On 17 May, Burundians will vote on constitutional amendments that would allow President Pierre Nkurunziza to prolong his stay in power. Those new provisions also could start to dismantle the carefully negotiated Hutu-Tutsi ethnic balance, defined in the 2000 Arusha agreement that helped end Burundi’s civil war. A major outbreak of violence in the country does not appear likely around the vote, despite a deadly attack on a village on 12 May; the status quo could even drag on for years. But the regime’s repression, the potential demise of power sharing in Burundian institutions and the crumbling economy are harbingers of instability.

Although the European Union (EU) has lost leverage over Nkurunziza’s government in recent years, it retains a strong interest in preventing such instability. The EU and its member states should closely watch developments before, during and after the referendum, and continue to explore channels for pressuring the government while supporting the population. These include encouraging African leaders and the African Union (AU) to renew mediation attempts between the regime and the opposition, while keeping Burundi in the international spotlight. As the Burundian economy collapses, the EU, which suspended direct budgetary support to the Burundian government in 2016, should also take steps to ensure that the aid it now channels through the implementing agencies of the UN, EU member states and international non-governmental organisations helps Burundians as best possible.

The government’s main intention with the forthcoming referendum is to lengthen presidential mandates from five to seven years. This change would restart the clock on the two-term limit – rather than annulling it – potentially giving President Nkurunziza a further fourteen years in power. The new draft constitution also stipulates that ethnic quotas in parliament, government and public bodies be reviewed over the next five years. These quotas, intended to protect the Tutsi minority by guaranteeing the Tutsi 40-50 per cent representation in different state institutions, including the army, were a key part of the Arusha agreement.

The regime has designed the constitutional changes primarily to remove any obstacle to its control of the state apparatus. But in the process it may also be laying the groundwork for reversing ethnic checks and balances. The same is true of the draft constitution’s provisions to reduce the number of vice presidents (currently there are two, one Tutsi and one Hutu) to one and to replace the two-thirds majority requirement for parliament to pass particularly significant legislation with a simple majority.

The regime, including the ruling party’s youth wing, the Imbonerakure, has carried out a campaign of intimidation against anyone who opposes the referendum or calls for a No vote. It is using threats of violence to push Burundians to register for the vote in hopes of minimising abstention, and identifying people in campaign meetings. The government has banned Western media outlets – the BBC and Voice of America – from radio broadcasting for the duration of the campaign, while its own propaganda machine is in full swing. It has forced citizens to make financial contributions that it claims will support forthcoming elections.

The forced march to the referendum has further accentuated divisions among President Nkurunziza’s opponents, despite opposition factions making a renewed attempt to align their positions at the start of 2018. The Amizero y’Abarundi coalition and the Sahwanya-Frodebu party, which remain in Burundi, have both declared they intend to campaign for a No vote. The exiled opposition, under the umbrella of the Conseil national pour le respect de l’accord d’Arusha (CNARED), is calling for a boycott. The divide over the referendum exacerbates the historical divisions over strategy and personal rivalries within the opposition.

Significant violence around the referendum appears unlikely, despite a 12 May attack on a village near the Democratic Republic of Congo border in which 26 people were reported killed by unidentified assailants. This attack comes after a relative absence of major security incidents since 2016, as armed opposition groups have suffered several setbacks. Some of their members were arrested by the Tanzanian government in 2017, sent back to Burundi, and have since disappeared. Those attacks that have taken place, which were launched from South Kivu in the neighbouring Democratic Republic of Congo, have failed to inflict significant losses on Burundian security forces or generate local support. But if the frequency of armed clashes between the army and insurgents has declined since 2016, human rights abuses continue. According to the human rights organisation la Ligue Iteka, 456 people were assassinated, 283 tortured and 2,338 arbitrarily arrested in 2017, the vast majority by the government.

