Meta-Borders and the Rule of Law: From Externalisation to ‘Responsibilisation’ in Systems of Contactless Control

Abstract

This article contests the strategic use of what I have called meta-borders. These are the array of border enforcement mechanisms implemented beyond the physical frontiers of States through different means and by different actors, for the purpose or with the effect of denying human rights protection to (unwanted) non-citizens. The ensuing ‘irresponsibilisation’ of States of destination, on whose behalf or for whose benefit the measures are executed, is anathema to the Rule of Law. My main contention is that prevailing understandings of jurisdiction and responsibility, as applied to externalised migration controls (the core feature of meta-borders), need to be revised. Currently, they allow for the emergence of a double standard, solely dependent on location, whereby the State may act abroad with impunity in relation to the human rights consequences of its conduct, exploiting geographical distance to create and legitimate ethical and legal detachment from its own wrongdoing. This article proposes an alternative model of ‘responsibilisation’ that tallies with the flexible spatiality of migration governance. The functional configuration of the meta-border is matched with an equally functional conceptualisation of jurisdiction that rejects unaccountable forms of power. The article thus problematises the localisation of the meta-border, mapping its multiple roles, modes, and dimensions, highlighting the significance of its legal manifestations, before exploring the impact of law on the de-territorialisation of the sovereign exercises of demarcation, delimitation, and exclusion that it implies. The meta-border, crafted by legal fiat, actively (re)orders space, curtailing the reach of human rights and disclaiming responsibility for related violations. To reconcile power with accountability, I advance the ‘responsibilisation’ model, premised on the acceptance that human rights, as fundamental components of the Rule of Law, track and constrain all exercises of State authority.

1 Introduction: Generating Unaccountability

The objective of this article is to contest the strategic use of what I have called meta-borders. These are the array of border enforcement mechanisms implemented beyond the physical frontiers of States through different means and by different actors, for the purpose or with the effect of denying human rights protection to (unwanted) non-citizens. The ensuing ‘irresponsibilisation’ of States of destination, on whose behalf or for whose benefit these measures are executed, typically in collaboration with or by delegation to third parties, is anathema to the Rule of Law. My point of departure is that unaccountable power is arbitrary power,¹ which is contrary to a Rule of Law-based conception of the international legal order post-World War II. A Rule of Law-based interpretation of the notion of ‘jurisdiction’, as applied in the human rights field, demands ‘conformity with the principles of justice and international law’²; all exercises of sovereign authority, to remain legitimate, need to conform to the requirements of basic legal guarantees that should be understood to attach to all manifestations of State control (whether projected within or beyond territorial confines). Public power must be bound by Rule of Law standards (including human rights) whenever it operates.³ That it is the State that acts is more important than where it may be that it is acting.

My main contention is that prevailing understandings of jurisdiction and responsibility, as applied to externalised migration controls (the core feature of meta-borders), need to be revised. Currently, through their fixation on territoriality, they allow for the emergence of a double standard, whereby the State may act abroad with impunity in relation to the human rights consequences of its conduct. They ‘permit a State … to perpetrate violations … on the territory of another State, which violations it could not perpetrate on its own territory’, generating ‘unconscionable’ arbitrariness and discrimination between rights holders solely depending on their location.⁴ The consequence is that—contrary to the Rule of Law—States may exploit geographical distance to create and legitimate ethical and legal detachment from their own wrongdoing,⁵ instrumentalising territory to selectively construe human rights liability. This allows them to arbitrarily generate gaps and vacuums that facilitate the expansion of ‘raw’ power—a power untrammelled by any obligations vis-à-vis those affected by States’ conduct and that deprives them of basic legal coverage. Meta-borders thus emerge as an oppressive, autonomy-restricting construct that undoes the essential purpose of human rights. The logic of human rights is not to extend protection selectively depending on location, but to limit the unfettered discretion of public authority subjecting it to basic Rule of Law premises.⁶

Migration controls in the European Union (EU) and elsewhere, particularly in the Global North, have been dispersed and expanded through externalisation techniques, such as visas, carrier sanctions, extraterritorial patrols, and similar methods.⁷ As a result, borders have become ubiquitous, multi-modal, and transnational systems of coercion,⁸ operationalised beyond the physical borders of the Member States. Cooperation and delegation arrangements serve to generate a (geographical and legal) distance vis-à-vis (unauthorised) migrants, effectively blocking their movement ‘upstream’, before they have arrived at the external frontier of the country of intended destination, while denying them legal shelter. If entry occurs, surveillance continues inland, ‘following’ the migrant throughout her journey, tracking her movement at all times, and constructing her as a ‘suspicious’ category to be constantly monitored. The distance that meta-borders generate allows States to disclaim or reduce the applicability of their human rights obligations and to restrict or negate responsibility for any infringements.

The violence implicated in these structures of de-territorialised, preventive enforcement of the border is justified for the sake of deterrence of the security threat that uncontrolled migration presumably represents.⁹ This generates critical challenges for the law. Assigning responsibility to a principal and its agents within multi-actor constellations is rendered difficult,¹⁰ because of the different State and non-State entities that collaborate in the enforcement of controls in extraterritorial settings that complicate causation and attribution lines. In externalisation systems, it is not only the border that ‘moves’¹¹ or ‘shifts’.¹² The geographical distance that externalisation interposes between the locus of power and the locus of surveillance is also used to ‘push away’ (or to simply reject) responsibility,¹³ thereby generating accountability gaps.¹⁴ The entire legal apparatus is (perceived as) ill-equipped to respond to the responsibility dispersion techniques¹⁵ that typically attach to offshored and outsourced means of policing.¹⁶ In regimes of cooperative extraterritorialised controls, while the border expands, legal protections tend to retract, causing a responsibility void—at least, at first sight.

In the following sections, I take issue with this conundrum. I problematise the localisation of the border through its multiple dimensions in Sect. 2, highlighting the significance of its legal and non-legal manifestations, before exploring in Sect. 3 the impact of law on the meta-territorialisation of the sovereign exercises of demarcation, delimitation, and exclusion that ensues. What I mean by ‘meta-territorialisation’ is the process of transformation of the techniques of control that ultimately alter the location, functionality, and morphology of the border. The resulting ‘meta-border’ constitutes a territorially transcendent (legal) construction that detaches enforcement from the physical confines of the geographical frontier. The meta-border is implemented beyond the territory it delineates—irrespective of any legal title expressly allowing States to augment and project their power in this fashion. While it preserves a physical position and a cartographical representation ‘on the map’, retaining a residual territorial feature, externalisation mechanisms of privatisation (e.g. to commercial service providers) and delegation (e.g. to third countries) that outsource and offshore surveillance undo the fixity of the border’s location. The target of these extraterritorialisation techniques are certain categories of (unwanted) non-citizens. And they are utilised for the specific purpose (or, at least, with the distinctive effect) of withdrawing important legal safeguards. Consequently, the locus of the meta-border is no longer the locus of Rule of Law protections, particularly vis-à-vis irregular migrants. The meta-border, crafted by legal fiat, actively re-orders space, determining the (non-)reach of human rights and establishing the circumstances under which they may or may not apply, relying on a variety of fictions and legal artifices. These externalised frontiers—qua mechanisms of migration control—play a productive role. They become the expression of a type of sovereign power that is (supposedly) unmoored from any human rights constraints, leading to the phenomenon that I have called ‘irresponsibilisation’.¹⁷

What is more, the legal process of meta-territorialisation of the border is selective (if not entirely discriminatory) and allows for the tailoring of responses and their adjustment to the characteristics of certain categories of people on the move whose movement is regarded as undesirable and illegal(ised). The same line, the same border, has, in consequence, different implications for different groups of persons. It assumes different functions, becoming a site of discrimination and exclusion for some, while not for others. The examination of the various techniques of ‘contactless control’ through which this is achieved,¹⁸ emerged especially in the aftermath of the ‘refugee crisis’ in the Mediterranean context, is undertaken in Sect. 4. Their effect, as mechanisms that ultimately disown responsibility for any resulting violations, is appraised in Sect. 5, where I criticise the prevailing conceptualisations of jurisdiction in relation to the extraterritorial application of human rights norms, drawing on my previous work.¹⁹ In Sect. 6, I seek to reconcile power with accountability—or ‘State spatial innovations with justice’,²⁰ proposing a ‘functional’ approach. To counter the impact of the ‘irresponsibilisation’ phenomenon, I advance a model based on a parallel meta-territorialisation of responsibility notions, adjusting the focus from the territorial to the functional plane. This is premised on the acceptance that human rights, as fundamental components of systems grounded in the Rule of Law, track and constrain all exercises of sovereign power, aligning control with responsibility. The gist of the argument is that human rights duties ‘follow the border’, wherever and whenever State authority may be exercised.²¹ I recount the advantages of this ‘responsibilisation’ model in Sect. 7, which also concludes and closes these reflections.

