Mediation, as it is defined here, is a form of bargaining in which the parties to a dispute seek the assistance of or accept an offer of help from, a party not directly involved in the conflict to resolve their differences without invoking or applying the authority of the law.[5] The transformation of a dyadic negotiation system into a triadic, mediated-negotiation system[6] results in numerous possibilities for mediators to influence the relationship and communication with and between the parties. The mediator also brings to the conflict management process a range of additional material and non-material resources, such as reputation, skills, and influence. Mediation, in other words, can "expand the pie," thereby creating inducements for peace. Mediation should not be conceived of as a uniform process or a set of discrete activities. Rather, mediation involves a continuous set of related processes involving actors, relationships, decisions, resources, and settings. Depending on the wishes of the disputing parties, it can be conducted formally or informally, in secret or in the open, by top leaders or low-level representatives, and with closed or open-ended agendas. It is therefore, a highly flexible form of voluntary, non-binding, decision-making.

The "post-modern" or new wars of the post-Cold War period pose unique conflict resolution challenges, not least because they frequently involve accompanying complex emergencies. There is an emerging consensus that managing or resolving internal conflicts involves multi-dimensional, long-term, coordinated approaches, and that there are key roles for peace-making activities, such as mediation, types of peacekeeping, and post-conflict peace-building.[20] Cultures of peace and reconciliation needs to be constructed at the levels of the power brokers (the top-level leadership), the power holders (the middle-range leadership), and civil society (grassroots leadership) for long-term stability and reconstruction to take place.[21] A bargain among the power brokers held together by inducements from powerful external actors is not on its own enough to reconfigure the underlying logic of internal conflict systems.

Mediation among the power brokers, therefore, should be seen as only one element in a coordinated approach to conflict resolution. Operating at only one level of society – the level of the power brokers, or the highest political and military leaders – it must be undertaken in such a way that it complements rather than complicates parallel processes at the subsidiary levels and the peace-building process that follows. Nonetheless, it is often a vital component of the overall process, since without an end to the large-scale, organized, and systematic violence other peace-building activities – peacekeeping, reconstruction, reconciliation – are all put at risk of failure. In other words, international mediation needs to be re-contextualized in an overall strategic plan or approach. This course implies that diplomatic mediation also needs to be coordinated with other concurrent activities, such as peacekeeping, reconciliation, national reconstruction, and activities antecedent to the mediation, such as peace-building. A coordinated approach also extends to avoiding the all too common situation of multiple, "crowded" mediation tracks, where a number of diplomatic actors each pursue their own (sometimes competitive) initiatives. More than one mediation track (at the top level) allows the protagonists to play the competing mediators off each other. Sometimes, new mediators need to be resisted, or at least convinced to support the primary track.

It is not enough, however, to re-contextualize mediation as part of a more holistic peace-building process. Rather, mediation itself needs to be re-conceptualized and re-formulated. In particular, three underlying orientations need re-visiting. First, the notion of mediation as primarily a diplomatic activity needs to be challenged. Diplomats are not the only possible mediators and limiting possible mediators to only one category of actors – recognized state and intergovernmental diplomats – is no longer judicious or appropriate. International mediation needs to mainstream and incorporate non-official mediators into its major peace initiatives, particularly those with a sustained presence in the zones of conflict and established communication channels with the protagonists. Local actors, such as churches or humanitarian agencies, are particularly important in this regard, as they are often trusted and respected for the relief work that they undertake. Additionally, partnerships between local and external mediators need to be fostered; it was the combined efforts of the local Catholic Church and an external faith-based community that precipitated the break-through in the Mozambique peace process.

