A Practical Guide to Mediation in Peace Processes

Any serious guide to mediation in peace processes has to begin with a sobering fact: most mediation efforts are improvised. Research on European engagement in peacemaking, carried out between 2007 and 2009 by the Crisis Management Initiative and its partners, concluded that approaches to peace mediation were largely ad hoc, with little systematic learning drawn from decades of practice. That finding still shapes how practitioners think about the craft today. Mediation works, but it works far better when it is prepared, staffed and evaluated like the professional discipline it has become.

This guide sets out what mediation in peace processes actually involves: the main models, the mediation process steps from conflict analysis to implementation, the division of labour across diplomatic tracks, and the qualities that separate effective mediators from political figureheads.

What mediation is, and what it is not

In the interest-based tradition, mediation is a voluntary and confidential structured process in which one or more impartial third parties assist conflict parties in reaching a mutually satisfactory solution. The mediator provides a framework but makes no substantive decisions. Two features distinguish it from other third-party interventions: it is not based on force, and the parties keep a degree of ownership over the outcome.

Mediation is often confused with dialogue, and the two overlap in practice. The distinction matters for design. Mediation aims at an agreement; dialogue aims at understanding.

“Unlike mediation, in which the goal is usually reaching a resolution or settlement of a dispute, the goal of dialogue is usually simply improving interpersonal understanding and trust.”

Dialogue processes, which are open-ended and facilitated rather than directed, build the confidence and relationships that later allow formal talks to hold. A longer treatment of how the two strands interact, officially and unofficially, appears in this analysis of mediation and dialogue processes.

Three models of peace mediation

Comparative research distinguishes three principal approaches, and the choice between them should follow from analysis of the conflict, not from the mediator’s habits.

• Power-based, deal-brokering mediation. The mediator brings sticks and carrots to bear, directs the process resolutely and pushes the parties toward a deal. The 1995 Dayton Agreement is the classic case. This model can deliver fast results, but parties often resent it and outcomes can prove brittle.
• Interest-based, problem-solving mediation. The mediator facilitates, promotes party ownership and generates creative options that satisfy underlying interests. The 2005 Aceh peace agreement between Indonesia and the Free Aceh Movement (GAM), brokered by Martti Ahtisaari with the Crisis Management Initiative, is the standard reference. Technical experts on land reform, power-sharing or reintegration of combatants are often brought in on parallel tracks.
• Transformative, long-term mediation. The mediator intervenes at several levels of society to change relationships and perceptions over years, not months. This is where sustained donor funding of local peace work matters most.

Mediation process steps, from analysis to implementation

Mediation is best understood as one moment in an extended peace process. In practice the mediation process steps run roughly as follows.

1. Conflict analysis and early warning. Effective mediation starts before the table is set. Analysis needs to be mediation-sensitive: it should identify entry points, potential spoilers and windows of opportunity, not just root causes. Research on the Nagorny Karabakh conflict found that a genuine opportunity for mediation was lost at the very start of the Minsk process, partly because international engagement was neither focused nor united. The lessons from the Karabakh mediation efforts remain a caution against slow starts.
2. Preparation and consent. The parties must accept the process and, ideally, the mediator. Preparedness is the first fundamental listed in the United Nations Guidance for Effective Mediation, issued in 2012 and available through the UN’s Peacemaker platform, alongside consent, impartiality, inclusivity, national ownership and coherence with international law.
3. Framework and ground rules. Confidentiality arrangements, agenda, venue, participation and the sequencing of issues are settled before substance is touched.
4. Negotiation and option generation. The mediator manages communication, tests options and brings in technical expertise on sticking points such as security arrangements, wealth-sharing or the reintegration of former fighters.
5. Drafting the agreement. Justice questions deserve particular care. Experience shows that accountability for war crimes is rarely fully settled at the peace table, so skilled mediators anchor general principles and commitments in the text while leaving detail open for future public consultation. Badly drafted provisions can block justice initiatives for years.
6. Implementation and monitoring. An agreement without a guarantee mechanism invites relapse. In Aceh, the European Union’s monitoring mission deployed in September 2005, immediately after the agreement, to supervise disarmament and troop withdrawal. The ability to monitor what was signed is itself a source of credibility for the mediator.
7. Evaluation. Seven parameters adapted from the OECD’s Development Assistance Committee guidance can be applied to mediation: relevance, effectiveness and impact, sustainability, efficiency, coherence, coverage of actors, and consistency with the values of mediation itself. Systematic comparison of successful and failed cases is still the exception, and the field is poorer for it.

Tracks: who mediates at which level

Peacemaking runs on multiple tracks. Track I covers high-level political talks between the conflict parties, usually led by states or multilateral bodies. Track II involves influential individuals and organisations that can shape Track I dynamics without formal standing. Where officials join unofficial formats, practitioners speak of Track 1.5. John McDonald and Louise Diamond expanded the scheme in 1996 into a nine-track model of multi-track diplomacy, running from business contacts to religious networks.