President Nkurunziza and his party are developing a doctrine that mixes personality cult, religion and historical mythology to justify his prolonged stay in power. The president is now referred to as “supreme traditional leader”. The president and his wife, both active in new Pentecostal churches and prayer crusades, adhere to a theocratic vision that blends traditional Burundian signs of power with divine attribution; tellingly, the government is planning to build a large prayer centre in Gitega where ruling party members will be required to attend lengthy retreats. More broadly, this emerging doctrine presents a Manichean view of history wherein a harmonious pre-colonial Burundi was later spoiled by the machinations of external powers, in particular Belgium, though language pointing the finger at foreigners also tends to contain veiled references to the role played by their supposed Tutsi allies.

The Burundian economy has been severely hit by the loss of overseas aid since 2015, and by the flight of human and financial capital. Gains made in health and education since the early 2000s – notably drops in infant mortality and increasing numbers of Burundian children in school – have stalled. Shortages of currency and fuel have afflicted all sectors. Some 430,000 Burundians have fled to neighbouring countries, principally Tanzania.

Though many Burundians already struggle to make ends meet, the government is introducing new taxes and ad hoc levies. As its relations with Western governments have worsened, it has turned to Turkey, China and Russia for aid. But while these countries might afford the government political support and some financial respite, they are unlikely to offer the sort of budgetary or technical help that Western donors provided. Meanwhile, the impact of private investment in the mineral sector on the wider economy is unlikely to be significant, at least in the short term.

After negotiations with the government under Article 96 of the Cotonou Agreement, the EU and its member states decided in March 2016 to suspend cooperation due to Burundi’s rights abuses. Instead, it now channels development aid through international NGOs, the implementing agencies of EU member states and UN agencies. The president and his top officials paint European aid policy and sanctions (which target a handful of those officials) as deliberately aimed at hurting the Burundian people. In some cases, the regime has cracked down on civil society groups that have worked with international donors, including by imprisoning NGO members on spurious charges.

The EU and its member states should take steps to help check Burundi’s repressive authoritarianism and alleviate deteriorating living conditions for its people.

On the former, Nkurunziza’s government has brushed off sporadic pressure from Western donors and actors such as the UN Office of the High Commissioner for Human Rights to open space for its opponents. Nor have mediation efforts of the sub-regional body, the East African Community (EAC), made progress. Indeed, some African leaders appear inclined to believe the government’s argument that there is no crisis to mediate.

That argument is flawed. The regime probably can keep dissent under wraps for some time. But the consolidation of its rule and dismantling of the Arusha power-sharing provisions augur ill for the country’s stability over time. The EU and its member states should press African powers and the AU to renew mediation attempts between the regime and the exiled opposition, with the aim of ensuring a credible election in 2020. They should strive to maintain international attention on Burundi, with EU member states on the UN Security Council pressing to keep Burundi on the council’s agenda. The EU also should uphold its position that conditions in the country do not allow for a free and fair referendum.

In light of its 2016 suspension of direct support to the government, the EU needs to redouble efforts to find ways to ensure its aid supports the population. In addition to the support it channels through international NGOs, it should continue pursuing its plan to directly support local NGOs, but with particular caution not to expose them to risk. This could mean providing them with adequate funding to reinforce their own management and legal capacity in case the government continues to harass them through the courts. The EU should also reinforce its delegation in Bujumbura and strengthen the tracking mechanisms with its implementing partners to prevent any misuse of its funds.

Late February and early March 2018 saw Sri Lanka’s most serious and widespread incidents of anti-Muslim violence since gaining independence in 1948. Police failed to contain Sinhala Buddhist mobs in central Kandy district; the rioting appeared close to spinning out of control before President Maithripala Sirisena declared a state of emergency on 6 March. Within 48 hours of army and other military units being deployed, order was restored, but not before more than two dozen mosques had been destroyed, hundreds of houses and businesses vandalised, and two people killed. The episode shredded the ruling coalition’s already tattered reformist credentials and hurt chances for post-war reconciliation across ethnic and religious boundaries. It revealed the depth of mistrust and fear between Sri Lanka’s Buddhists and Muslims, and underscored the risk of more violence to come.

