conflict becomes an all-out civil war; just about allpeacekeeping operations in the history of the United Nations did not start until at least one—and mostoften several—years after the initial fighting began.18 In contrast, the threshold of a conflict’sescalation to a full-scale civil war is usually reached within 12 months after hostilities began Thesetemporal patterns suggest that a reorientation toward preventing armed conflict escalation is bothnecessary and potentially fruitful Nevertheless, third-party involvement in nascent, pre–civil warconflicts has received only scant scholarly attention.

To be sure, much work by scholars and practitioners has focused on the role of third parties inpreventing conflict by improving structural conditions Many of these efforts focus on developmentassistance and institution-building.19 But while such long-term efforts may help mitigate the generalconditions that give rise to political violence, such efforts and the actors associated with them havelittle ability to help defuse concrete and time-sensitive armed hostilities between governments andopposition movements.20

We therefore focus on the early stages of domestic armed conflicts in this discussion In a typicalscenario, civil wars start as relatively small clashes between the government and a domesticopposition: that is, minorities, political movements, or other insurgent groups, all of whom canpotentially turn into organized rebel groups After initial eruptions of violence, governments andopposition groups typically face the important choice of mobilizing further, which risks conflictescalation to civil war, or of trying to return to peace by accommodating each other’s demandswithout significant further use of force Taking a cue from the bargaining theory of war,21 we see boththe rebels and the government as actors motivated by prospective gains, but also sensitive to the coststhat they may have to endure Thus, both actors follow cost-benefit calculations to determine whethercontinued fighting or a peaceful settlement would be more beneficial Importantly for our perspective,further large-scale organized violence can also be avoided if these cost-benefit calculations can bechanged to favor a peaceful resolution

We argue that certain international third parties can help change these calculations early in theconflict The answer, however, is unfortunately less straightforward than merely urging thatmediation, peacekeeping, or intervention efforts take place sooner As we explain in the next chapter,these approaches are a poor fit for civil war prevention They fall short on one or more of thefollowing dimensions First, deploying them quickly is often not feasible Second, their political-strategic nature introduces undesirable uncertainty into the calculations of both opposition andgovernment overwhether these activities are indeed forthcoming And third, they are temporary fixesthat are not all that well suited to addressing the particularly challenging long-term crediblecommitment problem that stands in the way of ending conflicts.22 This commitment problem inconflict bargaining occurs because, lacking a powerful, long-term guarantor of peace, each side fearsthat the other side will not hold up its end of a potential peace deal Instead, each party will usenegotiation phases or ceasefires to better prepare itself for a future offensive and strike when theconditions are favorable.23 Due to this problem and both actors’ anticipation of it, peace deals areoften difficult to achieve at the early stage of low-level armed conflicts Conflict parties may fail tosign them, or they may not proceed with their implementation That is, if signed, agreements oftenunravel before peace stabilizes

In this book, we argue that there is a significant opportunity in taking a closer look at the role ofinternational third-party actors that have much potential to be quite effective at dealing with these

challenges in civil war prevention We go beyond the “default” third-party response types ofinternational mediation, peacekeeping, and (military) intervention to focus on the role of a particularsubset of intergovernmental organizations ( IGOs) that share some key structural features and a naturalself-interest in conflict prevention These IGOs include multilateral financial organizations such asthe World Bank, various regional development banks, and the International Monetary Fund (IMF).While these organizations are well known, their role in managing the developmental trajectory ofdomestic armed conflicts is not Yet it is precisely these organizations and their influence — often inconcert with other regional organizations or third-party states — that has pushed both government andrebels in low-level armed conflicts toward settling their differences before their country slid intocivil war This influence has materialized in countries as different as Indonesia and Ivory Coast,which we explore in more depth in this book.

More generally, we argue that a certain type of IGOs can be particularly effective in steeringemerging, low-level armed conflicts away from becoming full-scale civil wars These IGOs can be auseful tool in preventing civil wars in the first place We identify these organizations as highlystructured IGOs They are a subset of the broader range of hundreds of IGOs that have come intoexistence since World War II While the primary task of most IGOs is to coordinate policies betweenmember states, IGOs also have a self-interest in keeping member countries free from civil war inorder to achieve the IGOs’ missions For instance, promoting economic development andinternational trade requires, among other conditions, that member countries maintain internal peaceand stability And a subset of IGOs, namely, highly structured IGOs, possess potent policy tools andtangible resources that can be used toward successful civil war prevention Key features of theseorganizations are that they are designed in ways that provide them with notable formal-legal leverageover member countries and that they often command valuable tangible resources that can be offered

to, or be withheld from, member countries Member states value these resources highly Access tothem is the reason why these states joined the organization and incurred the sovereignty costs ofsubmitting themselves to the institutional structures in the first place

The central thesis of this book is that international institutional structures are a key factor inpreventing civil war The extent to which a country experiencing a low-level domestic armed conflict

is embedded in highly structured IGOs can shape whether the emerging armed conflict on its territorywill escalate to full-scale civil war We briefly preview this argument here and then expand upon itthroughout the book Highly structured IGOs share the ability to act against the preferences orobjections of a member government, command over valuable resources, and an interest in memberstates’ stability These three factors enable highly structured IGOs to shape the incentives ofgovernments and potential rebels in the bargaining process before conflict escalation to civil wartakes place For instance, development banks and trade organizations alike have frequently promisedassistance with postconflict reconstruction to countries at the verge of, or in the midst of, civil wars

—but always under the condition of stopping the violence first And because domestic instabilityunduly affects the mandate of highly structured IGOs, governments and their opponents can count onthe involvement of these IGOs Otherwise, these institutions lose resources that they have previouslyinvested, risk having their missions compromised, and suffer reputational costs as well

This is not necessarily the case with other types of third parties Their involvement is often moreuncertain and subject to political decision making United Nations peacekeeping, unilateral ormultilateral interventions, and even mediation are all subject to political and strategic incentives oftheir own, as well as to the approval of multiple governments, or at least political actors within thethird-party itself Moreover, third-party involvement, especially in the form of other states or alliance

organizations, is often partisan.24 The costs of continued fighting may thus raise for one side (e.g., therebels), but decrease for another (e.g., the government) This gap in costs makes peace more elusive.

In contrast, any involvement of highly structured IGOs is biased toward peace because this is the mostdesirable outcome for these organizations

Highly structured IGOs can also address the credible commitment problem The threat of costsimposed by these IGOs not only alters the rebels’ and government’s incentives toward peace in thepresent time, but also constrains them from resuming the fight in the future The interest of theseinstitutions in maintaining the peace over the long run, as well as their continued ability to providetangible benefits and impose material punishments, help curtail potential future misbehavior by boththe rebels and the government By reducing uncertainty about future behavior, highly structured IGOsthus help mitigate commitment problems In this vein, they increase the odds of prolonged peace Thisargument mostly builds on the ability of highly structured IGOs to put pressure on membergovernments But, as we explain in Chapter 3, highly structured IGOs also provide a similar, yetindirect, incentive structure for the rebel side to pursue and commit to peace as well

Our argument implies that conflict management effects of highly structured IGOs will be greatestwhen a state is simultaneously a member of multiple such IGOs This helps reinforce the conflictmanagement function of each IGO It also makes penalties more multifaceted and severe and providesmore rewards for desirable behavior, such as honoring peace commitments It also increases thechances that one highly structured IGO will take a leadership role in helping resolve the conflict; and

it creates an opportunity for other IGOs to multiply the incentives that push both sides toward peace.The previous paragraphs constitute only a preview of the argument that we fully develop in

Chapter 3 In Chapters 4, 5, and 6, we carefully check the historical record for evidence on thisargument Across all low-level armed conflicts all over the world since 1945, we find that countriesthat are members of larger numbers of highly structured IGOs experienced substantially lower rates ofconflict escalation Follow-up analyses show that the conflict-mitigating influence of these IGOs alsoresults in a better record of conflict settlement, and that highly structured IGOs exercise theirinfluence by shaping the costs and benefits of conflict escalation In case studies of Indonesia in thelate 1990s and Ivory Coast in 2010–2011, we show in more detail the direct and indirect role thathighly structured IGOs took in these conflicts Conversely, we also investigate bargaining in theabsence of a strong influence of highly structured IGOs by considering the conflict between the Syriangovernment and opposition in 2011–2012 That case suggests that where highly structured IGOs areless present, an emerging conflict is more likely to escalate to a full-scale civil war

1.3 Contributions and Implications

This book makes multiple contributions to academic research First, by focusing on more subtle ways

in which IGOs can effectively shape domestic conflicts, we go beyond the traditional IGO conflictmanagement roles of restraining interstate conflicts.25 In doing this, we also demonstrate that domesticconflict management is not an exclusive domain of organizations such as the United Nations Instead,many IGOs have the potential to mitigate domestic conflicts, even though this is not their coremission These IGOs engage in shaping belligerents’ choices in domestic armed conflicts even beforeconflicts escalate to civil war Often, they do so effectively and prevent the escalation of tensions tocivil war This suggests that future research on civil war prevention should incorporate internationalinstitutions in both theoretical work and empirical investigations

Second, our findings show that there are positive externalities to certain institutional designs The

benefits of international institutional design of the type that we identify in highly structured IGOsextend well beyond the original core institutional mandates, such as trade promotion, to the politics ofdomestic conflict An important stream of the institutionalist literature has focused on the “rationaldesign” of international institutions.26 This research suggests that states design institutions to makecredible commitments and to reap the benefits of cooperation By demonstrating that IGO structurecan also affect domestic political developments, our findings show that high levels of internationalinstitutionalization can have benefits both beyond the international level of analysis and the corepolicy issue of the respective IGO Although few of the IGOs that we classify as highly structuredwere established to prevent the escalation of domestic armed conflict as their core mission, states’involvement in these IGOs is more likely to result in more effective domestic conflict management.That is, spillover effects and positive externalities from memberships in highly structured IGOs fordomestic conflict management are an important, non-trivial addition to well-established arguments ondesirable international effects of institutional design.