Second, mediation needs to move from its obsession with negotiated out comes to placing equal importance on the process and the relationship of the parties. As IDEA has observed, "Bad process will greatly impede agreement. It can even contribute to ultimate failure, no matter how well designed the outcome, simply because the way in which the talks were structured may cause friction and distrust and leave at least some parties questioning the legitimacy of the whole venture."[22] Processes are well established and accepted in interstate diplomacy, but they are not so well accepted – and rarely even articulated – in internal conflict resolution. In internal conflicts building a process that is perceived as fair, just, and sustainable is probably as, if not more, important than achieving a satisfactory outcome. After all, unlike interstate mediation where the parties do not necessarily have to live in close proximity to each other, in internal conflicts the parties may have to share power and manage an ongoing daily relationship. Building a relationship therefore is critical to long-term success. Part of constructing an effective and sustainable process involves devoting more care and resources to the pre-negotiation phase.

Third, international mediation needs to move from its short-term orientation toward a long-term orientation. In particular, focus needs to be sustained well into the implementation phase. All too often international mediation terminates when the parties sign the general peace agreements, since this step is perceived as being the official conclusion of the conflict. The problem is, of course, that the prevailing image of a conflict's official end "creates unrealistic expectations that the announcement of a tentative settlement means that 'it's all over bar the shouting', and fails to prepare constituencies to deal intelligently with the probable cycle of subsidiary disputes."[23] In fact, the implementation of the terms of the agreement can be the most difficult part of the whole process. The parties often have differing interpretations of the terms of settlement, all sides are maneuvering for position in the new dispensation, and there are multiple friction points. An alternative model of mediation – apart from adopting a more realistic mental map of the settlement process – would be to prepare for the implementation phase right from the beginning and put procedures in place to deal with the inevitable disputes that arise along the way.

Support for the efficacy of such a re-conceptualization of mediation can be found in several places, including an examination of the Mozambique peace process. In Mozambique, an alternative model of mediation emerged out of the non-official peace track. Non-official mediators took a long-term and coordinated approach that was as much oriented toward a just outcome as it was to constructing the foundation of a relationship between the parties. Most importantly, the approach worked. Although the peace in Mozambique is fragile, it has not fallen into the all-too familiar pattern of rounds of violence and diplomacy that have characterized similar conflicts in the DRC, Kashmir, Chechnya, Sri Lanka, Palestine, and Liberia to name a few.

Perhaps one of the most important general lessons from the Mozambique peace process is that non-official mediators – NGOs, churches, civil associations, prominent individuals – need to be mainstreamed into diplomatic mediation initiatives. Non-official or private mediators can play important roles in international mediation, including some that official diplomats cannot. In the first instance, employing non-official mediators can circumvent the "entry" problem of mediation in internal conflict.[28] Convening talks between the protagonists can sometimes be the most difficult part of the peace process for the simple reason that the government side may be unwilling to bestow diplomatic recognition on the rebels for fear of legitimizing their cause. In addition, official mediators, whether they represent another state or an intergovernmental organization, are perceived to have self-interested reasons for involvement and may be viewed suspiciously, especially by the non-state parties to the conflict. Entry to the mediation can be facilitated by the use of non-official mediators who do not automatically confer diplomatic status on the non-state side, and who have transparent and clearly benign – usually humanitarian – interests to pursue. Sant'Egidio, a very non-threatening and non-diplomatic actor, with strong humanitarian credentials and who already had a relationship with both sides, proved to be just such a mediator in Mozambique. Moreover, their ties to and partnership with the Catholic Church in Mozambique – an insider intermediary – was crucial to securing entry to the conflict.

Second, establishing informal (deniable), and trustworthy channels of communication is a vital step in the process, especially during the pre-negotiation phase when both sides are feeling each other out. Often, non-official actors like NGOs have a presence on the ground in zones of conflict that gives them access to both sides, and may be the most effective, regular, and reliable channel of communication between the protagonists. Official mediators may find it difficult to even contact rebel leaders, and may be distrusted when they do. In Mozambique, Sant'Egidio developed a relationship of trust and credibility with Renamo through aid and development activities, and a working presence in zones controlled by the rebels.[29] A network of Sant'Egidio members, linked to parishes and missionaries and working among war victims, provided a communication channel of a kind unavailable to most diplomatic actors. At the same time, Sant'Egidio had established important links with the Mozambique government as a result of facilitating contacts between the Holy See and the government of Samora Machel in 1985. While some governments in the region had private channels to Renamo, such as Kenya which was the only government willing to give Renamo travel documents, they were usually viewed with great suspicion by the Mozambique government as being Renamo patrons. From early 1989 to June 1990, as we have said, each side made extensive use of the good offices of both the Catholic Church in Mozambique and Sant'Egidio to establish contact at a non-official level, exchange proposals and ideas without being forced into commitments, and prepare the ground for later substantive talks. Similar efforts by South African and US diplomats in the 1980s had failed.

