Abstract
Despite the upsurge in the literature on third party intervention in recent years, particularly that relating to mediation, there is still a significant gap in the field. While some theoretical accounts content themselves with describing the qualities of an ideal mediator, other studies borrow examples from a variety of case studies to emphasise the wide range of the mediator's functions or the tactics it can use. What is missing is a systematic, case study-driven analysis that draws on the theoretical literature while generating some fresh propositions about the conditions that are propitious for successful mediation. The article aims to achieve this by proposing three hypotheses about the impact of power, impartiality and timing on the mediation process: (1) The more power (leverage) the mediator has over the disputants, the more likely it is to succeed. (2) The more impartial the mediator is, the more likely it is to succeed. (3) Mediation is more likely to succeed when the conflict has reached an escalatory stage. These hypotheses will be tested against the experience of four successful cases of third party intervention in the Arab—Israeli conflict, which over the past six decades has experienced a multitude of mediators of different types.
While literature on mediation has flourished in recent years, some significant problems remain. First, theories of mediation tend to be overly generic and self-evident; second, mediation research is often insufficiently driven by systematic case study analysis, thus reducing the treatment of cases to descriptive footnotes. These shortcomings result in stale theoretical discussions about the conditions for mediation success or the most propitious mediation strategies. Although mediation literature goes to some lengths to describe the various characteristics of an ideal mediator and the desirable functions and tactics, seldom is it followed by a rigorous examination of empirical examples.
The Middle East provides a particularly propitious environment within which existing theories can be tested. The Arab—Israeli conflict, in all its forms, is perhaps the most intricate and protracted conflict in the twentieth century, and as such has produced numerous case studies of third party mediation. The diversity of the mediators and their strategies over the years, as well as the changing circumstances and the nature of relationship between the parties, deserve a systematic analysis which could offer some broader conclusions about the desired conditions for effective mediation.
The aim of this article is to examine three hypotheses about the relationship between mediator characteristics and conflict attributes to chances of success in mediation. These concern the power and impartiality of the mediator, and the timing of the mediation effort. They will be tested against the experience of various intermediaries in the Arab—Israeli conflict over more than five decades.
At least with regard to the Arab—Israeli conflict the three hypotheses examined here are as follows.
Hypothesis 1: The more power (leverage) the mediator has over the disputants, the more likely it is to succeed.
Hypothesis 2: The more impartial the mediator is, the more likely it is to succeed.
Hypothesis 3: Mediation is more likely to succeed when the conflict has reached an escalatory stage.
Since the first Arab—Israeli War in 1948, powerful as well as impartial mediators have entered the conflict at various stages with the aim of reaching a settlement. As will be seen later, the vast majority of them have failed, despite the myriad of strategies and tactics employed. Accordingly, it is fairly clear that these conditions alone are not sufficient. This essay therefore poses the question – to what extent are they necessary?
Definitions of mediation vary, yet most include the notion of a process undertaken by a third party to promote a peaceful conclusion of hostilities. Thomas Princen's definition (1992) commands wide acceptance and will suffice for this discussion. Mediators, according to Princen, are ‘third parties who intercede for the purpose of influencing or facilitating the settlement of a dispute but who do not impose a solution. They are actors with incentives to be involved but without direct interests in the disputed issues’ (Princen, 1992, p. 3). Mediation is a distinct form of third party intervention. It is initiated and performed on a voluntary basis; it is non-violent; and its proposals or recommendations are non-binding. This stands in contrast to other types of third party intervention, such as arbitration and adjudication (Bercovitch, 1997; Zartman and Touval, 1996).
Power, Impartiality and Timing in Mediation
The empirical and conceptual dimensions of mediation success cannot be merely asserted. The intricacy of the issues, the human element and the variety and diversity of the case studies prevent the formulation of an agreed-upon definition for evaluating outcomes of international mediation. Indeed, ‘the issue of assessing outcomes of international conflict management is a tricky one’ (Kleiboer, 1996, p. 361).
Apart from adopting a minimalist approach and eschewing definition of success altogether (Assefa, 1987), the scholarship on mediation offers two contending approaches which aim to help the observer to understand success and failure in international mediation. The first approach offers seemingly objective criteria which look at the ultimate consequences of the mediation effort. These criteria are often broad and designed to compensate for the idiosyncratic nature of international mediation. This in turn allows the researcher a significant space to manoeuvre, and to retain a certain amount of freedom in analysing patterns of mediation while ignoring the motives of the disputants and the mediator. It is closely linked to the structuralist paradigm of mediation and rational choice theories, which explain mediation occurrence and success by important signposts such as timing and power. The antecedents as well as the consequences of mediation are viewed as objective and independent variables of the communication process (Haass, 1990; Princen, 1992). Accordingly, typical criteria of this approach will consider a ceasefire, a peace treaty or other political settlements, as well as the opening of a dialogue and a marked reduction in the level of hostilities, as indicators of successful mediation (Bercovitch and Houston, 1996; Kriesberg, 1991; Touval, 1982).
While this approach is useful in generating systematic analysis and measurement of success, it fails to equate mediation success with mediation effectiveness. The conclusion of political settlement (let alone the reduction in the level of hostilities) does not necessarily translate to effective results. The 1993 Oslo Accords is perhaps the most pertinent example of an objective mediation success (a binding political agreement), while the issue of its effectiveness to end the Israeli– Palestinian conflict remains highly contentious.
This is where the second approach attempts to bridge the gap between results and perceptions. While the first approach eschews any discussion of the subjective interpretations of the parties to the bargaining process, this approach explains mediation success by focusing on the process of communication as a means of change of attitudes, largely outside the formal negotiation structure. Here the occurrence of mediation and its result are interpreted as upshots of goals and values. This approach is linked to the social-psychological paradigm of mediation, and accordingly mediation success is defined in terms of the subjective perceptions of the disputants and the mediator that their respective aims at the initiation of the process have been achieved (Hopmann, 1995; Smith, 1985).