Third, the book demonstrates the importance of international institutional environments for thetrajectory of domestic conflict processes In doing so, it contributes to the emerging discussion on thetransnational dimensions of domestic armed conflict, advocated by Kristian Gleditsch, JeffreyCheckel, and others.27 In this context, scholars have investigated the role of external actors such asstates; ethnic, religious, or political diasporas;28 international mediation attempts;29 or changes in theinternational system.30 Focusing on states’ participation in networks of highly structured IGOs, ourstudy extends the concept of transnational aspects of domestic conflict, pointing to an understudiedaspect of more indirect transnational conflict management In combination with our findings, theproliferation of institutionalized cooperation between states31 and ever-increasing levels of IGOs’decision-making independence from member states32 imply that effects of centralized IGO structureswill likely become an even more influential factor in shaping the trajectories of domestic armedconflicts in the future

Finally, by highlighting the concept of domestic conflict escalation, we help bring together the level armed conflict and civil war strands of the domestic conflict literature And while our resultsimply that the determinants of lower-level political violence differ somewhat from those of full-scalecivil war—especially with regard to the role of a state’s institutional embeddedness at theinternational level—this is an important integrative step that can help spur new directions in thiswell-known research area While the extant literatures on low-level domestic political violence andcivil wars are both rich, less effort has been made to integrate the two Treating the two phenomena

low-as part of one trajectory provides the opportunity to approach conflict management efforts from thevantage point of conflict escalation This opens new possibilities for investigating potentiallyeffective ways of preventing full-scale civil war

Among the practical implications of our work, one important consideration is that we shed light on

an important third-party international actor whose contributions to civil war prevention have notreceived due attention As we demonstrate, highly structured IGOs have the potential to beparticularly effective in shaping the trajectory of low-level armed conflict While domestic conflictmanagement is typically not the mission of highly structured IGOs, these IGOs have the potential tomitigate conflicts Arguably, they are more effective at this task than traditional conflict-managingIGOs, such as the United Nations By focusing on highly structured IGOs and their roles, we pointpolicymakers interested in civil war management toward an already existing and potentially highlyeffective tool As civil wars continue to inflict their many human, societal, economic, psychological,

and material costs, having a better handle on civil war management is of the utmost priority Highlystructured IGOs can help here, as we find that their involvement can stop internal armed conflictsbefore they turn into civil wars.

Also noteworthy is the consideration that civil war prevention through highly structured IGOs isrelatively straightforward These institutions do not have to be persuaded that preventing a civil war

in a member country is something that should concern them Decision-makers and staff in theseinstitutions already know this, as domestic instability unduly affects their missions Thismakes themall the more likely to take an interest in emerging conflicts and seek to use their influence to preventtheir escalation In other words, these institutions’ awareness of and involvement in conflicts has thepotential to be automatic, precisely because domestic peace and stability in member countries is part

of the self-interest of these IGOs Their goals of peace and stability mean that the institutions’ interest

in emerging conflicts is neutral, not favoring one side or the other Highly structured IGOs insteadprioritize the return to peace and stability Furthermore, highly structured IGOs can effectivelyperform the long-termrole of a guarantor of peace that is critical to overcoming the crediblecommitment problem Performing this role is again in these institutions’ self-interest, meaning that theorganizations and their staff have sufficient motivation to engage in it These points all comparefavorably to conflict management efforts by peace-keepers, mediators, alliances, or third-party statescontemplating militarized intervention To greater or lesser extents, such efforts tend to be subject topolitical decision making, partisan preferences favoring government or rebel sides, delays, and lack

of long-terminterest These problems lead to an increased likelihood of escalation and prolongedcivil wars, with all their deleterious consequences Highly structured IGOs can and do circumventthese issues by acting fast, by being neutral, by their ability to impose penalties and offer benefits, and

by their interest in the long-term stability of member countries

Finally, our approach of treating low-level armed conflicts and civil wars as part of one trajectory

is not only of academic interest, but is also important for international policy External actors havelittle influence on spontaneous and rapid occurrences of political protest turning violent But there is

an opportunity for international institutions to affect the further development of such episodes ofpolitical violence Such opportunities come specifically from enhancing the conflict-endingcommitments of participating actors and signaling negative external ramifications of prolongedviolence As we demonstrate, highly structured IGOs are rather effective at preventing low-levelarmed conflict escalation to full-scale civil wars In this vein, our research also encourages emergingefforts of highly structured IGOs to coordinate their activities on armed conflict Some highlystructured IGOs have already established steady communication channels and share knowledge onthis issue Our findings about the effectiveness of highly structured IGOs in conflict prevention shouldfurther encourage these efforts

1.4 Outline of the Book

In Chapter 2, we continue with the themes raised here We further develop the argument that civilwars have been a particularly perplexing and important issue facing countries around the globe andthe international community This sets the stage for arguing that novel ways of examining civil warsand their management are needed Specifically, we advocate in favor of a developmental view ofcivil war and highlight the problem of conflict escalation The issue of how to avoid conflictescalation is key to preventing full-scale civil war Chapter 2 also surveys the existing literature oncivil wars, suggesting that both structural approaches and, especially, explanations focusing on

bargaining failure are central here Then, applying our insights to the bargaining failure approach, weidentify key problems that would need to be overcome in order for third parties to prevent full-scalecivil war These challenges include (a) swiftness of response, (b) the will and ability to imposetangible costs on (and offer benefits to) governments and rebels, and (c) long-term involvement Ouranalysis shows that the typical third-party civil war management approaches (mediation,peacekeeping, and intervention) fail to adequately address one or more of these issues, making theminadequate tools for the prevention of low-level armed conflict escalation to civil war Thismotivates our argument in favor of focusing on a different type of third-party that could arguably play

a particularly constructive role in civil war prevention Highly structured IGOs are these thirdparties

In the first part of Chapter 3, we define highly structured IGOs and list all organizations that matchthe definition Of particular importance are institutional features that give these institutions leverageover member states We also describe temporal and spatial patterns in the evolution of highlystructured IGOs and states’ participation in them The second part of the chapter presents the core ofour theoretical argument, focusing on the interplay between civil war development and highlystructured IGOs Here, we argue that highly structured IGOs have an inherent, vested self-interest inthe domestic peace and stability of member states Additionally, due to their institutional structure andcommand over substantial tangible resources, highly structured IGOs can act quickly to alter the cost-benefit calculations of both the government and rebel sides in their domestic bargaining interaction.The stakes of highly structured IGOs in member countries’ internal peace and stability are enduring.This means that these IGOs can continue to deter rebels and governments from steps toward conflictescalation long after an initial ceasefire Such enduring involvement is critical to overcoming thecredible commitment problem This satisfies all three of the conditions for successful conflictmanagement and civil war prevention that we identify in Chapter 2 Chapter 3 ends with a description

of exactly how highly structured IGOs engage in sanctioning behavior and with the exposition of ourmain hypothesis

The main purpose of Chapter 4 is to put the hypothesis to a systematic empirical test To this end,

we present a quantitative research design for examining the role of highly structured IGOs in a largenumber of low-level domestic armed conflicts We describe our empirical domain, unit of analysis,variables, and note our statistical methodology In keeping with standard practice, we define a low-level armed conflict as the occurrence of politically motivated violence within a country resulting in

at least 25 battle deaths within one year.33 Our main outcome variable, escalation to civil war,delineates whether a low-level armed conflict became a full-scale civil war and surpassed thethreshold of 1,000 casualties, a long-standing cut-off in the literature on civil wars.34 Since WorldWar II, roughly one-third of more than 260 separate low-level armed conflicts have escalated to civilwar The analysis provides strong evidence in favor of the hypothesis that countries that belong to alarger number of highly structured IGOs face a significantly lower likelihood that an emerging low-level armed conflict on their territories will escalate to a full-scale civil war The impact issubstantial: the likelihood of escalation for typical cases at the higher end of participation in highlystructured IGOs is cut to less than half compared to cases at the lower end

Chapter 5 further probes the validity, robustness, and additional implications of the main findingpresented in Chapter 4 We present a series of analyses to that end For instance, we investigatewhether the correlation of IGO participation and lower escalation risk is spurious and find that this ishighly unlikely to be the case We also show that low-level armed conflicts in countries thatparticipate in larger numbers of highly structured IGOs indeed end in the more desirable outcome, as

our theory suggests They are more likely to end through peace settlements than by one side defeatingthe other militarily before escalation to civil war took place Examining a key mechanism behind ourtheory, we also focus in particular on those highly structured IGOs that have the greatest ability toimpose costs and offer benefits to member governments and rebels This helps us isolate not just theorganizational structure, but also the ability and will to impose costs, as the factors that drive ourmain finding of a decreased escalation likelihood In a final set of different analyses, we expose ourinitial findings to more scrutiny by comparing them to other common correlates of armed conflict andcivil war.

