“Sisters in Peace”: Analyzing the Cooperation between the United Nations and the European Union in Peace Mediation

This article examines UN –EU cooperation over peace mediation. It compares their conceptual approaches to peace mediation and the evolution of their institutional capacities, demonstrating that the EU has learned from the UN , while actively supporting the strengthening of UN mediation capacity. The most important difference concerns the embeddedness of mediation in a broader foreign policy agenda in the case of the EU compared to the UN . The article also examines models of EU– UN cooperation in mediation practice. Drawing on an overview of cases of UN –EU cooperation, the article develops a typology of the constellations through which the two organizations have engaged with and supported each other. A case study on the Geneva International Discussions on South Ossetia and Abkhazia investigates the effectiveness of this coordination. The findings point to a high degree of effectiveness, although this has not yet translated into tangible mediation outcomes.

Contemporary international mediation has become increasingly crowded and complex, with the proliferation of external actors such as states, international and regional organizations, NGOs, and individuals serving as mediators (Lanz & Gasser 2013).Among international organizations, the United Nations (UN) has been one of the most frequent providers of international mediation since 1945 (Greig & Diehl 2012: 68).While the UN played important mediation roles in the immediate aftermath of the Cold War, its significance in mediation has been declining since the beginning of the 21st century, with its "track record as an international mediator appear[ing] to be low profile and obscure, occupying the center of global media attention less frequently" (Iji 2017: 83).Despite its involvement in the resolution of current conflicts, such as those in Syria, Yemen and Libya, the re-emergence of great power rivalry and challenges to multilateralism have rendered the environment and space for UN mediation much more complex (Troitskiy 2019).
At the same time, regional organizations have become increasingly engaged as peace mediators.Among them is the European Union (EU).After initial failures to broker agreements in Yugoslavia's conflicts in the 1990s, and a phase during which it consolidated its foreign and security policy, the EU has (re-)emerged as an important international provider of peace mediation, particularly in regions to the east and south of its area.Moreover, the EU has systematically strengthened its capacity for mediation, particularly after the adoption of the 2009 Concept on Strengthening EU Mediation and Dialogue Capacities (Bergmann 2020: 6-9).
The Concept on EU Peace Mediation adopted on 7 December 2020 -updating the 2009 Concept -strongly emphasizes the importance of the EU's partnership with the United Nations on peace mediation and "reconfirms its strategic solidarity with the United Nations" (Council of the EU 2020b: 2).In fact, there are a number of cases where the UN and the EU have closely cooperated on mediation engagements, such as in the Yugoslav civil wars of the 1990s, secessionist conflicts in the South Caucasus, Cyprus and, more recently, the ongoing civil wars in Libya and Yemen.Despite its empirical relevance, however, academic research has largely overlooked UN-EU cooperation in peace mediation.On the one hand, studies on UN mediation have extensively dealt with UN involvement in multi-party mediation efforts, but UN-EU cooperation has not featured prominently in those analyses.On the other hand, the emerging literature on the EU as an international mediator has largely focused on assessing the EU's effectiveness in mediation engagements, with little attention paid to how cooperation with other third parties influences the EU's mediation endeavors (Davis 2014;Niemann et al. 2018).Moreover, there is very little work comparing the EU's approach to that of other international organizations (for an exception, see Bergmann 2018).
Taken together, we know very little about how the two organizations' mediation approaches differ or converge, how they have learned from each other in terms of institutionalization of mediation structures, and how their cooperation works in practice.To address this research gap, this article explores how the UN and the EU cooperate on peace mediation, both in terms of policy development and specific mediation engagements.Looking at an individual case of UN-EU cooperation in peace mediation, the article also examines the extent to which their coordination on mediation is effective.Building on existing multi-party mediation research, effectiveness in mediator coordination is here defined as the degree to which the mediators cooperate in terms of (internal and external) communication, the pooling of resources, and the pursuit of a joint mediation strategy (see case study section below; Crocker et al. 1999Crocker et al. , 2001;;Böhmelt 2012).
The remainder of the article is structured in three parts.First, it compares the two organizations' conceptual approach to peace mediation and the evolution of their institutional capacities in this domain, demonstrating how the EU has adopted various features of the UN's mediation approach and learned from the UN's experiences, while actively supporting the strengthening of UN mediation capacity.Second, moving from the policy level to mediation practice, the article examines different models of EU-UN cooperation.Drawing on an overview of cases of UN-EU cooperation in mediation interventions, the article develops a typology of the different constellations through which the two organizations have engaged with and supported each other in mediation efforts.Third, a case study on the Geneva International Discussions (GID) on South Ossetia and Abkhazia investigates the effectiveness of UN-EU coordination as co-mediators.

Comparing the Approach of the UN and the EU to Peace Mediation: Concepts, Strategies, Institutional Structures
Both the UN and the EU have systematically built up their capacities and institutional structures for peace mediation over the past two decades.Despite International Negotiation 27 (2022) 418-447 differences in the institutional nature of the two organizations -the UN being a "pure" intergovernmental organization, while the EU's foreign policy is also partly shaped by supranational institutions and actors -both the UN and the EU have followed roughly similar trajectories concerning their approach to peace mediation (see Table 1).In this section, we discuss the evolution of the two organizations' conceptual and institutional structures for mediation before analyzing the extent to which they have influenced each other's capacity development in terms of institutional learning.

