Jump directly to page content

Mission Impossible?

UN Mediation in Libya, Syria and Yemen

SWP Research Paper 2018/RP 08, 15.10.2018, 60 Pages Research Areas

The upheavals in the Arab world since 2011 have led to civil wars in three countries: Libya, Syria and Yemen. In all three cases, the United Nations have tried to mediate agreements between the conflicting parties to bring about peace through power-sharing. In this endeavour, the UN can lean on its broad experience in mediation efforts to end civil wars.

In the three conflicts examined here, however, the UN’s attempts at reso­lution through power-sharing have failed. In Yemen and Libya, power-shar­ing agreements have not prevented conflicts from lingering on or violence from breaking out again. In Syria, the UN has not even managed to bring together the Syrian parties in the civil war for direct talks. The altered mili­tary balance of power has rendered the initial goal of a political transition unrealistic.

This study answers the following questions: Why are these three conflicts so resistant to resolution efforts? What are the specific aspects of the conflict configurations that impede UN efforts? What factors in the UN approach are obstacles to a successful conclusion? What lessons can be learned for future mediation efforts? And how can Europe contribute to progress in this area?

Dr Muriel Asseburg is a Senior Fellow in the Middle East and Africa Division at SWP

Dr Wolfram Lacher is a Senior Associate in the Middle East and Africa Division at SWP

Mareike Transfeld is a doctoral student at the Berlin Graduate School Muslim Cultures and Societies

Issues and Recommendations

Mission Impossible?
UN Mediation in Libya, Syria and Yemen

The upheavals in the Arab world since 2011 have led to civil wars in three countries: Libya, Syria and Yemen. In all three cases, the United Nations (UN) are trying to mediate agreements between parties in the conflicts to bring about peace through power-sharing. The UN has an established leadership role in media­tion efforts to end civil wars. Since the 1990s, some of the most protracted civil wars have been settled under the UN’s overall command or with its partici­pation. Most of these negotiated settlements were based on power-sharing agreements.

Not so for the three conflicts analysed here: all UN efforts to end them through power-sharing have failed. In Yemen and Libya, power-sharing agree­ments have not prevented conflicts from continuing or resuming. In Syria, the UN has not even managed to hold direct negotiations on power-sharing between the parties in the civil war. The changed military balance of power has now made the initial goal of a political transition unrealistic.

Why are these three conflicts so resistant to reso­lution efforts? Which aspects of these conflicts im­pede UN efforts? Which factors in the UN approach hinder progress? What lessons can be learned for future mediation efforts? And how can Europeans help to move the UN’s attempts at mediation for­wards?

The conditions under which the UN is trying to negotiate a resolution to the three conflicts are extra­ordinarily difficult. In all three conflicts, the balance of power and alliances between multiple actors change constantly. Many local actors in these con­flicts are not seriously committed to negotiations because they receive support from regional and great powers. All three conflicts are not only power struggles between local forces, but also offer an arena to rival foreign powers. This international dimension makes it more difficult to include all relevant local and external actors in the negotiations, because powerful states resist this. Furthermore, UN mediators are constrained by a Security Council (SC) that either disagrees about the right way to solve the conflict (as for Syria) or has associated itself with one party in the conflict (as in Yemen), which rules out the UN as an impartial mediator. Finally, enforcing agreements through peacekeeping forces was never a realistic option in these three countries – not only because there was no unanimous support from the five permanent mem­bers of the SC, but also because the majority of local actors in these conflicts reject the presence of international peacekeepers. Taken together, this meant that, from the outset, the UN’s efforts to bring about a negotiated settlement had only a very small chance of success – particularly in Syria.

However, alongside these adverse conditions, mis­takes and dilemmas in the mediation strategy itself as well as a lack of support from Western governments have also inhibited success. In Syria, the interests of regional and great powers thwarted negotiations that would have included all relevant parties, but in Libya and Yemen, UN mediators failed to involve important conflict parties in the talks. In both cases, the (inevi­tably unsustainable) agreements were pushed through with excessive haste. Insufficient support from local actors was offset by international legitimacy, gained from Security Council resolutions. At the time of signing, the agreements were seen as risky gambles, but during the unsuccessful attempts to implement them, it became clear that external actors either had little leverage over the conflicting parties, or failed to use the leverage they had. The use of targeted sanc­tions to deter potential spoilers in Libya and Yemen was particularly ineffective, if not counterproductive. Insufficiently inclusive agreements, combined with the subsequent failure to enforce them, led the unity governments of Libya and Yemen to become mere parties to the conflict. The UN thus turned from mediator to supporter of one side. In Syria, UN sanc­tions and other coercive measures to influence con­flict parties were out of the question anyway, due to the polarisation of the Security Council. Above all, however, Western governments were not prepared to exert serious pressure on the regional states that were key in preventing a negotiated solution or its imple­mentation in all three countries.

The conclusion obviously cannot be that it is futile to mediate in similarly complex future conflicts. Rather, the question is how mediation efforts can become more effective. The cases addressed here offer three conclusions. First, the mandate should not be limited to mediating between the local parties in a civil war. Instead, from the outset it should also pro­vide forums that allow for the reconciliation of the competing interests of relevant regional and great powers, or at least enable the UN to influence the rules of engagement in the conflict. Second, power-sharing agreements should be sufficiently inclusive; the negotiations should bring together actors who are truly representative of the political forces and con­stituencies on the ground; and the agreements should give these parties sufficient incentives to abide by the deal. Third, UN mediators should avoid taking sides in favour of unity governments if those governments them­selves become parties to the conflict and under­mine agreements.

Europeans often have little influence on the international power relations that constrain UN mediators. Nevertheless, they can help by refraining from doing anything that might undermine UN mediation ef­forts, such as circumventing sanctions or directly cooperating with conflict actors to pursue aims unrelated to a negotiated solution. Furthermore, for as long as UN missions do not have the mandate to reconcile com­peting regional and international interests, they should use their channels of dialogue with regional and major powers towards avoiding further escalation, establishing rules of engagement and focusing on protecting civilians.

The civil wars in Libya, Syria and Yemen are extremely complex in terms of the configuration of local actors, the military involvement of regional and great pow­ers, and the dynamic formation of alliances.1 But the three cases also raise general questions: when do negotiated settlements succeed in ending civil wars, and under what conditions can the UN successfully mediate peace agreements and guarantee their implementation? Previous cases allow us to draw a number of lessons and outline points of contention with regard to the roles of conflict dynamics, the mediation strategy, and the UN Security Council.

Conditions for Successful Mediation

There is no agreement on whether or not conflict dynamics can make a civil war ripe for resolution or ensure that all negotiations will fail in the absence of ripeness. According to an influential thesis, success­ful conflict resolution requires a stalemate between the conflicting parties in which the costs of continuing con­flict exceed those of making concessions for a peace agreement.2 But whether a conflict had reached this stage often only becomes clear in retrospect. The expectation that a civil war will be ripe for resolution at some future point is not only questionable because it can serve as a convenient pretext for insufficient efforts at mediation, but also because many civil wars come no closer to a solution as they grind on; rather, they produce ever new conflicts, as well as new actors and war profiteers with a strong interest in continu­ing conflict.

Clearly, however, a negotiated settlement will re­main out of reach as long as the military balance of power changes rapidly or conflict parties can count on sustained foreign backing – in other words, as long as one or more conflict parties expects to make military gains. Mediation efforts themselves can con­tribute to the formation of new alliances among those who oppose an agreement, thereby provoking escala­tion. The decisive role of external support, in turn, explains why, during the Cold War, civil wars – most of which were then also proxy wars – rarely ended through negotiated settlements, and why the “uni­polar moment” of the 1990s set in motion a series of peace agreements for old and new conflicts.3 How­ever, even during this heyday for conflict settlements, only a quarter of mediation attempts produced an agreement, and only a fraction of these agreements saw implementation.4 By contrast, the growing multi­polar disorder of recent years offers far more difficult conditions for conflict resolution. It drives a growing internationalisation of civil wars – an increasing number of great and regional powers are now in­volved5 – and a trend towards fragmented conflict landscapes: instead of binary struggles between a government and a rebel movement, we increasingly see multi-party conflicts in which continuously changing alliances prevent the emergence of a stale­mate.6 The three conflicts this study focuses on are prime examples for this development.

Which mediation strategy is successful depends not only on the configuration of the conflict, but also on the characteristics of the mediator. Mediators with little leverage – such as representatives of small states or non-governmental organisations – have to convince conflict parties that they are absolutely impartial. Their strategy will necessarily focus on creating trust between the conflicting parties. By con­trast, a mediator who represents world powers and can bring their influence to bear will rely more on international pressure to prod the conflict parties towards compromise, and on third-state guarantees to ensure the implementation of an agreement.7 Media­tion based on such “manipulation” of the bal­ance of power aims either to bring about a stalemate or to demonstrate to the conflict parties that they have already reached a stalemate.8

Either way, the minimum requirement is that the conflict parties agree to mediation. It is also impor­tant that all relevant conflict actors and their external supporters are directly or indirectly involved in the talks. There are two reasons why this has also become more challenging in recent years. First, the growing internationalisation of conflicts increases the number of external actors who need to sit at the negotiating table – and who will try to exclude other local or external actors. Second, jihadi groups have played an ever greater role in recent civil wars. They show little willingness to engage in talks, and their participation in negotiations remains an international taboo.9

Wherever the UN mediates, the approach pursued by mediators and its efficacy depend greatly on the Security Council. For UN mediation to be successful, a whole series of conditions have to be met. The Secu­rity Council has to provide the mediator with a clear mandate, but above all it has to be united in its sup­port for the mediation efforts and its interest in solving the conflict. The mediator also depends on the permanent members to back an agreement with guarantees – such as the UN or third states deploying troops – and threaten transgressors with sanctions. When these conditions are not met, even experienced UN mediators find it difficult to negotiate viable agree­ments.10 Finally, the UN Security Council’s posi­tion on a conflict determines whether the mediator can gain a reputation for impartiality, or is seen as biased towards one party in the conflict.

Peace through Power-Sharing?

Negotiated solutions to civil wars are notoriously fragile. It is well-established that when a civil war has ended with a negotiated settlement, a resumption of conflict is much more likely than if it has ended with a military victory by one side.11 It is the subject of lively debate under what conditions peace agree­ments survive or collapse. The debate particularly revolves around the issue of whether certain types of power-sharing agreement are more robust than others, or whether it is the conditions under which the power-sharing agreement was concluded that are crucial.12 Clearly, however, there is no straightforward formula for successful peace agreements. Only detailed analysis of individual cases allows an in­formed guess as to whether an agreement has a realis­tic chance of implementation or is likely to founder. Even in the best case, power-sharing agreements re­main a gamble for the local and international actors involved.

Power-sharing arrangements – whether political, military, economic or territorial – have been a com­ponent of most peace agreements since the 1990s. Even when implementation is relatively successful, critics question the sustainability of power-sharing agreements. They point out that power-sharing agree­ments signal to political actors that violence pays;13 that they are more of an obstacle to sustainable con­flict resolution, since power-sharing usually only comes about through massive international commitment, which often declines during implementation;14 and finally, that they primarily serve the interests of power-hungry politicians and warlords without addressing the actual causes of conflicts.15 On the last point, proponents counter that peace negotiations cannot succeed unless they bring together only those actors who can cause an agreement to fail.16 Include­ing a broader range of negotiating parties would dilute the influence of such veto players, who would therefore also lose interest in a solution.

As it stands, however, the interests of regional and great powers and the exclusion of some groups as ter­rorists often makes it difficult or even impossible to bring all relevant actors to the negotiating table. The case studies in this analysis underline this point. They also show that the prominent role of external powers in these conflicts necessitates a negotiating framework that involves both local and international actors. It is hardly surprising that such a complex under­taking rarely succeeds. Nevertheless, it is worth asking whether the configuration of a conflict made the failure of UN efforts all but inevitable, or whether the mediator’s miscalculations or bias contributed to failure. Answering this question will help in devising more successful mediation efforts for similarly com­plex future conflicts.

With this objective in mind, the following con­tributions analyse the UN’s mediation strategies in Libya, Syria and Yemen against the background of these conflicts’ trajectories, and the competing inter­ests of local, regional and international actors in­volved in them. The focus is on the following ques­tions: How do UN mediators deal with the structures and dynamics of these conflicts? What are the main obstacles to success for the UN’s efforts? What can European policymakers do to support UN mediation and make it more effective?

Overview 1 Mediation missions

United Nations Support Mission in Libya (UNSMIL)

The Office of the Special Envoy of the Secretary-General for Syria (OSE-Syria)

The Office of the Special Envoy of the Secretary-General for Yemen (OSESGY)


Resolution 1973 (March 2011) wel­comed the appointment of a Special Envoy of the UN Secretary-General to “find a peaceful and sustainable solution” to the crisis in Libya. UNSMIL was created by Resolution 2009 (September 2011). Several subsequent Resolutions, most recently 2376 (2017), have extended the mission mandate for mediation and provision of good offices, in­clud­ing (since December 2015) supporting the implementation of the Libyan Political Agreement.


The Special Envoy of the UN Sec­retary-General is mandated to bring about a conflict settlement on the basis of the Geneva Communiqué (June 2012) and SC Resolution 2254 (December 2015). This involves establishing a “transitional gov­erning body” (TGB) from among government and opposition repre­sentatives that will shape the political transition to a pluralistic, democratic, inclusive Syria.

Predecessor mission: joint LAS/UN mediator based on UNGA Resolution
66/253 (February 2012).


In April 2011, the UN Secretary-General appointed a Special Envoy for Yemen (without a mandate from the Security Council or General Assembly) to support with his good offices the efforts of the Gulf Co­operation Council (GCC) in Yemen.

SC Resolution 2014 (October 2011) welcomed the Special Envoy’s role as a mediator and reaffirmed his mandate to support Yemen in its political transition in line with the GCC initiative. SC Resolution 2216 (April 2015) called on the UN Sec­re­tary-General to intensify the envoy’s role. Accordingly, the UN Special Envoy was tasked with assisting the conflict parties in returning to a peace­ful process of transition based on the GCC initiative, the results of the National Dialogue and the rele­vant Security-Council resolutions.


Special Representative of the UN Secretary-General for Libya:

Abdelelah Al-Khatib
(Special Envoy)
(07/03/2011 – 20/08/2011)

Ian Martin
(11/09/2011 – 13/10/2012)

Tarek Mitri
(14/10/2012 – 30/09/2014)

Bernardino León
(01/09/2014 – 4/11/2015)

Martin Kobler
(04/11/2015 – 22/06/2017)

Ghassan Salamé
(since 22/06/2017)


Special Envoy of the UN Sec­retary‑General to Syria:

Staffan de Mistura
(since 10/07/2014)

Joint Special Representative for the United Nations and the LAS for Syria:

Kofi Annan
(23/02/2012 – 2/08/2012)

Lakhdar Brahimi
(18/08/2012 – 13/05/2014)


Special Envoy of the UN Sec­retary‑General to Yemen:

Jamal Benomar
(01/04/2011 – 15/04/2015)

Ismail Ould Cheikh Ahmed
(25/04/2015 – 27/02/2018)

Martin Griffiths
(since 17/02/2018)

Overview 2 Significant Military Involvement of Third Parties





logistical support, training, advice and arms for Khalifa Haftar;

several airstrikes
(2015–2017) against armed groups or cities which are hostile to Haftar.

member of the Saudi-led coalition;

marine support for the naval blockade.


deployment of Special Forces and intelligence assistance for Khalifa Haftar;

Assistance in training the Presidential Guard for the Serraj government in Tripoli.

financial and logistical sup­port, training for so-called moderate rebels and the SDF;

Special Forces to support the rebels and SDF.

logistical support for (and in­tel­ligence cooperation with) the Saudi-led coalition.


deployment of Special Forces in Misrata and intelligence assistance there, as well as for Khalifa Haftar in Benghazi.

financial and logistical sup­port, training for so-called moderate rebels and the SDF;

Special Forces to support the rebels and SDF.

logistical support for (and in­tel­ligence cooperation with) the Saudi-led coalition; deployment of military advisors in the Operation Room in Riyadh.


financial and logistical sup­port for the regime and Iran-led militias;

deployment of military advisors, special forces and ground troops to support the regime;

guidance for Iran-led militias (esp. Lebanese Hezbollah and militias from Iraq, Afghanistan and Pakistan).

financial support for Houthi rebels;

possibly arms deliveries, incl. long-range missiles;

military training for Houthi rebels by Hezbollah.


airstrikes against Hezbollah positions as well as Iranian and Syrian-regime positions, arms transports and factories.


deployment of military advisors and a field hospital to support armed groups from Misrata fighting IS in Sirte
(May – December 2016 and since);

training for the Presidential Guard of the Serraj government in Tripoli.

Overview 2 (cont.) Significant Military Involvement of Third Parties





airstrikes against the Qadhafi regime (2011).


no clear evidence for support for Khalifa Haftar; possibly indirect arms deliveries (via Egypt); advice or training pos­sibly via private companies.

financial and logistical sup­port; arms; military advisors for the Assad regime;

airstrikes (since September 2015) against rebels and IS;

ground troops and military observers monitoring the de-escalation zones.

Gulf Cooperation Council states

UAE: financial and logistical support and arms for Khalifa Haftar;

UAE: airstrikes in support of Haftar’s alliance (August 2014); construction of a UAE air-force base in eastern Libya for regu­lar airstrikes in Benghazi and Darna (2015-2018);

Qatar: financial support for Haftar’s opponents; funding of arms deliveries via third parties (Sudan).

Saudi Arabia, UAE, Qatar: financial and logistical support; arms; and training for rebels.

Saudi-led coalition against Houthi rebels with airstrikes, ground troops, blockade (since March 2015), incl. airstrikes by Saudi Arabia, UAE, Qatar, Bah­rain, Kuwait, Morocco, Jordan; UAE, Sudan and Qatar ground troops; financial and logistical support and training for various militias by UAE and Saudi Arabia.

Support also from Somalia and Eritrea by providing mili­tary bases and granting flyover rights.


several arms deliveries to opponents of Haftar; role of Turkish government unclear.

financial and logistical sup­port, arms and training for rebels;

airstrikes against PYD and IS;

ground troops in northern Syria (Operation Euphrates Shield 2016/17 and Operation Olive Branch 2018); military observers monitoring the de-escalation zone in the north.

Overview 2 (cont.) Significant Military Involvement of Third Parties





airstrikes against IS in Sirt (August – December 2016) in support of armed groups from Misrata;

Special Forces active in Benghazi and Misrata;

intelligence cooperation with various conflict actors.

financial and logistical sup­port; arms and training for so‑called moderate rebels (2013–2017) and SDF;

airstrikes in support of SDF offensives against IS;

Special Forces to support SDF.

airstrikes against AQAP since 2002; ground operations by special forces against AQAP in southern Yemen;

occasional airstrikes against Houthi positions (2017); deploy­ment of military advisors in the Operation Room in Riyadh; naval support in the Red Sea, logistical assistance and intel­ligence cooperation with the Saudi-led coalition; special forces at the Saudi-Yemeni border.

Anti-IS Coalition

airstrikes against IS

Wolfram Lacher

Libya: The Gamble That Failed

In December 2015, Libyan politicians signed an agree­ment in Skhirat (Morocco) to form a transitional gov­ern­­­ment. The agreement aimed at transcending the coun­try’s political divide after the eruption of civil war in mid-2014 had put an end to the post-2011 tran­sitional process. Negotiated under the aegis of the UN, the agreement not only had the support of Western governments but also of a united UN Secu­rity Coun­cil. The latter declared the Government of National Accord (GNA) formed on the basis of the agreement to be the only legitimate government of Libya, called on member states to cease contact with parallel insti­tu­tions, and threatened actors who obstructed the agree­ment’s implementation with sanctions.

