A Case of Undue Pressure:

This paper focuses on the mediator's strategy and tactics as variables that enhance or diminish the prospect of success. I argue that state and inter-governmental mediators frequently deviate from the logic of mediation and resort instead to power-based diplomacy. They are insensitive to the socio-pyschological dynamics of conflict and appear to be unfamiliar with the principles and techniques of mediation that provide a basis for managing and transforming those dynamics. As a result, the mediators' endeavours heighten the suspicion, fear and anger of beleaguered disputants and are consequently ineffectual if not counter-productive. A key assumption here is that conflict has both substantive and emotional dimensions that have to be addressed if the conflict is to be resolved.

The critique is organised around six strategic principles of mediation: mediators must not be partisan; the parties must consent to mediation and the appointment of the mediator; conflict cannot be resolved quickly and easily; the parties must own the settlement; mediators must be flexible; and mediators must not apply punitive measures. The significance of each of these principles is explained in terms of the political and psychological pressure experienced by parties in conflict, and the problems that flow from breaches of the principles are illustrated with reference to mediation efforts in African civil wars. The principles are formulated and motivated in the light of mediation conducted by staff at the Centre for Conflict Resolution (CCR) in Cape Town, although they have their roots in theory developed elsewhere. ¹ In the penultimate section and the conclusion, I suggest that international actors might be more effective if they regarded mediation as a specialist discipline and were proficient in its methodology.

CCR's understanding of mediation can be summarised as follows. Mediation is the process whereby an independent third party assists parties in conflict to reach a collectively acceptable settlement through dialogue and negotiation. The mediator's principal function is to help the disputants shift their stance from an adversarial `winner-takes-all' disposition to a more accommodating posture. The underlying premise is that the parties are interdependent and their co-operation is required to resolve the conflict; where this is not the case and the parties are able to walk away from the dispute, there may be little need for mediation. The premise does apply to civil wars where the protagonists have sizable constituencies that inhabit the same territory. The outright defeat of a community is seldom possible, and its suppression or exclusion from a settlement may lead to a resumption of hostilities at a later stage. Accommodating the needs of all the parties is therefore a prerequisite for durable peace and stability.

Mediators must not be partisan

Individuals and groups locked in conflict tend to regard each other with intense suspicion and animosity. They are reluctant to engage in dialogue even when they have begun to contemplate ways of resolving the conflict. Alternatively, they might enter into talks but be unable to move beyond mutual recriminations. The utility of mediation lies in creating a relatively calm and safe space for them to address their concerns with the assistance of a trusted third party. Mediation can thus be viewed as a confidence-building exercise, with the mediator acting as a bridge between the adversaries. Their trust in the mediator rests on the assumption that he/she will treat them fairly. This is a critical consideration especially for weaker parties which fear being outmaneuvered in the course of negotiations. If mediators are biased against one of the parties, they break that bond of trust and jeopardise the success of the process.

This emphasis on impartiality reflects an ideal that is not fully attainable since no-one is ever free of bias. CCR mediators naturally have personal, cultural and professional values, they are always concerned about the equity of agreements reached, and they invariably acquire positive or negative opinions of the antagonists in a specific dispute. Yet one of their professional values, declared expressly to the parties, is a commitment to facilitate the process in a non-partisan manner. If they are unable to honour that commitment for any reason in a given conflict, they will refrain from playing a mediating role.

Some scholars claim that the principle of impartiality does not apply to international mediation (e.g. Bercovitch 1996). William Smith (1985) presents the following version of the argument. Whereas the impartiality of mediators in domestic settings stems from the fact that they have no extended relationship with the parties and no interest in the dispute beyond its peaceful resolution, states have little motivation to mediate in international conflicts other than because they have a relationship with the adversaries and an interest in the details of a settlement. Consequently, international mediators are probably always biased to some degree. The bias may enhance the acceptability and effectiveness of the mediating state because its interest in its relationship with both of the disputants gives each of them a measure of leverage over it and vice versa. The less favoured party co-operates in the hope that the mediating state will extract concessions from the party with which it enjoys closer ties.

This argument is analytically and empirically incomplete. First, it does not distinguish between a partial mediator who is imposed on the parties and one whom they accept without duress; there can be no objection if they agree to use a mediator who is affiliated to one of them, and in this regard no distinction need be drawn between international and domestic mediation. Second, it considers the problem of bias primarily in terms of the mediator's interest in the dispute and prior relationship with the parties when the problem relates more to the mediator's conduct during the peacemaking process. Third, it disregards the fact that an international mediator may succeed precisely because of its lack of bias, as in the case of the Community of Sant' Egidio in Mozambique in 1990-2; according to Father Romano (1998:7), "our strength was exactly not having to defend any vested interest in the country but the one of a solid peace". Fourth, it ignores the evidence, illustrated below, of a mediator's acceptability and effectiveness being diminished greatly by its partisanship.

In 1989 Liberia was plunged into civil war when rebels led by Charles Taylor sought to oust Samuel Doe who had seized power in a coup ten years earlier. Five members of the Economic Community of West African States (ECOWAS) formed a Standing Mediation Committee to resolve the conflict. When its initial peacemaking bid failed, the Committee established the ECOWAS Ceasefire Monitoring Group (ECOMOG), a military force with a mandate for peacekeeping and peace enforcement. Over the next six years ECOMOG became embroiled in the fighting, prolonging the war and contributing to wider regional instability. Dominated by Nigeria which had previously backed the despotic Doe, it destroyed its claim to neutrality by targeting Taylor and arming rival factions (see Howe 1996/7; and Nyakyi 1998). According to Anthony Nyakyi (1998), former Special Representative of the UN Secretary-General to Liberia, the enmity between Taylor and Nigeria was the main impediment to securing a lasting peace agreement.