The European Union, its member states and other international partners should support efforts by the Sri Lankan officials, religious leaders and civil society groups to prevent further violence and address the underlying mutual misunderstanding between communities. It can do so by:

• sending strong messages to the government, through all available channels, that it supports the strict enforcement of laws against hate speech and religious violence, including through criminal prosecutions;
   
• offering financial and technical support to efforts by the government, civil society or media organisations to rapidly fact-check and counter rumours on social and traditional media; and by
   
• supporting efforts to strengthen local-level inter-religious committees, in particular by assessing the effectiveness of past initiatives and sharing lessons learned to help redesign such bodies in innovative ways.

Discussions of Sri Lanka’s long, bloody history of conflict tend to focus on the military campaign to crush the Liberation Tigers of Tamil Eelam (LTTE) and earlier leftist uprisings. Yet anti-Muslim violence is an enduring feature of modern Sri Lankan history and grew worse after the LTTE’s 2009 defeat. Many Sinhalese, including influential monks, have long feared that Sri Lanka’s foundational Buddhist and Sinhala character was under threat, with weak politicians incapable of protecting it. Whereas, before 2009, Sinhala nationalist insecurity centred on Tamil separatism, today nationalists point to Muslim “extremism” as the primary threat. Militant Sinhalese accuse Muslims of using clandestine means to suppress the Sinhala Buddhist population and gain economic and demographic dominance. Drawing in part on global Islamophobic discourse and events in Myanmar, Buddhist activists feed on Sinhala perceptions that over recent decades the country’s Muslims have grown more publicly and devoutly religious, and thus alien. Muslims make up under 10 per cent of the population, Sinhala Buddhists 70 per cent.

The deadly train of recent events began in the south-eastern town of Ampara, where Sinhala mobs attacked Muslim shops and a mosque on the night of 26-27 February. The violence was sparked by unfounded claims, spread through social media and backed by a video recording of an apparent “confession”, that staff in a Muslim-owned restaurant had placed a “sterilisation pill” in the food of Sinhala customers. Police were slow to react or make arrests and quickly released the alleged rioters on bail.

Activists and Muslim ministers had warned the prime minister and other senior officials that violence was brewing elsewhere. The police’s lax response in Ampara appeared to encourage militant Buddhist networks to strike again. Following appeals by Mahasohon Balakaya, an anti-Muslim group based in Kandy, and prominent monks, crowds turned out in Kandy on 5 March, angry at an earlier murder of a Sinhala Buddhist man by four Muslim men (there was no evidence the crime was communally motivated and all four attackers were promptly jailed). Over the next three days, Sinhala Buddhist mobs moved systematically from village to village burning and vandalising Muslim shops, houses and mosques. Police were again slow to respond, and in at least two cases, members of the Special Task Force – the elite police paramilitary unit – attacked Muslims, possibly in an attempt to falsely implicate them in the riots.

The violence came at a moment of confusion and weakness for the ruling coalition, weeks after a stinging defeat in local elections at the hands of former President Mahinda Rajapaksa’s newly formed Sri Lanka Podujana Peramuna (SLPP, Sri Lanka People’s Front), which ran a strongly nationalist campaign, including warnings of alleged abuses by Muslims. The changing political climate reinforces sympathy for ultra-nationalist agendas in the overwhelmingly Sinhala and Buddhist bureaucracy. It deepens the reluctance of police, bureaucrats and politicians to take action that could be seen as supporting Muslims.

In 2013 and 2014, under Rajapaksa and his powerful brother, Gotabaya Rajapaksa, who headed the police and military, security agencies were accused of supporting militant Buddhist groups. Among these, the most prominent was Bodu Bala Sena (Buddhist Power Force, or BBS), which incited deadly anti-Muslim riots in Aluthgama in June 2014. State support reportedly included facilitating large BBS rallies, allowing its cadres to publicly harass and intimidate critics with impunity, and intervening in criminal cases against militant Buddhists, including by pressuring victims to withdraw legal complaints – or not to file them at all. However, after January 2015, when Mahinda Rajapaksa was voted out, violent anti-Muslim campaigns were supposed to be a thing of the past.