Despite international support, the agreement could be considered a failure soon after it was signed. The country’s political and institutional divide persists. Instead of functioning as a power-sharing arrangement, the GNA has devolved into a façade for the capture of state institutions by a handful of Tripoli militias. The government as such exerts little author­ity in Tripoli, let alone beyond the capital city. The economic crisis has dramatically worsened since the GNA’s formation. Following unsuccessful attempts at isolating opponents of the agreement, Western governments have resorted to courting the most powerful challenger to the GNA: Khalifa Haftar, the leader of the so-called Libyan Arab Armed Forces.17 International efforts at brokering a new political ar­rangement that would include Haftar have remained stuck.

What explains the failure of the Skhirat agree­ment, and what are the prospects for negotiating a power-sharing deal in Libya?

Protracted Struggles

The defining feature of the Libyan conflict landscape is its fragmentation. There are few clear fault lines or constant actors, and few coherent political-military forces – with the exception of Haftar’s. Instead, count­less armed groups and political actors form ever new constellations. The continually changing balance of power makes for difficult conditions for negotiations.

The current conflicts have their origin in the 2011 revolution. During the civil war, numerous revo­lution­ary armed groups formed at the local level. Because of their fragmentation, the revolutionary forces failed to re-establish central authority after the regime’s demise. Rather, successive transitional governments were a front for the competition over resources between representatives of individual cities and groups. One focal point of these rivalries was the security sector, where representatives of competing factions used their official positions to build up mili­tias under the guise of state institutions.18

As the power struggles escalated, actors with common enemies gradually formed broader alliances. In the summer of 2014, this dynamic resulted in the formation of two rival camps and the eruption of a civil war whose epicentres were the country’s two largest cities. In Benghazi, rampant lawlessness cre­ated fertile ground for the establishment of a ren­egade army leadership under Haftar – formerly a high-ranking officer under Muammar al-Qadhafi; later his adversary in exile; and after the 2011 revo­lution an openly power-hungry militarist.19 In Ben-

Map 1 Situation in Libya (as of October 2018)

ghazi, Haftar’s forces fought a coalition of revolutionary and jihadi groups. In Tripoli, an alliance of mili­tias led by forces from Misrata drove its Zintani op­ponents, who had allied themselves with Haftar, out of the capital. A majority of the newly elected House of Representatives (HoR) met in the eastern city of Tobruk, lent its support to Haftar, and formed a gov­ernment in nearby al-Baida. A part of the General National Congress (GNC), elected in 2012, refused to recognise the HoR’s Tobruk sessions and appointed a government in Tripoli to represent the forces which had just taken control of the capital.

With the eruption of civil war, the UN Support Mission in Libya (UNSMIL) – established in 2011 to support the post-Qadhafi transition – began focusing on mediating between the conflicting parties to over­come the political divide. These efforts were initially unsuccessful. A first attempt to bring together rep­re­sentatives of the HoR in Tobruk with boycotting par­liamentarians failed in September 2014. The politi­cal climate was polarised. The Tobruk-based camp attacked its opponents as Islamists; they, in turn, accused Haftar and the Zintanis of leading a counter-revolution in league with regime loyalists. Regardless of the rhetoric, however, the two camps were politi­cally too diverse to fit into neat categories.20

In early 2015, a political and military stalemate between the two camps began to emerge. As a result, the two opposing alliances disintegrated. The UN-led negotiations – now primarily between representatives of the two parliaments – encouraged the grad­ual dissolution of the state of polarisation. Largely independently of the UN-led talks, initiatives by local commanders and mediators to negotiate local cease­fires in western Libya also had an important impact.21 The military stalemate and persistent political divi­sions did, however, provide space for Libyan affiliates of “Islamic State” (IS) to expand and gradually estab­lish Sirte as their Libyan headquarters.22

Despite the fragmentation of the two camps, the UN-led talks stuck to a binary framework, the principal negotiating parties being the HoR and the GNC. From mid-2015 onwards, rifts opened up between supporters and opponents of the negotiations, both within these two institutions and in individual cities that had previously been clearly associated with one or other of the camps. The politicians who signed the December 2015 Skhirat agreement therefore had shaky power bases. Repre­sentatives from eastern Libya, Misrata and Zintan who held positions in the nine-member Presidency Council often faced influential adversaries in their own communities. In eastern Libya, where Haftar set the agenda, opposition to the agreement prevailed.23

The GNA’s formation brought about further changes in the political landscape. Former Haftar sup­porters from eastern Libya who had obtained a posi­tion in the GNA now joined their former enemies in opposition to Haftar. Militias from Tripoli that had previously been hostile to the “puppet government” were now surprisingly pragmatic and exploited their new status as pro-GNA forces to move against local rivals. Former regime officials, who had been ostra­cised until the mid-2014 institutional divide, now became coveted allies for actors across the political spectrum. There was no longer any limit to opportunistic side-switching. During the same period, “Islamic State” disappeared as a prominent actor. Misratan militias, though divided in their attitudes towards the GNA, united against IS in Sirte. After several months of fighting, they received support in the form of sus­tained US airstrikes; by late 2016, IS was destroyed as a territorial force in Libya.24

The political survival of the Presidency Council is exclusively due to international recognition.

The GNA remained too weak politically to act as a magnet for rival factions. One reason was that it was a unity government in name only: two of the nine members of the Presidency Council began to boycott its meetings a mere month after the agreement was concluded; a third stepped down after a year; the remaining six were conspicuous mainly through their public spats and lack of cooperation.25 A second major weakness of the Presidency Council was that the Central Bank initially granted it only limited access to funds. Serraj, who increasingly monopolised the scarce resources and decision-making powers of the Presidency Council, gradually built up a precarious clientelist structure of rival politicians and militia leaders over whom he had no real control. Instead, the government increasingly came under the influ­ence of a handful of militias that divided up Tripoli between themselves.26 The political survival of the Presidency Council was exclusively due to inter­national recognition.

Throughout all twists and turns of Libya’s conflicts, Haftar was the only actor who steadily consolidated his power structure and expanded his territory. His alliance was initially very loose; many of his allies hoped to turn on him when they would no longer need him. But against such calculations, Haftar grad­ually eliminated all of his disloyal associates. His seizure of the eastern oil ports in September 2016 marked the definitive failure of attempts to under­mine him by integrating rival figures from eastern Libya into the GNA.27

A Regional Playing Field without Rules

Haftar owes his continuous ascent principally to levels of foreign support that no other actor in Libya’s conflicts enjoys. Military and political backing from Egypt and the United Arab Emirates (UAE) allowed Haftar to put up tenacious resistance to the Skhirat agreement and bet on long-term military gains against his opponents.

The Egyptian and Emirati support for Haftar is rooted in the regional antagonisms that developed after the overthrow of Egyptian President Morsi in July 2013: Haftar’s declared goal of “eradicating” the Muslim Brotherhood in Libya resonated in Cairo and Abu Dhabi.28 The oft-repeated accusations by Haftar and his allies that Qatar, Sudan and Turkey supported Libyan Islamists may also have encouraged the Emir­ati leadership to intervene. However, there is only sparse evidence for such assistance to Haftar’s oppo­nents. The extensive and barely concealed Emirati and Egyptian support for Haftar is far disproportionate to the few indications of Turkish and Sudanese arms deliveries or Qatari payments to Haftar’s adver­saries.

For regional powers, Libya in 2014 became a play­ing field in which no rules applied. The Emirates and Egypt delivered heavy equipment to Haftar, in fla­grant breach of the UN arms embargo.29 In mid-2014, the Emirates launched airstrikes on militia positions in Tripoli without ever admitting it – and without even consulting the US, let alone European governments.30 In subsequent years, the Emiratis secretly constructed an air base in eastern Libya from which they regularly launched strikes on Haftar’s opponents.31 In 2016 and 2017, there were at least three strikes by unknown attackers; in each case, the only plausible explanation is that they were Egyptian or Emirati aircraft.32 Neither the Libyan authorities nor Western governments or international organisations have so far been able, or willing, to identify the states responsible.

While Libya’s neighbours to the west, Algeria and Tunisia, supported the Skhirat talks, the UN Special Representatives Bernardino Léon (September 2014 – November 2015) and Martin Kobler (November 2015 – June 2017) could only bring their Egyptian and Emirati interlocutors to pay lip service to the agree­ment. Both states clearly put their support for Haftar above efforts for a negotiated solution. With­out this support, Haftar’s stubborn rejectionism would have been unthinkable and a power-sharing agreement would have had a realistic chance of suc­cess. The Egyptian and Emirati role was thus key to the failure of UN efforts.

The Growth of Vested International Interests

If Egypt and the UAE were able to obstruct inter­national efforts, this was not least because Western governments were unwilling or unable to prevent them from doing so. Western governments barely raised the issue of the continuous violations of the arms embargo. Their economic interests in both countries – especially regarding arms deals – and cooperation on conflicts in Iraq and Syria carried greater weight in the calculations of Western gov­ernments than the need to persuade regional powers to play a more constructive role in Libya.

In contrast to the regional powers, Western states had not contributed to the escalation of the conflicts in 2014. Indeed, they had united in support for a negotiated solution. However, the more protracted the situation in Libya, and the more remote any restoration of state authority, the more Western gov­ern­ments became directly involved. This initially occurred in pursuit of counterterrorism. From early 2016, French special forces supported Haftar’s forces in Benghazi.33 While French diplomats consistently played down this presence as merely for reconnaissance, the political signal was unmistakeable: despite his aggressive opposition to the Skhirat agreement, Haftar enjoyed Paris’s support. This made the official French support for the agreement meaningless, and shattered European unity on Libya.

In May 2016, the US, UK and Italy began to support Misratan forces – the biggest military counterweight to Haftar – in their offensive against the IS stronghold of Sirte, including with special forces.34 From August onwards, US fighter jets flew hundreds of sorties on Sirte. Officially, US airstrikes came on the request of the newly-formed GNA, and in support of forces loyal to it. In reality, politicians and armed groups in Misrata were divided in their attitude to­wards the GNA. The government had no influence over the course of the offensive, and the lack of GNA support alienated even militias that had backed the agreement.35 In turn, opponents of the GNA saw assis­tance for Misratan militias as proof that Western states used the cover of the GNA to extend support to their political adversaries.36

The victory of local forces over the IS affiliate in the eastern city of Darna in April 2016, the IS defeat in Sirte in December, and Haftar’s steady advances in Benghazi annihilated “Islamic State” as a territorial force in Libya – almost entirely without the GNA’s help. The threat of an IS emirate on the southern Medi­terranean coast receded, and Libya slid down the priority lists of the US, UK and France. At the same time, the increasing migration flows through Libya were a priority for Germany and Italy. Following the EU’s refugee agreement with Turkey in March 2016, the central Mediterranean route became the biggest challenge in attempts to seal off the EU’s external borders. The number of migrants arriving in the EU via Libya rose to a record high in 2016.37

Since early 2017, Europeans have come to terms with the fact the GNA as such lacks the capacity to curb migrant flows. As a result, networks of vested interests have formed between EU governments and local conflict actors. The Italian government quietly made arrangements with local militias that controlled the migration business in western Libyan port cities.38 Among the coastguard units that benefit from Euro­pean support are local groups that are directly in­volved in these criminal activities.39 The cooperation of the Serraj government with European efforts to seal off Libya’s shores has undermined the GNA do­mestically, provoking accusations that it was Europe’s puppet. The now overwhelming focus of EU Libya policy on containing migration has not only supplanted the original goal of contributing to the re-establishment of a functioning government – it is diametrically opposed to it.

Competing Forums

The growing entanglement of external actors in Libya’s conflicts also led to a proliferation of media­tion forums. The fact that the UMSMIL-led process had reached a dead end by mid-2016 paved the way for unilateral initiatives, which in turn was detrimental to the UN’s role. Egypt took the lead, organ­ising several rounds of talks from December 2016 onwards with Libyan politicians who were seen as acceptable by Cairo. Egypt subsequently began mediating between Haftar’s representatives and military officers from western and southern Libya. These Egyptian efforts conflicted with those of UNSMIL in that Cairo sought to negotiate an arrange­ment in which its Libyan clients would play leading roles, and Islamists were to be excluded. At the same time, Egypt resisted Algerian and Tunisian attempts to build a joint initiative, which would have required a more balanced approach.

Russia remained largely passive during the Skhirat negotiations, but it did not try to undermine the talks. Once the agreement was concluded, it weighed in as an advocate for the deal’s opponents. Haftar made several high-profile visits to Moscow. Subsequent­ly, however, Russian diplomats sought to diver­sify their contacts in Libya and dispel suspicions of Russian support for Haftar, trying to position Russia as a mediator.40

In July 2017, French President Emmanuel Macron received Haftar and Serraj near Paris, and subsequently gave the misleading impression that they had reached an agreement. Macron’s initiative was extraordinary both for the French leadership’s nonchalance in uni­laterally trying to reshape international mediation efforts, and for its willingness to make Haftar a legiti­mate interlocutor for European heads of state without extracting any concessions from him.41 Several visits by Haftar to Rome followed. However, the fact that European governments now courted Haftar did not increase his readiness to compromise. Finally, the African Union proposed yet another mediation initia­tive outside the UN framework with a summit for Libyan actors in September 2017. Haftar apparently did not consider that meeting sufficiently important to merit his attendance.

The UN Special Representative Ghassan Salamé, in office since July 2017, initially managed to breathe new life into the UN efforts and contain the drift towards unilateral mediation initiatives. However, Salamé still had to contend with the French-led push for quick elections. When Salamé was unable to demon­strate much progress by the spring of 2018, individual states re-appeared with their unilateral initiatives – including, once again, France. In May 2018, the Elysée surprised other Western governments and UNSMIL with a single-handed attempt to coax a handful of Libyan actors into accepting tight deadlines for parliamentary and presidential elec­tions.42 While several Western governments were sceptical, the Italians, in particular, openly opposed the French initiative.43 Salamé was left to juggle Macron’s insistence on the deadlines he introduced with other governments’ demands for progress on the conditions necessary to hold elections. While the French initiative was entirely unrealistic – both for its improbable deadlines, and because it extracted no firm commitments from Libyan actors – it nevertheless severely limited Salamé’s room for manoeuvre, depriving him of the time horizon he needed to devel­op a more sustainable negotiating framework.

Flaws in UNSMIL’s Process Design and Implementation

UNSMIL’s mediation efforts were able to build on fortuitous international circumstances, but also had to adjust to conditions it could not influence. The fortunate circumstances include clear Security Coun­cil backing and unqualified Western support for UN mediation, as well as – until the agreement was con­cluded – no attempts by Russia or China to under­mine these mediation efforts. Up to the moment the deal was signed, no state disputed the UN’s lead role as mediator in the Libyan conflict. At the same time, differences between Security Council members meant that UNSMIL was unable to mobilize international sup­port for a more muscular approach. Sanctions against so-called spoilers who sabotaged the agree­ment, the enforcement of the arms embargo, or the deployment of peacekeepers were out of the question.44

There were no easy answers to the question of who the real conflict actors were and who should represent them in negotiations.

Among the conditions that the UN Special Representative could not readily alter were the Egyptian and Emirati positions. Western governments were not committed enough to persuade both countries to adopt a more cooperative stance on Libya. This posed a massive obstacle to any attempts at mediation.

Of particular interest here, however, is how the UN dealt with these circumstances. The adverse con­ditions outlined above notwithstanding, which deci­sions contributed to the failure of efforts to find a sustainable solution?

The Issue of Representativeness

In late 2014, UNSMIL faced an extremely confusing political and military landscape in Libya. There were no easy answers to the question of who the real con­flict actors were and who should represent them in negotiations. This was not only due to the problem – common in peace negotiations – of how to involve those directly responsible for military action without bestowing such controversial figures with legitimacy or losing their adversaries as negotiating parties. An additional complication in Libya was that the armed groups were often only loosely associated with formal political figures such as parliamentarians, but at the same time they rarely had leaders empowered to nego­tiate. In most cases, these groups were organised on the basis of individual cities, and claimed to de­fend the interests of these cities – yet in few local­ities did political and military actors agree on repre­sentatives for negotiations.45

UNSMIL built its framework on the two rump par­liaments as negotiating parties, which corresponded to the fact that the dispute over legitimacy was cen­tral to the conflict. It also allowed UNSMIL to circum­vent the difficulties it would have faced if it had chosen the representatives of the conflicting parties itself – accusations of arbitrariness and prejudice would inevitably have ensued. However, this ap­proach also raised the question of how to involve a broader spectrum of actors and ensure that the actual forces on the ground were represented.

UNSMIL’s solution was twofold. First, it invited sev­eral figures to the political negotiations who were as­so­ciated with neither parliament. Second, it launched several other dialogue tracks in parallel to the politi­cal negotiations, including meetings of representatives of municipal councils, civil society activists and political parties, leaders of armed groups, prominent women and women’s right activists, as well as tribal representatives. It thereby hoped to garner broader support for the negotiation process without allowing these actors any actual influence over the outcome.

The implementation of this approach had considerable shortcomings. UNSMIL chose the independent figures it invited to the negotiations for their ability to build bridges between the conflicting parties and propose consensual solutions.46 However, they often had no power base of their own and distinguished themselves primarily by their good relations with Western diplomats. The same applied to several of the participants UNSMIL chose – just as arbitrarily – for the political parties and civil society track. The meetings of tribal representatives were discontinued after two unsuccessful attempts in Egypt. Fatal for the negotiations’ success, however, was the fact that UNSMIL in mid-2015 abandoned its efforts to mediate between armed actors, after several fruitless attempts. From then on, this key constituency was only in­direct­ly involved in the process, and the guessing game began as to how the armed groups would react to the agreement.

The Mediators: Facing Time Pressure and Conflicts of Interest

The time pressure Special Representatives León and Kobler applied while attempting to conclude the ne­go­tiations was seriously detrimental to the viability of the agreement. The rush was due to both León’s per­sonal ambitions and pressure from Western gov­ern­ments for a rapid conclusion in the autumn of 2015.

In mid-2015, León informed Western diplomats that he intended to leave his post within months and was determined to reach a deal before he handed over.47 In September and October, León tried to force an agreement to be able to take up his next post – director of the Emirates Diplomatic Academy – as soon as possible and with a success under his belt. This meant that León secretly negotiated his conditions with high-ranking Emiratis while at the same time mediating in Libya, where the UAE was blatantly intervening on one side of the conflict.48 When this came to light, León had to bring forward the planned handover to Kobler without having obtained a result in Libya.

Upon taking office, Kobler faced the choice of re­opening the draft agreement, which had been called into doubt by the León scandal, or bringing it to a quick conclusion. Libyan proponents of the agreement advocated the latter, not least due to self-inter­est. However, pressure from Western governments for an imminent conclusion proved decisive. The main reason was the steady expansion of IS, espe­cially in the region around Sirte. The Obama adminis­tration and Italy, in particular, saw an urgent need to act, but wanted a “legitimate” Libyan gov­ern­ment to request international support for action against IS. If the negotiations dragged on, however, military inter­vention against IS would happen without a unity gov­ernment – which, in turn, would negatively affect the talks. European governments also needed a legiti­mate counterpart so as to take military action against human smugglers in Libyan waters, since Russia was blocking the mandate for this in the Security Coun­cil.49 In late November 2015, Italy and USA set the pace by calling for a ministerial-level conference for mid-December to express its support for the agreement – when no agreement yet existed.50

The time pressure induced UNSMIL to try a risky manoeuvre. The negotiating parties, the HoR and GNC, were internally divided. According to the text of the agreement, the two parliaments were the parties to the deal, the institutional pillars of the agreement, and key to its implementation. However, neither par­lia­ment had a formal majority in favour of the agree­ment; the presidents of both parliaments had even with­drawn their delegations’ mandates. UNSMIL there­fore decided to let the negotiators sign the agree­ment in a personal capacity. To emphasise the sup­pos­edly broad support in Libya – as Kobler and West­ern diplomats asserted – UNSMIL flew dozens of parlia­mentarians, mayors and other political actors to Skhirat for the signing ceremony. Of the 21 figures who initialled the agreement on 17 December 2015, only 11 were elected members of one or the other par­liament; the other ten were handpicked by UNSMIL and mostly had no political base to speak of.