Since 1993 the Inter-Governmental Authority on Development (IGAD) has attempted to mediate in the civil war in Sudan. While the member states of this east African formation have a legitimate interest in ending the war because of its destabilising regional impact, three of them have bilateral military conflicts with Khartoum. Ethiopia, Eritrea and Uganda provide military support to Sudanese rebel movements, and Khartoum sponsors extremist groups accused of terrorism in each of these countries (see Africa Confidential 38(3), 31 January 1997; and Deng 1997). Francis Deng (1997) understates the problem in suggesting that these antagonisms complicate matters, raise questions about the mediators' credibility and threaten to undermine the regional initiative.

In 1996 the elected Hutu government in Burundi was overthrown in a coup by Major Pierre Buyoya and the predominantly Tutsi army. Neighbouring states immediately imposed sanctions on the country with the endorsement of former President Nyerere of Tanzania, appointed as the mediator for Burundi by regional leaders, the OAU and the UN Secretary-General. Three years later the embargo remains in place and no similar pressure has been put on any other party or rebel group. While the Buyoya regime has pursued negotiations with its internal opponents, it has resisted the external peace process under Nyerere's leadership on the grounds that he is anti-Tutsi; it has called for Nyerere's resignation and the appointment of a "team of neutral mediators" (IRIN Update No. 405, 1998). The tension between Buyoya and Nyerere, and the debate around sanctions and the mediator, now command as much attention as the conflict in Burundi (see International Crisis Group 1998:36-50; and Van Eck 1997).

In 1993 the second United Nations operation in Somalia (UNOSOM II) was launched with objectives that encompassed humanitarian assistance, peacebuilding and the promotion of political reconciliation among warring Somali factions. After Pakistani peacekeepers were killed in an attack, the UN embarked on a military campaign against General Aideed, the faction leader deemed responsible. In their bid to hunt him down, UN forces bombed a house and killed over fifty clan members. Ken Menkhaus (1996:59) asserts that "efforts to arrest or marginalize warlords failed to account for the deep-rooted notion of collective responsibility in Somali political culture... Actions taken against a clan's militia leader were seen by Somalis not as justice done to an errant individual, but as a hostile action against the entire clan". Having compromised its impartial standing, the UN became too discredited to pursue its mandate and departed Somalia in ignominy (see Menkhaus 1996; and Jan 1996).

In many of the situations described above, the mediator's bias may well have been justified. Moreover, although the complexity of the situations defies simple categorisations, a strong case can be made for resorting to punitive action in response to coups, armed rebellion and gross violations of human rights. Regardless of whether the bias is justified, however, a mediating body that undertakes such action will be mistrusted by the targeted party as surely as a soccer team mistrusts a jaundiced referee. It sacrifices its status as an `honest broker' and becomes a party to the conflict. Smith (1985) acknowledges that a biased domestic mediator will be viewed with suspicion and hostility by the disfavoured disputant and may make the conflict more intractable. This logic applies equally to international mediators. As argued further below, enforcement and mediation functions should be performed by different actors.

The parties must consent to mediation and the appointment of the mediator

Paradoxically, mediation is most threatening to disputant parties when it is most required. Where a conflict escalates to the point that its negative consequences are manifestly serious, independent observers might regard mediation and negotiations as obvious means of overcoming the impasse. At this stage, however, the adversaries are likely to hold entrenched positions and to view the conflict in zero-sum terms. From their perspective, mediation entails talking to `the enemy' and the prospect of compromising fundamental principles in order to reach a settlement. The parties fear losing face in the eyes of their supporters, being outmaneuvered by their opponent's negotiating tactics, and being pressurised by the mediator to abandon their goals.

Given these dynamics, CCR undertakes mediation only with consent of the disputants. It may be called in the first instance by one of the parties or by some authority with an interest in ending the conflict, but its staff will then meet with all the parties to ascertain their willingness to engage in mediation under CCR's auspices. This approach raises the disputants' confidence by giving them a strong measure of control over the process: they can select a mediator whom they trust, and any one of them can dismiss the mediator at any stage. As a result, the mediator acquires a keen sense of accountability to the parties. There is the further advantage of setting an early precedent of decision-making by consensus since the protagonists have to agree on the potential benefit of dialogue and on the appointment of the mediator.

The voluntary and consensual nature of mediation is so widely endorsed at the level of discourse and prescription that it can be regarded as a defining feature of the process. It is reflected in academic definitions of mediation (e.g. Bercovitch 1996:246); in perspectives on the peacemaking function of the UN secretary-general (e.g. Vance and Hamburg 1997:14); and in policy proposals on conflict resolution emanating from African quarters (e.g. Othman 1998:16). The principle of affording parties a `free choice of means' in the pacific settlement of international disputes is captured in Article 33 of the UN Charter and in other legal instruments, including the 1964 OAU Protocol of the Commission of Mediation, Conciliation and Arbitration (United Nations 1992:7 and 33-45). The principle implies that mediation cannot be imposed on the parties without their consent and acceptance of the mediator (United Nations 1992:42; and Puchala 1993:82).

The principle is often ignored in practice, however. When a conflict within or between states reaches a certain level of intensity, third party countries and multi-national bodies tend to assume the role of mediator and appoint envoys to that end without consulting the protagonists. The parties may not be receptive to mediation or they may be ready for talks but have no confidence in the host agency or its envoy. In either event, and especially if their belligerence stems from insecurity, they are likely to regard the endeavour as a form of interference and feel threatened thereby. More striking still is the refusal of mediators to step down when a disputant objects to their presence on the grounds of bias. As noted earlier, this is currently the case in Burundi where the main Tutsi party has called for the withdrawal of former President Nyerere as the mediator because of his involvement in imposing sanctions on that country. The mediator's persistence in these circumstances can become a significant secondary source of conflict and an obstacle to resolving the primary conflict.