The goal of Chapter 6 is to provide more detailed and nuanced qualitative case evidenceunderpinning our argument and main empirical finding We demonstrate that highly structured IGOsindeed undertake the particular conflict-preventing activities identified by our theory, and that rebelsand governments respond to these activities Specifically, we discuss evidence that highly structuredIGOs threaten and sanction member states that are at risk of conflict escalation Furthermore, highlystructured IGOs offer benefits conditional on successful resolution of low-level armed conflicts.Finally, we show that member-state governments are aware of IGOs’ concerns and respond to theirpressure—as do the rebels We use evidence from three different domestic armed conflicts andfollow their developmental trajectories We select conflicts with different escalation outcomes anddifferent levels of participation in highly structured IGOs to illustrate the link between these twovariables

The first case is Indonesia’s response to East Timorese demands for autonomy in the late 1990s,with Indonesia being comparatively well integrated into highly structured IGOs The case illustratesthe constructive influence of highly structured IGOs in managing the East Timor Crisis, during which

a very real possibility of a civil war was averted The second case is that of Ivory Coast in late 2010and early 2011 With this case, we show that even in instances in which there is a more traditionalmilitary response (i.e., a French intervention) in an emerging conflict, highly structured IGOs stillplay an important conflict-managing role As our narrative demonstrates, highly structured IGOs wereinstrumental in placing sanctions on a regime that was trying to hang on to power despite losingelections, as well as providing incentives to the rebels to settle the conflict While French forcesactively helped remove the old regime from power, highly structured IGOs were arguably key topreventing further violence and getting the opposing sides to settle Through this influence, theyprevented future conflict escalation and set the stage for subsequent peace This is no small feat, asresearch on military interventions demonstrates that such interventions often prolong domesticconflicts and thus actually contribute to civil wars.35 That such an outcome did not take place in IvoryCoast is arguably a result of the escalation-preventing involvement of highly structured IGOs

The early stages of the conflict in Syria in 2011 constitute our final case Despite the undesirableeventual outcome of conflict escalation and its dramatic consequences, this case is consistent with ourtheorizing It demonstrates that in a countrywith a shallow portfolio of participation in highlystructured IGOs, the international community had few ways of curbing the emerging violence Due tofew memberships, the Syrian government faced few constraints or potential benefits for avoidingescalation Thus, it had few incentives not to continue to fight the rebels and saw little point in settlingwith and accommodating the opposition Without highly structured IGOs’ counterweight to curtail thegovernment in its desire to fight the rebels, the rebels themselves saw little reason to stop their armedresistance In addition, the case illustrates the difficulty of obtaining the necessary agreement amongconventional external powers, either states or the UN Security Council, to take decisive action Theresult is a particularly brutal and deadly civil war

The final chapter of the book, Chapter 7, offers concluding thoughts We summarize our argumentand findings, as well as discuss five key lessons learned from our analyses These lessons includesuggestions for future research on political violence, conflict management, and internationalorganizations In the same context, we present additional implications for policy-makers concernedwith managing armed conflicts and point to opportunities for making use of our findings to furthercapitalize on the role of highly structured IGOs in preventing conflict escalation.

We divide the scholarly literature on the causes into two broad categories: the structural perspectiveand the bargaining approach, focusing on interactions between governments and rebels Along theway, we note what the findings from each perspective imply for policymakers interested in conflictmanagement The discussion shows that while existing theoretical approaches have provided muchvaluable insight and there have been many successful policy efforts to manage civil wars, these warsnevertheless continue to occur with disturbing frequency and consequences We argue that, therefore,

an ontological shift in how we think about civil wars and their management is necessary Managementefforts have largely focused on how to end already ongoing civil wars and how to prevent theirrecurrence after (interim) periods of peace Many of these efforts occur rather late, after civil warshave already caused severe suffering and damage Instead, we argue that it is necessary to think about

ways of managing domestic armed conflicts, with an eye toward civil war prevention.

In the second part of the chapter, therefore, we examine civil wars from the perspective of theirdevelopment Thinking about civil wars in developmental terms allows us to consider the issue of

escalation of low-level armed conflict into full-scale civil war Importantly, we can then ask what

can be done to prevent this escalation, thereby avoiding full-scale civil war Relying on insights fromthe bargaining approach, we analyze the dynamics in the bargaining between a government and anopposition group or potential rebels The analysis suggests that changing the government’s and rebels’cost-benefit calculi regarding the costs of escalation early in their emerging armed conflict can havethe positive effects of making a peaceful settlement more desirable in both the short and long run.Considering both short and long time horizons helps ensure that the peaceful settlement not only isreached, but also remains attractive even after the fighting has stopped The latter issue helps guardagainst escalation through conflict recurrence at a later point

Third parties in the international environment constitute one important set of actors that canpotentially affect the cost-benefit calculations of rebels and governments engaged in an armedstruggle, and thereby provide potent antiescalatory incentives Yet, as we discuss in the third part ofthe chapter, forms of third-party involvement that have been receiving much scholarly and practitionerattention—such as mediation efforts, peacekeeping, and diplomatic, economic, and militaryinterventions—suffer from some inherent shortcomings that often make them inadequate for the task ofpreventing the escalation of low-level government-rebel armed conflict into full-fledged civil war

We provide the logic underlying this conclusion It generally rests on three observations First, theseconventional types of policy efforts are simply not geared to address the conflict before escalation tocivil war has already occurred Second, third parties’ self-interests create time-inconsistencyproblems that undermine their efforts to stop armed hostilities That is, third parties involved in

mediation, intervention, and peacekeeping efforts tend to sit out the emerging conflicts on thesidelines all too often, or for too long They hence cannot be reliably counted on to become involved

in emerging armed conflicts at all, or, if they do so, to act quickly enough to prevent escalation tocivil war Third, these efforts’ long-term effects are questionable Most mediation, intervention, andpeacekeeping missions are involved only in the short to medium term, eventually leaving the conflictzone

Instead, a novel understanding of the available options for preventing escalation is needed Thecore of this understanding is that conflict prevention benefits a good deal from a third-party actor thathas a more automatic, consistent, and long-term self-interest in preventing civil wars We argue thatsuch an actor can be found in the form of a subset of intergovernmental organizations (IGOs) to whichthe state at risk of civil war already belongs While this topic is introduced in this chapter, its fullexploration follows in Chapter 3

2.2 Causes, Consequences, and Management of Civil Wars

Conservative estimates of civil wars’ cost in human life in the post–World War II period rangebetween 5 and 7 million deaths.1 And in addition to a continuously mounting toll of casualties, civilwars have devastating social and economic consequences For example, public health institutions andprocedures of post–civil war states are often severely disrupted,2 and subsequently, populations ofsuch states face higher risks of suffering or dying from various infectious diseases.3 Displacedpersons fleeing armed conflicts can destabilize politically not only neighboring countries, but alsowhole regions.4 Furthermore, civil wars inhibit economic growth within affected countries5 and theirneighbors6 by harming political stability,7 individual households, and the labor force;8 theseramifications of civil war are typically stronger than those suffered after interstate war Finally,recent studies report that civil wars trigger high levels of political and social intolerance.9

These consequences of civil wars are in part responsible for setting the stage for future civil wars,making civil war recurrence a major problem for countries that have experienced civil wars in theirpast That is, the economic, social, and political consequences of civil wars are also the core factorspredicting civil war onset, putting states into the so-called conflict trap from which an escape isdifficult The result is that nearly half of all civil war countries return to violence within five years ofthe cessation of hostilities, making civil war recurrence one of the greatest problems on the securitymap of the world.10 Accordingly, preventing civil war recurrence has taken on increased salience forpolicymakers and academics Yet, the fact that civil wars continue to occur, and recur, to this daywith grim regularity11 suggests that there is much room for improvement in how the internationalcommunity deals with civil wars—but not for lack of trying Given civil wars’ dire consequences, it

is not altogether surprising that scholars and policymakers have expended much effort over the pastdecades in order to both understand causes of civil wars and develop policies for managing them Tothese topics we now turn

2.2.1 Civil Wars from the Structural PerspectiveAfter over twenty years of intense research, scholars have identified a set of factors typicallyassociated with civil wars that tap into both the motive and opportunity to rebel For instance, lowlevels of economic development create grievances that motivate rebellion Poverty also renders the

opportunity cost for rebelling low, as well as generally depriving the state of resources needed tosuccessfully deter or quickly stamp out potential rebellions.12 Inaccessible terrain makes it easier forrebels to find sanctuaries and harder for the state to destroy rebel forces,13 whereas largerpopulations offer greater opportunities for rebel recruitment.14 Meanwhile, unstable orunderdeveloped political institutions, or institutional setups that systematically exclude ethnicminorities, may not be able to channel and resolve grievances,15 so they tend to build up; mountingfrustrations can turn into reasons to rebel.

Additionally, easily extractable but highly valued natural resources (e.g., oil, diamonds, andtimber) may provide both a temptation to rebel out of greed or may function as a means for funding arebellion that may be seeking to address a “legitimate” grievance (e.g., insufficient political rights).16Finally, though often investigated, the factor of ethnic diversity is only occasionally clearly linkedwith civil war.17 Ostensibly, ethnic divisions provide an opportunity for governments to decidewhom to provide and whom to deny political benefits or scarce economic resources And forpotential rebels, ethnic divisions help identify and recruit compatriots, mobilize populations, andhighlight grievances Even though scholars have had a hard time demonstrating a clear, systematiclink between raw ethnic diversity scores and civil war onset, more recent research shows that whatmatters is whether ethnic minority populations are systematically excluded from power and wealthsharing; if they are, this provides civil war–motivating grievances.18

A smaller subset of civil war research has started to tie in the issue of how transnational factorsmay be relevant to civil war onset as well For example, civil wars may spill over into neighboringcountries via refugee flows Additionally, rebels seeking sanctuaries across borders can carryviolence into neighboring countries, especially if they are pursued by their governments.19 Rebellions

in one country may encourage rebellions in others through the diffusion of norms, identities, ormotivations.20 In the recent Arab Spring, for instance, demands for political reforms quickly spreadacross states in North Africa and the Middle East

Foreign interventions, another transnational factor, are typically thought of as occurring in the midst

of civil wars, presumably often triggered by concerns over which side is likely to prevail But some

scholars point out that the expectation of a future partisan intervention also could encourage or

discourage the opposition in its decision-making overwhether to rebel, depending on whether itanticipates a third-party to intervene and, if so, on whose behalf.21 Returning to the Arab Spring,Western involvement in Libya may have created what Alan Kuperman has termed a moral hazard ofintervention,22 leading political opposition in other states in the region to falsely believe that foreignhelp would be forthcoming if they pressed their regimes for reforms

A case such as the Bosnian civil war from 1992 to 1995 illustrates many of these dynamics well.The fighting resulted in an estimated 200,000 casualties and 2 million displaced persons These fledmostly to the neighboring countries of Croatia and Serbia, severely taxing their resources andsomewhat destabilizing them as well Additional refugees fled to countries such as Germany,Sweden, Switzerland, Canada, and the United States This illustrates the broad reach of the war’sconsequences The fighting in Bosnia also severely damaged the country’s infrastructure while nearlycompletely wiping out the country’s economy, as well as its public health and educational structures.Many of the related aftereffects are quite pronounced to this day

Prewar Bosnia lines up relatively well with structural accounts of civil wars It belonged to thelower end of middle-income countries and had rather young and untested political institutions Thedesign of these new institutions implied a change to the prior setup, which had favored the Serbian

population The new design would allow two ethnic groups to ally and exclude the third Thisoccurred when the Muslim and Croatian sides pushed for a referendum on Bosnia’s independencefrom Yugoslavia, which the Serbian population did not want Once violence started, the heavilymountainous terrain made it relatively easy and safe for the Serbian rebel forces to lay long-termsieges on valley cities such as Sarajevo and Bihać, persistently shelling the trapped populations forover three years Reflecting transnational dynamics, the secession of Bosnia was part of a broaderwave of ethnic secessions in the region occurring in the aftermath of the end of the Cold War.Furthermore, one could speculate that the secessions were in part encouraged by the erroneousexpectation of Western military help with shedding socialist political systems And on the surface atleast, the war seems to mesh well with the idea that ethnic diversity is a contributing cause, asmembers of Serbian, Bosniak Muslim, and Croatian ethnic groups have all fought one another But adeeper analysis also reveals important instances in which ethnic and religious dividing lines wereless clear cut For example, Bosniak Muslims fought alongside Christian Croatians against ChristianSerbs; and Bosniak Muslims from the westernmost reaches of Bosnia allied themselves withChristian Serbs against the Muslim-dominated government in Sarajevo The latter example means thatthere was intra-Bosniak Muslim fighting—a phenomenon that occurred on a smaller scale on theCroatian side as well.