After negotiations that within this framework had lasted less than a year, the time pressure exerted by Western governments caused defects in the agree­ment that would prove fatal. The consent of the HoR, which was imperative for implementing the agreement and forming a government, failed to materialise after the signing, and the tug-of-war over that ques­tion opened up rifts that permanently paralysed the parliament. In November, to appease as many clienteles as possible, Kobler increased the soon-to-be Presidency Council from six to nine members whilst maintaining the requirement for unanimous Council decisions. This practically ensured that the Presidency Council would be unable to act. Moreover, at the time that the negotiations were hurriedly concluded, talks with armed actors over security arrangements had not even begun. This excluded key issues from the agreement – issues that would remain unresolved thereafter. All of these problems were already appar­ent in December 2015, but experts’ warnings against excessive haste went unheeded.51 That haste largely rested on assumptions that subsequently turned out to be erroneous.


During the negotiations, the UN mediators closely coordinated with Western governments that intended to support the agreement’s implementation. They thus shared the miscalculations that prevailed in West­ern capitals, regarding both their own influence on Libyan conflict actors and the GNA’s room for manoeuvre.

Even at the time of signing, Europeans were still planning a Libya International Assistance Mission (LIAM), for which Italy wanted to provide as many as 5,000 soldiers – ostensibly to train Libyan forces, but in reality to protect the GNA and the international presence in Tripoli.52 These plans were key to West­ern designs for the agreement’s implementation – but they were divorced from Libyan realities. It was clear that no Libyan government would agree to such a foreign presence. When the Italian government finally realised this, the plans were shelved.

In addition, both the UN Special Representative and Western governments overestimated the effect of the punitive measures at their disposal. León and Kobler repeatedly threatened sanctions against spoilers, but differences within the Security Council prevented UN sanctions against leading GNA oppo­nents, including Haftar. The EU and US therefore imposed separate sanctions against three politicians, including the presidents of the two parliaments.53 All three were figureheads rather than heavyweights, and it is debatable whether the term “spoilers” really ap­plied to them. After all, there were justified objections to the legitimacy of the accord and the government formed by it. Their opposition to an agreement that would lead to the dominance of their political rivals was also understandable. Besides, the sanctions had no significant impact, except for reinforcing the defiance of those opposing the agreement.

Western diplomats also based their calculations on the expectation that the Presidency Council and its government would overcome their initial legitimacy deficit by creating new facts on the ground. Since the GNA alone would have access to state funds, they argued, it would gradually buy off fence-sitters and opponents of the agreement. In March 2016, Western governments and the UN urged the Presidency Coun­cil to take office in Tripoli despite the fact that the par­liament in Tobruk had approved neither the agree­ment nor the government.54 Since there was no pro­gress on negotiating security arrangements, the Presi­dency Council depended for its security in Tripoli on local militias whom several GNA members viewed as enemy forces.

The move to Tripoli contributed to reinforcing the initial divergences between the parties to the agree­ment.55 Several members of the Presidency Council and government refused to work in Tripoli under these conditions, which essentially meant that the Presidency Council had already lost its claim to spear­head a government of national unity. Nor was it able to create the expected facts on the ground. The gov­er­nor of the Central Bank did not allow the Presidency Council access to regular budgets, as long as the par­lia­ment had neither approved the government nor a budget, let alone monitored state expenditure. Final­ly, members of the Presidency Council worked against rather than with each other, meaning that the cli­entelist networks of individual members did not amount to a power structure for the government as a whole. Consequently, the Presidency Council had no control over the various forces that were loosely associated with it.

The UN: From Mediator to Party in the Conflict

With the conclusion of the agreement in December 2015, the Security Council tasked UNSMIL with sup­porting the implementation of the agreement and coordinating capacity-building measures for the GNA. If the agreement had in fact been a power-sharing arrangement between the conflicting parties, this task would have been compatible with the role of impar­tial mediator. However, since the agreement was con­cluded between certain actors against the resistance of many others, the UN role was now to support the Presidency Council in asserting itself against its adver­saries. In Tripoli, this amounted to a tacitly acquiescent attitude toward the militias allied with the Presi­dency Council, who drove their rivals out of the capital in several rounds of fighting starting in mid-2016. The actors in Tripoli themselves even viewed this evolution as driven by UNSMIL’s senior security advisor, the Italian General Paolo Serra – an impres­sion the UN mission did nothing to dispel – and the Italian government.56 In eastern Libya, this meant that UNSMIL backed a group of controversial militia leaders, simply because they viewed the GNA as an opportunity to expand their influence at Haftar’s expense.57

Soon after the Presidency Council arrived in Tripoli, it became clear that the agreement would not see implementation and that the country remained politically divided.

Soon after the Presidency Council arrived in Tripoli, it became clear that the agreement would not see implementation and that the country remained po­liti­cally divided. As a result, Western officials empha­sised that a modified agreement would have to make room for Haftar. However, UNSMIL, being so closely associated with the GNA, could hardly mediate between the government and its opponents. In fact, Haftar refused to meet Kobler for more than a year after the agreement was reached – declaring it would be “a waste of time”.58 Only when Ghassan Salamé assumed office in July 2017 did UNSMIL begin to dis­tance itself from the GNA and once again seek a role as impartial mediator.

Conclusions and Outlook

UN Special Representatives León and Kobler in 2015 tried to negotiate a transitional power-sharing ar­range­ment to overcome Libya’s political divisions under challenging conditions. The UN mediators enjoyed the unanimous support of the Security Council and Western states, but they had limited influence on local actors because deploying peacekeepers to en­force the agreement was never a realistic option, and the great powers were unable or unwilling to dis­suade regional states from supporting the conflict parties.

Given these conditions, there is no definite answer to the question of whether a more patient approach by the UN mediators, and a stronger integration of armed actors, might have produced a more workable power-sharing agreement. However, poor decisions by the UN mediators and Western governments, particu­larly during the final phase of the talks, meant that UNSMIL’s efforts at that stage no longer amounted to a serious attempt to reach a viable agreement. Neither León’s personal ambitions nor concerns about the on­going expansion of IS justified the haste with which the agreement was pushed through. Already at the time, it was clear that the assumptions underlying this approach were unrealistic. The fact that UN Sec­re­tary-General Ban Ki-moon approved León’s move from Special Representative to a highly-paid post in the UAE – and that the UN as well as Western gov­ern­ments then simply swept the scandal under the carpet – also gives cause for concern that such ob­vious conflicts of interest may feature with future UN mediators as well.

From the perspective of Western governments, the Skhirat agreement was a partial success despite its failure. Although the internationally recognised gov­ernment in Tripoli was largely impotent, it al­lowed them to pursue key interests. The US, UK and Italy assisted Misratan militias in fighting IS in Sirte with GNA approval, sticking to the fiction that these militias were forces loyal to the GNA. The EU, and particularly Italy, used the GNA as a cover to con­clude arrangements with local militias to curb migra­tion. By late 2016, IS had lost all its territory in Libya, and in 2017 the number of migrants arriving in the EU via Libya dropped by a third compared to the pre­vious year – without a real unity government having been formed. Oil production also recovered – despite, rather than because of the Skhirat agreement. The urgency with which Western governments and the UN had forced through the formation of a unity gov­ernment then gave way to an accommodation to the status quo.

However, rapidly installing a government that served as a front for the pursuit of Western interests came with consequences. The repercussions of the Skhirat negotiations have created more obstacles to the UN’s ongoing mediation attempts. The contro­versy over the agreement irreversibly divided both parliaments, which lost all credibility in the eyes of the Libyan public. They are now even less suited to acting as negotiating parties than they were during the Skhirat talks. For better or worse, UNSMIL will have to try a new approach, one that relies on actors who are not elected officials. Designating representa­tives will be yet more difficult because the agreement has further exacerbated the fragmentation of the political landscape. And finally, with the failure of Skhirat, the last chance to negotiate a power-sharing agreement with a wider range of actors in eastern Libya than just Haftar was missed. Since the Skhirat agreement, Haftar has established himself so firmly there with his repressive tactics that other political forces can no longer organise in the east. His strengthened position also means that his ambitions are difficult for other Libyan political players to accom­modate. And as long as Haftar not only con­tinues to receive support from the UAE and Egypt, but is now being courted by Western governments, he remains in a position to veto any deal.

UN Special Representative Salamé was initially adept at recommitting Libyan and international actors to the UN-led process. He set out to negotiate amendments to the Skhirat agreement by the two parliaments; organise a conference at which all po­litical and social groups would be represented, but whose objectives remained unclear; and hold a referendum on the draft constitution presented in July 2017 by the constitutional assembly, as well as elections – all in 2018.59

Lack of progress on these points has since forced Salamé to revise his roadmap. His attempts to per­suade the parliaments to make a move that would spell their own dissolution have proven futile. UNSMIL turned the National Conference into a broad-based, local-level consultation process that offers no basis for new negotiations. Under pressure from the French push for rapid elections, Salamé has at times stated that voting could take place without prior amendment of the Skhirat agreement or a referendum on the constitution; at times he has dismissed the feasibility of elections in the near term, but has expressed confidence that a constitutional referen­dum could be held.60 And while Salamé has publicly recognised that negotiations between the two parlia­ments offer little prospect for progress, he has not proposed an alternative framework for negotiations. As a result, by mid-2018, there was no longer a clearly identifiable political process. In this vacuum, parallel initiatives began reappearing, among them Macron’s unsuccessful attempt to push through an agreement on elections at a Paris meeting in May 2018; an Italian plan for a Libya conference with unclear objectives, to take place in November 2018; and a “reconciliation meeting” between Libyan players the African Union scheduled for late 2018.61

A key obstacle to rapid elections is the absence of a legal and constitutional basis for a vote. In September 2018, the HoR claimed to have passed the necessary amendment to the 2011 constitutional declaration to pave the way for a referendum on the draft constitution. But such a referendum would be a risky propo­sition, since it could lead into a dead end if the draft is rejected or if the referendum is not held in all parts of the country. Both possibilities have been made more likely by the disputed manner in which the HoR drafted and adopted the referendum law.62 Moreover, the draft constitution itself is controversial: the Ama­zigh ethnic minority boycotted the constitutional pro­cess, and representatives of other important groups in the constitutional assembly refuse to recognise the draft. In addition, political and military power rela­tions in the country have fundamentally changed since the elections to the constitutional assembly in 2014. The balance of power is still so far from being settled that it is probably not realistic to adopt a per­manent constitution for the country at the moment. For Libyans to decide how to deal with the draft, a greater degree of stability is necessary. In the mean­time, however, the constitutional or political basis for elections is lacking, and the two parliaments have lost both the ability and the credibility needed to produce such a basis. Salamé has repeatedly spoken out against a new transitional arrangement, but that is exactly what is needed before elections can be held.

The most important lesson from the mistakes of Skhirat is that imposing seemingly quick solutions to the power struggles in Libya comes at the expense of their viability, and is ultimately counterproductive. Western governments and UNSMIL should avoid making the same mistake again. At the time of writ­ing, the objective of holding elections as soon as pos­sible continues to dominate international policies towards Libya, despite the lack of any progress on creating the conditions for successful elections. UNSMIL and most European governments see elec­tions as a solution to the demise of legitimate insti­tutions in Libya. But without a sufficiently broad agreement on the legal and constitutional framework for elections, voting is likely to deepen Libya’s crisis of legitimacy, not resolve it.63 Moreover, it is worth remembering that the idea of elections first gained traction among international players as a way of integrating Haftar into the political process, on the assumption that he saw elections as being in his interest. However, not only has Haftar made contra­dictory statements regarding his support for elections, but he may very well be interested in elections only in areas under his control, or he may expect to ben­efit from a failed attempt to hold elections.

For elections to resolve the conflicts in Libya rather than to fuel them, a range of conditions have to be met that cannot be created by international decrees. Only a constitution – or, indeed, an agreement that prevails in lieu of a constitution – can determine what institution elections will be held for, and what the competencies of that institution should be. Voters and candidates require assurances that warlords and militia leaders will not dictate election results in their respective areas of control; that elections will take place across the country; and that the majority of actors will accept the outcome. To move towards these conditions, a new, more effective power-sharing agreement is needed that also includes security ar­range­ments. Reaching such an accord requires persis­tent mediation efforts and consistent international support. Given that the dysfunctional institutions created by the Skhirat agreement have long blocked progress towards such an agreement, and that there now no longer is a political process, it may be time to build a new negotiating framework – one that brings together actors with real influence on the ground.64 To build such a framework, the UN will need Euro­pean support for a more sustainable approach, not unilateral initiatives for quick fixes.

Muriel Asseburg

Syria: UN Mediation at the Mercy of Regional and Major-Power Interests

A power-sharing arrangement between regime and opposition representatives, and a political transition to an inclusive, democratic and pluralistic political system ought to have pacified the Syrian civil war. At least, that is what both the 2012 Geneva Communiqué and Security Council Resolution 2254 of December 2015 stipulate. Yet, neither UN mediator Staffan de Mistura (in office since July 2014) nor his two predecessors Kofi Annan (February–August 2012) and Lakhdar Brahimi (August 2012–May 2014), who were joint UN and League of Arab States media­tors, managed to bring the Syrian conflict parties to the table for negotiations on substance. They were thus unable to make progress on power-sharing or a political transition; neither could they contribute to sustainably calming the conflict, or effectively protect the civilian population.

Instead, the conflicts of interest were initially fought out militarily between local actors, and then also between regional and international actors. Rus­sia, which has been directly involved in the fight­ing on the side of the regime and its allies since September 2015, has tried to use its military successes to resolve the conflict. In 2017 Moscow, together with Ankara and Tehran, established an alternative nego­tiation and conflict-management format in Astana, Kazakhstan, which competed with the UN approach. Yet, the Astana powers were equally unable to initiate direct negotiations, sustainably pacify the fighting, or make progress towards a political solution. Rather, the conflict escalated again on several fronts starting in the first half of 2018, after so-called “Islamic state” had largely lost its territorial base and rebels had been pushed back into a few enclaves. The leadership in Damascus under Bashar al-Assad was set to recap­ture additional territories; the regional powers (espe­cially Iran, Turkey and Israel) sought to militarily realise their interests in the future order of Syria; and Russia and the USA also continued to play the mili­tary card.

This contribution analyses the UN’s approach to media­tion in the Syria conflict against the backdrop of the conflict dynamics, the incompatible interests of the conflict parties, regional powers and international actors. It focuses on the following issues: how does Staffan de Mistura’s mediation mission deal with local, regional and international conflict actors? What are the main obstacles to the UN approach in Syria? What other factors in the mission’s approach make suc­cess more difficult? What can European policymakers do to support UN mediation and make it more effective?

Conflict Dynamics and Conflict Actors

The war in Syria began with civil protests against police violence, family rule and kleptocracy.65 From the outset of protests – usually dated 15 March 2011 – the regime resorted to a limited number of reforms, but above all to violence in order to quell the initially peaceful and unarmed demonstrations. In response, activists and deserters armed them­selves – first to defend demonstrations and opposition strong­holds, later to fight the regime actively. By the summer of 2011, the demonstrations were already turning into an armed uprising.

Countries in the region, first and foremost Turkey and the Arab Gulf states, quickly began to support various rebel groups logistically, financially or with training and arms, as did some European countries and the USA. This not only fostered the illusion that the opposition could prevail militarily, but also con­tributed to rivalries and continued power struggles within the ranks of the opposition. The armed oppo­sition nevertheless conquered neighbourhoods of provincial towns, villages and rural areas in many parts of Syria. The regime shifted to besieging, starv­ing and bombarding towns and districts captured by the rebels. It also withdrew its troops from the majority-Kurdish areas in the north and northeast of the country to concentrate on securing the capital, strategic locations and central transport routes.

The Russian military intervention in September 2015 turned the tide in favour of the Syrian regime.

When in 2012 the survival of the regime seemed seriously threatened, Iran and the Lebanese Hezbollah first came to its aid. Yet it was only the Russian military intervention in September 2015 that turned the tide in the civil war in favour of the regime, enabling it to gradually reconquer lost territory with the support of the Russian Air Force. In 2017, IS in Syria also came under huge pressure.66 Having con­quered large swathes of territory in Syria and Iraq since 2013, IS had proclaimed a caliphate in 2014 and founded a proto-state. Its representatives asserted the claim that this caliphate was to have a global reach, and encouraged the establishment of local offshoots in other fragile states. After successes against IS in Iraq, the US-led anti-IS coalition, established in 2014, shifted its efforts to Syria in 2017. In conjunction with the parallel efforts of Russia, Iran and the Assad regime, the coalition managed to push IS back and almost completely deprive it of its territorial base. In particular, the Syrian Democratic Forces (SDF), dominated by Kurds, recaptured areas in the north and east of the country with US air support and the help of Special Forces. They extended their control far beyond the majority-Kurdish areas. In August 2018, the regime and its allies controlled around 61 percent, the SDF around 25 percent, various rebel groups around 8 percent, Turkey with its rebel allies some 2 percent and IS some 4 percent of Syrian territory.67

The Local Level: The Assad Regime, Rebels and the PYD

From the very beginning, the Assad regime categorised all opponents and insurgents as terrorists. Damas­cus did carry out limited reforms to meet the demands of individual population groups. Yet, at no point was the Syrian leadership prepared to engage in substantial negotiations with representatives of the opposition on power-sharing and political transition. Since summer 2017 at the latest, and due to its suc­cess­ful reconquest of territory, the Syrian leadership has considered itself victorious. It declared the civil war over and heralded a phase of reconstruction.68 It also made clear its intention to continue the fight against terrorism – in other words, to reconquer those areas that were still under rebel, SDF or IS con­trol. At the same time, it felt less and less pressure to engage in negotiations on a nationwide ceasefire or humanitarian access to besieged areas. Although the regime temporarily allowed territories to remain under Kurdish or rebel control so as to pool its forces elsewhere, it had (and has) no intention of accepting a permanent territorial division of power. Accordingly, it continued to besiege and bomb even the de-escala­tion zones negotiated by the Astana powers. In the first half of 2018, it reconquered those in the eastern suburbs of Damascus (East Ghouta) and those north of Homs; in July it recaptured the one in the southwest; in September it prepared an offensive to retake the one in the northwest. The offensive was only averted last minute in mid-September by a Turkish-Russian understanding on establishing a demilitarised zone along the de-escalation zone’s boundaries.69

From the outset, the Syrian opposition not only fought against the Assad regime, but also struggled to establish inclusive and effective representation with a joint programme. The obstacles were substan­tial: opposition members represented a wide range of ideological trends, were divided into forces in the country and in exile, and there were serious divergences of interest between civilian actors and armed groups. Various opposition alliances came together in several phases, most recently in November 2017 under the umbrella of the Syrian Negotiating Com­mittee (SNC) in Riyadh. As the regime’s attitude hardened, the opposition’s stance became more prag­matic in the face of military losses. At the same time, the opposition members represented in the Syrian Interim Government (SIG) – established in exile in Turkey in March 2013 – and the SNC lost influence over the conflict on the ground. Instead, radical forces excluded from or unwilling to participate in nego­tiations gained in importance.70 For example, in the largest rebel-controlled enclave in the northwest of the country, Idlib Province, the jihadist al-Nusra Front (now the Levant Liberation Committee, HTS) prevailed over more moderate competing groups and establish­ed its own local government, the so-called Nation­al Sal­va­tion Government, in November 2017. It took con­trol of the border crossings and soon dominated the local councils in many places, either directly or by infiltration.71

For quite a while, it looked as if the Kurds would be the main beneficiaries of military developments. As early as 2012, the regime had withdrawn its troops from the predominantly Kurdish populated areas in the north and northeast of the country without any significant fighting. This allowed the Kurds to an­nounce a self-administered area there (in their own words: Rojava, i.e. Western Kurdistan) in 2012, in which the Syrian PKK offshoot, the Democratic Union Party (PYD), seized power with its fighters. In March 2016, the PYD not only announced extensive autono­my in the predominantly Kurdish areas, but also estab­lished a political structure in line with the ideol­ogy of Abdullah Öcalan.72 However, the PYD’s govern­ance was by no means as progressive and democratic as it had set out to be.73

The PYD was barely able to translate its military successes into political achievements.