While both approaches offer considerable advantages, inevitably a more synthesised approach that takes into account the tangible consequence of the mediation effort, as well as the perceptions of the parties involved, has more merit. Accordingly, it is suggested that the four cases examined in this study are associated with successful mediation efforts. Although some are less clear-cut than others (most notably the case of the Oslo Accords), they all are – and are the only – cases of binding agreements between Israel and the Arabs that were heralded as successes by observers as well as practitioners. Moreover, they were closely linked to some patterns of change in attitudes and perceptions on both sides.
The limited scope of this study precludes a comprehensive treatment of all the likely conditions for success in mediation. The three hypotheses examined here, however, touch upon three elements of mediation that have proved central in nearly every mediation effort in the Middle East. While the notions of power, impartiality and timing provide a constant source of debate between scholars, they are recognised as having considerable relevance to the outcome of any mediation effort.
The requisites for successful mediation are commonly divided into the circumstances of the intervention, and the mediator's qualities, or contextual and process variables (Bercovitch and Langley, 1993; Frei, 1976; Kleiboer, 1996; Touval, 1982). The contextual variables include, among others, the nature of the dispute, the level of intensity and the relationship between the disputants. Most importantly, however, there is a consensus that ‘Timing in launching negotiations is often decisive’ (Stein and Lewis, 1996, p. 467). There is little consensus however, as to what constitutes a ripe moment for mediation and when the mediator should enter the conflict (Crocker et al., 2003).
Similarly, the process variables, or mediator qualities, include numerous behaviours, roles and functions that may be assumed by the mediator during the mediation process – a direct result of the diverse nature of international mediation. Accordingly, James Wall alone has identified more than a hundred specific mediator techniques (Wall, 1981), while Joseph Stulberg provided a more modest list of nine mediator roles, ranging from a ‘catalyst’ to a ‘scapegoat’ (Stulberg, 1987). Others provide useful typologies of mediator roles and strategies (Zartman and Touval, 1985). Whatever the method used to evaluate these mediator qualities, it is evident that the mediator's power (leverage) and its perceived impartiality are of utmost relevance to its chances of success. As will be discussed below, the vast majority of definitions of mediation include either the mediator's active use of leverage or its perceived impartiality on the disputed issue. Furthermore, power and impartiality are intractably linked and often the occurrence of one in the mediation process can be successfully explained by the absence of the other (Smith, 1994).
Grouped together then, the themes of power, impartiality and timing can help to explain mediation success and failure, as well as to provide a better understanding of the process of mediation. Accordingly, they address some of the most important premises of international mediation: is mediation more likely to succeed when the mediator is a superpower, or rather a small state actor, and what are the effective sources of power (leverage) that different mediators enjoy? Can the disputants work with a mediator whose interests in the conflict are common knowledge, or should they approach an ostensibly neutral mediator? Should mediation be initiated early on in the conflict or should the mediator wait until the parties have exhausted all other options? What kind of mediator is likely to succeed in each phase of the conflict cycle? These are important questions, since the answers have practical as well as theoretical implications when one assesses the likelihood of future mediation success.
The broader nature of the relationship between mediation attributes and successful outcome of mediation is often debated by scholars. While some see the listing of general conditions for success as an exercise in futility, given the importance of regional, political and cultural idiosyncrasies (Meyer, 1960; Simkin, 1971), others tend to adopt a more prescriptive approach, seeking to suggest what conditions are conducive to successful mediation, in terms of the desired qualities of the mediator, as well as more systemic conditions such as the nature of the conflict and the intractability of the disputed issues (Bercovitch, 1997; Berridge, 2002; Mitchell, 1981). 1
Both approaches to the study of mediation are problematic. The first is excessively despondent about the benefit of constructing some broad lessons about theoretical and practical conditions for mediation success, and prefers to focus on the single case study. This view was summarised succinctly by William Simkin (1971, p. 118), who argued that ‘the variables are so many that it would be an exercise in futility to attempt to describe typical mediator behavior with respect to sequence, timing or the use or non-use of the various functions theoretically available’.
While this approach to mediation is almost one of resignation in the face of the task at hand, advocates of the second approach have offered overly simplistic alternatives. Studies of this nature content themselves with an extensive list of benign conditions and virtuous attributes that, if used appropriately, are likely to lead to a successful outcome, however it may be defined. These abstracted notions tend to propose ‘ideal-types’ of mediators that in reality rarely exist (see for example Berridge, 2002). Indeed, identifying an ideal mediator is no mean task, and these efforts have their merit in provoking discussion about the desirable qualities and functions of an appropriate mediator. Still, ideal mediators do not exist in the real world, and these descriptive attempts do little to further our understanding of the process of mediation and what is conducive to success in mediation.
This stark polarisation was addressed by Jacob Bercovitch and Allison Houston (1996), who in turn proposed a middle way known as the contingency approach, which views mediation outcomes ‘as the result of the interaction of context and process variables’ (Bercovitch and Houston, 1996, p. 17). However, Bercovitch himself is often guilty of producing all-inclusive definitions that have little value in providing policy-learning tools. Typically we are told that ‘only an appropriate mediator is likely to be effective’. He adds that ‘appropriate mediators should possess intelligence, tact, skills in drafting formal proposals, and a sense of humor, in addition to specific knowledge of the conflict at hand’ (Bercovitch, 1997, p. 147). We are then quickly moving back to the abstracted notion of ‘ideal mediator’. The contingency approach encompasses all the relevant variables of mediation, from the nature of the dispute and the mediator, to the initiation and strategies of mediation (Bercovitch and Houston, 1996); however, the conclusions drawn from these taxonomies about practical, helpful lessons for the future remain intangible.
