Who Mediates?: An Analysis of the Willingness to Offer
and Accept Mediation in Civil Wars
J. Michael Greig
Department of Political Science
University of North Texas
and
Patrick M. Regan
Department of Political Science
Binghamton University
Prepared for presentation at the annual meetings of the American Political Science Association, Sept, 2006, Philadelphia, PA
Introduction
Civil wars present an increasingly complex puzzle for the policy community today. It is not that these wars are new, but with the absence of serious, long running, interstate conflict, civil wars have become the destabilizing force in world politics. Over the past 50 years there have been on average about 20 ongoing civil wars at any one time, some of which are at risk of spilling over national boundaries, some result in massive refugee flows, and some result in complex humanitarian crises on the order of magnitude that gets them labeled genocide. In spite of the potentially grave consequences and relatively high frequency of civil wars, we know remarkably little about how best to manage them.
Today the civil war in Iraq poses particularly thorny problems for the United States. By almost all conditions scholars consider the violence in Iraq to be consistent with a civil war, but the political liabilities associated with calling it a civil war are debilitating. In effect, the United States doesn’t know what to do once it recognizes that Iraq is a civil war; the scholarly community has yet to provide much guidance.
The options that appear to be available to the United State in Iraq – or the world community in Darfur, Sudan – involve walking away and doing nothing, trying to use military and economic leverage, or engaging the warring side through mediation. One of the political problems faced by the US in Iraq is that if they acknowledge that it is a civil war, the public will increasingly demand the do nothing strategy, that is, they fear a swelling of demands to pull out. The military and economic approaches do not seem to be effective, which leaves the diplomatic channels as possibly the best hope.
Recognizing that diplomatic interventions might play a useful role, however, is a far cry from understanding the conditions under which the warring sides might accept an outside mediator. This is not a puzzle that is easily resolved. The decision to accept outside mediation, and in effect to attempt to resolve the conflict through a negotiated settlement, is a function of sunk costs, expected benefits, and estimates of the likelihood of victory. Moreover, to get to the negotiations both sides to the conflict must come to the conclusion that there are conditions under which a negotiated outcome is preferable to continued fighting. A mediator may be a necessary condition in civil wars.
In this paper we attempt to answer the question of under what conditions the warring parties will accept an offer by an outside party to mediate. Before success or failure of mediation or negotiation can be understood we have to consider the conditions that make them possible. One way to think about the acceptance of mediation is that both sides implicitly agree to make concessions, or at least that they have some concessions to give under certain conditions. It is only at this point that a negotiated outcome is possible.
We begin with a brief review of contemporary literature, followed by the development of a theoretical argument that describes the conditions under we would expect mediation offers to be accepted. We follow this with a discussion of the research design we use to test hypotheses derived from our theoretical framework, and finally with a discussion of our results. Our analysis suggests that mediation is more likely to be accepted when the costs to the government for continued fighting is high and when the political unit offering to mediate has had a prior history of mediating that particular conflict. Moreover, we find two situations that appear to approximate sufficient conditions for the acceptance of mediation: a prior colonial history with
the country making the offer, and an offer to mediate from a multiparty coalition.
The Acceptance of Mediation
An offer of mediation from a third-party provides an opportunity for the parties in a civil war to reconsider the means by which they pursue their goals, with the potential to move them away from the continued use of violence. Yet, despite this opportunity for a change in the conflict status quo between combatants, third-party offers of mediation are often rejected by one or more of the parties in conflict. At first glance, a third-party offer of mediation would seem to be an opportunity readily seized upon by civil war belligerents, providing them with a chance to forego the continued costs and pain of fighting in the hopes of producing a settlement that makes both parties better off than they would be from continued violence. The push toward the acceptance of mediation would appear to be all the more powerful given that a cease-fire between the parties is not necessarily a prerequisite for the acceptance of mediation. Thus, if civil war parties do not have to give up violence before accepting mediation, why do they say “no” to mediation in the first place?
A key impediment to the acceptance of mediation in civil wars is that it carries with it significant political costs for both rebel and government leaders. Even the consideration of a shift away from violence toward dialogue by the leaders of civil war participants risks their labeling as appeasers by their constituents, threatening their political survival (Specter, 1998). This risk is particularly pronounced in civil wars due to their tendency to occur across religious and ethnic lines, divisions that lend themselves quite easily to painting the opposing side as not only wrong, but evil. Furthermore, mediation, by involving a third-party in talks between disputants, invariably serves to impose constraints upon the actions of the belligerents that makes leaving negotiations more difficult or pushes them toward a settlement that they might not otherwise prefer. In this sense, if the acceptance of mediation entails political costs for civil war parties, then we should expect that they will be judicious in their acceptance of it, only doing so when they expect to achieve a better result with mediation than without it (Skjelbaek, 1991; Mitchell, 1981; Bercovitch and Jackson, 2001).
Fundamental to the identification of the point at which parties in conflict become receptive to offers of mediation is recognition that conflict exerts a dual effect upon its participants. On one hand, fighting serves to embitter combatants, making it more difficult for them to move toward negotiations (Nathan, 1999; Pearson, 2001; Skjelsaek, 2001; Pruitt and Olczak, 1995; McGinnis and Williams, 2000). Yet, at the same time, conflict also provides valuable information to disputants about both the costs of conflict and their ability to win it (Fearon, 1995; Wagner, 2000; Goemans, 2003; Reiter, 2003). As a result, conflict between disputants provides information that is instrumental in disputants’ ability to update their beliefs about the relative merits between fighting and talking (Powell 1990, 2004; Filson and Werner, 2002; Slantchev, 2002; Reiter, 2003).