Peacemakers naturally play uninvited roles where a party refuses to enter into negotiations, seeking to persuade it otherwise or acting as an interlocutor. Where the disputants are willing to engage in dialogue, in contrast, enabling them to choose the mediator is clearly preferable to imposing one on them. Each party could be asked to compile a list of acceptable mediators, the matter being settled quickly if there are any common nominations. If this is not the case, further iterations of the process would be required and, through a process of elimination, the parties might settle either for a balanced team of mediators or for a single mediator who is not completely unacceptable to any of them. This approach could also be followed where mediation is already underway and a party rejects the mediator. The rejection may be a pretext for avoiding talks, but nothing would be lost by calling that party's bluff and inviting it to propose alternative or additional mediators.

Conflict is complex and cannot be resolved quickly and easily

CCR mediators are seldom able to resolve community conflicts swiftly. Apparently irreconcilable interests and values, exacerbated by intense mistrust and by competition over scarce resources, defy simple solutions. The degree of complexity rises considerably where the conflict has a national character, the adversaries believe that their physical or cultural survival is at stake, there are multiple disputants and divisions within their ranks, large-scale violence has already occurred, and the principal causes of the conflict are structural.

Without discounting the UN's mistakes in Somalia in 1993-5, Menkhaus (1996) insists that certain of the organisation's goals and strategies were inherently incompatible and bound to generate conflict because of the objective circumstances of that country. He argues that Somali and international peacemakers were confronted by "unique political dilemmas and thus a menu of very unpalatable options, all of which posed a high probability of failure. There were, in short, no easy and obvious reconciliation strategies" (Menkhaus 1996:43; see also Friedrich Ebert Stiftung et al 1995:11-13). This argument has broad applicability to civil wars which, despite the distinctive features of each, present common dilemmas. For example, mediation and reconciliation efforts may entail courting groups responsible for atrocities and affording them formal recognition; this may alienate sectors of society and be perceived as rewarding violence. On the other hand, excluding the groups is likely to ensure their resistance to both the peace process and its outcome.

While these observations ought to be self-evident given the intractability of so many intra-state conflicts, the conceptual framework of international actors often suggests otherwise. In An Agenda for Peace, Boutros-Ghali (1992:11) uses the terms "conflict" to denote the presence of violence and "dispute" to depict a lower level of intensity. Yet the notion of a `dispute' scarcely captures the centrality and complexity of the root causes of civil wars in Africa and elsewhere: weak or failed states; authoritarian rule; a lack of coincidence between nation and state; the exclusion of minorities from governance; and acute socio-economic deprivation and inequity. The conventional approach to `early warning' and preventive diplomacy relegates these factors to `background conditions' and assumes naively that mass violence can be averted by reacting to its proximate rather than its structural causes.

In practice, the problem manifests itself in `quick fix' strategies that reflect little appreciation of the difficulty of achieving reconciliation and little familiarity with the intricacies of local dynamics and culture. The UN operation in Somalia is a case in point. When the Security Council launched UNOSOM II with a nine month mandate, Madelaine Albright, then US Ambassador to the UN, declared that "with this resolution, we will embark on an unprecedented enterprise aimed at nothing less than the restoration of an entire country as a proud, functioning and viable member of the community of nations" (quoted in Jan 1996:3). Given the extent of clan warfare and the absence of any state authority in Somalia, this goal was never likely to be attained within the time frame of the mission.

Although Albright's statement could have been intended simply to convey a message of hope, many analysts subsequently criticised the UN for being insufficiently sensitive and patient. There is a widely held view that the organisation was "more concerned with its own timetables in New York than with Somali political realities" and that, lacking an understanding of these realities and traditional reconciliation processes, it "worked against rather than with indigenous practices of conflict management" (Menkhaus 1996:57; see also Jan 1996; and Friedrich Ebert Stiftung et al 1995). In contrast, Assefa (1987:169) notes that the credibility (and no doubt also the effectiveness) of the World Council of Churches and the All African Council of Churches as mediators in Sudan in the early 1970s was enhanced by their detailed knowledge of the nature and complexity of the conflict.

In his study of mediation efforts to end the Sudanese civil war in the 1990s, Deng (1997:28-29) warns that diplomatic intercession which seeks "quick fixes in deep-rooted identity conflicts can only complicate the crisis". He concludes that "there is a tendency on the part of diplomatic peacemakers to look for aspects of a problem that lend themselves to relatively easy solutions and to postpone more difficult ones. While this is understandable, and perhaps even practical, it is probably the more difficult ones that eventually provoke people to violent confrontation, making them determined to kill and risk being killed". In these circumstances, as argued below, attempts to rush the process and compel the parties to conclude a settlement prematurely are unlikely to succeed.

The parties must own the settlement

In many situations of conflict, independent observers might view the stance of one or more of the parties as irrational or unreasonable, the issues in dispute as relatively trivial, or the solution to the underlying problems as fairly obvious. If the parties shared these views, however, there would be no need for mediation. Adversaries are typically motivated by an acute sense of injustice, by real or imagined threats to their security, or by unmet needs which they regard as fundamental. Moreover, basic human needs are not limited to material imperatives like food and shelter. Individuals and groups crave respect, affirmation and acknowledgement. They want to be involved in decisions that affect their lives and resent being treated as the object of some other body's plans. CCR has often mediated in conflicts where a community rejects a development project that was `imposed' on it without prior consultation.

In light of the above, a mediating body will alienate the parties and invoke resistance to its effort if it fails to take seriously the importance they attach both to their positions and to the process of resolving the conflict. Mediators should therefore refrain from prescribing solutions. Their job is not so much to solve problems as to facilitate problem-solving by the disputants. Agreements that are not shaped and embraced by the parties have little chance of enduring. For this reason too, mediators should avoid pushing the protagonists to reach a settlement prematurely.