Third, non-official mediators sometimes have an advantage in gaining the trust of the non-state protagonists in particular, precisely because they are not representative of any state or any state organization. States and intergovernmental organizations tend to stick together and look out for each other's interests, and are often perceived as being biased against non-state actors like secessionist or rebel movements. As Muthiah Alagappa has observed, the problem with intergovernmental organizations becoming involved in internal conflicts is that "their status quo character leads them to favor incumbents. Governments tend to support one another."[30] Low-level, non-official contacts, on the other hand, can build an atmosphere of trust and confidence in the crucially important first phase of the mediation. Non-official mediators are well equipped for facilitating this process as they are non-threatening to either side, lacking as they do any coercive capabilities or potentially threatening agendas. In particular, respected religious figures, like those from Sant'Egidio and Catholic Church officials, can "provide a level of reassurance that official diplomats are often hard-pressed to equal."[31]

Pre-negotiation is broadly speaking "talking about talking," and its importance cannot be overstated.[32] More specifically it may be defined as:

Pre-negotiation begins when one or more parties considers negotiation as a policy option and communicates this intention to other parties. It ends when the parties agree to formal negotiations (an exchange of proposals designed to arrive at a mutually acceptable outcome in a situation of interdependent interests) or when one party abandons the consideration of negotiation as an option.[33]

In other words, it is the period when the protagonists begin to shift from a combative to a collaborative orientation, and start to accept that their goals will be better achieved through cooperation. This change in outlook is a necessary precondition for conflict resolution and represents a profound paradigm shift in the parties' thinking, often greater than anything involved in the final agreed outcome. Mediators have an important, although often unrecognized role, during this phase. Their actions can reinforce the parties' cooperative mentality and help build the foundations for a relationship capable of withstanding the inevitable problems that arise during the substantive negotiations.

Two key lessons regarding mediator's roles during pre-negotiation emerge from the Mozambique peace process. First, mediators may need to engage in negotiator training, particularly of the rebel representatives who may be underprepared and lack the resources for diplomatic bargaining. As early as December 1990, it became clear to the mediators at Sant'Egidio that the Renamo negotiators were struggling with the demands of the negotiation and would need help at several levels.[34] In the early stages of the talks Renamo needed help to define its goals and objectives, a necessity for keeping the negotiations moving forward. In fact, the mediators suspended the talks in January 1991 for this purpose:

[T]he mediators were concerned that RENAMO was unprepared to negotiate the political items, such as the political party law, due for discussion next. RENAMO, while it espoused a commitment to democratic principles, had no experience in negotiating for the kind of practical arrangements and compromises needed to give these policies concrete meaning.[35]

Strong encouragement from the mediators motivated Renamo into preparing a number of documents that defined its position on constitutional issues and on the transition period. For this task Renamo looked to outside legal experts from South Africa and the US for help in conceptualizing and articulating their ideas. The role of helping the parties define their objectives and goals is vital not only to the smooth functioning of the negotiation, but also for the integrity of the process. If one side later feels that it was unprepared and unclear about its objectives, there may be a suspicion that they were taken advantage of, which in turn, could undermine their commitment to the final agreement. Later, of course, Renamo had to go through the necessary and more difficult process of transforming itself from a guerrilla movement into a political party. The difficulties of this process should not be under-estimated and mediators in internal conflicts must remain cognizant of the potential ways in which they can assist the nonstate parties in particular.