The UN's Mediation Capacity and Support Structures
Peace mediation has been a core function of the UN since its inception, being mentioned by the UN Charter as one of the key instruments for the peaceful settlement of disputes (UN 1945: Art. 33).In addition, it has been an incremental part of the good offices role of the UN Secretary-General (Day 2018: 69-73).Since the Cold War, in particular, the UN has significantly stepped up its mediation engagements around the globe (Iji 2017: 83).However, it was not before 2004-05 that the UN systematically strengthened its mediation capacity and "reached adulthood" as a peace mediator, Convergne (2016: 181), notes.In its December 2004 report, the UN Secretary-General's High-Level Panel on Threats, Challenges and Change called for the provision of "more consistent and professional mediation support" and proposed "a field-oriented, dedicated mediation support capacity" that should be financed through additional resources assigned to the Department of Political Affairs (United Nations 2004: 39).Responding to this report, the UN member states in the World Summit Outcome called on the Secretary-General (SG) to strengthen the organization's mediation support activities and thus provided an additional push for institutionalization of the UN's mediation capacity (Lanz et al. 2017: 7).
A year later, the Mediation Support Unit (MSU) within the Policy and Mediation Division of the Department of Political Affairs was created.While the MSU initially comprised only two staff members, it has around 20 today.The main functions of the MSU are (i) to offer technical and operational support to peace processes (operational role), (ii) to strengthen mediation capacity within and outside the UN through training (capacity-building role), and (iii) to develop and disseminate knowledge products on mediation (knowledge creation role) (Lanz et al. 2017: 8).In addition, a mediation expert roster was created in 2006, which is managed by the MSU.It comprises approximately 220 members from the UN, governments, academia and non-governmental organizations who have relevant professional expertise and experience in mediation (UNDPPA 2019).In 2008, the UN created a Standby Team of Senior Mediation Advisers (SBT) that comprises eight full-time mediation experts  that are on permanent standby and can be deployed within 72 hours to provide advice to UN officials and partners upon request and deal with a wide range of peace mediation issues -including constitution-making, power-sharing, gender issues and transitional justice (UNDPPA 2021).At the conceptual level, the adoption of the UN Guidance for Effective Mediation has been a key milestone of UN mediation, as it sets out in detail the logic and the eight fundamental principles that underpin the UN's mediation approach: preparedness, consent, impartiality, inclusivity, national ownership, international law and normative frameworks, coherence, coordination and complementarity, and quality peace agreements (UN 2012).Finally, the Secretary-General established a High-Level Advisory Board on Mediation in 2017 that is tasked with providing advice and guidance to the Secretary-General on issues related to peace mediation and conflict prevention, as well as contributing to the UN's further engagement with regional and sub-regional organizations.

The EU's Mediation Capacity and Support Structures
The EU has been a provider of peace mediation ever since the creation of its Common Foreign and Security Policy by the Treaty of Maastricht in 1992.Whereas initial EU mediation activities in the 1990s focused primarily on mediation efforts in the civil wars in former Yugoslavia, the EU has increased its mediation engagements both in quantity and geographic scope since the early 2000s.EU mediation engagements include High Representative Javier Solana's successful mediations of the conflicts in Macedonia (2001) and Montenegro (2002), EU mediation efforts in Ukraine (2003,, Georgia (since 2008), and conflicts beyond Europe, such as in Egypt (2012-14), Yemen (since 2011), and Myanmar (2011-12) (Bergmann 2020;Brandenburg 2017;Girke 2015;Natorski 2018;Pinfari 2018).
The EU's emergence as an international mediator has been accompanied by the evolution of a more systematic EU approach to mediation in terms of policy development and capacity-building since the 2000s.Following the call in 2008, in the Report on the Implementation of the European Security Strategy, for an expansion of EU mediation and dialogue capacities (Council of the EU 2008: 9), the Council of the EU, under the Swedish Presidency, adopted the Concept on Strengthening EU Mediation and Dialogue Capacities (Council of the EU 2009).Defining mediation as an EU foreign policy tool and spelling out the different mediation roles the EU is able to take up in a given conflict, the Concept became the main conceptual reference point for subsequent EU mediation activities in the 2010s.It also paved the way for a stronger institutionalization and professionalization of the EU's mediation capacity (Brandenburg 2017(Brandenburg : 998-1000)).To implement the 2009 Concept, the EU established a Mediation Support Team (MST) within the European External Action Service (EEAS) that became operational in late 2011 (Hauck & Sherriff 2013: 16-29).The European Parliament provided the MST with an initial administrative budget of roughly EUR 2 million, "which allowed for a small but highly motivated staff that has progressively inserted the relevance of mediation and mediation support into its overall policy practice" (Herrberg 2018: 303).The Mediation Support Team itself is not engaged as a third-party mediator in violent conflicts on behalf of the EU, but it is mandated to pursue the following four main tasks: (i) operational support to EU delegations and partners and expert deployments; (ii) inhouse coaching and training of EEAS officials and EU Special Representatives, (iii) knowledge management and provision, and (iv) sustaining partnerships with the member states, the UN, regional organizations, and civil society organizations.In addition, the MST receives support through the European Resources for Mediation Support (ERMES) project, which allows the provision of assistance to conflict parties and third-party mediators and groups at short notice (EEAS 2021: 1).Apart from the EEAS, the European Parliament has also set up its own European Parliamentary Mediation Service within its bureaucratic structure in 2014.The service provides support and advice for the European Parliament's efforts in the field of mediation and conflict prevention (Fonck 2018: 4).
In 2020, a decade after the adoption of the 2009 Concept, EU peace mediation saw another major boost in terms of capacity and strategic direction.In June, Secretary-General Helga Schmid launched an EEAS pool of mediators, drawn from all geographic Managing Directorates of the EEAS (EEAS 2021).In December, together with Council Conclusions on EU mediation, a new Concept on EU Peace Mediation was adopted that reinforces peace mediation as an important tool of EU foreign policy.It has adapted the strategic framework of EU peace mediation to the changed geopolitical context that has evolved since 2009 (Council of the EU 2020a, 2020b).A key emphasis is on the EU's values-based approach to mediation which is portrayed as complementing and aligned to the EU's growing geopolitical interests, and its ambition to become a more self-determined actor in global politics (Council of the EU 2020b: 4-5).Finally, the EEAS published Peace Mediation Guidelines to accompany the new EU mediation concept.The guidelines aim to provide direction to actors involved in EU mediation and mediation support activities -such as EU Special Envoys and Representatives, Heads of Delegations, and members of the EEAS pool of mediators.Interestingly, the Guidelines state that they are meant as an "EU-specific reference document" that should be consulted "as a complementary tool to already existing materials, in particular the United Nations (UN) Guidance on Effective Mediation" (EEAS 2020: 5; original emphasis).