A result of the war is that Bosnia seems to be a prime candidate for the conflict trap and civil warrecurrence, with a significantly damaged economy, fragile and problematic institutions, poor publichealth and educational systems, a tense and unstable political situation, and other problems That thisoutcome has not (yet) occurred may be due to over two decades’ worth of massive internationalassistance Yet, many observers have questioned whether the war in Bosnia will resume after theeventual departure of international personnel and suspension of aid.23 This suggests that one of thekeys to preventing future fighting may lie precisely in the long-term—ideally permanent—involvement of international actors We return to the feasibility of this scenario later in this chapter

Although the structurally oriented stream of research has increased our understanding of civil wartremendously, converting the related causes into potential solutions to actual civil wars isunfortunately not all that straightforward Policy implications stemming from conclusions based onstructural arguments, for example, would include eradicating poverty and developing stable, mature,and ethnically inclusive democratic institutions Although such goals are certainly noble, they are notthat easy to achieve The transformation of poorer countries with unstable political institutions intowealthy, stable, and inclusive democracies is challenging and likely to take generations.24 Decades ofefforts and resources provided by the World Bank and regional development banks in countries such

as Haiti and Uganda attest to the difficulty of successfully bringing about economic and politicaldevelopment Furthermore, altering other factors from the list—for example, preempting expectations

of future foreign military assistance, reducing large populations, decreasing ethnic diversity, andflattening mountainous terrain—may not be desirable or even possible In short, many of the commonstructural factors linked with civil war can generally be understood as either immutable, especially inthe short to medium run, or undesirable to change

2.2.2 Civil Wars from the Perspective of Bargaining Theory

A more recent stream of research, in contrast, takes a more dynamic view of why and how civil warsdevelop In its simple, abstract form, bargaining theory25 suggests that wars are a result of aninteraction between two actors that disagree over the distribution of some valuable resource Both

actors are rational, in the sense that they are motivated by prospective gains, but they are alsosensitive to the costs that they may have to pay to achieve their goals Thus, throughout theirinteractions, both actors follow cost-benefit calculations to determine whether fighting or a peacefulsettlement would be more beneficial Wars occur when the benefits of prospective gains from fightingexceed the anticipated benefits of peace.

But fighting can also be avoided, or stopped, if these cost-benefit calculations change This can bedue to, for example, mounting costs, changing relative power balance between the sides, or betterinformation about each other’s strengths and weaknesses Apeace deal that divides a contestedresource according to the sides’ relative strengths can then presumably be struck and (further) fightingavoided This bargain, however, is subject to the expectation that both sides will honor the peacedeal not only now, but also in the future The issue of whether each side can be trusted to be trulycommitted to peace—rather than simply waiting for a future opportunity to revise the new status quo

to be more in its favor—is particularly tricky This credible commitment problem helps explain why

potential peace deals fall through—that is, why fighting sometimes occurs or resumes even though theactors can agree in principle on how to divide contested resources

Civil wars and the bargaining framework

We follow this brief sketch of the bargaining framework by applying it to civil wars.26 Rebellionsleading to civil wars take place because the rebels desire to alter the status quo, which favors thegovernment, in terms of the distribution of political and/ or economic benefits The rebels thus seek tonegotiate a new bargain that would provide them with a more favorable distribution of political,economic, or some other resources If the disputants (the government and rebels) know each other’sstrengths, they can theoretically negotiate a revised status quo that reflects their relative power andhence avoid the costs of (prolonged) fighting

Initial stages of the fighting have the side effect of revealing information about the relative strengths

of the rebel and government sides For example, the rebels’ ability to recruit supporters, accessresources, develop sanctuaries, or secure financing is revealed as the fighting proceeds Similarly,the rebels learn more about the government’s will to fight, the skill of its soldiers, popular support,and tactics So while information problems that would reveal the true relative power of each side arenot necessarily trivial, as soon as the fighting begins, the two sides start to develop an understanding

of each other’s strengths and weaknesses This information can then be used to allow the two sides tostrike a bargain that would redistribute the contested resources in proportion to their relativestrengths Yet overcoming the information problem alone does not necessarily mean that the fightingwill be stopped all that easily.27

Although both the government and rebels may feel ready and willing to strike a bargain to end thefighting—and even agree over the settlement of grievances and demands—this process is seriously

hampered by what is known as the commitment problem In particular, the government often cannot

credibly commit to a fighting-ending agreement The specific sticking point is that such an agreementtypically calls for the rebels to disarm The rebels have to be more or less disarmed to fulfill one ofthe basic definitions and functions of a state: to have the sole monopoly over the legitimate use ofcoercive force on its territory Astate cannot have two different militaries, so the rebels have to laydown their arms; in some cases, the now former rebels are then integrated into the state militarystructure Yet, once the peace deal is signed and the rebels have disarmed, the government is placed

in a powerful position, as the sole armed actor Rebels likely may fear that such a scenario may

provide, in the future, a potent temptation for the government to renege on its promise not to harm thenow disarmed and vulnerable former rebels.

These potential future scenarios present a serious concern for the rebels when they arecontemplating signing a peace deal with the government at the present time Even despite thegovernment’s best intentions and sincerity during peace talks, it ultimately cannot be trusted to hold

up its end of the bargain over the long run That is, at some point in the future, the government couldrenege on the agreement and seek to alter the status quo to be more to its liking The government willhave the means to do this because the rebels will have disarmed It also has the motive because thenegotiated bargain represents a compromise that the government had to agree to Free fromconstraints, the government would presumably rather revise the bargain to be closer to what it trulywants if it could Knowing—or, perhaps more accurately, fearing—this possibility, the rebels are noteager to sign a peace deal with the government in the first place Alternatively, the rebels may signthe peace deal without an actual intention to follow through on its stipulations; they may do this to buytime or to create an appearance of goodwill on their part, all the while “knowing” that the government

will eventually renege The bargaining failure, therefore, occurs due to the government’s inability to

credibly commit to the terms of the proposed peace agreement In particular, bargaining fails due tothe expectation that the government will harm the rebels once they disarm, or else reverse a previousdeal away from the negotiated outcome Assumptions about the government’s insincerity consequentlymake the rebels less likely to trust and follow through on any peace deal

Civil war management

Although the previous scenario seems to paint a rather pessimistic picture, it simultaneously provides

an important insight into what it would take to successfully manage emerging civil wars In terms of abroad solution, one key ingredient appears to be the manipulation of the government’s and rebels’cost-benefit calculations to make the continued fighting costlier and the return to peace morebeneficial We discuss these comparatively simpler issues a bit later Of more immediate concern arepolicies that could help the government credibly commit to the peace deal; they would help signal tothe rebels the government’s sincerity to honor its commitments, especially the promise not to harm therebels once they have disarmed At issue is that there has to be some considerable cost for thegovernment if it reneges on the peace deal This cost would then decrease the government’s motive torenege and provide a potent deterrent With a dependable deterrent in place, the rebels will be morelikely to trust that the government will not renege on the peace agreement or harm them in the future

In turn, this also makes the rebels more willing to sign on to and abide by the peace deal An effectivebargain that ends the fighting (i.e., a peace treaty) can therefore be struck.28

Consistent with these insights, both researchers and conflict resolution practitioners haveinvestigated tools to figuratively tie the hands of governments Along these lines, one notable idea isthat there should be some guarantor of peace whose important task is to help deincentivize thegovernment’s interest in reneging on the peace deal and attacking the rebels in the future Scholarshave suggested that the credibility-enhancing “guarantor of peace” role can be performed by domesticinstitutional mechanisms For instance, former rebels or neutral domestic actors could be put incharge of institutions, such as an independent judiciary,29 so that they may act as a counterweight orotherwise have the ability to impose a costly punishment on government leaders who may (want to)renege on the peace agreement in the future.30

Alternatively, the task of keeping the government in check can be performed by third parties from

outside the country For example, Barbara Walter31 highlights the role of international peacekeepers.

