CIVIL WAR CONFLICT RESOLUTION FROM THE PERSPECTIVES OF THE PRACTITIONER AND THE ACADEMIC by Laurie Nathan, Karl DeRouen Jr., and Marie Olson Lounsbery Civil wars are particularly challenging to resolve via mediated negotiations. Practitioners and policy makers employ a variety of approaches to help move warring actors from war to peace. To assist in this process, peace researchers have examined civil war peace processes from a variety of perspectives. In doing so, scholars assume their research will be useful beyond simply accumulating knowledge. It has become evident, however, that the research has not in fact informed policy and practice. This article begins by examining the articles presented in this two-part special issue (April and July 2018) aimed at improving the scholar–practitioner divide. Practical lessons derived from the research are presented. These lessons are followed by a frank assessment of what practitioners need from scholars, as well as recommendations for better bridging this important theory-to-practice divide. INTRODUCTION Civil wars pose a particularly challenging problem for the international community. While such disputes can range from rather lowintensity affairs to all-out destruction, in most, the toll on civilians tends to be significantly high. As a result, human displacement often ensues and regions can become destabilized.1 With the increase in intrastate conflicts beginning in the early Cold War period, a wealth of scholarly attention was paid to trying to understand, first, the causes of these types of disputes, but also how they might be resolved more effectively. This became particularly important as it became evident that civil wars were prone to recidivism; cease-fires and peace PEACE & CHANGE, Vol. 43, No. 3, July 2018 © 2018 Peace History Society and Wiley Periodicals, Inc. 344 Civil War Conflict Resolution 345 agreements were signed only to fail. Yet, the alternative is often unacceptable.2 As a result, efforts to move parties toward a negotiated outcome rather than letting the conflict endure to its bloody conclusion are preferred. Experience has demonstrated, however, that attempts at resolving civil wars must be viewed as a process, one that is complex, varied, and evolving. Practitioners and policymakers employ a variety of approaches to help move warring actors toward a more peaceful set of interactions in the hope that their efforts will prevent conflict recurrence. For their part, peace researchers have examined civil war peace processes from a variety of perspectives. In doing so, scholars begin with the assumption that their research is designed to be useful beyond simply accumulating knowledge. Peace research aims to improve real-world outcomes. It has become evident, however, that the scholarly and practitioner communities too rarely interact with one another; a significant theory-to-practice gap remains. This disconnect exists for a number of reasons Andrew Mack aptly identifies.3 In particular, research outputs often use arcane methods that confound many in the policy world. A common misconception of policymakers is that probabilistic work done using empirical methods can be disproven by a few (or one) outliers. Relatedly, the data sets used by scholars in order to generalize their findings are very often met with intense skepticism. Mack concludes that little has been done to narrow the gap between the two communities. We continue this line of reasoning with more specifics from a mediation perspective below. This special issue emerged through an effort to bridge the divide between scholars and practitioners and begin a dialogue on improving civil war peace processes—one that recognizes the value of improving research outcomes based on the needs of those in the field. To do so, we present a two-pronged approach. The first involved bringing the perspectives of both scholars and practitioners into the same journal issue. We asked each to contribute in their own way to adding knowledge to the topic through research and experience. And second, we conclude this first approach with a research–practitioner dialogue aimed at better understanding how the field can move forward and overcome this important impasse. We have taken notice of Mack’s points raised above and from our own experiences as practitioner-scholars (Nathan) and scholars (DeRouen and Olson Lounsbery) to create a dialogue. In doing so, we 346 PEACE & CHANGE / July 2018 also recapitulate the two special issues that purposely include work from the scholar and policy communities. This concluding article begins with a synthesis of the scholarly literature (including the work in these two special issues) on civil war peace processes to understand how scholars have approached the topic thus far, including how this special issue aims to contribute to the discussion. We also explore the policy aspirations of the scholarly community, the types of policy recommendations that have emerged from their work, and a discussion of the impact it has had on policy. Laurie Nathan, from the perspective of the practitioner and policymaker, then discusses how the scholarly literature on mediation is perceived and poses challenges for scholars aimed at improving the usefulness of scholarly work. We conclude with a scholarly response to Nathan’s challenge by Karl DeRouen Jr. and Marie Olson Lounsbery. We encourage future practitioner–scholarly endeavors that will continue to improve civil war peace processes. THE SCHOLARLY APPROACH TO CIVIL WAR PEACE PROCESSES Scholars have worked to tackle the complex set of interactions involved in resolving civil wars by focusing intently on specific aspects and/or phases of the processes involved. The first step to any resolution process involves getting parties to the negotiating table. Previous work indicates territorial, internationalized, and longer wars are more likely to experience mediation.4 It also seems that more intense wars will gain the attention of potential mediators,5 although it would seem that mediation would potentially be more effective early on as the dispute emerges before harder lines are drawn. Mediation Svensson and Lundgren (this issue) explore a novel determinant of civil war. They show how nonviolent protest movements paradoxically come to be conflict antecedents. They theorize mediation is more likely when the conflicts yield negative outcomes for outside parties. In this event, mediators are inclined to intervene as a form of conflict prevention. Their empirical analysis reveals mediation is most likely when the movement is radical and/or has a propensity for violence. A repressive state also increases the likelihood of mediation. Over time these mediators are increasingly international. Civil War Conflict Resolution 347 The authors note the practical applications of the study. First, the study points to the need for a means of early warning prior to the outbreak of violence. The protests can fulfill this role. Other forms of third-party intervention prior to violence can also prove promising. This list includes fact-finding missions, monitoring, and the application of leveraged political pressure to reduce the threat of state repression. Once parties have agreed to engage in resolution dialogue, the negotiation and (often) mediation takes place. While unassisted negotiations do occur between intrastate conflict actors, the asymmetric power dynamics involved in such conflicts seem to deter effective negotiations that address underlying grievances.6 As a result, thirdparty-assisted negotiations (i.e., mediation) can be a rather important approach for bringing about effective resolution, a finding confirmed by previous research.7 The effectiveness of mediation can vary, however, depending on who is mediating8 and whether the mediator has a stake in the outcome (i.e., is partial to settlement).9 Research seems to suggest that mediators will be more effective if they are trusted by the parties involved or can bring significant leverage to the table. The practical implications of this work are that the selection of mediators, not surprisingly, needs to be made with attention to the conflict actors and dynamics involved. Mediation mandates are a neglected topic in the conflict resolution literature. Laurie Nathan (this issue) addresses this topic by outlining a mandate typology. His overarching point is mandates “constrain and empower” mediators, determine the contours for the process the mediator follows, and have a major influence on the outcome of the mediation. Mandates impact mediation in several important ways. Political