A Conceptual Typology of Kin‐State Mediation Open Access

In this study, I define “kin states” as those states whose dominant ethnic group identifies with a co‐ethnic population that transcends the borders of that state.¹ (In some cases, the “kinship” affiliation may be political or religious identity rather than ethnicity.) Mediation is generally defined as a form of third‐party intervention undertaken by those who have no direct involvement in the conflict and are not related to the disputants. Kin‐state actors, according to this view, would make unlikely candidates for mediating international conflicts that involve their kin communities because they are seen as likely to be biased toward them.²

In this article, however, I argue that kin states do, in reality, assume significant mediation roles, that these roles can be conceptually distinguished and categorized, and that their bias toward their kin may not necessarily undermine their involvement. The findings of my case analytic research support the argument that biased mediators can play a greater — and more beneficial — role in mediation than previously assumed (Kydd 2003; Savun 2008; Favretto 2009; Svensson 2009).

Many scholars have defined mediation, simply, as any third‐party intervention in a conflict designed to help the disputants reach a settlement (see, among others, Princen 1992; Sisk 1995; Bercovitch 2002). They have also argued that being a successful third‐party mediator requires having certain qualities, such as neutrality and status. Doyle Spencer and Huang Yang (1996: 1495), for instance, defined mediation as “the assistance of a third party not involved in the dispute, who may be of a unique status that gives him or her certain authority with disputants; or perhaps an outsider who may be regarded by them as a suitably neutral go‐between.” In an article reviewing recent mediation scholarship, James Wall and Timothy Dunne (1981: 219) observed that the definition of mediation remained essentially the same despite various attempts at fine‐tuning, shortening, or lengthening it over the years, and that mediation is simply “assistance to two or more interacting parties by a third party who—at that time—has no power to prescribe agreements or outcome.”

These definitions seem to assume that in mediation a clear distinction exists between the parties and the third party who seeks to assist them to settle their dispute. The disputants and the third party are thus conceptualized as mutually exclusive of each other, and the role that kin states or similar secondary party actors may play in peacemaking is typically not considered. The relationship between kin states and kin communities, directly or indirectly, would thus seem to disqualify from any mediation role those states that have such a relationship with a party in dispute.

It follows logically that a kin state's support to its kin community disputant could make a negotiated settlement less likely, and the existing research on kin states has often viewed kin states as backers of their kin communities, whose involvement helps prolong the conflict rather than resolve it (Brubaker 1996; Ganguly 1998; Caspersen 2008). Louis Kriesberg (1991), however, noted that some conflict situations have involved “quasi‐mediators,” factions or parties within a disputant group who have mediated between their group and the other disputant. Could kin states be involved in disputes as quasi‐mediators? If so, what specific roles do kin states assume in mediations led by third parties? And under what conditions is the involvement of kin states beneficial? These are the questions I focus on in this study.

First, I review general mediation studies and the existing research on kin states to consider the possible roles that kin states could play in mediation. I then develop a four‐part typology of kin‐state involvement in mediation, providing examples of actual cases of kin‐state involvement in peacemaking. According to the proposed typology, kin states assume one of four roles vis‐à‐vis their kin communities within the context of a mediation process: promoter, quasi‐mediator, powerbroker, or enforcer. Finally, I describe each type of kin‐state role, from the least to the most intrusive, using the examples of Croatia and Federal Yugoslavia in the Bosnia peace process, Greece and Turkey in Cyprus, and the Republic of Ireland and the United Kingdom in Northern Ireland.

In examining peacemaking efforts to address international conflict, Jacob Bercovitch (2002) divided most mediators into one of the three following categories: individuals, states, and institutions and organizations. Other mediation scholars have also noted that, while most mediators are unilateral actors, “collective” forms of mediation exist that include groups of states, individuals, and/or organizations acting together as a collective mediation body, such as the Contact Group in Bosnia or the United Nations Security Council (see Touval 1994; Leigh‐Phippard 1998). Although individual states often have more resource and leverage than such organizations, they are usually concerned with managing those conflicts that affect them directly — they face less pressure to pursue “altruistic” peacemaking in conflicts in which they have less at stake (Fretter 2002). The United Nations often thus becomes a mediator by default because states are unwilling to engage in particular conflicts. The prime weakness of U.N. mediation, however, is that, as an intergovernmental organization, it is incapable of exercising some of the basic functions required for effective mediation of complex international disputes such as leverage and coercion (Touval 1982).