The pressure on the mediator and the parties may be intense where a large number of people are dieing in on-going hostilities. Father Romano (1998:5-7) recalls that the Sant' Egidio team in Mozambique was "put under strong pressure to end the talks quickly... [Our] awareness that every additional day more of war meant more killings was an extremely hard burden to bear". The mediators nevertheless resisted the pressure because the "pathology of memory could not easily be cancelled" and because "there is no use in forcing people to agree on anything. The only way the process could have been successful and the reason that made it successful was that all the actors involved gained ownership [of] the process...".

State and multi-national mediators, in contrast, tend to be more concerned with securing a settlement than with the process by which this is done. They regard their function as one of persuasion rather than facilitation, relying on the authority and muscle of the body which appointed them to press for rapid results. This invariably gives rise to perceptions of partisanship and coercion. The use or threat of sanctions or force might compel the parties to reach to a settlement but it will be short-lived if they are not genuinely committed to it.

Even when there is no undue pressure on the parties, an overly prescriptive approach by the mediators is disadvantageous. By way of illustration, in 1994 IGAD formulated a Declaration of Principles that outlined the essential elements required to achieve a lasting peace in Sudan. The principles appear to provide a fair and pragmatic basis for addressing the root causes of the conflict but they run directly counter to the position of the Sudanese government. Not surprisingly, the document provoked an adverse reaction from Khartoum which, according to Francis Deng (1997:27), "now sees the mediators as no longer neutral".

A related problem arises when international actors seek to forge a settlement among political elites without fully engaging local communities, a criticism leveled at the UN operation in Somalia (Menkhaus 1996; and Jan 1996). Bethuel Kiplagat (1988) notes the absence of public participation in the failed Ugandan peace talks of 1985 and its presence in successful peacemaking at grassroots level in Sudan and Kenya. He describes a traditional approach to mediation through public gatherings that constitute a `community in discussion'. The disputants and the effected communities participate equally in the process; there is an emphasis on healing relationships rather than on bargaining; and a spiritual or religious dimension is introduced to promote reconciliation and higher values. Aside from the fact that communities will have useful ideas on resolving the conflict, their involvement in the discussion puts pressure on leaders to make peace and abide by peace agreements.

Insisting that mediators should avoid being prescriptive does not mean that they should never make proposals and tender advice. The point is rather that they should exercise great sensitivity and circumspection when doing so. No precise formulation can be presented in this regard, but the spirit of careful mediation is captured in the following account by Count Folke Bernadotte, the UN Secretary-General's special representative to Palestine in 1948: "In the course of the truce negotiations, the two parties had made it quite clear that they expected to receive from me, during the period of the truce, an indication of my ideas as to a possible basis of settlement. This, in their opinion, was the raison d'etre of the truce. Notwithstanding, therefore, the complete divergence of aims and the very short time left at my disposal, I decided to submit to the two parties a set of tentative suggestions, with the primary intention to discover whether there might be found at this stage a common ground on which further discussion and mediation could proceed" (quoted in Puchala 1993:88).

Mediators must be flexible

Whatever the validity of general propositions on conflict advanced in the academic and policy literature, the actual dynamics of conflict differ markedly from situation to situation as a result of historical developments and diverse personalities, social relations, cultural perspectives and material conditions. Since few of these factors are immutable, and because conflict is an open system, its thrust and contours change over time. As Chester Crocker (1992:204) observes with respect to intra-state and regional conflicts, each "has its own structure and should be approached with an appreciation of its uniqueness".

Mediators must therefore be sufficiently flexible to adapt their style and methods to the circumstances, the protagonists and the evolution of the conflict. Mechanical and formulaic models are bound to fail. The point is well made by Crocker (1992) in asserting that conflict resolution requires an empirical, case-by-case approach rather than preset recipes. In similar vein, Father Romano (1998:7) contends that Sant' Egidio's success in Mozambique should be regarded as an "example" and not a "model". The need for flexibility is reinforced by the view expressed earlier that it is not the job of mediators to craft and sell solutions to the parties. They may have an opinion on the substance of the conflict and its resolution, but they will provoke resistance from one or more of the disputants if they initiate the mediation process with a fixed position on the details of a settlement.

The UN and regional bodies like the OAU lack the requisite flexibility when their mediation efforts are subject to decision-making by member states. Saadia Touval (1994) observes that inter-governmental organisations tend to adopt only those measures on which consensus is possible, excluding ideas on which unanimity cannot be achieved; that such consensus is often ambiguous, reflecting compromises based on the lowest common denominator; that decisions cannot be modified easily in response to new developments; and that the compromises and discord weaken the coherence and credibility of a mediation exercise.

In some instances the organisation is rendered impotent by divisions within its ranks or the formal or informal veto of a powerful state. In other cases disparate state interests within the mediating body are exploited by the disputants and have the affect of exacerbating the conflict. Nyakyi (1998:2) draws this conclusion in assessing the role of ECOWAS during the Liberian civil war: "...although the lack of political will and commitment on the part of the warring factions to achieve a settlement was a major obstacle, the divisions among the key ECOWAS countries directly involved in Liberia, which supported different factions and failed to form a common policy on Liberia, was the underlying factor which emboldened the factions in their intransigence".

The inflexibility of a multi-national body can also stem from its principles and institutional predisposition. For example, a 1964 resolution of the OAU Assembly of Heads of State and Government forbids secession and irredentism, proclaiming the inviolability of African borders imposed by the colonial powers. This position derives partly from the justified concern that legitimising the redrawing of boundaries would provoke intra- and inter-state violence. Nevertheless, it offers little scope for easing the on-going tensions induced by incongruous borders. It ignores the possibility that ethnic groups might agree to partition a state, as happened in Czechoslovakia; it has precluded OAU mediation in civil wars, such as the Ogadon war in Nigeria; and it has led to the anomalous situation of OAU envoys attempting to resolve a successionist struggle while opposing the main demand of one of the protagonists, as occurred in the Federal Islamic Republic of the Comoros in 1998 (see Organisation of African Unity 1998:11).