2 Locating the Border: From Territory to Functionality

Borders, in the social sciences, have been described as more than simple enclosures demarcating geographical location and sovereign territoriality.²² They are understood as spatially disaggregated and conceptually complex projects of governance and belonging.²³ Instead of static boundaries, they constitute a fluid and shifting policy assemblage, delivered through bureaucratic practices and their continuous enactment and contestation by multiple actors, according to competing rationalities.²⁴ They generate differential power relations as well as hierarchies of inclusion and exclusion, becoming sites of ordering and policing,²⁵ rather than merely ‘lines on a map’. Today’s de-localised borders are more than ‘space makers’; they have become convoluted ‘borderlands’²⁶ or ‘borderscapes’,²⁷ continuously traversed by a plethora of practices, discourses, and imaginaries that constantly (re)produce changing notions of inside/outside, self/other.

Depending on the vantage point, borders assume different functions. They are deemed as historically contingent, multi-dimensional constructs that configure/reconfigure space and social relations depending on place-specific experiences and our engagements with them.²⁸ They are considered cultural artifacts constituted by the everyday uses that different social agents make of them. Marxist theorists, for instance, view borders as a function of capitalism, as a tool in the economic extraction arsenal that facilitates the accumulation of wealth and the perpetuation of material hierarchies across space. They are critical of the (selective) fixity of borders as an aspect that generates dependency and inequality, since they constrain and limit access to resources on a differential (if not, arbitrary) basis according to a logic that is detrimental to labour and favours capital.²⁹ A post-colonial perspective, by contrast, sees borders as the remnants of empire, as the vehicle through which its legacies and continuities are entrenched in today’s political order. Scholars of this tradition highlight the discriminatory character of borders, as the products of history-become-present, signalling persistent forms of political domination working along racial/racist lines.³⁰

Borders indeed have several dimensions. As locational markers of here/there, they ‘contain’ and ‘territorialise’ geographical space³¹ through processes that have cultural, social, political, and economic relevance in the configuration of us/them divisions.³² They have a functional as well as a normative slant, signalling where the boundaries of the State are/should be, while simultaneously constructing them in specific ways that contribute to nation-building and shared cultural affiliation.³³ Borders also have a symbolic character, because they contribute to building/unbuilding national identities, shaping (cross-border) conflict/cooperation systems, and embedding as well as disrupting structural injustices. In that regard, they have a framing aspect that ‘naturalises’ the resulting classifications of people and places.³⁴ Borders are, therefore, multifaceted, and co-implicated in the production of meanings and understandings of reality beyond their cartographical representations.

Legally speaking, borders play a constitutional role, functioning as a polity-building device that designates the place of the demos in relation to a specific, spatially-situated political community.³⁵ In fact, the border is a chief element of State sovereignty. It represents the State’s cartographical power in relation to its ‘own’ territory and population, since (in principle) it designates the outer confines of its authority and jurisdiction. Drawing from the Westphalian order, in international law, sovereignty is indeed defined as ‘supreme authority within a territory’.³⁶ The basilar principles of exclusive jurisdiction and non-intervention, for instance, articulated in the UN Charter,³⁷ considered key attributes of Statehood, pivot around the notion of (bordered) territory. ‘Territorial supremacy’,³⁸ or plenary power and control over all things and persons within that territory,³⁹ is what constitutes State sovereignty and, thus, plenary subjecthood in international law.

Borders, therefore, are fundamental delimitators of in/out in many different respects; they are signifiers that distinguish, filter, and classify spaces, objects, and populations in relation to a designated location that they contribute to define in a geographical, socio-political, and legal sense. They have State-making significance as designating both the place and power dimensions whereby the State (its territory and population) is made. Within this framework, the migrant qua non-citizen, to which the next section turns, is constituted by exclusion, produced by the effect of the very (b)ordering processes that render the border visible and meaningful in its different facets.

3 The Meta-territoriality of Borders: Governing Movement, Making the Non-Citizen

Globalisation, as a system of (supposed) ‘erasure’ of territorial distance and demarcations,⁴⁰ has contributed to the constant (re)negotiation of borders, (re)structuring ideas of space, time, and being across the world. Yet the intervention of security concerns, particularly since 9/11, has redirected the focus of sovereign control exerted through (b)ordering processes—processes of social ordering through bordering interventions—to encompass not only territories, but also populations, especially as and when they move.⁴¹ In this context, border technologies (re)emerge as technologies of biopolitical power that target mobility and transnational circulation as a form of government.⁴² The main function of borders is now to govern movement, rather than simply to demarcate territory. As a result, border ‘policing’, within and beyond borders, has ‘open[ed] up an entirely different spatial configuration of security’ in its relationship to territory and territorial conceptualisations of place,⁴³ which have become ‘flexible’⁴⁴ or ‘elastic’.⁴⁵ Border enforcement mechanisms now track the trajectories of non-citizens throughout their mobility course, from within their countries of origin, even before they start their journeys, up until they reach their destination, and even after they have crossed the geographical frontier of the State concerned, far beyond (and within) the actual borders of the countries of (intended or presumed) destination.⁴⁶ This is not to say that territorial borders have disappeared or lost their relevance. In fact, they have become fortified and militarised as ‘zones of exception’.⁴⁷ High-security walls, barbed wire fences, sensor motion and radar detection technologies have become commonplace markers of today’s physical frontiers.⁴⁸ But the emphasis is less on delimiting space than on precluding unauthorised access. Today’s (meta-)borders embrace a ‘pathological’ vision of human mobility,⁴⁹ as potentially dangerous, threatening the (desired) order(ing) of place and people, justifying their securitisation.

The shift to focusing on populations and their movement as the subjects of control has consolidated biopolitical means of discipline that categorise individuals at different points along the border continuum. The connection of crime and the terrorist threat with cross-border movement (at least at the rhetorical level) has hailed borders as the prime mechanism to prevent the security risks associated with disorderly and unauthorised transborder mobility, so that border control has risen (discursively and in practice) as the main instrument of ‘migration management’.⁵⁰ The securitisation of mobility has led to (b)ordering processes that transform the corporeal bodies of people on the move into a site of surveillance,⁵¹ relying on new (typically unsettled and diffuse) oversight techniques that outsource and offshore sovereign power. These range from measures of non-entrée that impede unauthorised arrivals,⁵² to techniques of ‘dataveillance’ that digitalise border controls through algorithms and AI to triage and exclude ‘risky’ migrants,⁵³ to strategies of ‘crimmigration’ that utilise the means of the criminal justice system for the purposes of deterrence and the prevention of irregular movement.⁵⁴ This transition has given rise to a ‘border overall’ phenomenon that diffuses (the importance of) physical frontiers and instead ‘follows’ the steps of non-citizens across international boundaries in a continuous and lasting mode.⁵⁵

But the ‘border overall’, or the meta-border generated through these processes, does not simply ‘shift’ with the movement of the migrant.⁵⁶ What happens is more pervasive and more fundamental. The end result is that the border inheres within the migrant. It becomes a constant feature of her foreignness, part and parcel of her migrant status. The border thereby ‘makes’ the migrant and renders her (legally) visible and cognisable to the processes of control. It is this configuration of the meta-border that constructs and regulates non-citizens—through their exclusion from citizenship and the full freedom of movement package attached to it. It is this configuration of the meta-border that enables their objectification as a risk and potential threat to be managed through (ubiquitous, status-sensitive) border enforcement.

A ‘constant border’ thus materialises,⁵⁷ latching on to migration status and distinguishing between authorised/unauthorised, wanted/unwanted sorts of mobility. The spatial/territorial border interlaces with the conceptual/categorical borders of belonging, sorting people, allocating rights, and establishing hierarchies of privilege and vulnerability. Thereafter, the (meta-)border infiltrates and defines the legal position of non-nationals, conditioning their possibilities of movement and settlement, and also their claims as rights-holders within the legal order of receiving States. It translates the unceasing ‘otherness’ of the migrant and her permanent differentiation/discrimination from ‘us’. The (meta-)border thus becomes status related, perceptive to nationality, security, and personal circumstances. It is no longer (or, at least, not only) territory based. It becomes individualised, tailor-made to the specific migrant, her background, and the potential risk she may represent in irregular migration terms. In a sense, it is the migrant herself, through her very foreignness, who embodies the (meta-)border and carries it with her.

In relation to these meta-borders, law plays a paradoxical, dual role that tends to be under-appreciated in legal literature,⁵⁸ as the tool through which certain conceptualisations of the border are institutionalised (and de-territorialised), demarcating legal(ised) from illegal(ised) transgressions. While law defines and legitimises borders (and the different modalities for their control), it can also undo them ‘from within’, providing the means to challenge the violence and exclusion implicated in border-making, appealing to dignity, human rights, and the Rule of Law. Simultaneously, therefore, the law constitutes a means of subordination and a tool of emancipation. The law is, at the same time, the maker and marker of borders, and the potential destabiliser and unraveller of the (b)ordering and othering constructs that envision, maintain, and may erase borders. The law promotes and sustains certain technologies, shapes operational responses, and justifies prevailing approaches to control as articulated in the multiple sites and scales in which (meta-)borders manifest.