UN and EU Mediation Approaches Compared: Similarities, Differences, Limitations
The evolution of UN and EU conceptual frameworks and institutional capacity for mediation indicates that both organizations have taken similar pathways concerning the development of their profiles in peace mediation.Indeed, there are some major similarities between the two organizations' approaches to mediation, but also important differences.These need to be analyzed in order to understand the conceptual and institutional dimensions that underpin UN-EU cooperation on peace mediation in concrete cases.
At the conceptual level, the UN's and the EU's definitions of mediation are very similar.Both understand mediation as a voluntary process through which third parties assist conflict parties in preventing, managing, resolving and transforming a conflict (UN 2012: 4; Council of the EU 2020b: 3).Although the EU's mediation concept2 goes into more detail concerning the difference between mediation, facilitation, mediation support and dialogue compared to the UN Effective Guidance for Mediation, both documents develop a very similar understanding of mediation.In addition, the UN and the EU share a strong commitment to major norms and values in peace mediation (interviews 5, 6).3In particular, both subscribe to inclusivity, a human rights-based approach to mediation, a conflict sensitivity built on rigorous conflict analysis, local and national ownership, the promotion of gender equality and women's empowerment through mediation, and the principles of coherence, complementarity, and coordination in peace mediation efforts (UN 2012: 6-20;Council of the EU 2020b: 4-9).As the new EU mediation concept highlights, an awareness of climate-related and environmental aspects of peace mediation and the protection of cultural heritage are additional normative commitments.
Moreover, another similarity of the conceptual evolution of UN and EU mediation is that both organizations had to invest strongly in anchoring mediation as a distinct practice and concept within their bureaucratic cultures.As Herrberg (2021: 134) notes, the underlying reason for this challenge is that "peace mediation -both as a concept and practice -suffers from a degree of ambiguity and opacity in the world of diplomacy, difficult to pin down as a Bergmann International Negotiation 27 (2022) 418-447 subject of enquiry or practice."Despite being mentioned by the UN Charter as a key instrument of peaceful dispute settlement (UN 1945: Art. 33), mediation in the UN has traditionally been "regarded as synonymous with diplomacy rather than as a specialized activity" (Nathan 2010: 1).Consequently, once established, the MSU staff members had to invest considerable time and energy in convincing colleagues across the UN that a mediation approach to conflicts requires more than traditional diplomatic and interpersonal skills.As demonstrated by Convergne (2016: 186-193), the strategy of portraying mediation as a specialized, apolitical and technical activity that requires generic knowledge was instrumental in the MSU acquiring authority within the UN system and vis-à-vis external partners, and in creating a specific thematic niche for itself.
For the EU, and the EEAS MST in particular, it has been equally if not more challenging to establish the relevance and conceptual distinctiveness of mediation within its own organization and among relevant EU institutions (Herrberg 2021: 136-139).In particular, raising awareness about the specific expertise and skills required for mediation, clarifying the differences between (direct) mediation and mediation support, and demonstrating the need to align mediation with other EU foreign policy approaches in a quest for coherence and complementarity were "a constant uphill battle" for MST staff members in the initial years following the adoption of the 2009 Concept (interview 8; see also Sheriff et al. 2013).The adoption of the new mediation concept in 2020, and the Council Conclusions on EU Peace Mediation, in particular, show, however, that this endeavor has started to pay off.Hence, it appears that both the UN and EU have successfully managed to establish mediation as a distinct concept and practice within their own cultures of conflict prevention and peacebuilding work.
There is one stark difference in the conceptual approach to mediation of the two organizations, however, which concerns an emphasis on the normative commitment to impartiality.The UN Guidance for Effective Mediation highlights impartiality as fundamental to successful mediation and emphasizes that "a mediator should be able to run a balanced process that treats all actors fairly and should not have a material interest in the outcome" (UN 2012: 10).In contrast, the EU's mediation concept does not mention impartiality as a norm in peace mediation.4Clearly, this omission does not mean that the EU does not aspire to adopt a balanced and fair approach towards mediating between conflict parties in any given conflict (EEAS, 2020: 26).Rather, it points to the important fact that in the case of the EU, mediation is part of its broader foreign policy toolbox and thus closely aligned to its foreign policy objectives.As the concept states, "mediation represents an effective avenue for the EU to pursue its foreign and security policy objectives" (Council of the EU 2020b: 4).This embeddedness of peace mediation in a broader foreign policy agenda implies that the EU cannot claim to follow the norm of impartiality as understood by the UN in terms of not having any material interest in the outcome of a mediation process (UN 2012: 10).Rather, it means that in many mediation cases the EU takes clear foreign policy stances that may potentially undermine its acceptability as a mediator vis-à-vis a particular conflict party, as will also be shown in the case study of the Geneva International Discussions (see below).At the same time, perceived partiality may not necessarily be an obstacle to effective EU mediation, as existing research on the role of bias and impartiality in international mediation demonstrates (Wallensteen & Svensson 2014: 320-321;Elgström et al. 2018).As an interviewee states, "it does not mean that the EU is doomed to fail as a mediator (…) It very much depends on the specific context whether the lack of impartiality will make a difference for EU mediation or not" (interview 7).However, an important implication of this difference in the extent to which mediation is embedded in foreign policy is that the UN may have access to conflict theaters in which the path for an EU mediation mandate might be blocked or at least be more difficult to pursue due to perceived partiality (interviews 6, 7, 8).
The extent to which mediation is conceptually embedded in a broader foreign policy agenda also has implications for the institutional and operational level of peace mediation.Given that the EU is able to combine its mediation efforts with other external policy instruments to influence the parties' willingness to compromise -such as the (threat of) imposition of sanctions or the promise of economic and political incentives in the context of EU development or enlargement policy -the EU brings a certain degree of leverage to the negotiation table.This is particularly the case for EU mediation engagements in its immediate neighborhood to its south and east.UN mediators, by contrast, largely rely on the UN Security Council and its members, or thirdparty countries, to "lend" their leverage over conflict parties to the mediation effort and thereby encourage disputants to find a peaceful settlement to their conflict.The effects of both limited availability of its own sources of leverage and the dependency of UN mediation efforts on external providers of leverage has been seen, for example, in the cases of Syria or Kosovo (Lundgren 2016;Bergmann 2018).Hence, when comparing the two organizations' potential limitations and weaknesses in peace mediation, it becomes clear that a lack of perceived impartiality may hamper EU mediation efforts in certain cases, but the availability of additional foreign policy leverage can prove to be an asset for EU mediators.In contrast, the fact that UN mediation is not embedded in a foreign policy agenda helps to promote a perception of the UN as an impartial mediator, which may facilitate its access to certain conflict arenas.At the same time, it also means that UN mediators are very much dependent on UN Security Council and its members, or third-party countries, to provide additional leverage if a more directive mediation engagement is necessary.
Concerning the leeway that the organizations' mediation support structures have vis-à-vis the member states of their organizations and within their bureaucracies, however, interview partners suggested that the UN MSU may enjoy a higher degree of autonomy compared to the EU MST in its daily mediation support practices.As one interviewee noted, a possible explanation for the EU MST having "its hands tied" more often on certain occasions is that EU member states may monitor very closely the EEAS' endeavors in a particular conflict case, particularly if member states pursue different interests and diverge in their analysis on how the EU should deal with a given conflict (interview 5).Hence, as EU foreign policy is still very much a domain dominated by member states, the strong embeddedness of mediation in the EU's foreign policy toolbox can, in some instances, work as a constraining factor for mediation support engagement undertaken by the EEAS, even if the latter is framed as a technical rather than political matter.
At the same time, being able to draw on an extensive network of foreign policy structures deployed in conflict zones -including EU Delegations and member states' embassies which closely coordinate on political affairs on the ground -is a clear strength of the EU, and the lack of similar structures is a challenge and limitation for the UN.Whereas EU Delegations have a clear political mandate and are supposed to contribute to conflict analysis and crisis prevention, the UN does often lack a similar, political on-the-ground structure.5 In sum, this comparative analysis of UN and EU mediation capacities and structures shows that both organizations have followed similar trajectories.The main difference between their mediation approaches is that the EU's mediation practice is more deeply embedded in a broader foreign policy agenda compared to that of the UN.Hence, the evolution of the EU's peace mediation capacity can also be interpreted as an adaptation of established approaches -such as that of the UN -to its own institutional context.This adaptation seems to have further progressed with the new EU mediation concept, as it elaborates more thoroughly on peace mediation's place within EU foreign policy than the 2009 Concept does.

