In most situations of armed conflict, external actors influence the course of the peacemaking efforts. These external actors may or may not have been involved in fuelling the conflict in the first place, or support one or more of the conflict parties. They are likely to include some combination of a wide variety of would-be peacemakers, including neighbouring and regional states, more distant powers or 'helpful fixer' donor states, multilateral, regional and non-governmental organisations, and private peacemakers and individuals. The various incentives and forms of pressure at their disposal can be called upon to reinforce the usually limited powers of influence and resources brought to the table by a mediator. However, that the incentives and pressure may themselves have policy ends somewhat distinct from peacemaking brings with it a new set of problems. It also helps explain why, while coordination of the various external interventions involving incentives, sanctions and conditionalities in peace processes would seem an obvious and uncontroversial goal, in practice it has proven surprisingly difficult.

This article will explore the ways in which coordination or complementarity between external actors can result in a coherent application of policy instruments. Its focus is on the obstacles to and potential for informal mechanisms employed to obtain coordination of diplomatic activity in support of peacemaking. Such mechanisms have flourished in the years since the end of the Cold War, in large part as a consequence of two inter-related factors: the marked upsurge in international conflict management, spearheaded by the United Nations (UN); and the nonetheless significant preclusion of the UN from many peace processes that has encouraged the emergence of other peacemakers (for reasons ranging from suspicion of the influence wielded by powerful members of the Security Council, to a lack of credibility in its ability to implement its own resolutions, or fears that the Council would either be too beholden to government interests, or promote an overly interventionist agenda). Sanctions offer other challenges, as in most cases (the imposition of unilateral sanctions by the United States in circumstances such as Sudan being the exception) they are the direct result of decisions taken by the Security Council or a regional organisation. As is explored elsewhere in this volume, whether their application is successfully coordinated with other actions is a different story.

Individual states and other actors engage in peace processes with widely differing interests, capacity and resources. Motives for engagement include a complicated mix of classic strategic and economic interests deriving from colonial or other ties; concerns regarding regional security and governance; 'softer' interests related to human rights and humanitarian issues; and, particularly in the period since September 11 2001, preoccupation with terrorism and its propensity to flourish in context of conflict or weak and failing states. This broad array of drivers contributed to the emergence of 'peace' as a foreign policy goal for many states and a global surge in conflict resolution activity by multilateral institutions, regional and non-governmental organisations. However, it provides slim grounds for optimism that harmonious 'orchestration' can easily emerge.

Outside a few exceptional cases, the conditions that allow a process to develop under the guidance of a lead mediator able to assume the role of 'conductor' of a coherent peacemaking strategy that includes the support of a group of states are generally lacking. Groups of 'Friends of the Secretary-General' formed to support UN-led peace processes in Central America and elsewhere gained currency in the early 1990s, but such clarity within the peacemaking architecture quickly eroded. Indeed in 1995 Secretary-General Boutros Boutros-Ghali warned in his Supplement to the Agenda for Peace that while the establishment of groups had become a "new trend" in recent years, it was not a panacea. It was necessary, he argued, to maintain "a clear understanding of who is responsible for what", as, if friends took initiatives on their own account, rather than in support of the Secretary-General's lead, there was "a risk of duplication or overlapping of efforts which can be exploited by recalcitrant parties."

The warning carried little weight. Processes in which the UN retained a clear lead were few and far between and, as more peacemakers pressed for involvement, the structures and purposes of the mechanisms formed inevitably grew more diffuse. Moreover, as the cases addressed within this volume illustrate, the conflicts with which the international community grappled in the post-Cold War era were complex, often involving multiple armed actors, each with their own relationships to a fragmented civil society and with supporters and detractors outside the immediate conflict theatre. Multilevel, and multiparty, mediations have become the norm and competition abounds, even amongst those who formally espouse the same ends.

Such situations are rife for exploitation by conflict parties who will understandably seek to extract the maximum advantage from any lack of unity amongst third parties. Except in a very few cases – those blessed with clear leadership, a benign regional environment, conflict parties with identifiable authority and an articulated strategy, the absence of spoilers, and an international community willing and able to bring sustained resources to the table – the development of conditions for the coherent application of sanctions, incentives, guarantees and conditionalities is remarkably difficult.

