Instruments of Evasion: The Global Dispersion of Rights-Restricting Migration Policies

“[N]o other source of tension and anxiety has been more powerful [in the global north] than fear, both real and perceived, of huge waves of future emigration from poor and weak states in the years and decades to come.”[2]

Introduction

Border control is often seen as the last bastion of sovereignty.[3] The term conjures up images of fortified walls and barbed wire separating one country from another. These fortifications determine access; entry can easily be denied to those on the other side. Defensive walls date back to prehistoric times. Think of Gilgamesh’s Uruk or the Walls of Babylon. The reconstructed Ishtar Gate, which once stood within the ancient city of Babylon and which is today housed in Berlin’s Pergamon Museum, is a sight of stunning grandiosity. Hadrian’s Wall marked the boundary of Roman Britannia and projected the empire’s expansive reach and might. Such fortifications have traditionally been built to mark a territorial boundary and to serve as manifestation, real or symbolic, of authority and control.[4] Today, however, governments seeking to restrict mobility rely only partly, and increasingly rarely, on brick and mortar. The border has become a moving barrier, an unmoored legal construct. Bordering activities are no longer collinear with the frontiers of a nation. The once-fixed location of the border has given way to agile bordering techniques that have broken away from the lines on the map. The consequent detachment of border control from any fixed geographical marker provides tremendous leverage and discretion for states seeking to limit (or conversely, facilitate) access to their territories and membership boundaries.[5]

This Essay traces the global spread of legal techniques that strive, as official government policy documents explain, to “push the border out” as far away from the actual territorial border as possible. This concept, enthusiastically embraced by governments worldwide, involves screening people “at the source” or origin of their journey, not the destination, and then again at every possible checkpoint along the way. My study is devoted to tracing the core countries and actors that have facilitated the adoption of such policies. I show that desired destination countries systematically learn from, and emulate, each other’s innovations in asylum-denying laws and policies. The lens of diffusion, which guides the analysis, emphasizes processes of inter-jurisdictional learning and emulation. It invites us to ask how and why “ideas travel” across jurisdictions and to trace the complex ways in which states are interacting with one another in shaping their own border control policies. These developments adopted by policymakers in major turning points are part of a major rethinking that extends—spatially, temporally, jurisprudentially—the sovereign authority of states to regulate mobility. While still operating in the shadow of a Westphalian image of fixed territoriality in which all nations fit neatly together like pieces of a jigsaw, prosperous nations increasingly rely on sophisticated legal tools to expand the reach of border control, limiting the rights of migrants both before and after they enter the country’s territory. The basic idea is that controlling the movement of people might begin “elsewhere,” as close as possible to the point of departure. The traditional static border is thus reimagined as the last point of encounter, not the first.

The globalization of mobility and the arrival of refugees to countries that are “non-contiguous” states (i.e., countries that are not in proximity to the conflict zone) have sharpened the political desire to limit the arrival of uninvited and unwanted migrants. Governments, in turn, have sought policy responses that regulate mobility beyond nations’ geographic boundaries, including within the spheres of influence of other states. The resulting reconceptualization of sovereignty as “shapeshifting” has played a major role in the extraterritorial reach of restrictive migration policies.

In this Essay, I show how nations have emulated one another in adopting such policies. I emphasize the processes of mutual learning and the diffusion of norms and legal policies that grant states and their deputized delegates (whether public or private) permission to act beyond their own territorial boundaries in regulating migration as well as to block movement from afar, prior to arrival. By relying on shifting border techniques, destination countries achieve buffer zones and “rings of protection,” allowing them to bar would-be immigrants from reaching their respective territories and concomitant legal protections while continuing to profess commitment to the protection of human rights around the world. The evidence presented here demonstrates that countries belonging to the richer club of nations proactively learn from one another through inter-jurisdictional “borrowing” and cross referencing of rights-restricting migration policies and regressive precedents.

Such emulation—the voluntary and strategic learning from comparable counterparts—helps to shore up legitimacy in charged political circumstances and provides a framework for policymakers to follow when faced with the horns of a pressing trilemma: (1) the surge in interregional movement of migrants, including asylum-seekers, out of violent conflict zones in the world’s poorer and less stable regions toward the greener pastures of well-off countries in the global north; (2) the moral and legal demands arising from the offering of international protection undertaken by countries committed to the Trinitarian values of human rights, rule of law, and democracy; and (3) the rise of anti-immigrant sentiment and political anxiety among voters due to what they perceive as a loss of control over borders, generating a “defensive reflex in recipient countries.”[6] These competing vectors are impossible to square, and so the trilemma has generated a new paradigm: the “shifting border.”

Unlike a reinforced physical barrier, the shifting border is not fixed in time and place; rather, it consists of nimble legal portals that can potentially be placed anywhere in the world. Governments have been developing the capacity to regulate and track individuals both before and after they reach their desired destination. This grants strategic space to skirt human rights obligations toward refugees and asylum seekers without formally withdrawing them. The conceptual and legal moves underlying this fragile balancing act, and the emulation patterns that help explain why and how such restrictive migration policies get introduced, adopted, rejected, or reacclimated, are at the heart of my inquiry.

By identifying a core repertoire of these processes, whereby restrictive migration policies “migrate” from one jurisdiction to another, I wish to emphasize that beyond familiar debates about the relationship between institutions and rights, we need to pay heed to the crucial work done by the law to “recast” space and place in the service of border control activities. I focus on the development and dispersion of restrictive migration policies adopted by desired destination countries in the global north that have spearheaded the shifting border paradigm: the United States, Australia, and Canada. I will also demonstrate how the European Union and its member states as well as the United Kingdom are responsible for lasting legal innovations that enable countries to evade legal obligations owed to refugees. Although there is a range of policy responses to the trilemma, all rely on severing the link between migration control and a fixed frontier location. This enables affluent countries to escape their legal protection obligations by deputizing partner countries, transit locations, and various actors operating along precarious travel routes to operate as “outpost border guards.” As a result, those on the move are often shut out long before reaching the promised lands. Instead of sharing responsibility, we find evasion.

The analysis highlights both the impetus for and diffusion of legal mechanisms designed to “stop unauthorised migrants and refugees in their tracks.”[7] I will trace the major actors and turning points in the development of an array of “instruments of evasion,” by which I mean the policies and judicial decisions that allow governments to formally uphold their obligations while exploiting the malleability and “stretchability” of the shifting border. There are many sources that reveal the near-obsession of wealthier countries with reimagining migration and border control through sophisticated legal instruments of evasion that increasingly require active cooperation among countries of origin and transit.[8] In particular, I draw upon government documents, case law, public statements by elected officials, the archival records of legislative debates, reports by non-governmental organizations, media coverage, and scholarly opinion and reflection. Bearing in mind the difficulties in “borrowing” policies from different jurisdictions and the need to adjust them to specific geopolitical circumstances rather than merely engage in a generic “cut and paste” exercise, I treat such variation as an opportunity to elucidate the challenges policymakers face when deciding whether to adapt (or reject, or partially implement) certain innovations of their counterparts—and to learn about the borrowed concepts themselves.

My discussion proceeds in three main steps. In Part I, I briefly elaborate the theoretical building blocks upon which my analysis rests: the move from unilateral action to interaction among states; scale jumping from the national to the supranational (and vice versa); and the growing reliance on “informalization” in border management arrangements that involve multiple actors and locations, governed by opaque accords without clear legal standing. I also introduce the concept of the “regressive precedent,” which does crucial work in providing legitimacy to the transnational transfer of restrictive immigration policies. Cumulatively, these instruments of evasion make it extremely difficult to hold governments to account in the court of law or of public opinion.

Part II, which constitutes the bulk of my discussion, provides a stylized account of the direct and indirect ways in which “ideas travel” across jurisdictions. I highlight situations where comparable policies are explored and adopted (with relevant local adjustments), as well as circumstances whereby they are ultimately rejected or kept on the backburner until an opportune set of circumstances arises. My account traces the major turning points in the development and implementation of rights-restricting policies in the post-1951 Refugee Convention era, focusing on the period from the 1980s onward when shifting border strategies begin to emerge as a shared script.[9] Here, avoidance or minimalization of responsibility under domestic, regional, or international refugee and human rights law is the goal or preferred outcome of states’ action.[10] I provide a diagnostic and comparative account of these instruments of evasion and track their diffusion. The analysis highlights both the innovation and risks attached to these rights-restricting policies. My narrative follows the twists and turns that impede the commitment to human rights protections and erode the robust asylum systems put in place by governments in the aftermath of WWII, and the surprising tenacity of ideas about gatekeeping, “return,” and processing claims outside the territories of desired destination countries. At the conceptual level, this article contributes to the nascent body of scholarship that emphasizes the importance of “spatial statism” and the need to account for the shortcomings of legal remedies grounded in traditional conceptions of territoriality and sovereignty in an age where borders, like people, have the capacity to move.[11]

In Part III, I shift the gaze from governments in recipient countries to the incentive structures and pressures in countries of origin and transit. Destination countries engaged in shifting border techniques must acquire the consent of their cooperating partners. In a world operating under the Westphalian commitment to formal equality, such consent is essential and cannot be waived if states wish to respect the sovereignty of the countries on whose territory they seek to establish “remote processing” centers and other migration control operations.[12] I emphasize the fragility of such Faustian bargains, which not only compromise the spirit of human rights and refugee law but also reveal the growing dependence of rich countries on partners with often dubious human rights records that extort material and political concessions by threatening to “open the floodgates.” Tragically, migrants are often caught in the crosshairs. By contrast, as we have seen in the 2022 Ukrainian refugee crisis, when countries amass the political will to cooperate and the solidarity to keep the gates open by sharing responsibility, the result is enhanced protection.

Table of Contents Show

I. Theoretical Building Blocks

It is useful to begin by identifying the key building blocks that have facilitated the diffusion of rights-restrictive policies in the age of shifting borders. These include: the move from unilateral action to interaction; the regressive precedent; scale jumping; and the proliferation and “informalization” of asylum and migration cooperation agreements.

A. From Unilateral Action to Interaction

Much of the scholarship on immigration and emigration has traditionally been country-specific, giving significant weight to domestic economic and cultural factors in explaining a country’s immigration law and policy. This Essay takes a different approach. While acknowledging these classic considerations, it considers both the explicit as well as the more subtle ways in which like-minded countries have engaged in a range of shape-shifting border techniques. In so doing, it reveals a kaleidoscopic picture: states are learning from and interacting with one another and with a host of delegated or “proxy” authorities in increasingly sophisticated ways. Nowadays, shifting border operations take place in countries of origin, transit, destination, readmission, and return, among others, as well as at “hot spots,” “pre-clearance” zones, and offshore remote processing centers, all the while maintaining presence at multiple checkpoints and entry gates along the travel continuum by air, land, or sea. This incredibly dense policy space and the legal “architecture” enabling it are at the heart of my inquiry.

B. The Regressive Precedent

One of the major themes in comparative constitutional law over the past few decades has been the celebration of the “migration of constitutional ideas.”[13] The gist of the argument is that apex courts in the world of new constitutionalism now regularly “borrow” progressive ideas from one another. An example of this would be citing precedents from comparable countries as a source of persuasive authority.[14] Through this transnational dialogue, judges are advancing a new and expanded catalogue of rights protections for citizens and non-citizens alike.[15] But there is a darker side to this increased cross-border dialogue. If progressive ideas about the scope and extent of rights protections can travel quickly, so can restrictive policies. To capture this dynamic, I introduce the idea of the “regressive precedent.”

Just as human rights and constitutional ideas can travel, so too can anti-constitutional or abusive constitutional measures. Once a reputable court or government adopts a highly restrictive policy toward irregular migrants, other governments facing similar predicaments may decide to follow it (even if legally and morally contentious) as a persuasive precedent or “model” to legitimize their own subsequent choices limiting substantive rights or procedural protections. A classic example of this pattern is the transregional influence of the U.S. Supreme Court decision in Sale v. Haitian Centers Council, which is a paradigmatic regressive precedent in legal debates about the extraterritorial reach of non-refoulement—the duty not to return migrants to face persecution at home.[16]

C. Scale Jumping

The prevalent assumption in the literature on human rights is that by shifting the locus and scale of governance structures—for example, by moving from the national to the supranational, or from the unilateral to the multilateral—migrants necessarily benefit from greater human rights protections and mobility options. Alas, this is not always the case. In Europe, to provide but one example, scholars have shown the multiple ways in which the “contemporary thought on [external] visas developed in parallel to the reflection upon the abolition of [internal] border controls within the Schengen.”[17] This juxtaposition of mobility and immobility, openness and closure, freedom in the interior and security threats from its exterior, is part and parcel of the shifting border paradigm. As official EU policy states, to remove internal barriers to internal movement within Europe, “the Union shall develop . . . an integrated management system for external borders.”[18]

A key to fulfilling this mandate is the development and implementation of a common policy on visas, which has been incorporated into consular instructions and manuals.[19] Even more significantly, it has facilitated the establishment of so-called blacklists of countries whose citizens must obtain visas to European member states in advance of travel.[20] Visa regulation of this kind is of course not new. It dates back to the nineteenth century, to the time of the “invention of the passport” (as John Torpey memorably put it), offering a prime example of unilateral nation-generated and enforced techniques of “remote control.”[21] Tracking the evolution of visa-free travel (or visa waiver programs) from the 1950s onward, recent studies have documented a global mobility divide, whereby visa-free travel and ease of mobility has increased for citizens of well-off countries, while it has stagnated or even decreased for those hailing from poorer or less stable countries. The creation of regional “blocs” such as the European Union has added yet another twist to the story. Shared or harmonized visa policies have paradoxically generated greater restrictions for those seeking admission from the outside, as “each [m]ember [s]tate must exhibit solidarity with the other [m]ember [s]tates and respect the presence on the blacklist of a country whose nationals do not necessarily present a problem for it.” [22] The cumulative list is thus more extensive and robust. This pattern is repeated in other contexts too. The power to enter into re-admission agreements, once reserved for member states alone and now expanded to the EU as well, provides another example whereby the regional bloc can gain greater concessions from its bargaining partners (primarily countries of origin and transit) than individual member states may, leading some commenters to dub such agreements as “repressive measures” of “pre-frontier” control.[23]

