Prof.Dr.Sibel Safi¹

Abstract

This article examines how the Court of Justice of the European Union’s judgment in Alace and Canpelli (Joined Cases C-758/24 and C-759/24) reshapes the legal parameters of extraterritorial asylum processing within the European Union. Building on Italy’s offshore cooperation with Albania for the processing and detention of asylum seekers intercepted at sea, the article argues that the safe country of origin concept and the use of accelerated border procedures cannot neutralise fundamental rights guarantees, even when codified in domestic legislation. The Alace and Canpelli judgment confirms that designations of safe countries of origin are subject to full and ex nunc judicial review and that applicants and courts must have access to the underlying country-of-origin information sources on which legislative presumptions of safety are based. Reading this jurisprudence together with ECtHR case law on non-refoulement, arbitrary detention and effective remedies—including Saadi v United Kingdom and S.H. v Malta—the article shows that extraterritorial schemes such as the Italy–Albania agreement risk producing “accountability gaps” and chain refoulement in violation of EU law, the ECHR and the 1951 Refugee Convention. It contends that offshore processing arrangements displace, rather than share, protection responsibilities and normalise legally opaque forms of migration governance at the EU’s periphery. The article concludes that Alace and Canpelli sets important limits on Member States’ ability to insulate externalisation policies from judicial scrutiny, but that stronger doctrinal and institutional safeguards are needed to prevent the transformation of extraterritorial territories into spaces of diminished rights.

Anahtar kelimeler

Extraterritorial asylum processing; safe country of origin; accelerated border procedures;

Italy–Albania agreement; judicial review

Özet

Bu makale, Avrupa Birliği Adalet Divanı’nın Alace and Canpelli kararının (Birleştirilmiş Davalar C-758/24 ve C-759/24) Avrupa Birliği içinde ülke dışı sığınma prosedürlerinin hukuki çerçevesini nasıl yeniden şekillendirdiğini incelemektedir. Denizde kurtarılan sığınmacıların sığınma başvurularının İtalya’nın Arnavutluk ile yürüttüğü offshore iş birliği kapsamında işleme alınması ve gözaltında tutulması uygulamasından hareketle, makale güvenli menşe ülke kavramının ve hızlandırılmış sınır prosedürlerinin—ulusal mevzuatta düzenlenmiş olsa dahi—temel hak güvencelerini ortadan kaldıramayacağını savunmaktadır. Alace and Canpelli kararı, güvenli menşe ülke belirlemelerinin tam ve ex nunc yargısal denetime tabi olduğunu ve başvuranlar ile mahkemelerin, güvenli ülke varsayımının dayandığı ülke-menşeli bilgi kaynaklarına erişiminin sağlanması gerektiğini teyit etmektedir. Bu içtihat, AİHM’nin geri gönderme yasağı, keyfi gözaltı ve etkili başvuru hakkına ilişkin kararları—Saadi/Birleşik Krallık ve S.H./Malta dâhil—ile birlikte okunduğunda, İtalya–Arnavutluk anlaşması gibi ülke toprakları dışı modellerin AB hukuku, AİHS ve 1951 Mülteci Sözleşmesi ile bağdaşmayan hesap verebilirlik boşlukları ve zincir refoulement riskleri doğurduğu ortaya çıkmaktadır. Makale, bu tür offshore düzenlemelerin koruma sorumluluklarını paylaşmak yerine başka yere kaydırdığını ve AB’nin çevre bölgelerinde hukuken göç yönetişiminde gri bir alan biçimini normalleştirdiğini ileri sürmektedir. Sonuç olarak çalışma, Alace and Canpelli’nin üye devletlerin dışsallaştırma politikalarını yargısal denetimden yalıtma kapasitesini önemli ölçüde sınırladığını, ancak ülke dışı alanların hakların zayıfladığı mekânlara dönüşmesini önlemek için daha güçlü doktrinsel ve kurumsal güvencelere ihtiyaç duyulduğunu savunmaktadır.

Anahtar kelimeler

Ülke dışı sığınma prosedürleri; güvenli menşe ülke; hızlandırılmış sınır prosedürleri; İtalya–

Arnavutluk anlaşması; yargısal denetim

Introduction

Offshore asylum processing has moved from the margins to the centre of European migration governance. Italy’s 2023 agreement with Albania, which establishes two facilities on Albanian territory to process people rescued at sea by Italian authorities, exemplifies this trend. Under the agreement, asylum seekers apprehended in the central Mediterranean are transferred to de facto detention centres in Gjadër and elsewhere in Albania, formally under Italian jurisdiction but physically controlled by Albanian authorities². They remain outside Italian territory for the duration of the procedure, with limited access to courts, lawyers and independent monitoring. This “territorial decoupling” of physical control and legal accountability creates precisely the kind of legal grey zones in which fundamental rights are most fragile³.

The Italy–Albania arrangement is part of a broader shift towards externalisation in EU asylum policy. By relocating procedures to countries with weaker institutional safeguards, Member States seek to control arrivals, accelerate removals and symbolically “regain control” over borders. Yet, as earlier externalisation experiments—such as the EU–Türkiye Statement and various North African cooperation frameworks—have shown, such strategies tend to displace protection responsibilities onto structurally weaker states while maintaining the decision-making power of the EU and its Member States³. The result is not shared responsibility but responsibility-shifting.