Yet, although President Sirisena and Prime Minister Ranil Wickremesinghe campaigned on a promise to crack down on militant groups, their coalition government has overseen not a single prosecution for previous religious violence. Militant Buddhist protests and attacks on mosques and Muslim businesses have continued, albeit at a lesser intensity. Online hate campaigns and militant organising have also proceeded apace, and fear and mistrust of Muslims remain as high as ever. Indeed, since coming to power, Sirisena himself and a senior cabinet member have met numerous times with BBS chief Galagoda Aththe Gnanasara and other militant monks. While officials claim the meetings were to encourage dialogue, most observers believe they aimed at shoring up the government’s credibility with Buddhist nationalists. Other ministers have called for BBS leaders to be prosecuted.

The government is running out of time to develop a strategy or build the political will to address two central issues underlying recent violence: Sinhala Buddhist nationalism and its politically powerful mix of entitlement and insecurity; and impunity for violence done in the name of protecting Sinhala and Buddhist dominance.

The current high tensions and suspicions are deeply rooted and cannot be resolved quickly. Nonetheless, much can be done by the government and other actors to address misconceptions and rebuild trust. The EU can support these efforts in a variety of ways, as suggested below, but it must do so with discretion, taking its cues from supporters of reform in the government and civil society, and recognising Sinhala sensitivities about foreign involvement.

Most important is for the government to conduct a quick, impartial investigation into the March violence and to speed up prosecutions for past actions. An expeditious inquiry would send a signal to those who might be tempted to commit violent acts in the future, and strengthen forces of reform in the judiciary, police and other state institutions. Some militant Buddhist leaders have been arrested for their role in organising the violence in Kandy. That is a positive first step, but it will bear fruit only if indictments and prosecutions follow. Ongoing cases against Gnanasara – for contempt of court and assault, among other charges – and other militant monks should be allowed to proceed to indictment or otherwise be concluded. The government should insist the police apply Sri Lanka’s International Covenant on Civil and Political Rights (ICCPR) Act, which has tougher penalties for hate speech and anti-minority violence than the regular criminal code.

The EU should send strong messages to the government that it supports criminal prosecutions for religious violence, including through the ongoing human rights monitoring process that accompanies EU Generalised Scheme of Preferences (GSP+) trade benefits, and in its regular dialogues with the government, including the meeting of the Working Group on Governance, Rule of Law and Human Rights in June 2018 and the EU-Sri Lanka Joint Commission meeting in fall 2018. Such prosecutions would be best framed as combatting not only anti-Muslim attacks, but also the larger phenomenon of impunity that has harmed all communities and places all citizens at risk of arbitrary violence.

The government can take other relatively easy, low-cost measures to develop a comprehensive and well-resourced information strategy to counter myths and misperceptions about Muslims. It could straight away establish an office tasked with rapidly fact-checking and countering rumours on social media, which, as the March incidents and earlier episodes have shown, are powerful sources of incitement. The office should liaise closely with the police and aim for quick distribution through all forms of state and private media, including social media. The EU and member states could offer financial and technical support to efforts by the government, civil society or media organisations on fact-checking.

In parallel, the government should strengthen existing district-level inter-religious committees to act more effectively as early-warning and mediation mechanisms. The committees should include police, influential monks, local government officials and local politicians, as well as Muslim representatives, and should report directly to the president and prime minister in emergencies to ensure that effective interventions are authorised when they are needed. Government-sponsored committees will need to coordinate better with civil society-led inter-religious groups, which have had limited impact over the years in part because they are often delinked from government and Buddhist religious authorities. The EU should support such committees, in particular by assessing the effectiveness of current and past initiatives and sharing lessons learned to help redesign such bodies.

The EU could also support existing projects by Muslim leaders to reach out to Buddhists, including monks, to explain and demystify Muslim teachings and practices, and familiarise them with what happens in mosques. Initial outreach programs have reportedly been well received by the monks involved. One option for EU support would be funding for European religious and community leaders and officials to share with leaders of all communities their experiences and lessons learned from their own inter-religious community work.