Power
Power or leverage in mediation refers to the mediator's ability to move a party in its intended direction (Zartman and Touval, 1996, p. 455). However, by its voluntary, non-binding and non-violent nature, the mediation process ensures that the ultimate power lies with the disputants – they have the power to initiate the process, and they hold the right to terminate it. Implicitly then, a first prerequisite for the initiation, let alone success, of any mediation effort is that the mediator be accepted by the disputant parties. Therefore, the ability to satisfy the expectations of both parties through the use of various mediation techniques and strategies is paramount. The more powerful the mediator is, the more resources it can employ to change the disputants' attitudes and perceptions.
The issue of a mediator's leverage over the disputants is central to our understanding of the nature of the bargaining process between the parties. Mediators, argues Jeffery Rubin (2003), enjoy six different bases of power: (1) reward power, which enables the mediator to offer the disputants side payments in exchange for changes in behaviour; (2) coercive power, which relies on threats and sanctions in order to change disputants' behaviour; (3) expert power, which derives from the mediator's knowledge and expertise on relevant issues; (4) legitimate power, based on legal authority or international law; (5) referent power, which stems from the relationship between the mediator and the disputants; and (6) informational power, which positions the mediator as a message carrier between the disputants.
In a similar fashion William Zartman and Saadia Touval (1996) offer five sources of leverage that may help moving the disputants in the mediator's intended direction: (1) persuasion, the ability to convince the disputants that an alternative course of action is preferable to the status quo; (2) extraction, the ability to persuade the disputants to produce an attractive position; (3) termination, the ability to withdraw from the entire mediation process – this should not be exercised too often, due to the risk of credibility loss on behalf of the mediator; (4) deprivation, the ability to withhold resources from one party; and (5) gratification, the ability to provide incentives to one party or both.
While the advantages of a powerful mediator who maintains leverage in the disputed issue are evident, this is not to suggest that less powerful mediators cannot be successful. As will be demonstrated in the discussion on impartiality, power is essential only to powerful mediators; but while less powerful mediators such as individuals or non-governmental organisations (NGOs) (described by Princen as ‘neutral mediators’, with no interest in the conflict or its related issues) cannot offer the disputants material incentives, or deprive them of certain resources, their perceived impartiality on the disputed issue is the decisive factor that makes them acceptable to the disputants.
Less powerful mediators therefore can often offer more procedural rather than substantive help. Some of this procedural contribution is what Peter Carnevale (2003) terms tactical strength in mediation. Carnevale's distinction between strategic and tactical strengths in mediation is useful here. Although ‘power’ as well as ‘pure’ mediators may use both strategic as well as tactical strengths, Carnevale is certain who is more likely to succeed: ‘Power is the centrepiece of mediation. There is a lot to recommend power. Powerful mediators are more likely to have greater access to influential decision makers’ (Carnevale, 2002, p. 33).
Strategic strength in mediation refers to the ‘social power that relates to the resources and relationships that the mediator brings to the conflict’ (Carnevale, 2002, p. 28). Tactical strength, on the other hand, ‘refers to what the mediator does at the negotiating table. It is about adroit manoeuvring, technique, and procedure’ (Carnevale, 2002, p. 30). These tactics may be used more frequently by less powerful mediators simply because often that is all they can offer the disputants. 2 As will be shown later on, it is often the ability of the powerful (‘principal’) mediator to combine strategic as well as tactical strengths that proves invaluable in producing an agreement.
Impartiality
Perhaps the most contentious aspect of mediation is that of impartiality, or bias, of the mediator. In essence impartiality is a matter of the perceptions of the disputants. Whereas some studies argue that mediation, by definition, requires the mediator to be impartial (Burton and Dukes, 1990; Skjelsbaek, 1991), others maintain that impartiality is not a necessary prerequisite for successful mediation and, in some cases, can even get in the way (Bercovitch, 1986; Princen, 1992; Touval, 1982). In many respects these divergent views are the result of contrasting definitions of mediation (Smith, 1994). Touval (1982, p. 4), for example, sees mediators as powerful actors who ‘seek to influence the parties to make concessions by exerting pressures and offering incentives’, while Christopher Moore assigns the mediator a more limited role. Here the mediator is ‘an acceptable, impartial and neutral third party who has no authoritative decision-making power to assist disputing parties in voluntarily reaching their own mutually acceptable settlement of issues in dispute’ (Moore, 1986, p. 14). What is often overlooked, however, is the fact that acceptance does not necessarily lead to success, as is often postulated. The mediator's tasks of eliciting information and exercising influence are often best achieved, not when a mediator is unbiased, but when it possesses certain resources that are valued by the disputants (Zartman and Touval, 1985).
An attempt to bridge the gap between the two views of impartiality is offered by Ronald Fisher and Loraleigh Keashly (1991) and James Smith (1994) who argue that in essence this is a debate between two divergent concepts of third party roles and functions. This in turn creates two common styles of mediation – low power, low stake mediator, and high power, high stake mediator (Smith, 1985, p. 446). The distinction between ‘pure’ mediation and ‘power’ mediation (Fisher and Keashly, 1991, p. 33) is useful as it does not advocate a ‘right’ approach to impartiality, but instead offers a general terminology to help us understand when and why pure and power mediators are likely to be successful. Thus, when Bercovitch and Houston (2003) argue that power is more important than impartiality in pushing the parties forward, it is helpful to understand that for Bercovitch a mediator is a ‘power’ mediator, which carries with it certain assumptions about the role and strategies that the mediator may employ. This terminology is closely linked to Princen's (1992) distinction between neutral and principal mediators. Neutral mediators, by their lack of direct interest in the disputed issue, are attractive because of their perceived impartiality. Principal mediators, on the other hand, are accepted not on the basis that they are impartial, but rather on their ability to push the disputants towards an agreement by rewarding or depriving of certain resources.
