Between the morning of 30 July and the evening of 31 July 2026, more than 70,000 people crossed into Ceuta, a Spanish city of 80,000 inhabitants on the North African coast, bordering Morocco. It was the largest single breach of the European Union’s external border on record.

At 1:52 pm on 29 July, an officer of Spain’s National Intelligence Centre in Ceuta informed the office of the government’s delegate in the city that calls were circulating online for a crossing “by fence and by sea” the following evening, on Morocco’s Throne Day. Facebook groups behind the call had around 180,000 members between them. That same day, the intelligence service alerted Morocco’s two intelligence agencies, the DGST and the DGED, and asked for their cooperation in stopping the plan.

We know all this thanks to 40 internal documents that the Spanish government later released, under pressure from both parliament and the courts. Yet the following morning, gendarmes on the beaches near Fnideq stood aside as tens of thousands of people – overwhelmingly young Moroccans – walked into the sea. At least 141 died, most of them in the sea off the Tarajal breakwater.

Within 72 hours, roughly 90 per cent of those who survived were back in Morocco. Most of them returned voluntarily, in cooperation with Rabat. Several thousand remained, among them more than 1,000 unaccompanied minors, who under Spanish law cannot be returned without a verified family placement, and applicants for asylum, whose removal is suspended until their claim has been assessed. Rabat has demanded that the minors be handed over.

Whether Moroccan authorities directed the crossing rather than merely failing to stop it remains contested. Pedro Sánchez told parliament on 3 September that during “the 10 central hours” of 30 July the Moroccan gendarmerie “permitted the crossing”, while maintaining that no proof had been offered that Morocco designed or executed the operation. But that is not the question that matters most. A state that polices its northern coast as tightly as Morocco does, and that two weeks later swiftly blocked a new online call to cross into Ceuta, does not just lose control for 10 hours by chance.

By design or by omission, the Moroccan state is responsible for the 141 people who died in the water. But other choices, too, made this tragedy possible.

A European failure

Two policy failures converged in Ceuta – one European, one Spanish.

The European policy in question is border externalisation. Since the agreement with Turkey in 2016, the EU has paid third countries to hold people back from its borders and has treated what happens to those people as a matter for the jurisdiction concerned. Libya in 2017, Morocco in 2019, Tunisia in 2023, Mauritania and Egypt in 2024; and now the Pact on Migration and Asylum, which extends the model rather than questioning it. Morocco receives more European border funding than any other state in North Africa.

Progressive forces have opposed this arrangement on humanitarian grounds for a decade. The countries the EU designates as safe are not safe for the people held in them, as the documented record of torture, rape, and abandonment in the desert makes plain. Moreover, defending the right to seek a life elsewhere means refusing to see people turned into a commodity exchanged between governments.

But Ceuta also exposes a second flaw in the model, one that governments indifferent to the first may find harder to ignore. Every externalisation agreement transfers to a third state a measure of control over the foreign policy of the state that pays. Turkey demonstrated this at the Greek border in 2020. Belarus demonstrated it against Poland and Lithuania in 2021. Morocco has now demonstrated it twice in five years, the second time on a scale seven or eight times larger than the first.

The EU believed it had purchased control over its external borders. What it has actually acquired is a set of dependencies, one per neighbour, each exercisable at that neighbour’s convenience.

Spain’s responsibilities

Spain, for its part, has pursued appeasement, attempting to secure a difficult neighbour’s goodwill through unilateral concession.

Morocco has used the border with Ceuta and Melilla as an instrument of pressure for decades. In 2002, it occupied Perejil, an uninhabited island close to the Moroccan coast but under Spanish sovereignty. In 2018, on the eve of Throne Day, it let several hundred people storm the Ceuta fence. In May 2021, after the leader of the Polisario Front – a Sahrawi nationalist liberation movement that fights for the independence of Western Sahara from Moroccan control – was admitted to a hospital in Spain under a false identity, some 10,000 people entered Ceuta in under three days while Moroccan police looked on.

