Must States Control Their Borders? The Ceuta Crisis and International Law

Ceuta, a Spanish city on the North African coast with a population of approximately 80,000, experienced an unprecedented mass influx of people during the last week of July. Around 70,000 people, mostly Moroccan nationals but also persons from other countries, entered the territory irregularly. Although the precise causes remain uncertain, the immediate trigger appears to have been a wave of false information on social media claiming that Spain’s border was open. That claim was partly based on a distorted reading of a 29 June 2026 judgment of the Spanish Supreme Court concerning the so-called “rejection at the border” (rechazo en frontera) regime applicable at the Ceuta and Melilla fences. The Supreme Court had clarified that this regime did not apply to persons intercepted at sea, who instead had to be dealt with under the ordinary removal procedure under article 58 of the Spanish Foreigners Act. The judgment did not establish an open border, but it was presented as doing so online. The consequences were dramatic. Many arrivals were hungry, dehydrated and without shelter. At least 82 people died, including persons who drowned or were crushed amid the panic. Thousands of unaccompanied minors were reportedly among the arrivals. Most subsequently returned voluntarily to Morocco or were escorted back through the border crossings. By 4 August, around thousands remained in Ceuta, including approximately 1,000 unaccompanied minors (among them, girls who have reported sexual abuse).

The crisis quickly became a diplomatic and political issue. Spain described the crossings as a violation of its territorial integrity. Comparisons were drawn with Morocco’s 1975 Green March, while some politicians and commentators invoked the more recent instrumentalization of migration by Belarus against Poland. Such statements should be understood against the geographical and political context of Ceuta and Melilla, whose sovereignty Morocco has historically claimed. In that context, the characterisation of the events as affecting Spain’s territorial integrity acquires a significance that goes beyond ordinary border control and must also be read against the longstanding dispute over the status of the two “autonomous cities” (ciudades autónomas). Morocco denied orchestration, attributing the movement to economic conditions, criminal networks and disinformation. At the same time, Morocco cooperated with Spain in subsequent returns. Indeed, the Ceuta incident is not the first episode of this kind. In April 2021, Brahim Ghali, President of the Sahrawi Arab Democratic Republic and Secretary-General of the Polisario Front, was admitted to a hospital in Spain for COVID-19 treatment. His presence in Spain triggered a diplomatic crisis with Morocco, which strongly protested the decision. A few weeks later, in May 2021, Morocco relaxed controls at its border with Ceuta, resulting in the arrival of around 10,000 migrants, including a large number of unaccompanied minors. Spain deployed security forces to the border and subsequently returned many of those who had entered. The episode was widely understood as an instance of migration being used as a means of exerting political pressure on Spain.

At this stage it is not clear that the July 2026 movements were deliberately instrumentalised by Morocco. Such a claim would have to be substantiated by evidence showing that the migratory flows were intentionally facilitated or exploited as a means of exerting political pressure on Spanish decision-making. If that were established, the conduct would raise concerns not only from the perspective of Spain’s territorial integrity, but also of its political independence, insofar as migration would have been purposely used as a tool to exert pressure on Spain and influence its political choices. Should that be the case, the Ceuta events could be situated within a broader debate about whether, and to what extent, the deliberate use of migration flows for political reasons may justify limiting, suspending, or otherwise recalibrating certain legal guarantees afforded to migrants. The recent Chișinău Declaration appears to point in this direction. While reaffirming that migrants remain entitled to protection under the ECHR, the Declaration expressly recognises the phenomenon of migration instrumentalisation and calls for “new approaches” to address the challenges posed by large-scale irregular arrivals. 

This raises a difficult question: can the exceptional circumstances created by the deliberate political use of migration affect the application or interpretation of safeguards such as the prohibition of collective expulsion and access to asylum procedures? The same question arises with respect to the specific guarantees applicable to unaccompanied minors. All of these safeguards derive, directly or indirectly, from the principle of non-refoulement and, in the case of children, from the best interest of the child principle. I do not intend to examine here the scope or implications of the Declaration in this regard, as this (see, for instance, here, here, here, here, and here) and other (see here and here) blogs have already published excellent analyses of this issue. I would, however, like to build on an issue identified in one of these posts. In his analysis of the Preliminary Draft Text for the Outcome Document Containing Elements for a Political Declaration to be adopted in Chișinău, Lorenzo Acconciamessa raises the question of whether States can be said to have an obligation to protect their borders. It is true, as he argues, that border protection can be understood as a sovereign right, one that does not appear to have been questioned either by the ECtHR or by international law more generally. The starting point is straightforward: Spain and Morocco are entitled to protect their territorial integrity and control migratory movements across their borders. There is nothing inherently unlawful about preventing irregular crossings, deploying security forces to a border, dismantling smuggling networks or arranging the return of persons who have no right to remain. But can this really be characterised as an obligation? If so, what would be its legal basis, and what would its content be? These are particularly relevant questions that, to a greater or lesser extent, will have to be addressed both by the ICJ in its forthcoming judgment in Lithuania v. Belarus and by the ECtHR in a series of pending cases. 