Practices of Cooperation, Institutional Learning and Competition
One factor explaining the conceptual and institutional similarities between the UN and the EU is that European states have been important drivers of strengthening mediation at the UN.Among others, Austria, Finland, Germany, Sweden and the United Kingdom have also been important advocates of building up peace mediation capacity within the EU (interviews 5, 8).Noteworthy in this regard is that, together with the Norwegian Ministry of Foreign Affairs, the European Commission has been the main funder of the SBT.Through its financial support of the SBT, "the European Union has also demonstrated a firm commitment for EU-UN cooperation" (Herrberg, Packer & Varela 2015: 11).
At the same time, there are also certain similarities concerning the institutional capacities of the UN and the EU that are a result of institutionalized cooperation between the two organizations.The latter has led to instances of learning, primarily in the way of knowledge and procedures transferred from the UN to the EU.
Since the EU MST's creation in 2011, it has developed close institutional ties with the UN MSU, resulting in different formats of cooperation.Regular exchanges have taken place between individual desk officers and focal points for each other's organization within the MST and the MSU, also involving colleagues from the UN representation in Brussels and the EU delegation at the UN in New York.Moreover, an annual strategic dialogue between the UN and EU on conflict prevention and mediation has been established, as well as an institutional culture of sharing experiences and knowledge -also in the context of the annual Regional Organizations Meeting on Mediation, the 2019 version of which was co-organized by the EU and the UN (College of Europe 2019).
The described institutional cooperation between the MSU and the MST has resulted in instances of institutional learning by the EU from the UN, often based on experiences and expertise of individuals who have served both in the UN and the EU, and accelerated by institutional pressures for professionalization.This mechanism of inter-organizational learning has been described by sociological institutionalists as "normative isomorphism" (DiMaggio & Powell 1983), explaining why international organizations often tend to have similar institutional structures for a specific policy task.Indications of "normative isomorphism" are also clearly observable in the field of peace mediation and mediation support.In setting up the working procedures of the EU MST and organizing the division of labor among the initial four colleagues, the team very much followed the model of the UN MSU and also drew on the experiences of one of their members who had worked at the UN MSU before joining the EEAS (Lanz et al. 2017: 13;interviews 5, 6, 8).Although the idea of setting up its own roster of mediation experts was initially considered, the EEAS decided instead to gather external support through the abovementioned ERMES project and an EEAS Framework Contract for Conflict Prevention and Mediation Support (Lanz et al. 2017: 14).Nevertheless, the question of how to set up structures to draw on the expertise and support of external mediation practitioners was very much debated in relation to the setup of the UN MSU, its roster of mediation experts, and the SBT (interview 5).
Another example of the EU MST, drawing inspiration from the UN MSU's experiences, is the recently adopted EEAS Peace Mediation Guidelines document.In the process of drafting the document, the EU MST also drew on the UN Guidance for Effective Mediation and consulted with UN MSU colleagues about their experiences in the implementation of the guidance (interview 6).The reference to the UN Guidance as being complementary to the EEAS Peace Mediation Guidelines is thus no coincidence, but rather both a result of this inter-organizational consultation process and the EU's interest in reconfirming the UN as a key reference point in international mediation (EEAS 2020: 5).
Notwithstanding these institutionalized forms of cooperation and learning, elements of competition and rivalry are not fully absent from UN-EU relations in the domain of peace mediation.In general, a certain degree of competition among providers of international mediation is not surprising.As Lanz and Gasser (2013:13-16) argue, conflicting interests between states offering mediation, clashing norms, as well as overlapping mandates between international and regional organizations can drive mediator competition.In the case of the UN and the EU, there are no indications of fierce competition or disputes between the two organizations on particular issues or concrete mediation cases.At the same time, the evidence suggests that the question of under what circumstances and in what cases the EU may draw on the UN Standby Team of Senior Mediation Advisers (SBT) has been an issue of debate between the two organizations in a few instances.Given the EU's strong financial International Negotiation 27 (2022) 418-447 support to the SBT, UN MSU officials sometimes had to work against the perception among some EU officials that the SBT was an instrument "owned" by the EU and the UN MSU's role was thus limited to that of an "implementation partner."As a result, it has happened that the EU requested support from the SBT for individual mediation efforts, but the UN MSU decided against the deployment of SBT members because of different understandings of when and under what conditions SBT deployment might be most helpful and appropriate (interviews 5, 6, 7).
Moreover, a certain degree of competition between the UN and the EU may also arise when both organizations are involved in mediation support efforts in the same conflict (see also section below).In those situations, parallel and uncoordinated efforts may lead to overlaps and inefficiencies in supporting ongoing peace processes.Yemen is an interesting case to look at in this regard.In the first phase of mediation efforts from 2011 to 2012, the EU was directly involved in mediation, in conjunction with the Gulf Cooperation Council and the UN, leading to an agreement with President Saleh on a National Dialogue Conference (Girke 2015).Since 2012, the EU has focused its activities on supporting the Office of the Special Envoy of the Secretary-General for Yemen (OSESGY) by facilitating informal meetings with conflict parties in Amman and Sanaa, and providing funding and technical expertise to Track II mediation efforts involving civil society organizations, in part, as a means to promote the latter's strategic engagement with the UN (Herrberg 2021: 144).However, at a certain stage of the conflict when the UN-led mediation efforts stalled in 2018, the EU's parallel activities led to Yemeni accusations of the EU being biased towards the Houthis, which also put strains on EU coordination with the UN and other international partners (Asharq Al-Awsat 2018).In the end, however, these tensions were defused and have led to even closer coordination between the UN and the EU since 2018 (Herrberg 2021: 144).
In sum, the findings demonstrate that the institutionalization of cooperation between the UN and the EU on peace mediation has fostered mutual understanding and facilitated institutional learning in terms of knowledge and procedures transferred from the UN to the EU.At the same time, there is also evidence of a moderate degree of competition between the UN and the EU, although it has primarily revolved around individual cases and is no indication of a systemic rivalry between the two organizations in the peace mediation domain.As the level of cooperation and competition between the UN and the EU very much depends on individual conflict contexts and mediation constellations, the next section develops a typology for mapping different arrangements of UN-EU cooperation in mediation practice.