mandates provide the mediator with legitimacy and resources. These mandates can also be associated with guidelines and boundaries the mediator must respect. Normative mandates are associated with greater inclusivity and respect for human rights and relevant portions of international law. Political mandates emanating from multilateral organizations may insist on certain outcomes, such as the termination of unconstitutional rule by holding free and fair elections. Donor mandates can result in pressure for an accelerated process and do not allow for general amnesties. The most important aspect of mandates is their ability to shape the mediation outcome. For example, if a mandate creates mediator 348 PEACE & CHANGE / July 2018 bias or impartiality—real or perceived—the process can be irreparably jeopardized. Antje Herrberg’s article (this issue) directly addresses the synapse between theory and practice from a practitioner’s perspective while pointing out the importance of understanding the mediation process in addition to the outcome. The process of which she speaks encapsulates direct talks, negotiation, trust building, dialogue, and socialization. Here, socialization is defined as the working relationship between the mediators and the involved local parties. She notes from experience the process can move in fits and starts. Based upon the work of Omer Shapira, Herrberg contends the effectiveness of the process is a function of eight mediator-centric variables: (1) competence, (2) political will and values (e.g., confidentiality and impartiality), (3) the ability to leverage financial and human resources, (4) legitimacy and a mandate (the mediator can create a sense of obligation in the disputants), (5) information and analysis (e.g., key intelligence), (6) the willingness to take a multitrack approach (provides multiple resolution strategies), and (7) suitable postmediation monitoring (e.g., support for implementation).10 There are several important takeaways from this article. First, care should be taken when identifying mediation cases and studying peace processes for evidence of success. Scholars and practitioners alike must be aware that it is possible to support and study mediation from a variety of “entry points.” Second, an overly rigid focus on outcome at the expense of process when analyzing mediation can be counter-productive. Theorists and mediators will do well to realize mediation is situational and varies across cases. Third, Herrberg emphasizes the actor-centric approach to the study of mediation effectiveness composed of the eight mediatorcentric variables. This approach affords an assessment of effectiveness beyond what one could get by only studying events taking place at the bargaining table. This approach also has relevance for ripeness theory as a mediation team replete with these eight virtues can better facilitate and identify that point in the conflict when peace initiatives will likely bear fruit. Civil wars can also pose a particular challenge to the negotiation process as the number of actors involved can raise questions of inclusivity. When multiple rebel factions are operating simultaneously, the potential for spoilers increases, particularly when negotiations do not include all factions.11 Warring governments may be reluctant to Civil War Conflict Resolution 349 include the most extreme elements, however, choosing instead to focus more intently on some level of compromise with more moderate elements.12 It does appear that “partial peace” is possible under certain conditions.13 Inclusivity poses a particular challenge for practitioners. Avoiding the potential for spoiling seems ideal, but doing so means that warring governments must be willing to engage even more extreme factions. Further, more groups in negotiations mean potentially more divergent interests, and therefore, a smaller window of opportunity for agreement.14 Mediation failure becomes more likely. Perhaps less inclusivity makes more sense when conflicts are less intense and agreements can be crafted that might discourage excluded groups from continuing the fight.15 Once again, this suggests that the resolution of civil wars is indeed a process. Some actors may be excluded at times then come into the fold later. Similarly, it may not be possible to get all parties to agree on all-important issues at the onset. A partial peace agreement, one that addresses certain but not all issues, may lead to subsequent negotiations and agreements that can become a continuing dialogue.16 Ripeness Previous work suggests that the timing of mediation initiatives is crucial to success. In particular, conflicts must be “ripe” for resolution to occur, according to I. William Zartman.17 A conflict will appear ripe when the parties involved feel that victory remains elusive and no one side is making progress toward achieving victory. In essence, the players feel a painful deadlock. In such a situation, negotiation may be perceived as a way out of that “mutually hurting stalemate.”18 How parties arrive at a stalemate will vary across cases. In some situations, battlefield advances might fail to gain traction. Stephen Stedman has suggested that the military element in the conflict is crucial when actors assess their situation relative to each other.19 Actors may arrive at the perception of a stalemate independently, but frequently external actors can help warring actors recognize negotiation as an attractive alternative to their currently painful situation.20 Further, third parties may actually work to move a conflict toward ripeness to bring warring parties to the table more effectively. In a collaboration between Zartman and U.N. mediator Alvaro de Soto, a “Peacemaker’s Toolkit” was produced bringing theory directly to practice.21 In doing so, the authors suggest that third parties may 350 PEACE & CHANGE / July 2018 employ threats/warnings, employ economic measures such as the promise of aid or the threat of sanctions, as well as offer expanded military support (such as peacekeeping) or threaten the withholding of military support. Doing so, however, runs the risk of forcing parties to the table to appease external actors without much willingness to negotiate in good faith.22 Pedro Valenzuela’s (previous issue) contribution focuses on a peace process that was playing out as we were preparing these issues in March 2018. The Colombian peace agreement is now in the early stages of implementation. Valenzuela looks at the events leading up to the agreement that ended the civil war between the FARC rebels and the government that had endured over fifty years. The article focuses upon the aforementioned issue of ripeness defined by Zartman. Valenzuela contends this was the situation in the years leading up to the peace talks in Havana. In addition, FARC’s decades-old central command structure was able to firmly control the rebel fighters during the negotiations lest they withdraw and form cells opposed to peace. Peace Agreements A significant amount of attention has been paid to the crafting of peace agreements with the understanding that the document itself is not only supposed to stop the violence, but also address underlying grievances to better avoid conflict recurrence. Scholars have focused intently on the provisions included in civil war peace agreements to better advise practitioners on the types of agreements that will lead to longer peace outcomes. Various forms of power sharing have been touted as particularly essential provisions that allow civil war actors to transition from war to peace. Political and military power sharing mechanisms have been embraced as valuable tools to provide voice and security to previously marginalized segments of society. Empirical research supporting political power sharing has been mixed, however.23 Ironically, by ensuring that particular groups are represented in the government and military, the solidification of those identities may make it difficult for crosscutting identities to emerge.24 The practical lessons learned through such research seem to indicate that power sharing can be an effective tool for expanding political and military power in warring societies, but mechanisms for moving beyond rigid identity-based politics should be explored. Civil War Conflict Resolution 351 In addition to power sharing, territorial autonomy and third-party guarantors have been found to make for more enduring peace agreements.25 Again, the empirical evidence on territorial autonomy as a form of power sharing is mixed.26 As a result, the application of this type of provision