The Pre-Removal Detention of Immigrants: A Return to Ordinary Meaning

Abstract

The EU Return Directive demands that immigrant detention be as short as possible, but, by logical implication, this also means that detention can be as long as necessary. What concerns the maximum length of detention, the Return Directive is remarkably generous: Immigrants can be detained for a period of up to eighteen months—a deprivation of liberty that is otherwise justified only as punishment for serious crimes. The practice of such long-term detention, now burgeoning, is highly questionable for moral, practical, and—our focus—legal reasons.

The European Convention of Human Rights (ECHR) provides the relevant yardstick. While discussions on the legality of immigrant detention have focused on requirements of necessity, we shift attention towards the surprisingly absent question of maximum duration. Our analysis delves into the drafting context of the ECHR to reveal that it only authorizes the pre-removal detention of immigrants for markedly short periods. Picking up the interpretative canon of the regime, we note that meanings can of course change, but we argue that it is a legal mistake to consider that long-term detention is now sanctioned by the Convention.

A. Introduction

In 2015, a pivotal year, the number of immigrants detained in Europe surged, and with around 100,000 immigrants detained annually by European authorities it has remained high since then. ¹ A part of these immigrants is detained specifically for the purposes of deportation. In the recent case of N.M. v. Belgique, for instance, an Algerian national was released by Belgian authorities after an extensive thirty-one-month detention period. The European Court of Human Rights (ECtHR), referencing Article 5(1)(f) of the European Convention of Human Rights (ECHR), which allows for “the detention of…a person against whom action is being taken with a view to deportation or extradition,” did not fault Belgium in this case. The Court maintained that the duration of detention alone does not constitute a violation of the applicant’s right to liberty, provided it does not surpass the reasonable time necessary for deportation purposes. ² The case is remarkable for the scarcity of its reasoning and the striking nonchalance for the actual reasons behind the detainee’s release: The detainee was ultimately released not because the deportation proceedings had concluded, but simply because a space had opened up in an immigrant shelter. But the Court’s approach in this case was in fact rather ordinary. ³

A consistent thread in the ECtHR’s jurisprudence holds that detention ceases to be reasonable when there is no longer a realistic prospect of deportation, or when authorities cease to pursue deportation with due diligence. ⁴ The legal discussions have focused on the requirement that detention must be “as short as possible,” and on the procedural enforcement of this requirement. It is, indeed, an uncontroversial requirement. Yet, paradoxically, this very requirement, allows for long detentions if they are deemed necessary.

But are there no absolute limits to the length of detention? Undoubtedly, there are. Article 5(1)(f) ECHR may not spell out such a maximum duration explicitly for pre-removal detention, but there is a consensus that detention under this provision cannot be indefinite, regardless of necessity. The precise boundaries of these limits remain uncertain and underexplored, which is precisely the focus of our present intervention: We dive into the intricacies of treaty interpretation to unpack the ordinary meaning of “detention” as articulated in Article 5(1)(f) ECHR. ⁵

For European Union Member States, the legal requirements of pre-removal detention are specified in the EU Return Directive. The Directive not only mandates that pre-removal detention of immigrants must be “as short a period as possible,” ⁶ it also provides that detention must cease when there is no reasonable prospect of removal. ⁷ Moreover, the Directive stipulates that the duration of detention should only extend “as long a period as…it is necessary to ensure successful removal.” ⁸ When it comes to the absolute maximum length of detention, however, the directive puts it at six months, with a possible extension of up to twelve additional months under certain circumstances, such as lack of cooperation by the detainee or documentation delays from third countries. ⁹ While the EU Return Directive is undergoing discussions for reform, the maximum limits on detention duration are likely to remain unchanged. ¹⁰

The Directive itself claims to conform with the ECHR—a claim that is widely accepted in legal commentary. ¹¹ We argue to the contrary that the authorization of detention in Article 5(1)(f) ECHR knows limits for the period of pre-removal detention that the EU Return Directive exceeds. It has notably been recognized that periods of detention of six to eighteen months are otherwise known only as penalties for severe crimes. ¹² The deprivation of liberty of up to six months is typically reserved for serious offenses against persons and property, such as assault and battery. ¹³ More extensive periods of detention, ranging from twelve to eighteen months, are imposed only for grave crimes like manslaughter, rape, or the production of child pornography. ¹⁴ The EU Return Directive thus prescribes such extended periods of detention for immigrants who have not been convicted of a crime. ¹⁵

The contrast is even more pronounced when considering pre-trial detention regulations and procedural criminal law. Given that they are intended not as punishment but to ensure compliance with a legal process, they are more directly comparable to pre-removal detention for immigrants. Pre-trial detention is available only for suspects of serious crimes, and for limited durations—in the Netherlands a maximum limit of 110 days, less than four months. ¹⁶ After this period, suspects must be released, even if further detention is necessary for the purposes of an ongoing criminal investigation. Such is the essence of absolute limits. They remain inviolable, not subject to practical necessity.

In Section II, we briefly reconstruct the legal debates centering on the requirements of necessity and the reasonable prospect of removal. We then shift attention to ask: What is “detention” within the context of Article 5(1)(f) ECHR? Which practices of detention does it authorize? Historical insights indicate that at the time of the ECHR’s drafting, “detention” referred exclusively to short-term pre-removal detentions of just a few days. Practices of detaining immigrants for periods extending beyond three months were non-existent and not meant by the drafters, a position further supported by the travaux préparatoires, as discussed in Section III.A.

Today, the notion of “detention” has seemingly been expanded by instruments like the EU Return Directive to now encompass long-term detention. However, the evolution of meanings, especially within the “living instrument” doctrine of the ECHR, is subject to doctrinal conditions and limitations. The ECtHR has historically steered semantic developments in a singular direction that favors the enhancement of human rights. As discussed in Section III.B, it does not permit their erosion.