Perceived impartiality then, is more crucial in the cases of ‘pure’ or neutral mediator or, as suggested by Smith (1994, p. 447), ‘the relevant characteristic for power mediators is their power, not their impartiality’. Their mandate to mediate rests upon the assumption – and expectation – that they will not favour a particular party during the process, and that their main interest is completely altruistic, that is to negotiate a settlement. In cases where the mediator is more powerful, however, impartiality is not necessary, and even not expected by the disputants. What matters here is the mediator's ability to deliver.
Timing
The historical record shows that ‘Knowing when to use mediation may be more important than how often it is used’ (Bercovitch, 1997, p. 145). Assessing when is the most propitious time for mediation is a difficult task, and various studies have attempted to identify the ripe moment for mediation (Haass, 1990; Zartman, 1985). Indeed, Geoff Berridge (2002, p. 200) concedes that ‘diagnosing “ripe moments” is not exactly a scientific exercise’. Nevertheless, some theoretical as well as empirical indicators may be useful in suggesting when the moment is ripe for mediation.
Mediation is more likely to succeed when both disputants have decided that the continuing costs of the status quo are no longer tolerable or, in other words, when the disputants have come to the conclusion that they will be better off with a settlement than without one. This ripe moment therefore opens for the mediator a narrow window of time that provides theoretically the best opportunity to succeed.
The question remains, however, how do we know that this is the ripe moment? There are conflicting views among scholars about what constitutes or how to recognise ripe moments (Kleiboer, 1994). The prevailing approach of late entry into the conflict relies on the assumption that perceptions of crisis or emergency will increase disputants' incentives to accept mediation (Rubin, 1981). Other scholars claim that mediation needs to be firmly initiated before a certain threshold of violence is reached and perceptions of the other side have hardened (Edmead, 1971).
The notions of a conflict cycle and a hurting stalemate are useful analytical tools for explaining mediation success in terms of timing. The notion of a conflict cycle suggests that most conflicts are characterised by sequential phases, namely, rising of tensions, confrontation, the outbreak of hostilities, followed by military clashes. The de-escalatory phases may include a ceasefire, settlement, rapprochement and reconciliation (Bercovitch and Langley, 1993; Crocker et al., 1999). In the Arab– Israeli conflict it is not uncommon to witness a reversal in the conflict cycle, or ‘leaps’ between escalatory and de-escalatory phases.
Successful mediation is therefore dependent on two variables: the level of violence (where the conflict is in its life cycle) and the identity of the mediator (neutral or principal, pure or powerful). In their discussion of the possible ‘entry points’ in the conflict cycle, Crocker et al. (1999) suggest that as long as the level of violence is not high, that is when perceptions on both sides have not completely hardened and alternatives to mediation still seem attractive, mediation is unlikely to succeed, and indeed the conflict may remain in this phase for years. This implies that the stakes for the mediator are not very high either, as failure to de-escalate the situation will not necessarily have a disastrous effect.
It is when the conflict is at its most violent phase, where ‘we—they’ images of the other have hardened and the parties are locked into a continuing struggle, that mediation is most likely to be successful. This is when the stakes are high and the opportunities for the mediator to engage the parties may diminish. Here, powerful mediators can be very effective in exercising substantive as well as procedural control – historical examples such as following the Yom Kippur War or the First Gulf War are relevant here. In short, when alternatives to mediation have worsened and become less attractive, the disputants are more likely to accept third party mediation (Crocker et al., 1999, pp. 25–9).
Accordingly then, the existence of this hurting stalemate provides perhaps the best benchmark to gauge when mediation should be initiated and when it is likely to succeed. Military defeats, changes in power relations or the threat of a deadline may convince the disputants that mediation is the best way out of the worsening situation.
The Arab—Israeli Conflict: Four Cases of Successful Mediation
Despite the numerous mediation efforts during nearly six decades of conflict, in only four cases were mediators successful in promoting substantive agreements between Israelis and Arabs. First, in the wake of the first Arab—Israeli war, UN mediator Dr Ralph Bunche successfully mediated four armistice agreements between Israel and Egypt, Lebanon, Jordan and Syria. Second, in the wake of the Yom Kippur War, US Secretary of State Dr Henry Kissinger skilfully orchestrated three disengagement agreements between Israel and Egypt and Syria. Third, some five years later, US President Jimmy Carter successfully negotiated the first Arab—Israeli peace treaty, between Israel and Egypt. Fourth, in a global environment of a new world order following the First Gulf War, a joint effort of the Norwegian government and an Oslo-based think tank resulted in a series of agreements between Israel and the Palestine Liberation Organisation (PLO), launching the Oslo peace process.
These four cases will be examined against the three hypotheses presented above. Can we identify similar patterns of mediator behaviour, which suggest, for example, that impartiality is an essential trait if third parties are to be successful? To what extent were functions of power conducive to the successful result of the four mediation efforts? Finally, to what degree were the successful outcomes in all four cases affected by the intermediaries' point of entry into the conflict cycle? The validation of all three hypotheses would suggest that mediators who are powerful (i.e. hold leverage over the disputants), impartial (have no direct interests in the dispute or its related issues) and intervene only after the conflict has reached an escalatory phase (meaning the breakout of hostilities and the existence of a hurting stalemate) are more likely to be effective than mediators who possess none, or only some of these traits.