As an armed force, they can physically protect the rebels against a potentially reneging government,thus helping assure their safety Having such a mechanism in place helps reassure the rebels that itwould be prohibitively costly for the government to attack them in the future; feeling safer about theirpeacetime prospects, the rebels are therefore less reluctant to sign and abide by a conflict-endingpeace treaty Furthermore, Beth Simmons and Allison Danner note that the International CriminalCourt (ICC) can function as a commitment device to reduce intrastate violence and human rightsviolations.32 That is, the ICC could prosecute government leaders who are tempted to harm the rebels

in the future

In all these and other instances, the function of third parties is to deter the government fromreneging on the commitment to implement the peace deal and not harm the disarmed rebels in thefuture, simultaneously offering the rebels assurances about their future safety With such mechanisms

in place, the rebels will be more willing to commit to and abide by a compromise agreement Thisscenario avoids the bargaining failure Both of these examples of third-party involvement also curtailthe rebels from violating the treaty terms later on, as peacekeepers and the ICC are not only expected

to police the behavior of the government; potential future agreement violations by rebels fall undertheir purview as well

The bargaining perspective also helps shed light on some important war dynamics in the case ofBosnia Backed by the Yugoslav People’s Army (JNA), the Serbian rebel forces were vastlymilitarily superior to their Muslim and Croatian enemies that controlled the central government Thus,though the Serbs constituted only about one-third of the country’s population, their military mightsuggested that they could extract disproportionate political concessions from the government.Unsurprisingly, the government (ostensibly representing the remaining two-thirds of the population)found compromises in which the Serbian side would control 80 percent of the country’s territory(among other examples) unacceptable Meanwhile, the Serb rebels had little incentive to acceptanything less than a deal that would reflect their military dominance A bargained compromise thatwould have prevented the fighting or stopped it once it begun thus was not possible, and the conflictcontinued for years

The situation started to change only when North Atlantic Treaty Organization (NATO) forcesintervened in 1995, altering the military balance radically in favor of the government The newsituation made the Serbian side much more willing to compromise It led to the signing of the DaytonAccords, which stipulated roughly equal sharing of territory and political power between thegovernment and rebel sides As bargaining theorywould anticipate, a particularly tricky part of theDayton Accords for the Serbian rebels was the requirement that their armed forces be dissolved Thisobstacle was generally overcome by the promise of on-the-ground NATO forces that guaranteed thesecurity of the Serbs once their forces disarmed Many institutional concessions providing theSerbian side with notable political influencewere also helpful in reassuring it that the governmentwould not be able to renege in the future and take over the Serb Republic In some form, many ofthese guarantees are in place to this day

Of course, these and other policies for overcoming the credible commitment problem are notinfallible For instance, internal institutional arrangements are subject to the willingness of thegovernment to engage in meaningful power sharing—that is, to giving (former) rebel enemies realpower with which they can curtail future government actions Neutral domestic parties, independentjudiciaries, and other similar arrangements can also become coopted or pursue agendas of their own,now with greater power, making them less than ideal solutions

Also, note that the Bosnian government accepted the Dayton Accords’ institutional compromisesonly under heavy international pressure, and these compromises are largely blamed for making thecountry ungovernable and dysfunctional to this day In terms of third parties, the ICC is still arelatively new institution and extradition of suspects may depend on the cooperation of the veryparties potentially involved in violations Furthermore, Terrence Chapman and Stephen Chaudoin33find that only those states that are unlikely to run afoul of the ICC anyway actually join it In otherwords, the ICC has little potential to change expected behavior Chapman and Chaudoin thus argueagainst Simmons and Danner’s34 conclusion that the ICC can function as a commitment device.

Meanwhile, peacekeepers are deployed in only a fraction of civil wars.35 Indeed, Western powerswere quite hesitant to become involved in Bosnia for over three years despite the media making itquite clear that serious brutalities were taking place there The eventual NATO intervention andpeacekeeping operation likely occurred only because the brutal war was creating persistent publicpressure and affecting Western Europe via refugees and economic destabilization for years Thus,while it proved helpful in Bosnia, counting on NATO to intervene in many other conflicts is notrealistic In a related point, there is a controversy over whether peacekeepers deployed into conflictzones are there to help the local populations or are actually pursuing agendas relevant to their sendingorganization or states.36

Similarly, evidence from the 1990s suggests that the willingness of the United Nations (UN) tosend peacekeepers does not make the signing of peace agreements more likely.37 Even ifpeacekeeping helps in some cases, peacekeeping mandates are rarely indefinite, so the third-partyability to deter agreement violations may well have an expiration date; this then calls into question therebels’ willingness to believe the government’s credibility in the long run, which brings us back tothe commitment problem In sum, the proposed solutions to the commitment problem may not beutilized all that frequently or are subject to shortcomings of their own The fact that roughly one-half

of civil wars tend to recur38 speaks to the issue that more broadly applicable and effective solutionsneed to be identified and employed

Furthermore, these approaches tend to focus heavily on ending ongoing, full-blown civil wars, andespecially on the post–civil war period Much as the case of Bosnia illustrates, the primary policy

objective seems to be averting the problem of civil war recurrence (i.e., relapse into fighting after

the original civil war is over) This is certainly a worthwhile goal, but an even more desirable

outcome would arguably have been achieved if the civil war could have been prevented in the first

place This would help minimize problems associated with civil wars, such as mass casualties,widespread human suffering, atrocities against civilians, refugees, high levels of political and socialintolerance, economic devastation, damage to the country’s infrastructure and educational and publichealth systems, etc Additionally, the problem of recurrence would be avoided, along with the quiteserious concomitant issue of the civil war trap.39

To be sure, ours is not just a critique of the approaches emanating from the bargaining perspectivethat focus on resolving the problem of credible commitment Standard conflict managing policies such

as mediation and intervention40 generally deal with revealing information, clarifying preferences,hammering out compromise specifics, or altering the relative power of the government and rebelforces Yet, they too are typically implemented only after the fighting has become quite intense andprolonged (i.e., once a full-fledged civil war is underfoot) This may be because it takes time beforethe fighting attracts sufficient international attention, before needed resources are mobilized, or before

it becomes clearer that the fighting is heading toward an undesirable outcome for one or more third

parties In any case, simply stating that peacekeepers or mediators should be deployed sooner isunfortunately insufficient; we explore this theme in more detail later in this chapter.

Bargaining theory provides several important insights for managing civil wars which are quiteuseful to us and which we utilize as key building blocks in developing our own theoretical logic.Next, we use three ideas in particular First, much as it is now broadly assumed in the civil warliterature, the government and rebels are thought to be rational actors, in the sense that they aremotivated by prospective gains but also sensitive to the costs that they may have to endure Thissuggests that manipulating the government’s and rebels’ cost-benefit calculations could be a usefulway of motivating them to strike a peace bargain rather than continue fighting Second, when properlymotivated to seek a peace deal as per the previous point, the commitment problem between thegovernment and rebels has to be resolved as well The key to securing a credible commitment is that

a long-term deterrent cost against reneging on the peace agreement needs to be present Third,we takeparticular note of the ability of outside third parties to serve as long-term guarantors of future peaceand thus push both the government and the rebels toward striking and maintaining their commitment tothe peace agreement

Even though we rely on key insights from the extant civil war management literature, we criticallydepart from much of this research (and policy) by shifting the ontological focus away from how to end

ongoing civil wars and minimize chances of their recurrence to the subject of civil war prevention.41

We argue that the insights afforded by bargaining theory–inspired research can be applied, with somemodifications, to the early stages of government-rebel confrontations—a time that we argue is muchmore ripe for settling differences and striking a lasting peace deal We hence maintain that the third-party cost-imposing and credibility-enhancing roles can be particularly effective when they arebrought to bear earlier in the life cycle of the conflict In that respect, the government and rebels can

be persuaded to end their dispute before the fighting reaches full-fledged civil war To achieve this

end, third parties that can be counted on to (a) react quickly, (b) impose costs so as to make thecontinuation and escalation of the fighting unattractive, and (c)maintain their interest in the disputeover the long run are needed

Later in this chapter, we discuss why the civil war management practices that academics andpolicymakers typically focus on generally fail to satisfy one or more of these criteria In Chapter 3,

we argue that a certain subset of IGOs to which the state experiencing the initial stages of an armedconflict may belong can react quickly, can successfully tip the rebels’ and government’s cost-benefitcalculations toward a peace deal, and can simultaneously address the commitment problem But weset the stage for these themes by first considering how civil wars develop Since they do not eruptovernight, we argue that taking stock of their development process provides opportunities for third-party efforts to be brought to bear on a budding conflict with an eye toward ending the dispute before

it escalates to full-fledged civil war

2.3 Civil War Development and Escalation

We approach civil wars from the developmental perspective, particularly taking stock of thedynamics relating to how initial skirmishes turn into sustained, high-intensity armed confrontationsknown as civil wars For the sake of simplicity, we conceptualize the development of civil wars asconsisting of two general stages: the initial, low-level armed conflict phase and the later, civil warphase The breakdown of a life cycle of an armed conflict into two segments allows us to study the

process of transition from the first phase into the second one That is, our focus is conflict escalation,

which is a concept employed in the research on interstate militarized conflict,42 but not necessarily inthe study of internal armed conflict.43

We see the low-level phase as the most opportune time to end an armed conflict In part, this isbecause many of the most serious consequences of a civil war have either not occurred yet or haveonly begun to materialize For instance, there are fewer casualties, less material destruction, and lessdamage to the country’s economy and public health and education systems Without the conflict havingintensified yet, there are also (temporarily) fewer incentives for revenge and reciprocation of attacks,less opportunity for feelings of enmity44 to become entrenched, and less need for ethnic unmixing.45 In

an atmosphere where some trust still remains, striking a conflict-ending bargain should thus be easierthan when all these consequences come to fruition, mount, and do so on a large scale Furthermore,once full-fledged civil wars are underway, they are difficult to stop, as the literature on civil warduration shows.46 For example, civil wars in Ethiopia, Sudan, Colombia, and Peru have raged fordecades And even when a civil war is eventually stopped, the danger of recurrence is so high thatKarl DeRouen and Jacob Bercovitch47 refer to many domestic conflicts as enduring internalrivalries.48 As we noted previously, policy tools to prevent recurrence are few, and not necessarilyfoolproof For both practical and normative reasons, such as avoiding large-scale loss of life, wetherefore focus on the question of how to prevent the escalation of low-level armed domesticconflicts into full-scale civil wars