We therefore find that the meaning of detention, including the present-day meaning, remains limited to short-term detention. The most compelling counterargument to this view acknowledges the collective nature of interpreting human rights, which involves a dynamic interaction among various political institutions and societal actors. The EU Return Directive is the product of an elaborate politico-legislative process and its stipulation of a maximum of six plus twelve months for detention is normatively significant. Still, as follows in Section III.C, this stipulation is at odds with the interpretation of the ECHR that arises from the regime’s interpretative canon.

We thus conclude that the long-term pre-removal detention of immigrants is an unjustified infringement of the right to liberty protected under Article 5 ECHR. Although this may be seen as a bold argument against the backdrop of the EU Return Directive and burgeoning practice, it underscores that there are legal limitations to how far the meaning of a human rights treaty can be stretched. These limitations must be respected by both the ECtHR and the EU legislator.

B. The Legality of Pre-Removal Detention—A Brief Reconstruction of the Debate

The legal debate has a shared starting point that is deeply embedded within the fabric of the law: Personal liberty is the rule, and its deprivation is the exception. ¹⁷ Detention is authorized only in a limited set of cases, with immigration detention being one of them.

I. The Return Directive and Article 5(1)(f) ECHR

When addressing the legality of the pre-removal detention of immigrants within EU Member States, attention turns to the EU Return Directive and the domestic laws implementing it. The prevailing assumption is that the Directive complies with Article 5(1)(f) ECHR. Article 1 of the Directive proclaims that it:

[S]ets out common standards and procedures to be applied in Member States for returning illegally staying third-country nationals, in accordance with fundamental rights as general principles of Community law as well as international law, including refugee protection and human rights obligations. ¹⁸

This self-professed compliance with the ECHR has, to date, hardly been questioned with regard to the maximum duration of pre-removal detention. ¹⁹ The legal debate has instead focused on specific questions of application and interpretation of the Directive itself. ²⁰ Important issues include the grounds for pre-removal detention, reasons for prolonging detention, conditions of detention, detention as ultimum remedium, procedural safeguards, and the effectiveness of detention. For instance, what does it mean if a third country is not cooperating or if there are delays in obtaining necessary documentation, as Article 15 (5–6) provides?

Recent legal discussions have focused on the requirement that detention be “as short as possible,” per Article 15(1) Return Directive, that detention must be reasonable and necessary, and on procedural safeguards, notably judicial review, to ensure enforcement of these requirements. The European Court of Justice (ECJ) has clarified the Directive’s requirement that detention must cease when there is no longer a reasonable prospect of removal. For the ECJ, this occurs when “it appears unlikely that the person concerned will be admitted to a third country” within the period of six to eighteen months. ²¹ The ECJ’s case law aligns with that of the ECtHR, which considers pre-removal detention to be unlawful if it “exceeds that reasonably required for the purpose” of effectuating deportation. ²² The duration ceases to be reasonable when the authorities do not diligently execute deportation procedures or if there is no realistic prospect of removal. ²³ If detention does not serve effective removal, it becomes unlawful.

The Directive’s self-professed conformity with the ECHR may not have been challenged regarding maximum durations because the Directive seemingly provides stronger protections than the ECHR itself. The Directive explicitly requires that immigrants may be detained only for “as long a period as…it is necessary to ensure successful removal.” ²⁴ This is a requirement that Article 5(1)(f) ECHR itself lacks in that clarity. Prior to the adoption of the Return Directive, Galina Cornelisse thus criticized the ECtHR for its interpretation of Article 5(1)(f) ECHR, which did not require pre-removal to be necessary for effective deportation. She expressed concern that this stance allowed unnecessary deprivation of liberty and could undermine the human rights character of the right to liberty. ²⁵ With the Return Directive, such a loophole would presumably be closed.

It may indeed be that the Return Directive has improved the legal position of immigrants in pre-removal detention compared to the standards set by the ECtR alone. However, whether the Directive also amounts to an improvement when compared to domestic legal arrangements is a much more complex issue—an interesting question that is however beyond the scope of our analysis. ²⁶

It has moreover been suggested that European Courts have engaged in a “judicial dialogue” concerning the implementation of the Directive, further enhancing the protection of immigrants in detention and strengthening the rule of law more broadly. ²⁷ Galina Cornelisse and Madalina Moraru thus welcomed the adoption of the Return Directive because it codified the necessity criterion for pre-removal detention. ²⁸ In addition, the authors note that judicial dialogue

has ensured the checks and balances of state powers that usually exist in other fields of law, but were less present in immigration law before the entry into force of the Directive. ²⁹

In line with this emphasis on the necessity of detention, Cornelisse and Moraru commend the ECJ for its ruling in El Dridi, where the Court subjected removal measures to the test of proportionality. ³⁰ The test dictates that authorities must always opt for the least burdensome and coercive measure available when attempting to remove a migrant. It is noteworthy that the proportionality test as applied in El Dridi is essentially equal to a test of absolute necessity: The ECJ evaluates whether detention is genuinely necessary for the removal, or if the authorities could achieve the same result with less coercive measures. In contrast, a proportionality test sensu stricto would have placed the benefits of a measure into relation to the burdens for the person affected by the measure. Regardless, the El Dridi test is vitally important when considering long-term pre-removal detention. Such detention must be deemed unlawful if it is not the least coercive option available. Therefore, concerns may arise less from European law itself but rather, as Moraru points out, from the domestic judicial system’s sluggish and incomplete implementation of ECJ case law. ³¹

In short, in several ways the Return Directive has indeed improved the legal standing of immigrants facing pre-removal detention and, in comparison with other instruments of immigration policy, it does not fare so badly. The inclusion of the requirement for detention to be “as short as possible” within the Return Directive was a decisive factor for the EU Parliament’s Committee of Civil Liberties, reinforcing its confidence that pre-removal detention would respect the rule of law. ³² Nonetheless, the fact that the requirement of “as short as possible” also implies “as long as necessary” harbors the ironic possibility of justifying long-term detention periods. Practices of long-term detention and cases such as N.M. v. Belgique bring us back to the pivotal question: When does “as short as possible” become too long?