Related to this issue, designing an effective framework during the prenegotiation phase involves dealing with the resource asymmetry of the parties. That is,

Good process design entails ensuring that resources on all sides are distributed equitably. That will mean allowing time for preparation, education and familiarization with the process of negotiation. A government with its full-scale administrative capacity, advisors and resources, is obviously at a huge negotiating advantage over a small insurrectionist movement with a handful of lieutenants more familiar with military tactics than political discourse. Suddenly, they must act as a fully fledged political party, when in fact they have had little opportunity to develop such skills.[36]

The mediators in Rome, lacking in material resources themselves, arranged for donors to provide Renamo with the kind of material support it needed for both the negotiations and its transformation into a political force.[37] The Italian government paid most of Renamo's expenses in Rome, while a coalition of states organized for a secure communications link between the Rome negotiators and Renamo headquarters. Later, the UN established a $15 million political trust fund to pay the parties to convert to politics.[38]

Another key area of training lies in helping the parties to communicate more effectively and in a less hostile manner. In deep-rooted internal conflicts, "parties who come to the negotiating table carry with them an abiding experience of conflict, struggle and war. The exercise of force has been their dominant, perhaps only, mode of engagement."[39] This is especially true for non-state actors, who may also simply be unaccustomed to the language of diplomacy. As Andrea Bartoli, a spokesperson for Sant'Egidio, stated:

[T]he essence of Sant'Egidio's work was to find ways to express RENAMO's ideas in terms consistent with the overall goal of reconciliation. Often the language of RENAMO's initial positions was either threatening or uncompromising. The government, accustomed to legal and diplomatic language, had less need of such assistance.[40]

Naturally, this kind of assistance does not apply just to the pre-negotiation phase. Throughout the negotiations, mediators must continue to facilitate communication and improve the quality of the dialogue. However, in the pre-negotiation phase, when trust is still fragile and a cooperative relationship has not yet been established, it is particularly vital that the parties learn to express themselves in ways that will facilitate openness and cooperation, rather than generate more hostility. On a practical level Renamo lacked the normal diplomatic channels of communication between the rebel leadership and the negotiators in Rome. This led to a level of inflexibility on the part of Renamo, who often maintained that they could not proceed until they had re-established contact with their leaders. In the end the mediators had to engage the services of several governments, and even Roland "Tiny" Rowland, the CEO of Lonrho, to create a more efficient and secure line of communication between the Rome representatives and the rebel leadership.

A final area of training that mediators may need to engage in to ensure the integrity of the process is educating the parties about how mediation works and how to employ more effective negotiating tactics. A fascinating instance of such training occurred in November 1991, when Renamo had to be assisted in their negotiating tactics:

Hume explained that when parties agree to a mediation, they are usually willing to work from a mediator's draft, especially after such a draft is already before the parties. If one of the parties then puts forward a competing draft, this action disrupts the negotiations and can be interpreted as a sign that the proposing party has lost faith in the mediation. Such a tactic is legitimate if one wants to get a position across in the strongest way, taking the risk of discrediting the mediation and perhaps breaking off the talks. To use the tactic after only a few days of talks seemed unnecessary [ . . . ] Hume suggested that a more effective tactic might be to insist that RENAMO's points be included in an amended mediator's draft. If RENAMO did not allow the mediators to resume the work of preparing the negotiating texts, it would give the unavoidable impression that RENAMO had decided to obstruct negotiations.[41]

Another key objective of the mediator during the pre-negotiation phase must include building a relationship of trust between the protagonists – and between the mediator and the protagonists – based on increasing levels of cooperation. In the first instance the mediators build trust by adhering to standards of strict impartiality in their dealings with the parties. This step is crucial in internal conflicts, even if it is a contested notion in the context of inter-state conflict.[42] Building trust and creating momentum can also be achieved through log-rolling, a process where agreement is sought on small, relatively easy issues first to create a sense of momentum. Atactic that the Sant'Egidio mediators frequently used to this end was the Single Negotiation Text, which could be passed back and forth between the parties. Small amendments by either side gradually built momentum and a history of compromise that assisted the mediators in obtaining compromises later on the more difficult issues. Often, the mediators would deliberately put aside or delay problematic issues, thereby preventing them from becoming a point of contention or an impasse.