International efforts to broker a solution to Venezuela’s implosion so far have not borne fruit. The crisis is spilling across Venezuela’s borders, with some 1.5 million Venezuelans fleeing the country over the past year and a half. Venezuelan President Nicolás Maduro’s government is unable or unwilling to reverse the economic and social collapse brought on by its misguided policies. It frustrated the last round of talks between it and opposition representatives by unexpectedly calling an early presidential election, even as those negotiators discussed reforms to help level the playing field. That vote, now scheduled for 20 May, is more likely to aggravate than resolve the crisis, as the EU’s April declaration on the situation in Venezuela identified. Most opposition leaders call for a boycott, arguing that Maduro’s re-election is predetermined. Latin American governments in the ad hoc Lima Group, as well as those of the United States, Canada and Spain, have declared they will not recognise the result should the elections proceed as planned. The European Parliament endorsed the same stance in a resolution adopted at the start of May.

The EU, U.S. and other Western governments have imposed targeted sanctions on dozens of government officials, including the president and vice president. The U.S. has also banned most loans to Venezuela and is considering some form of oil embargo. A solution to the crisis can only come through a negotiated transition, which will require new talks between the government and opposition and additional pressure on the government. Ideally, Lima Group members would use the threat of their own targeted sanctions – such sanctions from Latin American governments would be almost unprecedented – to help push the government back to the negotiating table. To contribute to such a strategy, the EU and its member states, should:

• Agree with Lima Group governments and the U.S. on clearly delineated steps the government should take to have Western sanctions lifted and avert Latin American sanctions.
   
• Caution against the oil embargo floated by the U.S. and called for by some opposition hardliners, which would worsen the humanitarian emergency.
   
• Encourage China, during engagements with Chinese officials, to help nudge Maduro to accept talks.

At the same time, efforts to contain the humanitarian crisis should continue. To this end, the EU and its member states should:

• Reinforce their support for migrants and refugees along Venezuela’s borders.
   
• Continue to seek out opportunities for delivering aid inside the country.

Particularly for the latter efforts, the EU will need to maintain a strict separation between the provision of humanitarian assistance and political demands on the government.

Venezuela is sinking ever deeper into a profound economic and social crisis. Annual inflation could reach upwards of 300,000 per cent by year’s end. Despite a government plan to strike three zeroes off Venezuela’s currency, cash is almost impossible to obtain, hitting the poor, many of whom have no other means of payment, particularly hard. Over eight million Venezuelans cannot afford three meals a day. Protein has disappeared from many of their diets. Essential medicines are lacking: for some such medicines only 20 per cent of the quantity needed is available; others have entirely run out. Many of those suffering chronic diseases like cancer, HIV/AIDS or haemophilia are dying for lack of treatment.

Most public hospitals cannot guarantee running water or working lifts, let alone equipment such as X-ray machines. Patients are forced to provide their own medical and surgical supplies. Many operations are cancelled because blood banks lack reagents to ensure transfusions are safe. Long-controlled diseases like measles and diphtheria are making a comeback. Parts of the country are in the throes of a malaria epidemic. Yet the Venezuelan government denies the humanitarian crisis exists, portraying any coverage of the crisis as misinformation designed to undermine its rule. It also rejects much humanitarian aid, arguing that such efforts are part of a foreign plot to oust it.

As many as 1.5 million people have left the country in the past eighteen months, and a similar number may leave in the course of this year. The exodus has placed public services in neighbouring countries under strain, with governments in countries as far away as Chile having to adapt immigration regulations accordingly. Temporary shelters and soup kitchens catering to Venezuelans have been set up in Colombian and Brazilian border towns. UN agencies and the EU are now beginning to provide international aid in those locations.

A presidential election is scheduled for 20 May, but is unlikely to provide a way out of the crisis. In February, the government brought forward the election by more than six months, thus sabotaging internationally facilitated talks with the opposition over electoral reforms that were underway at the time. Most opposition parties are boycotting the poll, but beyond that do not offer a coherent strategy for pressuring the government.