Critiques of U.N. mediation assume that to succeed the mediator must manipulate the negotiation context using leverage or coercion. That is power mediation, also known as manipulative or directive mediation, and scholars have identified it as one of the three main styles of mediation that address violent international conflict; the others are facilitation and formulation (Touval and Zartman 1995; Beardsley et al. 2006).³ “Mediation style” is broadly defined as “the overarching goals and definitions of the [mediation] role, sometimes implicit, that shape how mediators behave and what they consider the legitimate goals of intervention” (Kressel 2007: 251).

Facilitative mediation style essentially entails establishing communications between the disputants, arranging interactions between them, and helping them identify and clarify the issues at stake. The formulative mediator, by means of substantive suggestions and proposals, strives to devise a solution that is acceptable to the disputants.⁴ Ronald Fisher (2001: 11) defined facilitation‐ and formulation‐based mediation as “pure mediation” in which “the third party works to facilitate a negotiated settlement on substantive issues through the use of reasoning, persuasion, effective control information, and the suggestion of alternatives.”

Power mediation, on the other hand, includes elements of “pure mediation” but also the use of leverage or coercion by the mediator, which can occur in the form of promised rewards or threatened punishments (i.e., carrots and sticks). In addition, power mediators’ involvement may extend beyond the agreement if they, for example, take part in implementation and enforcement of the agreement (Fisher 2001; Beardsley et al. 2006).

Lindsay Reid (in press) has recently identified two forms of leverage in mediation: capability leverage, which involves the traditional use of power and resources, and credibility leverage, in which the mediator's historical and cultural ties play a role. Credibility leverage, I argue, is the form of leverage held by kin‐state actors who have ties with the disputing parties. Power, or leverage in whichever form, may enhance mediation efforts, but it has a significant downside. As Peter Carnevale (2002: 34) put it:

Nevertheless, many scholars consider manipulation of the negotiation context and the use of some coercive techniques to be essential elements of successful mediation of some conflicts, particularly if the mediator's facilitative and formulative efforts have proven insufficient to help the disputants reach a settlement. Kin states, which may have important national interests at stake because of the ethnic bonds they share with one or more of the disputants and because they are often geographically close to the parties involved, are well‐positioned to act in a manipulative or coercive capacity. The successful use of power in mediation and the involvement of kin states raises the question of whether the presence of bias in mediation may, in fact, be helpful.

The issue of bias is contentious in the mediation literature: scholars have debated the impact of a mediator's bias on the mediation outcome. Some have asserted that a mediator should be neutral toward disputants (Young 2007; Fisher 1995), but others have disagreed (Touval 2004; Carnevale and Arad 1996). More recently, some have argued that biased mediators, particularly major powers, have more credibility with the party toward whom they are biased, which enables them to mediate more successfully (Kydd 2003; Savun 2008; Favretto 2009; Svensson 2009). According to one study, biased mediators show more concern about the quality of peace agreements than do neutral mediators, who are more interested in reaching an agreement than in the agreement's durability (Svensson 1993).

Biased mediation is not only undertaken by major powers and states, but also by “quasi‐mediators” (Kriesberg 1991) and “insider‐partial mediators” (Wehr and Lederach 1993; Svensson and Lindgren 2015). Both terms refer to individuals and groups from the same communities as the disputants who become involved in mediation. Are kin states insider‐partial mediators or quasi‐mediators? Although both terms are essentially the same, the modifier “insider‐partial” puts the emphasis on the “insider” quality of the mediator, but not all kin states are necessarily insiders; hence, in this article, I use the term quasi‐mediator.