D. Proliferation and Informalization of Migration Accords

From travel bans to “transit zones,” recent years have demonstrated that the Orbans and Trumps of the world are willing to take action to dramatically restrict the rights of migrants and refugees—shaking up the liberal order in the process. However, even among countries and supranational entities that are outspoken critics of the anti-immigrant, anti-liberal, anti-human-rights backlash, we find reliance on policies that undermine the human rights and protections of migrants under the guise of “regaining” control over the border. Take the European Union. In the aftermath of the 2015 refugee crisis, the European Union has introduced a bewildering array of policy and funding instruments, relying on a “mix of positive and negative incentives and the use of all leverages and tools” to intensify cooperation with countries of origin and transit through which migrants arrive to Europe. Instruments known as “compacts,” “statements,” and “memorandums of understanding” have been signed with a plethora of non-EU countries, including a number of repressive governments known to have breached the legal protections of life and liberty that their own citizens and residents should enjoy.[24] These opaque instruments risk undermining the protection of those seeking safe haven and may put their lives at risk. Nevertheless, these tools are proliferating around the globe. Because these agreements are negotiated and signed outside the official rule-bound processes of legislation and treaty-making, these compacts remain shielded from democratic accountability and public deliberation. Operating on the spectrum of legal liminality, with no formal status, these instruments escape standard channels of approval and review. This lack of transparency and oversight has led members of the European Parliament, the only institution of the European Union that EU citizens directly elect, to issue a strongly worded resolution cautioning against this undemocratic way of working.[25] These hard-to-classify new compacts, statements, and memoranda—created under a cloud of secrecy—are less strictly regulated than standard instruments of international law once they are in operation.[26] Just as the shifting border represents a turn away from visible walls, enlisting instead invisible and mobile legal portals, these obscure agreements manifest a retreat from formal, visible lawmaking.

The agreement signed between the European Union and Turkey in 2016 to stem the flow exemplifies the trend. Defined by observers as “one of the most controversial policy steps taken by the EU in recent years,” it has substantively reduced the number of migrants entering Europe based on the premise, challenged by many, that Turkey is a safe third country for refugees and asylum seekers.[27] When this agreement was challenged before the General Court of the European Union, the Court dismissed the complaints, reasoning that this policy instrument, formally known as the “EU-Turkey statement,” was not subject to the Court’s review. This was because it was not an act of European Union institutions, but rather of the member states.[28] The murky status of these apparently non-justiciable agreements affirms concerns about “informalization”; as migration controls become detached from fixed and identifiable territorial borders, and indeed even from particular countries, a loss of accountability follows.

Across the Atlantic, we find another variant of this trend. In 2019, under the Trump administration, the United States pressured Guatemala, El Salvador, and Honduras into agreements whose very characterization was disputed between the countries. When the Guatemalan Interior Minister and the acting secretary of the Department of Homeland Security signed the first of these agreements in July 2019, U.S. officials presented it as a “safe third country agreement.” However, Guatemalan counterparts declined to characterize the accord as such, calling it instead a “Cooperation Agreement.”[29] Eventually, the label adopted was “Asylum Cooperative Agreements” (ACAs). These agreements and their implementation accords enabled the United States to “remove certain eligible migrants seeking humanitarian protection to each of the ACA countries” as part of a declared regional approach.[30] This is a classic shifting border strategy: it would have caused individuals who had actually made it to the U.S. border to be transferred back to a transit country (one of the ACAs) to seek asylum there, irrespective of whether these countries were indeed considered safe for the purpose of granting a fair process and hearing. Transfers under the U.S.-Guatemala asylum cooperative agreement were paused in mid-March 2020 due to Covid-19, and the agreements with El Salvador and Honduras were never implemented. (The Biden administration eventually suspended and terminated the ACAs.) I will later return to this episode, which combines features of bilateral and multilateral interaction, scale jumping, and informalization. It further reveals the danger of applying the rhetoric of regional cooperation and partnership to asymmetrical power relations, with the result that partner countries are turned into buffer zones, remote border guards, and “dumping grounds” for their richer and more powerful neighbors.[31]

II. How “Ideas Travel”: Key Moments and Actors in the Emulation of Rights-Restricting Immigration Policies

Today, we have forgotten that the 1951 Convention relating to the Status of Refugees (“Refugee Convention”) originally contained geographical and temporal limitations.[32] Although the Refugee Convention offered a universal international instrument for the protection of refugees, its scope of application was initially limited to the then “known groups of refugees,” namely, European refugees displaced in the immediate aftermath of WWII.[33] The Convention’s travaux préparatoires reveal that already at the time of drafting, the hope was that “all nations would be guided by it” and extend protections to refugees even in circumstances that were not in any way related to pre-1951 events.[34] The 1967 Protocol relating to the Status of Refugees formally removed the place and dateline limitations and made the Convention fully applicable to all refugee situations in all regions of the world.

The specter of signatory states relying on spatial barriers to escape their obligations to refugees has a longer history than typically assumed. The concern was already raised at the drafting stage of 1951 Refugee Convention. At the time, it was European states with “contiguous frontiers with other countries from where the stream of refugees [came]” that bore a disproportionate burden of hosting the vast post-WWII refugee population. Countries such as Australia, Canada, and the United States, on the other hand, were shielded from similar humanitarian obligations because “[t]he ocean protects the[se] countries” from the spontaneous arrival of the stream of refugees; they had no contiguous frontiers with the conflict zones.[35] Physical distance had traditionally “constituted a major obstacle for people seeking protection.”[36] As historian Gilad Ben-Nun has shown in his research, it was the countries that were remote from Europe—the epicenter of the post-WWII displaced-persons crisis—which “advocate[d for] non-refoulement [the principle of non-return at the time] because they did not have to bear its demographic and social consequences.”[37] It is one of history’s little ironies that this empirical assumption turned out to be deeply misguided, a realization that partly explains the subsequent invention and extraterritorial implementation of restrictive migration policies. The immediate period after WWII saw “extensive cooperation between the governments of almost all the countries of the ‘free world.’”[38] Alas, such cooperation broke down in subsequent years, especially after the adoption of the 1967 Protocol, which, as just mentioned, removed the original geographic and temporal restrictions that appeared in the 1951 Convention.[39] Post-1967, the Convention applied universally to any person seeking international protection, irrespective of country of origin or the timing of the conflict that led to their displacement. By the late 1970s, refugee movement was no longer contained to the region or conflict zone from which people fled. It had become global. Those operating under the “static” border conception had failed to predict the entrepreneurial desperation that has allowed at least some refugees and other people in need of international protection to “vote with their feet” and take action to reach certain desired destinations, including those very remote from their homelands.[40] In other words, refugee movement was no longer geographically contained. While the vast majority of the world’s refugees, 85 percent according to the latest count[41], are hosted in neighboring countries in the global south (for example, Afghans in Pakistan, Somalis in Kenya, or Venezuelans in Columbia), the shifting border paradigm has emerged in prosperous countries of the global north in response to a perceived “loss of control” and anxiety over the arrival of asylum seekers from the global south, often masking racialized exclusion through facially neutral legal institutions. Not only racial, ethnic, and religious biases but also socioeconomic disparities undergird new measures of exclusion and evasion. Once it became clear to policymakers in desired destinations in the richer parts of the world that they were no longer insulated from large-scale population movements originating from noncontiguous (i.e., non-proximate to a conflict zone) countries, they began to devise legal practices to avert refugee arrival in the first place.

A. The First Mover: The United States

In developing ever more innovative ways to bypass the domestic and international obligations to which they are committed, governments have sought to establish a wide degree of discretion when it comes to migration control. From the 1980s onwards, the United States developed one of the most influential instruments of evasion that has since travelled the globe: maritime interception strategies designed to stop and turn back migrants before they reach territorial waters.[42] As Bernard Ryan patently put it, such action is taken to “prevent sea-borne migrants from reaching their intended destination.”[43]

While there are predecessors of such policies occurring in the pre-1951 Refugee Convention era or by countries that have not committed themselves to non-refoulement, the United States was the first mover among wealthy, rule-of-law societies to adopt—and perfect—such maritime interception policies. The background is as follows. On September 29, 1981, invoking both constitutional and statutory authority, President Ronald Reagan issued Proclamation 4865 and Executive Order 12324 authorizing interdiction by U.S. Coast Guard cutters to prevent “the entry of undocumented aliens from the high seas.”[44] In a pattern that would later be replicated by other countries, the United States also entered a cooperative agreement with Haiti, under which the Haitian government agreed to assist the U.S. government to “stop the clandestine migration of numerous residents of Haiti to the United States.”[45] Initially, the United States pledged not to return anyone among those intercepted in rickety boats departing from Haiti “whom the United States authorities determine[d] to qualify for refugee status.”[46] The United States conducted the initial screening onboard U.S. Coast Guard vessels after they stopped the migrant boats on the high seas, where the vast majority of intercepted migrants were returned to Haiti.[47]

In 1991, following a coup d’état, a growing number of Haitians desperately tried to flee a repressive government. The United States intercepted close to forty thousand en route to safety. Initially, the United States “encamped” some of the intercepted migrants on Coast Guard vessels and directed others to the U.S.-controlled territory of Guantanamo Bay in Cuba. President George H.W. Bush then signed a new and even more ominous executive order in 1992.[48] Unlike the 1981 policy, the 1992 decree authorized the direct return of intercepted Haitian without prior inquiry into any claims of persecution, deploying notions of extraterritoriality and spatiality to avert the obligation of non-refoulement. As a result, the decree forcibly turned back those intercepted on the high seas to Haiti, where they faced the risk of “detention, abuse and death.”[49]

Return without screening for asylum claims has been described as the “ultimate” barrier or deterrent.[50] Government documents clarify the rationale for this policy: “[i]interdicting migrants at sea means they can quickly be returned to their countries of origin [or transit] without the costly process required if they successfully enter the United States.”[51] While well-to-do destination countries may see this sort of “pushback” policy as an efficient and convenient instrument of evasion, the measure has been controversial from the beginning.[52] The policy contradicts the clear and plain language of Article 33 of the 1951 Refugee Convention, which imposes a flat prohibition against the “return (‘refouler’) [of] a refugee in any manner whatsoever to the frontiers of territories where his (or her) life or freedom would be threatened.”[53] This non-refoulement principle, derived from the French word refouler (to “turn back” or “repel”), is the core principle of the post-WWII international refugee protection system. In addition to disregarding this basic humanitarian obligation, which is incorporated into domestic U.S. law, the direct return policy also marked a critical moment in modern bordering: it is one of the earliest exemplars of the shifting border, a moveable barrier dispatched to arrest the mobility of those who most need safe harbor. As sharply observed at the time, those escaping repression in Haiti were “met by a floating Berlin Wall that force[d] their return to the very captors they have fled.”[54]

Sale challenged the direct return policy before the U.S. Supreme Court.[55] The plaintiffs, Haitian refugees, had sought to limit the presidential powers that created the direct return policy. In its ruling, however, the Court deferred to presidential powers in foreign affairs and migration control and adopted a legal theory—vigorously defended by government lawyers—according to which, “as long as the Haitians were in international waters, they had no rights at all.”[56] Accordingly, the Court concluded that it could offer no judicial remedy to the refugees challenging the direct return policy. Here we witness how the shifting border paradigm dramatically affects the geography of human rights and their implementation. The Sale court upheld an image of the border as static and fixed, standing at the territorial edges of the country. The Coast Guard vessels’ interdiction of Haitian migrants operated as a “functional equivalent” of the border far away from the protected territory; as such it was not “visible” to a court that examined it under the classic Westphalian lens.[57] The Court held that U.S. Coast Guard cutters can “push back” Haitian asylum seekers if they are interdicted on the high seas, based on the argument that non-refoulement obligations do not apply extraterritorially. Former presidential advisors and national security officials who were responsible for shaping the U.S. government’s position on direct return claimed that the administration was not intent on breaching rights. “We could always do the right thing,” they said, meaning, in this context, to “allow the refugees to flee their country.”[58] But more important than that, as these officials saw it, was the need to maintain a “greater degree of flexibility. It’s a rare senior executive official who will surrender that flexibility.”[59]

Several national and international courts and tribunals critically rebuked the Sale ruling and the permissive U.S. stance on preemptive maritime interdiction. The English Court of Appeals, for example, broke the semi-sacred principle of international comity among courts when it referred to the case as “wrongly decided.” Going a step further, the Inter-American Commission held, contra Sale, that the non-refoulement provision in the Refugee Convention “has no geographical limitations,” thus giving legal responsibility and jurisdiction a more robust interpretation than ever before. The provision is, in this vein, no longer focused solely on territorial location (as it is seen under the static model) but also applies to states exercising “effective control” or “public power” beyond its borders. A growing number of international law and migration scholars echo this judgement call. However, these statements are non-binding, and Sale has proven more resilient than some have predicted. By absolving a destination country from any responsibility and accountability if the denial of rights it has authorized takes place beyond its territorial waters, Sale has emerged as a classic regressive precedent.[60]

As a judicial pronouncement by the apex court of a seasoned democracy and the world’s leading immigrant-receiving country[61], Sale has emerged as a source for guidance and legitimacy for comparable jurisdictions seeking to adopt similarly restrictive and obstructive interdiction policies.[62] Australia, which adopted its own maritime interdiction policy in the late 1990s, is a case in point. There, the fact that the United States had already provided justification for the approach was seen as “license” to follow suit.[63] And Australia’s hardline policy of interdiction at sea, combined with the transfer of asylum seekers to remote offshore processing locations, has in turn inspired discussions about implementing related measures by EU member states or by the Union as a whole. Most recently, Australia’s policy was directly invoked in Denmark and the post-Brexit United Kingdom, where the government has considered “offshoring” asylum seekers to remote locations in order to escape its own obligations toward refugees (at the time of writing, the policy has not yet materialized).