Considering a broadly generalized trajectory of domestic armed conflicts, most such conflicts arisebetween governments and domestic rebel groups: minorities, political movements, or other insurgentgroups Rebels may attack first to press their demands for ending government repression, betterminority treatment, economic benefits, territorial autonomy or independence, increased politicalpower, or other goals Alternatively, the government may use repressive force against a domesticgroup that is acting against the government, through organizing demonstrations, strikes, putting upbarricades, or refusing to implement a government policy Clashes where rebels violently challengegovernments or where governments use force against domestic opponents are likely to result in atleast a limited number of fatalities In purely quantitative terms, a threshold of, for instance, 25casualties, which satisfies the criteria for low-level armed conflict according to the UCDP/PRIOarmed conflict data set,49 can be reached relatively quickly.50

After such eruptions of violence, which characterize the first stage of conflict, governments andrebels typically face the important choice of mobilizing further, or of trying to accommodate eachother’s demands without significant further use of force If the sides want to press their case, they riskescalation of the conflict to the second, civil war stage While the first phase of conflict maysometimes be unexpected or unplanned, the transition to the second, civil war phase typically takesmore concerted action That is, the second phase takes more resources, planning, strategizing, andrecruiting

For an example of the opportunity for conflict prevention during low-level conflict phase and theescalatory spiral that we describe here, we summarize the early trajectory of the Sri Lankan conflict,

as reported by James Fearon andDavid Laitin.51 Fearon and Laitin describe the Sri Lankan case as anillustration of their argument about ethnically motivated conflict But the description also clarifieswhy our distinction between the low-level, early stage of conflict and full-scale civil war is animportant one Tensions between the Sri Lankan government and the Tamil ethnic group initiallystarted in 1956, after the government faciliated the passing of a Sinhala-only language bill In the later1950s and 1960s, the Sinhalese government suppressed periodical Tamil demands and riots.52 During

that time, the Sinhalese-led government negotiated continuously with the Tamil opposition,53 butduring these negotiations, the Tamils’ frustration grew, and smaller groups radicalized and beganpromoting violent resistance Both sides committed comparatively smaller acts of violence during the1970s In the early 1980s, government repression increased by conventional measures Sri Lanka’shuman rights record worsened significantly during that time.54 In 1983, what was previously a low-level armed conflict escalated to a civil war during an increasingly accelerating sequence of violentevents These events included the army killing 60 civilians and an ambush killing 13 soldiers.55Violent clashes like these increased in frequency and severity With brief interruptions, the resultingcivil war lasted until 2009 This case illustrates that civil wars do not occur overnight (see Chapter

1), but typically develop over time In the case of Sri Lanka, formal and informal bargaining betweenthe Sinhalese government and the Tamil opposition continued for more than a decade since the initialcontentious issue, the Sinhala-language law

Taking a cue from the bargaining theory of civil wars and insights developed in the earlier part ofthis chapter, we note that in their decision-making regarding the future steps in the developing conflictsituation, both rebel and government leaders will be sensitive to the costs and benefits of theirinteractions That is, each actor will compare the worth of the achievement of their respectiveeconomic, political, ideological, and/ or policy goals with the risks and costs involved in reachingthem If either party’s calculations suggest that continuing to fight will be more profitable than layingdown the weapons and finding a negotiated solution, the rebels, the government, or both may choose

to escalate the fight and pursue the path leading toward full-fledged civil war Alternatively, if thecosts of conflict escalation are high for both parties, then the rebels and government will be moreinterested in pursuing a negotiated bargain than in escalating hostilities toward civil war

Yet, a complication in this rather straightforward scenario is that, when facing the decision ofwhether to escalate the emerging conflict, the rebels and government find themselves in somewhatdifferent strategic positions The rebel-government interaction takes place between two at leastformally unequal actors Governments usually have access to resources that rebels typically cannotprocure: police forces, the military, fiscal authority, diplomatic channels, and others The asymmetry

of domestic politics may thus provide the government with some inherent advantages, creating serioushurdles for the chances of rebels winning a conflict According to recent estimates, the odds of arebel victory are only about 1 in 10.56 The government-favoring asymmetry constitutes an importantpoint because it suggests that the rebels’ strategy and resolve depend on their expectations about thegovernment’s preferred and most likely choices for dealing with the rebels The government is thesomewhat more decisive actor in our conceptualization, and we therefore focus much of our attention

on the incentive structure of the government Nevertheless, we also explore the rebels’ perspective inmore detail later in this chapter

Even though the government enjoys some inherent advantages vis-à-vis the rebels, it is also subject

to international influences Some of the government’s international environment can constrain thegovernment against pursuing the path of escalation to civil war Broadly speaking, this environmentconsists of other states, institutions that the state belongs to (IGOs), international commitments made

in treaties, or economic relations with other states This international environment may affect thegovernment’s decision on whether to negotiate with the rebels or escalate the conflict to a civil war.For instance, a country that is a party to all major human rights treaties and the ICC may sufferreputational and material costs should it try to engage in ethnic cleansing against a minority To besure, this does not mean that the government cannot engage in escalatory behavior Instead, the

international environment makes such activities more costly for the government If the violent conflictwith the rebels continues, and if government agents commit gross human rights violations, thegovernment may face costly international sanctions, apprehension of government leaders travelingabroad, or blockade of government funds on deposit with foreign banks All these effects are morelikely in countries that have more and deeper connections with their international environment than incountries that are comparatively isolated; we return to this difference later in Chapters 3 and 4.

Importantly, recall that when applying bargaining theory, costs are central They ultimately shapethe calculus of whether one set of actions (e.g., a peace settlement) is more desirable to pursue thananother (e.g., escalation toward civil war) Higher external costs of escalation toward civil war,provided by the state’s international environment, may well help tip the government’s calculustoward settling its differences with the rebels and away from engaging in escalatory actions Putbriefly, the greater the international costs of escalation toward civil wars, the more the governmentwill prefer a negotiated settlement to escalation toward civil war

This strategic calculus has an indirect impact on the rebels as well, even though they are typicallynot directly and extensively involved in international politics Recall the previous intuition that, atleast in a formal sense, the government is at an inherent advantage in a domestic bargaining setupagainst rebels Yet, the international environment provides a counterweight to the government Thiscounterweight helps the rebels and mitigates some of the government’s initial advantage But thisbenefit for the rebels comes with strings attached Those international actors that constrain thegovernment against escalating the conflict would likely also not find it desirable to have the rebelspush toward civil war Should the rebels attempt to exploit the government’s constraints, IGOs andother international actors would likely respond negatively, with noteworthy consequences for therebels Although rebels and opposition movements do not have diplomatic standing and thereforehave no direct connections to IGOs and other international actors, they may have expectations ofreceiving future benefits from abroad If the rebels continue to fight, these expected benefits willlikely not materialize, amounting to a future punishment For example, development IGOs andfinancial institutions can deny the rebels, or areas under their control, development assistance, or theymay freeze financial assets abroad that are related to the rebel organization or its benefactors Wereturn to exploring this issue in Chapter 3

Our logic suggests that the development of emerging domestic conflicts depends in a significantpart on the extent of the international-level costs and consequences that can be imposed on thegovernment should it choose to escalate a low-level conflict into a civil war Indirectly, as notedpreviously, this affects the rebels’ own calculations about whether to escalate or settle the low-levelconflict Governments facing fewer international constraints may find it less costly to keep fighting therebels (i.e., escalating the conflict to the civil war phase) Consequently, the rebels respond to thelack of external constraints on the government as well:without constraints on the government, thereare few incentives of international origin for the rebels to avoid escalation toward civil war

An example of this unconstrained path may be the original development of the civil war in Syria Agovernment largely isolated from international institutions could not be constrained significantly bythe international environment Free of international constraints, the Syrian government provided arather dismissive response to the opposition’s initial demands for reform And the government’s lack

of interest in the opposition’s demands, coupled with the international inability to force thegovernment to the negotiating table, meant that the antigovernment side saw conflict escalation as amore meaningful path toward reaching their objectives than negotiation with an unwilling opponent.The failure to seek and reach a negotiated end to the initial phase of the conflict ultimately resulted in

an escalation toward a particularly deadly, long-lasting, and complex civil war.

By contrast, a government facing strong constraints at the international level may find it too costly

to escalate its fight with rebels Escalating to civil war would have tangible, international-levelconsequences The latter type of government, therefore, has a greater incentive to settle the conflictearly and thus avoid escalation Additionally, the international constraints placed on the governmentwork in the rebels’ favor Yet, the rebels need to be careful not to overplay their hand by takingactions that would lead to escalation toward civil war, lest they lose the international counterweight

to the government This incentive structure, then, makes it desirable for the rebels to want to settlewith the government as well, thus avoiding a full-fledged civil war An example along these linesmay be the case of Indonesia and East Timor’s independence Under international pressure, both thegovernment and rebels pursued a settlement at a negotiating table before the fighting escalated intocivil war Both the Indonesian and Syrian cases are discussed at length in Chapter 6

Furthermore, recall that in the context of discussing the bargaining theory earlier in this chapter, wenoted that the mere willingness to compromise by both sides is not necessarily sufficient to lead to apeace deal being signed and implemented (i.e., to avoid conflict escalation toward civil war) Toavoid a relapse to fighting and the concomitant conflict escalation, away to deal with the commitmentproblem is needed as well We argue that under certain conditions, third-party actors from theinternational environment can help resolve this issue as well The key is whether the escalation-curtailing costs—or at least a credible threat of them—can be maintained over the long run Thecontinuing (threat of) costs ensures that the government (and therefore the rebels) continue to abide bythe peace deal This deterrent also makes the peace deal more likely to be reached in the first place.Later in this chapter, we examine the ability of common third-party civil war managing policies toovercome the credible commitment problem in the context of escalation

Finally, time is central to this approach to explaining and preventing escalation Precisely because

we take the perspective of preventing the escalation of low-level armed conflict into full-fledgedcivil war, international-level action needs to take place before the conflict escalates That is, it iscritical for third-party actors interested in civil war prevention to act quickly, to be able to alter thecost-benefit parameters of the rebel-government bargaining interaction before the fighting escalatesand reaches the civil war phase As we show next, civil war managing policies that are the commonfocus of academic investigations and practitioner efforts unfortunately do not necessarily allow swiftresponse by third parties This means that the answer to the question of what can be done to preventcivil wars is unfortunately not as simple as using standard tools of mediation, intervention, andpeacekeeping sooner A novel understanding of available options in preventing escalations is neededinstead; this is the topic of Chapter 3

2.4 What Does It Take to Prevent Low-Level Armed Conflict from Escalating to Civil

War?