Mediators deployed by international organisations labour under an onerous burden when they have rigid mandates and are kept on a tight rein. To all intents and purposes they are the spokespersons or negotiators of the host agency rather than mediators at the service of the disputants. They are unable to adapt quickly to changing circumstances and to shifts in the adversaries' bargaining positions, and their predicament becomes intolerable if the agency interferes with their endeavours while mediation is underway. In short, their relationship with their principal inhibits the development of a relationship of trust with the parties.

Arguing that these problems are inherent limitations of inter-governmental bodies, Touval (1994:45-6) concludes that the UN should refrain from mediating in complex international disputes and rather sponsor "unilateral mediation by great powers or other states who have a vested interest in conflicts within their sphere of influence". This is a strange conclusion as Touval acknowledges that third party countries may be unwilling to play that role, that disputants may be wary of such meddling, and that they sometimes choose a multilateral mediating body precisely to avoid a state mediator attempting to force on them an undesired settlement. Touval also ignores the extent to which the interests of powerful states can reduce their flexibility as mediators, and he underestimates the positive peacemaking role that UN secretaries-general and their envoys have played on numerous occasions (see Rivlin and Gordenker 1993).

In many situations the better structural solution would lie in the UN secretary-general undertaking mediation independently of the institution's deliberative organs, a strategy supported by Boutros-Ghali (1992:22-23). This approach would meet the dual imperatives of legitimacy and flexibility. Cyrus Vance and David Hamburg (1997) recommend strengthening the secretary-general's authority and capacity to utilise personal envoys and other representatives, without being undermined by second-guessing from Security Council members, as a low-cost and low-risk means of averting and ending crises. They note dryly that resistance from the US and other governments to such proposals has "more to do with the relationship of the Security Council to the UN Secretariat than with the prevention and resolution of deadly conflict" (Vance and Hamburg 1997:7). This option would have the further benefit, motivated below, of separating the secretary-general's mediation role from the Security Council's authority over enforcement measures.

Mediators must not apply punitive measures

Although the question of punitive action has already been canvassed, it requires further comment because one or more of the parties to a civil war is typically resistant to mediation and engages in conduct deemed unacceptable by the international community. Where oppression, abuse of human rights and other violations of international norms are features of an intra-state conflict, external actors might seek to apply pressure on the offending party through enforcement measures. As in the case of apartheid South Africa, sanctions can help to weaken an authoritarian regime to the point that it becomes receptive to negotiating a democratic settlement. In other situations the aim might be to impel a belligerent, such as Unita in Angola, to adhere to the terms of a peace agreement it has signed.

Further comment is required also because many scholars and diplomats believe that effective international mediation depends on the political power and leverage of the mediator. William Smith (1985:367), although citing only one successful case in support of this view, concludes that "it's therefore logical to expect that great power mediation will be generally more successful than small power mediation". Bercovitch (1996) draws the same conclusion while making contradictory assertions about the utility of `carrots and sticks' and the non-coercive nature of mediation. Saadia Touval (1994) maintains that the UN's lack of political leverage contributes to its ineffectiveness as a mediator, its threats of punishment and promises of assistance having little credibility because the institution has no readily accessible military or economic resources of its own. According to Vance and Hamburg (1997:14), envoys and representatives of the UN secretary-general should be familiar with "techniques to pressure parties to negotiate (e.g. sanctions or threats of force)".

This position is flawed in several respects. First, mediators can achieve a great deal where they have no formal political power and their credibility emanates instead from moral stature and unquestionable integrity. Examples here include the accomplishments of the World Council of Churches and the All African Council of Churches in Sudan in 1971-2 (Assefa 1987); of the Community of Sant' Egidio in Mozambique in 1990-2 (Romano 1998); and of representatives of the UN secretary-general in a variety of conflicts (Rivlin and Gordenker 1993). Father Romano (1998) attributes Sant' Egidio's achievement largely to the `weakness' of non-governmental mediating bodies that lack the capacity to threaten the disputant parties.

Second, the success of enforcement measures is hardly inevitable. A study of economic sanctions imposed between 1914 and 1990 found that the primary foreign policy goal was attained in barely one third of the cases (Hufbauer et al 1990, cited in Stremlau 1996:9). Aside from the practical difficulties of implementation, sanctions and the threat or use of force do not easily deter groups that believe they are fighting for their survival. Punitive action may even be counter-productive. The experience of ECOMOG in Liberia, for example, illustrates how peace enforcement operations can broaden, deepen and prolong hostilities (Nyakyi 1998; and Howe 1996/7). In the case of Burundi, sanctions have undermined Tutsi confidence in reconciliation, strengthened extremist positions within the army and the minority community by heightening their sense of persecution and vulnerability, and exacerbated poverty and inequality of wealth which are counted among the root causes of the conflict (International Crisis Group, 1998).

Third, general claims about the utility of leverage ignore a range of distinctions with respect to the nature, timing and purpose of external intervention. Consider, for example, a powerful state terminating military aid to a belligerent, offering a party financial inducements to abandon hostilities, attempting to bully it into peace talks through sanctions, providing resources to facilitate negotiations and the implementation of a settlement, and acting as a guarantor in respect of agreements reached. These interventions, all covered by the term `leverage', have such different strategic and psychological import that it is misleading to regard them as examples of a single category.

The problem is especially evident in the notion of `carrots and sticks' as a synonym for leverage (e.g. Touval 1994:55; and Vance and Hamburg 1997:4). The idiom implies that promises to reward a party for co-operation and threats to punish it for intransigence are similar strategies, when it would seem to be a matter of commonplace observation that individuals and groups respond to coercion and to offers of assistance in manifestly dissimilar ways. The distinction that should be drawn here is captured by Boutros-Ghali (1992:23) in referring to the mobilisation of UN resources in support of peacemaking ventures as "positive leverage" (emphasis added).