On the other hand, Bernd Beber (2012) disputed the idea that biased mediators are more successful in bringing about settlements. Beber claimed that a number of high‐profile cases are often cited as evidence of the effectiveness of biased mediation, but whether those cases are representative and whether their mediators were actually biased is questionable. Those cases were the American and British mediation of the conflict between Italy and Yugoslavia over the city of Trieste in 1954, Algeria's mediation during the 1979–1980 hostage crisis between the United States and Iran, the Soviet Union's mediation of the conflict between India and Pakistan in 1966 at Tashkent, and the American mediation at Camp David in 1978 between Egypt and Israel. None of the cases of kin‐state mediation that I examine in this article belongs to this set of high‐profile success stories of supposedly biased mediation—the cases discussed below thus provide substantial new support to the proposition that biased mediation can be effective.

In the aftermath of the Cold War and the democratization of Eastern Europe, kin states and their dealings with their kin communities have attracted international interest. According to the European Commission for Democracy through Law (also known as the Venice Commission), an advisory body of the Council of Europe, “the phenomenon of the concern of certain states for their kin minorities” (European Commission for Democracy through Law 2012) only became an issue for the international community as a result of the discussions that followed the Hungarian parliament's adoption of the Act of Hungarians Living in Neighboring Countries in June 2001.

Scholars have been discussing and developing theories about kin‐state involvement in ethnic conflicts since the mid‐1990s. Rogers Brubaker (1996) conceptualized a triadic model for kin state and kin community relationships in post–Cold War Eastern Europe, whose three sides are a nationalizing state (host state), a national minority population, and an external national homeland (kin state). Because kin states claim the right, even obligation, to protect the interests of their kin, he argued, they can play a significant role in the rise of extreme nationalism and ethnic conflict.

Researchers have studied kin‐state involvement in other geographical contexts as well. Rajat Ganguly (1998) studied some kin‐state interventions in secessionist conflicts in South Asia by drawing on George Modelski's (1961) theoretical model.⁵ Ganguly (1998) envisaged four different ways that kin states may respond to secessionist conflicts involving their kin: (1) diffusion and encouragement (the kin state allies itself with the kin group); (2) isolation and suppression (the kin state allies itself with the government of the host state); (3) reconciliation (the kin state becomes a third‐party mediator between the kin group and its host state); and (4) diffusion or isolation through inaction or nonintervention. If the host state is powerful, inaction would be regarded as isolation by the kin group, but if the kin group is stronger, inaction would be interpreted by the host state as support to the kin group, because the kin group is not intervening to stop it. Ganguly (1998) found that nonintervention by the kin state is often the least likely policy.

Sean Byrne (2000, 2006) used the term “external ethnoguarantors” to describe the kin states in the Northern Ireland and Cyprus conflicts.⁶ Byrne compared the role of ethnoguarantors to that of “primary mediators” in Northern Ireland and Cyprus, and stressed the need for cooperation between ethnoguarantors and mediators. Stefan Wolff (1991: 34) developed an analytical framework that takes into account “the complex relationship between host‐state, kin state, actors in the disputed territory and in the international context” to explain the transnational dynamics of ethnic conflict settlements in Alsace (now France), the Saarland (now Germany), South Tyrol (now Italy), Northern Ireland, Andorra, and the New Hebrides (now Vanuatu). Wolff argued that the kin state's ability to balance the interests of the external minority (the kin community) with those of the host state can help determine whether the conflict can be resolved or not. Wolff focused on the main developments in these conflicts, however, and did not examine the kin‐state involvement in any particular peacemaking process. For instance, he did not analyze the kin‐state dynamics of the Northern Irish all‐party negotiations, which culminated in the Good Friday Agreement.