Within this landscape, the Court of Justice of the European Union’s judgment in Alace and Canpelli (Joined Cases C-758/24 and C-759/24, 1 August 2025)⁴ occupies a pivotal position. The case concerned two Bangladeshi nationals who were intercepted by Italian authorities at sea and transferred to the Gjadër detention centre in Albania under the Italy–Albania Protocol. Their asylum claims were examined in an accelerated border procedure and rejected as unfounded on the basis that Bangladesh had been designated as a “safe country of origin” by Italian legislation adopted in October 2024. The applicants challenged not only the individual decisions but also the underlying legislative designation, arguing that they and the reviewing court had no access to the information sources on which the presumption of safety rested.

The referring court asked whether EU law allows a Member State to insulate such a legislative designation from effective judicial review and whether a state can treat a country as “safe” despite acknowledged risks for certain categories of persons. The Court’s answer was clear: while Member States may designate safe countries of origin by legislative act, such designations must remain subject to a full and ex nunc examination by the competent court under Article 46(3) of the Procedures Directive, read in the light of Article 47 of the Charter.⁵

The CJEU held that European Union Member States cannot classify a country as “safe” unless that country offers effective protection to all persons within its territory. This judgment directly challenges the Italian legal framework, which permits the designation of a country as “safe” while allowing for limited exceptions for certain minority groups.⁶Furthermore, the Court emphasized that any such designation must be open to effective and transparent judicial review, and should be grounded in publicly accessible and verifiable sources of information—both for the asylum applicant and for the judicial body responsible for reviewing the claim.

The Court of Justice of the European Union clarified in its judgment that European Union law does not, in principle, prevent a Member State from designating a third country as a “safe country of origin” through legislative means, provided that such a designation is subject to effective judicial oversight.⁷ The judicial review in question must assess compliance with the substantive criteria set out in Annex I of Directive 2013/32/EU, particularly in cases where asylum applications are rejected under accelerated procedures triggered by such designations.⁹

The Court further held that the evidentiary sources underlying a designation must be adequately accessible to both the applicant and the competent judicial authority. This requirement is essential to ensuring effective judicial protection—allowing applicants to meaningfully assert their rights and enabling courts to fully exercise their review powers. Moreover, domestic courts are permitted to consider information they have independently gathered, on the condition that such material is demonstrably reliable and that both parties are given an opportunity to comment on it within adversarial proceedings.⁸

Importantly, the Court stressed that, pending the adoption of the new EU Asylum Procedures

Regulation, a Member State may not designate a country as “safe” if it fails to meet the substantive protection criteria for certain categories of persons. This requirement underscores the obligation to guarantee that designations reflect a holistic and inclusive evaluation of risk across all population groups.

Judicial protection would be illusory if courts and applicants could not access the sources of information underpinning the presumption of safety. Moreover, Member States may not designate a country as safe if it fails to meet the material conditions in Annex I to Directive 2013/32 for particular categories of persons; where limited categories are at risk, those groups must be expressly excluded from the presumption.⁹This jurisprudence has direct implications for extraterritorial processing models. Offshore schemes frequently rely on legal shortcuts: broad safe-country presumptions, automatic detention and compressed procedures conducted at borders or in transit zones. Italy’s use of Albania has precisely those features. The agreement envisages automatic and potentially prolonged detention of all persons disembarked in Albania, combining the 28-day border procedure with other detention regimes that may extend up to 18 months.¹⁰ Such cumulative detention risks arbitrariness and clashes with both Article 5 ECHR and Article 31 of the 1951 Refugee Convention, as interpreted in Saadi v United Kingdom and subsequent ECtHR case law.¹¹ At the same time, distance from Italian territory—both geographic and symbolic—makes access to legal assistance, interpretation and effective remedies highly precarious.¹⁴

Against this background, the central question of this article is whether Alace and Canpelli meaningfully constrains the capacity of Member States to use extraterritorial arrangements to evade their protection obligations. More specifically, it asks:

1. To what extent does the judgment re-affirm the justiciability of safe country of origin designations and require transparency in the production and use of country-of-origin information?

2. How does this interact with ECtHR standards on non-refoulement, arbitrary detention and effective remedies in cases such as S.H. v Malta, Gebremedhin v France and Jabari v Turkey?¹²

3. What are the implications for offshore processing centres located outside the EU’s territory but under Member State “jurisdiction”, such as Italy’s facilities in Albania?

Methodologically, the article proceeds in three steps. First, it draws on the detailed doctrinal and normative analysis of the Italy–Albania agreement to reconstruct the extraterritorial architecture of the scheme, focusing on jurisdiction, detention and access to remedies. Second, it undertakes a close reading of Alace and Canpelli, situating the judgment within the broader evolution of the safe country of origin concept under EU law and highlighting the Court’s insistence on transparency and ex nunc judicial control. Third, it places the Luxembourg judgment in dialogue with Strasbourg case law, showing the convergences in their understanding of effective protection and the residual risks that remain for applicants held in offshore centres.

The article argues that Alace and Canpelli closes some of the most troubling loopholes in domestic attempts to legislate away judicial scrutiny over safe-country designations. By requiring access to sources, insisting on full judicial review and limiting the possibility of designating countries in which certain categories of persons remain at risk, the Court strengthens the procedural backbone of EU asylum law. However, without parallel limits on extraterritorial detention, distance-based obstacles to legal assistance and the use of cooperative agreements to fragment responsibility between Member States and third countries, offshore processing will continue to generate spaces of attenuated rights at Europe’s borders. The Italy–Albania agreement thus remains a test case—not only for the reach of EU jurisdiction, but for the future geography of refugee protection in Europe.