Former state Governor Henri Falcón of the Avanzada Progresista party, with the backing of two other small parties, is contesting the presidency. To do so, he has broken with the Democratic Unity (MUD) opposition coalition, which includes most of the more moderate opposition parties that had been negotiating with the government and are now planning to boycott the polls. The opposition’s harder-line wing, now represented by the Soy Venezuela movement, is calling for a “humanitarian intervention” – for the U.S. to intervene militarily, in other words – and for President Maduro to be impeached and tried for crimes against humanity. On 17 April, parliament, in which opposition politicians, mostly from parties in the MUD, hold a majority, voted overwhelmingly to approve Maduro’s trial for corruption by an ad hoc “Supreme Court in exile” – composed of judges appointed to the Supreme Court by the parliament and later forced into exile. But this initiative will have little practical effect. Parliament has been rendered largely powerless, especially after a new Constituent Assembly, dominated by ruling party loyalists, was elected last year in a vote the opposition also shunned.

Polls indicate that most opposition voters will abstain on 20 May, offering Maduro a clear chance of victory despite popularity ratings below 30 per cent. Even if Falcón were to win, the government’s control of electoral authorities, the Supreme Court – which has the final word on electoral disputes – and the security forces means it would have the power to block his victory. The absence of credible international observer organisations, which declined to deploy observers given the conditions in which the vote is being held, also gives Maduro a free hand.

Dozens of military officers, including commanders of key units such as the armoured Ayala battalion in Caracas, have been detained for allegedly plotting against the government. Their arrests lend credence to widespread accounts of unrest in the barracks. With the exception of a minority of mostly top military leaders, who are accused of benefiting from corruption and other criminal activities, members of the armed forces suffer the same deterioration in living standards as other Venezuelans. Military canteens often provide little or nothing to eat. That said, a coup attempt, while impossible to rule out, would be hard to pull off: the armed forces are fractured and extensively penetrated by counter-intelligence.

Venezuela’s international isolation has intensified markedly over the past year, with regional governments in particular turning their back on Maduro, especially after the breakdown of talks in February. Further sanctions are likely unless the president postpones the vote and takes measures to level the playing field. That said, exactly how the threat by Latin American and other governments to “not recognise the results” would be put into practice is unclear. Many governments already have withdrawn ambassadors from Caracas. But entirely severing diplomatic relations could reinforce the government’s siege mentality and backfire.

The Lima Group issued a fresh statement at the mid-April Summit of the Americas, which the summit’s host, Peru, barred Venezuela from attending. That statement called for free and fair elections and the restoration of democracy. The group also emphasised the need for humanitarian assistance, both within Venezuela and in neighbouring countries hosting Venezuelans that have left. Meanwhile, France, Germany, Italy, Spain and the UK, together with several Latin American governments, Canada, Japan and the U.S., have backed a joint initiative to locate and seize those assets of Venezuelan officials that they have reason to suspect have been acquired through corruption.

Venezuela’s crisis is now a grave threat not only for its own people, but also for the wider region. A lasting solution requires a negotiated transition. It also requires comprehensive economic reform, which can only be carried out by a government that enjoys international political and financial support. The starting point must be a return to negotiations between the government and opposition leaders.

Thus far, the threat of economic collapse has not persuaded the group around Maduro to participate in such talks, which would, in essence, be aimed at negotiating the end of one-party rule and the restoration of democracy. Top officials perceive potential exit costs as extremely high, and fear they would risk prosecution for alleged corruption, drug trafficking and human rights violations were they to lose power. For its part, the opposition is split into three main factions, each frequently adopting tactics that contradict those of the other two. Calls for military intervention by the harder-line Soy Venezuela faction are particularly counterproductive, fuelling the government’s accusations that humanitarian aid is a foreign plot.