Fourth, and most importantly, a mediating body will almost certainly be mistrusted by a disputant against whom it threatens or applies sanctions or military force. As argued earlier in respect of Somalia, Liberia, Burundi and Sudan, it loses credibility as a peace broker and becomes a party to the conflict. Giandomenico Picco (1994) contends similarly that when the institution of the UN secretary-general is involved in determining and managing the use of force, it forfeits the strength that it derives from having no vested interests of its own and compromises the impartiality that is critical to its role as a mediator; from the perspective of suspicious combatants, the authority to order killing renders the institution no different from major powers.

This is not to say that punitive action is intrinsically invalid. Rather, the point is that where it is deemed necessary, it should be undertaken by an agency other than the active or prospective mediator. My colleague Kent Arnold (1998:15) describes the necessity for an appropriate division of labour in peacemaking initiatives as a matter of "role integrity, which prevents actors from assuming contradictory roles which can impede or threaten the peace process". Picco (1994) thus proposes that the UN's management of force should not lie with the secretary-general but be sub-contracted to a military alliance or coalition of states, as occurred in the Gulf War; this would allow the secretary-general to play the `good cop' negotiator, with the Security Council playing `bad cop' if negotiations fail.

Some division of labour is probably inevitable in complex intra-state conflicts because the activities associated with mediation are too many and too varied to be conducted by the same agency (Mitchell 1993). The Mozambican peace process provides a good example of these activities being performed by different actors: partisan states put pressure on their allies to engage in talks (Zimbabwe in respect of Frelimo, and Kenya in respect of Renamo); a non-partisan religious body facilitated the talks (the Community of Sant' Egidio); powerful states provided resources and played the role of guarantor in respect of agreements reached (Italy, Portugal, the United States and the United Kingdom); and the United Nations oversaw the implementation of the peace plan.

Mediation is an art and a science

In CCR's experience, effective mediators have a particular combination of personal traits. For example, a high level of empathy and sensitivity is required to maintain the trust of the disputants and discern the issues that underlie their formal demands. Mediators must also have sufficient confidence to maintain control of meetings when tempers flare, avoid being bullied by powerful parties, and absorb the anger that may be directed at them when progress is slow. At the same time, they must be able to keep their egos in check and refrain from becoming too assertive or too anxious to achieve a settlement. Flexibility and creativity are prized because conflict is dynamic and no two cases are identical.

These attributes contribute to the art of the mediator. Yet while mediation is not a mechanical endeavour, neither is it a mystical and idiosyncratic affair as suggested by Arthur Meyer (1960:161, quoted in Bercovitch 1996:247): "[The] task of the mediator is not an easy one. The sea that he sails is only roughly charted and its changing contours are not clearly discernible. He has no science of navigation, no fund inherited from the experience of others. He is a solitary artist recognising at most a few guiding stars and depending on his personal powers of divination".

This passage captures the `loneliness of the long-distance mediator' but it is the antithesis of the present author's stand. Four decades after Meyer wrote, mediation can be regarded as a professional discipline in the sense that it encompasses a body of theory, comparative research, case studies and tested techniques. If the principles outlined above reflect the strategic dimensions of mediation, then the techniques are the tactical elements that constitute the essence of the profession. They relate to diagnosing the causes and dynamics of the conflict; analysing the interests of the protagonists; undertaking `shuttle diplomacy' where the parties refuse to talk directly to each other; designing and convening the mediation process; setting the agenda and conducting meetings; identifying common ground between the parties; generating options for meeting their needs and evaluating the merits and durability of these options; overcoming deadlocks; and determining means of monitoring adherence to a settlement and dealing with breaches thereof.

Skilled mediators attach particular importance to methods of facilitating `good listening' since adversaries are typically preoccupied with their own positions and filled with emotions that inhibit effective communication. Moreover, the parties become increasingly frustrated and obdurate if they feel that their concerns have not been heard. Acknowledgement of these concerns by an opponent helps to break down negative stereotypes and establish a measure of trust. This in turn provides a platform for the parties to address the causes of the conflict and to adopt more conciliatory postures. In short, professional mediators use a variety of tools to shift the parties from an adversarial `winner-takes-all' orientation to a more co-operative problem-solving approach.

In contrast to the above, states and multi-national bodies do not appear to view mediation as a specialised activity. Major policy statements on peace and conflict by UN secretaries-general invariably present a perspective on early warning, military deployment and other topics but say little about mediation beyond asserting its importance (e.g. Boutros-Ghali 1992 and 1995; and Annan 1998). Whereas discussion around humanitarian aid, sanctions and peace operations occurs within and between academic, political and activist circles, debates on mediation are confined mainly to scholars and professional mediators; one of the debates concerns the question of professionalising domestic mediation (see Morris and Pirie 1994), but the question is not posed in respect of international mediation. In practice, as illustrated in the previous sections, the principal methods of international mediation are persuasion, bargaining and the exercise of leverage. No clean distinction is drawn between mediation and enforcement, with the result that mediation is reduced to power-based diplomacy. It follows that mediators are appointed on the basis of their political stature rather than their experience and competence as mediators.

It is not the intention here to fetishise mediation or professional occupations, nor to suggest that there is any single correct approach to peacemaking. Rather, the aim is to highlight the emphasis that professions place on specialist training and qualification (see Greenwood 1957). Mediation of intra-state conflict is usually pursued in situations of imminent or actual violence where the stakes and risks are high. It seems absurd that states and inter-governmental bodies which would not deploy untrained soldiers, police or doctors in these situations, or in any other circumstances for that matter, are willing to utilise untrained peace brokers.