I. Safe Country of Origin Designations and Effective Judicial Protection

The safe country of origin concept was originally conceived as a procedural tool to streamline manifestly unfounded claims while preserving individualised assessment. Annex I to Directive 2013/32 defines a safe country of origin as one in which there is “generally and consistently” no persecution, torture or serious harm, and where the rule of law and fundamental rights are respected.¹³ In practice, however, the concept has often been used to justify accelerated procedures with reduced safeguards, particularly at borders and in transit zones. The Italian legislative designation of Bangladesh as a safe country of origin in 2024 is a paradigmatic illustration: it triggered automatic use of accelerated border procedures for Bangladeshi nationals, including those detained in Albania under the Italy–Albania Protocol.¹⁴

In Alace and Canpelli, the CJEU clarified that legislative designations do not escape the reach of Article 46(3) of the Procedures Directive. The Court recalled that applicants subject to accelerated border procedures must enjoy an effective remedy capable of a “full and ex nunc examination of both facts and points of law”. Building on earlier decisions such as Samba Diouf and CV,¹⁵ the judgment emphasises that this examination necessarily includes the lawfulness of the safe-country designation itself, including compliance with the substantive criteria of Annex I. If legislation were insulated from review, Article 46(3) would be deprived of practical effect. A central innovation of the judgment lies in its treatment of country-of-origin information (COI). The Court accepts that Directive 2013/32 does not expressly oblige Member States to publish the sources underpinning safe-country designations. Yet, reading Articles 36, 37 and 46(3) together and in the light of Article 47 of the Charter, it concludes that effective judicial protection requires applicants and reviewing courts to have access to those sources.¹⁶ Without such access, applicants cannot meaningfully challenge the presumption of safety and courts cannot verify the origin, authoritativeness, reliability, topicality and completeness of the information relied upon by the legislature. The Advocate General underlines that disclosure enhances both the credibility of the presumption and the efficiency of procedures: robust, transparent COI (Country of origin information) constrains arbitrary decision-making while allowing clearly unfounded claims to be processed swiftly.

The Court further accepts that Member States enjoy a measure of discretion in designating safe countries of origin, particularly in contexts of high migratory pressure. However, this discretion is not unlimited. Where a Member State knows that “limited but clearly identifiable categories of persons” face risks of persecution or serious harm, it may only designate the country as safe if those categories are expressly excluded from the presumption and processed under the ordinary procedure. In other words, general safety cannot be reduced to a legal fiction constructed on the back of large-scale, structural exclusion.

This doctrinal clarification has two immediate consequences for extraterritorial processing. First, it makes clear that Member States cannot hide behind legislative labels when examining claims in offshore centres. Even when procedures are conducted in Albania, Italian courts must be able to review both the individual decision and the underlying presumption that the applicant’s country of origin is safe. Second, the requirement of access to COI sources complicates efforts to compress time frames and restrict legal assistance in extraterritorial facilities. If applicants are detained in remote locations, lack effective legal representation and do not receive the COI underlying the presumption of safety, the accelerated procedure risks degenerating into a purely formal exercise, incompatible with Articles 46(3) Procedures Directive and 47 Charter.²⁰

II. The Italy–Albania Agreement: Legal Foundations and Implications for Extraterritorial Asylum Processing

The Italy–Albania agreement is a paradigmatic example of how extraterritorial asylum processing is used to reconfigure the territorial foundations of protection. Signed in November 2023 and ratified in early 2024, the agreement provides for the establishment of two migration centres on Albanian territory, intended to process asylum applications of individuals intercepted at sea by Italian authorities. While the Italian government emphasises that the centres remain under Italian jurisdiction, Albanian police are responsible for security and return operations, including in situations where persons abscond or lose their asylum claim. This hybrid model deliberately

²⁰ European Union (2013). Directive 2013/32/EU of the European Parliament and of the Council of 26 June 2013 on common procedures for granting and withdrawing international protection (recast). Official Journal of the European

Union, L 180, 60–95. (Article 46(3) on full and ex nunc judicial review.)

European Union (2012). Charter of Fundamental Rights of the European Union. Official Journal of the European

Union, C 326, 391–407. (Article 47 on the right to an effective remedy and a fair trial.)

separates formal jurisdiction from physical control, creating jurisdictional ambiguity at precisely the point where rights must be enforceable.