As the Yemen war enters its fourth year, prospects for military escalation and greater regional spillover are growing. The Saudi-led coalition’s military campaign along the Red Sea coast and in the Huthis’ home governorate of Saada, coupled with intermittent missile barrages fired by the Huthis at Saudi Arabia, threaten to quash the opportunity to revive the political process presented by the appointment of a new UN special envoy, Martin Griffiths. Military escalation could trigger direct confrontation between Saudi Arabia and its allies, particularly the United States, and Iran, which Riyadh accuses of assisting the Huthis in developing their missile program.

In this environment, the EU and its member states should:

• As an urgent priority, help prevent the looming Saudi-led coalition invasion of the Red Sea port of Hodeida, which would compound the already acute humanitarian crisis and could spark a wider war; such efforts would involve diplomatic engagement with Riyadh and Abu Dhabi, ideally in coordination with the United States; and publicly opposing such an invasion, while condemning and pressing the Huthis to end their missile attacks against Saudi Arabia. Quiet outreach to Tehran could help, urging Iran to use what influence it has with the Huthis to discourage such missile attacks.
   
• Assist the UN envoy in reviving a political process that is more inclusive and realistic. EU member states on the UN Security Council (France, the Netherlands, Sweden and the United Kingdom) could promote a new Security Council resolution that better supports the UN envoy’s efforts than the April 2015 Resolution 2216, which is outdated and places unrealistic demands on the Huthis. The EU delegation to Yemen is well placed to assist the new envoy if talks materialise, notably by encouraging the Huthis’ cooperation.
   
• Adopt a clear, public policy line on south Yemen, where separatist sentiment is increasing; such a line would oppose a unilateral move toward independence but recognise southern Yemenis’ grievances and the importance of revisiting the question of state structure and decentralisation.
   
• Continue urgent efforts to alleviate the war’s humanitarian fallout, including by demanding from the coalition unhindered humanitarian and commercial access to all seaports, including Hodeida, as well as the Sanaa airport.

On 4 December 2017, the Huthis killed their former partner, Ali Abdullah Saleh. Since then, the Saudi-led coalition and its Yemeni allies have acted as if the military and political tides have shifted in their favour. They have tried to pull former Saleh supporters to their side, encouraged rifts within the Huthi movement, stepped up efforts to target the group’s leadership and pressed the Huthis on a number of war fronts.

In these endeavours they have had some success. Between December 2017 and February 2018 the United Arab Emirates (UAE) and aligned Yemeni fighters won important tactical victories in Hodeida and Taiz provinces. Since then coalition-aligned forces have made small but steady gains, though not enough to shift the overall military balance. As in the past, the coalition has overestimated its ability to harm the Huthis in their northern highland strongholds. On 19 April, a coalition airstrike killed the head of the Huthi Supreme Political Council, Saleh Sammad, the de-facto president of the north and the highest-ranking Huthi killed thus far. Known as a moderate within the movement who could work with the late President Saleh’s party, his death is unlikely to reap significant military gains for the Saudi-led coalition but is a blow to peace prospects. Internal divisions within the anti-Huthi front continue to be its Achilles heel: some pro-Saleh fighters have joined the war against the Huthis, but many refuse to support President Abed Rabbo Mansour Hadi and his partners in Islah, an Islamist party. Islah and Hadi affiliates are at particular odds with UAE-aligned groups in areas such as Taiz and in south Yemen, which was an independent state prior to 1990.

After killing Saleh, the Huthis are simultaneously more open to diplomacy and more willing to up the military ante in response to coalition offensives. They have stated publicly and privately that they are ready to negotiate with Saudi Arabia over security concerns and to re-engage with the UN process under the new envoy. It is unclear if this readiness is a product of military pressure or an increased sense of security, as in the past the Huthis had cause to worry that Saleh would strike a deal behind their backs. Either way, their increased interest in talks offers hope of a political breakthrough.