PYD fighters also successfully offered their services as ground troops to the international anti-IS coalition for the fight against Jihadists. In return, they received military equipment and training from the USA within the SDF. With American air support, the SDF suc­ceeded in liberating a large part of the areas controll­ed by IS in Syria – well beyond the Kurdish-majority areas. However, the PYD was barely able to translate its military success into political achievements. It was not represented either in the UN-led negotiations or in the Astana format, since Turkey was vehemently opposed to its participation. The PYD’s territorial gains also remained precarious: they did not amount to a contiguous self-governing area. In the autumn of 2016, Turkey intervened militarily to drive a wedge between the Kurdish cantons and create a buffer zone. After the US announced its continued cooperation with the SDF,74 Ankara intervened again in Janu­ary 2018 to expel the PYD from the Afrin canton and turn the Kurdish cantons into enclaves. Simulta­neous­ly, PYD fighters in the east of the country came under pressure from regime troops seeking to bring areas liberated from IS under their control.

Map 2 Military control in Syria
(as of 20 August 2018)

The Regional Level: Turkey, the Arab Gulf States, Iran and Israel

From the outset, the regional powers in the Middle East shared few common priorities motivating them to engage in constructive joint conflict management in Syria. It is true that, like the international actors, they have regularly emphasised the need for a politi­cal settlement to the conflict that would preserve Syria’s territorial integrity and state institutions and serve the interests of Syrians. Yet, this has not pre­vented them from seeking to assert their own inter­ests and instrumentalise different parties within the conflict or even intervene militarily themselves.

At first, Turkey and the Arab Gulf states (above all Saudi Arabia, Qatar, and initially also the UAE) sup­ported various rebel groups to bring about a regime change in Syria. They thereby hoped to release Syria, a country of central geopolitical relevance, from its long-standing alliance with Iran and bring forces to power which would be favourably disposed to them. They did not agree, however, on which forces should prevail, and so each promoted groups ideologically close to them, instead of focusing on uniting the anti-regime forces.75 Since the direct Russian interventions in 2015 (at the latest), the main supporters of the rebels have significantly reduced their aid. Not only did armed rebellion no longer seem effective in the face of Russian supremacy, the Arab Gulf states also shifted their main focus to the war in Yemen, which they viewed as a threatening proxy conflict with Iran in their immediate neighbourhood.

In Turkey, on the other hand, civil war with the PKK, which had flared up again in 2015, and the run-up to the parliamentary elections brought forward to June 2018, focused attention on domestic policy objectives. As a result, the government increasingly concentrated on preventing the PYD from being armed with heavy weapons, enhancing its inter­national political standing and creating a contiguous Kurdish self-governing area on its southern border. It was concerned that the latter might serve as a model for Turkish Kurds and an area for PKK fighters to re­treat to. It also wanted Syrian refugees to be returned to a “safe zone” inside Syria. As a consequence of Ankara’s changed assessment of the situation and policy priorities, Turkey turned to previously unlikely partners (both Russia and Iran) and participated in the Astana format. It was thus able to militarily pur­sue its interests in the zone of influence attributed to it in the north of the country with Russia’s approval. Cooperation with Russia also opened up new scope for foreign policy beyond Syria. Given Turkey’s tense relationship with its NATO allies and the EU, this was a significant factor.

In Syria, Iran sided with the regime and condemn­ed the uprising as an externally instigated plot to weak­en the “axis of resistance” against Israel and the USA. Tehran feared losing a long-standing and important Arab ally in the Assad regime. During the fighting, the opportunity arose for Iran to: increase its presence in Syria; gain direct access to Hezbollah-controlled areas in Lebanon and thus to the Mediterranean; boost deterrence, particularly against Israel; and there­by immunise itself against possible attacks.76 Thus, Tehran not only sent military advisors and com­bat units, but also assumed a leading role in the mili­tary organisation and financing of domestic and for­eign militias – in particular from Iraq and Afghanistan, along with Lebanese Hezbollah. This support was crucial for the Syrian regime in recapturing territory as the capability of regular army units trusted by the regime would not have sufficed.77 As a result, Iran was able to significantly expand its influence in Syria. Inter alia, it began establishing a land corridor that it controlled along with its militia allies to provide unfettered access to troops and arms from Iran via Iraq and Syria to Lebanon and the Mediterranean.

Iran and Israel have incompatible visions for Syria’s future order.

Israel, Jordan and Saudi Arabia in particular viewed the growing presence of Iran with great con­cern.78 In 2017 indications that Iran and Hezbollah could establish themselves permanently near the border and thereby open up another front against Israel initially prompted the Israeli government to intensify its diplomatic activities, particularly vis-à-vis Russia and the USA. When Russia, the USA and Jordan agreed on a de-escalation zone in southwestern Syria in July 2017, this coincided with Israel’s (and Jordan’s) interest in pacifying the situation in the border region. Yet, the agreement did not exclude a permanent presence of Iranian troops and Hezbollah militias near the border with Israel.79 As Israel did not consider its security interests to be sufficiently safe­guarded, it adopted an active policy of containment in Syria80 to militarily prevent the establishment of Iranian bases, military installations and arms fac­to­ries and the transfer of strategic weapons to Hez­bollah. The situation has deteriorated since February 2018: Israel shot down an Iranian drone that had entered Israeli airspace from Syria, and the Syrian air defence system downed an Israeli fighter jet that was in Syrian airspace. In the wake of the US withdrawal from the nuclear agreement with Iran, direct armed confrontations between Israel and Iran took place in Syria for the first time. Israel has since continued to target Iranian and Iran-led militias’ installations in Syria. What is more, their visions of political order for Syria remain incompatible and are bound to clash in the future.

The International Level: Formulaic Compromises and Russian Dominance

In June 2012 in Geneva, representatives of the inter­national community agreed on a political solution to the conflict in Syria through political transition and power-sharing.81 Yet even the great powers Russia and the USA did not stop their diplomatic, financial and military support for Syria’s opponents. Thus, instead of promoting a political settlement, they encouraged both the regime and opposition in the belief that they could win the fight militarily.82

Russia’s over-riding concerns were to be recognised as a major power and thus an indispensable actor in conflict resolution; to signal to the international com­munity that – unlike the West – it was loyal to its allies and would not tolerate an externally forced regime change (as in Libya); and to secure or rather expand its influence in the Mediterranean. It there­fore used its veto in the Security Council to protect the Syrian regime from coercive measures and inves­ti­gations of war crimes and provided financial support and military advisors. Following international uproar about the use of chemical weapons by the Assad regime, Russia was able to prevent a US and French military intervention in 201383 by successfully urging the regime to sign the Chemical Weapons Convention and agree to the elimination of its declared chemical weapons stocks. When the regime came under in­creas­ing pressure despite Iranian support, Moscow decided on direct military intervention in September 2015. Moscow’s air support turned the civil war around and enabled its ally Assad to recapture stra­tegi­cally important areas.

Since then Moscow has assumed air supremacy over Syria (with the exception of areas east of the Euphrates), established itself as the dominant military force in the conflict and tried to pacify Syria together with Iran and Turkey within the Astana process. Yet, it has not used its weight to contain the regional pow­ers’ ambitions, achieve a balance of interests, enforce principles of non-violent conflict settlement, or pro­tect Syrian civilians. Rather, its own air force was involved in the bombing of civilian targets,84 and Moscow granted Iran, Turkey and Israel extensive freedom to fight out their competing interests mili­tarily: it had no intention of coming into conflict with one of the regional powers, and protecting the civilian population was not one of its priorities.

The Syrian regime was able to assume that it had nothing to fear from the US.

The USA positioned itself against Assad from the outset. The uprising was, at least initially, seen as an opportunity to establish a democratic system of gov­ern­ment and an internationally cooperative leadership in Syria. Nevertheless, its support for the rebels was hesitant; it was viewed from the outset as a means of exerting pressure on the regime for a nego­tiated transition, rather than as a military instrument for regime change. The more radical the uprising became, the greater the reluctance to train, finance and equip the rebels with heavy weapons. The USA under President Barack Obama also appeared fickle towards the regime. Red lines regarding the use of chemical weapons were drawn but not enforced, and announced airstrikes were cancelled at the last moment. In August 2013, Barack Obama’s top mili­tary advisor made it clear that the US would not pro­vide decisive support for the rebels. In April 2013, he had already ruled out establishing a no-fly zone or other kinds of direct intervention.85 This led the regime and its supporters to assume that they had nothing to fear from the US (apart from sanctions); no pressure to negotiate was generated.

Since 2014, the US has also focused its commitment in the region on the fight against IS, initially in Iraq. In so doing, it has largely left the field to sup­porters of the Syrian regime. Only under US President Donald Trump was the fight against IS in Syria fore­grounded – substantial successes having been achiev­ed in combating terrorism in Iraq. The USA mainly relied on the Kurdish-dominated SDF on the ground and thus expanded a cooperation that had already been established under US President Barack Obama. At the same time, the Trump administration showed no interest in further supporting the rebel­lion against Assad (discontinuing its respective pro­grammes by the end of 2017) or in progress at the Geneva negotiating table. In the first half of 2018, Washington sent contradictory signals: while Presi­dent Trump announced an imminent withdrawal of US troops, then-Secretary of State Rex Tillerson still referred to a permanent presence intended to prevent IS from regaining its strength and mitigate Iran’s influence.86 Airstrikes by the USA, France and UK in response to the Assad regime’s suspected use of poison gas in Duma in April 2018 did nothing to alter this uncertain strategic orientation. In early September 2018 newly appointed US Syria envoy James J. Jeffrey announced that the US would not withdraw its troops any time soon, but rather take a “more active approach” aimed at the exit of all Iranian military and proxy forces from Syria and a defeat of IS without outlining how to get there. At the same time, the US president threatened stepped-up military action in case of a use of chemical weapons, the creation of new refugee flows or attack on innocent civilians.87

UN Mediation in Syria

In February 2012, faced with escalating violence in Syria and a Security Council hamstrung by Russian and Chinese vetoes, the UN General Assembly called on the Secretary-General to appoint a special envoy to support the League of Arab States (LAS) in mediating between regime and opposition to agree on a “Syrian-led political transition to a democratic, pluralistic political system”.88 Since the beginning of the armed confrontations, the LAS had tried to prevent the spread of violence, but only achieved temporary local cease­fires. It was unable to initiate farther-reaching media­tion because the conflict parties had no interest in nego­tiations or a power-sharing arrangement, but in­stead were set on military victory. The Syrian regime also viewed the League as biased. After all, it had suspended Syria’s membership in November 2011 because the regime had not shown itself willing to implement an LAS action plan that it had initially approved.89 Moreover, the regional organisation was itself divided over Syria. Nor did other regional efforts see success: An “Islamic Quartet” initiated by the then-Egyptian President Mohammed Morsi in early autumn 2012 with representatives from Saudi Arabia, Iran and Turkey did not meet as planned at the for­eign-minister level. The Quartet would have included Iran, a supporter of the Syrian regime, but Saudi Arabia re­jected the idea and did not participate in the meet­ing.90

Joint UN/LAS Special Envoys

On the basis of Resolution 66/253 of February 2012, the UN Secretary-General appointed Kofi Annan as the first joint mediator of the UN and LAS in the same month. Yet, former UN Secretary-General Annan resigned less than six months later because the Assad regime had refused to implement his peace plan, the opposition hoped to make military gains, and he had received no tangible support from the UN Security Council for his approach.91 His successor as UN/LAS envoy, Lakhdar Brahimi, was a former Algerian foreign minister and high-ranking UN diplomat; he resigned after not quite two years. Bashar al-Assad had announced that he would stand again as presi­dential candidate in June 2014, thus making clear that he was not prepared to embark on a political transition. Against the background of a divided Secu­rity Council, Brahimi had also not made any progress in negotiations and accused the parties in the con­flict, especially the regime, of continuing to prefer military escalation.92

Both mediators’ approaches were essentially two­fold. On the one hand, they sought ceasefires to con­tain the violence and build trust between the conflict parties. Annan also tried to add an impartial element by sending in an observer mission. In April 2012, the UN Security Council unanimously decided to deploy the UN Supervision Mission in Syria (UNSMIS), to monitor compliance with a ceasefire agreed with the regime and the Syrian National Council, and the implementation of a six-point plan.93 However, as the conflict parties had no interest in building trust but rather sought to pursue their goals militarily, the approach was doomed to failure. The mission was forced to cease almost all its activities by mid-June 2012: violence had escalated again; the observers’ freedom of movement had been increasingly re­stricted; and their patrols had been attacked several times.94 The UN/LAS mediators on the other hand relied on negotiations with the major and regional powers as they rightly viewed an agreement at the international and regional level as a prerequisite for making progress at the local level. In late June 2012, Kofi Annan brought together the so-called Action Group for Syria in Geneva. The UN and LAS Secre­taries-General, the foreign ministers of China, France, Russia, the United Kingdom, the United States, Tur­key, Iraq, Kuwait, Qatar and the EU High Represen­tative were present. Iran, however, while also an important actor was not invited. In Geneva, regional and international supporters of regime and opposition agreed on the Geneva Communiqué – in other words, to pacify the conflict through a political tran­sition to a pluralistic, democratic Syria. This process was to be shaped by a transitional governing body (TGB) with full executive powers consisting of repre­sentatives of the regime and the opposition.

Despite an agreement to resolve the conflict politically, the major and regional powers continued their military support for the adversaries.

Both mediators were virtually predestined for the office.95 They enjoyed exceptional international renown, significant diplomatic experience and excel­lent relations (both in the region and internationally). Both also made a strong personal commitment. More­over, there were no competing mediation missions at the time. Yet, their missions consistently lacked the unanimous support of the UN Security Council, which was divided on key issues with regards to the conflict. This particularly concerned Assad’s role: should the president have to resign as a precondition for negotiations (as the opposition and its supporters interpreted both the Geneva Communiqué and the Annan Plan to stipulate, although the wording of the two documents does not provide for it),96 or should he be integrated into a power-sharing arrangement? Even when the Security Council adopted the Geneva Communiqué with a delay of about a year, the differ­ent interpretations remained. Russia and China did not support calls for Assad’s resignation and, not least due to the experience in Libya, rejected any coercive measures under Chapter VII of the UN Charter. Only these, however, could have provided the necessary teeth to the mediation missions. The major and regional powers also maintained their military sup­port for one or other of the adversaries; their agree­ment on political conflict resolution was thus mean­ingless. The two mediators also failed to bring all relevant local conflict parties and their regional sup­porters to the negotiating table. Thus, the dominant party in the Kurdish territories, the PYD, was exclud­ed by Turkey’s veto – a pattern that continues to this day in all international negotiating forums – and Iran was not allowed to participate because the oppo­sition, the USA and Saudi Arabia opposed it.

The UN Mediation Mission: Mandate, Approach and Outcome

On 10 July 2014, the experienced Italian-Swedish diplomat and long-time UN staff member Staffan de Mistura, who had inter alia led the UN missions in Iraq and Afghanistan, was appointed UN Special Envoy for Syria. In consultation with the LAS, the UN Secretary-General appointed the Egyptian career diplomat Ramzy Ezzeldin Ramzy as his deputy. Given the escalating violence, the UN mediator wanted to find ways to draw international attention back to Syria and make the international community – both major and regional powers, this time including Iran – assume responsibility for engaging in conflict resolution. In the absence of an appropriate international framework, he viewed talks on power-sharing between the Syrian conflict parties, which continued to strive for military victory, as less realistic.97

De Mistura also relied on conflict containment and confidence-building through ceasefires. He initially attempted to achieve a ceasefire in Syria’s most popu­lous city, Aleppo, which was partly controlled by the rebels and partly by the regime. Success in this sym­bolic city was meant to illustrate the possibility of progress. From the very outset, however, it was clear that the parties to the conflict were not prepared to compromise precisely because of the high symbolic value of the city.98

In addition, in the first months of his term of office de Mistura conducted extensive consultations with representatives of the Syrian regime, the opposition, women, civil-society groups and a great variety of rebel groups to explore possible agreements and points of consensus. Participants in the two permanent consultation forums – the Women Advisory Board and the Civil Society Support Room – were given the opportunity to highlight civil-society, humanitarian and women’s concerns, exchange views with mission representatives and come into contact with potential donors. Yet, the talks left many of them with the impression that they ultimately had little relevance for the behaviour of the UN mediator, the dynamics of the conflict or the indirect conversa­tions between regime and opposition.99

It was only after Russia’s direct military interven­tion that an international agreement on a ceasefire and a political transition was reached in Vienna in November 2015 on a Russian-American initiative: the two objectives were to be pursued in parallel.100 The members of the so-called International Syria Support Group (ISSG) – in which all relevant international and regional actors, including Iran, were represented for the first time – agreed to initiate formal nego­tiations between the Syrian conflict parties under UN mediation by early 2016 at the latest.101 A broad spectrum of Syrian representatives was to participate in these intra-Syrian negotiations. A resolution to the conflict was to ensure the “principles of Syria’s unity, independence, territorial integrity and non-sectarian character”; preserve state institutions; and guarantee the rights of all Syrians regardless of ethnicity or reli­gious denomination. Six months were set aside for the establishment of “credible, inclusive and non-sec­tarian governance” and for the setting of a time­table and modalities for a constitutional process. Within 18 months, free and fair elections were to take place under UN supervision and with the participation of all Syrians, including those living in the diaspora.

De Mistura’s main function at this stage was to advise the regional and international parties in such a way that an agreement could be reached even with­out a consensus on Assad’s future. In December 2015, the UN Security Council unanimously confirmed in Resolution 2254 the transitional roadmap presented by the ISSG on the basis of the Geneva Communiqué for a “Syrian-led, Syrian-owned political transition to end the conflict”. It instructed the special envoy to organise talks between the Syrian conflict parties in Geneva and to work out a comprehensive ceasefire and an appropriate monitoring mechanism.102 IS and al‑Qaeda offshoots, such as the al-Nusra Front, were to be excluded from the ceasefire. Jordan was tasked with reaching a consensus in consultation with the other ISSG members on which terrorist persons and groups would also be the targets of joint action. The consensus never came about. The ISSG members had incompatible definitions of who was a terrorist and who fought for legitimate causes, divergences which were characteristic of the differences in the wider international community, and which later turned out to be an important obstacle in ensuring compliance with ceasefires, often legitimising regime offensives.