Mapping UN-EU Cooperation Arrangements in Mediation Practice
In mediation research, the standard definition by Bercovitch et al. (1991: 8) refers to mediation as a "process of conflict management where disputants seek the assistance of, or accept an offer of help from, an individual, group, state or organization to settle their conflict or resolve their differences without resorting to physical force or invoking the authority of the law."Both the UN and EU in their own definitions of mediation follow this broad understanding of mediation, although both acknowledge that they may take different roles in their mediation activities -from direct mediation engagement to various mediation support roles (UN 2012; Council of the EU 2020b).Hence, when analyzing UN-EU cooperation on peace mediation in concrete cases, we would expect the UN and EU to cooperate with each other under different constellations, based on the individual mediation roles they have taken in a given conflict.
To develop a typology of different constellations of UN-EU cooperation in peace mediation, I propose to differentiate between direct mediation and mediation support, being aware that, in reality, third-party actors may adopt roles that lie in-between these two types of engagement.Direct mediation implies that the UN or the EU are recognized by the conflict parties as sole, lead or co-mediators, and are directly involved in the negotiations occurring in the context of the mediation process.In contrast, mediation support is understood as any activities to assist third-party mediators in the search for a negotiated settlement.In other words, it refers to activities to support an ongoing mediation effort led by another actor and can involve diplomatic, operational, technical and financial support to the third-party mediator and the mediation effort.Through these different means, providers of mediation support can assist mediators by providing diplomatic endorsement and legitimacy, lending leverage to them or empowering them in terms of operational and financial capacities for mediation (Müller & Bergmann 2020).6In a nutshell, one could roughly distinguish between "at the table" mediation (direct engagement) and instances of "around the table" or "beyond the table" activities (mediation support) (Herrberg 2021: 139;Lanz et al. 2017).
Following this simplified distinction of two different types of mediation roles, one can think of four different ideal types of UN-EU cooperation in peace mediation in practice (see Table 2).
International Negotiation 27 (2022) 418-447 Given the secrecy of many international mediation efforts, the full universe of cases of UN-EU cooperation in mediation efforts is hard to establish.Nevertheless, we can identify cases that more or less correspond with the four ideal types outlined in the table.First, cases of UN-EU co-mediation in a given conflict are rare.However, at least two mediation incidents correspond with Category 1.In summer 1992, the European Community and the UN called for the International Conference on the Former Yugoslavia in London.The mediation effort lasted for more than two years and broke down after the Bosnian Serbs rejected the joint EC-UN proposal for a comprehensive settlement: the Vance -Owen Peace Plan, proposed by the lead mediators David Owen (EC/ EU) and Cyrus Vance (UN) (Lucarelli 2000: 37-62).A second case of UN-EU co-mediation is the Geneva International Discussions on South Ossetia and Abkhazia, ongoing since 2008, in which the UN and the EU, together with the Organization for Security and Co-operation in Europe (OSCE), take up the role of co-chairs (see case study below).
Concerning the second ideal type -UN mediation efforts supported by the EU -Syria is a relevant example, where the UN has been (co-)mediating negotiations to end the civil war in the country, represented by Joint Arab League/ UN Special Envoys Annan (2012) and Brahimi (2012-14) and UN Special Envoys de Mistura (2014-19) and Pedersen (since 2019) (Zartman 2020).The EU has been supporting the mediation efforts of the UN, in particular the Geneva Process and related advisory bodies, through the Comprehensive Syria Peace Support Initiative, providing financial and technical support to Track I and Track II dialogue and mediation processes (European Commission 2016).As mentioned above, Yemen is another case in which the EU has supported a UN-led mediation process.Empirical examples of UN-EU cooperation that correspond to the third ideal type -an EU mediation initiative that is supported by the UN -are more difficult to establish.This is partly because, in cases where the EU has been in the lead as a mediator it has been less dependent on external resources for mediation due to its own capacity and expertise, and partly because in some cases where the UN was interested in providing mediation support to the EU, the member states rather than EU institutions and actors have been in the lead (interviews 5, 6, 7).One potential case of the UN being in a supporting role to EU mediation efforts is the EU-facilitated Belgrade-Pristine dialogue, ongoing since 2011, although this support has not involved the UN MSU.Rather, in some negotiation rounds and working groups, representatives of the United Nations Interim Administration Mission in Kosovo (UNMIK) participated in the discussions and thus supported the facilitation of agreements (Blockmans 2014: 434).Nevertheless, in this case, the EU's responsibility as a lead mediator, and the "division of labor" with the UN, was unambiguously clear, as it was the UN General Assembly Resolution 64/298 that mandated the EU to facilitate a dialogue between the parties (United Nations General Assembly 2010).
Finally, a fourth type of UN-EU cooperation in peace mediation occurs when both organizations provide mediation support to other third-party actors' ongoing mediation efforts.This can be understood as a case of UN-EU cooperation if both organizations coordinate their support, at least to a certain extent, even if the nature and scope of their individual support might differ.One example of this category is the mediation effort by the Intergovernmental Authority on Development (IGAD) in South Sudan's civil war (since 2013), which has been supported both by the UN and the EU.The UN has provided, inter alia, technical mediation support to IGAD via its Special Envoy for Sudan and South Sudan, who has been advised by the MSU on justice and accountability issues, "including on the specific language to be utilized in the agreement for addressing conflict-related sexual violence" (United Nations 2017: 12).The EU has supported IGAD-led mediation efforts through its Special Representative for the Horn of Africa and his team coordinating with IGAD mediators, member states, and other international partners, and has also provided financial support to the first phase of the mediation through the Early Response Mechanism (ERM) of the African Peace Facility (Müller & Bergmann 2020).
Taken together, these empirical illustrations show that the reality of UN-EU cooperation on peace mediation is multi-faceted and can take different forms, depending on the roles both organizations take in a given peace process.The constellation of mediator roles can also influence both the degree of coordination and competition between the two organizations in a given case.In a situation in which only one of the two organizations has a formal mandate to lead a mediation process, we may expect less challenges for coordination given a clear division of labor.We may expect more potential for competition between the UN and the EU in cases in which either both have a formal mediation mandate and need to arrange themselves in a multi-party mediation effort or find themselves both in a mediation support role.The latter constellation might bear the greatest potential for mediator competition between the UN and the EU, as mediation support is usually not based on formal mandates and may be offered ad hoc in parallel and often with different underlying objectives, which increases the need for coordination between the two organizations.
Although the overview of empirical examples is not exhaustive, it makes clear that UN-EU cooperation on peace mediation is very much contextspecific and that its effectiveness and challenges need to be studied on a caseby-case basis.In doing so, the next section takes a closer look at one particular case from which we can gain some empirical insights into how effectively the UN and the EU cooperate on mediation and what effects it has on mediation dynamics.For this purpose, we focus on the case of the Geneva International Discussions -a case of UN-EU co-mediation together with the OSCE.Clearly, the generalizability of findings from a single case study is limited and not easily transferrable to other cases of UN-EU cooperation in peace mediation.Nevertheless, as the GID have been ongoing for over a decade and are characterized by a formalized mediation setting since 2008, they seem to be an apt case for an initial study of the effectiveness of UN-EU cooperation in peace mediation.
The analysis builds on triangulation across multiple data sources, including official documents issued by the EU and the UN, policy reports produced by think tanks and NGOs, and academic literature, as well as nine semi-structured interviews with former and current EU and UN officials under a promise of anonymity.The interviews were conducted in two different time periods.A first series of face-to-face interviews on the GID case (interviews 1-4) were held in 2015 and 2016 when the author conducted a study on the EU's role in the mediation process.The second batch of interviews on the case (interviews 5-9) was conducted in February and March 2021 using videoconferencing to complement the initial findings and focus particularly on the issue of UN-EU cooperation in the GID framework.