Case One: Ralph Bunche: Power, Impartiality and Conflict at Escalatory Phase
One of the earliest mediators in the Arab—Israeli conflict, Dr Bunche was the deputy of UN mediator Count Folke Bernadotte, who was appointed almost simultaneously with the inception of the first Arab—Israeli war in May 1948. However, Bernadotte failed to promote ‘a peaceful adjustment of the future situation in Palestine’, as his terms of reference vaguely called for (Touval, 1982, p. 26). Bernadotte's lukewarm personality and mediocre qualities as a mediator made him unpopular with both sides, which rejected his two proposals. At the end, even his cherished asset as an impartial mediator could not save him from the tragic end of his mission, as he was assassinated in September 1948 by Jewish extremists who suspected that he was a British agent (Heller, 1979).
Against this background Bunche proved to be a much more able communicator, with an endearing personality that made him more acceptable to the disputants. With a long career in the US State Department, Bunche was also well acquainted with the idiosyncrasies of the conflict in Palestine. Finally, his well-known resourcefulness as a formulator helped him to secure four armistice agreements in a period of seven months. Starting in February 1949 on the island of Rhodes, Bunche effectively brought the first Arab—Israeli war to an end by overseeing complex negotiations between Israel and Egypt, Lebanon, Jordan and Syria. However, Bunche's personal qualities as a mediator go only a certain way to explain his success at the negotiation table. His leverage over the disputants and his perceived impartiality, as well as his entry to the conflict at an escalatory phase, undoubtedly contributed to the successful conclusion of his mediation efforts.
As an individual representing the UN, Bunche was not a powerful mediator to the extent that he could not directly reward or withhold incentives to influence the disputants' behaviour. However, as a former State Department official who could offer access to policy-makers in Washington, Bunche had a powerful leverage over the parties, both of which sought US aid and recognition. At the same time, however, Bunche was able to maintain a position of an impartial mediator, who had no interest in the disputed issues other than bringing an end to the fighting. While his predecessor aimed to solve the most politically intricate issues, such as the future of Jerusalem, Bunche's gradualist approach, which concentrated on the military aspect of the conflict and sought to bring an end to hostilities first, proved less threatening to Arabs as well as Israelis, who after nearly two years of fighting were militarily and economically exhausted. Diplomacy now looked more appealing as a means to bring an end to the continuing costs of fighting. The first case of successful mediation in the conflict, then, confirms all three hypotheses on power, impartiality and timing.
Case Two: Henry Kissinger: Power, Bias and Conflict at Escalatory Phase
The Yom Kippur War of October 1973 proved to be a watershed not only in terms of the role of the US in the conflict, but also in the profile and functions of the American mediator. US foreign policy in the wake of the war aimed to secure three objectives: mending relations with the Arab world; lifting the oil embargo of the Arab oil-producing countries; and pushing the Soviets out of the Middle East (Kissinger, 1982, p. 747). Tactically, the mediation efforts of Secretary of State Henry Kissinger, which produced three disengagement agreements between Israel, Egypt and Syria in eighteen months, were perhaps the first instance in the conflict of a mediator having learned the lessons of past diplomatic failures. While previous intermediaries produced overly ambitious and comprehensive plans for peace which instilled a certain rigidity in the parties (the experiences of Bernadotte; UN Ambassador Jarring following the Six Day War; and Secretary of State Rogers in 1969, are notable examples), Kissinger opted for a more limited, realistic approach to conflict resolution. Known as step-by-step diplomacy, Kissinger aimed to conclude a series of ‘small’ agreements that would help promote confidence and trust between the disputants, which could then be built upon at a later stage during negotiations on the final status of the more contentious issues. Although limited, the two disengagement agreements between Israel and Egypt in January 1974 and September 1975, and the one between Israel and Syria in May 1974, saw for the first time the withdrawal of Israeli forces from Arab territory.
By all accounts power was the quintessential element in Kissinger's mediation style. Effectively in control of American foreign policy, Kissinger was able to reward acquiescence in the form of military and economic aid. Indeed, these incentives were so central in Kissinger's mediation that some critics suggested that he was not exercising purposeful diplomacy, but rather ‘dollar diplomacy’. Furthermore, the American desire to bring an end to the energy crisis and to improve relations with the Arab world meant that on occasion (particularly during negotiations on the second Israeli—Egyptian disengagement) it seemed that Kissinger was much more eager to conclude an agreement than the disputants themselves (Quandt, 1977).
The fact that the mediator had a direct interest in the conflict and the outcome of any settlement precluded it from being impartial. Kissinger could not deny the American priorities in the Middle East (the preservation of Israel's military qualitative edge being one of them), nor could he hide his Jewish origins. However, despite the apparent bias towards Israel, Egypt and Syria willingly cooperated with Kissinger, who was in turn heralded as ‘Super-K’, the most successful diplomat of his generation. Almost paradoxically, it was the special relationship with Israel that enabled Kissinger to push the parties towards concessions. The lure of economic and military aid, as well as the promise of security guarantees, was attractive to Israel; at the same time, Egyptian President Sadat, who became disillusioned with Moscow, was eager to improve relations with Washington and, like Israel, to enjoy its economic and political patronage. In this case then, the mediator's ability to exert power was clearly more important than his perceived impartiality in the disputants' calculations of whether to accept its invitation (Stein, 1985).