De Mistura tried to counter an imminent obstruction of intra-Syrian talks on the question of whether to deal first with terrorism or with a transitional gov­ernment by proposing four parallel working groups to negotiate a framework document for the implementation of the Geneva Communiqué.103 Still, the intra-Syrian talks had a difficult start in early Feb­ru­ary 2016. De Mistura suspended them two days after the launch in the face of renewed escalating violence and severe restrictions on humanitarian access, and de­manded the major powers create an environment that would allow negotiations to take place. In fact, at his instigation, the ISSG set up a task force each on humanitarian issues and ceasefires in mid-February 2016 (the Humanitarian Task Force was chaired by the UN, whilst the Ceasefire Task Force initially had Russian and American co-chairs). On 26 February 2016, the Security Council adopted Resolution 2268 estab­lishing a nation-wide ceasefire which held tem­porari­ly, and improved humanitarian access to con­tested areas.

Consequently, de Mistura convened the second round of talks in Geneva for mid-March 2016, aiming to form a transitional government within six months. However, the representatives of the conflict parties refused to sit down together beyond the opening of these intra-Syrian talks. This meant that there were no direct intra-Syrian negotiations at any time; in­stead de Mistura and his team spoke separately with each side.104 Accordingly, the working groups were not set up either. Most of the content intended for the groups, however, was placed in various thematic “baskets”, which de Mistura and his team worked on with the respective delegations. In addition, escalat­ing violence repeatedly caused long breaks between the individual rounds of talks. In September 2016, cooperation in the ISSG between the USA and Russia broke down. The American Air Force had bombed regime positions in Deir el-Zor, and the USA blamed Russia for an air attack against a UN aid convoy near Aleppo.105 Washington and Moscow then resigned their joint presidency of the ISSG ceasefire task force, handing it over to the UN mission until further notice.

The UN mediator was unable to bring about either a lasting ceasefire or a political solution to the conflict.

Against this background, the talks made no substantial progress. In de Mistura’s words in July 2017, there was “no breakthrough, no breakdown, no one walking out, incremental progress”.106 The talks at least nominally committed the opposition and regime to the principles of conflict resolution, and in the tech­nical consultations many of the issues were thoroughly deliberated and thought through. Ulti­mately, how­ever, the UN mediator was unable to bring about a lasting ceasefire (or rather prevent an escalation of violence), enforce humanitarian access to besieged areas, effectively protect the civilian population or even come close to a political solution to the conflict. His involvement only allowed him to support the UN in exercising its humanitarian role in individual cases through diplomacy and negotiations on aid convoys, evacuations, ceasefires and the protection of civilians.

Towards an Explanation – Relevant Factors

International mediation efforts to date have shown that the timing of mediation and its acceptance by the conflict parties are important factors for success. Another relevant factor is the influence that the mission can exert as a result of backing by the inter­national community, primarily the UN Security Coun­cil. The latter decides what leverage the mission can apply. This influence diminishes when in competition with other negotiating forums. The specific me­diation approach (the mediator’s prestige, impar­tial­ity, inclusiveness) is also decisive for success or failure.107

Acceptance and Timing of the Mediation

The Syrian conflict parties accepted the UN mediator and engaged in the mission at least pro forma. They also took part in all rounds of talks, albeit with some delays and interruptions. However, they showed no genuine interest in mediation. At no time were the regime and opposition in a situation of mutually hurting stalemate in which they might have preferred a political conflict resolution to a military one; thus the conflict was never even close to being “ripe for resolution”.108 The regime did not even give its repre­sentative in Geneva, Bashar Jaafari, authority to nego­tiate; he merely acted as a spokesman.

The direct intervention of regional and major powers increasingly pushed their own interests to the fore.

On the contrary, external support meant the con­flict was permanently fuelled with financial assis­tance, arms, military advisors, etc.109 The significant shift in military power that resulted from Russia’s direct military intervention also did not push the par­ties closer to a political settlement. While the Syrian opposition became more willing to negotiate and take a more pragmatic approach due to the massive mili­tary setbacks it was suffering, the regime felt less and less under pressure to negotiate power-sharing and a po­liti­cal transition, or to enter into any kind of com­pro­mise due to its military success and Russia’s sup­port.

In addition, intra-Syrian negotiations did not begin until the conflict was already highly international­ised. The direct military intervention of regional and major powers increasingly pushed their own interests to the fore. However, these could not be dealt with dur­ing the Geneva talks between the Syrian conflict parties.

The Mediator’s Leverage

Whilst the talks mediated by the UN were the only ones that all relevant actors recognised as legitimate and there was ostensibly international consensus on the mediation’s mandate (Geneva Communiqué, ISSG statements, Security Council Resolution 2254), that con­sensus remained superficial. Also, regional and major powers consistently pursued military approach­es to conflict management. At the same time, largely because of Russia’s veto, the Security Council was unable to impose coercive measures (sanctions, a no-fly zone, an arms embargo, the prosecution of war crimes or crimes against humanity) that could have had an impact on the course of the conflict, served to protect the civilian population or helped to establish the responsibility of the conflict parties. The mission thus had no “coercive power”.

The efforts of the UN mediator were also challeng­ed by the establishment of an alternative format for ne­go­tiation and conflict management. After the regime had recaptured Aleppo in December 2016, and with a lack of progress in the Geneva negotiations, Russia sought to convert its military dominance into achiev­ing the pacification of Syria. Through the so-called Astana process, Russia alongside the regional powers of Iran and Turkey wanted to monitor a ceasefire announced in December 2016 and resume political conflict resolution. Moscow and its two regional partners therefore agreed on zones of influence and brought representatives of the regime and rebel groups to Astana in January 2017. Those talks were also indirect. In May 2017, the three guarantors agreed on four de-escalation zones in which all firing was to cease and humanitarian access was to be assured – with the exception of the fight against jihadists.110 In subsequent meetings, this de-escalation zone approach was further elaborated and specified. It did indeed initially lead to a significant reduction in violence and improve humanitarian access. How­ever, it collapsed after only six months, primarily because the regime continued to focus on reconquering territory. By mid-2018, only one de-escalation zone was left, after the other three had been retaken by the regime. A regime offensive to reconquer Idlib province was averted (for the time being) by Turkish-Russian compromise after international pressure and a Turkish show of force. Overall, Russia was unable to progress towards a nationwide ceasefire either. Its backing for the Syrian army offensives gave the strong impression that Russia had anyhow conceived of establishing de-escalation zones to allow the regime to concentrate its forces elsewhere, before attempting to reconquer the entire territory step by step.

Moscow’s efforts to convene a National Dialogue to legitimise conflict resolution through handpicked Syrians also failed. The SNC boycotted the meeting in Sochi at the end of January 2018 because it did not consider Russia to be an honest mediator and because Russia’s invitation policy guaranteed a pro-regime majority from the outset. The PYD was not represented following Turkey’s objection; other Kurds did not participate as a protest against the Turkish military operation launched just before in Afrin. In turn, the regime’s representatives opposed Russian proposals for a constitution. Ultimately, Russian mediators merely committed the participants to the principles of conflict resolution and handed the constitutional process back to the UN in Geneva.111 By mid-Septem­ber 2018, three lists representing the regime, the opposition and experts had been prepared, but the composition of the committee as well as procedural and substantative issues related to the constitutional process remained contested by the conflict parties.112

Although Russia stressed from the outset that its efforts complemented the UN mediation, it never­theless undermined it.113 First, Russia diverted local, regional and international attention and energy to Astana, devaluing the Geneva talks. Second, Russia upended the transition roadmap agreed in the Geneva Communiqué and Security Council Resolution 2254 (in the following order: ceasefire, establish­ment of a TGB of representatives of the government and opposition, drafting a new constitution, and finally parliamentary and presidential elections) by initiating the constitutional process without there having been a transition government first. The pres­entation of a draft constitution drawn up by Russia was contrary to the principle of local ownership. Staffan de Mistura nevertheless decided to play along with the game of complementarity, since only Russia and its Astana partners were capable of enforcing cease­fires on the ground.114 Simultaneously, he in­sist­ed that Astana was not a format for political negotiations but only for ceasefires and that the constitutional process would have to be brought back under the aegis of the UN. By attending most of the meet­ings in Astana and Sochi, he also intended to signal that the two processes were complementary. Yet, as a result, the political process was basically reduced to the constitutional committee’s work.

The major powers adopted different attitudes towards the UN mediator. While Russia sent the renowned expert Vitaly Naumkin as an advisor to de Mistura, the White House showed no interest in sending a representative to support or influence the UN efforts.115 Under President Trump, the US also downgraded its presence on the Ceasefire Task Force, no longer sending any military representatives.

Perceptions of the Mediator

By choosing de Mistura, the UN nominated a respect­ed and crisis-tested diplomat who had many years of experience as a “trouble shooter” at the UN.116 He was given as assistant an Egyptian diplomat with excellent contacts in the Arab world, including the Syrian regime. Despite his qualities, de Mistura did not have the prestige or prominence of his predecessors, a fact that initially seemed to make it difficult for him to access decision-makers in Turkey and the Gulf States. However, at the time those states had little interest in a negotiated conflict settlement anyway.117 De Mistura was criticised for not pushing the mediation forwards vigorously or systematically enough, but beginning instead by establishing a broad network of contacts and holding a large number of consultations, the ob­jec­tives of which were not always clear.118 Team mem­bers describe his style as improvisational, slow and lacking in ambition; he is characterised rather as re­active, following the initiatives of the major powers rather than taking the lead and accelerating the pace.119 Relations between him and UN headquarters were said to be tense at times. Secretaries-General (in particular António Guterres) have repeatedly criticised the stagnating process, but left open wheth­er their criticism was aimed primarily at de Mistura or at the lack of support from the Security Council and the ISSG.

Although opposition representatives recognised de Mistura’s negotiating skills, they questioned his impartiality,120 inter alia because the UN mission main­tains a presence in areas under regime control as well as contacts with the regime. Criticism of other UN agencies cooperating with Damascus was also blamed on de Mistura. There was also disagreement with statements made by the mediator and his en­tourage: their “neutral” presentation of the conflict dynamics on the ground, which did not clearly attri­bute responsibilities and concealed asymmetry, for example in war crimes, was seen as contributing to a distorted narrative.121 The fact that the mission was ultimately impotent in the face of war crimes dimin­ished its credibility, especially in the eyes of Syrian civil-society actors. On several occasions, however, the special envoy himself made remarks which the opposition considered biased and inappropriate.122 Moreover, his participation in the so-called National Dialogue in Sochi, organised by the Astana powers, was disapproved of by the opposition gathered in the SNC, which boycotted the meeting.123

In general, the mediator delegated relations with opposition bodies to members of his team and refrain­ed from attending even crucial opposition forums, such as the Riyadh I and Riyadh II Conferences (December 2015, November 2017). He was there­fore unable to establish close and trusting con­tacts with the opposition, and regime opponents gained the impression that he was exerting consider­ably more pressure on them than on the regime.124 Nevertheless, opposition representatives considered the UN mission to be much less biased than Russian mediation, Russia siding so clearly with the regime and thus becoming a party to the conflict. Accordingly, they argued that it would be advantageous to have an internationally legitimate format in which the details of conflict resolution could be negotiated in due course, even if current negotiations did not pro­duce results.125 The regime itself repeatedly tried to discredit de Mistura in the Syrian state press. It had no interest in establishing relationships of trust and frequently demonstrated its condescension – for example by leaking a video recorded at the indirect talks showing Syrian representative Jaafari appearing to lecture de Mistura, which it uploaded to YouTube in autumn 2017.

Inclusiveness and Representativeness

The indirect intra-Syrian talks in Geneva gradually included more and more groups, at least on the oppo­sition side, by extending the negotiating delegation in Riyadh under the umbrella of the SNC.126 Yet, they still excluded a significant number of the rebels (not just from IS and HTS, but others as well) because they refused to join or were not invited. This was particularly problematic because these groups were in con­trol of a significant proportion of the areas not held by the government. In fact, in the summer of 2018 the excluded Kurdish party PYD controlled consider­ably more territory than the opposition rebels. More seriously still, there was no forum for mediation where the regional powers would have been able to strike a balance between their conflicting interests or at least reach an agreement on a minimal code of conduct, for example on maintaining ceasefires.

Conclusions, Prospects and Recommen­dations for European Approaches

The approach of the UN mediation in Syria, as well as the UN/LAS mediation before it, was to contain the civil war and stabilise the country sustainably by establishing power-sharing between regime and opposition, as well as a Syrian-led political transition to a more inclusive and representative political sys­tem. In this vein, parallel talks were held on cease­fires on the one hand and power-sharing and political transition on the other. However, neither UN media­tor Staffan de Mistura nor his two predecessors, Kofi Annan or Lakhdar Brahimi, succeeded in bringing the Syrian conflict parties together for substantive nego­tiations. Even in indirect talks, de Mistura was unable to make any progress on power-sharing and political transition, or to contribute to a lasting pacification of the conflict and effective protection of the civilian population. Only selective agreements on ceasefires, aid deliveries and evacuations were concluded. Ulti­mately, military dynamics determined the scope and pace of UN mediation throughout. Given the inter­nationalisation of the conflict, it was largely obvious from the outset that the intra-Syrian talks would not see progress. Also, none of the regional or major powers focused on protecting Syrian civilians or on achieving sustainable conflict resolution. No respon­sibility for war crimes or crimes against humanity has been established.127

The UN mission was faced with the dilemma of either legitimising the Russian approach or distancing itself and giving up opportunities for influence.

While Russia prevented coercive measures against the Assad regime from the outset, its direct military intervention since September 2015 has increasingly made it a force shaping the Syrian order. It supported a military victory for the regime – in sharp contrast to its proclaimed goal of resolving the conflict by nego­tiating power-sharing and a political transition. It intended to legitimise itself on the ground as well as internationally by implementing isolated elements of Security Council Resolution 2254, namely con­sti­tutional reform and elections. Thus, the UN mission was confronted with a dilemma: if it served as a fig leaf for the Russian approach of predominantly mili­tary conflict resolution, it would at least remain “in play”, and be able to hold preliminary talks on the future order in Syria and selectively contribute to improving the humanitarian situation. If it ceased its efforts and clearly distanced itself from the Russian approach, it would deprive itself of the few opportunities to exert influence that it possessed. De Mistura decided on the first approach.

Three factors were (and are) key among the many factors blocking the success of UN mediation in Syria. First, the international community agreed on political conflict resolution in principle, but third parties de facto continued to back military support for the con­flict parties and increasingly enforced their own inter­ests through direct military intervention. This led to a situation in which a favourable mediation environ­ment could not emerge in the sense of the conflict becoming ripe for resolution. Rather, the conflict dy­namics made this possibility increasingly remote. Second, while the Syrian conflict parties accepted the UN’s mediation and participated in the Geneva talks, they did so purely pro forma. They were not willing to enter into substantive negotiations, but relied instead on enforcing their interests militarily, albeit with clear differences between the regime and oppo­sition. The opposition delegation acted much more constructively in the indirect talks than the regime and became increasingly pragmatic as its military strength diminished.128 Third, the disparate positions in the Security Council, particularly between Russia and the West, prevented effective coercive meas­ures – such as UN sanctions, an arms embargo, the imposition of a no-fly zone or the deployment of UN observers or even a peacekeeping force – that would have put pressure on the parties to the conflict, pre­vented human rights violations and helped to calm the conflict.

Factors originating in the approach chosen by the mediation mission clearly have less significance here. Thus, an earlier start to the intra-Syrian talks, a higher degree of inclusiveness and representativeness of the forums (for instance, by including a PYD delegation) or a more energetic mediator would not have made a decisive difference.

Due to these factors, a successful conclusion to the UN mediation under the terms of its mandate and thus a “Syrian-owned, Syrian-led” process can also be excluded for the future. To begin with, the issue of reconciling regional and international interests in Syria is key to ending the armed confrontations. How­ever, no format exists for this. The intra-Syrian talks are not suitable for it; important regional powers, espe­cially Israel and Saudi Arabia, lack representation in the Astana format; and Russia, itself a party to the conflict, has so far shown no willingness to bring about a comprehensive arrangement. Given the cur­rent US-Russian tensions, it is also unrealistic to expect that these conflicts of interest would be dealt with constructively in a revived ISSG or in a format resembling the 2016 Lausanne meeting.129

European members of the so-called Small Group130 should employ their respective diplomatic channels to push the regional and international players to­wards conduct that serves de-escalation and is in con­formity with international (humanitarian) law, thus avoiding further military spill-over effects and pro­tecting Syria’s civilian population. They should also make use of any opportunities by working towards conflict resolution: for instance, by using Iran’s in­terest in maintaining the nuclear agreement (JCPOA) to ensure that Tehran reduces its direct and indirect presence in Syria, especially in the border area with Israel and Jordan; refrain from provoking Israel; and exert pressure on the Assad regime to agree to a nego­tiated settlement. The incentives at the disposal of the EU and its member states towards Damascus – recog­nising it by once again posting ambassadors, suspend­ing sanctions, committing to reconstruction – should be used strategically to bring about a more inclusive political system and prevent serious human-rights abuses and discrimination in future. In this vein, the EU and its member states should also give political backing to legal proceedings to determine the respon­sibility for war crimes and promote their success by making resources available for fact-finding and legal assistance.

Mareike Transfeld

Yemen: Conflict Escalation despite UN‑Mediated Power-Sharing

Among the countries affected by the so-called Arab Spring, Yemen is the only one where a president in office was deposed by a power-sharing agreement. In March 2011, Saudi Arabia launched an initiative within the Gulf Cooperation Council to prevent an escalation of the violence. UN Special Envoy Jamal Benomar played a leading role in the negotiations. When the negotiating parties signed the Gulf Agree­ment in November 2011, thus launching a process of political transition, the sponsors of the deal believed civil war had been averted in Yemen. In spring 2014 the Gulf Cooperation Council Initiative and its im­plementation were still being feted as a model for the region. They were regarded as the UN’s showpiece for successful conflict mediation.

That the negotiating parties reached agreement over the Gulf Initiative is a UN mediation success. However, the civil war that it was meant to prevent was merely delayed by the deal. By the time the tran­sitional government (formed in December 2011) was overthrown by Ansarallah (also referred to as Houthi movement or Houthis) in January 2015, the Agreement had obviously failed. The UN special envoy’s attempts to persuade Ansarallah to return to the po­liti­cal process were unproductive.

In spring 2015, Saudi Arabia forged a ten-state coalition for military intervention, with the alleged goal of re-establishing the transitional government in Sana’a so as to allow the political process to continue within the framework of the GCC Initiative. Three years later, the intervention, intended as a short mili­tary operation, has got no closer to realising its stra­tegic aims. In the meantime, Yemen has developed into the world’s worst humanitarian crisis, with 22 million people in need of aid. The UN has made hard­ly any progress in peace talks. In September 2018, the new UN Special Envoy, the Briton Martin Griffiths, issued invitations to the most recent round of peace talks in Geneva. As the first attempt to launch talks since summer 2016, the talks were highly anticipated both by the Yemeni population and the international community. But the no-show of the Houthi mediators, after their demands were not met by the UN, proved the pessimists ahead of the talks right. While the current UN envoy was able to resume dialogue with the Houthis, who had refused for over half a year to speak to his predecessor, Ismail Ould Cheikh Ahmed, tangible mediation successes remain out of reach. Not least due to a lack of progress, the latter resigned in February 2018. This chapter focuses on the following issues: what factors have contributed to the failure of the power-sharing agreement? Why has the UN mission not been able to bring the conflict parties back to the negotiating table and pacify the conflict?

The first section will examine the conflict dynamics at the local, regional and international level. This demonstrates both the difficult conditions under which the UN mediates in Yemen and the network of interests that will have to be included in a successful conflict resolution. The second section is dedicated to the UN mediation mission, determining two phases: first, the negotiation and implementation of the Gulf Agreement, and then the attempt by the UN mission to encourage the conflict parties to resume the politi­cal process.