More recently, scholars have shown a growing interest in the relationships between kin states and their co‐ethnics, particularly in the citizenship laws and foreign policies of kin states (Wolff and Cordell 2002; Batory 2010; Pogonyi, Kovács, and Körtvélyesi 2010). Nina Caspersen (2007, 2008) analyzed the relationships between kin‐state leadership and kin‐community leadership in the conflicts in Bosnia, Croatia, and Nagorno Karabakh. She analyzed the Serb‐Bosnian, the Serb‐Croatian, and the Armenian‐Nagorno‐Karabakhi relationships in terms of their power dynamics, but has not assessed the roles of kin states in third‐party peacemaking processes. Caspersen's somewhat limited conclusion (2008: 370) was that “the degree and the form of kin‐state influence are likely to vary both between cases and, over time, within cases, and complete unity of kin‐state and local leaders should not, in any case, be assumed.”

Kin states’ roles in ethnic conflicts have not been fully explored, and, in particular, their role within third‐party mediation processes remains understudied. To assess whether this involvement supports or does not support the goals of third‐party mediators and disputants, scholars must first analyze and conceptualize kin states’ involvement in mediation in greater depth. Such scholarship should seek to determine how kin states influence the context and process of mediation, what specifically is their role in mediation, and how they interact with mediators and disputants within the context of peacemaking. Specific questions include: What roles do kin states assume in peacemaking interventions? Do they act as mediators? If they act as mediators, what specific roles do they assume? Do they assume tasks similar to the tasks of a third‐party mediator, such as facilitating communication between the disputants, formulating proposals, or acting as powerbrokers in the conflict?

Because kin states are directly or indirectly parties to the conflicts, we might expect that they should not be considered as peacemakers or mediators and that they would be more likely to act as counterforces in a mediation process, providing the parties with resources to sustain the conflict rather than help to resolve it peacefully.

But the range of available peacemaking roles is wider than a superficial analysis might suggest. Multiple actors provide a variety of mediation services; various groups and people can be officially designated as mediators but also as representatives of a particular party in the dispute and can function as intermediaries between the governments they represent and the other side (Kriesberg 1991). According to Kriesberg (1991: 23), the latter group can be conceptualized as quasi‐mediators, and

Kin states may thus act as quasi‐mediators, extending the mediation process beyond the disputants and the neutral third party.

Jacob Bercovitch, Theodore Anagnoson, and Donnette Wille (1991) proposed a basic contingency model of mediation that assumed two sets of variables in any given mediation: context variables and process variables. The context variables are the nature of the parties, the nature of the dispute, and the nature of the mediator; the process variables are the mediator's strategies. According to this contingency model, the interaction of the two sets of variables determines the outcome of the mediation. In mediations in which kin states are involved, as quasi‐mediators or in other ways, the involvement of kin states would qualify as another variable in the contingency model.

What other forms of kin‐state involvement in peacemaking can we identify? Kin states can promote the interests of their kin by providing them with support, in particular by “internationalizing” the conflict, that is, by bringing more international attention to it, drawing the involvement of the United Nations and third‐party mediators, and so forth.

A rarer form of kin‐state involvement in peacemaking is powerbrokering, which is more likely to be practiced by kin states that have particularly close ties with their kin group disputant. Turkey's involvement in the Cyprus conflict is an example of this kind of powerbrokering; its interests are closely entwined with those of its Turkish‐Cypriot kin on the island.

Finally, the most intrusive form of kin‐state involvement can be labelled as enforcement, in which the kin state dictates terms to the kin community. An example of intrusive involvement occurred when the Bosnian Serbs were forced by the leadership of their kin state, the Serbian dominated rump state of Yugoslavia, to accept the Dayton Peace Accords in 1995.

In sum, the four types of kin‐state involvement in peacemaking interventions are: promoter, quasi‐mediator, powerbroker, and enforcer. I will next examine each of these models in the context of the mediation processes in Bosnia (1991–1995), Cyprus (1999–2004) and Northern Ireland (1996–1998). Each of these conflicts involved ethnic or religious groups with neighboring kin states.