A duty to control borders based on the obligation to combat migrant smuggling?

Could such a duty to control borders be grounded in States’ existing obligations to combat migrant smuggling? This is perhaps one of the most promising avenues to explore as it is at the core of the above-mentioned case pending before the ICJ. International law does not appear to recognise, in general terms, a freestanding obligation on States to prevent all irregular movements across their borders. The situation may be different, however, where irregular migration is facilitated by organised networks engaged in the smuggling of migrants. States have undertaken specific obligations to prevent and combat migrant smuggling, including through the Protocol against the Smuggling of Migrants by Land, Sea and Air, supplementing the United Nations Convention against Transnational Organized Crime. This raises the question whether these obligations can be understood as requiring States not merely to prosecute smugglers after the event, but also to take reasonable measures to prevent their territory from being used as a platform for organised irregular migration. If so, the obligation to combat smuggling could provide a more concrete legal basis for a broader duty to exercise effective border control.

The Ceuta crisis offers an interesting test for this proposition. Morocco is not a party to the UN Smuggling Protocol, so Spain could not simply invoke the Protocol as the source of a binding obligation on Morocco. If Morocco knowingly allowed or facilitated the mass movement of migrants towards the Spanish border, whether or not motivated by political considerations, the question would instead be whether an obligation to prevent and combat migrant smuggling can be derived from general international law. This considerably complicates the argument: not only would the existence of such a general obligation have to be established, but so too would its precise content and, crucially, whether it encompasses a duty to prevent irregular departures or otherwise to exercise effective control over migration flows. The distinction between an obligation to combat migrant smuggling and a broader obligation to control irregular migration is therefore critical. Any obligation framed in these terms must, however, be assessed not only from the State’s standpoint but also from that of the individuals concerned. A duty formulated broadly as one to prevent irregular departures would necessarily entail restricting departure, and would therefore would have to be reconciled with the right of everyone to leave any country, including their own, guaranteed by articles 13(2) UDHR, 12(2) ICCPR, article 2(2) of Protocol No. 4 to the ECHR. As Jane McAdam shows, the right to leave any country, including one’s own, is firmly established. Yet it cannot be equated with a right to permanently migrate, since it remains an “incomplete” right: unmatched by any duty of admission, it cannot be equated with a right to migrate. That asymmetry, not any weakness in the right itself, is what channels migrant movement into irregularity, since those refused lawful entry commonly attempt unlawful entry (see here). Irregular departure does not thereby strip the migrant of protection: the Protocol against the Smuggling of Migrants may permit, even require, States to prevent unauthorized exit, but its article 5 shields the smuggled migrant from criminal liability under the Protocol for having been the object of smuggling and, as the Human Rights Committee insists in General Comment No. 27, any restriction on departure must satisfy article 12(3) ICCPR and not “swallow the rule”. Returning to McAdam, one might conclude that the right to leave, though now a fundamental principle of human rights treaty law, has never been an absolute right: it has always been subject to restrictions (criminal proceedings, military service, contractual or fiscal obligations, public health) and is in practice impeded through passports and exit visas, with a “surprising consistency” in the historical gap between principle and state practice. The right thus operates within its own limits and against a persistent backdrop of state control. In sum, the right to leave is neither absolute nor self-executing: recognising it as a fundamental right settles neither the scope of permissible restrictions nor the question of entry.

This question also illustrates why the distinction between a right and an obligation to protect borders matters. If border control is merely a sovereign right, its exercise remains principally a matter of State discretion, subject to the applicable human rights constraints. If, by contrast, international law imposes a positive obligation to prevent certain forms of irregular migration, perhaps because of States’ obligations to combat migrant smuggling, the legal consequences are rather different. The question then becomes not only what States may do at their borders, but what they are required to do.

A duty to control irregular migration based on a right to safe, orderly and regular migration?

A different possible basis for understanding a duty to control irregular migration lies in the international commitment to promote safe, orderly and regular migration (see here and here). At first sight, this might appear to provide a relatively straightforward foundation for States’ efforts to prevent irregular movements. Yet such an obligation cannot meaningfully be understood in isolation from the other side of the equation: the duty to facilitate effective legal pathways for migration.