Legally-mediated and legally-constituted borders (and the means of their enforcement) consolidate in mechanisms of control that provide differential access to systems of belonging, mobility and well-being. These fracture and diversify their impact by discriminating citizens from (certain categories of) non-citizens, with deep repercussions for the manner in which borders are ‘lived’ and experienced by people on the move.⁵⁹ In this sense, it is the law that renders the border meta-territorial and that legitimises the multifarious processes of control that constitute the resulting meta-border. It is the law that bans or allows certain forms of violence. It is the law that extends or retracts legal safeguards to/from certain locations, including or excluding individuals (or entire populations) from the scope of human rights and fundamental guarantees, stratifying dignity and access to legal protection. The extraterritorialisation of borders with which this Special Issue is concerned occurs through the law that tactically extends the reach of control whilst limiting the scope of (concomitant) obligations. But there is nothing ‘natural’ in this de-coupling. As the next sections unveil, it is the law that configures (de-)territorialised understandings of jurisdiction and that permits or forbids certain mechanisms of ‘remote control’ of migrant populations⁶⁰—typically on a selective, if not self-serving, basis.

4 The (Functional) Meta-Border: Extending Power through ‘Contactless Control’

Techniques of ‘contactless control’,⁶¹ as ‘movable legal barriers’,⁶² entail the enhancement of the State’s regulatory and operational reach, enabled by a malleable conceptualisation of legal spatiality that expands sovereign power, but without a correlative extension of sovereign responsibilities.⁶³ These techniques seek not only to deter, but also to preclude the movement of those identified or perceived as unwanted migrants—to whom the law has left only irregular mobility options through unsafe, unauthorised routes that typically engage smugglers or traffickers as facilitators. These techniques aim to proactively impede arrivals—and strive to equally impede the engagement of human rights obligations, by outsourcing border management to third parties, including partner States or private contractors. They are typically designed as strategies of collaborative containment that enlist the authorities or (semi-official) agents of countries of origin and/or transit, including paramilitaries, commercial carriers, transport companies, private security personnel and others, to cooperate in the implementation of controls, acting as surrogate enforcers of the border measures of the State(s) of destination. In this way, the (meta-)border is projected outwards to meet unwanted migrants at the point of movement with a view to blocking it before it takes place—or, in any case, before the person concerned succeeds in reaching the physical frontiers of her country of destination. At the same time, however, Rule of Law safeguards are given a strictly territorial construction, as applying solely within the geographical limits of the State concerned. The meta-border selectively and simultaneously expands and retracts to maximise power while minimising responsibility, manufacturing unaccountable spaces and mechanisms of control.

The control machinery of (b)ordering processes (mediated and constituted through law) is thus deployed vis-à-vis the (presumed) unwanted migrant already from within the country of departure or as close as possible to its territorial confines (on behalf and for the benefit of the country of destination). ‘Pullbacks’,⁶⁴ offshore detention/interdiction measures, and pre-emptive ‘rescue’ (or capture⁶⁵) at sea convert (pre)entry into (pre)exit controls, with the result of de facto neutralising the (effectiveness of the) rights to leave any country including one’s own⁶⁶ and to seek asylum,⁶⁷ to which non-citizens are entitled under international and EU law.⁶⁸ Execution is delegated and effected by proxy, but in accordance with the desires and instructions of the delegating State.⁶⁹ There are different iterations, but all these practices, in the words of the UN Special Rapporteur on the Human Rights of Migrants, González Morales, ‘manifest an entrenched prejudice against migrants and demonstrate a denial of States’ international obligations to protect [their] human rights’.⁷⁰ Characterised by an absence of an individualised assessment of their personal situations, no procedural safeguards or due process guarantees, they result in the summary expulsion to or violent containment within territories where the risks of harm, persecution, or ill-treatment are paramount.⁷¹

Dedicated financial and technical support to third countries by countries of destination in exchange for foreclosing (unwanted) departures is one of the main techniques of ‘contactless control’ deployed by the Global North, including particularly the EU Member States.⁷² Development aid and humanitarian assistance, visa facilitation and similar enticements are generally offered in return for the deputised enforcement of border exclusions. The objective is not to prevent entry but to impede exit at the very beginning of the migratory journey. These measures are all oriented towards curbing human trafficking and combating migrant smuggling and are (discursively) justified as based in a humanitarian concern for dignity and safety. Policy makers tend, indeed, to deviate attention towards ethically desirable goals,⁷³ like saving the lives of those in distress at sea and the prevention of injury during dangerous journeys where unwanted migrants are at risk—especially of abuse by unscrupulous mafias portrayed as the main source of harm. The structural forces underpinning the (b)ordering processes at play and that put people on the move in danger—which Grundler has called the ‘route causes’ of displacement⁷⁴—are not normally identified as constituting or compounding the threats they face and that force them into unsafe travel paths. The meta-border assemblage is portrayed as ‘neutral’, as a given, as part of the normative landscape sitting in the background that enables transboundary movement in the first place (supposedly through legitimate means and for legitimate ends). The thwarting of movement that ‘contactless control’ techniques involve is hence presented as a benevolent tool for the preservation of the integrity and security of those concerned; their broader human rights implications, beyond an immediate preoccupation with people’s (bodily) survival (through containment), are however normally neglected.⁷⁵

The coercive nature of the meta-border is, in fact, rarely acknowledged⁷⁶; it is rather considered an assortment of innocuous regulations and enforcement techniques, covered by the sovereign prerogative to govern access to national territory. However, if, as some political theorists maintain, the legal apparatus on which it rests is coercive per se—because law would not be law otherwise; it would not ‘officially’ command behaviour under the State’s authority—then the meta-border, as a legal construct, should be seen in the same light.⁷⁷ Insofar as the meta-border imposes the will of the State(s) and restricts choice structures on a non-consensual basis, in a way that reduces individual autonomy, its coercive nature should be beyond doubt.⁷⁸

In this connection, an equally problematic and seldom discussed element, is the fact that, while States have a recognised right to control their (territorial) borders, nowhere has it been bestowed on them by international law the power to govern the mobility of individuals throughout the whole of the world’s surface. This is why meta-borders, and their indirect ways of controlling trans-national journeys, through the engagement and cooperation of multiple actors, constitute an ingenious (and supposedly legally inoffensive) mechanism to project State power across jurisdictional boundaries without specific legal title.⁷⁹ The interfacing of various entities implementing single/several components of the meta-border that, taken separately, are legal/permissible, offers a great advantage. This ductile nature of the meta-border, severing and (re)combining different actions, allows States to exert control from afar by executing only part of the intervention(s) that impede (unwanted) movement, often without ever encountering the persons concerned, while (apparently) remaining within the confines of legality.

Recent modalities of ‘contactless control’ measures, that have become routine,⁸⁰ include mechanisms like ‘aerial refoulement’,⁸¹ consisting of refoulement practices assisted by aerial assets of destination States that detect migrant boats and relay location information to third country authorities for their intervention. This results in the interception and return of the persons concerned back to the point of (supposed) departure. The intermediation of the organ of an international organisation, such as the external frontiers agency of the EU (Frontex), functioning as a refoulement broker, offers yet another possibility of contactless control. According to the European Commission,⁸² Frontex has long been transmitting relevant details of shipwrecks and unseaworthy vessels carrying migrants to what they consider ‘the responsible RCC [Rescue Coordination Centre]’ in Libya, thus facilitating pullbacks by Libyan actors. The agency has, in fact, been found to have overstepped the limits of its powers, pursuing activities of dubious legality, being recently condemned by several entities for its failure to process non-citizens’ data in accordance with EU law and to comply with ‘its own mandate’.⁸³ Other variants of this practice rely on merchant or fishing vessels to carry out ‘privatised’ pushbacks, following guidelines provided by the authorities of the country of intended destination, carried out in collaboration with a country of origin/transit to which the migrants are forced back.⁸⁴ Scenarios implicating Italy, Malta, and Libya have been denounced by multiple observers.⁸⁵ In all cases ‘impunity … is prevalent’.⁸⁶