may require more consideration as to the future implications of such an approach. Third-party guarantors, however, appear particularly important for transitioning from war to peace.27 When external actors are present to oversee implementation, but also provide stability and security in the postconflict environment, militias and rebels may be more willing to trust the process and lay down their arms. It may also be necessary to address postconflict justice issues in the text of an agreement so that aggrieved parties can feel comfortable that transitional justice and human rights issues will be effectively addressed, though doing so involves a delicate balance between the need to come to an agreement through bargaining and the need to address abuses.28 While it was expected that more comprehensive and elaborate agreements would make for more enduring peace agreements, that does not appear to be the case. Based on all civil war peace agreements between 1975 and 2011 examined by Olson Lounsbery and DeRouen (previous issue), it seems that longer agreements with more provisions are more likely to fail. Further, those with outstanding issues do not seem to have any less success than those without remaining issues. It appears that peace agreements must be viewed as a process, and perhaps, as Herrberg indicates in her piece discussed above, that future work might consider other measures of success. Having established that peace agreements do not necessarily need to be comprehensive or elaborate for the peace to be maintained, Olson Lounsbery and DeRouen seek to better understand the factors that influence peace agreement structure. Interestingly, stronger rebels, although in a position to demand more concessions, tend to sign agreements with fewer provisions and outstanding issues. Conversely, peace agreements that are mediated tend to have more provisions, and seem more likely to end in the subsequent year. Of course, civil wars are particularly problematic due to asymmetric power relations, which tend to draw the attention of third parties. The less costly use of economic sanctions typically used by external actors to pressure warring actors to end the violence does not appear to have much of an impact on peace agreement structure. The exception is the finding that sanctions within five years of an agreement 352 PEACE & CHANGE / July 2018 tend to be associated with outstanding issue agreement, which does not appear to be a barrier to success. The more costly form of intervention is the use of force. Foreign military intervention benefiting warring governments (supportive interventions) seems to make for more elaborate agreements, while those benefiting rebels (hostile interventions) have fewer provisions generally. The prospects for peace agreement success would, therefore, be more associated with hostile interventions. Anna Jarstad and Desiree Nilsson (previous issue) also explore power sharing provisions. They demonstrate agreement design is in part dependent upon the regime type of the warring state that is party to the agreement. Autocracies are more likely to agree to political and military power sharing arrangements, allowing such regimes to keep power centralized. The authors theorize autocratic governments are wary of granting any real devolution of power to the opposition. Autocracies “co-opt” the opposition by allowing it just enough power and political spoils to keep it in check. The military and the inner circle of the leadership of the autocratic regime hold the real power and maintain a highly centralized system. The decentralized authority inherent in pluralistic societies is amenable to compromise through territorial power sharing. More power is devolved to the rebels in this arrangement. The authors test these propositions with a sample of agreements from the period 1989 to 2004. They report support for their theoretical argument. Interestingly, implementation rates are not significantly different across the regime types. Postagreement Issues The postsettlement environment needs particular attention if peace agreements are going to endure. It seems evident that agreement implementation is a crucial determinant of success. Previous work by Madhav Joshi and Jason Michael Quinn29 indicates that the more extensive the implementation, the more likely the agreement will lead to lasting peace. This work suggests also that implementation itself must be viewed as a process that assists in the normalization of intergroup relations and helps to address commitment problems as well as underlying grievances. When peace agreements include new elections, the timing of those elections may be of concern. The tendency is to press for early elections as a means to transition toward new Civil War Conflict Resolution 353 governance. Scholarly research seems to indicate that practitioners should be wary of pushing elections too early. Doing so may ironically undermine the democratization process.30 Of course, the ability to implement peace agreements may require a level of state capacity that may not exist. Postconflict states that have higher levels of capacity appear better able to implement their agreements, although third parties may be able to bolster state capacity in some cases.31 Similarly, addressing poor economic conditions, as well as the hurt and pain caused by the conflict itself, may also make for more effective transitions to peace. Civil wars significantly undermine, if not destroy, economies and infrastructure. Normalizing, then growing, postwar economies not only builds state capacity but also diminishes the need or desire to resort to arms over inequitable grievance issues.32 Civil wars can also lead to high levels of targeted civilian fatalities.33 Mechanisms to address this victimization so that postwar societies can begin to heal can help make peace more lasting. Scholarly research on postwar societies provides important implications for practitioners in this regard. Joakim Kreutz (previous issue) also takes up a relevant issue in his piece on postconflict society. If a war ends through attrition or unexplained reason, dyads with a history of fighting tend to survive. However, if a document is signed or there is a decisive victory on the battlefield, new actors sometimes emerge to threaten the peace. These new actors can be skeptical of their postwar prospects and as they have battle and organizational experience, can relatively easily revert to a fighting force. His results are derived from an exploration of civil war actor data and country case studies. This can be explained by the actions of the government. These actors tend to maintain their hold on power after a negotiated settlement but the situation is different for rebel leaders who are more prone to lose their status. These fighters who no longer bear allegiance to the rebel command may again take up arms. Obtaining peace in these situations requires the new incarnation of the rebels to recognize the legitimacy of the government if war over the same issue is to be avoided. Taken together, the articles in these two special issues have addressed some important issues in the research–practice divide. It seems clear that the scholarly work presented in these special issues explicitly aims to inform civil war resolution processes, whether through empirically derived advice on how to approach and better 354 PEACE & CHANGE / July 2018 craft agreements or through an examination of mediation timing and approach. Practitioner contributions seem focused intently on better understanding mediation mandates and how one approaches evaluating mediation occurrence. Bringing these differing approaches together has helped highlight the need for further collaborative efforts between scholars and practitioners as we both work to improve policy and practice. Of course, more work needs to be done. Next, we turn to the practitioner’s perspective and then conclude with a response from the scholars. WHAT KINDS OF KNOWLEDGE DO MEDIATION POLICYMAKERS AND PRACTITIONERS WANT? From its inception, the field of peace studies has been driven by a normative commitment to peace and to produce knowledge that is useful for practical purposes.34 In this section, we explore that commitment specifically with respect to mediation. Scholarly research on mediation frequently concludes with policy recommendations.35 But it has had little impact on policy and practice. As Peter Wallensteen and Isak Svensson put it, “mediation research seems to remain remote from the world in which actual mediators find themselves.”36 They conclude that there