Promotion is the most common form of kin‐state involvement in peacemaking. In this model, the kin state's role is limited to backing its kin community at the international level. For the kin state, this type of involvement does not require very close ties or direct engagement in the conflict. In these cases, the kin state uses its international network and resources to help its kin community at the international level by providing diplomatic and financial support, particularly by attracting international attention, which can put pressure on the state. Because the kin community typically lacks the resources necessary to voice its concerns at the international level, the kin state's assistance can represent a significant contribution.

For instance, Greece has worked to internationalize the Cyprus conflict for the benefit of Greek Cypriots both during the conflict in the 1950s and also later at the European Union level. In the early 1950s, Greece repeatedly appealed to the United Nations to hold a referendum concerning the island's international political status (Faustmann 2001). In the 1990s, Greece lobbied and campaigned for Cyprus's membership in the European Union, a plan opposed by the island's other main community, Turkish Cypriots, and its kin state, Turkey (Tocci 2013).

The Republic of Ireland's involvement in the Northern Ireland conflict, particularly in 1970s and 1980s, represents another case of kin‐state promotion of its kin community's interests. Ireland offered its diplomatic support to its kin community, the Northern Irish nationalist community (mostly Catholics), throughout this period (McKittrick and McVea 2012). The involvement of kin states as promoters of their kin communities’ interests usually occurs in the early phase of the ethnic conflict and seems to be an essential element of the kin community's efforts to internationalize its cause.

Kin‐state involvement as promoter seems less obviously connected to peacemaking efforts. But the kin community often depends on its kin state to advance its cause internationally, which can be an important element of the peacemaking process. A kin state acting as a promoter is one of the actors in the process—the third‐party mediator must take the kin state into account and establish relations with it in order to further a peace process. Should the third‐party mediator fail to establish contact and cooperation with the kin state, the kin community, who relies on the kin state's support, would be less likely to support the mediator's efforts to resolve the conflict. The kin state could also choose to go beyond its promoter role and become a spoiler of the mediation process, something the mediator would seek to avoid. The kin state's role as promoter is a conceptually significant aspect of kin‐state involvement in peacemaking interventions, but it is not a mediation role. Rather, it is an additional aspect of the dispute that third‐party mediators should consider when formulating their mediation approach.

As I noted earlier, Kriesberg (1991) coined the term “quasi‐mediator” to refer to individuals or groups who often have close ties with one of the disputants, but, nevertheless, assume some mediation roles. Kin states are most likely to assume this role when their close ties with the kin community can improve the prospects for successful mediation, particularly when continuation of the conflict threatens the state's wider interests. In the quasi‐mediator role, the kin state seeks to bridge the gap between its kin and the other disputants. For instance, the kin state may serve as a go‐between among the disputants or as a mediator between its kin community and an officially designated neutral mediator. The kin state may also help formulate proposals.

The kin state will not be seen as impartial by the other disputants, but its affinity with its kin community may nonetheless prove beneficial. The kin state may already enjoy close ties to its kin community, and may already have—or could easily establish—communications and trust with its kin community. Finally, because quasi‐mediation is essentially a nonintrusive mode of engagement, it creates no severe dependency or legitimacy issues that could weaken whatever resolution emerges from the mediation. Because no coercion is involved, the deal may be seen as more legitimate by the local population and the agreement will, as a result, require less outside support and enforcement.

Examples of quasi‐mediation by a kin state include the participation of both the United Kingdom and the Republic of Ireland in the Northern Irish all‐party talks (1996–1998), Turkey's involvement in the Cyprus conflict (1999–2002), and the involvement of Federal Yugoslavia during the Vance‐Owen Plan negotiations (1993) to resolve the Bosnia conflict.

