WHAT DO MEDIATORS DO WHEN THEY MEDIATE?

What is it that mediators do when they intervene in a conflict? Like many questions about mediation, the answer to this one is far from simple or obvious. We must clarify what we mean by mediation behavior, and how best to interpret it.

Alternatively, mediator activities were organized conceptually to describe mediator behavior in terms of various preordained roles and tactics (Gulliver 1979) or phases (Folberg and Taylor 1984; Mitchell 1981; Moore 1986) that govern mediator intervention behavior. While these may be interesting classifications, they bring us no closer to understanding the underlying dynamics of the mediation process and the reality of the changing nature of a conflict (Bercovitch 1992, 103).

In essence, the practice of mediation revolves around the choice of strategic behaviors that mediators believe will facilitate the type of outcome they seek to achieve in the conflict management process. A mediator may be less reactive and more practical and systematic in his/her behavior than previously thought. That is, mediators may be seen as skilled practitioners of a learned craft. A mediator’s behavior is dependent on the perceived role or purpose, and the resources and the techniques available to him/her within the specific dispute context. Mediation behavior can thus be understood as an overall plan or approach to conflict management to achieve a specific end: the settlement of the dispute, the stopping of violence and destruction, or the overall resolution of the conflict.

As such, mediation is not an “art” that is highly idiosyncratic, based on intuitive insights, and resistant to systematic analysis (Meyer 1960); it is rather a coherent and planned activity. Consequently, it is possible to explain and understand a mediator’s behavior in terms of the identification and conceptualization of various roles, tactics, processes, and strategies which can be exercised in the practice of mediation.

Mediation strategies

The most useful way of describing and inter­preting mediator behavior is to conceptualize their activities in terms of broad strategies. While the analysis of the roles and stages of mediator behavior provide perfectly valid and feasible explanations of single cases, the categorization of mediation behavior into broad strategies is the most practical and useful option when studying a large number of conflicts. This approach provides a simple yet logical structure within which the extensive inventory of mediator behavior can be organized and understood.

For our purposes, the most useful taxonomy of mediator behavior that can be applied to international mediation analysis is based on the identification of three fundamental mediator strategies along a continuum ranging from low to high intervention. These are: (a) communication-facilitation, (b) procedu­ral, and (c) directive strategies (discussions of these can be found in Bercovitch 1992; Bercovitch and Wells 1993; and Bercovitch et al. 1991). These strategies are based on assumptions derived from Sheppard’s (1984) taxonomy of mediator behavior that focuses on the content, process and procedural aspects of conflict management.

(1) Communication-facilitation strategies describe mediator behavior at the low end of the inter­vention spectrum. Here, a mediator typically adopts a fairly passive role, channeling infor­mation to the parties, facilitating cooperation but exhibiting little control over the more formal process or substance of mediation. Norway's mediation role in the Oslo agreement between Israel and the PLO of 1993 exemplifies this approach.

(2) Procedural strategies enable a mediator to exert a more formal control over the mediation process with respect to the environment of the mediation. Here a mediator may determine structural aspects of the meetings, control constituency influences, media publicity, the distribution of information, and the situation powers of the parties' resources and commu­nication processes. New Zealand's efforts in the Bougainville conflict in 1995, where it brought both parties to a military camp in New Zealand, exemplify this form of mediation.

(3) Directive strategies are the most powerful form of intervention. Here, a mediator affects the content and substance of the bargaining process by providing incentives for the par­ties to negotiate or by issuing ultimatums. Directive strategies deal directly with and aim to change the way issues are framed, and the behavior associated with them. Richard Holbrooke's efforts at Dayton are typical of this approach.

Although mediators have a wide array of tactical choices at their disposal, there is no suggestion here that they may use any of the strategies they wish with its associated tactics in any conflict they intervene. Clearly, there are some conflicts that will show greater amenability to some forms of mediation behavior, and of course there will be mediators who will feel more comfortable with, or have the resources and determination to implement, one strategy rather than another. Analyzing which strategies and which tactics work in which conflicts has been a dominant, if inconclusive, theme of mediation research (for an interesting example of this research, see Beardsley et al. 2005).