is “a need for more bridge-building between practitioners and researchers to make this research useful in a world which is still full of significant conflicts of different types.”37 A comprehensive understanding of the gap between mediation research and mediation policy and practice would require identification and analysis of the following: the knowledge needs of policymakers and practitioners; the mediation community’s perspectives on academic research and products; the content, practicality, and accessibility of mediation research; and existing modes of knowledge transmission and acquisition. This section concentrates on the knowledge needed by mediation policymakers and practitioners, the utility of that knowledge, and the means by which it is acquired. These issues will be addressed in terms of the following types of knowledge: conflictspecific knowledge, comparative thematic knowledge, knowledge on mediation best practice, and generalizations based on statistical analysis. The empirical focus is on international mediation undertaken by multilateral bodies and states in high-intensity conflict. We argue that while mediators make extensive use of different kinds of knowledge and research, they are least receptive to the large- Civil War Conflict Resolution 355 n statistical studies that have become the most common methodology employed by mediation scholars.38 This is because mediators view every conflict as distinct, dynamic, and complex. They therefore have an aversion to overly generalized approaches to peacemaking and a preference for responsiveness and flexibility.39 As the United Nations Guidance for Effective Mediation puts it, “an effective mediation process responds to the specificity of the conflict” and must take account of its causes and dynamics; the positions, interests, and coherence of the parties; the needs of the broader society; and the regional and international environments.40 Consequently, mediators find value in comparative knowledge and generalizations that appreciate variations between cases and the intricacies of each case, but they have little confidence in generalizations based on aggregated data that ignore the challenges and specificities of mediation in different cases. This argument is informed by a review of the policy literature on international mediation, cited throughout this section, and by the practitioner and policy experience of one of the authors.41 The policy literature provides a strong indication of practitioner and policymaker preferences for certain kinds of knowledge, although there are no doubt differences of opinion between and within organizations that engage in mediation. Conflict-Specific Knowledge Mediation teams working on a particular conflict have an obvious desire for knowledge about the history and dynamics of that conflict, its structural causes and dimensions, the issues in dispute, the conflict parties and their allies and patrons, and the goals and interests of other relevant actors. They are not concerned solely with hostilities and belligerent forces but also need knowledge of historical and contemporary peace dynamics. This includes learning about previous peace processes in the country, about domestic groups and individuals that play peacemaking roles, and about local approaches to conflict resolution and reconciliation.42 Much of this knowledge is acquired through the mediation team’s engagement with local actors, including researchers at local universities and policy institutes, and with the diplomatic corps and multilateral organizations based in the country undergoing conflict. The teams are also interested in policy research conducted by credible think tanks, such as the International Crisis Group (ICG), and in scholarly 356 PEACE & CHANGE / July 2018 research by academics. Members of mediation teams circulate ICG reports and academic work relating to the conflict, the UN Department of Political Affairs commissions research on particular conflicts, and UN political affairs and mediation support staff in New York attend forums that think tanks convene with experts on a given conflict. The conflict-specific knowledge that is arguably most valuable to mediators is also the hardest to obtain. To fine-tune their approach, mediators are keen to discern the internal strategizing and debates of the conflict parties. As one UN mediator put it: “What I want to know, above all, is the hidden agenda, that is to say, the real agenda, of the parties.”43 This information is confidential and sensitive, however. Conflict parties are typically engaged in a game of bluff with each other and with the mediator, and they are loath to disclose their private debates, bottom lines and tolerance for compromise.44 The importance of knowledge about the conflict parties’ internal deliberations is well illustrated by the British government’s mediation of the Rhodesia/Zimbabwe negotiations at Lancaster House in 1979. Stephen Stedman observes that the success of the talks was due in part to the mediators’ extensive cumulative knowledge of the parties and their leaders.45 This knowledge was built by studying failed mediation initiatives for Zimbabwe over the previous decade. It enabled the mediators to apply bargaining leverage within and between the parties with great precision and forethought. By the time of Lancaster House, the mediators had “compiled information that directly affected their choice of strategies and tactics. Their learning was extensive and was divided into knowledge about the actors, politics within the actors, possible solutions, bargaining tactics, and sources of leverage.”46 This learning was supplemented by information that the British intelligence service had obtained by clandestine means.47 Mediators do not expect that conflict-specific knowledge will provide solutions to the problems that confront them. Rather, its utility lies in building a sound foundation for decision making, strategizing, and process design. The knowledge helps to reduce the risk of error; identify opportunities for progress and the reasons for lack of progress; make sense of the interlinked local, national, and regional conflict dynamics; and shed light on the complex relationships between actors, interests, issues, and objective conditions. Civil War Conflict Resolution 357 Comparative Thematic Knowledge In addition to conflict-specific knowledge, mediators need thematic knowledge because the content of peace agreements typically covers a standard set of issues. On the security front, these issues include disengagement of forces; demobilization, disarmament, and reintegration of combatants; deployment of third-party peacekeepers; and long-term security sector reform. On the political front, the critical issues in comprehensive agreements include power sharing arrangements; respect for human rights and the rule of law; constitutional, electoral, judicial, and governance reform; and transitional justice. There are also thematic issues related to mediation processes. A prominent example of this is the inclusion of civil society, and women in particular, in negotiations to end intrastate conflict.48 On these and other thematic issues, mediation practitioners and policymakers welcome knowledge of comparative experience. Since every case is unique, mediators are not looking for formulaic or offthe-shelf answers to their questions. But they can benefit immensely from drawing on the variations, insights, and lessons of other experiences. This comparative knowledge is especially useful when it highlights measures that have often proven to be either advantageous or detrimental. Mediators are not opposed to generalizations per se, provided they are derived from close scrutiny of relevant cases. By way of illustration, one of the authors was responsible for drafting the chapter on security arrangements in the Darfur Peace Agreement of 2006.49 A vital overarching lesson from historical experience is that these arrangements have to be context specific. They must take into account, for example, the size, location, mobility, armaments, logistics, cohesion, and communication systems of all the belligerent armies. The security chapter was also enriched by reviewing research and talking to experts about security arrangements in other conflicts. The resultant lessons included the necessity to map the location of the belligerent forces, avoid unrealistic deadlines for implementation, develop contingency plans for the eventuality that deadlines are missed, consult the peacekeepers about the viability of the arrangements before the agreement is