B. Australia I: Stopping the Boats

Once buttressed by Sale’s regressive precedent, the U.S. practice of maritime interdiction “travelled” to Australia and subsequently elsewhere, inspiring stringent rights-restricting policies. In the late 1990s, Australia saw a rise in the number of people seeking asylum who had reached its shores by way of unauthorized boat arrival. This became a perennially charged political issue. In 1999, Prime Minister John Howard launched a coastal surveillance task force, which concluded that there was a need to strengthen Australia’s maritime investigatory and enforcement powers in encounters with unflagged or foreign-flagged vessels.[64] This recommendation was the precursor for the Border Protection Legislation Amendment Act 1999, which amended the 1958 Migration Act by enhancing the control powers of immigration officials operating at sea.[65] Much like the U.S. policy, the new Act gave legal authorization to intercept vessels carrying unauthorized migrants. But the Australian Act had a caveat: “push backs” at sea were initially restricted to vessels suspected of illicit people smuggling. Seeking to shore up support for the government’s initiative, the Minister for Immigration and Multicultural Affairs, the Hon. Philip Maxwell Ruddock, reassured the Australian House of Commons by stating:

It should be understood that Australia is not alone in adopting a more active approach . . . The United States of America, for example, has announced its intention to broaden the scope of its border enforcement powers beyond its territorial waters. Canada is also reviewing its laws to combat this criminal activity [human smuggling]. I expect that other countries will soon follow. The fact is that if we are not at the forefront in dealing with these issues through legislation of the sort that I am proposing, and other measures, we will be seen as a more attractive destination to the people smugglers who are arranging this sort of trafficking.[66]

The reference to comparable countries’ experiences serves here not only as justification for shifting outward the location of Australia’s maritime interception activities. It also instills a fear that if no such action is taken, Australia will lose out: it will acquire the reputation of the “weakest link,” bearing the risk that it (rather than countries with “tougher” border regimes) will become the target of unauthorized arrivals, many of which rely on clandestine passage and people smugglers. This spiraling race-to-the-bottom is a mirror image of the race-to-the-top rationale that I elsewhere termed “competitive immigration regimes.”[67] With competitive immigration, rich countries compete to attract highly skilled migrants, engaging in a global race for talent that is played at multiple levels. Destination countries must satisfy domestic interest groups as well as respond to (or preferably preempt) their international counterparts’ own recruitment efforts.[68] Conversely, when it comes to deterring uninvited and unwanted migrants, countries similarly engage in complex multilevel games; indeed, additional layers are incorporated when a supranational entity is involved, such as horizontal negotiations among member states or vertical ones with the European Union. But unlike in immigrant attraction, where destination countries engage in fierce competition, these same countries have proven more amenable to cooperation when it comes to immigration deterrence. The result is a transnational web of restrictive policies that aim to deter unwanted immigrants from arriving in sought-after destinations.[69]

In his parliamentary speech, Mr. Ruddock also referred to U.S. President Bill Clinton’s newly implemented Presidential Proclamation 7219, by which the United States extended its “contiguous zone” from twelve to twenty-four nautical miles.[70] The contiguous zone refers to an area extending seaward of a given country’s territorial waters. This “stretching outward” enabled coast guards and federal law enforcement agents to outspread the geographic enforcement range and reach of U.S. law. Vice President Al Gore explicitly noted that one of the policy targets of this expansion was to put “would-be [people] smugglers . . . on notice that we will do everything in our power to protect our waters and our shores.”[71]

In that very same summer of 1999, Canada also witnessed an increasing number of unauthorized vessels heading toward its western shores. The main source of these vessels was the Chinese province of Fujian. This rise in arrivals created a debate among policymakers, both domestically and transnationally (the latter occurred in venues such as the influential Inter-Governmental Consultation on Asylum, Refugee and Migration Policy (IGC), to which I will return later, a closed-door forum organized around a small membership of senior bureaucrats of advanced industrialized states focused on migration management), about whether Canada, rather than Australia, was in fact the weakest link, and whether this apparently lax refugee system rendered Canada “open to abuse.”[72] The New York Times reported at the time that “[i]n Australia, jail terms were doubled last month for the crime of smuggling humans into the country. This summer, the United States and Canada have stepped up patrols in the Western Pacific, seeking to intercept boats filled with clandestine human cargo.”[73] Against this backdrop, Toronto Conservative Senator William Kelly, who chaired Canada’s Senate Committee that reported on illegal immigration, argued that Canada could follow the stricter approaches taken by the United States and Australia in dealing with unauthorized boat arrivals.[74] Even the editorial board of The Globe and Mail, typically a tribune of the establishment, decried that: “[t]he back door’s open, come o-o-n in.” Parliamentary debates followed. While Canada eventually declined to adopt a comprehensive policy of high seas interdiction (primarily due to logistical rather than principled reasons), the record is interlaced with references to the legitimacy of such tactics based on reference to similar practices already adopted by the United States and Australia.[75]

C. Canada’s Interlude: Airborne Interdiction

Instead of “stopping the boats,” Canada invested its enforcement resources in developing another variant of the shifting border strategy: the technique of passenger prescreening overseas. This practice, which is also known as interdiction abroad, effectively relocated much of Canada’s migration regulation activities to overseas gateways. Located primarily in Europe and Asia, migration integrity officers, or “liaison officers,” conduct border control activities as a matter of course despite being nowhere near the frontier of Canadian territory. The Canadian government describes itself as a “world leader in developing interdiction strategies against illegal migration.”[76] Given the country’s geographical location and topography, most travelers seeking to enter Canada reach the country by international flight. This explains Canada’s reliance on airborne, rather than maritime, interdiction. Nevertheless, the basic logic remains the same. By strategically dispatching migration liaison officers to “key foreign embarkation, transit and immigration points around the world,” Canada enhances the spatial and operational reach of its border control functions. [77] It also gains a “greater degree of flexibility,” much as the United States did through sea-borne interception. As the Canadian Border Service Agency has repeatedly stated, its operations are geared toward “moving the focus of control of the movement of people away from [the territorial] border to overseas, where potential violators of citizenship or immigration laws are interdicted prior to their arrival.” [78] As official documents described, this is part of a Canadian border strategy that strives to “‘push our borders out’ . . . as far away from our [territorial] borders as possible.”[79] Being turned away before reaching Canadian territory is crucial for sharply redefining—downward—substantive rights and due process that migrants are due.

Just as in the United States, in Canada the act of “touching base” on the territory significantly affects the scope of protections granted to asylum seekers according to domestic and international legal obligations. In 1985, one of the earliest and most revered cases of the Supreme Court of Canada during the Charter era dealt with the rights of refugees. In Singh v. Minister of Employment and Immigration, the Court held that undocumented or irregular migrants who reach Canadian territory are entitled to the protection of the Charter of Rights and Freedoms. As such, migrants have a consecrated right to a full refugee status hearing before facing potential removal from the country.[80] No similar protections are triggered if one is interdicted or intercepted before reaching Canada’s shores. As part of a comprehensive study on immigration-triggered detention and removal, Canada’s Senate Standing Committee on Citizenship and Immigration concluded that the “interdiction abroad of people who are inadmissible to Canada is the most efficient manner of reducing the need for costly, lengthy removal processes.”[81] By intercepting migrants before their arrival on Canadian soil, Canada can avoid triggering the constitutional provisions that prospective migrants would otherwise enjoy.[82] Investing so much legal meaning in the distinction between “inside” rather than “outside,” between regular and irregular migration, and between the territorial and the extraterritorial, has created an incentive to “push out the border.” By decoupling the “effective border” from its territorial analog, Canada can avert the physical arrival of unauthorized migrants and therefore prevent the engagement of protected rights and procedures. While Singh continues to operate as good law, the wide net cast by Canada’s interdiction strategy effectively bars asylum-seekers from presenting their case to state authorities.[83]

Other countries have emulated Canada’s strategy of placing liaison officers in overseas airports.[84] In 2007, when the U.S. Department of Homeland Security recommended the expansion of the Immigration Advisory Program (IAP), which up until then operated only as a pilot, it explicitly compared the features of this U.S. program to the selective characteristics of the interdiction overseas programs adopted by Canada, Australia, New Zealand, and the United Kingdom. It noted the success that other countries had achieved by “creat[ing] airline liaison officer (ALO) programs and plac[ing] officers in foreign countries to reduce the number of improperly documented passengers traveling into their respective countries.”[85] It is not accidental that these countries were used as the relevant counterparts. They are the longstanding partners with whom the United States has formed the “five-eyes” alliance (FVEY), a wide-ranging intelligence and data-sharing network with extensive capabilities. Next, the EU followed suit, creating an expanded transnational network of immigration liaison officers operating under an EU directive framework that bound them all. In 2019, the EU upscaled its network through new regulation designed to further “enhance cooperation among officers operating within the same third country or region.” This regulation proposed that, “[w]here immigration liaison officers [were] deployed to the Union’s diplomatic missions in a third country by the Commission or Union agencies, they should facilitate and support the immigration liaison officers’ network in that third country. Where appropriate, such networks may be extended to liaison officers deployed by countries other than member states.”[86] As a result, today’s interdiction programs have grown into massive information-gathering operations dispatched globally through a “network of contacts with host-country officials, officials from other governments in the designated region, airline personnel and law-enforcement agents.” These programs operate as soon as travel begins, identifying improperly documented travelers at the earliest point possible and as remotely as possible from the actual border.[87] The International Organization for Migration (IOM) succinctly summarized the runaway success of this Canadian innovation: “[m]any states which have the ability to do so find that intercepting migrants before they reach their territories is one of the most effective measures to enforce their domestic immigration laws and policies.”[88]

D. Australia II: Excision and Offshore Processing—‟Perfecting” the American Model

Turning our gaze back to Australia, we may recall that in August 2001 the “Tampa Affair” became a hotly debated issue in Australian politics. Prime Minister Howard refused to permit a Norwegian freighter, MV Tampa, carrying 433 rescued refugees (predominantly Hazaras of Afghanistan rescued from a distressed fishing vessel in international waters) and five crew members, to enter Australian waters. The government insisted that the refugees disembark elsewhere. When the Tampa refused these orders, Prime Minster Howard ordered the Special Air Service Regiment to board the ship. Subsequently, the government of Norway censured this action and claimed the Australian government had failed to meet its obligations under international law. A few days after the incident, the Australian government hastily introduced the Border Protection Bill 2001, intending to affirm Australia’s sovereignty to “determine who will enter and reside in Australia.”[89] Going further than the U.S. model of interception and pushback on the high seas, Australia’s new border regime (known as the “Pacific Solution”) authorized uniformed officers to remove any ship that entered the territorial waters of Australia. Moreover, Australia gave itself the liberty to deny asylum to people who arrived in its territory by boat without authorization to do so. However, this created a challenge. Where would these asylum claimants be sent instead? How could Australia evade its responsibility towards those who reached its shore? Technically, those who “landed” in Australia were entitled to a full asylum hearing. And here evolved one of the most pernicious innovations of the shifting border paradigm.

Australia, even more explicitly than Canada or the United States, has officially re-located its border through words of law, creating—as its government readily admits—a distinction between the country’s “migration zone” and “Australia” as we know it on the map.[90] This “excision” policy was created through the Migration Amendment Act of 2001, and was expanded in 2005, and then again in 2013.[91] This legislation authorized Australia’s immigration officials to remove asylum seekers that managed to land on Australian soil as though they had never reached Australia.[92] Put differently, those who reach the excision zone cannot make a valid asylum claim in Australia, because they never entered Australia in a legally cognizable way. The territory that they had reached was no longer “Australia” for immigration law purposes. This legal fiction further limits the procedural and substantive rights to which asylum seekers and other irregular migrants are entitled.[93] It also eliminates the possibility of judicial review, thus not only redrawing the territorial border but also attenuating legality in the process. In 2013, the excision zone was expanded, through legislation, to include the entire Australian mainland. In effect, this means that the border applies everywhere and nowhere at the same time.

The legal consequences of arrival at Australia’s “erased” territory are both far-reaching and irreversible; those falling under the spell of excision are denied the opportunity to secure status in Australia, even after their claims have been adjudicated. Excision provides a hocus-pocus way to keep out those who were never wanted or invited. This legal fiction makes them ineligible to claim protection under Australian immigration laws. By erecting an unlimited line of defense against unauthorized maritime arrivals, excision creates a legal barrier. The possibility of passing through the proverbial entry gates, even for those who have managed to reach the country’s (actual) territory, dissolves like a mirage.