II. Excursus: The Prospect of Removal

The Return Directive does include a provision that might alleviate concerns about the maximum duration of detention. Article 15(4) stipulates that detention is no longer justified when there is no “reasonable prospect of removal.” The ECJ interprets a prospect as reasonable if removal is feasible within 18 months. ³³ However, this equation of “reasonable prospect” and the maximum allowed duration is far from clear. Izabella Majcher thus contests the ECJ’s interpretation of Article 15(4), suggesting that it should rather be understood as meaning removal “in the reasonably near future.” ³⁴ While we sympathize with Majcher’s argument for a shorter maximum detention duration, we doubt its legal grounding. We engage with it in some detail because, if solid, it would also alleviate our concerns about maximum durations. Because we question her argument, however, we ultimately find that it supports the need to return to examining the kind of detention that is authorized under Article 5(1)(f) ECHR, which is how we then continue.

According to Majcher, Advocate General Ján Mazák adopted a stricter interpretation of Article 15(4) Return Directive than the ECJ in the Kadzoev case. ³⁵ She cites Mazák, who paraphrases the reasonableness requirement as detention being justifiable “within a reasonable period” and “in the reasonably near future.” ³⁶ Majcher could have also added that, according to the Advocate General, “the existence of an abstract or theoretical possibility of removal, without any clear information on its timetabling or probability, cannot suffice.” ³⁷

While the Advocate General seemed to advocate for a stricter standard than what the ECJ ultimately adopted, Majcher’s interpretation reads too much into his remarks. The Return Directive does explicitly permit detention in instances when “the third-country national concerned avoids or hampers the preparation of return or the removal process” (Article 15(1)(b)), is uncooperative (Article 15(6)(a)), or when there are “delays in obtaining the necessary documentation from third countries” (Article 15(6)(b)). Given that detention is allowed in these circumstances, it is difficult to agree with Majcher’s interpretation of the reasonableness requirement of Article 15(4), and it seems doubtful that the Advocate General’s stance provides support for her interpretation.

Rather, the Advocate General’s opinion distinguishes between types of non-cooperation that affect the prospect of removal. Certain non-cooperative behaviors may indicate that removal is impossible from the outset, precluding the authorities from obtaining “any clear information on its timetabling and the probability.” For example, if the administration of the destination country is entirely dysfunctional or unresponsive to immigration matters, and this structural issue is unlikely to be resolved, removal prospects are null. Conversely, some non-cooperative actions may still allow insights into the likelihood and timing of removal. This could occur during a diplomatic or economic dispute with a third country that is using non-cooperation in deportation proceedings as a negotiation tactic. In such cases, there may still be a reasonable prospect of deportation if there are indications of an impending resolution to the diplomatic disagreement. Contrary to Majcher’s viewpoint, interpreting the Advocate General’s opinion as requiring the end of detention within a reasonable period of time or in the near future would in fact be offering an interpretation contra legem, contrary to the permissions granted by Articles 15(1)(b) and 15(6).

Majcher also relies on the ECtHR’s case law concerning the notion of a “realistic prospect of removal,” which parallels Article 15(4) of the Return Directive. She argues that, when a removal delay is due to the destination country’s lack of cooperation, then the detaining authorities must take the necessary measures to ensure collaboration. In her words, if “there are no objective signs that the authorities of the destination country change their position “quite soon,” a realistic prospect of removal ceases to exist.” ³⁸ However, the ECtHR’s support that Majcher seeks is not as solid as she implies. The phrase “quite soon” used by the Court does not pertain to the period within which the destination country may collaborate or when the removal is likely to take place. Instead, the ECtHR refers to the point at which the detaining authorities realize that effective removal is unlikely. ³⁹ In the focal case at issue, it “quite soon” became apparent to the authorities that removal would be impossible. Consequently, detention should have ceased because they knew “quite soon” that there was no prospect of removal, either imminently or in the future. The ECtHR deemed the detention to be unlawful due to this knowledge, not because the removal did not occur “quite soon.” If the latter had been the criterion, the Court could have easily said so, especially because the individual was detained for up to three years and eleven months. But the Court did not set such a criterion.

III. Maximum Duration

The parameters of reasonableness and necessity have dominated the legal discourse concerning the duration of detention. By contrast, the absolute maximum period of detention—of six plus twelve months, according to the Return directive—has itself remained largely unchallenged.

The ECtHR’s case law shows that duration is a pertinent issue. The Court employs an escalating scale when addressing detention length: Detention spanning only a few days is described as simply “short” detention, and it is deemed unproblematic. ⁴⁰ A few weeks to a month or two is still categorized as “relatively short” and is still fairly unproblematic. ⁴¹ However, when detention approaches five to seven months, the Court no longer refers to is as “short.” ⁴² For even longer detentions, the Court’s concern visibly grows, prompting a more rigorous examination. ⁴³ The ECtHR dedicates more time and applies greater scrutiny to assess whether there was indeed a realistic prospect of removal and if the authorities were diligent in executing deportation procedures.

What the Court does not directly address, however, is a head-on examination of the maximum duration of detention that Article 5(1)(f) permits. It is worth reiterating that the Article itself does not define a maximum limit. But this absence should not be misconstrued as allowing indefinite detention. In both theory and practice, the existence of a maximum duration is entirely undisputed. What is that maximum length of detention? Answering this question hinges on the legal meaning of “detention” itself.