finalized, and design mechanisms for rapid response to violations of the agreement. The mediation community acquires and transmits comparative knowledge through different methods. One of these is training, at which technical experts present state-of-the-art knowledge on a 358 PEACE & CHANGE / July 2018 particular topic to mediators and support personnel. For example, the UN Mediation Support Unit offers specialized training in ceasefires; women, peace, and security; and religion and mediation.50 Another method of transmitting knowledge is through the deployment of experts to advise and support mediation teams in the field—in 2018 the UN Standby Team of Senior Mediation Advisers covered expertise in process design, gender and inclusion, security arrangements, power sharing, constitution making, wealth sharing, and natural resources.51 The mediation community also commissions and publishes comparative research on a given theme. An example of this is the UN guidance for mediators dealing with conflict around natural resources.52 This is an excellent guide because it reflects expertise on both mediation and national resource conflict, distinguishes between different mediation phases and different categories of natural resources, draws on practitioner insights, and captures diverse experiences through case studies. Knowledge on Mediation Best Practices There is an abundance of best practice material on international mediation, much of it prepared by or for the UN and other multilateral organizations that engage in peacemaking.53 This material covers, among other things, conflict analysis for the purpose of peacemaking, establishing a mediation team, designing a mediation process, preparing an agenda for talks, getting the parties to the table, facilitating dialogue and negotiations, dealing with spoilers, building regional and international coalitions, drafting and finalizing agreements, and supporting implementation. There is also more specialized advice on mediation in relation to specific themes, such as cease-fires;54 women, peace, and security;55 and natural resources and conflict.56 The best practice material cited here is either written by mediators themselves or based on discussions with them and on the after-action reports prepared by mediating organizations. These methodologies generally ensure that the material is credible and not formulaic. But because the methodologies are not rigorous and systematic, there is no certainty that the documents identify best practice accurately. No studies have been undertaken on whether the documents are in fact used by mediation teams and, if they are, whether they enhance the effectiveness of mediation. Civil War Conflict Resolution 359 Generalizations Based on Large-N Statistical Analysis As discussed above, mediators find comparative thematic knowledge useful when it offers general lessons while appreciating the complexities and distinctiveness of each case. By contrast, they doubt the validity and utility of general findings and recommendations based on statistical studies when these studies treat dissimilar conflicts or peacemaking strategies as homogenous, ignore the political and normative goals and constraints of international mediation, and misconstrue what mediators actually do. These problems are illustrated below with reference to quantitative work on negotiated settlements, which is usually the goal of mediation in intra- and interstate conflict, diplomatic interventions in civil wars, mediation identity, mediation strategy, and mediation bias. Our aim is to highlight policy concerns that scholars might take into account in future work. In a study of peace durability following different kinds of civil war termination, Monica Duffy Toft questions whether policymakers are correct to have a strong preference for negotiated settlements.57 She finds that these settlements are largely ineffective in preventing further war, that wars ending in military victory by one side are less likely to recur, and that rebel victories produce the most durable outcomes. These findings are rendered dubious by the heterogeneous nature of the categories “civil war” and “rebel victory” in Toft’s data set. Rebel victories include the Bolivian popular revolt of 1946, the coup in Argentina in 1955, the North Yemeni civil war in 1962, Frelimo’s struggle against Portugal, the Khmer Rouge, the Taliban, the Tutsi invasion of Rwanda during the 1994 genocide, and other dissimilar cases.58 For policy purposes, coups, national liberation movements, genocidal armies, and antigenocide forces are entirely different phenomena. In her policy recommendations, Toft usefully draws attention to the importance of well-designed and robust security sector reform in peace agreements. Failing such reform, she proposes that “support in pursuit of victory, especially victories by rebels, may be a worthy objective.”59 This proposal ignores the vital contextual questions of norms, politics, and strategy. For example, in the case of the Sudanese civil war that ended through a mediated settlement in 2005, policymakers did not consider a military victory by either the ruling National Islamic Front or the Sudan People’s Liberation Movement to be remotely desirable. As with many negotiated settlements, the 360 PEACE & CHANGE / July 2018 mediation and resultant agreement were the choice of the belligerent parties, which recognized that outright victory was unattainable. Even where military victory is attainable, policymakers are constrained by political and normative considerations. For example, a victory by the Sudanese government was probably feasible at the start of the Darfur conflict in 2003 but it was not a tenable option for Western policymakers because of Khartoum’s brutal campaign of ethnic cleansing.60 For present purposes, the key point is that the actual facts of this case, as with every case, are more relevant to policymakers than general recommendations derived from statistical analysis of diverse cases.61 At face value, studies that focus on the identity of mediators may appear to be on safe ground when they distinguish between international organizations, regional organizations, states, and individuals.62 But these are not homogenous categories. There are significant differences in strategy and style between discreet UN mediation authorized by the Secretary-General and more high-profile and prescriptive UN mediation mandated by the Security Council; between mediation conducted by different regional organizations, such as the AU, the Arab League, the European Union, and the Intergovernmental Authority for Development; between mediations undertaken by different heads of state representing the same regional body; between state mediators in different parts of the world; and even between different individual mediators representing the same state.63 Research on mediator identity that ignores these distinctions is too far removed from the reality of international mediation. The same kind of problem can arise in relation to assumptions of homogeneity in peacemaking strategy. An example of this is the study by Patrick Regan and Aysegul Aydin on the effect of external military, economic, and diplomatic interventions on the duration of civil wars.64 The study investigates the effects of these interventions and is especially concerned with the effects of mediation.65 Four categories of diplomatic intervention are considered, namely, mediation, international forums, the recall of ambassadors, and third-party offers to mediate that were not accepted by the conflict parties.66 In the statistical analysis, these categories are combined to create a single, dichotomous variable for diplomatic initiatives by third parties.67 From a policy perspective, this move is not plausible. The four types of diplomatic intervention are not sufficiently alike to warrant any general policy finding or recommendation. Mediation policymakers would be Civil War Conflict Resolution 361 especially skeptical about the assumption that mediation offers that are rejected by the parties have similar effects to actual mediation that is accepted by the parties. The theoretical framework constructed by Regan and Aydin rests on the premise that mediation is effective, or at least useful, principally because it facilitates the transfer of information to the conflict parties: At the core of the ideas about mediation is the notion that mediators transfer information between the combatants, information about relative capabilities, prospects for victory or defeat, possible concessions, reservation points, and other