The centres are expected to operate as de facto detention facilities. Under the combined effect of the border asylum procedure and existing Italian detention legislation, individuals disembarked in Albania may be subjected to automatic and cumulative forms of detention lasting up to eighteen months. The agreement contemplates a 28-day detention period linked to the accelerated border procedure, which can be layered with administrative detention for removal.¹⁷ This stacking of detention regimes undermines safeguards designed to limit deprivation of liberty and risks transforming the centres into spaces of prolonged confinement without meaningful individual assessment.¹⁸ Moreover, the automatic and prolonged detention of individuals disembarked in Albania appears to contravene Article 31(1) of the 1951 Geneva Convention, which prohibits penalizing refugees for their irregular entry, provided they present themselves without delay and demonstrate valid grounds for seeking protection.¹⁹ The cumulative application of various detention regimes—ranging from identification procedures to administrative detention for removal—amounts to a system where asylum seekers may be deprived of their liberty for extended periods without individualized judicial assessment.²⁰ Such practices contradict the ECtHR’s guidance in Saadi v. United Kingdom, which emphasized that immigration detention must not only pursue a legitimate aim but also comply with principles of necessity, proportionality, and good faith.²¹

Italy’s reliance on Albania also aggravates existing obstacles to access to justice in the domestic asylum system. Even within Italy, asylum seekers face financial, linguistic and procedural hurdles in accessing courts and contesting detention order. These difficulties are amplified when applicants are held in remote facilities on Albanian soil, dependent on remote communication technologies, and removed from the ecosystems of lawyers, judges, civil society organisations and interpreters that ordinarily support asylum procedures. As Amnesty International and other observers have noted, remote legal assistance and videoconference interviews increase the risk of misunderstandings, poor interpretation and truncated narratives, particularly for vulnerable applicants.²² The extraterritorial location thus transforms formal rights into fragile, conditional entitlements.

The agreement is also emblematic of contemporary externalisation politics. It is framed by Italian officials as a “pilot” project for Europe, signalling its potential replication by other Member States.²³ This signals a dangerous shift toward the normalization of asylum externalization as a policy template, particularly appealing to governments seeking to restrict access to asylum without formally breaching EU or international law. However, as demonstrated by ECtHR jurisprudence in cases such as Hirsi Jamaa v. Italy and Sharifi and Others v. Italy and Greece, outsourcing migration control responsibilities does not absolve states of their human rights obligations.²⁴ If unchecked, the Italy–Albania model may set a precedent that entrenches legal fragmentation, offloads protection duties to structurally weaker countries, and accelerates the erosion of Europe’s normative asylum framework.

Set against the background of earlier arrangements—such as the EU–Türkiye Statement and cooperation with Morocco, Tunisia, Libya, Mauritania and Egypt—the Italy–Albania deal extends a governance model that combines informal instruments, legal opacity and structural asymmetry between contracting parties. Albania, an EU candidate state with limited asylum infrastructure and experience with large-scale reception, becomes a buffer zone that absorbs the human consequences of European border control while lacking commensurate oversight or capacity. In this sense, the agreement is not merely a bilateral experiment but a laboratory for a new geography of European asylum.

III. Facts in Alace and Canpelli (C-758/24 and C-759/24)²⁵

The Alace and Canpelli judgment directly arises from the operationalisation of the Italy– Albania Protocol. The applicants, two Bangladeshi nationals, were rescued at sea by Italian authorities and transferred to the Gjadër detention centre in Albania. Their applications for international protection were examined under an accelerated border procedure on the basis that

Bangladesh had been designated as a safe country of origin by a legislative act adopted in October 2024. In line with that legislation, their claims were rejected as unfounded on the ground that their country of origin was presumed safe. The applicants, however, were unable to rebut this presumption because they and the reviewing court did not have access to the country-of-origin information (COI) underlying the designation.

The referring court raised several questions about the compatibility of this framework with EU law. It asked, first, whether Articles 36 and 37 of Directive 2013/32 permit a Member State to designate a safe country of origin by legislative act, and, second, whether such a designation can be insulated from effective judicial review when the sources of information on which it is based are not disclosed.²⁶

The CJEU held, in essence, that EU law does not preclude legislative designations of safe countries of origin, provided that such designations fully respect the obligations and objectives of Directive 2013/32.²⁷ A Member State may enact a law listing safe countries, but this law implements EU law and therefore remains subject to the principle of primacy and to effective judicial protection under Article 47 of the Charter. Crucially, the legislative nature of the act cannot remove the designation from the scope of judicial review; otherwise Article 46(3) of the Directive, which requires a “full and ex nunc examination of both facts and points of law”, would be deprived of real effect.

On the transparency of COI, the Court acknowledged that Directive 2013/32 does not expressly require Member States to disclose the sources used to construct safe-country lists. Nonetheless, read in light of Article 47 of the Charter, Articles 36, 37 and 46(3) demand that both applicants and courts have access to those sources. Without access, applicants cannot understand the reasons for rejection or effectively rebut the presumption of safety, and courts cannot verify the origin, authoritativeness, reliability, relevance, timeliness and completeness of the information. Disclosure thus becomes a functional precondition for the effectiveness of the judicial remedy.

The Court further clarified that Member States retain a degree of discretion to designate a country as safe even if one or more limited and clearly identifiable categories of persons are at risk, but only if these categories are expressly excluded from the presumption of safety and their claims are processed under the ordinary procedure. Generalised or extensive exceptions—encompassing numerous groups—would indicate systemic deficiencies and are incompatible with Annex I to the Directive. The judgment concludes by affirming that where sources are not disclosed, national courts must themselves rely on reliable public information, including from the actors listed in Article 37(3),²⁸ to assess the lawfulness of the designation. In sum, Alace and Canpelli reinforces that safe-country designations are rebuttable, reviewable and embedded within a framework of robust procedural guarantees.

A. Legislative Lists and the Limits of Domestic Autonomy

A core issue before the CJEU was whether a safe-country designation enacted through legislation enjoys a heightened presumption of validity or reduced judicial scrutiny. The Court rejected any such distinction. According to the CJEU, the manner in which a Member State formalises the designation—administratively or through legislation—does not diminish the obligation to ensure full judicial review. Article 46(3) of the Procedures Directive requires national courts to reassess both facts and law ex nunc, including the substantive correctness of the safecountry designation itself.