Coordination problems can be rooted in four broad areas:

• The interests of the external actors may not align, as some international actors may favour the stability of one or more of the conflict parties, their own influence over them, or access to their trade or resources, more highly than the goal of a just and sustainable resolution to the conflict.
• International actors approach a given conflict with widely divergent institutional cultures and funding streams that inhibit the potential for the flexible and responsive measures that a peace process will require. That competing agendas can commonly be found within individual states and organisations as well as between only exemplifies the seriousness of the challenges involved.
• State, multilateral or other international agencies engage in peace processes on the basis of varying experience of the requirements of peacemaking, particularly with regard to knowledge of the conflict parties, a familiarity with peace process design, the patience required for a successful process, and readiness to engage with non-state armed actors whose practices they may abhor.

They consequently may favour policies that embrace both coercive measures, such as sanctions, and incentive-based approaches, promising peacekeepers, humanitarian, technical and other assistance, that together amount not to a coordinated strategy in favour of a peace effort, but a confusing, and perhaps even contradictory basket of actions with unpredictable consequences for the peace effort as a whole.

The stakes involved in the messy situations that develop are high. Incoherence in a mediation effort generally dooms it to failure, while incoherence between the mediation of an agreement and its subsequent implementation, or within implementation itself (when the distinct priorities, competition or flagging attention of donors may supersede a more needs-driven approach) will reduce effectiveness, increase costs and sap the credibility of international actors. In a worst-case scenario, it may also undermine the peace process more directly.

It is a peculiarity of the various informal structures and coordination mechanisms created to further conflict resolution that they are self-selecting. Their existence is, in the first instance, the product of external interest in a conflict. Yet how that interest manifests within a group structure – whose formation involves no hard commitment to the provision of human or financial resources, or a particular set of policy actions – varies greatly. By the mid-2000s, more than thirty 'friend,' 'contact' and 'core' groups – and monitoring or other structures to further implementation of a peace agreement or peacebuilding more broadly – could be identified. The differences between them with respect to their goals, functions and impacts on individual peace processes are marked. It is, nevertheless, possible to distinguish five broad kinds of structure, several of which may be involved in a given process:

• Contact groups have represented vehicles for the direct diplomacy of major powers in a variety of different peace processes. A Contact Group first appeared in Namibia, crafting the plan that became the basis for the Namibian settlement. The Contact Group on the former Yugoslavia was created in 1994, in part to circumvent the UN, and since then has allowed for differences between the states with the most obvious interests in regional stability to be hammered out away from the glare of Security Council attention. A related mechanism is the ad hoc Quartet of the European Union, Russia, the UN and the United States, formed to coordinate international action on the Middle East, but increasingly perceived as a vehicle for the projection of the leadership of the United States.
• Groups of friends, whether of the UN Secretary-General (as in El Salvador, Georgia or Haiti) or a specific peace process, are more informal structures, generally formed to provide support to peacemaking in contexts that elicit a middle level of international attention. Groups of friends may be engaged throughout a peace process, although will fulfil different functions during peacemaking and in helping to implement any subsequent agreement. As with contact groups, friends' engagement with non-state conflict parties has varied in accordance with the level of international acceptability of the latters' demands (ideological, decolonialist, or secessionist), practices (more or less abusive of human rights or identified as 'terrorist'), and the degree of international engagement they have pursued in the conflict and efforts to end it. Related mechanisms include the Core Group formed to support the UN's role in the transition of East Timor, as well as the Troika (Norway, the United Kingdom and the United States) that provided reinforcement to the regionally-led mediation of the north-south conflict in Sudan.
• Friends of a country have also proliferated, particularly within the UN, although often without direct articulation with a specific peace process. They tend to be larger than the groups of friends more operationally articulated with peacemaking, and concentrate their activity in New York. Their purposes have ranged from the sharing of information to attempts to mobilise attention and resources on conflicts further removed from 'high politics.' The impact has often been less than hoped, such as in the cases of the Friends of Angola, the Central African Republic and Guinea-Bissau.
• Implementation and monitoring groups are distinguished by a mandate establishing their responsibilities in a peace agreement, but have varied greatly in the extent to which they are directly engaged in monitoring activities. In most circumstances, mechanisms have followed a model established in Namibia, where a Joint Monitoring Commission was chaired by the representative of the UN Secretary-General and included representatives of the parties to the conflict as well as key external actors.
• Coordination mechanisms for assistance beyond the parameters of the monitoring of an agreement have also proliferated. The Ad Hoc Liaison Committee for Assistance to the Palestinian People was created to support the Oslo Peace Accords, while the Peace Implementation Council in Bosnia brought together a large number of actors to oversee assistance and decision-making after the Dayton agreements. The Co-Chair group of donors for Sri Lanka (European Union, Japan, Norway and the United States) was more modest in scope, established at a moment at which – over-optimistically as it turned out – coordination of assistance for an advancing peace process seemed the priority.