In this sense, key to the acceptance of mediation is the timing of the offer to a point in a conflict at which combatants begin to focus upon the costs and pain of the conflict rather then the hostility it imparts upon them. Offers of mediation that occur too soon, before the parties have had the opportunity to experience significant conflict costs between them, are likely to be spurned as each side continues to believe that it can achieve a better outcome through violence than through mediation. Even at a middle point in the conflict, when the parties have begun experiencing high conflict costs, they may still resist offers of third-party diplomacy. At this middle stage of conflict, not only may the combatants continue to believe that they retain the ability to impose their preferred settlement upon the other side, but their continue reliance upon violence as their preferred strategy in the pursuit of their objectives tends to be further driven by the self-reinforcing nature of conflict. The pain and hostility that conflict imparts upon its participants perpetuates it, not only be encouraging the disputants to see the imposition of pain upon the other side as an objective in and of itself, but also by encouraging them to think of previous violence as sunk costs. When these conditions prevail, disputants are unlikely to see mediation as a better means of achieving their goals than violence. Only when they reach a point when they begin to question the continued viability of violence as the preferred strategy in the pursuit of their goals do disputants become more likely to consider other less conflictual means, such as mediation, of achieving their objectives.
This willingness to accept mediation should be particularly pronounced as a hurting stalemate develops (Walter, 2002; Zartman 1985; 2000). A hurting stalemate exists when disputants ultimately reach a point in their conflict such that they begin to experience the costs and pain of conflict with little real hope of achieving their goals through force. That is, costs accumulate while the prospects for meaningful benefits from fighting diminish. Thus, when a hurting stalemate is present, disputing parties are most likely to begin searching for new solutions that provide a way out of the conflict (Young, 1967; Holbrooke, 1998; Regan and Stam, 2000; Greig, 2001, 2005). In this sense, movement toward mediation becomes more likely as an impasse develops between the parties such that victory becomes more illusory and bilateral efforts at conflict management have proven unsuccessful (Bercovitch and Derouene, 2005).
Without the presence of a hurting stalemate, the task of a third-party seeking entry to a conflict as a mediator is doubly hard. While the hostility that conflict fosters between parties makes the initiation of mediation problematic for any disputing parties, acceptance of mediation is particularly difficult for governments involved in civil wars. Even if a settlement cannot be obtained, participation in mediation by a rebel group challenging a government at least has the benefit of conferring a sense of legitimacy to the rebel group both domestically and internationally. Yet, this very feature of mediation provides a powerful disincentive for governments to accept mediation with rebels. Israel, for example, has resisted negotiation and mediation with the PLO, and now Hamas, precisely because it feared that doing so would represent a de facto recognition of the organization and an acceptance of the legitimacy of its demands (Watkins and Lundberg, 1998). In this sense, governments face particularly strong incentives to seek to marginalize groups challenging their power by refusing to negotiate with the group in the hopes that, with force, rebel groups and their demands can be defeated.
As a result, rebel groups face a key challenge in getting to the bargaining table; they must achieve a sufficient level of military and/or political credibility that forces the government to accept them as a viable actor to be negotiated with and not eliminated, ignored, or marginalized. In most cases, however, the only way for a rebel group to achieve this credibility is by demonstrating through force sufficient capacity to challenge the government and its interests that it makes it difficult for the government to marginalize the rebel group. In effect, rebel groups must use violence as a bargaining tool to bring the government to the negotiating table. The rebel group need not have sufficient power to actually win a conflict with the government, but must demonstrate enough capacity to inflict sufficient costs to show at least the potential to produce a stalemate with the government. By doing so, the rebel group can impose enough costs upon the government that it forces the regime to not only overcome its initial unwillingness to grant recognition and legitimacy to the group by beginning dialogue, but actually begins to consider making concessions to the group.
Call’s (2002) discussion of the civil war in El Salvador in the 1980s provides an example of the way in which the capacity of a rebel group to establish credibility for itself coupled with high conflict costs and inability of either side to prevail can encourage a government to accept mediation with the group. In November 1989 the FMLN launched a series of coordinated attacks across El Salvador, attacking numerous key government military facilities. At the same time, the FMLN also brought 2000 combatants into the capital, demonstrating a capacity to mount a significant and widespread challenge to the government. In the end, while the attacks were subsequently turned back by the government, highlighting the inability of the rebels to win a decisive victory, the attacks served to convince key officials within both the government and the country’s economic elite of the scope of the challenge from the FMLN and the economic costs it could continue to impose. As a result, sentiment within both the government and the FMLN moved toward dialogue such that UN-sponsored mediation began in April 1990. In this respect, by imposing unacceptable costs upon the government, the FMLN was able to push it toward negotiations.
Stanley and Holliday’s (2002) discussion of the transition to mediation in the Guatemalan civil war is consistent with this argument. Stalemate on the battlefield functioned to convince the rebels of the URNG that victory was impossible and, ultimately, that negotiation was preferable to conflict. At the same time, the civilian-led Guatemalan government was also motivated to accept mediation as a means of ending the costs the conflict was imposing upon the country. In effect, an interaction between costs and prospects for victory results in an incentive structure conducive to negotiation.