A frequent problem in international mediation is what has been called mediator "crowdedness,"[43] a situation where numerous third parties each pursue their own strategy in a series of disparate and uncoordinated attempts to dominate the peace process. In many cases, mediators' competing agendas provide the parties with the opportunity to play the mediators off against each other, most often with disastrous results for the peace process. The mediation in the Mozambique conflict was remarkable for the degree to which mediator "crowdedness" was avoided through early and continuous consultation, and coordination between different types of mediators – state, UN, and non-official – was maintained throughout the negotiations. Such a sustained and focused mediation by a wide array of mediators has rarely been recorded in any recent internal conflicts.

As early as August 1990, it was agreed that the Sant'Egidio representatives would be the sole mediators, while the other parties with an active interest in pushing forward the peace process – Italy, Kenya, Zimbabwe, the UN, the US – would take on other important non-mediating roles, such as providing hospitality and logistic support, advising either side, supplying experts, and guaranteeing any eventual agreement.[44] In fact, meetings had to be held with Kenyan officials to try and convince them not to mediate, and to allow the Sant'Egidio representatives to take full control of the process. Later, in October 1991, the Kenyans again tried to insert themselves into the mediation process, this time by offering to facilitate direct talks between Dhlakama and US officials. The US resisted this move, stating that they still wanted to "support the negotiations in Rome and had no intention of doing anything to replace the Italian mediation."[45] On 31 October 1991, the US met with the mediators to try and coordinate the gradual involvement of additional governments into the process with the least disruption.[46]

Such continual and careful cooperation between potential mediators is rare. In the former Yugoslavia, for example, multiple diplomatic interventions by the UN, the EU, the Contact Group, NATO, and at one stage, former US president Jimmy Carter, created confusion and opportunities for manipulation by the Protagonists.[47]Avoiding mediator "crowdedness," therefore, is an important goal of any intervention in internal conflicts. At times, mediators need to be as aware of potential interference from other third parties as they are of spoiling tactics by the parties themselves, and may need to forcefully restrain other interested actors from starting their own competing mediation track.

Another important lesson involves coordinating the activities of different types of third parties, usually according to their capabilities and strengths. Throughout the 11 rounds of talks, different actors undertook different roles in an unprecedented display of international coordination. Rarely does one mediator have the necessary resources, abilities, or relationship with the parties to play all the roles needed for keeping the process moving, especially in complex and intractable internal conflicts. The primary role of the Italian government, for example, was to provide hospitality and logistic support to the parties, something the mediators would have been unable to do on their own given their lack of resources. Even Roland "Tiny" Rowland provided vital transportation on his corporate jet for Renamo representatives.[48] The UN and other agencies, such as ICRC, were also brought into the talks at various points to coordinate different levels of assistance, such as ongoing humanitarian assistance. In particular, the UN was given the task of establishing a peacekeeping force and implementation plan for the post-settlement transition period. Finally, a number of states collectively expressed their willingness to act as guarantors of the agreement. This afforded a degree of reassurance to the parties that the mediators on their own – as non-official actors – could not provide, and gave them some measure of confidence in the overall process. In short, the Mozambique case demonstrates the clear potential of a constructive interface between official and unofficial diplomacy, and in the range of approaches to dealing with internal conflicts "the use of track II in seriatim with track I appears to be one of the most useful."[49]

Both the Mozambique peace process and empirical studies suggest that high-ranking officials, like heads of state or rebel leaders, can be effective as impasse-breakers in both negotiation and mediation,[50] yet another important part of coordinating the activities of different kinds of mediators. At various points during the different rounds of negotiations, the US, Kenya, Zimbabwe, and Italy backed up the mediators' efforts by applying pressure on the parties to make concessions. This coordinated use of political and material leverage, which the nonofficial mediators distinctly lacked, complemented the mediators' use of moral leverage. For example, at the 5-7 August 1992 summit when President Chissano and Dhlakama met for their first face-to-face talks, the mediators brought in President Robert Mugabe of Zimbabwe and Italian officials to reassure and pressure the parties into agreeing to a final timetable for concluding the talks.[51] On the negotiation side, the involvement of the two main leaders – Chissano and Dhlakama – helped overcome blockages that lower-level representatives were unable to circumvent.