In other words, national legislation cannot create an unreviewable presumption. This principle is particularly significant in externalised procedures, where access to legal counsel and courts is already compromised, as illustrated by Italy’s use of Albanian territory for processing and detaining applicants. The CJEU also addressed the informational basis of safe-country designations. Although Directive 2013/32 does not explicitly mandate disclosure of country-of-origin information (COI), the Court held that Articles 36, 37 and 46 must be read with Article 47 of the Charter: both the applicant and the reviewing court must have access to the material used to support a safecountry listing.

Lack of access to the underlying reports or data makes it impossible:

• for applicants to contest the presumption, or

• for courts to verify the reliability, timeliness or relevance of the sources used.

This requirement is especially critical in offshore or remote facilities where communication barriers and restricted contact with lawyers already pose structural obstacles to effective remedies.

The CJEU further clarified that a third country may still appear on a national safe-country list even if certain identifiable groups face potential persecution or harm. However, this is only permissible where such groups are expressly carved out of the presumption and their claims are channelled into the ordinary asylum procedure. This prevents Member States from using the safecountry doctrine as a blanket, unqualified tool to accelerate and streamline procedures when specific vulnerabilities are clearly documented.

1. The Strasbourg Perspective: Individual Risk as a Non-Negotiable Standard

The CJEU’s position aligns closely with Strasbourg jurisprudence. In S.H. v Malta,²⁹ the ECtHR examined a Bangladeshi journalist’s asylum claim processed in an accelerated procedure heavily influenced by Malta’s safe-country list. The Court identified significant procedural defects—including superficial reasoning, absence of effective legal assistance, delayed notifications and lack of automatic suspensive effect—and held that removal without a fresh assessment would violate Articles 3 and 13 ECHR.

Similarly, in D.L. v Austria, the ECtHR³⁰ explicitly stated that the safe-country classification is irrelevant when evaluating an individual’s risk upon return. The protection duty is non-delegable.

Across both courts, certain principles converge:

• Safe-country status is merely a rebuttable presumption (§ 48; § 91).

• Risk assessment must be individualised and evidence-based.

• Remedies must be genuinely effective and, in both systems, accompanied by an automatic suspensive effect of appeals (§ 79; Article 46(5)).

• Adequate access to information and reasoning is indispensable (§ 80; § 86–87).

• Authorities must examine risk proactively when alerted to relevant facts, even if incompletely articulated by the applicant, consistent with F.G. v Sweden.³⁵

When these standards are applied cumulatively, the implications for extraterritorial or accelerated border procedures are far-reaching. Requirements such as full ex nunc review, disclosure of COI, automatic suspensive effect, and proactive assessment of risk impose strict limits on any attempt to rely on safe-country lists as the basis for fast-tracked rejection.

In facilities like Italy’s Albanian centres, where procedural rights are already attenuated due to distance, detention, and limited legal access, the proper application of Alace and Canpelli becomes particularly challenging. Without structural safeguards, the doctrine risks transforming into a mechanism that obscures rather than illuminates protection needs.

2. How Safe Is the Safe Country of Origin Doctrine

The combined lessons of Alace and Canpelli and S.H. v Malta demonstrate that the safecountry concept is considerably less stable and less operationally useful than its title suggests. Its viability hinges entirely on transparency, individualized assessment, and effective judicial oversight. Where these conditions are weak—such as in externalised, detention-heavy or accelerated procedures—the doctrine becomes precarious and risks facilitating summary rejection rather than ensuring genuine protection.

The safe-country concept is therefore “safe” only to the extent that Member States maintain the procedural and informational guarantees required by both Luxembourg and Strasbourg. When these safeguards erode, the doctrine becomes not a tool of efficiency but a potential conduit for rights violations.

3. Jurisdiction, Detention and Non-Refoulement in Offshore Centres

The Italy–Albania agreement sits at the intersection of several regimes: EU asylum law, the ECHR and the 1951 Refugee Convention. Its most troubling features concern jurisdiction, detention and non-refoulement. Although Italy asserts that individuals transferred to Albanian centres remain under Italian jurisdiction, the practical displacement of applicants outside EU territory, and the role of Albanian police in enforcing security and return measures, create a fragmented responsibility structure.³¹ This fragmentation risks generating “accountability gaps”, where neither Italy nor Albania assumes full responsibility for ensuring access to remedies or preventing refoulement.

From the perspective of detention, the automatic and prolonged confinement envisaged by the agreement is difficult to reconcile with Article 5 ECHR and Article 31 of the Refugee Convention³². The ECtHR has consistently held that detention of asylum seekers must not be arbitrary and must be closely linked to a legitimate aim, pursued in good faith and for no longer than strictly necessary. This provision assumes that asylum seekers should not be penalized for their mode of entry nor be isolated from the host society where they seek international protection.³³ In the Italian context, asylum seekers already face considerable obstacles—financial, linguistic, and procedural—in accessing judicial remedies to appeal negative asylum decisions and contest detention orders.³⁴

A legal framework that neither guarantees a right to lawful entry nor ensures the feasibility of lawful access to another state’s territory places refugees in a legal limbo, indefinitely deferring their access to the substantive rights enshrined in the Geneva Convention.³⁵This risk of arbitrary detention is exacerbated by the situation of asylum seekers transferred by Italian authorities to detention centres in Albania, where their access to procedural rights is even further constrained.³⁶Linked to the legality of detention for removal purposes is the principle of nonrefoulement.