In numerous instances no such group has been formed: efforts to create a friends mechanism for Somalia foundered until an International Contact Group was established in 2006; discussion of a Group of Friends of Darfur to support the talks in Abuja in 2005-2006 came to naught. In other cases a decision was made not to create a group to provide direct support to peace¬making, as in Sri Lanka. The explanatory factors range from a lack of strategic interest on the part of major powers in traditionally 'orphaned' conflicts (Somalia or Burundi), to differences in engagement and under¬standing of the problem at hand (Darfur), or the presence of a regional power, such as India, averse to diluting its influence within a group structure.

Moreover, even in cases when coherent groups have been present, they rarely embrace the totality of the peacemaking effort. As primarily state-centric bodies, their engagement with non-governmental actors or private sector groups pursuing different avenues for peacemaking has been sporadic at best, with opportunities for track two linkage to track one efforts rarely fully explored. This is despite some well known examples of private peacemaking – by the Community of San'Egidio in Mozambique, or the Centre for Humanitarian Dialogue and then the Crisis Management Initiative in Aceh – nurturing agreements whose implementation was subsequently monitored by more formal bodies.

The potential benefits to be gained from the engagement of a small group of states in an ongoing peace process are considerable. In a best case scenario – as seen, for example, in the negotiation of peace agreements in Central America, the role played by the Core Group in East Timor, or the engagement of the Troika in southern Sudan – they bring: leverage, information and practical help to the lead mediator (including through coordination of action in the Security Council as appropriate); legitimacy and influence to the states in the groups; a level of equilibrium, as well as technical and other assistance, to parties to the conflict that may otherwise be characterised by their asymmetry; and attention, resources, and the potential for coherence in the international intervention as a whole.

The circumstances within which this potential has been achieved have, of course, differed widely in accordance with the unique characteristics of each peace process. However, some common elements can be identified.

These include:

• Clear and accepted leadership of the peacemaking initiative;
• A favourable regional environment, represented by significant regional participation with the group mechanism – as was seen, for example, by the role played by Mexico within the Central American groups, or that of Australia and New Zealand in the Core Group on East Timor;
• Conflict parties with a history of engagement with the international community, with the non-state actors in possession of effective leadership, control of territory and/or a defined political agenda;
• A select (four to six states) membership, like-minded in holding the settlement of the conflict as their highest goal; and
• An acute sense of the timing of a mechanism's involvement in a peace process, derived from an understanding that this will determine what a friends or related group may be able to contribute.

Complementarity within a group is critical to its utility. Differing relations with the conflict parties in the successful cases, for example, allowed members of the respective Friend, Core and Troika mechanisms to divide incentives and points of pressure upon the parties between them behind a common vision of what the peaceful settlement of the conflict might look like. Moreover, that vision was one rooted in the demands of the conflict parties themselves, as they had evolved within negotiations: it was encouraged, but not arbitrarily imposed, by outside actors. Late in the day on the negotiations on El Salvador, for example, the Friends worked hard to encourage both parties to accept the recruitment of a significant number of former guerrillas into a new national police force. This was a clear compromise between the guerrillas' original demands for the merging of the two armies and the government's rejection of any such outcome, but also a solution that neither the Friends themselves, nor the UN mediator they supported, would have foreseen or believed possible when the negotiations began a year and half earlier.