Alongside the spatial expansion of the zone of excision, Australia has adopted another means of border shifting. Since 2013, all “asylum seekers who unlawfully arrive anywhere in Australia” must be transferred to third countries for offshoring processing.[94] According to the Australian government’s “regional processing” policy, colloquially known as offshoring, those who have reached the excision zone must be transferred to offshore locations in remote islands in the Pacific. Such offshore locations include Nauru, a tiny microstate island-nation that is 4,500 kilometers away from Australia, which recently signed a bilateral agreement with the Australian government to establish “an enduring regional processing capability in Nauru.” On the occasion of signing the memorandum of understanding with the Nauruan government, Australia’s Minister of Home Affairs, Karen Andrews, reiterated her country’s hardline border protection policies: “There is zero chance of settlement in Australia for anyone who arrives illegally by boat. Anyone who attempts an illegal maritime journey to Australia will be turned back, or taken to Nauru for processing. They will never settle in Australia.”[95] Until the end of 2021, Australia also operated regional detention centers in Manus Island in Papua New Guinea. [96] Asylum seekers had languished for years in these remote processing locations while their claims were processed and assessed. Currently, Australia is the only country in the world that uses other countries to process asylum claims, although this may soon change. Like the United States during the Haiti pushback years, Australia has negotiated agreements with third countries to serve as remote locations in which to hold asylum seekers while they await determination of their claims.[97] Moreover, even those whose claims have proven credible are forbidden for life from settlement in Australia due to the “original sin” of arriving on its excised territory. The erased territory thus becomes a legal black hole, with a gravitational field so intense that no unauthorized migrant can ever escape it. This ironclad policy—the one-way ticket away from Australia—has recently attracted the interest of European policymakers desperately seeking answers to their respective challenges of responding to uninvited migration flows. As I explain in the following Section, the idea of offshoring is not novel on the continent. It has roots dating back to the 1980s and reiterated in the 1990s, after the “US established maritime interception strategies that included a variety of ad hoc offshore refugee screening processes” (such as those in Guantanamo) and reappears periodically.[98] In the era of resurgent populism, the temptation to follow the “Australian way” is immense, despite the denial of constitutional and human rights that it entails.

E. Denmark, the Netherlands, and the United Kingdom: Putting Extraterritorial “Zones of Protection” on the Agenda

The idea of creating regional refugee processing centers outside of Europe in which asylum seekers’ claims could be processed has been percolating for years among European policymakers.[99] It dates to the 1980s, before the fall of the Berlin Wall. At that time, Denmark and Sweden faced an influx of non-European refugees, many of whom hailed from Iran and traveled via Turkey before reaching (what was then) East Germany.[100] From there, the refugees crossed to West Germany. Given the country’s constitutional protection of the right to asylum, the refugees were not turned back. However, the German authorities were growing concerned by the increasing number of applications. Barred from directly stopping the inflow, they “attempted to make the country as unattractive as possible by the introduction of ‘deterrence measures.’”[101] These included prolonged stays at the reception center and very lengthy waiting periods—potentially lasting eight to ten years—for a determination of refugee status. Deterred by these measures, thousands of asylum seekers moved to Denmark, which at the time had a generous refugee regime. Unlike Sweden, which took the preemptive step of denying passage to asylum seekers who did not have an entry visa to Sweden, Denmark accepted refugees even if they arrived from another country in which they had sought, or could have sought, international protection.

As the number of entering refugees continued to climb, the political climate changed. The legislature scrambled to find a solution that would allow Denmark to turn back asylum seekers at the border if it considered another state through which they passed to be a safe country in which asylum could be sought.[102] This legal change, adopted in 1986, came to be known as the “Danish clause.” The basic idea animating this new provision was that “states could remove an asylum seeker to another jurisdiction, on grounds that protection could be sought elsewhere.”[103] It is a classic instrument of evasion: a measure that aims to restrict access to territory and allows states to deflect or “disclaim” responsibility.

Given the major gap in international refugee law of “neither ascrib[ing] a positive right to seek asylum nor set[ting] out the conditions in which states can share responsibility for refugees,” the Danish clause created a domino effect.[104] European states saw the danger that the responsibility to resettle migrants could be “pushed onto” them by neighbors; this created an incentive to adopt restrictive practices. As these unilateral “flanking” measures proliferated in European countries, a regional approach was introduced. The safe third country concept was incorporated into a multilateral intra-EU agreement, formalized as Art. 3(5) of the 1990 Dublin Convention.[105] From Europe, the concept “traveled” globally.[106] Here, a combination of unilateral and multilateral interactions enhanced the ability of states to exert control over movement. It further placed a growing and arguably unfair burden on countries of first asylum or those deemed as safe third countries while decreasing the autonomy of refugees to exercise choice by engaging in what is pejoratively known as “secondary movement.”[107]

Another initiative, entitled “international procedures for the protection of refugees,” received little attention but proved to be a precursor. In 1986, Denmark submitted a draft resolution to the United Nations General Assembly, which called upon the United Nations High Commissioner for Refugees (UNCHR) to set up refugee processing centers in origin regions. This shifted the responsibility away from the world’s rich destination countries toward poorer and less stable regions. This is the antithesis of responsibility sharing. The resolution never passed. Despite its resounding failure, the logic undergirding this initiative has proven more resilient than its initial dismissal may have foretold. In 1993, the Netherlands proposed that the IGC, the closed-door intergovernmental consultation forum we encountered earlier in our discussion, “consider the possibility of sending all asylum seekers back to reception centers in their region of origin to process claims.”[108] The members of the IGC rejected the Dutch proposal on both principled and pragmatic grounds. There were feasibility concerns related to the complex transnational cooperation that this would require countries of origin, transit, and destination, among other actors.[109] However, the IGC did consider other models, which it dubbed “protected areas.” In Reception in the Region of Origin, we find direct reference to the U.S. experience of “processing of Haitians in Guantanamo as a partly successful model, though one difficult to repeat elsewhere due to the unique legal status of Guantanamo, the U.S. military’s unusually high capacity for logistics and detention, and the possibility of intercepting asylum seekers before they reached U.S. territory.”[110] As David FitzGerald recounted, the Danish government brought the idea of offshore processing to the ICG again in 2001, this time around “circulat[ing] favorable reports on the U.S. interdiction of Haitians and Australia’s ‘Pacific Solution.’”[111] In 2003, moving beyond the “talk shop” environment of the ICG, the U.K. government turned to the European Council with its “new vision” for the global management of asylum seekers, refugees, and other migrants. The gist of the proposal was that asylum seekers, refugees, and other migrants arriving in the United Kingdom would be sent automatically to “regional protection zones” abroad. Human Rights Watch reported that a leaked copy of the United Kingdom’s proposal acknowledged “that Australia’s refugee policy [was] its source of inspiration for its ‘new vision’ plan.”[112] In line with Australia’s offshoring processing model, the United Kingdom’s proposal would deny entry to those awaiting determination of their asylum claims and keep them in regional protection zones. The proposal was discussed, but ultimately rejected, at a summit of the European Council.[113] A year later, in 2004, then German Interior Minister, Otto Schilly, brought it back to the agenda by proposing that asylum-seekers “be intercepted in the Mediterranean and returned to EU-financed camps in North Africa.”[114] This plan was even more restrictive than the United Kingdom’s “new vision” proposal. It asserted that only a fraction of those granted refugee status while waiting in the processing zones would be accepted by EU states on a voluntary basis. This edged closer to Australia’s contentious policy of externalizing both asylum processing and refugee protection to third countries.

F. The EU’s Post-2015 Regional Approach

After the Syrian refugee crisis that peaked in the summer of 2015, Europe extended its regional approach, further “externalizing” the reach of its borders deep into the heart of the African continent. A frenzy of cooperation initiatives with key African countries followed. The European Union launched the EU Emergency Trust Fund for Africa (EUTF), used to provide countries of origin and transit with financial and technical assistance to address the internal causes of irregular migration and displaced persons in Africa. As of January 2020, €4.9 billion have been approved under the EUTF.[115] The following year saw the launch of the European Migration Partnership Framework, financed by the EUTF.[116] Currently, the partnership framework includes six “priority countries of origin and transit”—Mali, Nigeria, Niger, Senegal, and Ethiopia. In addition, member states revisited the call for establishing offshore processing centers. France’s President, Emmanuel Macron, publicly floated a proposal to create “hotspots” in Libya where migrants would be pre-screened for asylum claims. This proposal sparked rebuke given Libya’s dismal human rights record. In 2016, the issue was raised at the EU supranational level when the European Council put forward the idea of “regional disembarkation platforms.” Again, this provoked criticism by human rights advocates from around the world. Here, too, we saw a process of inter-jurisdictional learning, whereby those denouncing regional offshore processing drew on the lessons learned from Australia’s record.[117] These accounts noted “the difficulty the EU would face in ensuring that disembarkation platforms in North Africa follow humanitarian standards. At the notorious Australian island centers on Nauru and Manus, legal experts have described the mental health of children as being ‘at crisis point.’”[118] We find a similar pattern later repeated in the United Kingdom, where opponents of the government’s proposal to have asylum claims processed outside the United Kingdom referred to the risks it posed by referencing the lessons learned from Australia. As the IGC had predicted back in the 1990s, the difficulty of finding countries that would agree to host the intercepted migrants has proved to be a major obstacle. Italy, Germany, Austria, and France called for refugee processing centers to be set up in countries as varied as Niger, Jordan, Tunisia, Morocco, Sudan, and Libya. However, these African countries staunchly rejected such proposals. In 2018, the Moroccan foreign minister, Nasser Bourita, explicitly rejected the idea of processing asylum seekers bound for Europe, saying that “Morocco is generally opposed to all kinds of centers.”[119] This did not deter the United Kingdom from approaching the Moroccan government again in 2020 to enter a bilateral agreement, instead of an EU-led multilateral agreement. Morocco firmly declined the offer again.[120]

In the post-Brexit United Kingdom, the debate over offshore asylum processing centers has been renewed. In 2021, after months of speculation about a planned immigration overhaul, the government introduced the Nationality and Borders Bill. The proposed bill incorporated provisions for return and readmission agreements to safe third countries and makes it possible to process asylum claims outside the United Kingdom, pending the development of capacity for offshore processing. It was reported that, like Australia’s turn to Nauru for offshoring, the Home Secretary, Priti Patel, had asked officials to explore the construction of an immigration center on Ascension Island, over four thousand miles away from Britain in the south Atlantic.[121] Prime Minister Johnson expressed his support for the idea, and the foreign office was specifically asked to “offer advice on possible options for negotiating an offshore asylum processing facility similar to the Australian model in Papua New Guinea and Nauru.”[122] Former Australian Prime Minister, Tony Abbott, who was appointed an official U.K. trade advisor to the Johnson government, was cited as an influence. Yet again, the reconceptualization of borders as amorphous and moveable legal constructs, stretching far beyond the country’s geographic location and frontier, was readily emulated from one country to another.

Parliamentary debates revealed the centrality of the inter-jurisdictional learning process. During these debates, proponents and opponents of offshore processing repeatedly brought up the Australian case. The proponents asked, “Is the Home Secretary prepared to do what Prime Minister Abbott of Australia did? He ensured that all arrivals were put in a secure location and left there until their claims were assessed and then either deported or allowed to stay.” [123] The opponents asked, “What person and what Government with an ounce of compassion or respect for international law would even consider casting vulnerable people off to an island using an offshoring system that, in Australia’s case, has been described by the UN as an affront to ‘common decency’?” [124]

The increased acceptance for offshoring came full circle in 2021 when Denmark, the country that put forward the (failed) draft proposal for regional processing before the UN General Assembly in 1986, authorized its government to move asylum seekers to other countries if partnership agreements were signed to that effect. Prior to the vote, Danish parliamentarians referred to the Australian model of offshore processing as an “excellent model.” The Minister of Foreign Affairs and Integration, Matthias Tesfaye, presented the amendments on behalf of the government. His comments referred directly to the precedents provided by the United States, Australia, and the United Kingdom to justify the practice of relocating asylum-seekers as an emerging “international trend.”[125] The Danish legislation provoked a strong response from the African Union. The African Union condemned Denmark’s efforts to enter bilateral agreements with African countries as the potential hosts for such remote processing and highlighted the deeply imbalanced global allocation of responsibilities toward refugees: “Africa . . . generously shoulder[ed] the burden of the world’s 85% of the refugees, often in protracted situations, whereas only 15% [were] hosted by developed countries.” What is more, the “African Union perceive[d] such [European] attempts as an extension of the borders of such countries and an extension of their control to the African shores. Such attempts to stem out migration from Africa to Europe is xenophobic and completely unacceptable.”[126]

Whereas Denmark has not yet entered any agreement to implement its vision of extraterritorial asylum processing and protection, the United Kingdom has signed a deal with Rwanda to send asylum seekers to that East African country.[127] In announcing this offshoring scheme, Prime Minister Boris Johnson stated that “From today, anyone entering the U.K. illegally . . . may now be relocated to Rwanda.”[128] The Prime Minister further stated that “the deal between London and Kigali was ‘uncapped’ and could potentially see Rwanda ‘resettle tens of thousands of people in the years ahead.’”[129] Just like Australia’s hardline policies, those offshored to Rwanda will not gain protection in the United Kingdom, irrespective of the merit of their claims. Instead, they will be given the option to stay in Rwanda, to return home, or to try to secure protection in any country other than the United Kingdom. The U.K. government described the new scheme as an important tool in the fight against human smuggling, providing a policy response to the irregular arrival of migrants on trucks, ferries, and mostly, in small boats across the English Channel. Human rights organizations, by contrast, warned that the agreement violated international law, increased the vulnerability of asylum seekers, and may ultimately backfire by “lead[ing] to more smuggling, not less.”[130] A senior legal advisor for the UNHCR in the United Kingdom condemned the plan as a way for the United Kingdom to “shift its responsibilities towards refugees, not share them.”[131] This latest instrument of evasion is likely to be challenged before the courts, and it remains unclear whether, and if so, how and when, this latest instrument of evasion will be implemented. However, the undercurrent has certainly changed; shifting borders measures are now found everywhere. The trilemma I identified earlier is just as pressing today as it has ever been. The pressures that led to the invention of interception, interdiction, remote processing, safe third country agreements, and related rights-restricting migration control policies still exist today. Responsibility sharing efforts are often evaded. They have thus far fallen short of providing adequate solutions.[132] As one commentator opined with caution:

It may take years for the policies relating to ‘containment in the region’ that are currently being developed to become institutionalised within the EU or internationally, but they are taking shape and being implemented in a piecemeal fashion that lacks coherence, cooperation and courage. When the ‘vision’ thing gets rejected, it is not always dumped—instead pilot or exploratory projects are launched, or the vision is used to transform structures already in place, so that a few years down the line, one discovered that what had been rejected is now well-established.