Although Albania is a signatory to both the 1951 Refugee Convention and its 1967 Protocol,³⁷ its national asylum framework remains underdeveloped and lacks the institutional capacity required to implement Convention standards effectively. In practice, Albania has received relatively low numbers of asylum applications and has limited experience with large-scale reception or long-term refugee integration.³⁸

In Saadi v United Kingdom, the Court accepted that states may detain individuals who enter irregularly and seek asylum, but insisted that detention be a measure of last resort, subject to individualised assessment and proportionality. By contrast, the Italy–Albania scheme envisages routine detention for all persons disembarked in Albania, combining border procedures and removal-oriented detention in a way that can extend well beyond a year. Such cumulative detention, particularly when coupled with uncertain prospects of removal, is at high risk of being deemed arbitrary.

Yet under the Italy–Albania model, individuals intercepted at sea are not only removed from EU territory but are subjected to automatic detention in facilities they cannot leave, with limited access to asylum procedures and judicial review. In practice, this amounts to a form of collective punishment for attempting to seek protection by sea, rather than an assessment of individual protection needs.³⁹ Non-refoulement concerns are further aggravated by the possibility of “chain refoulement”. ECtHR case law, including Gebremedhin v France and Jabari v Turkey, underlines that remedies against removal must be practical and effective, not illusory.

4. Luxembourg–Strasbourg Dialogues: Alace and Canpelli and S.H. v Malta

The Court of Justice’s reasoning in Alace and Canpelli does not develop in isolation; rather, it forms part of a broader judicial dialogue between Luxembourg and Strasbourg concerning the procedural limits of accelerated examinations and the use of safe country of origin designations. In this regard, the European Court of Human Rights’ judgment in S.H. v Malta provides an important normative and doctrinal counterpart.⁴⁰

In S.H. v Malta, the applicant—a Bangladeshi journalist—had his asylum claim examined under an accelerated procedure premised on Bangladesh’s designation as a safe country of origin. The Strasbourg Court identified several structural deficiencies in the Maltese procedure, including the absence of effective legal assistance, a cursory credibility assessment, failure to consider relevant country-of-origin information, and a stereotyped judicial review completed within twentyfour hours.⁴¹The Court concluded that such shortcomings deprived the remedy of its practical effectiveness and held that removing the applicant without a sufficiently rigorous reassessment would violate Articles 3 and 13 ECHR.⁴²

The parallels with Alace and Canpelli are striking.⁴³ Both courts underscore that accelerated procedures based on presumptions of safety must remain compatible with robust procedural safeguards, and that national safe-country designations cannot replace the obligation to conduct an individualised and up-to-date assessment.⁴⁴

The two courts also converge on the importance of information and transparency. The ECtHR has criticised inadequate reasoning and lack of proper notification of decisions, emphasising that applicants must understand why their claims were rejected and on what basis the safe-country presumption is applied. The CJEU complements this by requiring access to the COI sources underlying safe-country lists, both for applicants and courts. In both systems, the effectiveness of the remedy hinges on the availability of reasons and information; without them, judicial review risks degenerating into a ritualistic confirmation of administrative decisions.

Finally, both courts recognise a proactive element in the duty of protection. The ECtHR has held that when a state becomes aware of facts suggesting a real risk of ill-treatment, authorities must assess that risk of their own motion, even if the applicant does not articulate it fully. In Alace and Canpelli, the CJEU similarly stresses that national courts must, on the basis of information in the file and other reliable sources, raise potential non-compliance with Annex I criteria ex officio.⁴⁵ This convergence suggests an emerging European standard: safe-country presumptions cannot be mechanistically applied; they must be embedded within an active, inquisitorial approach to risk.

For offshore processing models, this dialogue is particularly significant. It implies that courts reviewing decisions taken in extraterritorial centres must not only assess individual claims but also scrutinise the structural assumptions—safe-country lists and externalisation agreements—on which those decisions rest. In contexts such as Italy’s centres in Albania, where applicants are geographically and procedurally remote from reviewing courts, ensuring that this standard is met will be a major challenge.

IV. From Responsibility-Sharing to Responsibility-Shifting

The Italy–Albania arrangement is often presented as a pragmatic contribution to “burdensharing” within a stressed European asylum system. Yet its design and effects more closely resemble responsibility-shifting. Albania, a country with limited experience in large-scale asylum processing and modest institutional capacity, is tasked with hosting detention facilities, providing security and managing aspects of return, while Italy retains primary decision-making power. This asymmetry reflects and reproduces broader global patterns in which wealthier states externalise the human and political costs of migration control to less powerful partners.

Within the EU, this model builds upon a decade of externalisation policies. The Italy–Albania agreement adds a new layer by combining externalised detention and procedure with the formal retention of Member State jurisdiction.