The Disintegration of a Nation-State

Seven years after the countrywide protests against the rule of Ali Abdullah Saleh, the political map of Yemen has permanently changed. A power vacuum at the national level and constantly changing local, regional and international alliances have greatly limited the influence of the former political elite. A unified nation-state has de facto ceased to exist.

Map 3 Situation in Yemen (as of September 2018)

In March 2011, the Gulf Cooperation Council focused on the conflict amongst the elite in Sana’a, neglecting conflicts on the country’s periphery.

The root causes of today’s conflict are a power struggle within the political elite in Sana’a, which has descended into violence since 2011; and con­flicts between that elite and the Houthis in northern Yemen as well as the independence movement in southern Yemen, the Southern Movement (also known as al-Hirak). In March 2011 the Gulf Cooperation Council focused on the conflict amongst the elite in Sana’a, neglecting conflicts on the country’s pe­riph­ery. The GCC Initiative merely provided for the deposition of President Ali Abdullah Saleh in exc­hange for his immunity, and power-sharing between government and opposition parties. At the time, the opposition parties took the form of an alliance headed by the Islah Party. The latter consists mainly of groups that were close to the regime – including the Yemeni Muslim Brotherhood, Salafists, northern Yemeni tribes and conservative businessmen. The UN special envoy to Yemen, the EU and the US backed the negotiations. In November 2011 the government and opposition parties signed the agreement and an implementation mechanism for a two-year-long tran­sition process.131 On the basis of the agreement, Abdu Rabu Mansour Hadi was nominated as the consensus candidate for interim president, and a unity government was formed. The implementation mechanism of the agreement also included measures on transitional justice, security-sector reform and a national dialogue aimed at deciding the structure of the state and paci­fying intrastate conflicts.

Although the Yemeni elite agreed to abide by the rules stipulated by the GCC Initiative, during the implementation of the agreement each faction of the elite pursued its own interests so as to maintain or expand its position of power. They used their networks in the military, political parties, tribes and media. The fact that not one of the elite groups was really interested in sharing power was one of the main reasons for the failure of the agreement.132 Local groups, first and foremost the Houthis and the Southern Movement, seized the chance to create new facts on the ground. As a result, the claim to power of the internationally recognised government under Hadi is being challenged throughout the country, and state and armed forces are fragmented.

Meanwhile, actors not involved in the GCC Initiative now dominate and therefore neither feel re­spon­sible for its success nor champion its implementation. In the northwest of the country and in its capital, Houthis are the dominant force today. Following an alliance with former President Saleh and his networks in the military, tribes and media, the group was able to spread rapidly in northern Yemen using force from mid-2014, and in September 2014 they conquered the capital. This cooperation with Saleh was new: from 2004 to 2010, the Houthis had been at war with the Saleh government, rebelling against the marginalisation of Zaidi133 Saada in the north of the country and condemning the government’s involvement in the US’s war on terror. In 2001 the Houthis had joined the countrywide protests against Saleh, and from 2013 onwards they were involved in the formal politi­cal process as part of the National Dialogue. However, by September 2014, Saleh had entered into an alli­ance with them in order to weaken his opponents – interim president Hadi and the Islah Party – by taking advantage of the ideologically charged power conflict between the nominally Sunni Islah Party and the Zaidi Houthis. Since 2011, the Islah Party had prof­ited from the changes and gained in political influence, which was one of the reasons why it be­came a target for the Houthis. In autumn 2014, the Houthis destroyed the military capacities of the Islah Party almost entirely.

The Houthi-Saleh alliance also mobilised the popu­lation against Hadi. Accusations of corruption and the argument that Hadi’s legitimacy had expired with his term of office (limited to two years) in February 2014 met with broad approval. The Houthis were able to portray themselves as a popular force and deprive the Hadi government of power in a coup d’état, which was consummated with Hadi’s provisional resigna­tion in January 2015 and the creation of the Houthi Revolutionary Committee in February 2015. The Saudi-led military intervention against the Houthi-Saleh alliance in the following month welded the alliance even more tightly. The Houthis and the Gen­eral People’s Congress under Saleh further con­soli­dated their rule in the north and presented them­selves as the rightful government.134 However, their alliance was merely pragmatic, and has been visibly fracturing since mid-2017. Saleh repeatedly signalled to the Saudis that he was prepared to talk. After re­voking his alliance with the Houthis, he was murder­ed on 4 December 2017. Many of his supporters were detained or killed by the Houthis.135

The regions outside of the Houthi-controlled northwest are politically and militarily fragmented.

The regions outside of the Houthi-controlled north­west are politically and militarily fragmented.136 The dominant actors there, such as local tribes, militias and other groups, although hostile to the Houthis, do not unconditionally recognise the Hadi government. What united them at the beginning of the Saudi-led intervention was defence against the expanding Houthi-Saleh alliance. This also explains their sup­port for the military intervention by the coalition. In turn, the Saudi-led coalition supplied local militias with money, weapons and training, thereby de facto under­mining the authority of the Hadi government. Al‑Qaeda in the Arabian Peninsula (AQAP) benefited from the conflict, which pushed resistance groups into entering into strategic alliances with it against the Houthis.137 Since summer 2015, Aden has been con­sidered the nominal capital of the Hadi government, but as Hadi himself enjoys hardly any support in the Southern Movement-dominated city, security concerns prevent him from actually governing from there.

The southern independence movement supported the intervention not out of loyalty towards Hadi, but primarily to protect the south from a takeover by a northern Yemeni force, i.e. the Houthi-Saleh alli­ance.138 The Southern Movement was formed in 2007 out of a loose movement of retired officers from the formerly independent South Yemen, which has been demanding independence since at least 2009. In 2011 the movement also joined the countrywide protests against Saleh. However, when the National Dialogue began in March 2013, the Southern Movement was marginalised. It saw the implementation of the GCC Initiative as a northern Yemeni process that did not acknowledge the south’s interest in an independent state. After Hadi had fled from the Houthis to Aden in February 2015 and withdrawn his resignation, he and the Southern Movement appeared to form an alli­ance.139 In reality, however, the Southern Movement has used the conflict configuration to shore up its own position and has taken further steps towards an autonomous South Yemen since the military in­tervention. In May 2017, the Southern Transitional Council was established in Aden, which claims power over the entire south. This fragmentation of the national state has been decisively driven forward bySaudi and UAE support for the various groups fighting the Houthis.

The Gulf Cooperation Council: From Mediator to Conflict Party

The states of the Gulf Cooperation Council were the first external actors to seek a resolution to the con­flict, from 2011 onwards. Saudi Arabia, whose royal family had an ambivalent relationship with Saleh and was concerned about the escalating violence between elites in the neighbouring country, took the lead. It saw itself as a mediator and regularly offered to host negotiations. However, it was the UN special envoy who undertook the detailed negotiation of the agree­ment and mediated during the implementation of the GCC Initiative. At the same time, Saudi Arabia did not limit its action in Yemen to this multilateral frame­work provided by the GCC Initiative, but also tried to influence Yemeni politics unilaterally, especially as regards Saudi opposition to the Muslim Brotherhood and its rivalry with Iran. The military intervention – which continues to be of questionable legitimacy140 since Riyadh merely refers to Hadi’s call for help of 23 March 2015 – should also be seen in this context. After Hadi had fled the capital to the south, he feared for his life when fighters of the Houthi-Saleh alliance were entering Aden, and appealed to the Gulf Co­opera­tion Council to use military means against the Houthis. Saudi Arabia claimed to be pursuing the goal of re-establishing the legitimate president, Hadi, in Sana’a and continuing the UN-led political process. This stated aim secured Saudi Arabia the support of the international community for its military inter­vention. However, the royal family is also pursuing its interest in preventing the spread of Iran’s influence in Yemen. Domestic political motivations also play a part. The Saudi King Abdullah bin Abdulaziz had died two months before the invasion; the defence minister and son of the new king, Mohammed bin Salman, manoeuvred himself into a more favourable position during the succession conflict through the Yemeni military campaign.141 Saudi Arabia expected to reach its war objectives relatively quickly,142 but more than three years later, not one of them has been realised.

The coalition conducting the military intervention in Yemen since March 2015 mainly consists of Sunni Arab states, many of whom are involved only sym­bolically. They tend to be more interested in good relations with Saudi Arabia than in Yemen itself. The main purpose of the coalition is thus to provide a multilateral framework for the intervention and strengthen Saudi Arabia’s position vis-à-vis Iran.143 Saudi Arabia and the UAE form the core of the coa­lition, but pursue different strategies, operate in different deployment zones, and support divergent groups.

Saudi Arabia’s priority in Yemen is to defeat the Houthis in order to reinstall Hadi in Sana’a. It has primarily relied on air attacks and a sea, land and air blockade that aims to put pressure on the Houthis and prevent them from receiving weapons from Iran. However, Iranian arms have nevertheless reached Houthi hands even since the 2015 intervention. UN experts view the blockade as a deliberate measure to bring about starvation as an instrument of war.144 The blockade has contributed significantly to the dramatic humanitarian crisis, since the coalition has also dras­tically restricted food imports and the distribution of humanitarian aid. The air raids on military targets have weakened Saleh’s network in particular. Yet Saudi Arabia has not been able to deprive the Houthis of their power; not even after the death of Saleh, which meant the loss of the Houthis’ strongest backer. Although the Hadi government continues to exist in Saudi exile, it has now lost support in Yemen itself.

The UAE’s main focus is on the south, where they have troops on the ground. A key interest is control­ling the port of Aden. UAE troops managed to recap­ture the town from the Houthis in summer 2015 with the support of Southern Movement fighters. Since the Islah Party mainly consists of the Yemeni Muslim Brotherhood and is utterly unsuitable for the UAE as a cooperation partner in the south, the Emirates rely on the Southern Movement. They thus promote its quest for independence and contribute to the creation of parallel structures. The UAE also maintained a rela­tionship with former president Saleh through con­tacts with his son Ahmed Ali, who is based in Abu Dhabi.145 Unlike Saudi Arabia, the Emirates continued to see Saleh as part of a possible solution in northern Yemen, despite his alliance with the Houthis. The UAE also take more decisive action against AQAP in southern Yemen than their Saudi partner. Saudi Arabia watched the expansion of AQAP more passive­ly, inter alia no doubt because AQAP opposed the Houthis. The two Gulf states are united in their wish to push back Iran’s influence in Yemen.

Since Saleh’s death in late 2017 and the conflict between the Southern Transitional Council in Aden and the Hadi government in early January 2018, Saudi Arabia and the UAE have been increasingly forced to find common ground. Since May 2018, they have jointly powered an offensive against the port of Hodeidah, which the international community fears will dramatically worsen the humanitarian situation. Griffiths’ attempts to stop the offensive through media­tion were inconclusive. With the failure of the Geneva talks in early September 2018, violence around Hodeidah has once again escalated.146 The coalition hopes the offensive on Hodeidah port will bring the Houthis to their knees, as it would allow cutting off supply routes from the port to Sana’a. The port is an important transhipment centre for relief goods, but is also considered a major source of in­come for the Houthis and a possible entry point for smuggled Iranian weapons.

The Houthis are not merely carrying out orders from Iran.

Saudi Arabia sees the Houthis’ success in capturing the capital Sana’a and in threatening Riyadh with their missiles as the direct result of Iranian support. Since the 2011 protests, ties between the Houthis and Iran have intensified. The former have received media and military training from Lebanese Hezbollah, which is funded by Tehran. However, the Houthis are not merely carrying out orders from Iran. While Tehran welcomed the Houthi takeover of Sana’a, it warned them off further expansion south of Sana’a.147 In fact, Saudi policy in Yemen has also contributed to strengthening the Houthis ever since Riyadh ended its support for the Houthis’ greatest adversary, the Islah Party, in March 2013 and weakened the Party by halt­ing regular payments to it and its supporters among the tribes.148 Direct relations between the Houthis and Iran were expanded especially after the Houthi take­over in September 2014: Iran announced regular flights between Tehran and Sana’a as well as infra­structure projects for electricity supply. UN experts also assume that Iran supplies the Houthis with weapons, including missiles.149 Ever since the dis­so­lution of the nuclear agreement by President Trump first loomed in February 2018, Iran has shown itself more cooperative towards Europe as regards a politi­cal solution in Yemen.150

Western Priorities in Yemen: AQAP and the Partnership with Saudi Arabia

After the military intervention began in March 2015, the US, UK and France soon declared their readiness to provide political and military support. Germany did not question the legitimacy of the deployment. All four states had already been substantially in­volv­ed within the EU and UN in advocating the nego­tiation and implementation of the GCC Initiative. The US and UK assist the coalition with logistics and intel­ligence, and advise Saudi officers in a joint opera­tions centre in the Saudi capital. Among Western countries, the US is traditionally most involved in Yemen. Its priority has been to continue the anti-terrorism cooperation in Yemen against AQAP, which has been ongoing since 2002. Its commitment since 2011 has been devised accordingly; but its involvement in the current conflict is counterproductive to its own for­eign policy goals in Yemen.

Hadi’s overthrow also deprived the US war on terror in Yemen of its foundation.

Hadi’s overthrow also deprived the US war on terror in Yemen of its foundation. Although the US was able to continue some anti-terrorism operations in Yemen even after the coup d’état, it quickly lost contact with partners in the Yemeni armed forces, which were now under Houthi control.151 In order to continue the fight against AQAP, the US threw its weight behind Hadi and simultaneously sought con­tact with the Houthis, who had previously flatly rejected Yemeni cooperation with the US in fighting AQAP. However, at the same time, the Houthis were foregrounding their own fight against the terrorist group to justify their advances into southern Yemen, which made them seem a natural ally for Washington. Whilst there was an exchange between the US and the Houthis, and whilst some attacks on AQAP even looked coordinated, no long-term cooperation was established.152

The US military was unsure about the prospects of success for the Saudi-led operation and about wheth­er it could be reconciled with the US’s interest in fight­ing AQAP. There were also doubts as to whether the Houthis were really being directed from Tehran. The US administration nevertheless declared its support for the intervention.153 This should first and foremost be seen as a signal to Riyadh that it aimed to continue the historically close partnership between the two coun­tries – at the time of the intervention, the US ad­ministration and the EU were negotiating the nu­clear agreement with the Iranian leadership, and the prospects of a rapprochement between the West and Tehran were viewed with concern in Riyadh.154 Under President Donald Trump, the US position since Janu­ary 2017 has moved closer to the Saudi position: the new US administration also sees the Houthis as proxies of Iran. In May 2018 the New York Times re­ported that US special forces were stationed along the Saudi-Yemeni border to pinpoint Houthi missile sites.155 Trump also reinforced the US commitment against AQAP by increasing drone deployments and ground operations.

The GCC Initiative and the UN’s Efforts

The UN Secretary-General has been involved in Yemen by providing “good offices” since the Moroc­can Jamal Benomar was appointed special envoy (April 2011 – April 2015). Benomar was succeeded by the Mauritanian Ismail Ould Cheich Ahmed (April 2015 – February 2018) and then the Briton Martin Griffiths (since February 2018). However, there is no explicit mandate from the Security Council or Gen­eral Council for the UN mission OSESGY in Yemen. The GCC Initiative continues to be the authoritative framework. UN mediation in Yemen can essentially be divided into two phases. The first phase dealt with supporting the political transition process sketched out by the GCC Initiative. The second phase, which started in April 2015, focused on the special envoy’s efforts to persuade the conflict parties to return to the political process.

The GCC Initiative and the Agreement on the Im­plementation Mechanism for the Transition Process in Yemen in Accordance with the Gulf Cooperation Council Initiative can be seen as a success for UN mediation. Benomar quickly became the main media­tor in talks, enjoying the backing of the European, US and Arab Gulf state embassies. The broad outline of the Gulf agreement – namely Saleh’s step-down and the formation of a unity government – had already been set out by the Gulf Cooperation Council and the Yemeni parties. But Benomar worked out the imple­men­tation mechanism with the negotiating parties and thus contributed substantially to the accord. In­ter­estingly, it was precisely the restraint shown by the Security Council members that allowed Benomar to present himself as an independent and trustworthy mediator in the conflict.156 The US hesitated to take an openly anti-Saleh stance for fear of losing a partner in the fight against AQAP; Russia shied away from clear demands for his resignation as a matter of principle in the context of the so-called Arab Spring.

Agreement was made possible by the fact that a balance of power between the regime forces and the opposition had emerged in the Yemeni capital in autumn 2011. However, the choice of an apparently weak consensus candidate – the then-vice president, Hadi, from the governing party of many years, the Gen­eral People’s Congress – created for both sides the prospect of shaping the transition process in their own interest. For Saleh it was obvious that approving the agreement would not mean renouncing power entirely and that his immunity would allow him to remain politically active. Nevertheless, the Security Council had to apply pressure before Saleh finally consented to sign the agreement. In its Resolution 2014 of October 2011, the Security Council emphasised the express support of the international com­mu­nity for the Gulf agreement. US and European threats of sanctions against Saleh and his family were also effective bargaining tools.

Once the agreement had been signed, the special envoy, backed by a united Security Council, actively accompanied its implementation. The implementation mechanism of the initiative rested on three pil­lars: transitional justice, security-sector reform, and a National Dialogue to set out the future structure of the state and settle the conflicts with the Houthis and the Southern Movement. Power-sharing and national dialogue had already been used or considered in solv­ing previous conflicts in Yemen.157 In other words, the UN approach built on principles which were already familiar in Yemeni politics. That the approach never­theless failed was mainly due to the formats’ lack of inclusiveness, an unrealistic timeframe, attempts at sabotage by the elites, and institutions pervaded by patronage networks.

Inclusiveness and Representativeness

The talks were not sufficiently inclusive from the outset and throughout the process. The Gulf states and UN negotiated with the leadership of the political parties with which the international community was already in contact. This made it easier to reach an agreement, but actors such as the Houthis, the South­ern Movement and the independent youth movement, who gained in relevance through the nationwide pro­tests in 2011, were excluded from the agree­ment.

These actors could only become involved in the political process under the National Dialogue, which took place between March 2013 and January 2014. The UN funded the talks, provided support for del­egates, motivated individual representatives to take part and took particular care to ensure high levels of participation among women. The moderate elite camp, which centred on President Hadi, dominated the process due to its international support and was thus able to co-decide the format and results of the dialogue to a large degree.158 The traditional parties were represented in greater numbers than new forces. However, figures with significant political and mili­tary influence, such as the Houthi leader Abdulmalik al‑Houthi and the Ahmar and Saleh families, did not take part in the conference. This allowed them to ab­dicate any responsibility for the success (or other­wise) of the National Dialogue;159 however, they were largely responsible for the escalating violence following the conference.

Although half the delegates were from the south, the Southern Movement did not feel represented at the forum. It demanded separate talks on Yemeni unity as a precondition for its participation in a dia­logue on pan-Yemeni issues. A few representatives of the Southern Movement agreed to attend, often after Benomar’s urging. Moreover, many of them were asso­ciated with Hadi, who hails from the south. Some of them were branded traitors for their willingness to engage in talks and were unable to return to the south after the conference; others were replaced short­ly before the conclusion of the dialogue by people close to Hadi.160 In other words, the forum was neither rep­resentative nor did it bring together the figures able to facilitate real political change in the country.

The consequences of inadequate inclusiveness were particularly clear in the Houthis’ conduct. Due to their exclusion from the transition government and the conception of the National Dialogue, they were able to shun all responsibility for the political process. In addition, there was no serious discussion either in the run-up to the National Dialogue or during the talks themselves about the role of armed nonstate actors – or even how to handle them. There were calls on the Houthis to found a political party, but since they only participated reluctantly in the con­ference, they were never required, for instance, to put down their weapons as a precondition. By con­trast, AQAP as a nonstate violent actor and enemy of the Western supporters of the dialogue was explicitly excluded. Although the Houthis were able to assert many of their interests, they ultimately rejected the way in which Yemen was supposed to be divided into six federal regions. They used their resistance to this key conference outcome as a pretext for military expan­sion and demands for a new, more inclusive gov­ernment.