Spain did not respond by reducing its exposure. Instead, in March 2022, Sánchez sent an official letter to King Mohammed VI endorsing Morocco’s Autonomy Plan for Western Sahara, reversing half a century of Spanish neutrality over the decolonisation of its former colony and doing so without a parliamentary debate and without consulting the coalition partner then in government. In June that year, at least 23 migrants and asylum seekers died in a crowd crush at the fence in Melilla – Spain’s other city on the North African coast – on the Moroccan side of the border, as Moroccan and Spanish security forces moved to repel a mass crossing.

Four years on, the customs posts at Ceuta and Melilla still do not operate normally, and Spain and Morocco have not resolved their airspace dispute. That concession taught Rabat that pressure at the border produces results.

In August this year, Morocco’s justice minister, a royal appointee, went on state television to claim Rabat’s “historic and geographic right” over Ceuta and Melilla, demanding a bilateral commission on their “return to the homeland”. In September, responding to the documents declassified by the Spanish government, Morocco’s foreign minister, Nasser Bourita, issued a statement accusing Spanish politicians and media of “strategic blindness”, warning that this would produce “recurring tensions”, and reminding Madrid that “geographic proximity is an immutable fact”. The two cities have since featured in Morocco’s parliamentary election campaign.

Yet until the hours preceding this summer’s crisis, the Spanish state still trusted Morocco to act on its warnings that a crossing was being organised. Madrid’s posture was to fully rely on the goodwill of a neighbour that had already used the border against it, five years earlier, to extract the largest foreign policy concession in decades – and to double down on a policy that had already caused unnecessary deaths.

Responsibility for this failure lies with Spain’s Socialist Party (PSOE). Sumar, the junior coalition partner of which I am a member, has warned for years that the relationship built since 2022 would produce a tragedy like this summer’s. After 30 July, it has repeated in parliament that Spain suffered an attack in which people in extreme vulnerability were used as the instrument, that the explanation attributing the crossing to smuggling networks is not credible, and that the response was belated and inadequate. It has done so not with the intention to bring down the government, but to change the policy.

The reticence of the Socialist side is structural rather than personal. To name Morocco’s responsibility is to concede that the 2022 concession bought nothing. A government that builds its regional policy on keeping a neighbour appeased loses the capacity to criticise that neighbour’s actions. Five weeks passed before Sánchez addressed parliament. The summoning of the Moroccan embassy’s representative on 21 August was kept secret for two weeks, while similar episodes in disputes with Israel, Italy, and Argentina were widely publicised. The prime minister’s most recent formulation is that he “exculpates no one” and will act “with the utmost firmness” if solid proof emerges. Morocco, for its part, has informed Madrid that it is conducting its own investigation. It has not said who is conducting it or when it will report back.

Why Europe did not help

Madrid’s reticence explains why no case for European solidarity was made, but does not justify how the EU and some of its member states reacted.

On 31 July, Italy’s right-wing government threatened to suspend Schengen with Spain and blamed the deaths on Spain’s extraordinary regularisation of undocumented migrants, a process opened in January which had drawn more than 800,000 applications by July. No emergency Council was called. The Treaty’s solidarity clause was not invoked. A member state had suffered a breach of its external border on a scale never seen before, and the response from the other capitals was silence.

The reasons are not hard to find. Spain recognised the state of Palestine in 2024, has been the most outspoken European voice on Gaza, has declined to make its bases fully available for offensive operations in the US war on Iran, and has decided to regularise migrants rather than deporting them. For many European powers and for the US, these count as serious transgressions.

In the months before the crossing, several of Spain’s allies moved closer to Rabat. On 15 July, two weeks before the mass crossing, the US House of Representatives passed the State Department funding bill for 2027. The report the Appropriations Committee had published in April to accompany it described Ceuta and Melilla as “administered by Spain” but “located in Moroccan territory”, remaining “the subject of Morocco’s longstanding claim”. It also endorsed State Department efforts to facilitate dialogue on their future status. The report carries no legal force, and its sponsor has since described the wording as purely geographical. Still, it was the first time the US Congress put the Moroccan claim in writing.