In a similar fashion to Bunche, Kissinger's intervention also came at a propitious time. Kissinger deliberately avoided launching a peace initiative during the first days of the fighting as it was clear that the Egyptians and Syrians would reject it until they accomplished their military objectives; the Israelis too were expected to reject any political settlement until they regained the territories taken by Egypt and Syria in the first week of the war. After three weeks of war and thousands of casualties, however, all parties were more susceptible to third party intervention, which seemed more attractive than any military option. Just like Bunche, Kissinger entered the conflict when it reached a hurting stalemate and the parties could no longer tolerate the continuing costs of war. The experience of Kissinger's diplomacy, therefore, confirms the hypotheses on power and timing, but refutes the proposition that an impartial mediator is necessarily more likely to succeed than a biased one.
Case Three: The Camp David Accords: Power, Impartiality and Conflict at De-escalatory Phase
To some extent the Israel—Egypt peace treaty which was signed in 1979 was the inevitable result of the disengagement agreements that came in the wake of the Yom Kippur War. Although Henry Kissinger was no longer in office in 1978, the Camp David Accords that were concluded that year certainly built on his earlier achievements and the momentum for peace they had created.
Despite the diplomatic triumphs in the aftermath of the war, Israeli and Egyptian leaders found it difficult to make the crucial step towards a final, comprehensive peace settlement. The conflict between Israel and Egypt was undoubtedly in its de-escalatory phase, as witnessed by President Sadat's historic visit to Jerusalem and his address to the Israeli Knesset in November 1977, and a series of bilateral talks in Ismailya. This case therefore refutes the hypothesis that successful mediation will necessarily take place at the conflict's escalatory phase.
Perhaps because of the waning tension, the Egyptian and Israeli teams failed to make significant progress. This in part stemmed from the inability to read correctly the other side's position due to conflicting cultures, and the need to save face when the time came to ‘sell’ to their respective publics a peace treaty that no doubt included some painful concessions.
President Carter's intervention at this point (upon Sadat's request) in order to save the talks from reaching a deadlock aimed to provide two basic mediator functions: to minimise misunderstandings and miscommunications between the disputants, and to act as a helpful source for pooling information on the other side. By inviting Prime Minister Begin and President Sadat to Camp David in September 1978 with no preconditions and with a mandate to make decisions on the spot, President Carter decisively changed the nature of the Israeli—Egyptian rapprochement.
A powerful mediator, Carter was adroit in using a variety of techniques to accommodate the culturally different styles of negotiation the two leaders brought to Camp David. The American president needed to harmonise the incompatible needs of a hierarchical and traditionalist Egypt with a democratic and westernised Israel. Sadat and Begin had culturally divergent personalities, and contrasting negotiation styles. These differences were deeply rooted in national cultures. Sadat tended to talk about general principles and broad strategic concepts. He was not subject to democratic controls and could disregard the advice of members in his delegation. Begin, on the other hand, wanted to discuss every little detail; as a democratically elected leader he was first among equals and could not ignore or override the views of other members in his delegation, or those of the Knesset (Cohen, 1990). Carter acknowledged that his role as mediator would be shaped by these cultural differences, and was indeed successful in reconciling Sadat's philosophical and intuitive expositions and Begin's point-by-point debating technique.
That only a powerful mediator could have been successful in this case is evident by the special relationship that gradually evolved between Sadat and Carter. Coming from a society dominated by patron—client relations, Sadat, in return for his trust in Carter, expected ‘special’ favours in return. Sadat's willingness to place his own fate and that of Egypt in the hands of Carter was perhaps the single most noteworthy feature of the Camp David talks. Unlike Begin, Sadat did not nit-pick over details but preferred to talk about the broader picture of a future Egyptian—Israeli peace, as well as establishing personal rapport with the mediator – first with Kissinger and now with Carter. He also left it up to Carter and Secretary of State Vance to decide how to deal with Jerusalem. Carter, in turn, rewarded Sadat for his trust and confidence in him, in a manner that a ‘pure’ mediator could not offer – most notably by providing generous financial and military aid together with improved relations with the most powerful country in the world.
While the mediator had not changed since the last successful intervention in terms of its leverage over the disputants, it did go a certain distance with regard to its perceived impartiality. Whereas Israeli—American strategic relations bloomed during the early 1970s, the election of Carter in November 1976 was received with some suspicion in Jerusalem. Notwithstanding the customary pro-Israeli pledges during his presidential campaign, Israeli leaders were concerned by the growing influence of the ‘Brookings group’ over the new president. In the aftermath of the energy crisis a number of scholars from the Washington-based think tank the Brookings Institution, most notably William Quandt and Zbigniew Brzezinski, who was appointed national security adviser, rose to prominence in Washington. This emerging pro-Arab school was influenced by the consequences of the Yom Kippur War and the oil embargo, as well as the rise in power of the Saudi lobby in Washington. Carter adopted the recommendations of the group's report from December 1975, which concluded that the time had come for the US to move beyond interim agreements and to pursue a comprehensive and final peace in the Middle East; and that in this context an Israeli withdrawal to the pre-1967 borders was necessary and the creation of a Palestinian homeland in those territories was the ultimate objective.
Not surprisingly, these views did not make Carter a popular figure in Israel, and they struck a considerable blow to Israeli hopes that the nearly decade-long American ‘tilt’ towards Jerusalem in any peace negotiations would be maintained. At the same time, the Carter administration was far from showing a bias towards Egypt or the Arab cause in general. Carter could not, and did not wish to, overturn the ongoing special relationship with Israel that had first emerged during the Kennedy administration. Despite being the first American president to call publicly for an Israeli withdrawal to the pre-1967 borders and to acknowledge the rights of the Palestinians, Carter was committed to the survival of the state of Israel and to the strengthening of the special relationship between the two countries.