The previous discussion suggests that third parties in the international environment can play aconstructive role in preventing civil wars Consistent with this observation, various internationalactors have been trying to manage civil wars for some time using a variety of techniques Forexample, many organizations and high-profile individuals have conducted numerous mediationattempts57 in conflicts ranging from the Democratic Republic of Congo to Northern Ireland Otherpopular tools include peacekeeping missions,58 foremost by the United Nations, but also by other

organizations such as the African Union, the Economic Community of West African States(ECOWAS), or NATO These missions have attempted to keep peace after conflicts in countries likeBosnia, Rwanda, and Sierra Leone Commonly employed as well are diplomatic and economicsanctions,59 such as those placed on the former Yugoslavia in the 1990s, via multilateralorganizations or groupings of states Finally, though not necessarily neutral, militarizedinterventions60 such as the NATO mission in Kosovo are also sometimes employed with an eyetoward managing civil wars.

Yet, while there seems to have been no shortage of international involvement in civil wars, justabout all past efforts have focused on either ending already ongoing, full-fledged civil wars orpreventing their recurrence By contrast, we are interested in the topic of civil war prevention Forthis, a key question is whether these policy efforts can be brought to bear successfully on emergingarmed conflicts before they escalate to civil wars

As explored in the discussion in the preceding section, successful escalation prevention effortsneed to satisfy three fundamental conditions The first is the immediacy of an international response.Time is of the essence Although for the purposes of our analysis, we have divided the development

of civil wars into two phases, in practice there is not necessarily some notable pause, like theintermission of a play, that affords international actors the opportunity to notice the conflict, decidewhat to do, and mobilize and deploy their resources—all the while the government and rebels arepatientlywaiting for the actual international response In reality, the transition from a low-levelconflict to a civil war is a continuous process that rarely waits for other parties to become involved

When an opposition movement is trying to extract concessions from its government and thegovernment responds with repression and brute force, the historical record suggests that theopposition is ill advised to wait and endure repression and violence until another country or aninternational organization intervenes and attempts to stop the government And even in cases wherethe opposition movement is initially a coherent organization whose leadership prefers nonviolentprotest, it is unlikely that leaders will be able to continue to keep all wings of the organization fromusing force to defend themselves against government violence.61 Similar observations hold for thegovernment side as well A government facing a rebel movement that uses force to push for itsdemands will have strong incentives to use all means available (including force) to contain the rebelmovement In other words, neither the government nor the rebels can be expected to wait, but ratherwill press their agendas quickly; waiting could well mean providing the other side with a notablestrategic advantage Therefore, if low-level armed conflicts are to be stopped before they escalate,any external influence on the rebel-government bargaining process would need to be applied withoutdelay

The second set of conditions dealing with a successful prevention of escalation by a third-party isthe ability and will to impose tangible costs on the government and rebels This has to do with thealteration of the rebels’ and government’s cost-benefits calculi, presumably in a manner that would

make escalation more costly for both parties—and peaceful settlement therefore more attractive.

Ideally, the threat of third parties’ cost-imposing involvement will be so credible as to assure boththe government and rebels that their choice to resort to large-scale violence will be met with aguaranteed reaction from the external conflict management agent Yet, while there are several third-party actors that could potentially impose significant costs on the government and rebels (for instance,globally or regionally powerful states or organizations with military forces at their disposal, such asNATO or ECOWAS), few third-party actors are actually interested in becoming systematicallyinvolved in emerging armed conflicts around the world and suffering related costs of their own The

lengthy and public debate over whether the United States and its allies would punish the Syriangovernment militarily for its alleged use of chemical weapons during the Syrian civil war is only onehigh-profile example of this inconsistency.

The uncertain (if not downright weak) threat signals from third parties that they will becomeinvolved in a manner costly to the rebels and government does little to dissuade conflict parties fromescalation To use an analogy, in traffic safety, permanently installed radar controls combined withwarning signs routinely lead to reductions in speeding.62 The effect is that the credible threat of apenalty changes behavior In contrast, a reliance on random enforcement by a sometimes-presentmobile radar gun is a lot less effective; the costs for misbehaving are uncertain, thus doing less tocurtail speeding effectively Similar dynamics can be said to apply to the issue of conflict escalation:when the imposition of punishment is far from guaranteed, one can expect the behavior associatedwith escalation to occur more frequently

Finally, the third condition to prevent escalation of low-level armed conflicts into civil wars is thedurability of third-party involvement As noted in our previous discussion, the government’s inability

to credibly commit to a potential peace deal is a serious stumbling block in ending armed conflictbefore escalation to civil war Without a credible commitment, the rebel-government peace deal isunlikely to be reached in the first place, or it may unravel in the future The literature suggests thatthird parties can be helpful with this issue by providing a check on the government’s will and ability

to harm the rebels in the future.63

This is certainly helpful, but only up to a point If the terms of the peace agreement hinge onguarantees made by a third-party, the natural challenge for the rebels and government is to prepare forthe time when the third-party withdraws or relaxes its involvement Going back to our traffic safetyanalogy, speeding would likely resume in an area from which the fixed radar surveillance isremoved The impermanence of third-party involvement changes the incentive structure vis-à-visabiding by a peace deal and can lead to strategies where the rebels, the government, or both aresimply waiting the third-party out before resuming hostilities The solution, then, would seem to bethat the third-party involvement would need to be quite lengthy, if not permanent Few third parties,however, are in a position to make an indefinite commitment to sustaining peace deals, as their owninvolvement is costly and there may be new emerging conflicts to manage elsewhere Witness, forexample, the withdrawal of U.S forces from Somalia after a costly mission, as well as the UnitedNations Operation in Somalia (UNOSOM) mission’s end in March 1995

2.5 The (In)adequacy of Common Third-PartyConflict Managing Policies in

Preventing Conflict Escalation

Third-party civil war management strategies have attracted much attention from scholars andpolicymakers We now evaluate these strategies based on the three conditions for preventing theescalation of low-level armed conflict to civil war Our analysis suggests that each third-partyconflict management policy may be able to address one, or maybe two, of these conditions, but

generally not all three Starting with mediation, its obvious advantage is that it can be fast Mediator

teams can be deployed quickly to conflict zones, as they are highly mobile and require few resources,

at least initially Mediation, however, fails to satisfy the second and third criteria Mediators cannottypically impose tangible costs on the government and rebels and in fact often depend on theirgoodwill to continue operating Without significant leverage, therefore, the mediators would have a

difficult time affecting the rebels’ and government’s cost-benefit calculations This shortcoming alsomakes them ineffective on the third issue Although the mediators can remain involved in a conflictover the long run, their inability to curtail the government means that they are an ineffective device forovercoming the credible commitment problem Indeed, recent research on the lasting effects ofmediation in civil wars suggests that while external mediation attempts in intrastate crises can help

with short-term reduction of violence, they frequently do not result in lasting settlements.64

Furthermore, skeptics may point out that mediators often do not become involved in a conflict thatquickly.65 In fact, Michael Greig’s findings on Africa are quite telling: “For an average low-intensityconflict, there is less than a 01 probability that it will be mediated within the first three years of theconflict and less than a 02 probability of mediation within the first five years.”66 This might be thecase because in the early stages of the conflict, the parties are overly optimistic about winning, somediators are less likely to be invited until the conflict has become entrenched Alternatively, themediators themselves are strategic actors, so they pick and choose among conflicts and selectivelydeal with the more devastating and high-profile ones Addressing a conflict with more notoriety andgreater media coverage provides the mediators with more public exposure as well, helping toadvance their statures and careers Whatever the actual cause of potential mediation delay is, we putthis premise to the test in Chapter 5 Our findings reveal that mediation attempts generally take placeonly after a low-level armed conflict has already escalated to a civil war

Turning to peacekeeping, research clearly shows that peacekeepers are generally deployed inattempts to stabilize post–civil war countries.67 The primary concern of peacekeeping seems to be theprevention of civil war recurrence This obviously goes against our first recommendation that thethird-party responses occur before escalation to civil war takes place The possibility of apeacekeeping mission is usually not even discussed by potentially interested states or organizationsuntil the fighting is well underway and has become large scale Furthermore, administrativeprocedures related to authorizing a mission, gathering resources, and ultimately deploying thepeacekeepers take significant amounts of time, which is measured in months to years Changing thisstrategy toward proactive, early deployments into early conflict stages would require a fundamentalrethinking regarding what peacekeeping mandates are about Additional requirements would includemaking significant changes in administrative procedures and having a standby peacekeeping force.While such changes have occasionally been broached at the United Nations, there is little reason tobelieve that they will advance beyond the idea stage in the foreseeable future

For the second condition—the will and ability to impose tangible costs—peacekeeping seems to

be handicapped by its typical use in the (post–) civil war conflict phase rather than in thepreescalation phase Yet, organizations such as the United Nations or NATO have on occasionembarked on peace-enforcement missions, where peacekeepers are deployed into active conflictzones in order to make the fighting stop An example of such a mission would be Somalia With(usually) well-armed troops and military hardware on the ground, this would seem to satisfy thesecond escalation-preventing condition regarding the imposition of tangible costs Yet, peace-enforcement missions are unlikely to take place before the conflict has already become large scale,such as in Somalia Furthermore, peace-enforcement (and, more generally, peacekeeping) operationsare a matter of choice States and organizations (first and foremost the United Nations) make strategicdecisions about the cases into which to send peacekeepers This has led to arguments that the actualdeployment of peacekeeping troops by multilateral institutions is heavily influenced by their mostpowerful members and their strategic and economic interests.68 Even if one disagrees with such

notions,69 the fact remains that, historically, only about one-third of civil wars have receivedpeacekeepers.70 So although the ability to impose tangible costs on the government and rebels in theearly stages of conflict can be said to exist in theory, the track record relating to the frequency of theactual peacekeeping, let alone peace-enforcing, deployments casts serious doubt on peacekeepingbeing a reliable tool for preventing conflict escalation.