The most dangerous types of these techniques, falling within the family of measures of so-called neo-refoulement that instrumentalise geography to restrict access to territory and legal protection,⁸⁷ are accomplished ‘by omission’.⁸⁸ They typically involve a constellation of active and passive behaviours coalescing in ‘composite’ conduct that, ‘in aggregate’,⁸⁹ generates harm—if not outright loss of life. These practices have flourished and acquired new proportions, particularly in the Mediterranean in the aftermath of the ‘refugee crisis’. They could be defined as a sort of necro-policy⁹⁰ that integrates mechanisms ‘based on deterrence, militarisation and extraterritoriality’, which deliberately incorporate or, at least, tolerate ‘the risk of migrant deaths as part of an effective control of entry’.⁹¹ This modality is based on the negation of rescue (and the associated increased probability of dying), including through the outright abandonment at sea of persons in distress,⁹² the removal of naval assets from naval missions, or the restriction of operational areas covered by maritime operations to avoid direct encounters with potential ‘boat migrants’ that require the provision of assistance.⁹³

While ‘pushbacks’ are typically performed by the authorities or agents of a country of destination to impede arrivals, ‘pullbacks’—in any of their manifestations—are undertaken by the authorities or agents of a third country of origin or transit to counter departures at the behest of the authorities of the country of destination. By transferring the coercive management of border checks to third countries there is a legally prominent effect that ‘contactless control’ techniques achieve: they eliminate any direct physical contact between the third-country nationals concerned and the authorities of the countries of intended destination. Ultimately, if not completely severed, the jurisdictional link that would normally give rise to human rights obligations (and responsibility in case of any violations) becomes blurred and difficult to determine.⁹⁴ The advantage of ‘contactless control’ mechanisms for EU countries (and other Global North States with similar concerns and the resources to implement them) is that they allow them not to perform any containment directly or by themselves. It is the organs or agents of partner countries that carry out the maritime interceptions, exit denials, and migration detentions that keep unwanted third-country nationals at bay within their territorial domains or close by—in any event, far away from the territorial jurisdiction of the delegating State.

When Turkey, under the EU-Turkey Statement,⁹⁵ or the Libyan Coastguard, under the Italy-Libya MoU,⁹⁶ prevent departures or undertake pullbacks at sea, they are the ones directly performing the relevant acts that may involve international legal responsibility.⁹⁷ The point, however, is to elucidate whether their interventions may be characterised as proxy actions/omissions of the EU countries with which they collaborate, and on whose behalf, or for whose benefit, they undertake them. This requires an evaluation of the prevailing (border-tight, territorially-bounded) conceptions of ‘jurisdiction’ in relation to (extraterritorial) human rights obligations, which I undertake in the next section. This sets the ground, as a preliminary step, to explore the alternative Rule of Law-based ‘functional’ model proposed to avoid ‘irresponsibilisation’.

5 Retracting Obligations: Territorial(ised) Jurisdiction and ‘Effective Control’

Collaborative (meta-)border control infrastructures are vested with the capability to (violently) coerce at a distance. As already mentioned, externalised mechanisms of ‘contactless control’ negate access to territory and protection regimes in countries of destination at an early stage, before any physical contact has been made with the authorities of the State concerned. They are, indeed, matched with responsibility dispersion mechanisms that either diffuse or negate legal accountability.⁹⁸ ‘Responsibility diffusion’ relates to the relational trait of externalisation mechanisms and the multi-actor constellations they tend to involve, which obfuscate the attribution of conduct and causation determinations in the legal liability chain, impeding the establishment of responsibility. This is because of the multiplicity of agents and organs of different States and organisations intervening in ‘contactless control’ collaborations. Conversely, ‘responsibility denial’ refers to situations in which the main beneficiary of ‘contactless control’ ventures openly disclaims any responsibility for any potential/occurrent violations, submitting that it lies instead with the executing actor(s)—to which it has delegated the remote enforcement of its border measures. The ultimate effect in both scenarios is the ‘irresponsibilisation’ of the State(s) of destination.⁹⁹

Irresponsibilisation, as a legal phenomenon, is facilitated by the reconfiguration of the relationship between law and territory that meta-borders entail. It is allowed by the disjuncture that exists (or is actively interposed) between understandings of sovereign power, which is routinely projected abroad in various forms and fashions—including through legal means of ‘remote’ and/or ‘contactless’ control—and conceptualisations of ‘jurisdiction’ for the specific purposes of human rights compliance, which typically remain bound by territorial conceptions of State authority and responsibility. Prevailing international human rights law interpretations propound an understanding of ‘jurisdiction’ as ‘primarily territorial’, while extraterritorial manifestations are considered ‘exceptional and requiring special justification’.¹⁰⁰Where the State acts has become more significant than the fact that it is the State that is acting, whatever the location.¹⁰¹ This is especially the case for the European Court of Human Rights (ECtHR/Strasbourg Court), which has the richest, most detailed, and influential body of case law on the extraterritorial reach of human rights obligations under the European Convention on Human Rights (ECHR).¹⁰² The core idea, defended by some individual judges, that ‘jurisdiction means actual authority, that is to say, the possibility of imposing the will of the State on any person, whether exercised within the territory … or outside’,¹⁰³ does not hold in the post-Banković landscape. The basic intuition that jurisdiction amounts to the exercise of ‘political power’ over ‘human activities made relevant by human rights treaties’ is not the line followed by the Court in extraterritorial situations.¹⁰⁴ The Court has elaborated a case-by-case piecemeal approach, which does not always seem to tally with the Rule of Law foundation underpinning the system of international (human rights) law. Overall, it has embraced a narrow construction of the circumstances that may trigger the extraterritorial applicability of the ECHR and the determination of responsibility for any related violations of Convention obligations. The two steps—the ascertainment of the existence of an exercise of jurisdiction, followed by the establishment of responsibility for any infringements—are not systematically differentiated. The Court has recently admitted that ‘there may be areas of overlap’ that render the two steps undistinguishable ‘in so far as the Court is invited to examine whether any acts of the perpetrators are to be attributed to the State in the context of its jurisdiction assessment’.¹⁰⁵ In these situations, the Articles on the Responsibility of States for Internationally Wrongful Acts (ARSIWA),¹⁰⁶ codifying customary international law on the matter, are said to be ‘clearly relevant’ and to provide a source of inspiration.¹⁰⁷

However, the Court’s adherence to ARSIWA has not been fully consistent. While the ARSIWA’s provisions could lead one to believe that State conduct of any sort should per se be considered a manifestation of State jurisdiction, whether taking the form of ‘legislative, executive, judicial or any other functions’,¹⁰⁸ whatever the character or position of the State organ or agent concerned,¹⁰⁹ and regardless of its exercise inland or abroad, this is not what the Strasbourg Court recognises. The Strasbourg Court recognises only two main models where the necessary threshold may be attained for the triggering of ECHR obligations in extraterritorial settings: the ‘State agent authority’ or personal model and the ‘control over an area’ or territorial model.¹¹⁰ An incipient third model, which can be denominated as ‘due diligence’ jurisdiction, is progressively taking hold in situations regarding the procedural obligation under the right to life, whereby a ‘jurisdictional link’ may be established ‘between the respondent State and the victim’s relatives’.¹¹¹ Each of the models, as explained below, has its own limitations and incongruities.

5.1 The Territorial, Personal and Due Diligence Models

The territorial model refers to scenarios in which jurisdiction takes the form of State military action abroad¹¹²—whether with or without legal title.¹¹³ Human rights duties derive in these situations from ‘the fact of such control’ as the State may exert over foreign territory, either directly or via a subordinate local administration.¹¹⁴ Establishing whether jurisdiction has been exercised in a human rights-relevant way is seen as a ‘question of fact’, depending on the strength of the military deployment of the State concerned or the level of ‘decisive influence’ wielded over the proxy actor through which control is maintained.¹¹⁵ There is an equation between extraterritorial jurisdiction and the notion of ‘effective control’—shaped by de jure and/or de facto elements—which also applies in the context of the personal model. Recently, though, the Court has considered bombing and shelling during active hostilities between Georgia and Russia as insufficient to establish jurisdiction (by themselves). According to the Court, ‘the very reality of armed confrontation and fighting between enemy military forces seeking to establish control over an aera in a context of chaos means that there is no control over an area’.¹¹⁶ Rather than taking violence as a sign of ultimate (potentially lethal) control over the individuals and places affected,¹¹⁷ the Court interpreted it as a lack thereof. And instead of considering the relationship between the State(s) and the individuals within their might—which is the relationship that matters under Article 1 ECHR,¹¹⁸ the Court heeded the relationship between the competing sovereigns vying to establish control over the area in dispute—which is not determinant of the jurisdictional assessment under the Convention.¹¹⁹ The judges attached such ‘decisive weight’ to the prevailing violence and the ‘context of chaos’ that they also precluded the activation of the personal model of jurisdiction.¹²⁰

There are at least three variants of the personal model that the Strasbourg Court has accepted in its jurisprudence. The first relates to the acts of diplomatic and consular agents ‘present on foreign territory in accordance with provisions of international law’ whenever they ‘exert authority and control over others’, thus potentially emphasising aspects of de jure jurisdiction.¹²¹ However, in later jurisprudence, the Court seems to have limited the relevance of de jure factors to situations where State representatives, when acting ‘in their official capacity’, ‘exercise abroad their authority in respect of that State’s [own] nationals or their property’.¹²² In relation to non-citizens, these same agents need to ‘exercise physical power and control’ for the establishment of State jurisdiction—de jure authority alone does not suffice.¹²³ The State’s Convention obligations are thus modulated depending on nationality—paying no attention to the discriminatory effect that this implies.