The fact that both Israel and Egypt were not overly enthusiastic following Carter's election, and that neither party was completely satisfied with the terms of the peace treaty (mostly on the issue of the Palestinians – the Israeli government reluctantly accepted a clause mentioning the Palestinian right for ‘autonomy’; Egypt complained that it did not go further than that), perhaps made Carter the ultimate impartial mediator, as neither party failed to feel ‘favoured’. The case of Camp David, then, confirms the two hypotheses on power and impartiality, but refutes the claim that mediation is more likely to succeed during the conflict's escalatory phase.
Case Four: The Oslo Process: Pure Mediation, Impartiality and Conflict at Escalatory Phase
Perhaps the most surprising aspect of the success of the Oslo peace process, which included the conclusion of two historic agreements between an Israeli government and the PLO in 1993 and 1995, and an unprecedented mutual recognition, is the identity of the mediator. Indeed, the small Nordic nation could not be further removed from the tragic, intricate realities of the Arab—Israeli conflict. Nevertheless, the joint effort of the Norwegian Ministry of Foreign Affairs and the Oslo-based Institute for Applied Science (FAFO) was surprisingly successful in mediating an Israeli—Palestinian agreement for the first time in the history of the Arab—Israeli conflict. The Norwegian experience was also the first quasi-informal case of successful mediation, pointing to the combination of formal and informal processes of mediation (Aggestam, 2002). While the Ministry of Foreign Affairs formally provided the financial resources, the locations of meetings and the official aura of the talks, the informal element provided by the Oslo think tank added flexibility and much-needed secrecy for the success of the talks, as well as a comfortable platform to develop trust and confidence around the negotiation table (Egeland, 1999).
Norway's tactics and functions as a mediator were markedly different to the diplomatic arsenal deployed by the US. It could not exert its power over the disputants because it was what Princen termed a ‘neutral’ mediator – it had no direct or indirect interest in the conflict or the disputed issues per se. It could not reward the parties for acquiescence, nor could it threaten them with coercive measures in cases of intransigence. Instead, the Norwegians could offer a ‘pure’ alternative to the power strategies they had grown accustomed to. Given their limited leverage over the disputants, the Norwegian mediators' role was adroitly adjusted to provide more procedural and communicative services, rather than manipulative strategies. However, the Oslo process was empowered by the quality of secrecy, providing the ultimate back channel for talks between the two parties. This meant firstly that the media was not involved. Secondly, the secret negotiations enabled the parties to actually talk and negotiate, rather than spending valuable time on diplomatic protocols or preparing speeches; here all efforts were concentrated on substantive negotiations. Thirdly, the seclusion in Oslo of seven individuals allowed for the development of close personal friendships and an atmosphere of mutual trust and confidence. Finally, the participation of a Norwegian NGO (the Oslo think tank) provided the deniability factor that was so important in the early stages. All these factors more than compensated for the mediator's lack of concrete, US-style, leverage over the parties.
The mediator's seemingly ‘powerless’ position became more attractive given Norway's unique position as an impartial actor in the Israeli—Palestinian dispute. Although not directly interested in the disputed issues, Norway had enjoyed the trust and confidence of Israel as well as the PLO since the 1970s. The Norwegian Labour party forged close ties with its Israeli counterpart, while at the same time showing great sympathy for the Palestinian cause. In fact it was Yasser Arafat himself who first proposed using Norway as a back channel to Israel in 1979. The idea finally materialised as a series of secret meetings between Israeli academics and Palestinian officials in January 1993, following the failure of the public talks in Washington that came in the wake of the 1991 Madrid Peace Conference.
Notwithstanding the mediator's favourable attributes, the timing of the talks was particularly propitious. First, in global terms, the ending of the Cold War and the demise of the Soviet Union created new conditions for peace. The danger of a regional crisis developing into a superpower confrontation had reduced, and so had the central role of American mediation in the region. Moreover, in the new global environment the Clinton administration was more reluctant to underwrite the financial commitments that a Camp David type of peace would bring, and so the willingness on behalf of the traditional mediator to reward the parties was considerably more limited. Finally, in regional terms, the Israeli—Palestinian conflict in the early 1990s had gone through an escalatory phase, which began in 1987 with the outbreak of the first Intifada. The daily clashes between Palestinians and Israeli forces had taken their toll and after five years of fighting both sides had reached a hurting stalemate in their ability to sustain the mounting costs, in terms of human lives and infrastructure (for the Palestinians in particular), the economic ramifications and the corrosion in domestic and international support (mainly in the case of Israel). After fifteen years of the intransigent Likud party in power, the election of Prime Minister Rabin and the Labour party in Israel, along with the declining power of the Tunis-based PLO leadership as a result of the Intifada and the rising power of Hamas, meant that Israeli and Palestinian leaders alike believed that the time was ripe for talks. The entry of the Norwegians into the conflict at this stage could not have been more auspicious. The case of the Oslo peace process therefore confirms the hypotheses on impartiality and timing, but refutes the hypothesis on power mediation.
Conclusion
The experiences of the four successful intermediaries confirmed only in part the hypothesis that powerful and impartial mediators who intervene at the escalatory phase of the conflict are more likely to be successful. Apart from the case of Ralph Bunche which confirmed all three hypotheses on power, impartiality and timing, the remaining three confirmed two of the three hypotheses: Kissinger's step-by-step diplomacy confirmed the hypotheses on power and timing; the Camp David Accords confirmed the hypotheses on power and impartiality; and the Oslo case confirmed the hypotheses on impartiality and timing. Still, despite the mixed record, the fact that every hypothesis was confirmed at least three times suggests that the three elements go a long way in explaining mediator success and failure. Keeping both the theoretical perspectives and the empirical evidence in mind, some generic conclusions can be drawn which could provide useful policy learning tools for future intermediaries.