Case Study: UN-EU Cooperation in the Geneva International Discussions
This article defines effectiveness in mediator coordination as the degree to which the mediators cooperate in terms of (internal and external) communication,

Bergmann
International Negotiation 27 (2022) 418-447 the pooling of resources, and the pursuit of a joint mediation strategy.Effectiveness is best understood as a continuum of degrees.It takes a high value when there is intensive communication between the mediators, joint messaging towards the conflict parties and external audiences, and close interaction that involves the pooling and sharing of resources, knowledge and expertise in the pursuit of a joint strategy.A medium degree of mediator coordination means that communication and interaction between the mediators have been established to a certain extent, but it does not exclude the possibility of the mediators taking unilateral steps, sending contradictory messages or withholding information and resources.A low degree of mediator coordination implies that there is very limited and infrequent -or even no -communication and interaction between the mediators, as well as contradictory messaging to external audiences and the conflict parties.Hence, for the case of the Geneva International Discussions (GID), we investigate the extent to which the co-mediators have shared relevant information, pooled their resources to increase their leverage vis-à-vis the conflict parties, and ensured that the composite mediation team has spoken with a "single voice" and moved into the same strategic direction.

Assessing the Effectiveness of Mediator Coordination in the GID
Since October 2008, the UN, the EU, and the OSCE have been co-chairing the GID, in which government representatives of Georgia, Russia, the United States, and representatives of South Ossetia and Abkhazia participate.The launch of the GID was a result of agreement on a six-point plan signed by the presidents of Georgia and Russia on 15/16 August 2008, following their countries' five-day war in August 2008.In subsequent agreements on implementing measures brokered on 8 September by French President Sarkozy, the parties agreed to the launch of international discussions so as to deal with "modalities of security and stability in the region, the question of refugees and displaced persons on the basis of recognized international principles, and the practice of post-conflict order, and any other issues on the consensus of the parties" (Le President de la Republique Francaise 2008b: 2; translation from the French original).
Given the prominent role played by France in brokering the ceasefire agreement, it was evident to policymakers in Brussels and European capitals that the EU had to continue its engagement in managing the conflicts in Georgia.Although it had not been specified in the agreements who should chair the mediation process, the EU and its member states decided to go for an inclusive approach vis-à-vis the OSCE and UN, being particularly responsive to the OSCE, which had also called for the establishment of a mediation platform to ensure security and stability in the Caucasus (Mikhelidze 2010).In addition, including the OSCE and the UN in the talks was considered as a way of strengthening the legitimacy of the format and as a possible opportunity to tap into the rich field experience of these two organizations on the ground (Merlingen & Ostrauskaite 2009: 23).
Despite variations over the different phases of the mediation effort ongoing since 2008, the overall empirical evidence points to a high degree of coordination between the EU, the UN, and the OSCE, based on regular and intensive communication, common positions on the management of the GID, and the beneficial use of joint resources and knowledge.
Concerning regular communication within the multi-party mediation team, the three co-chairs and their teams meet four times a year in the context of the GID meetings, conduct four to six joint missions every year to the conflict region and Moscow, and have an additional two to three meetings held exclusively for strategic mediator coordination (interviews 1, 2; Panchulidze 2020).In addition, mediators also hold regular telephone conferences to keep each other updated on the latest developments in the conflict region (interview 2).The communication and coordination between the organizations in these meetings have been described as very cooperative and mutually beneficial, as they have ensured that the co-chairs turn up with a joint position on how to manage the negotiations at every meeting of the GID (interview 1, 9; Salber 2016: 4).Moreover, interviewees reported that the three organizations have been circulating papers and drafts of proposals for compromises among themselves in order to benefit from each other's input before presenting the proposals to the conflict parties (interviews 1, 2).