The second situation relates to State acts that amount to an exercise of ‘public powers normally to be exercised by [the domestic] Government [of the foreign country concerned]’, if this is underpinned by ‘the consent, invitation or acquiescence’ of the territorial sovereign, and provided that ‘the acts in question are attributable to it [i.e. the ECHR party] rather than to the territorial State’.¹²⁴ This is the paradigm that was applied, e.g., in Jaloud, where an individual died after being shot at a military checkpoint manned by Dutch soldiers in Iraq.¹²⁵ Although the Netherlands was not an occupying power, it was officially supporting the US and the UK, so its national forces were considered to be ‘asserting authority and control over persons passing through the checkpoint’¹²⁶ in a way similar to the ‘public powers’ model of Al-Skeini.¹²⁷

However, what tends to be ‘decisive’ in the determination of extraterritorial jurisdiction under the personal model (necessary for the triggering of Convention obligations, without which the ECHR is not deemed to apply) is ‘the exercise of physical power’ over individuals abroad.¹²⁸ This constitutes the third and most frequent variant of the personal model, underscoring the importance attached to de facto elements of control by the Strasbourg Court. This variant has indeed been accepted in cases concerning the arrest/detention,¹²⁹ abduction,¹³⁰ or extradition/surrender of persons abroad.¹³¹ In exceptional situations entailing what Strasbourg denominates ‘an element of proximity’—without clarifying whether this is to be understood in geographical or causal terms—‘isolated and specific acts of violence’ have also been construed as being covered under the personal model.¹³² This means that ‘beating or shooting by State agents of individuals outside that State’s territory’,¹³³ as well as ‘the extrajudicial targeted killing of an individual by State agents’,¹³⁴ may qualify as an exercise of jurisdiction. Yet the Court appears to accept this only in situations where the killing takes place ‘in the territory of another contracting State’,¹³⁵ which generates an arbitrary distinction between extrajudicial targeted killings perpetrated in a fellow ECHR party vis-à-vis extrajudicial targeted killings executed in a non-ECHR party. The rationale for this discrepancy has not been stated, which in any event is hard to reconcile with the Rule of Law.

A third emergent model of extraterritorial jurisdiction arises from the investigative obligations attaching to the right to life, when ‘a death occurs outside the territory of the Contracting State in respect of which the procedural obligations under Article 2 [ECHR]’ are considered to apply.¹³⁶ According to the Court, ‘the procedural obligation to carry out an effective investigation under Article 2 has evolved into a separate and autonomous obligation that can be considered to be a detachable obligation arising out of Article 2 and capable of binding the State even when the death occurred outside its jurisdiction’.¹³⁷ In such situations—and in a somewhat circular reasoning—‘if the … authorities of a Contracting State institute … their own criminal investigation or proceedings concerning a death’, whether of their own free will or possibly out of a sense of duty ‘by virtue of their domestic law’, then ‘this may in itself be sufficient to establish a jurisdictional link’.¹³⁸ When ‘no such investigation or proceedings have been instituted … special features [that the Court has not defined] may [still] trigger the existence of a jurisdictional link’.¹³⁹ Whether such ‘special features’ exist or not ‘will necessarily depend on the particular circumstances of each case’,¹⁴⁰ which inhibits predictability. The suspects of a murder fleeing to an area within State jurisdiction, thus preventing the territorial sovereign from conducting its own investigation¹⁴¹; a failure to cooperate with the territorial sovereign, who had sought to institute proceedings¹⁴²; or principles of customary international law intervening in the case, coupled with a lack of resources on the part of the territorial sovereign to investigate the deaths by itself,¹⁴³ have been deemed to constitute ‘special features’ in this regard. In most cases the obligation is established to somewhat compensate for the lack of capacity of the territorial State to perform its own investigation, but on the understanding that the ‘compensating’ ECHR party does not have substantive obligations of its own, nor has it even established jurisdiction rendering Article 2 ECHR directly applicable on its behalf, with the main action concerned not being attributable to it either.¹⁴⁴ How a Contracting State can be held responsible for investigating an incident not attributable to it and which, otherwise, did not come within its jurisdiction under Article 1 ECHR is striking. Why such a special jurisdictional link and autonomous obligation should exist only for the purposes of investigating death, rather than to protect life when still possible also remains unexplained. Such an expansive, if incongruous, understanding of positive duties in this framework clashes with the approach pursued by the Court in other domains.

5.2 Missing Definitions

Silences and inconsistences constitute a main limitation of the Strasbourg jurisprudence. The exact impact of de jure and de facto bases for the establishment of extraterritorial jurisdiction remains unclear, as does the degree of directness of the control to be exerted and whether physical contact is ever/always essential, especially when non-citizens are concerned. There are some examples involving maritime blockades¹⁴⁵ or the forcible rerouting of a foreign ship where the jurisdictional link was not contested,¹⁴⁶ even if no direct physical contact had been made between the State and the individuals affected (or their vessel). Yet, in other situations of killing from a distance, including through a pre-planned, NATO-sanctioned bombing operation,¹⁴⁷ the link has not been deemed to have been established, so the Convention was said not to apply and no responsibility could be determined.

In truth, the Court has never provided a detailed definition of ‘jurisdiction’ or ‘effective control’. Neither has it specified the criteria on which they depend on a general basis or explained the principles that sustain these doctrines—a position which has attracted criticism, even from within the Court itself,¹⁴⁸ especially on account of the role these notions have been made to play as a ‘threshold’ without which the Convention is considered not ‘activated’.¹⁴⁹ Indeed, without jurisdiction—understood as ‘effective control’ in extraterritorial scenarios—ECHR obligations are regarded as not being triggered at all and as not binding the conduct of State parties, whatever their consequences on the ground. The Court has recently asserted in this connection that ‘Article 1 of the Convention [does] not accommodate the theory that anyone adversely affected by an act imputable to a Contracting State, wherever in the world …, is thereby “brought within” the “jurisdiction” of that State’.¹⁵⁰ Such an understanding—in the eyes of the Court—would be ‘tantamount to equating the determination of whether an individual falls within the jurisdiction of a Contracting State with the question of whether that person can be considered to be a victim’ under the Convention.¹⁵¹ This contrasts sharply with the approach followed by the former European Commission on Human Rights, who espoused a construction of Article 1 ECHR as being triggered, precisely, whenever individual rights were ‘affected’, considering that this made them fall within the State’s ‘authority’.¹⁵²

This is why, amid this ambiguity and to counter related risks of arbitrariness, my proposal is to embrace a ‘functional’ approach to jurisdiction that matches the equally functional nature of (meta-)borders, giving rise to the applicability of human rights obligations whenever and wherever State power is exercised in the performance of border control ‘functions’. This interpretation, as expounded in the next section, is supported by elements of extant interpretation by other international Courts and Treaty bodies, and even by parts of the Strasbourg’s own jurisprudence. A rationalisation of these components in light of the basic tenets of a Rule of Law-based construction of ‘jurisdiction’ motivates this effort.

6 ‘Functional’ Jurisdiction: Towards a Responsibilisation Model

I understand ‘functional’ jurisdiction as stemming from the governmental ‘functions’ through which the use and deployment of State sovereignty manifests itself in individual cases. What the Strasbourg Court calls ‘public powers’¹⁵³—which can be discharged via legislative, executive, judicial activity, or a combination thereof—are recognised vehicles of State authority. A ‘functional’ approach to jurisdiction, therefore, builds from that and takes into account the underlying sovereign-authority nexus that connects the ECHR State Party to those within its power—irrespective of the manner in which and the place from which that (public/State) power may be exercised.¹⁵⁴ What matters is the public character of the power being exerted, the fact that it is the State that is acting. The basic premise is that human rights obligations—as Rule of Law-based guarantees against arbitrariness—track sovereign jurisdiction and require that any exercises thereof be performed in conformity therewith, regardless of the location. This requires that ‘human rights be [placed] at the centre of [State] efforts to address migration in all its phases’, accepting their applicability in ‘any operations [by States] where they exercise effective control or authority over an area, place, individual(s) or transaction’ in whatever way this may take place.¹⁵⁵

From an international legal perspective, domestic legislative action ‘express[es] the will and constitute[s] the activities of States, in the same manner as do legal decisions or administrative measures’.¹⁵⁶ Adjudicative, administrative, and legislative measures amount to an expression of State jurisdiction, denoting a concrete implementation of State sovereignty, which should, accordingly, be understood to equally trigger State responsibility in case of a violation of pre-contracted obligations.¹⁵⁷

What determines responsibility in this framework is an instantiation of effective control understood not solely on the basis of the intensity or directness of any physical force that may be applied. Grounded in a construction of jurisdiction as ‘functional’, what makes control effective is its material impact to determine the course of events unlocked by its exercise, even when the relevant activity of the State is carried out from a distance and applying ‘minimal’ coercion.¹⁵⁸ In this context, the effectiveness of control is to be judged against its influence on the ensuing situation and the (legal/factual) position in which those affected by it are left as a consequence of that control. This means that both de facto elements (like force) and de jure factors (like legal title) should be considered in tandem as the conduits of expression of ‘public powers’ by the State that it may deploy via legislative, executive, or adjudicative activity (including ultra vires). Whenever that activity—once undertaken and actually carried out in the real world—interferes with the enjoyment of the rights that ‘[ECHR] Parties shall secure to everyone within their jurisdiction’,¹⁵⁹ the relevant obligations should be considered applicable and constrain State conduct.