G. America’s Invisible Border Wall

In 2016, Donald Trump made a campaign promise to build an “impenetrable, physical, tall, powerful, beautiful southern border wall.”[133] While that border wall was never built, the Trump administration erected a different kind of wall: an “invisible” border wall made of hundreds of legal, regulatory, and administrative changes deliberately designed to halt immigration and end protections for asylum seekers.[134] This required reliance on both temporal and spatial techniques, limiting the rights of migrants both before and after they reached the country’s territory. Take “metering.” Under this turnback policy, once a daily quota is reached, the U.S. government directs asylum seekers arriving at the US southern border away from official ports of entry. These asylum seekers are placed on waitlists (turning time into a barrier) as they wait in border towns in Mexico to have their number called, which may take months.[135] By 2019, it was estimated that more than twenty thousand were awaiting the opportunity to begin their asylum process. In 2020, with the onset of the Covid-19 pandemic, all processing at border entry points stopped under Title 42.[136] This order remained in effect, although modified to exempt unaccompanied children, under the Biden administration. In addition to “metering,” the Trump administration introduced another signature initiative: the Remain in Mexico program. Formally known as the Migrant Protection Protocols, the program authorized officials at the border to return certain new entrants to the United States to Mexico while they awaited a hearing of their asylum claim. Close to seventy thousand asylum seekers were removed to Mexico under the program.[137] The legal process takes place in the United States, typically at temporary hearing facilities, but the asylum seekers are removed to Mexico, where they remain for months in dire circumstances and where they may be exposed to substantial safety risks. Combined with lack of access to legal representation, the Remain in Mexico program was described as a death knell for asylum.[138]

Next came an American variant of the tried-and-tested European border externalization and responsibility shifting tactic. We saw above how, in 2019, the U.S. government relied on the rhetoric of the safe third country principle when it entered asylum cooperative agreements (ACAs) with Guatemala, El Salvador, and Honduras. Prior to signing the ACAs, the Trump administration tried to pressure Mexico into entering such an agreement. This would have allowed the United States to turn away asylum seekers who travelled through Mexico to have their claims processed there, in effect “outsourcing” to Mexico the responsibility to protect and evaluate migrants who reach the southern U.S. border. Despite the threat of heavy tariffs, Mexico resisted being designated a safe third country. Instead, in classic shifting border fashion, it agreed to expand the terms of the Remain in Mexico program and to deploy the newly established Mexican National Guard (Guardia Nacional) on its southern border, granting the United States an added strategic depth of 1,864 miles. By these measures, the distance between the U.S. southern border and the Mexican southern border became a buffer to prevent the passage of migrants arriving by land. Following the path of regressive precedent, and learning from the European “neighborhood” approach, the United States signed its first regional compact, known as the memorandum of cooperation.[139] The newly established regional approach aimed to reach even further into Central America, beyond enlisting Mexico to help impede the passage of uninvited migrants to the United States. The U.S. government stated this goal explicitly. When the agreement with Guatemala was signed in Washington, D.C., the highest official of the U.S. Department of Homeland Security described the agreement, and the proposed agreements with Honduras and El Salvador, as part of the drive to “turn the region into a buffer zone for U.S.-bound migrants.”[140] Had the three ACAs been implemented, the enforcement tentacles of the shifting border would have reached deep into Central America.[141] In this example, as in the European antecedents upon which it rests, scale-jumping from the national to the regional may end up capitulating rights protections rather than safeguarding or enhancing them.

In parallel with the ACAs, the Trump administration’s boldest move was to introduce a rule that would have banned anyone who had passed through another country before reaching the U.S.-Mexico border from applying for asylum.[142] Using this Kafkaesque logic, safe passage or mere passage would have become a booby trap, a dangerous, explosive burden on both the individual and the country through which that migrant passed. Further restricting the rights of asylum seekers, the focus was on passage through a third country, without the requirement that the transit country be “safe.” The effect of the rule would have been “to categorically deny asylum to most persons entering the United States at the southern border if they had not first applied for asylum in Mexico or another third country through which they passed.”[143] If such a sweeping overhaul of asylum procedures were implemented on a regional scale, a person escaping gender violence in Honduras would have to seek asylum in Guatemala if she passed through that country before reaching Mexico and ultimately the United States. Whereas the ACAs were a matter of bilateral negotiation, however asymmetrical, the Transit-Country Rule was a matter of domestic policymaking, a unilateral measure. But there was nothing standard about the adoption of this rule, which was originally issued as an interim final rule. This procedure does not require the ordinary notice-and-comment period that otherwise occurs before significant regulatory change takes place, again highlighting concerns about informalization and erosion of democratic input.

In a classic manifestation of diffusion as the process through which policy choices in one country affect those made in other countries, and the resulting spread and adaptation of law and policy across jurisdictions, then-President Trump tweeted flyers of the Australian border protection policy saying, “[t]hese flyers depict Australia’s policy on Illegal Immigration. Much can be learned!”[144] These flyers were part of Australia’s already-discussed hardline border protection campaign. They were inserted into the U.S. debate to build legitimacy and to emulate a regressive precedent that is already in operation. The tweet went viral that same day.

Image 1: Screenshot of Trump’s tweet on Australia’s border protection policy (27 June 2019).

Trump’s reference to Australia’s hardline policies as justification for the ACAs completed a full circle of emulation. We saw earlier that maritime interception and remote processing was an American invention that Australia emulated and embellished. For a President guided by the “Make America Great Again” motto, Trump unknowingly repatriated one of its most lasting inventions in the instruments of evasion’s policy universe.

III. Piercing into the Future: Emulating Restrictive Migration Policies or Advancing Responsibility Sharing?

Part II provided insights into the mechanisms through which restrictive migration policies spread. We saw that as countries in the global north increasingly rely on a complex web of surrogate border guards and outpost locations in the global south to “stem the flow,” lofty claims of responsibility sharing give way to harsher realities of responsibility shirking. By relying on shifting conceptualizations of borders and the offshoring of government functions, policymakers have responded to political realities when public opinion sours on the acceptance of refugee and asylum seekers. But what happens when public opinion shifts in favor of supporting people escaping harm’s way? Under such circumstances, is it possible to counter the dynamic of emulation of restrictive migration policies with the diffusion of protection-enhancing mechanisms, i.e., promoting more inclusive and welcoming initiatives rather than hostile policies from the perspective of international protection seekers? I illustrate this dynamic in action in the context of the 2022 Ukrainian refugee crisis. But before I do so, it is imperative to identify a new stage in the diffusion game. Where in the past we have seen destination countries learn from one another, nowadays transit countries strategically located on gateways leading to rich nations are “appropriating” shifting border techniques to advance their own policy goals or to seek leverage vis-à-vis countries of origin or destination.

A. Reverse Engineering the Shifting Border: Gatekeepers “Weaponizing” Migrants

As the evidence presented in previous Parts has shown, countries faced with the trilemma have invented, emulated, and fined-tuned restrictive measures to deter—if not outright prevent—the arrival of “spontaneous” migrants through policies such as interdiction, interception, readmission agreements, safe third country provisions, offshoring, and the like. Legally and conceptually, these shifting border strategies ultimately facilitate the “protection there, not here” approach.[145] This approach carries serious risks with it, however. The more that power is delegated to partner countries along travel routes, the greater the incentive to avoid criticism of these countries lest they “open the gates.” And this is the case even if these “middleman” regimes put migrants at risk, repress their citizenry, or take a dive toward authoritarianism. An unholy grand bargain has emerged. Global north countries provide financial support, technical know-how, capacity building, even the promise of visa waivers. In exchange, these wealthy countries gain the cooperation (or coercion) of partner countries when it comes to blocking movement, establishing “remote processing” centers, or hosting refugees on their territories.[146] However, even ignoring the risky calculations involved, this strategy rests on shaky normative ground. An unintended consequence of this approach is that it may lend muscle to transit countries to “weaponize” migrants, treating them as bargaining chips. As of 2021, Turkey is the nation hosting the largest number of refugees worldwide—among them close to four million Syrian refugees. When the 2016 EU-Turkey accord was up for renewal, the Turkish government did not hesitate to play the migration card.[147] In 2020, it opened the Pazarkule gate at the Turkey-Greece border—allowing access to EU territory—ferrying and bussing undocumented migrants to the unlocked gateway and generating a “border spectacle.” [148] The EU caved almost immediately to Turkey’s demands for greater funding. The “gatekeeper” had successfully turned the tools of the shifting border paradigm against its architects.

Another flashpoint emerged in 2021 when Belarus’s repressive ruler manufactured a migrant crisis at the Belarusian borders with Poland and Lithuania (both EU member states) in retaliation for sanctions imposed against his country. Like Erdoğan before him, Lukashenko was “reverse engineering” the shifting border. In summer 2021, reports have emerged that Belarus briskly increased air traffic to Minsk, the country’s capital, from countries as far afield as Morocco, Somalia, South Africa, India, Sri Lanka, Algeria, Libya, and Yemen, including a daily route from Damascus. This has been taking place with the aid of a private Syrian air carrier, the previously defunct Cham Wings. The regime also relied on other intermediaries, such as travel agencies in the region, that were authorized to offer packages that included airfare and entry visas.[149] This fast-tracked procedure averted the standard procedure of applying for an entry visa at a regional Belarusian consulate and thus facilitated the sharp rise in the number of entrants. Many hailed from troubled conflict zones in the Middle East, such as the Kurdistan Region of Iraq.

Lest we think that the Lukashenko regime has become a symbol for freedom of movement and migrants’ rights, footage has shown uniformed and armed Belarusian soldiers pressing frightened men, women, and children westward toward the borders with Lithuania and Poland. There, the migrants were met with barbed wire and Lithuanian and Polish border guards blocking their passage. Thousands of people camped in the brutal cold of a no man’s land—prevented from entry into the EU bloc and unable to return to Belarus. Deprived not only of access to an asylum procedure but also of access to water, food, and shelter, the migrants are trapped. Tragically, several deaths have been reported. Fearing a political backlash and the creation of a “pull” magnet for refugees from far afield, European leaders closed ranks with Poland and Lithuania, condemning the Belarusian “orchestrated attempt to use human beings as weapons.”[150] The calculative behavior of the Belarusian regime notwithstanding and its appropriation of language of human rights (Belarusian officials reached out to media organizations such as the BBC to point out that Polish border guards were deploying “unjustified means to repel people seeking protection”), many of the individuals who flew to Minsk never wanted to remain there.[151] They hoped to cross the borders into Poland and Lithuania—into the European Union—to gain asylum there or simply escape their fate in the birthright lottery.[152] Europe, despite its humanitarian commitments, has not been extending open arms.[153]

B. Expanding Protection? Lessons from the Ukrainian Refugee Crisis

Fast forward to 2022. Russia’s invasion of Ukraine triggered the largest displacement of civilian populations in the last eighty years. Over five million people have already fled Ukraine,[154] exceeding the UNHRC’s worst-case prediction.[155] Before Russia invaded Ukraine, 2015 was dubbed the “year of Europe’s refugee crisis.”[156] At that time, the arrival in Europe of close to 1.3 million refugees, many of them escaping war-torn Syria, was described as the largest refugee influx since the end of World War II.[157] Unlike the current crisis, which is seeing Ukrainians crossing land borders to enter neighboring countries in Europe, Middle Eastern asylum seekers had to pass through third countries and then embark on dangerous crossing routes in the Mediterranean (primarily from Libya to Italy) and across the Aegean Sea, via Turkey, to Greece. Their arrival was met with conflicting responses: in Lesbos, generosity and humanitarian solidarity; and in Germany—at least, initially—a warm “welcome culture” spearheaded by Angela Merkel’s famous words, “Wir schaffen das!” (“We can do it!”).[158] Elsewhere in Europe, refugees faced discrimination, stigmatization, and within a few short months, “Fortress Europe” began to close its gates. Hungary’s Viktor Orbán demonized Muslim refugees hailing from the Middle East as a threat to Europe’s “Christian civilization.” Across Europe the far-right was emboldened, using the influx to galvanize anti-immigrant sentiment into a rallying cry of resurgent populist nationalism.