Britain's involvement in the Northern Irish all‐party talks (1996–1998) is complicated. The Republic of Ireland's role as a kin state to the Northern Irish nationalist community is obvious, while the relationship between Britain and the Unionist community is more complicated. Northern Ireland is part of the United Kingdom and therefore, strictly speaking, Britain is not a kin state to any Northern Irish community. The relationship between the British state and Northern Ireland's Unionist community, however, largely resembles one between a kin state and its kin community. The Unionists, as the term itself indicates, strongly identify themselves with Britain. But Northern Ireland's status within the United Kingdom has been disputed so frequently that British governments have contemplated withdrawing from Northern Ireland. According to Adrian Guelke and Frank Wright (1990: 52), that is “because Northern Ireland is not perceived as an integral part of the national territory, nor is the British identity of its Unionist majority accepted at face value by most people on the UK mainland.” The Unionists identify as British. Although the British do not share exactly the same affinity, they maintain a concern for, and feel some responsibility for, the Unionists. In this way, Britain's relationship with that segment of the Northern Irish population is similar to the relationship of a kin state to its kin community.

An explicit example of Britain's involvement as a quasi‐mediator occurred when Tony Blair and Bertie Ahern, the British and the Irish prime ministers, joined the Northern Irish all‐party talks, chaired by the U.S. mediator, former Senator George Mitchell. These talks were instrumental to achieving the Good Friday Agreement in 1998 (Mitchell 1999; Powell 2008; Blair 2010). Britain and Northern Ireland continued as quasi‐mediators after 1998, on those occasions when the local parties were unable to concur upon aspects of the agreement.

Turkey and Federal Yugoslavia are both kin states to their respective kin communities, the Turkish Cypriots and the Bosnian Serbs. In both cases, the quasi‐mediation efforts of the kin state followed a similar pattern. Initially, the kin state supported the demands of the kin community. But both kin states saw their national interests threatened by the continuation of the conflict and their roles changed.

Turkey's quest for EU membership was the catalyst for Turkey to support the United Nations–mediated peace talks in Cyprus. Ultimately, Turkish involvement in the talks helped get the Turkish Cypriots back to the table and led them to say “yes” in 2004 to the U.N.'s blueprint for peace on the island, which was rejected by the Greek Cypriot community. The Turkish government focused on the Turkish Cypriot leader Rauf Denktash, acting as quasi‐mediator between him and U.N. Secretary‐General Kofi Annan's team of peacemakers when the talks were deadlocked between 2002 and 2004. Diplomats who were involved in the process have noted that the Turkish government's efforts did more than put pressure on its kin community's leader to accept the U.N.'s blueprint. The Turkish government was, in fact, deeply engaged in efforts that fall under the category of facilitative mediation, such as re‐establishing communications between the intransigent kin community's leadership and the U.N. and then, with the U.N., helping develop alternative proposals that were more acceptable to the Turkish Cypriots (United Nations Security Council 1985; Hannay 2005).

The Bosnian Serbs challenged Bosnia's declaration of independence from Yugoslavia in 1992, hoping to secede from Bosnia and form a separate entity. The Serbian leadership of the Yugoslav Federation, which at this time consisted of Serbia and Montenegro, actively supported the Bosnian Serb cause. The Serbian leadership initially provided both military and diplomatic support to their kin community in Bosnia. The Yugoslav Federations were threatened by U.N.‐sanctioned embargoes, however. The kin state reduced its support to the Bosnian Serbs, and Yugoslavia's Serbian leader Slobodan Milosevic took on an intermediary role between the Bosnian Serbs and the mediators, David Owen and later Richard Holbrooke (Owen 1996; Holbrooke 1998). The mediators needed Milosevic because they were sometimes unable to establish direct communications with the intransigent Bosnian Serb leadership, and also because U.S. mediator Holbrooke was unwilling to directly deal with the Bosnian Serb leadership, who the Americans regarded as war criminals (Holbrooke 1998).

The Milosevic government's quasi‐mediation role was apparent during the Vance‐Owen Plan negotiations in 1993. Then, facing growing international pressure, Milosevic, along with Dobrica Cosic, the federal president of Yugoslavia, and Momir Bulatovic, the Montenegrin president, held meetings with the Bosnian Serb leadership to convince them to accept the plan in the spring of 1993. These efforts culminated in the decision of Bosnian Serb leader Radovan Karadzic to agree to the Vance‐Owen Plan. Milosevic's efforts continued when the plan was brought before the Bosnian Serb Parliament (Silber and Little 2008; Owen 1996). But the plan failed to pass in a referendum and the Bosnian Serb leadership's continuing intransigence lead Milosevic to go beyond a quasi‐mediation role, which I will discuss below.