It is evident that ‘power’ mediators have been markedly more successful than ‘pure’ mediators in the Arab—Israeli conflict. Pure mediators may enjoy success in the early stages of negotiations, however only powerful mediators can successfully monitor and ensure the implementation of the agreement. Putting aside the cases of Kissinger and Carter, where power was the quintessential element of mediation, the more multifaceted cases of Bunche and Oslo also demonstrate the necessity of power to the success of mediation. Notwithstanding Dr Bunche's ostensibly ‘pure’ position as the UN mediator during the armistice talks in 1949, it was his American nationality and long experience in the State Department that made him so attractive to both Israelis and Arabs. In addition to his charismatic personality and ability as a communicator, his credentials in Washington and the ability to offer almost immediate access to policy-makers made him a very powerful mediator, much more than his terms of references in the UN mandate suggested. There is much to recommend power in the Norwegian case as well. The mediation role assumed by the Norwegians was evidently ‘pure’, that is, restricted to communication and procedural functions with no resort to manipulative or coercive strategies. However, perhaps the lack of power on behalf of the mediator in this case can explain the ultimate failure of the Oslo peace process. The Norwegians, helped by their status as a small, neutral and disinterested actor in the conflict, were successful in bringing the disputants together and in establishing trust and confidence – this in part explains the successful conclusion of the Oslo Accords in 1993. However, at the same time, it was precisely this powerless position that was partly to blame for the failure in implementation – this is why more than a decade after the start of negotiations in Oslo, we still talk about a ‘process’. The inability of the mediator to monitor and ensure the implementation of the agreements is a lesson that is often ignored. While the US were able to offer security guarantees to Israel or political assurances to Egypt in the 1970s in return for their respective implementation of the disengagement agreements or the Camp David talks, the Norwegians were in no position to ensure implementation. This was evident early on when both parties were slow to act on their promises – the Oslo peace process quickly collapsed. It was only following the active support of the Clinton administration that a minor agreement over Hebron was reached in 1997, for example.
The conclusion, then, is that while pure mediators may be successful in the procedural, more functional aspects of mediation, powerful mediators are still essential in every mediation effort in the Arab—Israeli conflict in order to help the parties jump over the last hurdle, and to ensure the implementation of the agreement. There may be a case for engaging with a more powerful actor as a monitor and enforcer to support the process once it has been initiated by a ‘pure’ mediator.
The historical record also suggests that the mediator's impartiality or bias has no definitive bearing on the likelihood of mediation success, although three out of the four successful mediators used a variety of power strategies to exert leverage during negotiations. At least with reference to the Arab—Israeli conflict then, the four cases hint that power matters more than impartiality. Ralph Bunche used his perceived impartiality as UN mediator, but also his close links with Washington to conclude the armistice agreements. Henry Kissinger had never made an effort to hide his Jewish origins, nor could he reliably deny the special relationship between the US and Israel – yet he still managed to produce three agreements in eighteen months. As for the Norwegians, their perceived impartiality certainly helped to open communication channels between Israel and the PLO, but at the same time their lack of power proved a detrimental factor in their inability to monitor the implementation of the Oslo agreements. Moreover, the evidence suggests that the disputants are likely to engage constructively with the mediator if they believe it can positively change the nature of their interaction; trust does not necessarily stem from impartiality.
Finally, the timing of intervention has proved crucial to the disputants' inclination to accept the mediation effort in the first place. Perhaps more than any other factor, good timing is likely to contribute to a high likelihood of success. The historical record suggests that all four successful mediation efforts took place at propitious times. The armistice agreements were concluded after nearly two years of fighting with all the parties militarily exhausted, a situation which repeated itself nearly 25 years later in the wake of the Yom Kippur War, with Egyptian and Israeli forces entangled and the Egyptian Third Army on the brink of annihilation; in both cases third party mediation came at the right time and was perceived by the disputants as a favourable alternative to the continuation of fighting. The timing of the Camp David talks was also ripe. Although it did not come at an escalatory phase and there were no explicit signs of tension or hostilities between Israel and Egypt, President Carter's invitation of Sadat and Begin to Camp David followed Sadat's historic visit to Jerusalem and the subsequent deadlock in the bilateral talks between the Egyptian and Israeli delegations. As for the Oslo experience, the ‘new world order’ following the Gulf War and the demise of superpower rivalry created favourable conditions for Israeli—Palestinian negotiations. The first Intifada, which began in 1987, had an important corrosive effect on the disputants' resilience and desire to continue the hostilities and made third party mediation seem attractive. By stark contrast, diplomatic initiatives in the 1950s were doomed to fail from the outset since the conditions on the ground did not constitute a ‘hurting stalemate‘, as the disputants preferred to maintain the relatively quiet status quo rather than enter into risky negotiations. This was also the case during the 1980s, when initiatives by President Reagan and Secretary of State Shultz failed to make an impact.
To sum up, it is evident that any of the three parameters alone cannot guarantee a successful outcome; nor does the existence of all three necessarily guarantee success. Often the mediation process is shaped by psychological, personal and cultural attributes on behalf of the mediator as well as the disputants, the impact of which is difficult to measure. Nevertheless, understanding the importance of power and impartiality in the process of the mediation, and the impact of the timing of the intervention on the disputants' motivation to acquiesce, can help us to contextualise past and present mediation efforts, and to extract valuable lessons about the conditions for successful mediation in the future.
Footnotes
I am grateful to Richard Aldrich for his comments on earlier drafts of this article.
1
Underlying this debate are more fundamental tensions between perspectives that are rooted, respectively, in area studies and political science.
2
These include communication tactics (separate the parties, shuttling between them, bringing them together); image tactics (affecting parties' perceptions, helping them to save face); momentum tactics (creating momentum by controlling the agenda); and relational tactics (modifying relations with and between the parties, avoiding alienating them in order to achieve an agreement).