C. The Definition of “Detention” in Article 5 ECHR

This section interprets “detention…with a view to deportation,” as stated in Article 5(1)(f) of the ECHR. ⁴⁴ The current legal discourse may gloss over the question of what detention means rather quickly because it appears self-explanatory: Detention means the act of detaining somebody or the state of being detained. Our focus, however, is on the duration of detention and whether long-term detention aligns with the concept of detention as contained in Article 5. We are after the specific legal meaning of detention. Once again, it might appear evident at first glance: Article 5 does not qualify or specify the kind of detention it permits. So, why would long-term detention be excluded?

First of all, the fact that detention is not qualified in Article 5 is inconclusive. There are numerous instances where unqualified legal terms are understood to encompass only a subset of its broader, literal meaning. ⁴⁵ For example, the cardinal prohibition of the use of force in Article 2(4) of the UN Charter refers to “force” without further qualification. Leading interpretations of this provision restrict the term to “armed force,” excluding political or economic force. ⁴⁶ This interpretation draws on the historical context of the UN Charter’s drafting and on how the notion of force was used at the time. Similarly, when examining practices of pre-removal detention at the time of drafting the ECHR, we find no precedent of long-term pre-removal detention of immigrants. At that time, “detention” was understood to mean short-term detention.

Second, while we acknowledge that the current ordinary meaning of detention has evolved to include longer detention periods, the question arises whether the legal meaning within the context of Article 5 ECHR can similarly shift. This leads to discussions on evolutionary interpretation and the Convention’s nature as a living instrument. We argue that within the Convention’s specific regime, semantic changes can only expand human rights protections, not diminish them.

Third, the best counterargument against this one-directional view of semantic change is that societal priorities shift, and that the democratic politico-legislative process should determine the application and interpretation of human rights. We agree in principle. The EU Return Directive, which allows for up to 18 months of pre-removal detention and notably asserts its conformity with the ECHR, is normatively significant. ⁴⁷ Nonetheless, it is the role of the ECtHR to assess this self-professed claimed conformity, and there are compelling reasons to challenge it. Our interpretation of the ECHR leads us to the view that the Directive’s claim of conformity cannot hold. This is not because a detention period of six plus twelve months is unnecessary or disproportionate per se, but rather because such a lengthy maximum detention period extends beyond the short-term detention that Article 5(1)(f) ECHR authorizes.

II. Ordinary Meaning

1. Contemporaneous Meaning

Article 31 of the Vienna Convention on the Law of Treaties (“VCLT”) establishes the well-rehearsed “general rule of treaty interpretation”: “A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.” ⁴⁸ The ECtHR has developed a regime-specific approach to treaty interpretation and, in particular, formed the doctrine according to which the ECHR is read as a “living instrument” ⁴⁹ and as a “constitutional instrument of European public order.” ⁵⁰ From its inception, the ECtHR specified its interpretative methodology, emphasized that the rights of the Convention must be given full effect. ⁵¹ It has stressed the significance of the preamble and rule of law principles in guiding its interpretations. ⁵² Specifically, the concluding paragraph of the preamble, where signatory states identify themselves as “Governments of European countries which are like-minded and have a common heritage of political traditions, ideals, freedom and the rule of law,” reinforces the collective commitment to these principles. ⁵³

The interpretative canon as it has developed under the influence of the ECtHR has thus some specificities. To start, however, the Court generally acknowledges in line with the general rule of treaty interpretation that the ECHR should be interpreted to reflect a term’s ordinary meaning within the treaty’s framework. ⁵⁴ This ordinary meaning is tied to the term’s common use, which is to say, its “natural and ordinary meaning in the context in which they occur.” ⁵⁵

Granted, if we consider how the term “detention” is currently used in the context of pre-removal immigrant detention, it encompasses various types, including long-term detentions lasting two years or more. Had the ECHR been drafted in today’s context, without additional qualifications, it would be difficult to argue that the ordinary meaning of “detention” does not extend to long-term detention. At the time of the drafting of the ECHR, however, long-term pre-removal detention was not a recognized official practice. And that makes a difference. It therefore stands to reason that the ECHR’s drafters did not envision, let alone intend, to authorize long-term pre-removal detention. The contemporaneous ordinary meaning of “detention” in Article 5(1)(f) ECHR can thus be interpreted as encompassing only short-term pre-removal detention.

The available historical sources indicate unanimously that long-term pre-removal immigration detention was not practiced. ⁵⁶ The main reason for the absence of long-term pre-removal detention practice is straightforward: There was no need for authorities to detain migrants for a long time, simply because there were no migrants whose deportation procedures took a long time. During 1901–1950, immigration was primarily an intra-European affair, with non-European immigration being extremely limited. ⁵⁷ The common deportees were European citizens, usually from neighboring countries, deemed undesirable due to socio-economic factors—often labeled as “vagrants” and “vagabonds.” ⁵⁸ Their deportation was typically swift, frequently resolved within days. ⁵⁹ Migrants arriving from non-European countries came from former colonies. Most had a legal title to stay or could be repatriated relatively easily. ⁶⁰ Also when larger numbers of non-European migrants came to Europe during that period, for instance to France, it was part of a deliberate policy to attract manual labor, especially during war time. The same goes for illegal migrants arriving after the WWII: They were not deported but regularized as they once more constituted a much-needed labor force. ⁶¹ Not only was there no need for long-term pre-removal detention, also the proper legal framework for long-term pre-removal detention was missing. Some countries had statutes authorizing only short-term pre-removal detention. ⁶² Other countries lacked any explicit statutory basis for pre-removal detention so that authorities relied on criminal law provisions instead when carrying out short-term pre-removal detention. ⁶³