Another lesson related to the need for coordination and utilizing the strengths of different third parties is that a wide range of experts – in the military, constitutional, electoral, and economic development fields – need to be included in the agreement design phase of the mediation. The US, for example, provided experts in demobilization and ceasefire observation, along with constitutional experts, to help the parties construct workable and effective mechanisms and processes. Often these technical experts would operate parallel to the main mediation. They provided assurances to the parties that the mediators' proposals were workable. Additionally, agreements on technical issues sometimes had a flow-on effect to the main negotiations. As has been suggested, for non-state actors like Renamo expert help was vital in transforming their organization from a guerrilla force into a political party and for translating their loosely formed ideas into clear political objectives.

A key lesson that emerges from the Mozambique peace process is that there needs to be long-term engagement into the implementation and post-conflict reconstruction phases. It is at this point that mediators are most needed, and yet frequently, as in the Middle East, it is at this stage that they are most often absent. As we have said, there was an early commitment by the UN and the wider international community to guarantee the Mozambique pact. On conclusion of the GPA in October 1992, the UN authorized the extensive United Nations Operation in Mozambique (ONUMOZ). Over the next two years, thousands of peacekeepers, aid, and development specialists, and over one billion dollars were invested in relief and reconstruction.

At the same time a key feature of the Rome peace agreement had been the establishment of a "Supervisory and Monitoring Commission" (CSC). Its purpose was to "supervise the cease-fire and monitor respect for and implementation of the agreements between the Parties within the framework of these negotiations."[52] Composed of representatives of Frelimo, Renamo, France, Italy, the UK, the US, the OAU, and the UN, the Commission became one of the key mechanisms for sustaining the implementation process. In particular, Aldo Ajello, the UN's Special Representative in Mozambique, managed to build real trust with the parties and played a critical role at crucial moments in overcoming political obstacles to the implementation of the GPA.[53] Along with Ajello, the CSC actively engaged with the parties in problem-solving, confidence-building, and, when necessary, the direct mediation of disputes over implementation issues. In this way the international guarantors were able to ensure the proper and full implementation of the agreements – unlike the implementation of the Oslo Accords in Israel-Palestine or the Lusaka Accords in Angola.

The sustained engagement of the international guarantors in the process extended up to and even beyond the national elections held in October 1994, two years after the Rome Accords. When Renamo announced that it was withdrawing for 24 hours during the actual balloting, the response was swift and forceful. Special Representative Ajello, Western ambassadors, the heads of neighboring states, and a senior representative of the new South African government all applied the greatest diplomatic pressure on Dhlakama.[54] Almost at once Renamo announced that it was back in and the crisis was averted. It was continual intervention like this that ensured the largely successful implementation of the Rome Accords.

There is little question that during the implementation phase, any number of disputes or problems could have derailed the entire process. The long-term and sustained commitment shown by the international guarantors was a key factor in forestalling such a return to open conflict. In too many other cases – Chad, Angola, Israel-Palestine, Liberia – the failure to sustain any such long-term commitment into the implementation phase played a large part in the failure of peace to take hold. In addition, the kind of post-settlement engagement practiced in Mozambique was also important:

By including a strong humanitarian component in ONUMOZ, the UN established a critical linkage between conflict resolution and peacebuilding in Mozambique. The various programmes – dealing with issues such as the reintegration of former soldiers into civilian life, humanitarian assistance, and the resettlement of refugees and displaced people – were undertaken without looking at the political affiliation of the beneficiaries. This served to reunify and reconcile individuals who had been separated for years, many for no other reason than the fact that they had held different political opinions. By helping to establish relationships between individuals supporting different parties in the conflict, the UN ensured that the social base of the conflict became a strong peace constituency.[55]