Since the stature of a mediator is an important factor in the context of intra-state crises, it is understandable that heads of state and senior diplomats are entrusted with this task. In some instances they may be effective because of their inter-personal skills, the attractiveness of the `carrots' offered to the parties and/or the ripeness of the conflict for resolution. Yet the success rate might be higher if the mediators were proficient in mediation techniques. It is not possible to adduce hard evidence in support of this proposition since failed peacemaking ventures cannot be replayed in real life with a different mediator or style of mediation. Nevertheless, African diplomats have expressed discomfort at their lack of expertise and confidence when engaged in complex mediation, and a number of ambassadors and foreign affairs officials have raised the need for comprehensive training in mediation and related skills ².

In addition to meeting this need, units comprising expert mediators could be established in the offices of the UN and the OAU secretaries-general. Their functions would include fact-finding, monitoring, analysis, confidence-building, mediation, and early warning of both emerging conflicts and opportunities to resolve long-standing ones. Where adversaries are unwilling to engage in negotiations, the mediators would seek to facilitate dialogue indirectly through shuttle diplomacy. They would also serve as technical advisers on preventive diplomacy and peacemaking to the secretaries-general and their staff, to member governments on request, and to heads of state and senior diplomats appointed to mediate in high-profile cases. As standing entities, the units would ensure greater continuity, consistency and knowledge of specific situations than occurs with ad-hoc missions.

A host of benefits might accrue if the units functioned independently of the other organs of the UN and the OAU. The mediators could be deployed long before a conflict reaches crisis proportions and attracts the attention of these organs; while obviously constrained by international norms, their flexibility would not be restricted by formal resolutions and the vested interests of member states; and they could more easily make contact with regimes and rebel movements that had acquired pariah status. Most importantly, their impartiality, low-profile and lack of coercive power might render their efforts more acceptable to the parties and less likely to be viewed by governments as interference in domestic affairs.

Conclusion

The experience of peacemaking in African civil wars suggests that international mediators are ineffective, if not counter-productive, when they deviate from the logic of mediation and apply undue pressure on the disputants. The mediators' motives may be sound but their approach is deeply flawed. Individuals and groups tend to resist coercion under most circumstances. This is especially the case where they are in conflict over issues related to freedom, justice, identity, security and survival, and where the conflict has engendered a high level of enmity and fear. While external pressure may be unavoidable because of a disputant's intransigence or aggression, a mediator whose conduct is threatening to a party will lose that party's trust and inhibit the resolution of the conflict.

Before concluding this argument, it is necessary to dismiss three possible inferences that would not be justified. First, it would be incorrect to infer from the critique that the mediators were responsible for the lack of success in the cases cited above, or that a different style of mediation would definitely have borne fruit. Establishing the reasons for failure in each case would entail a more thorough investigation than has been presented in this paper. Moreover, even the most experienced mediator can do little if one of the parties has no inclination to reach a settlement. Nevertheless, the more modest point is that an inexperienced mediator can blow opportunities for progress and exacerbate the conflict.

Second, comparing international mediation with the approach of a non-governmental organisation (NGO) does not imply that NGOs engaged in peacemaking are necessarily competent and effective. Like governments, they can behave unethically, lack respect for local communities, and intervene in complex situations with a superficial analysis and misguided strategy. Such problems arise in particular, and are most serious, when NGOs operate outside their own country and cannot easily be called to account for their mistakes and misconduct (see Sorbo et al 1997). Their potential to cause harm may be less than that of states but it exists nonetheless.

Third, states and multi-national bodies face challenges substantially greater in scale and intensity than those typically addressed by NGOs. Whereas the latter tend to pursue peacemaking at a sub-national level, the former seek to manage intra-state conflict that has escalated to crisis proportions. International actors have to navigate a precarious course between a range of conflicting pressures: the humanitarian imperative of preventing and ending mass slaughter; resistance to external intervention from some of the adversaries; and the diverse interests of neighbouring and other countries. These factors do not diminish the validity of the mediation model outlined above but they pose significant obstacles in applying it.

Because the dynamics of international mediation derive in large measure from the organisation of the international system around states and power, it could be argued that the problems described in this paper are inescapable. It might therefore be concluded that peacemaking by states, inter-governmental bodies and non-state actors should be regarded as complementary, suitable in different settings or appropriate at different stages in the resolution of a conflict. This is a widely held view that has contributed to the proliferation and legitimacy of `multi-track' initiatives in recent years (e.g. Rupesinghe 1997; and Sorbo et al 1997:1-4).

The thrust of this paper, however, is that mediators diminish the prospect of ending conflict when they deviate from the principles of mediation and are unfamiliar with its techniques. Mediation is a specialised activity that is not a mystical affair, reducible to common sense or synonymous with power-based diplomacy. Assuming good faith on the part of the mediator, strategic and tactical errors are not inevitable. The stronger conclusion, then, is that international actors should acquire greater proficiency in the art and science of mediation. This could be achieved at little expense through comprehensive training; by deploying qualified mediators alongside prominent personalities involved in peacemaking; and by establishing expert mediation units within the UN and the OAU.

REFERENCES

Annan, K. 1998. `The causes of conflict and the promotion of durable peace and sustainable development in Africa', Report of the United Nations Secretary-General, 16 April.

Arnold, K. 1998. `Cast of thousands: managing interveners' role integrity and role co-ordination'. Track Two 7(1): 15-17, Centre for Conflict Resolution.

Assefa, H. 1987. Mediation of Civil Wars: Approaches and Strategies - the Sudan Conflict. Boulder and London: Westview.

Bercovitch, J. 1996. `Understanding mediation's role in preventive diplomacy'. Negotiation Journal 12: 241-258.

Boutros-Ghali, B. 1992. An Agenda for Peace. New York: United Nations.

Boutros-Ghali, B. 1995. `Supplement to An Agenda for Peace: position paper of the Secretary-General on the occasion of the fiftieth anniversary of the United Nations', A/50/60 - S/1995/1, 3 January, United Nations.

Crocker, A. 1992. `Conflict resolution in the Third World: the role of the superpowers'. In: Brown, S.J. and Schraub, K.M. (eds). Resolving Third World Conflict: Challenges for a New Era. Washington, DC: United States Institute for Peace, pp. 195-210.