There has been state practice in 20^th-century Europe that, on first glance, may seem to contradict our argument, notably special camps for aliens, internment of foreign residents, and other forms of alien detention. But upon closer scrutiny these practices simply do not amount to official long-term pre-removal detention preceding the drafting of the ECHR. Many of the special camps for aliens were not detention facilities but served as temporary housing for refugees and stateless persons following the world wars. The aliens were not deprived of their liberty. Neither were the aliens in the camps awaiting their deportation; they had received asylum, albeit in very harsh reception conditions. ⁶⁴ Other instances did constitute a form of long-term alien detention, but they did to pursue purposes of immigration policy or removal. Foreign residents were interned in times of international conflicts for security and military reasons, not migration policy. ⁶⁵ Poor undocumented aliens could end up in long-term internment, in so-called vagrant colonies were they had to work to earn enough to cover the costs associated with their return. While authorities arranged the necessary documentation for the return of the aliens to their country of origin, the legal basis for internment was not removal but destitution. ⁶⁶ Similarly, sometimes aliens were put in long term detention not as a measure to ensure deportation but as a punishment for violating migration rules, for example, rupture de ban d’expulsion: Imprisonment as a punishment for illegally re-entering the territory. ⁶⁷ For sure, in the later 20^th century, European states also practiced long-term pre-removal detention in Europe. But all those practices date more than a decade after the ECHR was drafted. ⁶⁸ Moreover, sometimes these long-term pre-removal detention practices were not even officially recognized but clandestine operations, meaning these secret practices could not inform the contemporary legal understanding of detention that was to be authorized under the ECtHR. ⁶⁹

In short, the evidence from the available studies and sources suggests that long-term pre-removal detention was not practiced. Typically, those detained for deportation were European nationals, and their repatriation faced few obstacles. Instances of long-term alien detention found in 20^th-century Europe were either not related to pre-removal processes, were carried out illegally and in secret, or occurred well after the ECHR was drafted.

2. Context

The ECHR is structured in such a way that articles’ first paragraphs often define the specific rights and freedoms, with subsequent paragraphs detailing permissible limitations. For instance, Article 2(2) restricts the lawful deprivation of life to a limited set of circumstances, subjecting any infringement to the very high threshold of absolute necessity. In stark contrast stands Article 3, whose categorical ban on torture is articulated in a singular, uncompromising sentence, devoid of qualifiers or subsequent paragraphs. ⁷⁰ Several articles, including Articles 8, 9, 10, and 11, employ a shared “paragraph 2” mechanism. This mechanism permits infringements upon the protected rights only when such actions are “in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.” ⁷¹ Beyond these specifications, the only other exception is a derogation during states of emergency as outlined in Article 15, which is likewise bound by rigorous procedural and substantive requirements.

Article 5 of the ECHR, which safeguards the right to liberty and security, stands out within the Convention’s framework. Unlike other rights with a “paragraph 2” clause allowing for state interference under specific conditions, Article 5’s first paragraph sets out the right and articulates the exclusive circumstances under which liberty may be restricted. Specifically, item (f) pertains to “detention…with a view to deportation,” thus narrowing the scope of this singular administrative action. The subsequent paragraphs, 2 through 5, of Article 5 confer procedural rights upon individuals who are arrested or detained, enabling them to contest the legality of their detention and, where appropriate, seek compensation.

The absence of a “paragraph 2 mechanism” within Article 5 of the ECHR points to the drafters’ intention to strike a specific balance between the fundamental human right to liberty and pressing societal interests. These interests, as permissible grounds for the deprivation of liberty, are exhaustively enumerated within the Convention. Thus, Article 5(1)(f) does not recognize any societal interest analogous to those in “paragraph 2” provisions; it exclusively addresses the administrative measure, namely the action of deportation. Consequently, no societal interest—regardless of its perceived urgency—can justify an expansion or intensification of the constraints on the right to liberty and security beyond what the Convention permits. This interpretation suggests that any necessity for detention that might arise cannot legitimatize actions beyond the scope explicitly authorized by the Convention.

Furthermore, the lack of a “paragraph 2 mechanism” in Article 5 signifies that the Convention’s drafters chose to prioritize personal liberty over societal needs within a democratic society. This prioritization operates almost mechanically: Societal interests associated with effective removal initially supersede personal liberty, requiring only that authorities demonstrate the detention’s contribution to effective removal. Justification for outweighing personal liberty is not required. However, as detention continues and ceases to remain short, the priority shifts—personal liberty takes precedence over removal-related interests. Therefore, Article 5 does not permit authorities to independently determine that a prolongation of the detention is necessary for society. A systematic reading of Article 5, alongside the Convention, thus supports the view that detention was intended to be short-term, negating the need for a “paragraph 2 mechanism” because only brief pre-removal detention is permissible.

It remains a separate question whether detention must be necessary for effective deportation. This necessity requirement is common in various domestic laws and is explicitly required by the EU Return Directive. Conversely, the ECHR and ECtHR’s interpretation are themselves not as clear-cut on this matter, which has elicited notable surprise and critique from several commentators. ⁷² Our interpretation posits that the ECHR’s drafters, having already balanced the relevant interests, specifically intended the detention authorized under Article 5(1)(f) to be short rather than long-term.

3. Travaux Préparatoires

The preparatory documents related to Article 5 are scattered and brief, particularly concerning subsection (f). Nevertheless, the existing records on immigration detention provide significant insights and, overall, the travaux préparatoires further support our interpretation that Article 5(1)(f) sanctions solely short-term detention. ⁷³

In developing Article 5, the drafters of the Convention based their work mainly on a text template derived from the Universal Declaration of Human Rights (UDHR) regarding the right to life, liberty, and security. This draft, omitted in the final, succinct version of Article 3 UDHR, was copied with minimal modifications into the definitive text of Article 5 ECHR, without much debate. In the Committee of Experts that was tasked with drafting the Convention, only Luxembourg’s delegate proposed revisions concerning the detention of aliens. He aimed to ensure that the Convention would permit States the discretion to expel “undesirable aliens” and expressed concerns that, without amendments, such actions might be construed as arbitrary detention. ⁷⁴ This proposed amendment, however, was overlooked and not adopted. The Convention’s travaux préparatoires otherwise provide no further clarification, leaving only the earlier UDHR drafts to shed additional light.