But, when it is employed, peacekeeping does well on our third condition, the durability of party involvement Although peacekeeping mandates are typically not that long, they can be extended

third-if the circumstances on the ground call for continued peacekeeper presence Long-running UNmissions such as Bosnia, Kosovo, Cambodia,Cyprus, and the Golan Heights serve as good examples

of the ability and will to provide a long-termpresence in (still unstable) conflict zones

Turning to military, economic, and diplomatic interventions again provides a mixed ratingaccording to our three conditions for preventing conflict escalation Concerning the first condition,interventions seem to occur only after the rebel-government fighting has affected third partiesmaterially or is negatively influencing their policy goals and interests.71 Motivations along these linesinclude fighters or refugees crossing borders and causing regional instability, concerns about the

“wrong” side winning, or gross violations of human rights, including ethnic cleansing or genocide.Military interventions and/ or diplomatic or economic sanctions are not automatic, and therefore thisresponse is not quick Our empirical analyses in the remaining chapters of this book confirm thatinterventions generally take place only after the civil war stage of the rebel-government conflict hasbeen reached

On the second issue, the ability and will to impose tangible costs, the situation is similar to the onerelating to peacekeeping Interventions can impose substantial tangible costs, but they generally occuronly after conflict escalation has already taken place Furthermore, interventions are also a matter of(strategic) choice72—suggesting that the will to impose costs is not present systematically In fact,Patrick Regan reports that states or international organizations did not intervene in about one-third ofall civil wars since 1945, often with ample time to make that choice as conflicts were raging on.73Another important complication is that military interventions are often not conducted with conflictmanaging goals in mind Instead, many interventions are partisan, hoping to assure victory for aparticular side This has an added unfortunate side effect of actually prolonging civil wars.74

These complications come as a disappointment to those who note that multilateral interventions—perhaps coordinated by an IGO—appearwell suited to divide the burden of military, diplomatic, andeconomic involvements.75 And over the last decades, the number of international institutions,particularly regional organizations with security mandates, has grown substantially.76 One mightexpect that this development would lead to a higher frequency of third-party interventions But thishope is misplaced So far as actual interventions go, the potential benefit of third-party burden-sharing is offset by bargaining over the distribution of costs and potential conflicts of interest betweenthe third parties: third-party states are, after all, the ones that carry the cost of deploying troops and ofeconomic sanctions Hence, interventions as a tool for imposing costs in the context of low-levelarmed conflicts to prevent their escalation do not appear to be a policy tool that can be counted onconsistently

Again similar to peacekeeping, interventions have the ability to provide enduring incentives againstfuture government-rebel fighting Although military interventions typically tend not to be long termdue to their high costs, sponsoring countries and organizations have shown the will to employdiplomatic isolation and/or economic sanctions over the long run Cases such as the economic

sanctions against and diplomatic isolation of South Africa, Iran, North Korea, or Serbia come tomind Although not all of these deal with civil wars, the implication is that the diplomatic and/ oreconomic intervention policy tool may be available for civil war prevention, so far as the crediblecommitment condition is concerned.77

Table 2.1 An Assessment of Common Third-Party Conflict Management Policies

Conflict Management

Policy

Swiftness of Response

Ability to Impose Costs Before Escalation

Ability to Overcome the Credible Commitment Problem

present

but seldom used

Present

Intervention (Diplomatic,

Economic, and/orMilitary)

Lacking Theoretically present,

but seldom used

Present

Bringing together these points paints a picture where the common third-party conflict managingtools typically fail to satisfy one or more of the three key conditions necessary to prevent low-levelarmed conflict escalation to civil war identified previously Table 2.1 summarizes this analysis Anatural question is whether some combination of various policies could overcome this shortcoming.For example, theoretically at least, mediation can be used to offer a quick response, satisfying thefirst condition These efforts can then be linked with diplomatic and economic sanctions to providetangible antiescalatory incentives for the government and rebels, addressing the second condition.Finally, peacekeepers could be deployed to assure the long-term survival of the peace agreement thatprevented escalation to civil war, supporting the third condition Such a scenario would be highlydesirable After all, we have witnessed coordinated responses, whereby mediators, sanctions, andpeacekeepers have all been utilized in situations such as the Bosnian conflict

Yet we have serious doubts about the practical feasibility of coordinated conflict managementstrategies when it comes to preventing the escalation of a low-level armed conflict between thegovernment and rebels into a full-scale civil war Our primary concern is the quick timing of theresponse, the first condition The closest that any of the common conflict managing policies comes tosatisfying this condition is mediation But mediators unfortunately are not all that quick to react Even

if they were, it would still take much time to bring interventions/sanctions on line in order to affectthe rebels’ and government’s cost-benefit calculi in real time (i.e., before the conflict spirals intocivil war) And coordination across various strategies may not be seamless The seemingly chaoticand indecisive approach to managing the Bosnian conflict again comes to mind as an example.Finally, the use of costly resources such as economic sanctions or peacekeepers still remains subject

to strategic calculations by the actors potentially providing them Given the statistics from priorresearch given in this chapter, there are serious concerns about whether the ability to manage aconflict via tangible cost-imposition can be assumed to automatically and quickly translate into thewill to do so The inconsistency of the response again generates the undesirable uncertainty overwhether the third-party help will be forthcoming so far as a particular conflict is concerned In short,while a combined policy strategy could work in theory, practical obstacles and the track record

suggest that it would not be a realistic tool for preventing low-level armed conflict escalation.

2.6 Our Theoretical Framework for Escalation Prevention

As we discuss in the following chapter, our focus for the constraining function of the internationalenvironment rests on international institutions We identify a specific subset of IGOs that hasdemonstrated much ability and will to impose significant costs on member states, do so in a timelymanner, and remain involved long after the fighting subsides The involvement of these IGOs, in turn,means that the member government’s incentive to escalate the conflict with the rebels will be reducedsubstantially Consequently, the government will be more willing to strike a bargain with the rebelsthat changes the country’s status quo to a position that is more favorable to the rebels Meanwhile, therebels’ incentive to escalate the conflict toward a civil war is also decreased by a promise of arevised, more favorable status quo Alternatively, the rebels would lose this opportunity, as well asthe international counterweight to the government’s power, should they exploit their situationaladvantage and try to escalate the conflict By altering the cost-benefit calculi of both the governmentand rebel sides, a constraining international environment in the form of certain IGOs therefore has theunique ability to push the government and rebels toward reaching a bargain and avoiding escalation tocivil war

I n Chapter 3, we also argue that the same IGOs have a vested interest in seeing peacemakingagreements be both reached and maintained in the long run That is, the government and rebels alikeface a credible threat of future punishment for reneging on the agreement This is important because itdecincentivizes the government’s interest in harming the rebels once they disarm Put differently, thecontinued attention of IGOs and threat of costly future punishments has the effect of resolving thecommitment problem in the long run This problem has been identified in this discussion as one of themajor sources of bargaining failure and, consequently, of civil wars Therefore, the IGO-basedconstraints emanating from the international environment have the effect of making both reaching andsustaining a conflict-ending compromise easier

In sum, examining civil wars through a developmental perspective provides us with an opportunity

to analyze the political process that leads to the escalation of low-level armed conflict toward fledged civil war This approach, in turn, brings into focus the issues of whether escalationsometimes can be prevented, and, if so, how Linking bargaining theory to these issues provides theinsight that settlement is possible in a bargaining scenario in which escalation is costly for thegovernment and rebels alike Yet, wanting a settlement is not the same as actually reaching andsustaining it: potential settlements are prone to the commitment problem A way out of thecommitment problem depends on whether the incentive structure that made the initial settlementattractive will continue to hold We argue that a state’s membership in international organizations thatcan impose tangible costs on member states helps resolve these dilemmas For reasons specified in

full-Chapter 3, certain IGOs find civil wars in member states particularly detrimental to their missions.They therefore have a strong self-interest in becoming involved in the conflict early on, with the goal

of preventing its escalation to a more damaging civil war Equally important, they can crediblythreaten to punish member governments for not settling with the rebels before escalation to full-fledged civil war and for reneging on conflict-ending agreements in the future We turn to theexploration of these ideas in greater detail in the next chapters

(a) Making conflict escalation sufficiently costly for governments and rebels.

(b) Doing so quickly, before the conflict escalates.

(c) These costs and the related change in the escalation-deterring incentive structure are enduring.

In other words, preventing escalation to civil war requires the long-term involvement of a credible

third-party with a vested interest in the conflict Establishing and maintaining these costs, then, are

key strategies for outside actors to help prevent the escalation of low-level armed conflict to civilwar

The theory and evidence presented in this book suggest that certain intergovernmental organizations(IGOs) can effectively address all three of these issues simultaneously The post–World War II era,and especially the past few decades, have witnessed a proliferation of IGOs While the primary task

of most IGOs is to coordinate policies between member states, many of them also directly affect politics within member states This is the case because their primary task of effective cooperation on

the international level depends on member countries’ domestic policies IGOs’ international goalscan be adversely affected by political and economic developments within member states In order toadvance and protect their missions, IGOs therefore often request that member countries alter theirdomestic policies Examples of this behavior are plentiful and span issue areas such as intellectualproperty rights, rule of law, human rights, fiscal and monetary policies, and others But what isperhaps less obvious is that IGOs’ missions often depend on member states being stable and peacefulwithin For instance, economic development and international trade are seriously harmed by civilwars.1 This dependence of IGOs’ successful operations on domestic peace and stability of memberstates in particular is key to understanding that IGOs have a self-interest in keeping member countriesfree from civil wars And, as we argue in this book, a subset of IGOs actually possesses potent policytools that can be deployed to address all three conditions necessary for successful civil warprevention

These IGOs are organizations with established and standing institutional structures, or, as we call

them, highly structured intergovernmental organizations These organizations, by virtue of their

structure and their tasks, have the capacity to (a) respond swiftly, if not automatically, to risingtensions within member countries Additionally, they are in a position to (b) impose tangible costs ondomestic conflict parties And, finally, due to their structure, role, and interest in the long-term