Distinction in the roles pursued by different friends was evident in Central America and East Timor, as well as in southern Sudan. In El Salvador and Guatemala, for example, the privileged relationship enjoyed by Mexico with the insurgents, and the United States with the governments, allowed each to exert pressure at key moments of the negotiations. Meanwhile, the Core Group on East Timor was composed of states with specific and quite distinct roles to play. Regional actors (Australia and New Zealand, especially, but also Japan) had legitimate interests in security of their neighbourhood and contributed significant resources to ensure that it be preserved. More distant members of the UN Security Council (the United States and the United Kingdom) welcomed the regional lead and provided diplomatic and other support as appropriate. In Sudan, Troika states were able to work together to calibrate their various interventions and leverage upon conflict parties with whom they had deeply rooted but distinct relationships: the United Kingdom for historic reasons drew on greater knowledge of the north, the sympathies and clout of the United States gave it more leverage in the south, while Norway fell somewhere in between.

Positive results from the involvement of a group structure are not guaranteed. Internal differences or other factors related to a group's membership, most of them deriving from incompatibility in members' interests in a given conflict, can limit its utility in a process, creating sensitivities to be managed and negotiated in addition to those of the conflict parties. In the Georgian/Abkhaz case, differences between the group's European members (France, Germany and the United Kingdom), the United States and Russia have plagued the group of Friends throughout its fifteen-year existence. In other cases groups assume an identity of their own that can sustain the status quo – such as for Western Sahara, where a group of Friends manages action within the Security Council in accordance with priorities distinct from the settlement endorsed by the Council itself. Dynamics beyond the immediate context of a particular conflict (ranging from preoccupations with terrorism to an issue such as accession to the European Union) can also take their toll on a mechanism's efficacy.

Sensitivities regarding composition – reflecting a perennial balancing act between the efficiency of a small group and the legitimacy offered by a broad representation of states – are an ongoing problem. Members of a group will stress the flexibility, trust and cohesion that can be developed among a small number of states. Yet the influence that such groups can amass – usurping the authority of the UN Security Council, and/or excluding regional actors – bears a cost. In some cases the creation of a two-tier structure has helped address these issues: in East Timor, for example, a larger 'Support Group' complemented the small Core Group. In others, pressure for inclusion has led to large groups that cannot play an effective role. Unsurprisingly, experienced peacemakers have at times eschewed a group altogether (Cyprus in 1999, Afghanistan after 2001) , preferring to pursue the coordination of and complementarity among the multiple external actors involved in each case by different means.

In their interactions with non-state conflict parties groups of states face a series of challenges rooted in the state-centric biases of international peacemaking. A state that is also a conflict party engages with external actors with obvious advantages: the legitimacy afforded by membership of regional and multilateral organisations, familiarity with diplomatic norms and the rules of the system, and greater access to international resources than non-state counterparts. Such a state may not always welcome a coordination structure. However, it will be able to resist its pressures through invocation of the threats to sovereignty those pressures may appear to constitute, as well as the threat it faces from non-state actors it holds as illegitimate, criminal and, most likely, terrorist as well. Except in circumstances (such as southern Sudan or East Timor) where the non-state party enjoys the sympathy of international actors, the relations of group structures to non-state actors are inevitably more complex. Coordination mechanisms can appear as a means by which the international community has united against them, and attempts to introduce conditionalities, as in the case of Sri Lanka, may go awry.

It comes as no surprise that there is no easy answer to the orchestration of international action in peace processes. Best practice involves the recognition that, however attractive the prospect of a group may be, it may not always be the answer. Moreover, as form should follow function, flexibility will be key: strategies and mechanisms employed during peacemaking may not be adequate either to the demands of implementation and peacebuilding or to a process that has suffered a violent reversal.

Developing effective complementarity among state actors involved in a peacemaking effort, between state and non-state peacemakers, and, more ambitiously, in order to try to channel or rationalise the incentives and sanctions being applied by other actors and structures, is likely to remain an ongoing struggle. Yet not to try is not an option. Critical in any such endeavour will be the recognition that a group structure or mechanism, however effective, must remain at the service of, and not a substitute for, strategies for international engagement in a peace process.