The draft UDHR provided for a number of exceptions to the general prohibition against deprivation of liberty, including the case of immigration detention, which was applicable solely to immigrants who unlawfully entered a country. An amendment proposed by the Indian representative sought to extend this provision to include the detention of immigrants unlawfully residing within a country for the purpose of deportation. He immediately added that such detention, while lawful, must not be indefinite and called for explicit safeguards. ⁷⁵ The Chairman from the United Kingdom indicated that general “habeas corpus” protections were sufficient. ⁷⁶ The debate still continued, with the Chinese representative, representing the Republic of China in 1947, echoing the necessity for time-bound detention, cautioning against indefinite confinement. ⁷⁷ The U.S. delegate readily agreed, suggesting that despite the general “habeas corpus” clause, particular safeguards for immigrant detention remained essential. ⁷⁸ In subsequent discussions, the Chinese representative reiterated concerns about the potential for excessively lengthy detention, advocating for particular safeguards for immigrant detention. ⁷⁹ The U.S. representative still shared the concern, but then deemed the “habeas corpus” provision a sufficiently robust safeguard. ⁸⁰

The drafters of the UDHR thus expressed a shared view that immigration detention should not result in long-term detention. This consensus stemmed from concerns about the potential misuse of a state’s authority to detain. Although the maximum duration of immigrant detention was linked to “habeas corpus” protections and the right to judicial review, it was also apparent that a limitation of the period of detention was necessary with those protections. Even if an upper limit was not made explicit, it was implied, and its necessity was undisputed.

There may still be a cynical, yet plausible, reason why the drafters of the ECHR were content with limiting pre-removal detention to short-term periods. At the time, there was a shared assumption, supported in law and fact, that the rights and freedoms of the Convention would apply almost exclusively to Europeans—unless contracting parties expressly decided otherwise. The protection was confined to European territory (Article 56 ECHR) and did not extend to overseas territories. In the European territories of the contracting parties, there were scarcely any non-Europeans. ⁸¹

Regarding the ordinary meaning of Article 5(1)(f) ECHR, it is important to acknowledge that reliance on the Convention’s travaux préparatoires has diminished in the interpretive practice of the Convention. ⁸² We also recognize that the meanings of terms can evolve over time: The concept of pre-removal detention has expanded to also encompass long-term scenarios. This raises the question of whether, and if so, how, such semantic changes should influence treaty interpretation practices. In the interpretative practice of the Convention, any changes in meaning over time have typically been resolved in favor of the current, contemporary interpretation. ⁸³ The diminishing role of the travaux and the preference for current meanings suggest an assumption that semantic changes are progressive and enhance human rights protection. However, with regard to “detention…with a view to deportation,” that has not been the case.

II. Which Ordinary Meaning: Current or Contemporaneous?

1. Semantic Change and Generic Terms

The extensive debates on treaty interpretation leading to Article 31 VCLT converged on textualism as a common denominator, even if theoretical and practical disagreements on interpretive approaches persisted. ⁸⁴ Writing in 1963, Rudolf Bernhardt, who would serve as a Judge at the ECtHR from 1981–1998 and briefly as its President in 1998:

If the text of a treaty is recognized as the object of interpretation [Auslegung], then nothing stands in the way of assuming that the text can experience a change of meaning in the course of its development even without any tacit modification of the treaty. ⁸⁵

We readily agree with this observation, which is, as such, virtually beyond debate. But if laws evolve through changes in meaning, what is then the role of parties’ intentions? Did parties perhaps intend the potential evolution of meanings over time? ⁸⁶ Legal doctrine has reframed this question, focusing on the deployment of generic terms within treaties and formulating criteria to gauge the legitimacy of semantic change. ⁸⁷ The International Court of Justice (ICJ) has shaped this debate with its dictum that:

[W]here the parties have used generic terms in a treaty, the parties necessarily having been aware that the meaning of the terms was likely to evolve over time, and where the treaty has been entered into for a very long period or is ‘of continuing duration’, the parties must be presumed, as a general rule, to have intended those terms to have an evolving meaning. ⁸⁸

Determining the original intentions of treaty drafters can be challenging, especially when the final text still reflects a dissensus. ⁸⁹ The text itself thus remains the main point of reference, also for determining parties’ intentions—prompting the question such as why were certain terms chosen rather than others? Although the pursuit of such answers often involves conjecture, drafters can typically expect that choosing broader terms allows for greater interpretative flexibility compared to more precise language. However, what is considered specific at one point may later be deemed generic. For instance, when China committed to liberalize “sound recording distribution services” in anticipation of its World Trade Organization (“WTO”) accession in 2001, could it have foreseen that this would encompass digital distribution methods? Facing a dispute on this issue, China contended that the scope of its legal commitments could not simply change by changes in the meaning of terms it could not possibly have anticipated. ⁹⁰ But the WTO Appellate Body rejected this argument, reasoning that the terms in China’s General Agreement on Trade in Services (“GATS”) Schedule were “sufficiently generic that what they apply to may change over time.” ⁹¹

2. Conditions of Change: Detention Makes a Qualitative Difference

Debates on whether parties to a treaty could have intended certain terms to be subject to evolutive interpretation, regardless of their persuasive power in general or specific contexts, miss a vital point: Not all changes bear the same qualitative weight. It is pivotal to discern whether the evolving interpretation of a term may include elements that are qualitatively distinct from those initially intended. This distinction has significant implications in both legal theory and application, as it is crucial to determine if new particulars that fall under a term due to its evolving interpretation are equivalent to those covered at the time of the treaty’s inception. ⁹²

The ICJ’s stance on generic treaty terms that we just mentioned, for instance, revolved around the right of treaty parties to navigate a river “for purposes of commerce.” Nicaragua maintained that “commerce,” based on its contemporary meaning when the treaty was signed in 1858, pertained solely to the transport of goods. ⁹³ Conversely, Costa Rica argued—quite persuasively—that “navigation for purposes of commerce” should now also encompass passenger transportation, or tourism commerce. The ICJ concurred with Costa Rica, finding the semantic change acceptable because the transportation of goods and tourists are equivalent in the sense that they are both “commercial in nature.” However, the use of government vessels for navigation, which does not equate to commercial activity, remained outside this interpretation. ⁹⁴ In essence, broadening the scope of “navigation for purposes of commerce” to include tourism did not represent a qualitative change but rather an expansion of particulars within an established category.