Can we, in any way, link strategies to outcomes? Few studies attempt to assess the effectiveness of different strategies. Those that do so have found that the strategies at each end of the intervention spectrum appear to dominate actual mediator intervention in international conflicts (Bercovitch and Houston 1996). Further analyses of medi­ation revealed that while communication­facilitation strategies are the most frequently utilized by international mediators, directive strategies appear to be the most successful (e.g.

The choice of a strategy in any situation is clearly affected, inter alia, by the nature of the relationship between the parties, the context of the conflict, and their historical experience. Mediators adapt their style of intervention to meet the requirements of the situation, and we think that certain styles or strategies of mediation will be generally more effective in certain situations. An intense conflict with high fatalities may require more intense interventions than a low-level conflict (see Hiltrop 1985, 1989; Rubin 1980). The costs of no agreement in the former are dangerously high. If a mediator is involved in such a conflict, they will use any stick or carrot at their disposal to nudge the parties toward a zone of agreement. However, given the entrenched and intense nature of the conflict, it is more than likely that the most that can be achieved is a partial cessation of violence. In a low-intensity conflict, disputants are likely to view those same sticks and carrots as overbearing and too directive in nature - making them less effective, despite the likely lower stakes involved.

Factors affecting the choice of a strategy

A number of factors affect the choice of a mediation strategy and its potential for success. Amongst the most important factors are the following:

(1) The intensity of a conflict is recognized as a major factor affecting the nature of conflict management, and any evolving pattern of mediation. But how exactly does the intensity of a conflict influence the implementation of a particular mediation strategy? Conflict intensity usually refers to such factors as the severity of conflict, the level of hostilities, the number of fatalities, the level of anger and intensity of feeling, the types of issues at stake, and the strength of the parties' negative perceptions (Kressel and Pruitt 1989). When conflict intensity is low, Rubin (1980, 389) suggests that the parties are concerned with “mending their own fences" and do not want third-party intrusion.

(2) The type of issues in conflict may be examined to identify the “essence" of a conflict. Intuitively, we would expect this to be an influential factor in a mediator's choice of strategy, as issues represent the focus of what separates the parties, and what the conflict is all about. But what exactly are issues in conflict, and how can we identify them and conceptualize their presence? Conflicts can, in the first instance, be internal or interstate. When they are internal, they are often focused on issues such as identity, autonomy, and ethnicity. These are subjective and emotional issues often including fear, resentment, and distrust that are hard to negotiate over, and harder still to mediate. The best that a mediator may do in such contexts is to resort to communication strategies that build the confidence and trust of the parties, and give them an incentive to pursue peaceful conflict management. In contrast, conflicts over issues such as security, resources, and defense, involve more concrete issues that are easier to work with. Here, a mediator may press for concessions on tangible issues to achieve a conflict termination or resolution. Each type of issue in conflict elicits a different form of mediator behavior.

(3) The internal characteristics of the parties enable us to examine how each party's political and economic structure affects the process of conflict management and mediation. Parties with similar political systems or social structures (ethnic, cultural, or religious groups organizing society) may be more amenable to serious, active mediation.

(4) The previous relationship and experience of the parties can be examined to gauge how past experiences of conflict and conflict management affect current behavior and determine choice of mediation strategy. Any social relationship is affected by previous experiences between the same parties. Similarly, any current conflict management is affected by previous conflict management efforts and any learning which may have taken place (on the role of learning in conflict management, see Leng 2000). The past does indeed cast a shadow on the present. Repeated mediation efforts by the same mediator may establish some norms of interaction and to a large extent determine what each party may expect and how it should behave. In an environment of risk and uncertainty, mediators may use information from previous efforts, or build on any rapport they may have had with the parties. Here, I want to suggest that previous conflict experience and mediation may exert a strong influence on the choice of a current strategy. Previous mediation efforts can establish norms and a certain rapport between the parties, and these can affect their current disposition and behavior. There is an element of reinforcement and learning occasioned by previous experience of mediation which influences how mediation is currently conducted. Past conflict management behavior is a pretty good indicator of current and future behavior. Although conflict management is not a linear process, there is certainly an element of learning at work here, and this element affects the choice of a mediation strategy.