6.1 Drawing Inspiration from the Law of the Sea

Several international Courts and Tribunals dealing with different branches of international law embrace an analogous position on jurisdiction and its ‘functional’ reading. They understand it (often implicitly) as a threshold criterion that determines the applicability of the relevant Treaty obligations, with any concomitant violations leading to the establishment of legal responsibility. For example, the Norstar ruling by the International Law of the Sea Tribunal (ITLOS) follows this interpretation,¹⁶⁰ when judging on a case engaging the UN Convention on the Law of the Sea (UNCLOS).¹⁶¹ The issuance of a decree of seizure by Italy vis-à-vis a foreign ship on the high seas suspected of illegal activity was taken as an exercise of jurisdiction sufficient to engage Italy’s responsibility for the resulting outcome (i.e. the unlawful arrest that followed, which was subsequently enforced by a partner country within Italy’s territorial waters), not because it produced physical control by itself, but because it generated the conditions for its actual (wrongful) enforcement.¹⁶² It was the combined force of the decree issued by Italy, coupled with its subsequent enforcement, performed by Spain on its behalf, alongside the actual seizure of the vessel by the Spanish authorities, that engaged Italy’s jurisdiction and determined its responsibility in the eyes of the Tribunal.¹⁶³ It was not Spain, but Italy which was considered responsible for the final outcome—Spain had acted upon Italy’s request, by Italy’s delegation and as its proxy, in this case. The sovereign-authority nexus that the action had generated was vis-à-vis Italy, as the ordering State for whose benefit the arrest was to be effectuated (even if unlawfully, as ITLOS concluded in the end). The ensuing breaches of the freedom of navigation principle and related UNCLOS provisions were attributable to Italy on account of the ‘contactless control’ it had deployed and exercised via Spain in relation to the ship concerned.

The Tribunal specifically considered whether direct ‘physical interference or enforcement [action]’ impinging on the freedom of navigation applicable on the high seas was necessary for an exercise of jurisdiction to be identified.¹⁶⁴ It concluded that those elements were not essential and that ‘any act which subjects activities of a foreign ship on the high seas to the [authority] of States other than the flag State’ not only amounts to an exercise of jurisdiction but also ‘constitutes a breach of the freedom of navigation’.¹⁶⁵ In the Tribunal’s view, the principle of exclusive flag State jurisdiction governing the high seas (Art. 92 UNCLOS) ‘prohibits not only the exercise of enforcement jurisdiction on the high seas by States other than the flag State but also the extension of their prescriptive jurisdiction to lawful activities conducted by foreign ships on the high seas’.¹⁶⁶ For the Tribunal, ‘if a State applies its criminal and customs laws to the high seas and criminalises activities carried out by foreign ships thereon’, it not only (unduly) exercises its jurisdiction, but the action ‘would [equally] constitute a breach of article 87 of the [UN Law of the Sea] Convention, unless justified by the Convention or other international treaties’.¹⁶⁷ And this remains the case ‘even if the State refrained from enforcing those laws on the high seas’ and did so only subsequently and within territorial waters.¹⁶⁸ While the locus of enforcement may be relevant in the determination of the existence of a violation, according to the Tribunal, this does not constitute the ‘sole criterion’,¹⁶⁹ nor is it always decisive to establish the (separate, though related, question of the) applicability of the Convention in the first place. Accordingly, the Tribunal considered that the ‘Convention [i.e. UNCLOS] [was] applicable in the present case’ and, in addition, it concluded that ‘Italy, by extending its criminal and customs laws to the high seas, by issuing the Decree of Seizure, and by requesting the Spanish authorities to execute it—which they subsequently did—breached the freedom of navigation [provision in Art. 87 UNCLOS]’.¹⁷⁰ The ultimate authority under which and for the benefit of whom the conduct was performed was Italy, so it was Italy’s jurisdiction and responsibility that was engaged.

6.2 Examples within the Human Rights Domain

In the human rights field, many have adopted a similar, arguably equally ‘functional’, approach—in the sense that I posit. The Inter-American Commission on Human Rights (IACHR), for instance, has espoused a ‘causal link’ model. In the Alejandre case, Cuba was deemed to have exerted sufficient control (and thus jurisdiction) through the shooting down of two aircraft flying beyond its domestic aerial space because ‘the victims died as a consequence of direct actions of agents of the Cuban State’.¹⁷¹ The Inter-American Court on Human Rights (IACtHR) has endorsed and generalised this understanding, concluding that ‘a person is under the jurisdiction of the State … if there is a causal link between the action that occurred within its territory and the negative impact on the human rights of persons outside its territory’.¹⁷² What counts is that the State ‘exercises effective control over the activities carried out that caused the harm and consequent violation of human rights’,¹⁷³ which entails an active decision to do/not do on the part of the authorities that amounts to an expression of their (public) power.

The Committee Against Torture (CAT) has also adopted this model in a case concerning the interdiction of a migrant at sea by the Spanish authorities and his subsequent abandonment close to the Moroccan shore. Having ordered him to jump overboard, disregarding pleas that he could not swim and allowing him to drown, the Committee concluded that the ‘undeniable cause-effect relationship … between Mr Sonko’s death and the actions [and arguably also the omissions] of the … officers’ established the jurisdictional link that triggered the applicability of the Convention against Torture and determined its violation.¹⁷⁴

The Human Rights Committee (HRC) has adopted an elaborate stance in this regard, expanding on the causal-link approach to construct an impact-based conception of jurisdiction. The Committee considers that Parties to the International Covenant on Civil and Political Rights (ICCPR)¹⁷⁵ are subject to ‘respect and to ensure the Covenant rights to all persons who may be within their territory and to all persons subject to their jurisdiction’, understood to encompass ‘anyone within the power or effective control of that State Party, even if not situated within the territory of the State Party’.¹⁷⁶ In this context, both ‘power’ and ‘effective control’ count towards the determination of jurisdiction, as equivalent manifestations of the exercise of State sovereignty with relevance to human rights adjudication. The Committee has built on this general premise, in relation to the right to life, to specify that ‘subject to its jurisdiction’ should be construed to include ‘all persons over whose enjoyment of the right to life [the State Party] exercises power or effective control’.¹⁷⁷ As a result, ‘[t]his includes persons located outside any territory effectively controlled by the State, whose right to life is nonetheless impacted by its … activities in a direct and reasonably foreseeable manner’.¹⁷⁸ Such ‘activities’—performed whether as actions or omissions of a legislative, executive, or adjudicative character—constitute expressions of the State’s ‘functions’ and, therefore, of its ‘jurisdiction’ for the purposes of establishing the applicability of the ICCPR and its potential violation.

This ‘impact model’ draws from pronouncements in previous decisions, where the Committee considered the ICCPR to be applicable and the State Party concerned to be responsible when its conduct constituted ‘a link in the causal chain that would make possible violations in another jurisdiction’.¹⁷⁹ The mere ‘risk of an extra-territorial violation’ was deemed to trigger the action of the Covenant and to possibly lead to a violation if it could be considered ‘a necessary and foreseeable consequence … judged on the knowledge the State party had at the time’ of the events.¹⁸⁰ So, knowledge of and predictability of the potential impact of State conduct of whatever kind not only activates Covenant obligations but must also be taken into consideration in both the jurisdictional and responsibility assessments.

This is why failing to adequately respond to a distress call by individuals in danger of being lost at sea was determined to amount to an exercise of jurisdiction that violated the Covenant. In S.A. v. Italy, ‘negligent acts and omissions in the rescue activities’ of the defendant ‘directly affected’ the situation of the individuals concerned ‘in a manner that was reasonably foreseeable’, especially considering Italy’s duties under the rescue Conventions.¹⁸¹ The subjection of the individuals affected to Italy’s jurisdiction was through this ‘impact’, which ensued from a combination of de jure and de facto elements, including ‘the initial contact made by the vessel in distress with the [Rome] MRCC, the close proximity of [the Italian warship] ITS Libra to the vessel in distress, and the ongoing involvement of the [Italian authorities] in the rescue operation as well as relevant legal obligations incurred by Italy under the international law of the sea’.¹⁸² All these elements combined generated ‘a special relationship of dependency’ between the victims and the Italian State. Such a holistic understanding of Italy’s jurisdiction is what led to the establishment of its responsibility.