Moreover, on 7 August, hours after taking office, Colombia’s new president, pro-Trump right-winger Abelardo de la Espriella, recognised Moroccan sovereignty over Western Sahara in the presence of Morocco’s foreign minister, and three days later recognised Israeli sovereignty over the Golan Heights, which international law considers part of Syria. Morocco is Israel’s closest military partner in North Africa. No coordination needs to be alleged for these facts to matter: aligned interests move in the same direction without a masterplan.

The Spanish Right has managed to play both sides. The right-wing Partido Popular (PP) and the far-right Vox spent the first days of the crisis in studied ambiguity, then rounded on Rabat, accusing Morocco of organising the crossing and Sánchez of covering for it. At the same time, they demanded the immediate return of the people who crossed, and said nothing about the US report on the “future status” of the two exclaves.

Severity towards those in the water, deference towards the governments backing the state that put them there: that combination is not a Spanish peculiarity, and progressives elsewhere in Europe will recognise it in their own opposition.

Another way forward

Spain, and the EU with it, built a border policy on the assumption that a neighbouring autocracy would act in their interest, and that assumption has now failed at a cost of 141 lives. What replaces that policy must start from a different premise.

The first needed change concerns the bilateral relationship. Every Spanish government has accepted that Morocco’s capacity to cause harm makes a working relationship necessary, and that is a reasonable starting point. But there is a difference between maintaining a relationship and organising an entire regional policy around keeping the neighbour placated. A relationship with Rabat that can survive Ceuta must rest on reciprocity that can be verified and consequences that can be enforced: accountability for 30 July before ministerial contacts resume at their previous level, auditable conditions on border cooperation, and diversification of Spain’s regional ties so that no single neighbour holds a veto.

There have been steps in the right direction: Sánchez visited Algiers in July to mend a diplomatic rift around Western Sahara, and on 10 September, the Spanish parliament approved a law extending nationality to Sahrawis born in the territory before August 1977, when the option to claim Spanish nationality granted at the end of colonial rule expired. However positive, these steps are not yet a policy.

The second necessary change is European. When Belarus pushed people towards the Polish and Lithuanian borders in 2021, the EU called it a hybrid attack, imposed sanctions on Minsk, and adjusted its legal framework accordingly. Applying the same standard to Morocco is the minimum. Beyond that, every externalisation agreement in force should be reopened to attach conditions and consequences, because funding without them amounts to a retainer rather than cooperation.

More fundamentally, the EU should reconsider its outsourcing model. That means offering safe routes that people can use without crossing a sea or a fence: work and family-reunification visas issued in countries of origin at a scale that corresponds to actual demand rather than to political tolerance; the possibility of applying for protection at European consulates and through humanitarian corridors, so that the right to asylum does not depend on surviving the journey; and mobility agreements negotiated with the countries migrants come from, as partners with interests of their own, rather than with the transit states through which they are pushed.

A shared European system of reception and relocation that does not leave the state where people first arrive to manage the consequences alone is also needed, alongside asylum procedures that take weeks rather than years, and the regularisation of those already present, because an undocumented population is the raw material of every form of exploitation. None of this is quick or cheap, but it removes the dependency at the root: a route that does not run through Morocco is a route Morocco cannot open or close.

The third change concerns the people at the fence. Treating the migrants as the threat shifts the responsibility away from the actors who sent them. The far right across Europe supplies that narrative, in Spain with demands for immediate pushbacks, and elsewhere with the argument that people used as a weapon may be handled as one. The two interests converge at the fence, and each relies on the other to keep the argument there. Refusing that argument is the only coherent answer. The people who crossed into Ceuta on 30 July have rights that do not depend on why they crossed, and a border policy that does not begin from that fact ends by rewarding the state that used them.

These three changes are not in tension. Defending the people who cross borders and defending the borders’ integrity are the same task. Any policy that does not recognise this is a recipe for more tragedies and blackmail by autocratic powers.