(5) Mediator identity and rank describe the official position of a mediator. These will clearly affect the choice of a strategy. At the most basic level, some mediators have the potential to utilize resources, use leverage and influence; others can rely only on their legitimacy or reputation. Who the mediator is determines to a large extent what a mediator can do. Mediators' use of a strategy is not random; it is the result of many complex factors. One of these factors relates to the official position and status of different mediators. Some have the full range of resources, and thus the full range of strategies available to them. Others (individual mediators, NGOs) can only use communication strategies as they simply do not have access to expensive resources. Who a mediator is determines what a mediator can do, and which strategies are used.

(6) The initiation and timing of mediation inter­vention. While mediation is ultimately a voluntary process, it may be initiated, that is, suggested, appealed for, or offered, by either the disputants, the mediator, or various other concerned parties. The perceived need or justification for mediation is influenced by the disputants' and mediator understanding of what the role of the mediator should be in managing the conflict. The mandate of the intervener and the legitimacy and authority of mediator behavior are to some extent determined by who initiates the mediation process (Kaufman and Duncan 1992), and the timing of intervention in terms of the conflict phases and the state of the parties' current negotiations.

These factors may determine the acceptability of a specific mediator, and the role, bounds, and expectations within which a mediator may act to manage the conflict, and the type of strategies employed (Kolb 1983; Raiffa 1982). The initiation of mediation, and choice and acceptance of a mediator, are largely dependent on the parties', the mediator's, or other third parties' perceptions of the resources and skills a mediator may have on offer, their expectations of how the outcome of the dispute may be influenced by the intervention of a specific mediator, and the level of commitment and urgency of the parties to achieving a settlement.

(7) The mediation environment. An important dimension which may influence mediation behavior and choice of strategies is the mediation environment. The choice of media­tion environment may be determined by the demands of the parties, their powers, resources and goals, and their willingness to negotiate, by the extent of constituency and media pressures, or it may be the product of a mediator's strategy to control a particular conflict situation. In turn, the specific environment in which mediation takes place may determine the type of behavior a mediator employs. As such, the mediation environment, with the various opportunities and constraints that it provides, may be a powerful factor in understanding the dynamics of mediation behavior. An ideal mediation environment will support rather than hinder parties' conflict management efforts and interactions, and provide the mediator with opportunities to manage and control the whole process (Touval 1982).

The structure imposed on mediation by the environment provides opportunities for both parties and the mediator to be empowered and manage their conflict competently and productively, and to avoid or mend any dysfunctional behavior that may regress the parties’ mediation efforts. The physical context of the mediation event establishes the bounds that dictate, and perhaps constrain, the ability of the parties and the mediator to express their status, authority, power, leverage, and assertiveness within mediation and how their efforts are represented to the external constituencies, media, and interna­tional audiences. A mediation environment may also determine the situational powers of the participants, their proximity, and social interactions. Clearly, a party’s legitimacy, standing, and integrity are integral character­istics that must be protected and maintained if mediation is to be successful (Rubin and Brown 1975) and are dependent on the nature and urgency of the dispute being managed.

Mediation behavior and choice of strategies cannot be foreordained, nor can these be prescribed in advance. They are part of the overall structure of a mediation event and context. Mediators choose strategies that are available, feasible, permissible, and likely to achieve a desired outcome. Mediation behavior is adaptable; it reflects, to a large extent, the context in which it takes place. I have highlighted some of the important contextual dimensions which may have an impact on mediation behavior and outcomes. We ignore these dimensions at our peril.