These developments mark a broader trend within EU policy: the rise of informal, legally opaque cooperation mechanisms that carry significant implications for the rights of asylum seekers. The EU’s continued pursuit of external migration control—evident in deals with countries such as Morocco, Tunisia, Libya, Mauritania, and Egypt—reflects a migration governance model increasingly untethered from transparent legal obligations and grounded instead in political expediency.⁴⁶This dynamic was further institutionalized through the establishment of the European Pact on Migration and Asylum, which strengthens border control frameworks and seeks to preempt future “crises” by solidifying, externalized deterrence mechanisms.⁵²

Within this context, the Italy-Albania agreement emerges as a paradigmatic example of contemporary external processing. While framed as a practical response to migration management, the agreement raises multiple human rights concerns. First, the relocation of asylum procedures outside EU territory results in a legal grey zone where the procedural and substantive rights of applicants become precarious. Such arrangements deviate from the core assumption of EU asylum law that protection claims should be processed within the jurisdictional framework of the Member States. This is especially evident in the erosion of rights enshrined in Article 9(1) of the EU Asylum Procedures Directive, which stipulates that applicants must be allowed to remain in the territory of the Member State during the examination of their claim.⁴⁷

Second, the Italy-Albania agreement reflects broader challenges of extraterritorial human rights protection. If asylum seekers processed in Albania are not afforded the same procedural guarantees as those within Italy, this raises questions of discriminatory treatment and undermines the universality of human rights. Moreover, it invites scrutiny over the extent to which fundamental rights obligations apply extraterritorially—especially when detention and removal practices occur under the authority of a third state.⁴⁸

Third, comparative legal analysis illustrates that external processing models in other jurisdictions have frequently led to systemic human rights violations, particularly concerning arbitrary detention, lack of access to effective remedies, and non-refoulement breaches.⁴⁹ The ItalyAlbania arrangement bears striking similarities to the United Kingdom’s controversial Rwanda scheme, which has already been subject to strong condemnation by the European Court of Human Rights (ECtHR).⁵⁰ Whether the ECtHR will take a similarly critical stance toward the Italy-Albania agreement remains to be seen, particularly as Albania is itself a party to the European Convention on Human Rights.

Finally, differential treatment among asylum seekers—those processed in Albania versus those in Italy—risks violating the principle of equality before the law and may also contravene obligations under the Charter of Fundamental Rights of the European Union.⁵¹ The lack of judicial review mechanisms, the informal character of the agreement, and the potential violation of key procedural safeguards all suggest that the Italy-Albania arrangement could serve as a precedent for the normalization of extraterritorial asylum governance that is both legally and ethically problematic.

If unchallenged, this could normalise a spatial division of labour in which protection responsibilities are progressively displaced to the EU’s periphery or beyond, while the core retains strategic control. In this sense, externalisation is not simply a technical policy instrument but a transformative project that re-draws the moral and legal boundaries of Europe.

Normatively, Alace and Canpelli offers an important, but incomplete, corrective to the risks inherent in externalisation. On the one hand, the judgment strengthens the internal coherence of EU asylum law by affirming that safe-country designations—whether legislative or administrative— remain subject to robust judicial scrutiny, transparency requirements and individualised assessment.⁵² It thus limits Member States’ ability to weaponise presumptions of safety to justify accelerated procedures that sidestep fundamental guarantees.

On the other hand, the judgment does not directly address the extraterritorial dimension of schemes such as the Italy–Albania agreement. It does not, for instance, set clear limits on the use of automatic detention in offshore centres, nor does it articulate a comprehensive doctrine of extraterritorial jurisdiction under EU law. These gaps leave room for Member States to continue experimenting with territorial decoupling, as long as they formally maintain jurisdiction and provide some form of judicial review. Without additional safeguards on detention, access to legal assistance and non-refoulement, offshore centres risk remaining spaces of attenuated protection.

A rights-compatible model of extraterritorial processing would require, at minimum, five conditions: (1) unequivocal recognition that EU and Member State human rights obligations apply wherever they exercise effective control; (2) strict limits on detention, with a presumption in favour of liberty and robust judicial oversight; (3) guaranteed access to high-quality legal assistance, interpretation and information, including COI sources; (4) automatic suspensive effect of appeals against negative decisions and removal orders; and (5) independent monitoring by international and civil society actors with full access to facilities.⁵³ The Italy–Albania agreement, as currently designed, falls short on each of these counts.

Conclusion

The Alace and Canpelli judgment marks a significant step in clarifying the legal parameters of safe country of origin designations and effective judicial protection within EU asylum law. By affirming that legislative safe-country lists are subject to full and ex nunc judicial review, requiring access to COI sources, and limiting the use of presumptions where specific categories of persons remain at risk, the Court reinforces the procedural integrity of the asylum system. These clarifications have direct implications for extraterritorial processing schemes, where accelerated procedures and distance from courts heighten the risk of violations.

Yet the Italy–Albania agreement demonstrates that externalisation continues to push refugee protection to the edges of legal imagination. Through territorial decoupling, cumulative detention and jurisdictional ambiguity, the agreement creates a space in which rights are formally recognised but practically constrained. While Alace and Canpelli constrains some of the most problematic uses of the safe country of origin concept, it does not fully address the structural dangers of offshore processing.

Ultimately, the future of European asylum law will depend on whether courts and policymakers treat extraterritorial space as part of the legal order, rather than as an exception to it. If offshore centres are allowed to function as laboratories of diminished rights, the normative foundations of the 1951 Refugee Convention, the ECHR and the EU Charter risk being gradually eroded. If, however, Alace and Canpelli is read as a starting point for re-asserting jurisdiction, transparency and individualised protection in all spaces under European control, it may yet open a path towards a more principled and accountable asylum governance.