Turning to the notion of “detention…with a view to deportation,” it is clear that the drafters of the ECHR chose not to elaborate further on this notion. At first glance, “detention” might seem as generic as the term “commerce.” However, a deeper legal analysis prompts the question: Does long-term detention fundamentally differ from short-term detention? Is expanding the scope of “detention…with a view to deportation” to encompass long-term detention merely extending the same category, or does it introduce a substantive, categorical shift?

We claim that the duration of detention is not a marginal matter, but that long-term detention is qualitative and categorical different from short-term detention. What distinguishes them, exactly? Unlike the transport of goods or people, which do not differ qualitatively from the perspective of commerce, the distinction between short- and long-term detention is significant from the perspective of human rights. The spirit of the Convention is to protect life and security, considering the increasing hardship on the detainee and the expanding executive power over time, both of which Article 5 ECHR aims to balance and regulate. ⁹⁵

Moreover, we refer once more to the comparisons with pre-trial detention and scales of prison times: Short and long-term detention are just not equivalent particulars. As the duration of liberty deprivation extends, the level of authority required to sanction the detention also escalates. Beyond a specific threshold, the nature of the detention shifts categorically. This aligns with the drafters’ intentions and their concerns regarding the temporal limits of detention, as they sought to prevent the distinction and adverse implications of protracted confinement. The travaux préparatoires reflect this concern, signaling no expectation or intention for the term “detention” to evolve to include prolonged periods that are qualitatively different from the original conception.

This intention is reflected in the objectives and purposes of the ECHR. ⁹⁶ Endorsing an increase in states’ powers to detain immigrants, thereby facilitating greater encroachments on the right of liberty, contradicts the fundamental goal of human rights protection and the commitment of signatory governments to uphold freedom and the rule of law. While interpretations of various ECHR concepts have indeed evolved over time, such evolution has consistently aligned with the Convention’s foundational aim of enhancing human rights protections.

3. The ECHR as a Living Instrument

The interpretative approach to the ECHR has evolved to acknowledge the Convention’s dynamics nature. The meanings of treaty terms are not static, legal obligations have naturally shifted over time, and the law has evolved through interpretation to reflect new developments. Such progressive interpretations are both recognized and encouraged within legal theory and jurisprudence. A landmark in this evolutionary journey was the case of Tyrer v. United Kingdom, where the ECtHR asserted that “the Convention is a living instrument which…must be interpreted in light of present-day conditions.” ⁹⁷ The case involved a fifteen-year-old student subjected to birching—a form of physical punishment once considered ordinary but later deemed by the ECtHR to be a degrading punishment in violation of Article 3. ⁹⁸ This is but one of many instances. In the words of the Judges Sicilianos and Raimondi, since its formulation in Tyrer, the living instrument doctrine “has spread throughout the Strasbourg case-law and…has enabled the Court to adapt, over time, the text of the Convention to legal, social, ethical or scientific developments.” ⁹⁹ In their analysis of trends in the case law, Larry Helfer and Eric Voeten noted the ECtHR’s progressive development of ECHR interpretation, even when it entails overturning previous judgments. They point to the Court’s stance:

While it is in the interests of legal certainty, foreseeability and equality before the law that the Court should not depart, without good reason, from precedents laid down in previous cases, a failure by the Court to maintain a dynamic and evolutive approach would risk rendering it a bar to reform or improvement. ¹⁰⁰

Helfer and Voeten further inquire whether, in light of recent challenges to the Court’s authority and its progressive judicial practices, there might be a shift towards more conservative human rights interpretations. They observe a rise in dissenting opinions within the judiciary, some of which suggest that the Grand Chamber’s interpretations have regressed. Yet, as of now, the Court has not reversed any previous judgments in a manner that diminishes human rights protections, at least not explicitly and certainly not in the name of the living instrument doctrine. ¹⁰¹ The question remains: Could it?

Governments and stakeholders can influence international courts and judicial practices in a variety of ways, including through judicial appointments, financial control, politico-legislative input, and other subtler or informal forms. ¹⁰² It is possible for courts to be swayed by such influences. Our concern, however, lies not in the political science or sociological realms regarding judicial behavior. Instead, we examine a question of legal doctrine: Is it within the capacity of the ECtHR to legally reverse progress on human rights protection? Could the doctrine that regards the ECHR as a “living instrument” facilitate a retrogressive interpretation?