Importantly, the decision to demobilize the opposing forces before the national elections was critical to the overall success of the peace process. In Angola, the UN's failure to demilitarize the electoral process led directly to renewed fighting when UNITA lost the elections in 1992. The timely and judicious enactment of a disarmament, demobilization, and reintegration (DDR) program circumvented the possibility of a similar outcome in Mozambique.

A final lesson from Mozambique, oft repeated but nonetheless true, is that there is no substitute for the parties' own willingness to seek a peaceful solution. Without it, all outside interventions are likely to fail. This willingness is thought to come from what has been called conflict "ripeness," a term describing the changes in circumstances that seem to propel the parties toward peace rather than war. It does not imply that a peaceful solution is automatic, but rather that the external and internal conditions are propitious for pursuing certain kinds of peace-oriented policies. The hard work of conflict resolution must still be done with great care and diligence, but under "ripe" conditions it will more likely lead to success than when the conflict is not "ripe."

It is clear that the situation in Mozambique in late 1989 was "ripe" for an attempt at conflict resolution. Several factors had conspired to propel the parties toward a peaceful settlement. First, by the late 1980s, the parties had clearly reached what has been termed a "mutually hurting stalemate,"[56] where neither the government nor the insurgents could attain their objectives through military action. After more than one million fatalities, more than two million displaced people,[57] a devastated national infrastructure, and the end of external support for both sides, exhaustion had set in and the belligerents were compelled to re-evaluate their war strategies and seriously consider a negotiated settlement. The pressure only increased in 1991 and 1992 with a severe drought in the region. Combined with ongoing economic crisis, the drought created a disastrous humanitarian situation in the countryside – Renamo's main zone of operation – and threatened the rebel movement's ability to sustain itself in the rural areas.[58]

Second, a change in the external environment of the war took place. At the wider global level the end of the Cold War resulted in declining superpower sponsorship of ideologically driven conflicts like the Renamo-Frelimo conflict. Regionally, peace was being negotiated in Namibia and Angola, and surrounding states were growing weary of having to deal with the consequences of the war, particularly the vast numbers of refugees and disruptions to regional transport systems. In South Africa, the de Klerk government visited Maputo in July 1989, and agreed to end aid to Renamo, effectively cutting off its main source of external support.[59] At the same time, Mozambique lacked the vast mineral wealth that continued to drive the war in Angola. Rather, it was almost entirely dependent on external aid flows and by the late 1980s, donors were showing greater and greater levels of willingness to use this leverage over the Frelimo government to sue for peace. If Mozambique wanted acceptance as a Western economic partner, peace with Renamo would be have to be the quid pro quo.[60]

Third, there were a number of internal changes within the parties themselves. For example, in the late 1980s a new generation of Frelimo leaders showed greater willingness to contemplate a dialogue with Renamo. There is little doubt that the emergence of the more moderate Joachim Chissano as president of Mozambique after 1986, replacing the hard-line Machel government, also gave impetus to the change in attitude.[61] Similarly, Frelimo as a party underwent a profound transformation after 1989, changing itself from a vanguard party to a mass party and widening its top-level ethnic representation to include groups historically associated with Renamo. It also abandoned Marxism, committed itself to multi-party democracy, and opened the way for direct talks with Renamo.[62] A similar transformation took place in Renamo under the leadership of Dhlakama, and in 1989 Renamo abandoned many of its preconditions for negotiations, unlocking the way for direct talks.

In short, the success of the Mozambique peace process was the "result of a gradually growing and mutually painful stalemate, which was gradually transformed from a ripe moment to a comprehensive set of agreements by a complex array of mediators."[63] The context, in other words, provided the opportunity for peace, but an appropriate model of mediation was needed to translate the opportunity into success.