bits of vital information that antagonists cannot credibly convey. This information is critical to understanding the effect of interventions on the duration of internal conflicts. . . .The most important element in facilitating the movement toward a compromise outcome is to provide information.68 The idea that the provision of information constitutes the essence of mediation is evident in several quantitative (and some game theoretic) articles on mediation bias.69 The idea is not substantiated with evidence, however, and it is unlikely to resonate with mediators. Three problems stand out. First, mediators themselves are often acutely aware of how much information about the parties they lack and of the difficulty in obtaining this information.70 Second, the parties get information from many sources, not least their own intelligence capabilities and those of their allies, and there is no basis for believing that mediators necessarily have better information than these sources. Third, the essence of mediation is the facilitation of negotiations, not the provision of information. The mediator’s core functions include providing information but they also encompass confidence building to reduce enmity and suspicion between the parties; assisting the parties to find common ground, engage in joint problem solving, and accept the need for compromise; and helping the parties address the causes of the conflict and craft viable and sustainable solutions in peace agreements. In short, the preoccupation with information provision in certain scholarly studies does not capture accurately what mediators actually do. Another manifestation of the failure to see mediation as facilitated negotiations is the tendency in many studies to investigate the 362 PEACE & CHANGE / July 2018 relationship between mediation strategies and outcomes without taking account of the power, character, objectives, and deliberations of the conflict parties.71 Yet it is the parties, and not the mediator, who are the principal protagonists in the drama of conflict. It is they, and not the mediator, who decide whether to perpetuate the conflict or terminate it through a negotiated agreement. These decisions may be influenced by the mediator’s intervention but they are also based on the parties’ perception of the balance of power, the likely trajectory of the conflict, and the relative risks, costs, and benefits of further conflict versus negotiations.72 Studies that ignore these considerations may be substantially incomplete and end up overstating the role and influence of the mediator. The mediator cannot be studied in isolation from the parties for the additional reason that their relationship is dynamic, the moves of each influencing the moves of the other. As noted in the introduction to this article, Andrew Mack observes that few policymakers appreciate that probabilistic theories cannot be refuted by a small number of counterexamples, leading them to dismiss important findings for the wrong reasons.73 This may be true, but the authors of quantitative studies should anticipate the kinds of questions that probabilistic theories raise for policymakers. If, say, a study concludes that mediation strategy X is generally more effective than mediation strategy Y, this is the beginning and not the end of the story. The prudent policymaker would pose the following questions: Under what conditions is strategy X likely to succeed and under what conditions is strategy Y likely to succeed? What are the respective failure rates of the two strategies? What are the consequences of failure in the two categories? And do we understand the reasons for success and failure? In light of such questions, scholars would be well advised to make their policy recommendations more nuanced, elaborate, and sensitive to context and risk. CONCLUSION Several recommendations emerge from the above discussion. The need for conflict-specific knowledge suggests that scholarly work should continue and even expand case study methodologies. In addition, it would be beneficial to expand the examination of civil war peace processes beyond simply the period of conflict. Relationships evolve and may ultimately ebb and flow in and out of periods of violence and tend to involve repeated resolution attempts. All of this Civil War Conflict Resolution 363 informs subsequent resolution efforts. Scholars have been moving in this direction in some areas of research,74 but more can be done in this regard. The need for understanding the hidden agenda of the conflict parties is indeed challenging—understanding the historical relationships involved could help, but this is more the work of those that know the parties intimately. Skilled mediators may also be able to elicit such agendas. The need for comparative thematic knowledge and moving beyond formulaic answers also suggests that scholars can and should work from detailed case studies when crafting data sets for analysis. The comparative politics methods are appropriate here.75 Being able to select cases with similar characteristics and examine those more closely when needed would be helpful. It is then possible to compare cases that differ only on the dependent variable (most similar system design). This will allow the researcher to ascertain which of the explanatory (independent) variables are important. Similarly, researchers can employ a most different systems approach in which the explanatory variables differ but the dependent variables are the same. The method proceeds by trying to figure out why the dependent variables are the same despite a variety of independent variables. The trend in the field toward disaggregating civil wars76 and the parties involved77 is moving in the right direction. It looks like scholars could make an important contribution to knowledge on best mediation practices by examining the methodologies employed more systematically for the purpose of evaluation. Similarly, research on whether the types of materials and training provided are helpful could be carried out. Program evaluation methodologies could be employed for this purpose. The largest divide between research and practice is likely over the use of quantitative methodologies frequently employed by mediation scholars.78 When creating relevant data sets, we need to code the diversity of cases more effectively so that we are not treating the cases as homogenous when they are not. As the field has developed, scholars have been working in this direction, but more can be done. We as scholars also seem to do a poor job of explaining often complex data operations and statistical analyses; if we are aiming to better inform practice, then we need to write in a way that is accessible to those without quantitative backgrounds or interests. We also need to use case information to better explain the applicability of our approach. 364 PEACE & CHANGE / July 2018 Doing so will take what often seems abstract to the concrete and, therefore, become more informative. Academics, as pointed out by Andrew Mack above, must also explicitly note that their statistical methods produce probabilistic, rather than law-like, results. There are exceptions to more or less all of the empirical studies. Errors arise from variable coding, faulty assumptions, overgeneralization, and imperfect use of statistical modeling. For example, if autocorrelation (a trend in data whereby successive observations are correlated, which can exaggerate statistical significance) is not correctly diagnosed, results can be spurious. The coding of some variables can be difficult if not impossible. For example, how does one code mediation strategy? Doing so can prove vexing as mediation strategy can change over time. What if there are a number of mediations of a war that are unsuccessful and then a different mediator comes along and facilitates peace? Would this be coded as a single success or several failures followed by one fruitful effort? These are challenging questions that can be met head-on if scholars work with practitioners who know the practicalities of the field. Too often these real-world events remain shrouded in mystery to scholars who sometimes exaggerate the belief that mediation is a panacea. The previous section observes that the Toft study lumps civil war types together to make generalizations. Academics must be careful of this trap. One way to resolve this situation is to control for war type. For example, coups tend to be relatively short, thus creating outliers in the data if one is analyzing civil war duration. This can be remedied by specifying a dichotomous