Regarding the pooling of their resources and expertise, EU officials have frequently emphasized that one key asset of the UN and the OSCE has been their longstanding experience in managing the conflicts over South Ossetia (OSCE) and Abkhazia (UN) (Morel 2009: 1;Salber 2016: 4;interview 3).As for the UN, its Observer Mission in Georgia (UNOMIG) had, by 2009, involved a maximum of 459 staff members, who had been engaged for almost 16 years in monitoring the Abkhaz-Georgian ceasefire agreement and brokering a peaceful settlement of the conflict (UNOMIG 2009).The UN's experience in managing the Abkhaz-Georgian conflict has been helpful in various ways.For example, the establishment of Incident Prevention and Response Mechanisms (IRPMs) was modelled on similar mechanisms organized by the OSCE and the UN in the period before the August war of 2008 (interview 3).Moreover, the UN Representative to the Geneva International Discussions (UNRGID) and her team are also responsible for preparing, convening and facilitating the meetings of the Gali IPRM for Abkhazia, in which European Union Monitoring Mission (EUMM) representatives participate as observers.The UN's role in the Gali IPRM and its ability to regularly visit Abkhazia and exchange with Abkhazian authorities make the UN an important information provider for the whole mediation team (interview 9;Herrberg 2021: 143).In this context, the UN's history of being involved in the management of the Georgian-Abkhazian conflict has helped the UN to establish a trusted working relationship with Abkhazian authorities and helped to facilitate a perception of the UN as an impartial mediator.To a certain extent, this balances the perceived partiality of the EU as a mediator, a perception which is particularly shared by Abkhazian representatives, given that the EU has unambiguously stated its support for Georgia's territorial integrity and has adopted a non-recognition policy (interviews 1, 2, 9).
Another example is the co-moderator role played by the United Nations High Commissioner for Refugees (UNHCR) in the second working group of the GID.Having provided emergency humanitarian assistance to refugees and Internally Displaced Persons (IDPs) that have fled Abkhazia since 1993, the UNHCR has gathered tremendous expertise on refugee/IDP-related matters and has served as an important provider of objective information on the situation of IDPs and refugees in Georgia (UNHCR 2009;United Nations General Assembly 2014, 2016).
A third example for close UN-EU cooperation in the broader context of the GID is the Confidence Building Early Response Mechanism (COBERM), a fund managed by UNDP and financed by the European Commission (Herrberg 2021: 143).The COBERM fund has supported initiatives to build some level of confidence and trust between the parties.An example of such an initiative is the Archives Without Borders project.In the context of this project, Georgian authorities, in a meeting of the GID, handed over copies of Abkhazian archive documents restored in the Georgian State Archives to the Abkhazian authorities, thus helping them to rebuild their archives in Sukhumi that were burnt during the 1992 war (Mikhelidze 2012: 12).
Overall, there are very few indications of competition or rivalry between the mediators in the GID setting.Yet, some tensions behind the scenes between the EU and the UN emerged in 2012 when Abkhazia suspended its participation in the IPRM meetings in Gali due to the accusations of bias against EUMM Head of Mission General Tyszkiewicz.Following the breakdown of the IPRM meetings in Gali in 2012, the "UN felt threatened" in its mediator role concerning the Abkhazian conflict, given the EU's refusal to replace Tyszkiewicz immediately and thus preventing the UN to continue to play its important role as facilitator of the IPRM Gali meetings (interview 1).According to the interviewee, the tensions between the EU and UN were defused when a International Negotiation 27 (2022) 418-447 compromise was found -not extending the mandate of General Tyszkiewicz in 2013 -thus both saving the EU's face and allowing for a continuation of the IPRM meetings thereafter.
Moreover, in the course of the mediation process, both the UN and the EU launched some their own initiatives to move the negotiations forward, although these attempts to broaden the agenda sought to complement, rather than contest, the joint approach of the multi-party mediation effort.For example, both EU Special Representatives Pierre Morel (2008-2011) and Philipp Lefort (2011-2014) sought to introduce cooperation on economic issues as a third agenda item of the GID, but were blocked in their attempts by the conflict parties who did not want to open up another agenda item (interviews 1, 2).Another example is the increased focus on the implications of the Women, Peace, and Security (WPS) Agenda for the GID since 2020, which was introduced by an initiative of UN Representative to the Geneva International Discussions, Ayşe Cihan Sultanoğlu, but endorsed and supported by all co-chairs (interview 9; UN Women 2020).
Taken together, these findings suggest that the cooperation between the UN, the EU, and the OSCE has been very smooth and beneficial for the overall organization and steering of the mediation effort.The findings show that UN-EU coordination occurs at multiple levels -at the level of UN/EEAS headquarters staff, their regional bureaus, the level of co-chairs, and at the level of staff members of the UN and EU Special Representatives participating in the GID.As EU Special Representative Salber (2016-19) noted in a public speech in 2016 on the degree of cooperation between the three organizations in this co-mediation case: I see only advantages in having the EU, the OSCE and UN Co-Chairing the GID together […] I am convinced that the close EU, OSCE and UN cooperation that we have established for the conduct of the GID, is something that can serve as an example in other conflict management and resolution processes.