The European response to the Ukraine conflict has proven different from that adopted in previous crises, and importantly so. Instead of discord, the EU has offered a coordinated response that could serve as a blueprint for future incidents of mass influx, whether from war or climate crises or any manner of catastrophe. Instead of endless bickering, the EU and its member states decided to implement a multilateral framework that highlights burden sharing and individual choice. This legal procedure is known as the Temporary Protection Directive. [159] The Directive is__ __designed specifically to address circumstances of “mass influx,” in which the standard asylum procedures that require individualized assessment of each claim become unfeasible. It activates instead an “exceptional measure to provide immediate and temporary protection to displaced persons from non-EU countries and those unable to return to their country of origin.” [160] Those escaping a particular region or country (in this case, Ukraine) instantly qualify for temporary protection. Those benefiting from this solidarity mechanism are entitled to stay in an EU member state and gain important rights and benefits, including access to employment, housing, social welfare, emergency medical treatment, education for school-age children, and opportunities for family reunification. This protection status is initially granted for a period of one year and may be extended for up to three years.

The EU Directive was originally drafted in 2001 in response to the bloody conflicts in the Balkans.[161] It has been on the books for more than twenty years but has never been activated prior to the Ukrainian refugee crisis. The decision to dust off this procedure was adopted unanimously by Europe’s capitals. This legal mechanism shares basic elements with earlier temporary protection programs adopted elsewhere, including, most recently, Colombia’s bold humanitarian initiative to provide temporary protection status to 1.7 million displaced Venezuelans. But it exceeds previous country-specific responses by offering a comprehensive regional approach that is designed to increase protection rather than restrict mobility. Unlike the ACAs that the United States signed, which were designed to push people back to the countries through which they passed, the EU temporary protection measures ushered them in. Ukraine is an immediate neighbor of EU eastern frontier countries. There were no “safe third countries” for people to transit through as they escaped harm’s way; the EU border gates have remained open. “Every refugee must be helped” has become the new guiding motto. Beyond this humanitarian sentiment, which was more widely pronounced across Europe, the innovation of the EU mechanism lied in two additional features.

First, there was an agreement to promote responsibility sharing among European states (backed up by a financial commitment by the EU to support such solidarity). This was a sticking point in the 2015 refugee crisis—and globally the lack of responsibility sharing among states remains a major obstacle to providing adequate protection to millions who are “locked up” in refugees camps or in the territories of the first country of asylum they could reach. Today, 85 percent of the world’s refugees are hosted in developing countries; least developed countries provide asylum to 27 percent of the total. This is patently unfair. The preamble of the 1951 Refugee Convention calls for international cooperation and burden sharing. These commitments were reaffirmed in the global compacts. The EU Temporary Protection Directive breaks new ground in providing an operational regional model that demonstrates how to put these promises to work in a concrete, solidaric fashion.

The second feature is the decision to allow the displaced to have a say in where to enjoy the rights and protections provided to them by the directive. Allowing individual choice is a breakthrough for which scholars and advocates have long called. The refugee system is designed to protect individuals, but it gives them little voice in deciding which country will host them. Proximity becomes destiny, as the principle known as “territorial arrival” dictates that people who arrive at the border or territory of a given country must seek asylum there. The European Directive, by contrast, allows individuals to choose the member state in which they will enjoy the rights that attach to temporary protection. They may wish to follow pre-existing connections and networks to join family members (defined broadly to include relatives that go beyond the nuclear family) and acquaintances across diasporic communities that exist within the Union.

Cooperation is planned not only among member states within the bloc, but also in collaboration with resettlement countries outside the EU, such as Canada, which has a large Ukrainian diasporic community. These resettlement countries are opening channels for family members to join their relatives abroad, setting up procedures for private sponsorship, and, following the European Union’s lead, introducing their own variants of temporary protection mechanisms. As stated by Germany’s Interior Minister, Nancy Faeser, the goal is to establish a responsibility-sharing “system based ‘on solidarity’ to fairly distribute refugees” among participating states.[162]

Concluding Remarks: Out of Devastation May Come Innovation

The 1951 Refugee Convention was born out of the tragedy of World War II and the Holocaust. Despite many shortcomings, it is widely considered “one of the outstanding achievements of the 20th century in the humanitarian field.”[163] The Convention established the “principle that the refugee problem is of concern to the international community and must be addressed in the context of international cooperation and burden-sharing.”[164] Although they formally uphold these commitments and offer access to asylum procedures to those who reach their shores (unless the asylum seekers reach “excised” territories), desired destination countries have also introduced sophisticated legal maneuvers to skirt their obligations toward refugees. The invention of shifting borders—the evolution from well-defined physical facts to amorphous legal constructs—has played a key role in enabling the emulation and enforcement of restrictive migration policies that nowadays reach beyond nations’ geographic borders. My diachronic and comparative analysis has shown how and why rich nations have “borrowed,” learned from, emulated, and in certain cases exceeded, the regressive precedents they adopted from one another.

While we cannot read the tea leaves, we have seen that migration policies travel fast across borders. There is nothing intrinsic, however, to the technique of diffusion that associates it with restrictive policies. It is possible to imagine that more solidaric responses may also “migrate,” if and when the right circumstances arise. And, here, we must return to the trilemma.

Some of the very same countries that refused to open their gates in the past, when those seeking entry belonged to racialized and stigmatized populations, are now leading the relief effort. Back in 2015, Poland, together with its partners in the “Visegrád group”—Czech Republic, Hungary, and Slovakia—staunchly opposed the EU plan to share responsibility among European states in order to relieve the disproportionate burden then carried by Greece and Italy, the frontline states of Europe’s southern frontier. Now, with a war on Europe’s soil, they have changed course and embrace the cause of providing refuge. As they have borne the brunt of responsibility, so has their position on burden sharing shifted.

What has changed? One line of explanation focuses on discrimination, race, and xenophobia. Bias at the border has been and remains real and deeply disturbing, and must be called out and confronted. Concerns about the impact of race, ethnicity, religion, and related prohibited grounds of discrimination are embedded in a broader critique of the inequities baked into the global refugee system. Compare 2022—which has seen an outpouring of solidarity with “blue eyed, blond haired,” Ukrainians—with 2015, when some of the very same countries refused to open their gates to refugees from Syria, Iraq, and Afghanistan, to mention a few examples. Here, the claim is about hypocrisy and double standards in humanitarian protection. Poland is a case in point. It has heroically accepted more than two million displaced persons since Russia invaded Ukraine, while opting to build razor wire in the past.

It is notable that in this present crisis, unlike previous ones, Poland shares a border with Ukraine. Globally, 73 percent of the world’s refugees are hosted by neighboring countries.[165] Turkey, as mentioned earlier, hosts the largest number of refugees worldwide—close to 3.7 million, many of whom escaped into Turkey from neighboring Syria.[166] In this crisis, Poland is Europe’s Turkey. On this account, the explanatory factor for the diverging treatment is not merely racial, ethnocultural, or religious animus but also a story of political will and shared historical, regional, and geopolitical links. Whichever narrative one adopts, the fact remains that depending on where you are fleeing from, your race, gender, religion, nationality, and physical ability to escape harm’s way (among other prohibited grounds of discrimination), individuals whose basic safety and dignity the international refugee system is designed to protect receive decidedly disparate treatment. This disparate treatment—barbed-wire versus empathy and solidaric admission—underscores the inequality these refugees face on account of arbitrary factors beyond their control: which geopolitical crisis they are escaping and from which country they originate.

To break this cycle, greater cooperation, resettlement, and mobility choices need to take heed. While not a panacea, such a burden-sharing approach offers an alternative to a responsibility-shirking one. It offers hope in lieu of despair. To change hearts and minds, we must recognize the structural failings that have led to great disparities in the application of the global refugee regime. The invention of shifting borders has proven a powerful tool in both preserving and reproducing such inequalities. Today’s moment of solidarity provides an opportunity to change course. The lessons to be learned from these more solidaric responses offer guiding lights for a better tomorrow. Instead of continuing to disperse instruments of evasion, it is time to expand measures of protection.

© 2022 Ayelet Shachar, Professor of Law & Global Affairs and the R.F. Harney Chair in Ethnic, Immigration and Pluralism Studies, Faculty of Law & Munk School of Global Affairs and Public Policy, University of Toronto; Director of the Leibniz Research Group, Transformation of Citizenship, at the Normative Orders Research Centre, Goethe University Frankfurt. I am grateful for insights and feedback I received from participants at Berkeley Law’s Sharing Responsibility for Refugees Conference, the Global Migration Law Workshop, and the Philosophy and Borders Conference at Nuffield College, Oxford. Special thanks are due to E. Tendayi Achiume, Alex Aleinikoff, Jacquie Bhabha, Seyla Benhabib, Rebecca Buxton, Seth Davis, Jamie Draper, David Singh Grewal, Ran Hirschl, Jeremy Kinsman, Loren Landau, Katerina Linos, Noora Lori, Saira Mohamed, Hiroshi Motomura, Moria Paz, Jaya Ramji-Nogales, Ben Roswell, and Leti Volpp for most helpful comments and suggestions. Yiwei Jin, Zachary Rosen, Samuel Schmid, and Mira Seyfettinoglu provided outstanding research assistance. This project benefited from generous research funding provided by the Max Planck Society. DOI: https://doi.org/10.15779/Z381C1TG73.

2. Bimal Ghosh, Toward a New International Regime for Orderly Movements of People, in Managing Migration: Time for a New International Regime? 6, 10 (Bimal Ghosh ed., 2000). ↑

3. See Catherine Dauvergne, Making People Illegal: What Globalization Means for Migration and Law 2 (2012). ↑

4. Fortification is derived from fortis (Latin: strong) and facere (Latin: to make). Fortify, Online Etymology Dictionary (2022), https://www.etymonline.com/word/fortify#etymonline_v_11825 [https://perma.cc/FS6B-GLSM]. ↑

5. For further elaboration, see Ayelet Shachar, The Shifting Border: Legal Cartographies of Migration and Mobility 40–41 (2020) [hereinafter Shachar, Shifting Border]. ↑

6. Alexander Cassela, Asylum Seekers in Europe: A Humanitarian Quandary, 44 World Today 187, 187 (1988); see also Etienne Piguet, The ‘Refugee Crisis’ in Europe: Shortening Distance, Containment and Asymmetry of Rights–A Tentative Interpretation of the 2015-16 Events, 34 J. Refugee Stud. 1577, 1577–80 (2020) (describing the rise in interregional movement of refugees); Hiroshi Motomura, Whose Alien Nation? Two Models of Constitutional Immigration Law, 94 Mich. L. Rev. 1927 (1996) (articulating the intersection of race, ethnicity, and migration law and policy). Similar themes are explored by the contributions to this Symposium. See Symposium, Sharing Responsibility for Refugees, 110 Calif. L. Rev. 873 (2022). ↑

7. Thomas Gammeltoft-Hansen, Extraterritorial Human Rights Obligations in Regard to Refugees and Migrants, in The Routledge Handbook on Extraterritorial Human Rights Obligations 153 (Mark Gibney, Gamze Erdem Türkelli, Markus Krajewski & Wouter Vandenhole eds., 2022). ↑

8. Thomas Gammeltoft-Hansen & James Hathaway, Non-Refoulement in a World of Cooperative Deterrence, 53 Colum. J. Transnat’l L. 235, 236 (2015). ↑

9. For further discussion, see Shachar, Shifting Border, supra note 4. ↑

10. See Gregor Noll, Visions of the Exceptional: Legal and Theoretical Issues Raised by Transit Processing Centres and Protection Zones, 5 Eur. J. Migration & L. 303, 304 (2003). ↑

11. Of this emerging body of scholarship, see, for example, Seyla Benhabib, The End of the 1951 Convention: Dilemmas of Sovereignty, Territoriality, and Human Rights, 2 Jus Cogens 75 (2020); Ran Hirschl and Ayelet Shachar, Spatial Statism, 17 Int’l J. Const. L. 187 (2019); Shachar, Shifting Border, supra note 4; The Routledge Handbook, supra note 6. ↑

12. This raises another tension between formal equality and substantive inequality in the asymmetrical exercise of power and sovereignty, which scholars of international law and international relations, Third World Approaches to International Law, and related movements have critically explored. ↑

13. For a concise overview, see Vlad F. Perju, Constitutional Transplants, Borrowing, and Migrations, in Oxford Handbook of Comparative Constitutional Law 1304, 1305 (Michel Rosenfeld & András Sajó eds., 2012). ↑

14. See H. Patrick Glenn, Persuasive Authority, 32 McGill L.J. 261, 263–65 (1987). ↑

15. See, e.g., Anne-Marie Slaughter, A New World Order (2004). ↑

16. Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 155 (1993). For a differing interpretation, see Jamaa v. Italy, App. No. 27765/09, 2012 Eur. Ct. H.R. 198–200. ↑

17. Nora El Qadim, The Symbolic Meaning of International Mobility: EU-Morocco Negotiations on Visa Facilitation, 6 Migration Stud. 279, 282 (2018). ↑

18. Consolidated Versions of the Treaty on European Union and the Treaty on the Functioning of the European Union, art. 77, para. 1(c), 2012 O.J. (C 326) 1, 75–76. For comprehensive analysis of the development of these legal measures, see Violeta Moreno-Lax, Accessing Asylum in Europe 47–69 (Paul Craig & Gráinne de Búrca eds., 2017). ↑

19. Consolidated Versions of the Treaty on European Union and the Treaty on the Functioning of the European Union, art. 77, para. 2(a), 2012 O.J. (C 326) 1, 76. For those hailing from the world’s less stable and poorer regions, securing a visa is a mandatory first step to gain lawful admission to Europe—and it must take place outside the continent, prior to travel. ↑