The third role that kin states may play is powerbroker. This term suggests the extensive control over the mediation process that the kin state often has through its manipulation of local actors using both threats and bribes. Kin states, as powerbrokers, can play significant roles in the conflict and are able to shape political aspects of the kin community's involvement in the mediation process. A kin state will typically only take on this role when the kin community is both heavily dependent upon it and has limited means to oppose it. As powerbrokers, kin states are the dominant actors in the relationship and their actions are often unilateral. The kin community may negotiate with the kin state about the mediation process, but where their interests clash with the former, those of the kin state will take precedence. Powerbrokering is an intrusive form of kin‐state intervention in peacemaking, and agreements generated this way can suffer because the kin community does not feel ownership of the agreement and because such agreements are seen to lack legitimacy.

A striking example of kin state powerbroking occurred during the 1973 Sunningdale Conference. The conference generated a power‐sharing agreement between Unionist and the nationalist communities in Northern Ireland and called for a Council of Ireland, which would establish cooperation with the government of the Irish Republic. The Council of Ireland was anathema to the Northern Irish Unionists who saw it a step toward the Republic of Ireland's takeover of Northern Ireland, but the British government was able to overcome the Unionist objection using powerbrokering: reportedly British Prime Minister Edward Heath heavily pressured Brian Faulkner, leader of the Ulster Unionist Party (UUP), telling him that the Council of Ireland was the price to be paid for preservation of the union (Kerr 2005).

Greece and Turkey played powerbrokering kin‐state roles in the 1958–1960 Cyprus negotiations, during which they internationally represented the Greek and Turkish Cypriot communities, who had no means to oppose their kin states. Thus the Greek and the Turkish governments first agreed to independence for Cyprus through bilateral negotiations that culminated in treaties in 1960. The treaties were subsequently accepted by the leaders of the kin communities who were essentially coerced into accepting them (Varnava 2004).

Turkey played a similar role during the Annan Plan negotiations (2003–2004). When opposed by the hardliner Turkish Cypriot leader Rauf Denktash, Turkey shifted its support to the Turkish Cypriot opposition and forced Denktash to agree to a referendum to ratify the Annan Plan (Eralp and Beriker 2005; Ker‐Lindsay 2005; Palley 2005). Because the Turkish Cypriots were heavily dependent on Turkey for financial and diplomatic support, Turkish involvement in the Annan Plan negotiations (2002–2004) can be considered as powerbrokering. As I noted above, Turkey's early intervention (1998–2002) as a kin state was mainly in a quasi‐mediation role. Turkey's desire to join the EU, however, grew, and by late 2002 the new Turkish government of Recep Tayyip Erdogan was unwilling to limit itself to a quasi‐mediation role. Instead the government sought to change the course of Turkish Cypriot politics to better align with Turkey's objectives (i.e., EU membership). The Turkish government publicly criticized Denktash for his intransigence and emphasized the Turkish Cypriot state's dependence upon Turkish financial assistance for its survival (Hürriyet Daily News 2004).

Powerbrokering can be an effective form of kin‐state involvement in peacemaking, and the third‐party mediator may see it as beneficial. The intrusive nature of kin state powerbrokering, however, means that the process often meets opposition from kin communities. The leadership of the kin community in particular is likely to object to the kin state's overpowering involvement. As I noted above, relations between the Federal Yugoslav and Turkish governments and the leaders of their respective kin communities deteriorated as their kin‐state involvement intensified. In such cases, third‐party peacemakers must balance the benefits and disadvantages of effective powerbrokering by kin states. The most significant drawback is the threat to an agreement's legitimacy — if a peace deal is struck via kin‐state powerbrokering, its long‐term viability may require further powerbrokering.