The 2018 Global Compact for Safe, Orderly and Regular Migration is particularly relevant in this respect. Its underlying premise is not simply that States should reduce irregular migration, but that migration should be made safer, more orderly and more regular through international cooperation, including by expanding regular pathways and addressing the factors that drive people towards dangerous and irregular routes. In this sense, an obligation to combat irregular migration cannot simply mean an obligation to prevent people from crossing borders irregularly. It must also entail an effort to ensure that those who have legitimate reasons to migrate have viable alternatives to irregular routes. This is particularly important where the absence of such alternatives is itself one of the factors driving irregular movements. From this perspective, the concept of safe, orderly and regular migration does not simply strengthen the case for border control; it also places that control within a broader framework of shared responsibility for creating lawful avenues of mobility. The implication for Ceuta is that the existence and content of any duty to control irregular migration cannot be assessed solely by looking at what Spain is entitled to do at its border. It must also be by considering what Spain, Morocco and the European Union are doing to make regular migration genuinely possible. 

Where, then, does the analysis leave the question of a binding obligation? Neither the Smuggling Protocol, which does not bind Morocco, nor the Global Compact, a political commitment rather than a source of binding obligations, establishes a freestanding duty to prevent irregular migration. That does not mean that general international law is silent. Other rules may be engaged on particular facts: the Corfu Channel principle, under which a State must not knowingly allow its territory to be used for acts contrary to the rights of other States, and, where migratory flows are deliberately deployed to influence another State’s political choices, the principle of non-intervention. On Milanovic’s distinction between coercion-as-extortion and coercion-as-control, it is the former that would be engaged here, since bending Spain’s sovereign decisions through the threat or infliction of harm supplies the coercive element that a prohibited intervention requires, in a way that a mere failure to control irregular departures would not. The more accurate conclusion may therefore be not that States bear a general duty of border control, but that international law imposes particular obligations in particular circumstances, whose existence and content depend on how, and to what end, a neighbouring State has acted.

When border control obligations are breached: State responsibility and enforcement

If such an obligation exists, a further question concerns the legal effects of its breach. Under the law of State responsibility, the responsible State must cease the wrongful conduct, make full reparation and, where appropriate, provide assurances and guarantees of non-repetition. These consequences, however, may not be forthcoming spontaneously, raising the question of how compliance can be secured through the mechanisms of international law. Retorsion measures, being lawful, remain available, while countermeasures may also be envisaged within the limits imposed by international law. Where the scope or even the existence of the obligation is disputed, peaceful means of dispute settlement, including judicial mechanisms, become particularly relevant, as illustrated by Lithuania v. Belarus before the ICJ. What seems clear is that, particularly in cases of instrumentalisation of migration, the consequences of a breach should be borne by the responsible State (see here). A separate, though connected, question is whether that wrongdoing can itself furnish a legal basis for relaxing the guarantees that the allegedly injured State independently owes to the individuals within its jurisdiction. It cannot, without more. An obligation such as non-refoulement is not owed to the neighbouring State, so its breach by that State neither engages a circumstance precluding wrongfulness nor authorises a countermeasure, which in any event may not affect obligations for the protection of fundamental human rights. This is a temptation to which States tend to succumb, and it seems to lie at the heart of what the promoters (see here and here) of the Chișinău Declaration appear to be suggesting.

At the present stage of international law, there is little basis for arguing that a State may invoke an exception to its obligations under the principle of non-refoulement or the best interests of the child on the ground that another State has previously failed to comply with its own obligation to control its borders in the context of combating irregular migration. Nor can N.D. and N.T. v. Spain be read as supporting such an exception. EU law had already faced this scenario. In M. A. (C-72/22 PPU), against the backdrop of Belarus’s instrumentalisation of migration, the European Court of Justice held that neither a mass-influx emergency nor irregular status could justify denying access to the asylum procedure or detaining an applicant for irregular entry alone, rejecting Lithuania’s general invocation of article 72 TFEU Though the Directives it interpreted have since been replaced by the 2024 Pact on Migration and Asylum, the logic holds: exceptional circumstances do not displace the guarantees owed to those within a State’s jurisdiction. This is particularly important because the language of border protection can easily obscure the distinction between preventing irregular entry and returning persons who have already come within a State’s jurisdiction. A State may take measures to prevent irregular crossings, cooperate with countries of origin and transit, combat smuggling networks and strengthen its external borders. But the existence of a duty to control the border cannot create an exception to non-refoulement. Even if it were established that a neighbouring State had facilitated or encouraged irregular movements to put pressure on Spain, this would not alter the individual protection obligations owed by Spain to persons within its jurisdiction. The appropriate legal response to the conduct of the State responsible for the instrumentalisation would have to be sought through the rules governing State responsibility, diplomatic and political measures, cooperation mechanisms, or other forms of international enforcement, not through the removal of safeguards protecting individuals from refoulement. Put in general terms, the wrongdoing of one State does not, in itself, provide a legal basis for relaxing the obligations that another State independently owes to individuals within its jurisdiction. Crises such as Ceuta inevitably place States and international law under considerable pressure. Yet precisely because of that pressure, decisions about the future direction of international law, and about whether some of the safeguards painstakingly developed since the 1950s should be weakened, call for careful reflection, removed from the urgency and political tension of the crisis itself.