Unofficial tracks carry real advantages. Non-governmental mediators can engage armed groups and unrecognised parties without conferring legitimacy, which official mediators struggle to do. They can also reframe talks around human security rather than state-centred security. Private diplomacy organisations, from the Centre for Humanitarian Dialogue to Sant’Egidio, have carried significant parts of recent peace processes, a division of labour examined further in this account of private diplomacy in crisis management.

What makes an effective mediator

Impartiality is the guiding principle, though the record is more complicated: a large share of international mediation has in fact been conducted by biased mediators who were accepted because they could deliver one side. Status matters too. State mediators are often perceived as closer to state parties and can find less acceptance among non-state actors.

The professional consensus that emerged from European practice in the late 2000s was blunt about the gap between political appointment and mediation skill.

Political figureheads were being chosen for empathy and seniority, researchers found, while “professional and technical expertise in mediation are not considered decisive factors”, with some envoys tempted to impose a solution simply to move on to the next conflict.

The emerging remedy separates two roles: a high-level mediator with convening power, and a mediation support team of process and thematic experts. The Mediation Support Unit in the UN Department of Political Affairs, now part of the Department of Political and Peacebuilding Affairs, pioneered this model with standby teams of specialists. The personal attributes that matter, patience, discretion, creativity under pressure, are explored in more depth in this piece on what makes a good mediator, drawing on Nobel laureate Martti Ahtisaari’s practice.

Common pitfalls

Three failure patterns recur. First, competition among mediators: peacemaking is a crowded field, and when several channels operate at once, belligerents engage in forum shopping, playing mediators off against each other until none of them retains real influence over the parties. Coordination, as the France-Russia-US troika eventually brought to the Karabakh talks under the OSCE, is the counter-measure. Second, missed early windows: tensions are cheapest to defuse at the start, yet early-warning signals are routinely ignored. Third, agreements without guarantors: parties who fear exploitation of their concessions cannot credibly commit without third-party guarantees, which is why mediation should be planned together with monitoring and broader conflict prevention policy from the outset.

Frequently asked questions

What is mediation in a peace process?

Mediation is a voluntary, confidential and structured process in which an impartial third party helps conflict parties reach a mutually acceptable agreement. The mediator manages the process and communication but does not impose decisions, and the parties retain ownership of the outcome.

What are the main mediation process steps?

The core steps are conflict analysis and early warning, securing consent and preparing the process, agreeing a framework and ground rules, negotiation and option generation, drafting the agreement, implementation with monitoring, and evaluation. The steps overlap in practice and rarely run in a straight line.

What is the difference between mediation and dialogue?

Mediation aims at a settlement: the third party leads a process designed to produce an agreement. Dialogue is open-ended communication facilitated to build mutual understanding, empathy and trust. Dialogue often prepares the ground for mediation and sustains agreements afterwards.

What are Track I and Track II mediation?

Track I refers to official, high-level political negotiations between conflict parties, typically involving states or multilateral organisations. Track II involves unofficial actors such as NGOs, academics and religious networks who influence the process without formal standing. Mixed formats are called Track 1.5.

Does a mediator have to be neutral?

Impartiality is the guiding principle, and perceived bias limits what a mediator can achieve. In practice, however, a large share of international mediation has been conducted by biased mediators who were accepted because they had influence over one of the parties.

What is interest-based mediation?

Interest-based mediation focuses on the underlying interests of the parties rather than their stated positions. The mediator facilitates rather than directs, promotes ownership of the process and helps generate creative options. The 2005 Aceh agreement is a frequently cited example.

What is power-based mediation?

Power-based mediators use rewards and pressure to broker a deal, directing the process firmly. The Dayton Agreement that ended the Bosnian war in 1995 is the standard example. The model can produce quick results but risks resentment and fragile implementation.

What is the UN Guidance for Effective Mediation?

Issued in 2012 by the UN Secretary-General, the Guidance for Effective Mediation identifies eight fundamentals for professional mediation: preparedness, consent, impartiality, inclusivity, national ownership, international law and normative frameworks, coherence of effort, and quality peace agreements.

Why do multiple mediators cause problems?

When several mediation channels operate without coordination, conflict parties can engage in forum shopping, choosing whichever channel seems most favourable at the moment. This weakens every mediator’s bargaining position. Coordination structures, such as a joint chairmanship, reduce the risk.

How are peace mediation efforts evaluated?

A practical framework adapts seven OECD-DAC criteria: relevance, effectiveness and impact, sustainability, efficiency, coherence, coverage of actors involved, and consistency with mediation values such as confidentiality. Systematic evaluation remains rare, which slows learning across cases.

When should mediation start in a conflict?

As early as possible. Initiating mediation and dialogue while tensions are still escalating has considerable potential to defuse a conflict before positions harden. Several studied cases, including Nagorny Karabakh, show how missed early opportunities can lock in decades of stalemate.

What is a mediation support team?

A mediation support team pairs a high-level mediator with specialists in process design, thematic issues such as security arrangements or transitional justice, logistics and research. The UN’s Mediation Support Unit and its standby team of experts are the best-known model.