Lack of local ownership is an even greater problem when kin states act as “enforcers of their own will” in the mediation process. Kin states act as enforcers by taking over the kin community's decision‐making bodies and imposing their will on their kin communities with little or no consultation with the kin‐community leadership. This form of involvement is usually problematic for mediation. It is found in particular types of conflict, for example, when the kin community is wholly dependent on the kin state because it is waging a war and has no other sponsors. Enforcer kin states often treat kin‐community leadership as a mere extension of their own policy making. Third‐party mediators will not normally welcome such interference.

Kin states acting in this capacity may help bring about a ceasefire or a peace agreement, but the lack of political input from local actors will likely undermine the prospects for a lasting agreement. This form of involvement can only serve as temporary means to a limited end—without the kin community's genuine involvement in the peacemaking process, neither the third‐party mediator nor the kin state can achieve a sustainable peace.

An example of a case in which kin states acted as enforcers involves the participation of the governments of Federal Yugoslavia's President Slobodan Milosevic and Croatia's President Franjo Tudjman in the negotiations that took place in Dayton, Ohio in 1995 to end the war in Bosnia. Representatives of Croatia and Federal Yugoslavia spoke for their kin communities (Bosnian Croats and Bosnian Serbs) whose leaders were either not present or present but not involved in decision making.⁷ Seeing no other viable alternatives, Richard Holbrooke's team of mediators promoted such a form of kin‐state involvement as they sought a peace agreement for Bosnia (Holbrooke 1998; Daalder 2000). The fact that the kin‐community leaders were almost totally ignored and the kin states were able to negotiate on their behalf proved helpful in the short term but undermined the accords in the longer term. This would prove to be a major reason why local kin communities failed to strongly support the Dayton Peace Accords. A third‐party mediator should be wary of accepting such overwhelming participation by a kin state in a mediation process. When kin states act as enforcers of their own will, they threaten their own relationships with their kin communities. Moreover, any settlements that might arise from the mediation are likely to lack legitimacy and the kin‐community disputants are less likely to see themselves as “owners” of the peace agreement.

Kin‐state involvement in peacemaking processes presents third‐party mediators with opportunities and challenges. The most challenging situation occurs when the kin state assumes an enforcer role, leading the local kin community to see the mediation as an imposition. Although the kin‐community actors’ objectives can often be easily identified, a kin state's strategy can be more difficult to determine. The kin community's political goals could include a power‐sharing deal, establishing dominance over the other group, and secession. A kin state, however, may see a broader array of policy options that could include annexation, a power‐sharing regime, a majoritarian democracy, and political defeat of one or the other side in the conflict. The kin state's policy agenda can thus become the most dynamic component of a peacemaking process, and that agenda often reflects the kin state's wider international interests.

Kin‐state involvement in mediation appears to be widespread. The conceptual typology I have proposed in this article comprises the four major roles, sometimes distinct, sometimes mixed, that kin states typically assume in a peacemaking process. They are promoter, quasi‐mediator, powerbroker, and enforcer. My analysis of several cases strongly indicates that biased mediation, when it takes particular forms, such as quasi‐mediation, can indeed be effective and helpful. This is a relatively new area of study and additional empirical research into the forms that kin‐state involvement can take would be illuminating.

Mediators can manipulate the relationship between a kin community and its kin state with incentives and disincentives to influence the kin state to become involved for the benefit of the peacemaking intervention. In cases in which kin states act as powerbrokers or enforcers in particular, I advise third‐party mediators to be aware that the mediation is far more likely to succeed if the process takes into account the needs of kin‐community actors as well as the interests of kin states.

Finally, I note that cooperation between kin states, as in the case of Northern Ireland, can itself become a crucial dynamic of the peacemaking process, and an option that peacemakers should consider. Kin‐state and kin‐community relations across borders are strong in many politically turbulent regions (such as the Middle East); thus, kin‐state involvement is likely to become an important aspect of mediation, requiring further scholarly attention.