Some scholars and judges assert that the Court could indeed make such a shift in compliance with legal doctrine and methodology. They argue that there is no doctrinal restriction that mandates that interpretive changes must always be progressive. ¹⁰³ Former ECtHR President Wildhaber has robustly presented this viewpoint:

The idea that a “ratchet” should be attached to the interpretative process . . . has neither a treaty nor a customary law basis, nor a reasoned justification in the Court’s jurisprudence or the States’ practice and reactions to this jurisprudence. Democratic societies may and sometimes must change their values and laws, especially when they are confronted with a widespread and vicious life-endangering terrorism. ¹⁰⁴

Bracketing the validity of this argument, it is essential to note that Wildhaber’s perspective is not directly relevant to our interpretation of “detention” within the context of Article 5(1)(f) of the ECHR. The current debate on regressive interpretations revolves around whether the Court has the authority to overturn its own, once progressive interpretation. The controversy concerns instances where the Court has used the living instrument doctrine to depart from the original, contemporaneous meaning of the Convention in favor of an evolutionary and progressive interpretation that expands the scope of human rights protection. Proponents of the possibility of regressive interpretations, such as Wildhaber, argue for the Court’s discretion to relinquish its own progressive rulings and revert to the contemporaneous ordinary meaning. In this scenario, “regressive interpretation” implies that the Court would diminish the extent of human rights protection that it had previously enhanced.

However, the ECtHR’s case law concerning pre-removal detention presents a different scenario. The Court did not expand the initial protection provided for by the Convention. There was no progressive interpretation of Article 5(1)(f) on the basis of the “living instrument” doctrine. On the contrary, the Court has arguably lessened the original scope of human rights protection by deeming long-term pre-removal detention lawful—an interpretative error, in our opinion.

Interpreting Article 5(1)(f) ECHR to include long-term detention constitutes a misinterpretation by overlooking the special characteristics of human rights treaties. ¹⁰⁵ Unlike many other treaties, the object of human rights treaties is not about reciprocal obligations that create a quid pro quo situation between contracting parties, turning the contracting parties into both duty holders and beneficiaries. Instead, the primary beneficiaries are individuals who are entitled to the rights and freedoms enshrined in the treaty, and it is the duty of the contracting parties to protect these entitlements. Ironically, it is often these same contracting states that pose the greatest risk to these rights and freedoms, through direct actions or failures to act.

Furthermore, a human rights treaty serves a dual function: It is both a recognition of inherent human rights and liberties and a legal framework that establishes them. This dual nature underpins the characterization of the ECHR by the ECtHR as a “constitutional instrument of European public order.” ¹⁰⁶ Consequently, due to the treaty’s inherently non-reciprocal, pro homine (in favor of the individual), and constitutional character, no legal authority—be it the ECtHR, the EU, or domestic bodies—has the authority to decrease, through its interpretations, the level of human rights protection provided by the Convention. ¹⁰⁷

Upon closer inspection, proponents of regressive interpretation appear not to be discussing what constitutes a legally permissible interpretation at all. For instance, Judge Wildhaber, in the previously mentioned statement, concludes with an assertion that regressive interpretations are permissible under existential threats, citing the Convention itself, specifically Articles 15 to 18. ¹⁰⁸ Article 15 does indeed permit the temporary suspension of certain rights and liberties in times of emergency, but such suspensions are not a matter of interpretative changes. They are suspensions of the applicable rights, without altering the legal interpretation of those rights. Therefore, the ordinary meaning of “detention…with a view to deportation” within Article 5(1)(f) must adhere to its contemporaneous meaning, which pertains to short-term detention.

D. Conclusion: Limits of Interpretation

The ECHR establishes a non-negotiable core of human rights, while also allowing for societal discretion in defining further human rights protections. Within this legal framework there is room for legal actors, especially the ECtHR and legislators in Europe, to adapt the ECHR’s content and scope of application to societal and political needs. This is the business of legal interpretation. But there are legal limitations to the legal actors’ discretion when filling out the content and scope of the ECHR. These limits result from the interpretative canon, which is, in turn, shaped by the specific regime of the ECHR. Granted, these limits are again to some extent malleable, but they are none the less real. There are breaking points—the practice of interpretation is both creative and it is constrained. ¹⁰⁹

We found that in Europe, at the time of drafting of the ECHR, the ordinary meaning of “detention…with a view to deportation” was restricted to short-term detention only. The reasons were relatively straightforward: There was neither a need nor a sufficient legal basis for long-term pre-removal detention. The ECHR’s drafters could not have envisaged let alone intended detention to also include long-term detention. We further found that there was no intention to allow the meaning of “detention” to evolve in a way that it would eventually include long-term detention, nor would such a semantic change be in line with the understanding of the Convention as a living instrument. The living instrument doctrine has never been used to reduce the level of human rights protection initially provided for by the ECHR. It has been developed and used to expand the initial human rights protection under the ECHR, not to undercut it.

One approach to challenge the lawfulness of long-term pre-removal detention would be to apply the test of proportionality—a hallmark of constitutional human rights adjudication—to the specified maximum detention duration of six plus twelve months, thereby contending that such long-term detention is excessive and disproportionate. This perspective would align with the view of several other commentators who have also criticized the practice of long-term detention. ¹¹⁰ However, our argument is different. Neither the Court nor the legislators in Europe possesses the authority to assess the proportionality of long-term detention, because long-term pre-removal detention is simply not authorized by the Convention to begin with. It exceeds the authorization expressed in Article 5(1)(f) ECHR with “detention…with a view to deportation.” Comparable to other elements relating detention like minimal sanitary standards, for instance, the maximum detention period is absolute, non-negotiable, and impervious to considerations of necessity.

In conclusion, we note that it is remarkable that, to the best of our knowledge, the ECtHR and other legal actors like the EU or national legislators, have not conducted, or anyway not published, a proper interpretation of Art. 5 (1)(f) ECHR in accordance with the legal rules of interpretation. They assume that the meaning of “detention” under Art. 5 (1)(f) ECHR encompasses long-term pre-removal detention. Accordingly, they assume and profess that long-term pre-removal detention authorized by the Return Directive complies with the ECHR. Our legal analysis contradicts this view. Long-term pre-removal detention is unlawful as it violates Art. 5 (1)(f) ECHR. Consequently, administrative orders authorizing long-term pre-removal must be quashed and long-term detainees must be released, lest there is another basis for detention.