Implications of Mediator Coordination for Mediation Dynamics and Outcomes
Despite a high degree of effectiveness in terms of mediator coordination, it is important to note that the GID has not yet led to any major tangible outcomes with respect to its mandate -to find agreement on modalities for security and stability in the region, and with respect to the question of refugees and internally displaced persons (Panchulidze 2020).In light of this fact, one may conclude that the close cooperation between the co-mediators has not had any relevance in terms of the outcome of the mediation.Indeed, one could argue that the high degree of mediator coordination in terms of information sharing and pooling of resources has not made much difference to the conflict parties' calculations and their willingness to compromise.The latter is indeed the most important factor explaining the lack of success of the GID, as a zone of agreement between the parties is virtually non-existent.On the issue of the non-use of force, Russia strongly refuses to sign an agreement with Georgia or make a unilateral pledge on the non-use of force based on the argument that it is not a conflict party.In turn, Georgia refuses to sign bilateral agreements with Abkhazia and South Ossetia.On humanitarian issues, Abkhazian participants have blocked any serious discussions on the return of refugees because they perceive Georgia's claims for refugee return as a threat to Abkhazia's security (Crisis Group 2013: 13).Hence, due to a low level of willingness to find common solutions on those substantial issues, a compromise agreement between the conflict parties is out of reach.
Despite these obstacles to reach any tangible outcomes in the mediation, there is also evidence to suggest that the close cooperation among the mediating organizations has positively influenced their ability to jointly steer and conduct the mediation efforts in Geneva over more than a decade.With the three organizations taking joint positions against the walk-outs of participants from GID sessions and calling jointly for the continuation of discussions despite perceived deadlocks (OSCE 2012a(OSCE , 2012b)), mediator coordination has strongly contributed to the continuation of the mediation process (interviews 1, 2, 4, 9).From a counterfactual point of view, it appears that a low degree of mediator coordination could have actually been exploited by the conflict parties and substantially increased the chances of the breakdown of negotiations.The contributions of the UN to the mediation effort have been particularly felt in the field of security management and stabilization.It has played an important role in facilitating the IPRM Gali meetings and provided important input for the discussions in the first working group on security modalities (Salber 2016: 4; interview 3).Thus, the pooling of resources and sharing of information has positively contributed to the stabilization of the security situation and the prevention of renewed escalations of violence.
In sum, the case of the GID is an illustrative example of how cooperation between the UN and EU (together with the OSCE) looks like in practice and how it influences the steering and organization of a co-mediation effort.At the same time, the case clearly shows that close cooperation between the mediators is only one of several ingredients of successful mediation, and is certainly not sufficient on its own to achieve conflict settlement.Moreover, in other cases of UN-EU cooperation in which the organizations' mediation mandate is less clear-cut and formalized, additional challenges to coordination may arise.Hence, this speaks again to a plea for context-specific research on additional cases of UN-EU cooperation in peace mediation, which might be taken up by future research.

Conclusions
This article set out to explore the under-researched topic of UN-EU cooperation in peace mediation, analyzing its conceptual and institutional foundations as well as its empirical manifestations in concrete cases.In doing so, we provided an overview of the evolution of the two organizations' approach to mediation, focusing on their main strategic and conceptual tenets and their institutional capacities for conducting mediation and mediation support activities.The results of this systematic comparison demonstrate that the UN and the EU have followed similar trajectories in this evolution and share a similar conceptualization of mediation, as well as a similar institutional architecture for mediation support within their own bureaucracies.Institutional learning, accelerated by pressures for professionalization and transmitted through experiences and expertise of individuals who have served both in the UN and the EU, is a key explanatory factor for these similar pathways.A major difference concerns the embeddedness of mediation in a broader foreign policy agenda in the case of the EU compared to the UN.This may lead to advantages on the side of the EU in terms of the possibility of drawing on an established field presence on the ground, and its ability to combine mediation efforts with other foreign policy instruments, such as sanctions or development assistance in order to create additional leverage.At the same time, it may also undermine the EU's acceptability as a mediator vis-à-vis conflict parties if it is perceived to be biased, a challenge that is potentially less relevant to the UN, which has credibly stressed its commitment to impartiality as a cornerstone of its mediation engagements.
Turning to the cooperation of the two organizations in concrete cases of peace mediation efforts, the article has provided a first systematization of the different constellations in which the UN and EU may interact with each other, depending on the individual mediation roles they have taken up in a given case.The case study on the Geneva International Discussions -a case of UN-EU cooperation within a multiparty mediation effort -shows what effective coordination can look like, but also reminds us that the degree of cooperation among mediators is only one piece of the puzzle that is successful mediation.

Table 1
Key steps of the conceptual and institutional evolution of UN and EU mediation since the 2000s Source: Author's compilation Downloaded from Brill.com11/29/2022 11:28:41PM via free access

Table 2
Ideal types of UN-EU cooperation in peace mediation Source: Author's compilation