variable coded as one if the war is classified as a coup. Another option would be not to include this war type in the data set. Practitioners can inform other coding issues. Some civil wars follow a start and stop pattern over decades. This is especially true of separatist wars fought in the periphery of the state. Examples include several of the civil wars in Myanmar, Indonesia, Sri Lanka, and India. Should these wars be considered episodes of one long war or should they be treated as separate wars when modeling? Sometimes these wars recur when rebel armies fighting over the same territories begin the fight anew but with a different name. Should these recurrences be considered the same war or new ones? A case could be made for either decision but input from practitioners, especially country-specific ones, would be beneficial. Civil War Conflict Resolution 365 It seems also necessary for scholars to read the accounts and the writings of practitioners and their experiences in civil war peace processes. Perhaps we rely too heavily on the scholarly literature as we formulate theoretical models and build data sets. We can learn from practitioner experiences so that we can avoid some of the misunderstandings identified above. NOTES 1. Idean Salehyan and Kristian Skrede Gleditsch, “Refugees and the Spread of Civil War,” International Organization 60, no. 2 (2006): 335–366. 2. See the work of Roy Licklider, “The Consequences of Negotiated Settlements,” American Political Science Review 89, no. 3 (1995): 681–690. 3. Andrew Mack, “Civil War: Academic Research and the Policy Community,” Journal of Peace Research 39, no. 5 (2002): 515–525. 4. Karl DeRouen Jr., Jacob Bercovitch, and Paulina Pospieszna, “Introducing the Civil Wars Mediation (CWM) Dataset,” Journal of Peace Research 48, no. 5 (2011): 663–672. 5. DeRouen et al., “Introducing the Civil Wars Mediation (CWM) Dataset.” 6. Marie Olson Lounsbery and Alethia Cook, “Negotiating Your Way Out of Rivalry: Unassisted Conflict Resolution Efforts in Southeast Asia,” International Negotiation Journal 20, no. 2 (2015): 242–268. 7. See, for example, Marie Olson Lounsbery and Karl DeRouen Jr., “The Viability of Civil War Peace Agreements,” Civil Wars 18, no. 3 (2016): 311–337; Patrick M. Regan and Aysegul Aydin, “Diplomacy and Other Forms of Intervention in Civil Wars,” Journal of Conflict Resolution 50, no. 5 (2006): 736–756. 8. Karl DeRouen Jr. and Frida Moller, “The Short Effects of Mediation on Low-Intensity Civil Wars,” Negotiation Journal 29, no. 4 (2013): 413–438; Scott Signmund Gartner, “Signs of Trouble: Regional Organization Mediation and Civil War Agreement Durability,” The Journal of Politics 73, no. 2 (2011): 380–390; Tobias Bohmelt, “The Effectiveness of Tracks of Diplomacy Strategies in ThirdParty Interventions,” Journal of Peace Research 47, no. 2 (2010): 167–178. 9. Paul Wehr and John Paul Lederach, “Mediating in Central America,” Journal of Peace Research 28, no. 1 (1991): 85–98; Isak Svensson, “Who Brings Which Peace? Neutral versus Biased Mediation and Institutional Peace Arrangements,” Journal of Conflict Resolution 53, no. 3 (2009): 446–469. 10. Antje Herrberg, “Effective Mediation Diplomacy and Its Seven Virtues: The Case of the European Union,” Peace & Change 43, no. 3 (2018): 292–317; Omer Shapira, “Exploring the Concept of Power in Mediation: Mediators’ Sources of Power and Influence Tactics,” Ohio State Journal on Dispute Resolution 24, no. 3 (2009): 535–569. 11. Stephen John Stedman, “Spoiler Problems in Peace Processes,” International Security 22, no. 2 (1997): 5–53. 12. Marie Olson Lounsbery and Alethia Cook, “Rebellion, Mediation and Group Splintering: An Empirical Investigation of Competing Hypotheses,” Journal of Peace Research 48, no. 1 (2011): 73–84. 366 PEACE & CHANGE / July 2018 13. Desiree Nilsson, “Partial Peace: Rebel Groups Inside and Outside of Civil War Settlements,” Journal of Peace Research 45, No. 4 (2008): 479–495; Marie Olson Lounsbery, “Examining the Success and Failures of Less Inclusive Peace Agreements,” unpublished manuscript. 14. David E. Cunningham, “Veto Players and Civil War Duration,” American Journal of Political Science 50, no. 4 (2006): 875–892. 15. Olson Lounsbery, “Examining the Success and Failures of Less Inclusive Peace Agreements.” 16. J. Michael Grieg and Paul F. Diehl, International Mediation. (Cambridge, U.K.: Polity Press, 2012). 17. William I. Zartman, Ripe for Resolution (New York: Oxford University Press, 1986). 18. William I. Zartman, “Ripeness: The Hurting Stalemate and Beyond,” International Conflict Resolution After the Cold War, eds. Paul C. Stern and Daniel Druckman (Washington, D.C.: National Academy Press, 2000). 19. Stephen John Stedman, Peacemaking in Civil War (Boulder, Colorado: Lynne Rienner, 1991). 20. Zartman, Ripe for Resolution. 21. William I. Zartman and Alvaro de Soto, Peacemaker’s Toolkit: Timing Mediation Initiatives (Washington, D.C.: United States Institute of Peace, 2010). 22. Patrick Johnston, “Negotiated Settlements and Government Strategy in Civil War: Evidence from Darfur,” Civil Wars 9, no. 4 (2007): 359–377. 23. Matthew Hoddie and Caroline Hartzell, “Civil War Settlements and the Implementation of Power Sharing Arrangements,” Journal of Peace Research 40, no. 3 (2003): 303–320; Matthew Hoddie and Caroline Hartzell, “Power Sharing in Peace Settlements: Initiating the Transition from Civil War,” in Sustainable Peace: Power and Democracy After Civil Wars, eds. Philip G. Roeder and Donald S. Rothchild (Ithaca, New York: Cornell University Press, 2001); Anna K. Jarstad and Desiree Nilsson, “From Words to Deeds: The Implementation of Power Sharing Pacts in Peace Accords,” Conflict Management and Peace Science 25, no. 3 (2008): 206–223; Karl DeRouen Jr., Jenna Lea, and Peter Wallensteen, “The Duration of Civil War Peace Agreements,” Conflict Management and Peace Science 26, no. 4 (2009): 367–387. 24. Timothy Sisk, Power Sharing and International Mediation in Ethnic Conflicts (Washington, D.C.: United States Institute of Peace, 1996). 25. Caroline Hartzell and Matthew Hoddie, “Institutionalizing Peace: Power Sharing and Post-Civil War Conflict Management,” American Journal of Political Science 47, no. 2 (2003): 318–332. 26. Frederic S. Pearson, Marie Olson Lounsbery, Scott Walker, and Sonja Mann, “Rethinking Models of Civil War Settlement,” International Interactions 32, no. 2 (2007): 109–128. 27. Barbara Walter, “A Critical Barrier to Civil War Settlement,” International Organization 51, no. 3 (1997): 335–364. 28. Christine Bell, Peace Agreements and Human Rights (New York: Oxford University Press, 2000). 29. Madhav Joshi and J. Michael Quinn, “Implementing the Peace: The Aggregate Implementation of Comprehensive Peace Agreements and Peace Duration Civil War Conflict Resolution 367 after Intrastate Armed Conflict,” British Journal of Political Science 47, no. 4 (2017): 869–892; Madhav Joshi and J. Michael Quinn, “Is the Sum Greater than the Parts? The Terms of Civil War Peace Agreements and the Commitment Problem Revisited,” Negotiation Journal 31, no. 1 (2015): 7–30; Matthew Hoddie and Caroline Hartzell, “Civil War Settlements and the Implementation of Military Power Sharing Arrangements.” 30. Dawn Brancati and Jack L. Snyder, “Time to Kill: The Impact of Election Timing on Postconflict Stability,” Journal of Conflict Resolution 57, no. 5 (2013): 822–853; Thad Dunning, Dawn Brancati, and Jack L. Snyder, “Rushing to the Polls: The Causes of Premature Postconflict Elections,” Journal of Conflict Resolution 55, no. 3 (2011): 469–492. 31. Karl DeRouen Jr., Mark J. Ferguson, Samuel Norton, Young Hwan Park, and Jenna Lea,“Civil War Peace Agreement Implementation and State Capacity,” Journal of Peace Research 47, no. 3 (2010): 333–346. 32. Sean Byrne, Chuck Thiessen, and Eyob Fissuh, “Economic Assistance and Peace Building in Northern Ireland,” Peace Research 39, no. 1/2 (2007): 7–22; Helga Malmin Binningsbø and Siri Aas Rustad, “Sharing the Wealth: A Pathway to Peace or a Trail to Nowhere?” Conflict Management and Peace Science 29, no. 5 (2012): 547–566; Michael W. Doyle and Nicholas Sambanis, “International Peacebuilding: A Theoretical and Quantitative Analysis,” The American Political Science Review 94, no. 4 (2000): 779–801. 33. Kristin Eck and Lisa Hultman, “One-Sided Violence Against Civilians in War: Insights from New Fatality Data,” Journal of Peace Research 44, no. 2 (2007): 233–246. 34. Johan Galtung, “Twenty-Five Years of Peace Research: Ten Challenges and Some Responses,” Journal of Peace Research 22, no. 2 (1985): 141–158. 