Deng, F. 1997. `Preventive diplomacy: the case of Sudan'. Preventive Diplomacy Series 1. Durban: ACCORD.

Friedrich Ebert Stiftung, Life and Peace Institute and Norwegian Institute of International Affairs in co-operation with the Lesson-Learned Unit of the UN Department of Peace Keeping Operations, 1995. `Comprehensive report on lessons-learned from the United Nations Operation in Somalia: April 1992 - March 1995', December, Sweden.

Greenwood, E. 1957. `Attributes of a profession'. Social Work 2 (3): 44-55.

Howe, H. 1996/7. `Lessons of Liberia: ECOMOG and regional peacekeeping'. International Security 21(3): 145-176.

Hufbauer, G., Schott, J. and Elliott, K. 1990. Economic Sanctions Reconsidered: History and Current Policy, 2nd ed. Washington DC: IIE.

International Crisis Group, 1998. `Burundi under siege: lift the sanctions; re-launch the peace process'. Burundi Report 1, 28 April.

IRIN Update for Central and Eastern Africa, No. 405, 28 April 1998. Integrated Regional Information Network for Central and Eastern Africa, United Nations Office for the Co-ordination of Humanitarian Affairs.

Kiplagat, B. 1988. `Is mediation alien to Africa?'. Track Two 7(1): 4-7, Centre for Conflict Resolution.

Menkhaus, K. 1996. `International peacebuilding and the dynamics of local and national reconciliation in Somalia', International Peacekeeping 3(1): 42-67.

Meyer, A. 1960. `Functions of the mediator in collective bargaining'. Industrial and Labour Relations Review 13: 156-165.

Mitchell, C. 1993. `The process and stages of mediation: two Sudanese cases'. In: Smock, D.R. (ed). Making War and Waging Peace: Foreign Intervention in Africa. Washington DC: United States Institute for Peace.

Morris, C. and Pirie, A. (eds) 1994. Qualifications for Dispute Resolution: Perspectives on the Debate. Victoria, British Columbia: UVic Institute for Dispute Resolution.

Nyakyi, A. 1998. Untitled paper. Presented at Learning from Conflict Resolution in Africa: Workshop on the Experience of Individual and Institutional Mediators, Mwalimu Nyerere Foundation and the Tanzanian Ministry of Foreign Affairs and International Co-operation, Arusha, 21-23 January.

Othman, H. 1998. `Summary Report' by the Rapporteur General. Presented at Learning from Conflict Resolution in Africa: Workshop on the Experience of Individual and Institutional Mediators, Mwalimu Nyerere Foundation and the Tanzanian Ministry of Foreign Affairs and International Co-operation, Arusha, 21-23 January.

Organisation of African Unity, 1998. Resolving Conflicts 2 (4), March-April, OAU Political Department, Addis Ababa.

Picco, G. 1994. `The UN and the use of force: leave the Secretary General out of it'. Foreign Affairs 73(5): 14-18.

Puchala, D.J. 1993. `The secretary-general and his special representatives'. In: Rivlin, B. and Gordenker, L. (eds). The Challenging Role of the UN Secretary-General: Making "The Most Impossible Job in the World" Possible. Westport, Connecticut: Praeger, pp. 81-97.

Rivlin, B. and Gordenker, L. (eds) 1993. The Challenging Role of the UN Secretary-General: Making "The Most Impossible Job in the World" Possible. Westport, Connecticut: Praeger.

Romano, A. 1998. `Peace is possible: lessons from the Mozambique peace process'. Presented at Learning from Conflict Resolution in Africa: Workshop on the Experience of Individual and Institutional Mediators, Mwalimu Nyerere Foundation and the Tanzanian Ministry of Foreign Affairs and International Co-operation, Arusha, 21-23 January.

Rupesinghe, K. 1997. `The general principles of multi-track diplomacy'. Preventive Diplomacy Series 2. Durban: ACCORD.

Smith, W. 1985. `Effectiveness of the biased mediator'. Negotiation Journal 1(4): 363-371.

Sorbo, G.M., Macrae, J. and Wohlgemuth, L. 1997. `NGOs in conflict - an evaluation of International Alert'. CMI Report Series 6. Bergen: Chr. Michelsen Institute.

Stremlau, J. 1996. `Sharpening international sanctions: toward a stronger role for the United Nations'. A Report to the Carnegie Commission on Preventing Deadly Conflict, Carnegie Corporation of New York.

Touval, S. 1994. `Why the UN fails'. Foreign Affairs 73(5): 44-57.

United Nations 1992. Handbook on the Peaceful Settlement of Disputes between States. New York: United Nations.

Vance, C.R. and Hamburg, D.A. 1997. `Pathfinders for peace: a report to the UN Secretary-General on the role of special representatives and personal envoys'. A Report of the Carnegie Commission on Preventing Deadly Conflict, Carnegie Corporation of New York.

Van Eck, J. 1997. `Brokers in Burundi: broadening the Nyerere mediation initiative'. Track Two 6(1): 17, Centre for Conflict Resolution.

-------------------------------------------------------
NOTES

1. The Centre was established by the University of Cape Town in 1968. Its mission is to contribute to a just and sustainable peace in South Africa and other African countries by promoting constructive, creative and co-operative approaches to the resolution of conflict and the reduction of violence. CCR's approach to conflict and mediation has been influenced, in particular, by the writings of John Burton, Johann Galtung, John Paul Lederach, Manfred Max-Neef and Chris Moore.

2. The discomfort noted here was raised with the author in off-the-record comments by Tanzanian officials who facilitated the Arusha peace process for Rwanda in 1993-4. As a result of their remarks, in 1997 CCR established a conflict resolution training programme for senior African government officials. The participants frequently request follow-up training and propose that such programmes be included in the formal training of foreign affairs staff.