Duties under the Search and Rescue (SAR) and Safety of Life at Sea (SOLAS) Conventions were crucial de jure components to determine the jurisdictional nexus vis-à-vis the persons in distress.¹⁸³ These Conventions conceptualise the jurisdictional link expressly in a capability-based manner, requiring ‘[t]he master of a ship at sea which is in a position to be able to provide assistance … to proceed with all speed to their assistance’.¹⁸⁴ The duty starts to apply ‘on receiving information from any source that persons are in distress at sea’,¹⁸⁵ whether a distress call, data from radar detection, or a visual of the vessel in peril. This replicates the terms of the customary rescue obligation codified in Article 98 UNCLOS, according to which ‘[e]very State shall require the master of a ship flying its flag, in so far as he can do so without serious danger to the ship, the crew or the passengers to render assistance to any person found at sea in danger of being lost [and] to proceed with all possible speed to the rescue of persons in distress …’.¹⁸⁶ Such an understanding may well amount to ‘collapsing the ability to engage in … search and rescue … with the notion of jurisdiction’ in this domain,¹⁸⁷ but it is what the maritime Conventions explicitly require (per the contracting States’ own accord) to preserve human life at sea and maintain the effectiveness of the rescue regime.¹⁸⁸

This capability-based approach to jurisdiction (as a threshold criterion activating international obligations and determining their possible violation) is not unheard of in other areas of international law where vital interests are at stake. The Committee on the Rights of the Child (CRC) has recently ‘broadened’ the concept to consider France’s ‘capacity’ and ‘power’ to protect the rights of several children detained in Syria with their parents as akin to ‘jurisdiction’. A failure to offer effective consular assistance and to repatriate them as a means to avoid the risk of harm was considered an expression of France’s jurisdiction by omission that also established its responsibility under the Convention on the Rights of the Child.¹⁸⁹

Control over risks and sources of harm has been determinant in another case concerning the climate change impact of State decisions on children’s rights. In a ground-breaking pronouncement—and despite declaring the communication eventually inadmissible on other grounds—the CRC determined that ‘when transboundary harm occurs, children are under the jurisdiction of the State on whose territory the emissions originated … if there is a causal link between the acts or omissions of the State in question and the negative impact on the rights of children located outside its territory’.¹⁹⁰ This was anchored in the understanding that ‘the State of origin exercises effective control over the sources of the emissions in question’.¹⁹¹

Also the International Court of Justice seems to have developed a somewhat flexible construction of human rights jurisdiction in extraterritorial situations, without attaching paramount importance to ‘effective control’. In the Wall Opinion, concerning the applicability of the ICCPR in the Occupied Palestinian Territories, the Court ascertained that ‘while the jurisdiction of States is primarily territorial, it may sometimes be exercised outside the national territory’.¹⁹² Adopting a teleological approach, ‘[c]onsidering the object and purpose of the [Covenant]’, it concluded that ‘it would seem natural that, even when such is the case, State parties [such as Israel in this case] … should be bound to comply with its provisions’.¹⁹³ While this flexibility may well be attributed to the fact that adjudication occurred against the background of military occupation, in a situation not reaching that threshold, the Court has also displayed an openness to determining the applicability of human rights obligations in a different context. In Georgia v. Russia, it found that the provisions of the Convention on the Elimination of Racial Discrimination (CERD)¹⁹⁴ ‘generally appear to apply … to the actions of a State Party when it acts beyond its territory’, without further caveats.¹⁹⁵ While in this case the finding may have been facilitated by the absence of a jurisdictional clause akin to Article 1 ECHR in the body of the CERD, the wording employed denotes a general appreciation that States remain bound by their international human rights obligations when acting abroad, which they cannot arbitrarily undo through a tactical use of their powers.

6.3 Persisting Controversies

These decisions have ignited an intense debate, particularly regarding whether effective control over the enjoyment of rights,¹⁹⁶ or over the source of harm,¹⁹⁷ should be construed as an exercise of jurisdiction. For some, this approach relies on ‘a more coherent’ and ‘more basic’ idea than the effective control over persons or territory convey, aligning with the overarching ethos of international human rights law.¹⁹⁸ Others, by contrast, see an unwarranted conflation between capacity to act and obligation to do so, which unduly expands extraterritorial duties.¹⁹⁹ The ‘capacity-impact’ model, as it has been referred to,²⁰⁰ is said to put ‘the cart before the horse’ and to distort the ‘ought implies can’ principle, turning it upside down.²⁰¹

This strand of scholarship presumably discards the possibility of triggering positive obligations in extraterritorial settings, since, for the authors concerned, the model relies on ‘consequentialist arguments by which duties are owed because of the negative repercussions that their absence would engender’²⁰²—a premise that is, nevertheless, accepted and well established in non-refoulement cases, for instance, both in territorial and extraterritorial contexts.²⁰³ In any case, this stance presupposes that negative and positive obligations are easily distinguishable and, somehow, that duties to do something are radically different from duties not to do something.²⁰⁴ The reality, however, is messy and more complex than simple categorical exclusions would concede. For example, is the duty not to generate the conditions that would endanger human life at sea a positive or a negative obligation? Also, why should the reversal of ‘ought implies can’ into ‘can implies ought’ under certain circumstances be untenable? Rather than ‘can’—in the abstract and without more—systematically implying ‘ought’, what about contextualised ‘can implies ought’ scenarios, especially in situations where duties would have been activated if it were not for the State’s decision to go extraterritorial? If the State has—of its own free will—resolved to go extraterritorial, influencing the course of events and impacting individuals’ rights with its decision (for its own benefit and in pursuance of its sovereign interests), why should that power that it wields not amount to an exercise of jurisdiction? And why should that same power not engage positive duties, given the State’s capacity to affect final outcomes—potentially coupled with relevant knowledge of attendant risks, taking account of proximity and foreseeability considerations? Why should the fact that these elements are typically invoked to harness the action of due diligence obligations and define their content within territorial settings prevent extraterritorial application in absolute terms? The exclusion appears all the more arbitrary when faced with extraterritorial positive obligations, like the duty to rescue at sea, that have already been codified (and expressly consented to by States or emerged through the conduits of customary law). In relation to these already-codified extraterritorial positive obligations, how else should jurisdiction be defined if not through a capacity-impact analysis?

What these critiques omit is the element of sovereign volition that underpins all forms of extraterritorial intervention and that engages the State’s system of coercion, limiting individual freedom in ways not (always) explicitly authorised by international law and with a potentially harmful impact on human rights. Extraterritoriality does not just ‘happen’. It does not fall from the sky. It is the result of a decision by the State to operate (or laissez-faire) abroad—a decision over which it has plenary and exclusive control. So, the proposition that this model confounds the concept of State power/authority over an individual with the notion of power/influence over the source of harm is formalistic and misses the point.²⁰⁵ Harm is harm when caused to someone (to their interests/rights/property/integrity). Power/influence over an entity or a specific situation, at the bottom of it, engenders power/authority over an individual/group/population, albeit in an indirect way—the former is the means, the latter is the end that is thereby being pursued, both of which constituting the product of the State’s determination. Indeed, ‘every time a State undertakes pre-planned extraterritorial actions involving the use of instruments of State power directly affecting private parties’, including any form of force/coercion via active/passive conduct, ‘[t]he process of planning and deciding about general methods and specific actions [or omissions] … creates a jurisdictional link’.²⁰⁶ It is ‘the carrying out [of] the decisions taken’, as well as their tactical suspension, that ‘places the persons affected under the public power of the State in question’ and, therefore, ‘under the control of that State’.²⁰⁷

When a State decides to go extraterritorial, it should not be allowed to rely on that very decision to be able to evade obligations that would otherwise apply, had the State performed its conduct within its domestic territory. That there is, supposedly, no prior State consent to such an ‘expansion’ of duties as the one entailed by the ‘capacity-impact’ model, denies the chief relevance to be accorded to the fact that it is the State in the first place that—with or without specific authorisation—has elected to project its power abroad. It is this (sovereign) decision that should be taken to trigger human rights obligations and constrain State conduct whenever it affects individuals and their fundamental freedoms, as the UN Treaty bodies and the Inter-American Court propound in their case law.