That said, 2018 has seen an unprecedented uptick in Huthi missile attacks on Saudi Arabia. There is growing evidence of Iranian supply of Huthi weapons, including missile and drone technologies. For the Huthis, coalition attacks on Hodeida, the main port in the territories they control, and Saada, their home governorate, represent existential threats. Hodeida in particular is a red line. The coalition’s blockade, ostensibly to prevent weapons smuggling to the Huthis, has made the port a chokepoint for goods entering the north; prolonged fighting there could compound Yemen’s humanitarian disaster manifold. The Huthis have proclaimed they are willing to sink commercial ships to deter an attack. In April, Saudi Arabia accused the Huthis of firing on a Saudi-flagged oil tanker in the Red Sea, the first attack of its kind.

To avoid this scenario and the regional escalation it could trigger, the EU should take a clear public position against a coalition attack on Hodeida for both humanitarian and political reasons, and engage in vigorous diplomacy, in Riyadh, Abu Dhabi and Washington, to help prevent it. Diplomatic efforts also should be directed toward encouraging both sides to de-escalate the conflict ahead of a possible resumption of talks. This could include the Huthis halting missile strikes at Saudi Arabia and ships in the Red Sea in return for the Saudi-led coalition stopping their offensive moves into Saada and along the Red Sea coast in Hodeida and Taiz provinces. The new UN envoy, with the help of the EU delegation and member states, could broker such an agreement.

If military escalation can be held at bay, the envoy will have a chance to revive negotiations over a cessation of hostilities and a return to an internal Yemeni political process. To be successful, these efforts will need a new framework that improves the one set forth in UN Security Council Resolution 2216. That resolution sets out a bilateral structure for talks between the Hadi government and the Huthi-Saleh bloc, which has become outdated and which never represented the range of Yemeni forces with influence on the ground. It also places unrealistic preconditions for a political settlement on the Huthis, including requiring them to withdraw from territories gained and hand over weapons. The EU, and in particular Security Council members France, the Netherlands, Sweden and the United Kingdom – the latter being the penholder on the Yemen crisis – should press for a new resolution that would support the UN envoy’s efforts based on his plan for reviving the political process, which he will present in June 2018.

The EU delegation is uniquely placed to assist the UN envoy in improving the structure and substance of potential negotiations. As a non-belligerent in the Yemen war, the EU has access to all sides, including the Huthis. The delegation could assist in communicating with and encouraging Huthi cooperation at the various stages of talks. Information and lessons from EU-sponsored Track II events during the course of the war, particularly with local security stakeholders, could help guide the process of improving intra-Yemeni negotiations. The EU and its member states should work with the UN envoy to produce a negotiating framework that more effectively includes women and other civil society representatives in decision-making roles early in the process, a deficiency during the last three rounds of UN-sponsored talks.

South Yemen, where separatist sentiment is strong and the UAE is supporting separatist-leaning groups, is a critical flashpoint. In effect, the south is moving toward independence, but not all southern stakeholders support the idea. Nor do Yemenis in the north. The EU and its member states should have a clear, public policy line that opposes a unilateral move toward independence but recognises southern Yemenis’ grievances and the need to revisit the question of state structure and decentralisation, which remained unresolved in Yemen’s 2014 National Dialogue Conference. The EU delegation and member state representatives should also prioritise engaging with the UAE-supported Southern Transition Council and other southern political groups, and support their inclusion in intra-Yemeni negotiations.

Finally, ameliorating the war’s humanitarian impact should remain a top priority. The numbers are staggering. Over 22 million Yemenis – three quarters of the population – need humanitarian assistance. Of those, 8.4 million are at risk of starvation. Three million are internally displaced, mostly women and children.

The EU and member states should continue to demand unhindered humanitarian and commercial access to all seaports, including Hodeida, as well as Sanaa airport. To assist in their full opening, the EU is well placed to offer assistance to the UN in negotiating and possibly implementing security checks that address the Saudi-led coalition’s legitimate concerns regarding arms smuggling. They should also press the Huthis to allow unhindered humanitarian access to areas they control and to ease restrictions on aid workers operating in these areas. Beyond physical access, the EU should work with the Yemeni Central Bank to stabilise the value of the Yemeni riyal and promote a political compromise by which the Hadi government pays salaries to all civil servants nationwide, including in Huthi-controlled territories.