The Peace and National Partnership Agreement that was subsequently negotiated by Benomar and signed by the Hadi government and the Houthis on 22 September 2014 could be seen as an extension of the GCC Initiative. It was signed after the Houthis had captured the capital by force. The Agreement aimed to put an end to the Houthis’ escalation of violence through a new power-sharing agreement. Accordingly, it stipulated a new, more inclusive government with Houthi and Southern Movement participation, along with reforms and the implementation of the National Dialogue outcomes.161 However, the Houthi’s non-compliance with the agreement was foreseeable. Unexpected empowerment and Saleh’s support en­couraged the Houthis to continue expanding despite having signed the agreement, creating new facts on the ground.

An Unrealistic Time Frame

Following the ten-month-long Dialogue, the 565 del­egates submitted a catalogue of outcomes in January 2014, which offered a good basis for a constitution with its provisions on issues such as civil rights, edu­cation and health. The international community celebrated the conclusion of the talks as a model for a UN-led transition process. However, the formal closure of the dialogue merely diverted attention from the fact that the actual conflicts had not been resolved. Ultimately the conference foundered on the southern issue.162 The Southern Movement del­egates at the dialogue were not representative figures; however, no solution could be found for the future handling of the south even with those who attended. A key outcome of the dialogue was the subdivision of Yemen into federal regions to make the distribution of state resources more uniform through decentralisa­tion. However, conference attendees were unable to agree on the number or shape of the federal regions. The northern-Yemeni elite insisted on a six-region solution because the division into two of the resource-rich south complicates the latter’s quest for independ­ence. Southern-Yemeni representatives demanded a five-state solution and sought to establish one federal region within the borders of the formerly independ­ent South Yemen.

By December 2013 the envisaged time frame for the dialogue of six months had already been exceeded by three months. Outside the talk forums, tensions were rising between the political actors: several mod­erate delegates were murdered, and violent conflicts grew on the periphery between the Islah Party and the Houthis. Along with Benomar, Hadi persuaded the delegates to sign a document agreeing to a “fairer solution for the south”. The decision to divide the country into six regions was ultimately confirmed by a committee outside of the dialogue, which had been appointed by Hadi for that very purpose.163 This cir­cum­vented the National Dialogue with the aim of expediting closure – at the cost of a sustainable con­sensus among the delegates.

Neither Hadi nor the UN mission nor the international sponsors of the GCC Initiative acknowledged that the Dialogue had missed its objectives.

It was evidently foreseeable that the Southern Move­ment would reject this solution. The Houthis complained about the unfairness of the decision-making process. They were also dissatisfied with the region that they had been allocated, since they sought access to the sea. Despite this opposition and the grow­ing protests in the south, the transition government – backed by the UN – attempted to continue the political process on the basis of the Dialogue out­comes.164 President Hadi’s term in office, which ended in February 2014, was extended for a further two years by the National Dialogue Conference without elections. His legitimacy was subsequently questioned throughout the country, with Houthi and Saleh sup­porters in particular demanding a new government ever more emphatically. Armed conflicts also esca­lated all over Yemen immediately after the conclusion of the Dialogue, which therefore merely delayed – but did not prevent – the outbreak of violence. Nei­ther Hadi nor the UN mission nor the international sponsors of the GCC Initiative acknowledged that the Dialogue had missed its objectives. This has shaped both their engagement with conflict actors and the negotiation formats to this day.

Resistance to Reform

The transition government, formed in December 2011, mainly consisted of Saleh’s GPC and the Islah Party. Since the main responsibility for implementing the GCC Initiative lay with President Hadi and re­forms were not going to be discussed until the Nation­al Dialogue, the transition government barely tried to improve the population’s living standards through reform efforts of its own. Instead, its members used their positions to cultivate patronage networks and expand their own power vis-à-vis rivals’.

The reform of the security sector launched by the GCC Initiative was meant to create a national mili­tary loyal to the state, and not to individual figures. Although members of the Saleh family were pushed out of their official military positions through the threat of sanctions against spoilers contained in Secu­rity Council Resolution 2051 of June 2012, they con­tinued to exert influence via informal networks. It was almost impossible to institutionalise and pro­fes­sion­alise the security sector, given that it was built on patronage networks. Ultimately, only a rearrange­ment of patronage networks within the institution was achieved.165 This enabled Hadi to gain influence within the armed forces, but not over the entire secu­rity sector.

As a consequence, the Houthis with Saleh’s support managed to capture the capital in autumn 2015. Hadi’s control over the military was too limited to deploy it against the Houthis and protect himself against the coup d’état. With the Houthis’ accession to power, the political transition process and the paci­fication of conflicts through power-sharing had failed under both the GCC Initiative and the Peace and National Partnership Agreement.

Resolution 2216: Partial UN Mediation

In the second phase of UN mediation, since April 2015, the Secretary-General tasked his special envoy with mediating between the conflict parties to induce them to return to the political process within the frame­work of the GCC Initiative, the outcomes of the National Dialogue and the relevant UN Resolutions.166 By this time, the UN mission had already lost most of its standing among Yemenis. Jamal Benomar was publicly held responsible in Yemen for the failure of the political process since his intense commitment had made him the face of international support for the GCC Initiative.167 Benomar eventually stepped down as special envoy in April 2015, after Yemeni criticism of his alleged partiality to the Houthis became increasingly fierce and even the Gulf states turned their backs on him. He was replaced by Ismail Ould Cheikh Ahmed.168 Talks between the Hadi gov­ern­ment and representatives of the Houthi-Saleh alli­ance, mediated by the new special envoy, took place in June and December 2015 in Geneva and between April and August 2016 in Kuwait. They led to an exchange of prisoners and several truces. However, although different draft agreements were discussed, none was concluded. The Hadi government was loath to agree to the formation of a new government before the Houthis had negotiated their withdrawal and dis­armament. The Houthis, however, saw a new gov­ern­ment as the precondition for negotiating the with­drawal and disarming.169 With the new special envoy, there was some hope within the international com­munity, that the peace talks might gain new momen­tum. In contrast to his predecessor, Griffiths has approached all conflict parties and has been able to secure assurances from all parties that they are ready to solve the conflict.170 The peace plan he submitted in June 2018 provides for a ceasefire as well as the handover of the Houthis’ medium and heavy wea­pon­ry, including long-distance missiles. In return, the coa­lition is to cease its air attacks. A unity government is to be formed thereafter.171 After a two-year ebb of UN engagement, Martin Griffiths issued invita­tions to a new round of talks in Geneva to be held in the first week of September 2018. While both the Houthis and the internationally recognised government were supposed to be present at the talks, they had no intention of speaking directly to each other, but of staying in separate rooms, with Griffiths mediat­ing between the two sides to define the frame­work for the talks. Unfortunately, it did not come even this far, as the Houthis did not travel to Geneva after their demands had not been met (i.e. that the injured and sick be transferred from Sana’a to Oman and that the mediators travel to Geneva with an Omani plane). Griffiths responded to the failure by initiating shuttle diplomacy between Riyadh, Muscat and Sana’a.172 However, along with the UN’s failure to create the conditions for the Houthis to travel to Geneva, a number of factors contribute to the failure of UN talks in the Yemen conflict. First, the events since mid-2014 have demonstrated the limits of ex­ternal influence, particularly on the Houthis. Con­nect­ed to this is the circumstance that the Yemeni negotiating parties may not take the UN talks very seriously.173 They sent no representatives with real influence over military actions. Consequently, the truces were short-lived. Second, the UN approach based on UN Resolution 2216 of April 2015 rests on false assumptions.

The Limits of External Influence

When the Security Council stood by Hadi and de­mand­ed the unconditional withdrawal of the Houthis, it took a stance that contradicted the facts on the ground. The Houthi coup d’état and the power-shar­ing arrangement of the Peace and National Part­ner­ship Agreement had changed the balance of power and shown the Houthi-Saleh alliance that it would be better off by not acceding to the rules of the GCC Initiative. The Houthis now had access to state insti­tutions, including the military and the state arsenal. Simultaneously, through his alliance with the Houthis, Saleh attained his objective of disempowering Presi­dent Hadi and the Islah Party. The narrative promoted by the Houthi-Saleh alliance – that the coalition inter­vention was an attack on the Yemeni nation-state – resonated widely in northern Yemen.

Meanwhile it had become clear that the Security Council had no means at its disposal to change this balance of power. During the reforms of the military in 2012, the threat of sanctions against Saleh had been effective. However, the strong increase of pres­sure through the creation of a sanctions committee in February 2014 already had a less positive impact. The looming sanctions neither prevented Saleh from enter­ing into an alliance with the Houthis nor the Houthis from taking a stance against Hadi and bank­ing on military expansion. That the UN had little real influence on the conduct of the Houthis became clear in summer 2014 at the latest, when Benomar repeat­edly travelled to Saada to persuade the Houthis to take part in the political process. An agreement – which the Houthis did not comply with – was not reached until the Houthis had captured Sana’a and thus achieved unprecedented levels of power.

In November 2014 the Security Council eventually imposed sanctions against Saleh and two Houthi com­manders. For the Houthis, the financial sanctions and travel ban were trivial since they neither travelled abroad nor depended on foreign bank accounts.174 The Saleh family, by contrast, was impacted, since the former president had invested his fortune amassed over the years abroad and had tried since 2011 to hide those riches via transfers to his sons and via an inter­national financial network.175 However, the sanctions did not make Saleh relent. He tried to negotiate for them to be lifted, but simultaneously expanded his alliance with the Houthis.176 The Security Council had thus forfeited its only leverage against Saleh.

False Assumptions

A further fundamental problem of the UN approach to the talks is that it is based on the parameters of Resolution 2216. In UN Resolution 2216 of April 2015, initiated by the Gulf Cooperation Council, the Security Council urged the Yemeni conflict parties to return to the framework of the political process, assuring both Hadi and the Gulf Cooperation Council of its support. It called on the Houthis to withdraw immediately and unconditionally from government and security institutions and the capital, and refrain from actions that belonged to the “legitimate gov­ern­ment” alone. Moreover, the Security Council imposed sanctions against Ahmed Ali Saleh and the Houthi leader Abdulmalik al-Houthi, as well as an arms em­bargo against three Houthi commanders, President Saleh and his son. The arms embargo was used by Saudi Arabia to legitimise its sea, air and land blockade and (after the fact) its intervention.177

What renders this problematic for the UN talks, is that it provides a framework resting on the assumption that the political process could be resumed at the point where it had been interrupted by the Houthi takeover of power. The UN mission still considered the National Dialogue a success and limited his efforts to pacifying the conflict between the Houthis and the Hadi government so as to be able to continue the political process. The negotiating parties thus con­sisted only of the Hadi government on one side and representatives of the Houthis and the GPC on the other. However, the UN approach failed to recognise that President Hadi enjoyed barely any legitimacy in Yemen even at the outset of the military intervention and had little influence over the actors there. It also ignored the fact that the Southern Movement rejected the National Dialogue entirely, did not consider itself represented by the Hadi government and did not seek to be part of a pan-Yemeni solution.178 The Southern Movement was unable to make itself heard at this point in time and was excluded from the talks along with other forces that had recently emerged or been strengthened in the context of the military intervention, such as militia or tribal leaders.

The binary talk format enabled Saudi Arabia to con­tinue to portray itself as a mediator rather than a conflict party, by throwing its weight behind Hadi as the only legitimate president. The latter thus re­mained an important element of a political solution despite his lack of influence. Simultaneously, Hadi’s privileged position in exile in Riyadh now rested on continuing the war.179 Following the talks in Kuwait, the special envoy and international diplomats for the first time considered proposing the disempowerment of Hadi as part of an agreement. The plan submitted by the UN special envoy in October 2016, for instance, mentions this.180 However, the West continues to stand by Hadi, as does Saudi Arabia, since his role is crucial for the Riyadh-led intervention. Hadi is there­fore in a position to keep blocking a political solution and rejecting proposals that would lead to his dis­empowerment.

For the West, good relations with Saudi Arabia are more important than Hadi.

For the West, good relations with Saudi Arabia are more important than Hadi. This predominant interest results in a biased attitude by the UN Security Council and other UN institutions, an imbalance that has been aggravated by the influence exerted on Security Coun­cil resolutions by non-permanent members Jordan (2014–15), Egypt (2016–17) and Kuwait (2018–19), which are also part of the Arab coalition, and by the Gulf Cooperation Council itself.181 Saudi Arabia has used its influence to prevent an investigation into human-rights violations in the Yemen conflict; gain a seat on the UN Human Rights Council; and threaten the UN with cancelling its financial support if it was not removed from the Black List, to which it was added for its responsibility for the killing of children in Yemen.182 Not least for these reasons, the Houthis considered the UN mission to be partial; since June 2017, they have refused even to meet the special envoy Ismail Ould Cheich, who stepped down in February 2018.

The new special envoy, Martin Griffiths, was able to resume proceedings with the Houthis and is in regular contact with the group’s leaders. Griffiths announced at the start of his term in office, however, that he would continue to use Resolution 2216 and the GCC Initiative as the framework for talks. This will dramatically limit his room for manoeuvre, including for integrating relevant conflict actors.183 Nevertheless, Griffiths has begun dialogue with the Southern Movement represented by the STC, which promises a more inclusive approach. While Griffiths vows to continue dialogue, it is as yet unclear wheth­er and how the Southern Movement will be integrated into the peace process. Although it is a requirement for a comprehensive solution, it is debatable whether all conflict parties can be integrated into negotiations on a new government, with both the internationally recognised government and the Houthis vying to dominate the peace process.


The failure of the UN-accompanied transitional pro­cess and the subsequent escalation of violence are largely due to the interests of the Yemeni conflict actors. Each faction of the elite believed that the sus­tained support it received from domestic and inter­national allies meant it could sustain its own position of power. That the elites, and above all former presi­dent Saleh, were even able to sabotage the political process was due to Saleh’s immunity and the exclu­sion of important actors by the GCC Initiative. Inter­national support for Hadi as president of the transi­tion government strengthened a faction of the elite that would have been incapable of asserting itself against local actors without external assistance. The Houthis and Southern Movement, by contrast, were excluded from the negotiations on the Gulf agreement and thus could not contribute to shaping the transition process. Both therefore blocked the process and created new facts: the Houthis used force to expand in the north, while the Southern Movement began to eject representatives from northern Yemen from the south. The omissions of the GCC Initiative ultimately led to the failure of the political process and paved the way for a war that is increasingly being used by regional and international powers to further their own interests.

The negotiating format of the UN special envoy does not do justice to the complexity of the conflict.

The negotiating format of the UN special envoy does not do justice to the complexity of the conflict since it concentrates on two conflict parties, while leaving out other influential actors as well as the regional conflict parties. It is also based on the false assumption that a return to the political framework set out by the GCC Initiative will lead to a pacification of the conflict. The actors who currently decide con­ditions in Yemen are not interested in this political process. Moreover, external actors do not have suffi­cient influence to bring about an agreement in this framework, let alone compel the conflict parties to comply with it.

Although both parties reportedly showed themselves willing vis-à-vis the special envoy to negotiate, neither appears to be willing to compromise. The Houthis are unwilling to participate in a process in which they stand only to lose. The new escalation of violence along the Red Sea after the failure of talks in September 2018 shows that the Saudi coalition is trying to force the Houthis into a weaker position to enable negotiations. The attempts of UN Special Envoy Martin Griffiths to bring the parties to the con­flict back to the negotiation table and his inclusive approach are commendable. Whilst Griffiths has already let it be known that he will adhere to Resolu­tion 2216 and the GCC Initiative, he should nevertheless interpret these documents in a way that affords him more room for manoeuvre. Resolution 2216 should be seen as a goal rather than a path. The new special envoy should define interim goals that make a long-term implementation of the resolution possible. Thus, agreements could be concluded in an inclusive negotiating setting between local actors, before the “big” conflict between the Houthis and the Saudi coalitions is resolved. Thus, the Southern Movement and representatives of other regions – for example, the Hadhramout tribal conference or militia leaders from Taiz – could be integrated into a multiforum process along with the Hadi government, the GPC and the Houthis. First and foremost, the talks format should reflect the balance of power in situ and the interests of the individual actors. A flexible negotiation format could enable the special envoy to inte­grate the interests of both local and regional actors in different rounds of talks. This will require winning over the UAE and Iran along with Saudi Arabia to a comprehensive peace process.

The UN special envoy should also evaluate to what extent a resumption of the GCC Initiative process is even possible. It would be advisable to concentrate on the outcomes of the National Dialogue on which con­sensus can be reached, leaving out those that have contributed to the escalation of the conflict. Specifi­cally, this concerns the division of the country into six regions. Instead of insisting on this point, the spe­cial envoy should attempt to discuss, within an inclu­sive talks format, territorial power-sharing based on the actual balance of power in Yemen. The basis could be a federal system, as the National Dialogue stipulated. However, the number and delimitation of the federal regions would have to be reviewed in a transparent and inclusive process. Competences in government and the security sector could then be allocated on this basis, and the future of nonstate violent actors, including the military wing of the Houthis, be negotiated. The biased role of the UN Security Council will complicate Griffith’s efforts to convince the conflict parties of his impartiality. As long as he behaves accordingly when dealing with both the conflict parties and Security Council mem­bers, he could nevertheless succeed.

Europe should seek a neutral approach to the Yemen conflict.

Europe should seek an impartial approach to the Yemen conflict, even if that is complicated by bilat­eral relations with Riyadh. The EU declaration of 26 June 2018, calling on all conflict parties to prevent an escalation of the violence in Hodeidah, was a positive step. But Europe must become more active in pre­venting further deterioration of the world’s worst humanitarian crisis. EU member states must re-assess their direct and indirect involvement in the Yemen war. Rather than continuing to fan the flames of the war by selling arms to the parties of the conflict, they should back the EU parliament’s call for an arms em­bargo against Saudi Arabia and put a stop to weapons deals with countries involved in the war.184 Simultaneously, Europe should use the talks on Yemen be­tween the EU, Germany, France, the UK, Italy (EU/E4) and Iran, which began in February 2018, to discuss what solutions could lead to the Houthis exercising moderation.185 The EU should use its influence and diplomatic tools to assist the UN special envoy in finding the urgently needed political solution to this devastating war. In the meantime, the EU must con­tinue to support local organisations in providing humanitarian assistance and security. The disastrous humanitarian situation is already degenerating dras­tically towards a point of no return, with millions now at risk of dying of starvation.186

In the three cases studied here – Libya, Syria and Yemen – the UN has tried to mediate power-sharing agreements to pacify (Libya, Syria) or avert (Yemen) civil wars that had broken out or were looming fol­lowing Arab Spring protests. However, in none of these has the UN special envoy achieved a negotiated resolution to the conflict or even effective conflict management, or contributed significantly to protect­ing civilians. Although the UN envoys in Yemen and Libya were able to negotiate power-sharing agreements, these did not overcome the political divide in Libya or lastingly end armed confrontations in Yemen. In Syria, no negotiations on power-sharing even took place, merely separate talks between the UN mediator and the regime and opposition delega­tions.