References

Amnesty International, Italy–Albania Agreement: A Dangerous Precedent (2024) EUR30/7102/2024 https://www.amnesty.org/en/documents/eur30/7102/2024/en/ accessed 10 September 2025.

Amnesty International, ‘The Italy–Albania Agreement on Migration: Pushing Boundaries, Threatening Rights’ (19 January 2024) https://www.amnesty.org/en/documents/eur30/7587/2024/en/ accessed 4 August 2025.

Anker D, ‘The US Border, Migration and Human Rights’ (2018) 30(1) International Journal of Refugee Law 1.

ASGI (Associazione per gli Studi Giuridici sull’Immigrazione), Legal Opinion on the Italy– Albania Agreement (2024) https://www.asgi.it accessed 5 August 2025.

Broerse I, The Italy–Albania Agreement: Externalising Asylum Procedures in Violation of Human Rights (ACMRL Migration Law Series No 26, 2024) https://doi.org/10.71881/51abe5cb92a2-4de9-8395-9323bdc7f75c.

Carrera S and Vosyliūtė L, EU External Migration Policies: A Comparative Perspective (CEPS 2019).

Charter of Fundamental Rights of the European Union [2012] OJ C 326/391.

CJEU, Alace and Canpelli v Italy (Joined Cases C-758/24 and C-759/24, Judgment of 1 August 2025).

CJEU, C.V. v Ministerstvo vnitra (Case C-63/15, Judgment of 18 December 2015).

CJEU, M. Samba Diouf v Ministre du Travail, de l’Emploi et de l’Immigration (Case C-69/10, Judgment of 28 July 2011).

Directive 2011/95/EU on standards for the qualification of third-country nationals or stateless persons (Qualification Directive).

Directive 2013/32/EU of the European Parliament and of the Council of 26 June 2013 on common procedures for granting and withdrawing international protection, OJ L 180/60.

Directive 2013/32/EU, Annex I — List of Safe Countries of Origin, OJ L 180/60.

ECRE, ECRE Comments on the Italy–Albania Protocol: Legal and Human Rights Concerns (2023) https://ecre.org/ecre-comments-on-the-italy-albania-protocol-legal-and-human-rightsconcerns/ accessed 4 September 2025.

ECtHR, D.L. v Austria (Application no 34999/16, Judgment of 7 December 2017).

ECtHR, F.G. v Sweden (Application no 43657/11, Judgment of 23 March 2016).

ECtHR, Gebremedhin [Gaberamadhin] v France (Application no 25389/05, Judgment of 26 April 2007).

ECtHR, Hirsi Jamaa and Others v Italy (Application no 27765/09, GC, Judgment of 23 February 2012).

ECtHR, Interim Measures in the Context of Rwanda Removals (Application no 13156/22, June 2022).

ECtHR, Jabari v Turkey (Application no 40035/98, Judgment of 11 July 2000).

ECtHR, M.S.S. v Belgium and Greece (Application no 30696/09, Judgment of 21 January 2011).

ECtHR, Saadi v United Kingdom (Application no 13229/03, Judgment of 29 January 2008).

ECtHR, S.H. v Malta (Application no 37241/21, Judgment of 20 December 2022).

ECtHR, Sharifi and Others v Italy and Greece (Application no 16643/09, Judgment of 21 October 2014).

ECtHR, Xhavara and Others v Italy and Albania (Application no 39473/98, Decision of 10 January 2001).

European Commission, New Pact on Migration and Asylum COM(2020) 609 final.

European Commission, Proposal for a Regulation on Asylum Procedures COM(2020) 611 final.

European Union, Regulation (EU) 2024/1348 on Asylum and Migration Procedures (entered into force 11 June 2024).

Giuffré M, ‘Outsourcing Protection? Legal Implications of the Italy–Albania Model’ (2024) 36(1) International Journal of Refugee Law 1–25.

Goodwin-Gill GS and McAdam J, The Refugee in International Law (3rd edn, OUP 2007) 386.

Government of Italy and Government of Albania, Agreement on Cooperation in Migration and Asylum Management (signed November 2023, ratified 2024).

Government of Italy and Government of Albania, Protocol on Strengthening Cooperation in Migration Matters (signed 6 November 2023).

Gleeson M, Offshore: Behind the Wire on Manus and Nauru (NewSouth 2016).

Human Rights Watch, ‘Italy–Albania Asylum Pact Threatens Rights’ (10 November 2023) https://www.hrw.org.

Italian Ministry of the Interior, ‘Press Release on Albania Agreement Implementation’ (November 2023) https://www.interno.gov.it.

Marzagalli S, ‘Access to Justice for Asylum Seekers in Italy’ (2020) 22(1) European Journal of Migration and Law 35–57.

Moreno-Lax V, ‘The Architecture of Functional Jurisdiction: Unpacking Contactless Control’ (2020) 21(3) German Law Journal 385.

United Nations General Assembly, Convention Relating to the Status of Refugees (28 July 1951).

United Nations Treaty Collection, Status of Treaties: 1951 Convention and 1967 Protocol https://treaties.un.org accessed 4 August 2025.

UNHCR, Observations on the Italy–Albania Agreement on Migration Cooperation (2024) https://www.unhcr.org accessed 4 September 2025.