6.4 Lessons from (Some of) the Strasbourg Jurisprudence

A similar perspective adopted in the migration/border control realm would lead to mechanisms of ‘contactless control’ being deemed human rights-relevant and possibly trigger responsibility in cases of a violation, even if perpetrated by any proxy actors with whom destination countries might collaborate. A functional jurisdiction-based reading of the ECHR and similar instruments of human rights protection would take account of the ‘sufficiently proximate repercussions’ of State (enforcement and/or prescriptive) action ‘on rights guaranteed by the [European] Convention [on Human Rights]’ and related human rights and refugee law safeguards, ‘even if those repercussions occur[red] outside’ national territory.²⁰⁸ The predictable consequences of an exercise of ‘public powers’, including when undertaken beyond national territory, would entail an exercise of jurisdiction thus engaging the State’s responsibility for any resulting infringements. This would also apply when the State acted through a third actor that came under its ‘decisive influence’, be it ‘by virtue of the military, economic, financial [or] political support given to it’.²⁰⁹

In fact, the Strasbourg jurisprudence, though unsystematically, has already embraced elements of this paradigm in cases involving multiple States/actors whose joint action was considered to amount to an exercise of jurisdiction (in the form of ‘effective control’) in breach of ECHR obligations, engaging a mix of its own territorial and personal models, applying in extraterritorial scenarios, whether in the context of the actions/omissions of military operations in Iraq by the UK²¹⁰ or in relation to cooperation between Russia and the local separatist movement controlling the Transdniestria region in Moldova.²¹¹

For example, in Al-Skeini, the Court found that the UK had exercised ‘authority and control’ (i.e. jurisdiction) over individuals accidentally killed during an exchange of fire with a local armed group in Basra during a security operation. Although ‘it [was] not known which side fired the fatal bullet’,²¹² the death of the spouse of one of the applicants was considered attributable to the UK and to trigger the action of the Convention. This was because the death had occurred ‘in the course of a United Kingdom security operation’ which was deemed to establish a ‘jurisdictional link between the United Kingdom and this deceased also’.²¹³ What counted was the ‘functional’ connection established between the deceased and the British forces via their security operation, regardless of whether they had exercised any direct physical coercion. The operation itself constituted an exercise of ‘public powers’ of those ‘normally … exercised by a sovereign government’.²¹⁴ In this case, the British presence in Basra had been authorised by a series of UN Security Council Resolutions, which lent a de jure basis to the exercise of those ‘public powers’, which presumably had to be discharged in line with human rights obligations. The combination of de jure and de facto elements of State authority served to holistically determine the applicability of the Convention and to establish the responsibility of the UK for the resulting violations in the particular case, against a ‘functional’ understanding of jurisdiction.

In the Ilaşcu line of cases, by contrast, the Strasbourg Court held Russia and Moldova to be jointly responsible for the human rights violations perpetrated in the region of Transdniestria. The passivity of Moldova vis-à-vis the human rights violations endured by the applicants living in the region was judged to engage its responsibility by omission, on consideration that, even if not de facto, de jure it still retained territorial jurisdiction over Transdniestria, involving positive due diligence obligations to ‘ensure’ human rights in all parts of the country.²¹⁵ Moldova’s responsibility qua the territorial State notwithstanding, Russia’s indirect intervention, through the separatist local administration de facto controlling the region, was considered sufficient to activate its liability under the Convention. The actions and omissions of the local administration, albeit a third actor with which Russia had no ‘direct involvement’,²¹⁶ were considered to come under the ‘decisive influence’ of the Russian government.²¹⁷ Such an influence was decisive on account of the level of dependency on Russian support of the separatist local administration, which operated ‘by virtue of the military, economic, financial and political support given to it’ by Russia.²¹⁸ As a result, Russia bore responsibility for the third actor’s conduct, given the ‘continuous and uninterrupted link of responsibility … for the applicants’ fate’ that its support involved.²¹⁹ In reality, the local separatist government acted as a proxy—as a vehicle of ‘contactless control’—allowing Russia to indirectly exercise ‘functional’ jurisdiction over Transdniestria from a distance, triggering its ECHR obligations, and determining its responsibility for any concomitant breaches of the Convention.

Such an understanding of jurisdiction—as I envision it—inspired by the jurisprudence of the various international Courts, Tribunals and Treaty bodies explored above, but also based on existing pronouncements of the Strasbourg Court in different subject areas, serves to counter the ‘irresponsibilisation’ phenomenon referred to earlier with a principled and sustainable approach. A ‘functional jurisdiction’ model, like the one I propose, is adapted to today’s globalised world of meta-borders and its mechanisms of ‘contactless control’. It is not based on wishful legal thinking, but on a systematisation of existing ECHR rulings, other relevant international jurisprudence, and (already) accepted bases of extraterritorial jurisdiction and legal responsibility. What it does is to rationalise these precedents to reconnect law with power and responsibility, re-establishing the foundational premise against arbitrariness underpinning the Rule of Law, whatever the spatial location of the conduct (actions/omissions) concerned.

7 Conclusion: Rejecting Unaccountable Power

The externalisation of migration management highlights the need to consider composite notions of agency and responsibility de-coupled from territory and territorially-anchored definitions of border policing in order to avoid accountability gaps. Physical distance-creation through techniques of ‘contactless control’ should not translate into the negation of legal obligations. Mechanisms that (purposively) offshore and outsource border enforcement constitute self-serving effectuations that (by themselves) should not be taken to alter the applicable standards or negate their effectiveness. The proposed ‘functional’ approach to the definition of jurisdiction in international and European human rights law constitutes a step towards a ‘responsibilisation’ model that reconciles law, space, and power in a principled manner.

As the previous sections have shown, borders in today’s globalised world assume different forms and functions. They are complex products of the socio-cultural, economic, and legal processes that constitute them. The law’s implication in the construction of what I have called the meta-border and the inclusion/exclusion effects it produces is paramount to the understanding of how mechanisms of externalisation through strategies of ‘contactless control’ work. It is through the law that the space that separates geography from the exercise of power is produced and legitimised. It is through the law that the migrant as a subject of policing and control is ‘made’. And it is through the law that State sovereignty extends beyond territorial jurisdiction, while concomitant Rule of Law constraints are (tactically) restricted and disapplied. However, this situation, whereby State authority is implemented outside national territory, but with no human rights accountability, needs to be contested. Unaccountable power is incompatible with a Rule of Law-based international system. Arbitrariness is incompatible with international law.²²⁰

Therefore, my main line of argument herein has been that human rights guarantees, in their function as checks on State power, should be deemed to restrain exercises of public authority anywhere they may take place. So, in a situation where sovereign power strategically extends beyond territorial dominion, the State cannot free itself from the legal constraints that would apply in the absence of such a tactical extension—the State’s prerogative to govern migration remains subordinate to its international law commitments.²²¹ The opposite would be tantamount to awarding States the freedom to select, at will, those worthy/unworthy of being included in the Rule of Law framework, unilaterally undoing the universality, inalienability, and erga omnes character of human rights.²²²

Instead, a ‘responsibilisation’ model that is principled and predictable, aligned with the basic safeguards pertaining to the Rule of Law, should be adopted. A ‘functional’ conceptualisation of jurisdiction that allows for the attribution of responsibility for ‘contactless control’ measures, based on an understanding of State functions pursuant to international law, preserves the role of Article 1 ECHR as a ‘threshold’ criterion, but giving it a coherent interpretation across territorial divisions. If/when jurisdiction is construed in this functional sense, as an expression of the ‘public powers’ of the State,²²³ there is no longer a need for arbitrary distinctions of in/out, territorial/extraterritorial, personal/spatial manifestations of sovereignty. The functional approach engages with the sovereign-authority nexus that connects individuals to the State, whether via an exercise of legislative, executive, or judicial ‘functions’, as the main trigger of human rights duties. All de jure and de facto applications of State power, whether ascertained through physical force or indirect forms of control, are considered holistically, to evaluate their aggregate impact on the situation of those concerned. What makes control effective in this context is its ability to effectuate a change in the legal/material position of the individuals affected with human rights-relevant repercussions.

Cooperation with third States/actors, including through techniques of ‘contactless control’, do not become a priori prohibited by the ‘responsibilisation’ model. What the responsibilisation model precludes is recourse to bilateral/multilateral agreements in a tactical fashion in order ‘to bypass human rights obligations’.²²⁴ (Genuine) collaboration can still be pursued but remains subject to human rights compliance.²²⁵ However, if the continued abidance by human rights standards is not possible or cannot be guaranteed, ECHR parties cannot consider themselves exempted from their obligations. On the contrary, they cannot ‘enter into an agreement with another State which conflicts with [their] obligations under the Convention’.²²⁶ These need to be taken into account when designing and executing any form of collaborative remote/contactless controls.

Such a conceptualisation matches the flexible spatiality of migration governance and tallies with the contemporary design of (meta-)borders and their processes of enforcement by rejecting unaccountable power. It thereby closes the ‘unconscionable’ accountability gap that alternative notions leave unaddressed,²²⁷ bringing people on the move (back) within the Rule of Law framework.