20. El Qadim, supra note 16, at 282. ↑

21. John Torpey, The Invention of the Passport: Surveillance, Citizenship and the State 3 (2000). The term “remote control” is drawn from the influential work of Aristide Zolberg. See Aristide R. Zolberg, Matters of State: Theorizing Immigration Policy, in The Handbook of International Migration: The American Experience 71 (Charles Hirschman, Philip Kasinitz & Josh DeWind eds., 1999). ↑

22. El Qadim, supra note 16, at 282. ↑

23. Marion Panizzon, Readmission Agreements of EU Member States: A Case for EU Subsidiarity or Dualism?, 31 Refugee Surv. Q. 101, 106 (2012). ↑

24. Communication from the Commission to the European Parliament, the European Council, and the European Investment Bank on Establishing a New Partnership Framework with Third Countries Under the European Agenda on Migration, at 6, COM (2016) 385 final (June 7, 2016). For a concise overview, see Mark Akkerman, Expanding the Fortress: The Policies, the Profiteers and the People Shaped by EU’s Border Externalisation Programme 14–15 (Nick Buxton & Wendela de Vries eds., 2018). ↑

25. See Addressing Refugee and Migrant Movements: The Role of EU External Action, 2018 O.J. (C 298) 39. ↑

26. Ramses A. Wessel, Professor, Univ. of Twente, Presentation at the European Consortium for Political Research Standing Group on the European Union Conference: ‘Soft’ International Agreements in EU External Relations: Pragmatism over Principles?, 1 (June 13–15, 2018), https://www.utwente.nl/en/bms/pa/research/wessel/wesselconf17.pdf [https://perma.cc/8VGG-X9XB]. ↑

27. Narin Idriz, Taking the EU-Turkey Deal to Court?, Verfassungsblog (Dec. 20, 2017), https://verfassungsblog.de/taking-the-eu-turkey-deal-to-court/ [https://perma.cc/RJ9N-4FKL]. ↑

28. Id.; see General Court of the European Union Press Release 19/17, Orders of the General Court in Cases T-192/16, T-193/16 and T-257/16 (Feb. 28, 2017); Case T-192/16, NF v. Eur. Council, ECLI:EU:T:2017:128 (Feb. 28, 2017); Case T-193/16, NG v. Eur. Council, ECLI:EU:T:2017:129 (Feb. 28, 2017); Case T-257/16, NM v. Eur. Council, ECLI:EU:T:2017:130 (Feb. 28, 2017). Such scale jumping, distinguishing EU court jurisdiction from member state jurisdiction and competences, also played a key role in a recent decision focusing on humanitarian visas. See Case C-638 PPU, X and X v. État belge, ECLI:EU:C:2017:173 (Mar. 7, 2017). In both cases, the painstaking technical legal maneuvers adopted by the architects of the shifting border proved vital to blocking the substantive protection of vulnerable migrants. ↑

29. One of the explanations offered to this discrepancy was that the Guatemalan Constitutional Court ruled that the safe third country deal required legislative approval, whereas the signed agreement was negotiated in secret by the executive branch. Lauren Carasik, Trump’s Safe Third Country Agreement With Guatemala Is a Lie, Foreign Pol’y (July 30, 2019), https://foreignpolicy.com/2019/07/30/trumps-safe-third-country-agreement-with-guatemala-is-a-lie/ [https://perma.cc/38S2-PH95]. ↑

30. Press Release, U.S. Dep’t of Homeland Sec., DHS Announces Guatemala, El Salvador, and Honduras Have Signed Asylum Cooperation Agreement (Dec. 29, 2020), https://www.dhs.gov/news/2020/12/29/dhs-announces-guatemala-el-salvador-and-honduras-have-signed-asylum-cooperation [https://perma.cc/XV2B-LHCA]. ↑

31. I here draw on the terminology developed by Katerina Linos and Elena Chachko in their contribution to this Symposium. Katerina Linos & Elena Chachko, Refugee Responsibility Sharing or Responsibility Dumping?, 110 Calif. L. Rev. 897, 900 (2022). ↑

32. See Convention Relating to the Status of Refugees, art. 1, para. B(1), July 28, 1951, 189 U.N.T.S. 137 [hereinafter 1951 Refugee Convention]. ↑

33. U.N. High Comm’r for Refugees, The Refugee Convention, 1951: The Travaux Preparatoires Analysed with Commentary by Dr Paul Weis 4 (1990), https://www.unhcr.org/en-us/protection/travaux/4ca34be29/refugee-convention-1951-travaux-preparatoires-analysed-commentary-dr-paul.html [https://perma.cc/RM8A-DK82]. ↑

34. Id. ↑

35. Gilad Ben-Nun, The British-Jewish Roots of Non-Refoulement and Its True Meaning for the Drafters of the 1951 Refugee Convention, 28 J. Refugee Stud. 93, 100 (2014). ↑

36. Piguet, supra note 5, at 1580. ↑

37. Ben-Nun, supra note 34, at 100 (italics added). ↑

38. Göran Melander, Nordic Refugee Policy in a European Perspective, 9 Current Rsch. on Peace & Violence 183, 183 (1986). ↑

39. These geographical and temporal limitations allowed the contracting states to determine whether the Convention applied only to events occurring in Europe or to events occurring anywhere in the world. See 1951 Refugee Convention, supra note 31, art. 1, para. B(1). ↑

40. Several accounts equate such movement with an expression of agency by refugees, a point of view that I share. ↑

41. Refugee Data Finder, UNHCR: UN Refugee Agency (2022), https://www.unhcr.org/refugee-statistics/ [https://perma.cc/ELZ9-5M6N]. ↑

42. For further elaboration, see Bernard Ryan, Extraterritorial Immigration Control: What Role for Legal Guarantees?, in Extraterritorial Immigration Controls: Legal Challenges 3 (Bernard Ryan & Valsamis Mitsilegas eds., 2010); David Scott FitzGerald, Refuge Beyond Reach: How Rich Democracies Repel Asylum Seekers (2019). On the legal and ethical dimensions of the encounter at sea, see Itamar Mann, Humanity at Sea: Maritime Migration and the Foundations of International Law (2016). ↑

43. Ryan, supra note 41, at 22. ↑

44. Proclamation No. 4865, 46 Fed. Reg. 48,107 (Sept. 29, 1981); Exec. Order No. 12,324, 46 Fed. Reg. 48,109 (Sept. 29, 1981). ↑

45. Agreement on Interdiction of Haitian Immigration to the U.S., Haiti-U.S., Sept. 23, 1981, 33 U.S.T. 3559, 3559. ↑

46. Id. at 3560. ↑

47. Laws. Comm. for Hum. Rts., Refugee Refoulement: The Forced Return of Haitians Under the U.S. - Haitian Interdiction Agreement 3 (1990). ↑

48. See Exec. Order No. 12,807, 3 C.F.R., Comp. 303 (1992). The Executive Order was eventually continued under the Clinton administration. ↑

49. Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 155, 208 (1993) (Blackmun, J., dissenting). ↑

50. David A. Martin, The New Asylum Seekers, in The New Asylum Seekers: Refugee Law in the 1980s 1, 6 (David A. Martin ed., 1988). ↑

51. Daniel Ghezelbash, Refuge Lost: Asylum Law in an Interdependent World 74 (2018) (citing Alien Migrant Interdiction, U.S. Coast Guard (Oct. 31, 2014)). ↑

52. Non-refoulement is nowadays widely considered a jus cogens from which states cannot deviate. ↑

53. See 1951 Refugee Convention, supra note 31, art. 33. The United States is a signatory to the 1951 Refugee Convention so it bears this obligation under both international and domestic law. The non-refoulement principle was also incorporated into U.S. domestic law with the passage of the Refugee Act of 1980. Pub. L. No. 96-212, 94 Stat. 102 (1980). For an excellent overview, see Andrew I. Schoenholtz, Jaya Ramji-Nogales & Philip G. Schrag, The End of Asylum (2021). ↑

54. William Raspberry, Clinton’s Floating Berlin Wall, Wash. Post, Feb. 14, 1994, https://www.washingtonpost.com/archive/opinions/1994/02/14/clintons-floating-berlin-wall/14137212-84c8-46db-bb8e-66788019d7bf/ [https://perma.cc/789R-WFW4] (quoting exiled Haitian President Jean-Bertrand Aristide). ↑

55. See Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 155 (1993). ↑

56. For a captivating account of this legal struggle and the oral argument before the Supreme Court in particular, see Brandt Goldstein, Storming the Court: How a Band of Law Students Fought the President–and Won 232 (2005). ↑

57. On the functional equivalent of the border argument in the Sale case, see Hiroshi Motomura, Haitian Asylum Seekers: Interdiction and Immigrants’ Rights, 26 Cornell Int’l L.J. 695 (1993). ↑

58. Goldstein, supra note 55, at 293 (quoting national security official Eric Schwartz). ↑

59. Id. ↑

60. The U.S. position was reiterated in response to UNHCR’s Advisory Opinion on Extraterritorial Application of Non-Refoulement Obligations, released in 2007. Press Release, U.S. Dep’t of State, Observations of the United States on the Advisory Opinion of the UN High Commissioner for Refugees on the Extraterritorial Application of Non-Refoulement Obligations under the 1951 Convention Relating to the Status of Refugees and its 1967 Protocol (Dec. 28, 2007) (U.S. Dep’t of State Archive Jan. 20, 2001–Jan. 20, 2009), https://2001-2009.state.gov/s/l/2007/112631.htm [https://perma.cc/9RJ5-STMY]. ↑

61. In absolute numbers, the United States has the largest immigrant population per country, representing about 20 percent of the total global migrant population. See Abby Budiman, Key Findings About U.S. Immigrants, Pew Rsch. Ctr. (Aug. 20, 2020), https://www.pewresearch.org/?p=290738 [https://perma.cc/VB34-UR8E]. ↑

62. See infra Parts II.B-II.D. ↑

63. As in the United States, Australian courts have determined that practices of interdicting migrants and asylum seekers on the high seas does not violate non-refoulement. See, e.g., Ruddock v Vadarlis [2001] 110 FCR 491 (Austl.). ↑

64. Press Release, John Howard, Australian Prime Minister, Launch of a Coastal Surveillance Task Force (Apr. 12, 1999), https://pmtranscripts.pmc.gov.au/release/transcript-11236 [https://perma.cc/BD5U-78WM]. ↑

65. Border Protection Legislation Amendment Act 1999 (Austl.). ↑

66. Second Reading of the Border Protection Legislation Amendment Bill 1999 (Austl.) (Sept. 22, 1999). ↑

67. *Ayelet Shachar, The Race for Talent: Highly Skilled Migrants and Competitive Immigration Regimes*, 81 N.Y.U. L. Rev. 148, 153 (2006). ↑

68. See Robert D. Putnam, Diplomacy and Domestic Politics: The Logic of Two-Level Games, 42 Int’l Org. 427, 429–30 (1988). ↑

69. For further discussion, see Ayelet Shachar & Ran Hirschl, On Citizenship, States, and Markets, 22 J. Pol. Phil. 231 (2014); Ayelet Shachar, Citizenship for Sale?, in The Oxford Handbook of Citizenship 789 (Ayelet Shachar, Rainer Bauböck, Irene Bloemrand & Maarten Vink eds., 2017). ↑

70. Proclamation No. 7219 on the Contiguous Zone of the United States (Sept. 2, 1999), https://www.govinfo.gov/content/pkg/CFR-2000-title3-vol1/pdf/CFR-2000-title3-vol1-proc7219.pdf [https://perma.cc/4QB9-259Z]. The extension applies to U.S. states, the Commonwealth of Puerto Rico, Guam, American Samoa, the U.S. Virgin Islands, and the Commonwealth of the Northern Mariana Islands. ↑

71. Philip Shenon, U.S. Doubles Offshore Zone Under Its Law, N.Y. Times, Sept. 3, 1999 (quoting then-Vice President Al Gore), https://www.nytimes.com/1999/09/03/us/us-doubles-offshore-zone-under-its-law.html [https://perma.cc/5BJU-27LU]. ↑

72. J.A. Sandy Irvine, Canadian Refugee Policy: Understanding the Role of International Bureaucratic Networks in Domestic Paradigm Change, in Policy Paradigms: Transnationalism and Domestic Politics 10 (Grace Skogstad ed., 2011). ↑

73. James Brooke, Vancouver Is Astir Over Chinese Abuse of Immigration Law, N.Y. Times, Aug, 29, 1999, https://www.nytimes.com/1999/08/29/world/vancouver-is-astir-over-chinese-abuse-of-immigration-law.html [https://perma.cc/2Z3L-QV9H]. For a comparative account, see Michael Flynn, There and Back Again: The Diffusion of Immigration Detention, 2 J. on Migration & Hum. Sec. 165 (2014). ↑

74. Canada’s Open Door, Maclean’s, Aug. 23, 1999, https://archive.macleans.ca/article/1999/8/23/canadas-open-door [https://perma.cc/8Y5S-KNT2]. ↑

75. Santosh Sirpaul, Clerk of the Parliamentary Standing Committee on Citizenship and Immigration, Minutes of Proceedings and Evidence, Meeting 2 (Nov. 3, 1999) (Can.). ↑

76. Janet Dench, Controlling the Borders: C-31 and Interdiction, 19 Refuge 34, 37 (2001). ↑

77. Can. Border Servs. Agency, Admissibility Screening and Supporting Intelligence Activities – Evaluation Study (2009). ↑

78. The quotation is drawn from the preamble to the Canada-U.S. Statement of Mutual Understanding (SMU) signed in the aftermath of 9/11 to enhance information sharing, reflecting the premise that “border security and border management are based upon cooperation and collaboration.” Statement of Mutual Understanding on Information Sharing, pmbl., Can.-U.S., https://www.canada.ca/en/immigration-refugees-citizenship/corporate/mandate/policies-operational-instructions-agreements/agreements/statement-mutual-understanding-information-sharing/statement.html [https://perma.cc/E6BN-T3ZU]. ↑