35. For example, Patrick Regan and Aysegul Aydin, “Diplomacy and Other Forms of Intervention in Civil Wars,” 736–756 at 754; Isak Svensson, “Who Brings Which Peace?” Neutral versus Biased Mediation and Institutional Peace Arrangements in Civil Wars,” in The Journal of Conflict Resolution, 53, no. 3 (2009): 446– 469 at 464–465. 36. Peter Wallensteen and Isak Svensson, “Talking Peace: International Mediation in Armed Conflicts,” Journal of Peace Research 51, no. 2 (2014): 315–327 at 324. 37. Wallensteen and Svensson, “Talking Peace.” 38. For a review of the quantitative literature on international mediation, see Wallensteen and Svensson, “Talking Peace.” 39. United Nations, United Nations Guidance for Effective Mediation (New York: United Nations, 2012); Kjell Skjelsbaek, “The UN Secretary-General and the Mediation of International Disputes,” Journal of Peace Research 28, no. 1 (1991): 99–115 at 105; Lakhdar Brahimi and Salman Ahmed, In Pursuit of Sustainable Peace: The Seven Deadly Sins of Mediation (New York: Center on International Cooperation, New York University, 2008), 9–10; Johannes Botes and Chris Mitchell, “Constraints on Third Party Flexibility,” Annals of the American Academy of Political and Social Science 542 (1995): 168–184 at 171. 40. United Nations, United Nations Guidance, 4. 368 PEACE & CHANGE / July 2018 41. Laurie Nathan is a Senior Mediation Advisor to the UN, a member of the UN Academic Advisory Council on Mediation, and the lead designer and trainer of the UN High Level Mediation Course. He has served as a mediation adviser to the African Union, the Economic Community of West African States, the Intergovernmental Authority for Development, and the Southern African Development Community. 42. For example, “Briefing Note on Conflict Analysis and Ripeness for Mediation,” 2017, presented at United Nations High Level Mediation Course. 43. Quoted in Laurie Nathan, “The Intelligence Requirement of International Mediation,” Intelligence and National Security 29, no. 2 (2014): 208–226 at 218–219. 44. See, for example, Lazaro Sumbeiywo, To Be a Negotiator: Strategies and Tactics (Zurich, Switzerland: Centre for Security Studies, ETH Zurich, and swisspeace, 2009), available at www.css.ethz.ch/box_feeder/negotiator.pdf. 45. Stephen Stedman, Peacemaking in Civil War: International Mediation in Zimbabwe, 1974–1980 (Boulder, Colorado: Lynne Rienner 1991), p. 224 46. Ibid., p. 225. 47. Jeffrey Davidow, A Peace in Southern Africa: The Lancaster House Conference on Rhodesia, 1979 (Boulder, Colorado: Westview 1984). 48. United Nations Department of Political Affairs, Guidance on Gender and Inclusive Mediation Strategies (New York, United Nations, 2017). 49. Darfur Peace Agreement, 2006; Laurie Nathan, “No Ownership, No Peace: The Darfur Peace Agreement” (Working Paper no. 5, Crisis States Research Centre, London School of Economics, 2006). 50. United Nations, “United Nations Activities in Support of Mediation. Report of the Secretary-General” (UN doc. A/72/115, 2017), 18–19. 51. United Nations Department of Political Affairs, “2018 Standby Team of Senior Mediation Advisers,” available at https://peacemaker.un.org/sites/peacemake r.un.org/files/Factsheet_2018_Standby_Team_English_0.pdf. 52. UN Department of Political Affairs and UN Environment Program, National Resources and Conflict: A Guide for Mediation Practitioners (New York: UN DPA and Nairobi: UNEP, 2015). See also United Nations Development Program, Supporting Insider Mediation: Strengthening Resilience to Conflict and Turbulence (New York: UNDP, 2014). 53. For example, United Nations, United Nations Guidance; Organization for Security and Co-operation in Europe, Mediation and Dialogue Facilitation in the OSCE (Vienna: OSCE, 2014); African Union and African Centre for the Constructive Resolution of Disputes, African Union Mediation Support Handbook (Durban: African Union and ACCORD, 2014); Connie Peck, A Manual for UN Mediators: Advice from UN Representatives and Envoys (Geneva, Switzerland: UN Institute for Training and Research and UN Department of Political Affairs, 2010); United Nations Department of Political Affairs, Mediation Start-up Guidelines (New York: United Nations, 2011); United Nations, “Handouts and Briefing Notes for the UN High Level Mediation Course” (2017); Brahimi and Ahmed, In Pursuit of Sustainable Peace. 54. See the information provided on the UN Peacemaker website, at https://peacemaker.un.org/mediation-support/featured-projects/ceasefires-project. Civil War Conflict Resolution 369 55. For example, UN Department of Political Affairs, Guidance for Mediators: Addressing Conflict-Related Sexual Violence in Ceasefire and Peace Agreements” (New York: United Nations, 2012). 56. UN Department of Political Affairs and UN Environment Program, National Resources and Conflict. 57. Monica Duffy Toft, “Ending Civil Wars: A Case for Rebel Victory?” International Security 34, no. 4 (2010): 7–36. 58. Monica Duffy Toft, “Data Appendix for Monica Duffy Toft”, Securing the Peace: The Durable Settlement of Civil Wars (Princeton, New Jersey: Princeton University Press, 2010), available at http://belfercenter.ksg.harvard.edu/files/MTc odebook2010.pdf, accessed February 11, 2011, pp. 2–4. 59. Toft, “Ending Civil Wars,” 36. 60. Nicholas Kristof, “Genocide in Slow Motion,” New York Review of Books 53, no. 2 (February 2006): 14–17. 61. For a more detailed critique of Toft’s article, see Laurie Nathan, “Correspondence. A Case for Rebel Victories? Reviewing the Evidence and Method,” International Security 36, no. 1 (2011): 202–210. 62. For example, Jacob Bercovitch and Scott Gartner, “Is There Method in the Madness of Mediation? Some Lessons for Mediators from Quantitative Studies of Mediation,” International Interactions 32, no. 4 (2006): 329–354. 63. On different approaches adopted by U.S. mediators, see Daniel Curran, James Sebenius, and Michael Watkins, “Two Paths to Peace: Contrasting George Mitchell in Northern Ireland with Richard Holbrooke in Bosnia-Herzegovina,” Negotiation Journal 20, no. 4 (2004): 513–537. 64. Patrick Regan and Aysegul Aydin, “Diplomacy and Other Forms of Intervention in Civil Wars,” Journal of Conflict Resolution 50, no. 5 (2006): 736–756. 65. Ibid., 736–737. 66. Ibid., 747. The authors do not provide an explanation of the term “international forums.” 67. Ibid., 748. 68. Ibid., 741. 69. For example, Robert Rauchhaus, “Information, Mediation, and Conflict Management,” World Politics 58, no. 2 (2006): 207–241; Burcu Savun, “Information, Bias, and Mediation Success,” International Studies Quarterly 52, no. 1 (2008): 25–47; Andrew Kydd, “Which Side Are You on? Bias, Credibility, and Mediation,” American Journal of Political Science 47, no. 4 (2003): 597–611. 70. Nathan, “The Intelligence Requirement of International Mediation.” 71. For example, Jacob Bercovitch and Allison Houston, “Influence of Mediator Characteristics and Behavior on the Success of Mediation in International Relations,” International Journal of Conflict Management 4, no. 4 (1993): 297–321; Isak Svensson, “Mediation with Muscles or Minds? Exploring Power Mediators and Pure Mediators in Civil Wars,” International Negotiation 12, no. 2 (2007): 229–248. 72. For example, David Mason, Joseph Weingarten Jr., and Patrick Fett, “Win, Lose, or Draw: Predicting the Outcome of Civil Wars,” Political Research Quarterly 52, no. 2 (1999): 239–268; I. William Zartman, “The Timing of Peace 370 PEACE & CHANGE / July 2018 Initiatives: Hurting Stalemates and Ripe Moments,” Global Review of Ethnopolitics 1, no. 1 (2001): 8–18. 73. Mack, “Civil War: Academic Research and the Policy Community.” 74. For a comparative case study approach to this work, see Alethia Cook and Marie Olson Lounsbery, Conflict Dynamics: Civil Wars, Armed Actors, and Their Tactics (Athens, Georgia: University of Georgia Press, 2016). For an example of how a quantitative data set uses this approach, see David E. Cunningham, Kristian Skrede Gleditsch, and Idean Salehyan, “It Takes Two: A Dyadic Analysis of Civil War Duration and Outcome,” Journal of Conflict Resolution 53, no. 4 (2009): 570–597. 75. Carsten Anckar, “On the Applicability of the Most Similar Systems Design and the Most Different Systems Design in Comparative Research,” International Journal of Social Research Methodology 11, no. 5 (2008): 389–401. 76. See the Journal of Peace Research’s special issue (53, no. 4) on Disaggregating Civil War with an introduction by Lars-Erik Cederman and Kristian Skrede Gleditsch. 77. Cunningham et al., “It Takes Two.” 78. Mack, “Civil War: Academic Research and the Policy Community.”