All three cases have clear parallels regarding con­flict dynamics and the international context, which should be considered the main causes of the lack of success in mediation efforts. First, in none of the three situations was a stalemate possible, whereby the local conflict parties would have preferred a negotiated solu­tion to continued armed confrontations, thus making the conflict “ripe for resolution”. Instead, the local conflict parties were able to rely on external support, albeit to varying degrees, which strengthen­ed their trust in a military, rather than political, solution to the conflict. External backers thus fuelled the con­flicts even when – as in the case of the US support for rebels in Syria – their intention was to enable negotiations. Moreover, with time the conflicts be­came increasingly internationalised, and third parties engaged in ever more energetic military interventions on the side of one or several conflict parties to pursue their own respective objectives. The main conflict parties also changed during the conflicts: new actors featured, old actors lost relevance; alliances were formed and splintered. As a result, the original parties in the civil war were not interested in the talks offer­ed by the UN to solve the conflict. Nor were negotiations between them any longer a meaningful way of ad­dress­ing the increasingly complex conflicts. In addi­tion, the UN missions were not mandated to balance the interests of external actors.

Second, coercive measures that would have put pres­sure on the conflict parties to agree to a negotiated settle­ment were only used to a very limited extent – or not at all, as in Syria due to the Russian veto in the Security Council. Where such measures were decided by the UN Security Council, they were not consistently applied. Responsibility for this lies with the inter­ests of external actors. Arms embargoes, for example, were undermined by Egypt and the UAE in Libya, and only imposed on one conflict party in Yemen, namely the Houthi-Saleh alliance. UN sanctions targeting those who tried to undermine power-sharing agree­ments remained a mere threat in Libya and were only imposed on one Yemeni conflict party, the Houthi-Saleh alliance. In any case, their impact was negli­gible where spoilers, such as the Houthi rebels in Yemen, neither travelled abroad nor invested assets abroad. The sanctions imposed by the USA and EU against conflict parties – for instance representatives of the Syrian regime – had an extremely limited effect and were also repeatedly circumvented by EU member states. Deploying peace-keeping troops to accompany and secure the implementation of an agreement on conflict resolution was not a realistic option in any of the three cases – both because of dis­agreements within the Security Council and be­cause of the resistance by local conflict parties.

Furthermore, the approach of the UN mediators in Yemen and Libya contributed to the power-sharing agreements being short-lived and not resulting in pacification. They put excessive time pressure on the negotiation of agreements, which led to the conflict parties not assuming any personal commitment for their implementation and left significant issues unresolved. Important conflict actors were left out (because of vetoes by regional or major powers or controversially being defined as terrorists). The UN mediators backed unity governments even when the latter turned into parties in the conflict, thus for­feiting the special envoys’ impartiality. In Syria issues of representativeness and inclusiveness moved into the background since no negotiations on power-sharing took place – or are likely to take place in the near future. In all three cases, the mediators focused on internationally legitimised objectives, which had been rendered unrealistic or obsolete by the changed balance of power. In Syria, these objectives were power-sharing and a political transition despite the fact that the rebels had lost the civil war and been pushed back to a few enclaves. In Libya, the UN spe­cial envoy adhered for far too long to the negotiation framework based on the two parliaments, despite them no longer representing the increasingly frag­mented actor landscape. And in Yemen, the UN mission stubbornly insisted on the outcomes of the National Dialogue Conference as the basis of negotia­tions and on Hadi as “legitimate” president although the latter had barely any support left in the country. The approaches pursued by the UN mediators thus offered fewer and fewer chances of bringing about a negotiated resolution to the conflict.

On one issue the three cases diverge. In Syria alone an external conflict actor has been active since the autumn of 2015 who is not only militarily dominant but also has the power of veto in the Security Coun­cil: Russia. The UN mediation therefore cannot cir­cumvent Russia. This has presented the UN mission with a dilemma: it cannot bring about pacification without or against Russia, but any cooperation with Russia grants legitimacy to an approach that runs counter to a resolution to the conflict negotiated between the Syrian conflict parties according to Secu­rity Council Resolution 2254, and violates fun­da­men­tal tenets of international law, for example through the bombing of civilian targets. In the case of Yemen, it is Saudi Arabia that is militarily and politically domi­nant. The Security Council members have allow­ed the Kingdom to largely dictate the rules of engage­ment in the conflict and a negotiation mandate opposed to a negotiated solution that integrates the interests of the Houthis and other local actors.

Clearly, the conclusion from UN mediations in Yemen, Libya and Syria that have seen little success cannot be that mediation should not even be at­tempted in similarly complex future conflicts. After all, the world community has a duty to act to re-estab­lish peace and security and protect human lives. The issue is therefore how to make such mediation efforts more effective even where the circumstances com­plicate a successful resolution to a conflict.

Three conclusions can be drawn from the cases studied here. First, the mandate should not be limited to mediation between local parties in the civil war, but also from the outset provide forums that would allow for the reconciliation of competing interests of all relevant regional or major powers, or at least enable the UN to influence the rules of engagement in the conflict. As a point of principle, the UN should not permit the exclusion of actors based on other actors’ vetoes. Second, it must ensure that power-shar­ing agreements are inclusive, that the representatives involved in the negotiations are truly representative, and that adhering to agreements is incentivised. All these elements should take precedence over a quick conclusion. Third, UN mediators must avoid taking sides with unity governments generated by power-sharing agreements when these turn into parties in the conflict, circumvent agreements and fuel the conflict rather than pacify it.

The EU and its member states have little influence over the regional and global configurations that con­strain UN mediators. Nevertheless, two general rec­om­mendations can be made. First, Europeans should refrain from anything that undermines UN mediation efforts, be it through circumventing unilateral or multilateral sanctions or through direct cooperation with conflict actors not aimed at a negotiated resolu­tion to the conflict. This is the case, for example, with the upgrading of Libyan militia to control sea borders to prevent unlawful migration or with arms deliveries to Saudi Arabia or Turkey that are not effectively con­ditioned. Rather, the EU and its member states should use their influence to support a negotiated resolution to conflicts. This includes making deliberate use of any potential influence (through tightening or loosen­ing sanctions, or granting reconstruction assistance or recognition), preferably within the framework of broad international coalitions. Second, where UN mis­sions are not mandated as stipulated to reconcile the interests of all relevant regional and major powers, Europeans should make use of their own diplomatic contacts so as to contribute to regional and inter­national arrangements that prevent further escalation of the war, establish rules of engagement, uphold inter­national humanitarian law, and are centred on protecting civilians.



Al-Qaeda in the Arabian Peninsula


Center for Applied Research in Partnership with the Orient (Bonn)


European External Action Service


European Union


Euro-Mediterranean Studies Commission


Gulf Cooperation Council


General National Congress (Libya)


General People’s Congress (Yemen)


House of Representatives (Libya)


Hessische Stiftung Friedens- und Konflikt­forschung (Hessian Foundation for Peace and Conflict Research, Frankfurt am Main)


Hay’at Tahrir al-Sham (Levant Liberation Committee, Syria)


International Crisis Group


International Institute for Strategic Studies


“Islamic State”, also ISIS, ISIL


International Syria Support Group


Joint Comprehensive Plan of Action


League of Arab States


Libyan International Assistance Mission


Middle East Research and Information Project


National Dialogue Conference (Yemen)


Office of the Special Envoy of the Secretary-General for Syria


Office of the Special Envoy of the Secretary-General for Yemen


Partiya Karkerên Kurdistan
(Kurdistan Workers’ Party)


Project on Middle East Political Science


Partiya Yekîtiya Demokrat
(Democratic Union Party)


(UN) Security Council


Syrian Democratic Forces


Syrian Interim Government


Syrian National Council


Syrian Negotiation Committee


Transitional Governing Body (Syria)


United Arab Emirates


United Nations


United Nations General Assembly


United Nations High Commissioner for Refugees


United Nations Support Mission in Libya


UN Supervision Mission in Syria


United States Institute of Peace (Washington, D.C.)



 See also the overview of the military involvement of third parties in the three conflicts, pp. 11ff.


 I. William Zartman, Ripe for Resolution (New York, 1985).


 Lise Howard and Alexandra Stark, “How Civil Wars End. The International System, Norms, and the Role of External Actors”, International Security 42, no. 3 (2017/18): 127–71.


 Virginia Fortna, Peace Time. Cease-Fire Agreements and the Durability of Peace (Princeton, 2004).


 Sebastian von Einsiedel, Civil War Trends and the Changing Nature of Armed Conflict, Occasional Paper 10 (Tokyo: United Nations University Centre for Policy Research, March 2017), https://collections.unu.edu/eserv/UNU:6156/Civil_war_trends_
(accessed 16 May 2018).


 Kathleen Gallagher Cunningham, Understanding Frag­men­tation in Conflict and Its Impact on Prospects for Peace, Oslo Forum Paper 6 (Geneva: Centre for Humanitarian Dialogue, Decem­ber 2016), https://www.hdcentre.org/wp-content/uploads/ 2017/02/Understanding-fragmentation-in-conflict.pdf (accessed 16 May 2018).


 Kyle Beardsley, “Using the Right Tool for the Job: Media­tor Leverage and Conflict Resolution”, Penn State Journal of Law & International Affairs 2, no. 1 (2013): 57–65; Lindsay Reid, “Finding a Peace That Lasts: Mediator Leverage and the Du­rable Resolution of Civil Wars”, Journal of Conflict Resolution 61, no. 7 (2017): 1401–31.


 William Zartman, “Ripeness: The Hurting Stalemate and Beyond”, in International Conflict Resolution after the Cold War, ed. Paul Stern and Daniel Druckman (Washington, D.C., 2000), 225–50; Allard Duursma, “A Current Literature Review of International Mediation”, International Journal of Conflict Management 25, no. 1 (2014): 81–98; Nita Yawana­rajah, “Ripe, Ready or Strategic? The Timing of Peace Initia­tives”, Oxford Research Group, October 2016, http://www. oxfordresearchgroup.org.uk/publications/articles_multimedia/ripe_ready_or_strategic_timing_peace_initiatives (accessed 16 May 2018).


 Véronique Dudouet, Mediating Peace with Proscribed Armed Groups, Special Report 239 (Washington, D.C.: United States Institute of Peace [USIP], May 2010), https://www.usip.org/ sites/default/files/SR239Duduoet.pdf (accessed 16 May 2018); Howard and Stark, “How Civil Wars End” (see note 3); Barbara Walter, “New Civil Wars”, Annual Review of Political Science 20 (2017): 469–86.


 Fen Hampson, “Can the UN Still Mediate?”, in The United Nations and Global Security, ed. Richard Price and Mark Zacher (London, 2004), 75–92; Kyle Beardsley, “The UN at the Peace­making-Peacebuilding Nexus”, Conflict Management and Peace Science 30, no. 4 (2013): 369–86.


 Center for Humanitarian Dialogue, Charting the Roads to Peace. Facts, Figures and Trends in Conflict Resolution, Mediation Data Trends Report 2007 (Geneva, 2007), http://www. conflictrecovery.org/bin/oslo_charting_FINAL.pdf (accessed 12 March 2018).


 Caroline Hartzell and Matthew Hoddie, “Institution­alizing Peace: Power Sharing and Post-Civil War Conflict Management”, in: American Journal of Political Science 47, no. 2 (2003): 318–32; Karl DeRouen, Jr., Jenna Lea and Peter Wallensteen, “The Duration of Civil War Peace Agreements”, Conflict Management and Peace Science 26, no. 4 (2009): 367–87; Thorsten Gromes, Machtteilung nach Bürgerkriegen: Verbreitung und Erfolgsbilanz seit 1990, Report 3 (Frankfurt a.M.: Hessische Stiftung Friedens- und Konfliktforschung [HFSK], 2015), https://www.hsfk.de/fileadmin/HSFK/hsfk_downloads/
(accessed 16 May 2018).


 Denis Tull and Andreas Mehler, “The Hidden Costs of Power-Sharing: Reproducing Insurgent Violence in Africa”, African Affairs 104, no. 416 (2005): 375–98.


 Alex De Waal, “Mission without End? Peacekeeping in the African Political Marketplace”, International Affairs 85, no. 1 (2009): 99–113; Beardsley, “The UN at the Peace­making-Peacebuilding Nexus” (see note 10).


 Andreas Mehler, “Peace and Power Sharing in Africa: A Not So Obvious Relationship”, African Affairs 108, no. 432 (2009): 453–73; Chandra Sriram and Marie-Joëlle Zahar, “The Perils of Power-Sharing: Africa and Beyond”, Africa Spectrum 44, no. 3 (2009): 11–39.


 David E. Cunningham, “Who Should Be at the Table? Veto Players and Peace Processes in Civil War”, Penn State Journal of Law & International Affairs 2, no. 1 (2013): 38–47.


 Haftar and his officials also – but less frequently – use the term ‘Libyan National Army’.


 See Wolfram Lacher and Peter Cole, Politics by Other Means: Conflicting Interests in Libya’s Security Sector, Working Paper 20 (Geneva: Small Arms Survey, October 2014), http://www. smallarmssurvey.org/fileadmin/docs/F-Working-papers/SAS-SANA-WP20-Libya-Security-Sector.pdf (accessed 8 March 2018).


 Frederic Wehrey, “Whoever Controls Benghazi Controls Libya”, The Atlantic, 1 July 2017; Aidan Lewis, “Veteran Com­mander Vies for Power in Libya’s Shifting Sands”, Reuters, 26 February 2018.


Frederic Wehrey and Wolfram Lacher, “Libya’s Legitimacy Crisis”, Foreign Affairs, 6 October 2014, https://www.foreign affairs.com/articles/middle-east/2014-10-06/libyas-legitimacy-crisis (accessed 22 May 2018).


Virginie Collombier, “Dialogue, Mediation and Reconcilia­tion in Libya’s Local Conflicts”, in Inside Wars. Local Dynamics of Conflicts in Syria and Libya, ed. Luigi Narbone, Virginie Col­lombier and Agnès Favier (Florence, 2016), 30–37.


 David Kirkpatrick, Ben Hubbard and Eric Schmitt, “ISIS’ Grip on Libyan City Gives It a Fallback Option”, New York Times, 28 November 2015.


 International Crisis Group (ICG), The Libyan Political Agree­ment: Time for a Reset, Middle East and North Africa Report no. 170 (Brussels, 4 November 2016), https://d2071andvip0wj. cloudfront.net/170-the-libyan-political-agreement.pdf (ac­cessed 8 March 2018).


 Frederic Wehrey and Wolfram Lacher, “Libya after ISIS”, Foreign Affairs, 22 February 2017, https://www.foreignaffairs. com/articles/libya/2017-02-22/libya-after-isis (accessed 22 May 2018).


 ICG, The Libyan Political Agreement (see note 7).


 Wolfram Lacher, Tripoli’s Militia Cartel. How Ill-Conceived Stabilisation Blocks Political Progress, and Risks Renewed War, SWP Comment 20/2018 (Berlin: Stiftung Wissenschaft und Politik, April 2018).


 ICG, The Libyan Political Agreement (see note 7).


 “General Khalifa Haftar: My Goal Is to Cleanse Libya of Muslim Brothers and Oppressors” (in Arabic), Al‑Sharq al‑Awsat, 20 May 2014; “Haftar: No Leniency in Eradicating Muslim Brotherhood” (in Arabic), Al‑Ahram, 7 July 2014.


 UN Security Council, Final Report of the Panel of Experts in Accordance with Paragraph 24 (d) of Resolution 2213 (2015), S/2016/209, 4 March 2016.


 David Kirkpatrick and Eric Schmitt, “Arab Nations Strike in Libya, Surprising U.S.”, New York Times, 25 August 2014.


 UN Security Council, Final Report of the Panel of Experts in Accordance with Paragraph 13 of Resolution 2278 (2016), S/2017/ 466, 1 June 2017.


 In addition, an unidentified aircraft – most likely a CIA drone – carried out at least one targeted strike in Ubari in 2018. Joe Penney, Eric Schmitt, Rukmini Callimachi and Christoph Koettl, “C.I.A. Drone Mission, Curtailed by Obama, Is Expanded in Africa Under Trump”, New York Times, 9 Sep­tember 2018, https://www.nytimes.com/2018/09/09/world/ africa/cia-drones-africa-military.html (accessed 27 September 2018); “Unidentified Aircraft Bombs Libya’s Derna, 3 Dead – Witness”, Reuters, 7 February 2016; UN Security Council, Final Report of the Panel of Experts (see note 15), 35; “Libya: 16 Civil­ians Killed in Derna Airstrikes”, Human Rights Watch, 5 No­vember 2017, https://www.hrw.org/news/2017/11/05/libya-16-civilians-killed-derna-airstrikes (accessed 8 March 2018).


 Cyril Bensimon, Frédéric Bobin and Madjid Zerrouky, “Trois membres de la DGSE tués en Libye, le gouvernement libyen proteste”, Le Monde, 20 July 2016.


 Missy Ryan and Sudarsan Raghavan, “U.S. Special Operations Troops Aiding Libyan Forces in Major Battle against Islamic State”, Washington Post, 9 August 2016.


 Author’s interviews with officers and leaders of armed groups involved in the offensive, Misrata, September 2016.


 Author’s interviews with political representatives, Bani Walid, September 2016.


 United Nations High Commissioner for Refugees (UNHCR), Italy Sea Arrivals Dashboard, January – December 2017 (Geneva, 11 January 2018, https://data2.unhcr.org/en/ documents/download/61547 (accessed 8 March 2018).


 Daniel Howden, “The Central Mediterranean: European Priorities, Libyan Realities”, Refugees Deeply, October 2017, http://issues.newsdeeply.com/central-mediterranean-european-priorities-libyan-realities (accessed 8 March 2018).


 Amnesty International, Libya’s Dark Web of Collusion, December 2017, https://www.amnesty.org/download/ Documents/MDE1975612017ENGLISH.PDF (accessed 8 March 2018); Mark Micallef and Tuesday Reitano, The Anti-Human Smuggling Business and Libya’s Political End Game, North Africa Report 2 (Institute for Security Studies/The Global Initiative against Transnational Organized Crime, December 2017), https://issafrica.s3.amazonaws.com/site/uploads/nar2.pdf (accessed 22 May 2018).


 Yuri Barmin, Russia in Libya: From Authoritarian Stability to Consensus Settlement, Alsharq Forum Expert Brief (August 2017), http://www.sharqforum.org/wp-content/uploads/dlm_ uploads/2017/08/Russia-in-Lybia-01.pdf (accessed 8 March 2018).


 Marc Semo, “Emmanuel Macron parraine un accord en Libye, sans garantie de résultat”, Le Monde, 25 July 2017.


 Author’s interviews with Western diplomats, May 2018; Frédéric Bobin and Marc Semo, “A Paris, les Libyens s’accor­dent sur des élections en décembre”, Le Monde, 30 May 2018.


 “Italy’s PM Conte Sees No Rush for Libyan Election”, Reuters, 8 August 2018.


 “Briefing and Consultations on Developments in the Political Dialogue in Libya”, What’s in Blue, 25 August 2015, http://www.whatsinblue.org/2015/08/briefing-and-consultations-on-developments-in-the-political-dialogue-in-libya.php (accessed 8 March 2018).


 Wolfram Lacher, “The UN Wants to End Libya’s Civil War. Here’s the Big Challenge They Face”, Monkey Cage, 5 November 2015, https://www.washingtonpost.com/news/ monkey-cage/wp/2015/11/05/why-libya-isnt-just-divided-but-fragmented/ (accessed 22 May 2018).


 Author’s interview with UNSMIL official, Tripoli, April 2018.


 Author’s interviews with Western diplomats, Berlin and London, September 2015.


 Randeep Ramesh, “UN Libya Envoy Accepts £1,000-a-Day Job from Backer of One Side in Civil War”, The Guardian, 4 November 2015.


 Somini Sengupta, “Russia Resisting Europe Request for Tough U.N. Anti-Smuggling Step”, New York Times, 15 Septem­ber 2015.


 Author’s interviews with Western diplomats, Tunis and London, November and December 2015.