79. Gov’t of Can., Canada and the European Union: Border Management, https://www.international.gc.ca/world-monde/international_relations-relations_internationales/eu-ue/index.aspx?lang=eng#a2 [https://perma.cc/MZJ7-ZHX]. ↑

80. See Singh v. Minister of Emp. & Immigr., [1985] 1 S.C.R. 177 (Can.). ↑

81. Standing Committee on Citizenship and Immigration, Immigration Detention and Removal: Summary of Recommendations No. 18, (June, 1998) (Can.), https://www.ourcommons.ca/DocumentViewer/en/36-1/CITI/report-1/page-81 [https://perma.cc/RGX4-VALM]. ↑

82. See Singh, 1 S.C.R/ 177, supra note 79. ↑

83. For an illuminating analysis on Canada’s interdiction strategy, see James Hathaway, The Emerging Politics of “Non-Entrée,” 91 Refugees 40 (1992); François Crépeau & Delphine Nakache, Controlling Irregular Migration in Canada: Reconciling Security Concerns with Human Rights Protection, 12 Immigr. & Refugee Pol’y Choices 1 (2006). ↑

84. See Ryan, supra note 41, at 20–21. ↑

85. U.S. Gov’t Accountability Off., GAO-07-346, Aviation Security 35. ↑

86. Regulation 2019/1240 of the European Parliament and of the Council of 20 June 2019 on the Creation of a European Network of Immigration Liaison Officers (Recast), 2019 O.J. (L 198) 88, art. 13 (EU). ↑

87. See Can. Border Servs. Agency, 2008-2009 Report on Plan and Priorities (2009), http://www.tbs-sct.gc.ca/rpp/2008-2009/inst/bsf/bsf02-eng.asp [https://perma.cc/47M5-5LKR]. ↑

88. Hum. Rts. Watch, Int’l Cath. Migration Comm., World Council of Churches, NGO Background Paper on the Refugee and Migration Interface (2001), https://www.hrw.org/sites/default/files/media_2021/08/202108global_ngo_refugeeinterface.pdf [https://perma.cc/KT62-S57V]. ↑

89. Border Protection Bill 2001 (Austl.). The bill was eventually defeated in the Senate. ↑

90. This Section is based on my discussion in Shachar, Shifting Border, supra note 4, at 41–43. ↑

91. Migration Amendment (Excision from Migration Zone) Act 2001, para. 1 (Austl.). For background on this act, see Gov’t of Austl., Department of Immigration and Multicultural Affairs, Fact Sheet 71: Border Measures to Strengthen Border Control, (Consequential Provisions Act) (2001) (Austl.). ↑

92. Instead, they are immediately directed to third countries declared safe, such as Nauru and Papua New Guinea (until the latter’s court ruled the practice unconstitutional according to PNG law), where Australia has funded detention centers. Even if the Australian government considered these unauthorized arrivals refugees, they could not settle in Australia, and must remain in Nauru or PNG, or be resettled elsewhere. See infra notes 94–96. ↑

93. Even before the creation of the excision zone, Australia introduced a mandatory detention policy for all arrivals without valid visas. See Migration Act 1958 s 196 (Austl.). ↑

94. Migration Amendment (Unauthorised Maritime Arrivals and Other Measures) Act 2013 (Austl). For a concise overview of these legislative changes, see Australia’s Offshore Processing Regime: The Facts, Refugee Council of Austl. (May 20, 2020), https://www.refugeecouncil.org.au/offshore-processing-facts/ [https://perma.cc/M2LL-QL6L]. ↑

95. Joint Media Release, Hon. Karen Andrews & Hon. Lionel Rouwen Aingimea, New Agreement to Secure Our Region from Maritime People Smuggling, Minister of Home Affrs. (Sept. 24, 2021), https://minister.homeaffairs.gov.au/KarenAndrews/Pages/maritime-people-smuggling.aspx [https://perma.cc/MB8B-47QL]. ↑

96. Joint Media Release, Hon. Karen Andrews & Hon. Westly Nukundj, Finalisation of the Regional Resettlement Arrangement, Minister of Home Affrs. (Oct. 6, 2021), https://minister.homeaffairs.gov.au/KarenAndrews/Pages/finalisation-of-the-regional-resettlement-arrangement.aspx [https://perma.cc/GKB5-TGSC]. ↑

97. Elibritt Karlsen, Parliament of Austl., Australia’s Offshore Processing of Asylum Seekers in Nauru and PNG: A Quick Guide to Statistics and Resources, Rsch. Paper Series 2016–2017, at 2 (Dec. 19, 2016). ↑

98. Sergio Carrera, Nora El Qadim, Maryellen Fullerton, Blanca Garcés-Mascareñas, Susan York Kneebone, Ana López Sala, Ngo Chun Luk & Lina Vosyliūtė, Open Soc’y, Offshoring Asylum and Migration in Australia, Spain, Tunisia and the US: Lessons Learned and Feasibility for the EU 35 (2018). ↑

99. The earliest such proposal was raised by the Danish government as a draft resolution to the UN General Assembly. See U.N. GAOR, 41st Sess., 3d comm., U.N. Doc. A/C.3/41/L.51 (Nov. 12, 1986). ↑

100. The fact that the refugees came from outside of Europe, and were non-White, non-Christian, fits with the narrative developed by Tendayi Achiume in this Symposium issue. See E. Tendayi Achiume, Empire, Borders, and Refugee Responsibility Sharing, 110 Calif. L. Rev. 1011 (2022). Contributions to the Symposium on Undoing Discriminatory Borders provides further perspective on this topic. See Symposium, Undoing Discriminatory Borders, 115 Am. J. Int’l L. Unbound 328 (Catherine Briddick & Cathryn Costello eds., 2021). ↑

101. Melander, supra note 37, at 183. ↑

102. See id. at 186. ↑

103. Rosemary Byrne, Gergor Noll & Jens Vedsted-Hansen, Understanding Refugee Law in an Enlarged European Union, 15 Eur. J. Int’l L. 355, 360 (2004) (emphasis added). Others have noted that Switzerland, too, is among the pioneers of the safe third country concept. ↑

104. Nikolas Feith Tan, Denmark’s Extraterritorial Asylum Vision, Refugee L. Initiative Blog (Apr. 19, 2021). ↑

105. The Dublin Convention was later replaced with the Dublin Regulations and then Dublin II and Dublin III. It has been criticized as unfairly shifting the responsibility onto the EU’s coastal and other border states and has caused friction within the block. See, e.g., M.S.S. v. Belgium & Greece, App. No. 30696/09, 2011 Eur. Ct. H.R. 89 (holding that an Afghan asylum seeker had suffered a violation of his rights by virtue of his transfer to Greece by Belgium authorities who acted on the strength of the Dublin II regulation); see also Joined Cases C-411 & 493/10, N.S. and M.E, Joined Cases C-411/10 and C-493/10, N.S. v. Sec’y of State for the Home Dep’t & M.E. v. Refugee Applications Comm’r, 2011 E.C.R. I-13905. ↑

106. A classic example is the U.S./Canada Safe Third Country Agreement (STCA). See Audrey Macklin, A Safe Country to Emulate? Canada and the European Refugee, in The Global Reach of European Refugee Law 99, 100 (Hélène Lambert, Jane McAdam & Maryellen Fullerton eds., 2013). The STCA has been the subject of ongoing litigation in Canada. ↑

107. A policy brief by the European Parliamentary Research Service provides a concise overview of this position: “Secondary movements of asylum-seekers can put pressure on host countries, including their reception capacities, asylum systems, economy, and security. Multiple asylum applications lodged in different countries can lead to inefficiencies, administrative duplication, delays, and additional costs. They may be perceived as a form of misuse of the asylum system and may thus reduce political and public support for refugee protection.” See Anja Radjenovic, Briefing: Secondary Movements of Asylum-Seekers in the EU Asylum System, Eur. Parliamentary Rsch. Serv. (Oct. 2017). The earliest pronouncement at the international level against secondary movement dates to 1989, with the adoption by the Executive Committee (ExCom). Conclusion No. 58 (XL) of the Executive Committee of the High Commissioner’s Programme identified the “phenomenon of refugees and asylum seekers ‘who move in an irregular manner from countries in which they have already found protection, in order to seek asylum or permanent resettlement elsewhere’” as a growing concern. See María-Teresa Gil-Bazo, The Safe Third Country Concept in International Agreements on Refugee Protection: Assessing State Practice, 33 Neth. Q. Hum. Rts. 42, 47 (2015). ↑

108. FitzGerald, supra note 41, at 213. ↑

109. Id. at 214. I am indebted to the account offered by David FitzGerald of the ICG’s early 1990s discussion of these ideas. Details about these discussions are also reported by Noll, supra note 9; Irvine, supra note 71. ↑

110. *FitzGerald, supra *note 41, at 214 (citing IGC Secretariat, Geneva, Reception in the Region of Origin: Draft Follow-up to the 1994 Working Paper (1995)). ↑

111. Id. ↑

112. Hum. Rts. Watch, An Unjust “Vision” for Europe’s Refugees 2 (June 17, 2003). ↑

113. Marit Ruuda, Strong Opposition to UK Asylum Plan, EUObserver (June 17, 2003), https://euobserver.com/news/11741 [https://perma.cc/N4WF-UH8L]. ↑

114. FitzGerald, supra note 41, at 215. ↑

115. Eur. Comm’n, EUTF For Africa: The EU Emergency Trust Fund for Stability and Addressing Root Causes of Irregular Migration and Displaced Persons in Africa (2020), https://ec.europa.eu/trustfundforafrica/sites/euetfa/files/factsheet_eutf-for-africa_january_2021_0.pdf [https://perma.cc/FZJ7-JJYN]. ↑

116. Eur. Comm’n, Migration Partnership Framework: A New Approach to Better Manage Migration (2016), https://eeas.europa.eu/sites/eeas/files/factsheet_migration_partnership_framework_update13_12_2016_final.pdf [https://perma.cc/477G-E9EJ]. ↑

117. For more information, see Press Release, Eur. Council, “Further Action on Main Migratory Routes into the EU,” European Council Conclusions on Migration, Security and Defence, Jobs, Growth and Competitiveness, Innovation and Digital, and Other Issues (June 28, 2018), https://www.consilium.europa.eu/en/meetings/european-council/2018/06/28-29/ [https://perma.cc/A4AT-WEAJ]. ↑

118. *Melanie Ward, Why the EU Can’t Outsource Its Migration Crisis, *Politico (Sept. 23, 2021), https://www.politico.eu/article/eu-cant-outsource-migration-crisis-disembarkation-platforms-centers/ [https://perma.cc/ZC7L-RYU4]. ↑

119. *Esther King, Morocco Rules out Building EU Offshore Asylum Centers, *Politico (Oct. 3, 2018), https://www.politico.eu/article/morocco-rules-out-building-eu-offshore-asylum-centers-migration/ [https://perma.cc/T2EM-K3HR]. ↑

120. See Robert Verkaik, Opinion, British Plans to ‘Offshore’ Asylum Seekers Have a Long and Grubby History, Guardian (Oct. 1, 2020), https://www.theguardian.com/commentisfree/2020/oct/01/british-plans-offshore-asylum-seekers-australian-refugees-criminals-uk [https://perma.cc/J6ER-W2PN]. ↑

121. Sebastian Payne, Peter Foster, Robert Wright & George Parker, No 10 Confirms UK Offshore Asylum Plan Under Consideration, Fin. Times (Sept. 30, 2020), https://www.ft.com/content/9baaf989-f64d-417d-90c5-b0ea8f78bf0c [https://perma.cc/R3XS-2CLH]. ↑

122. Paul Lewis, David Pegg, Peter Walker & Heather Stewart, Revealed: No 10 Explores Sending Asylum Seekers to Moldova, Morocco and Papua New Guinea, Guardian (Sept. 30, 2020), https://www.theguardian.com/uk-news/2020/sep/30/revealed-no-10-explores-sending-asylum-seekers-to-moldova-morocco-and-papua-new-guinea [https://perma.cc/3NQY-Y26L]. ↑

123. HC Deb (24 Mar. 2021) 691 col. 932 (statement by Sir Edward Leigh (Gainsborough) (Con)), https://hansard.parliament.uk/Commons/2021-03-24/debates/464FFFBB-ECA5-4788-BC36-60F8B7D8D9D1/NewPlanForImmigration#contribution-557EE55D-8D13-4E64-8C12-D99868E1B499 [https://perma.cc/99EV-TCNH]. ↑

124. HC Deb, supra note 122, col. 926 (statement by Anne McLaughlin (Glasgow North East) (SNP)). In October 2021, Australian immigration law experts and advocacy groups provided additional testimony before the U.K. Parliament Joint Committee on Human Rights. The submissions directly addressed the humanitarian consequences of the Bill’s proposed scheme of overseas processing by invoking Australia as a “regressive precedent.” The UNCHR also published a report detailing its observations on the proposed Nationality and Borders Bill. The report mentioned Australia seven times as an example of the concerns raised by extraterritorial processing. See U.N. High Comm’r for Refugees, Observations on the Nationality and Borders Bill, Bill 141, 2021-22 (Oct. 2021), https://www.unhcr.org/uk/615ff04d4.pdf [https://perma.cc/6PPC-4RTD]. ↑

125. Parliamentary speeches, Proposal for an Act amending the Aliens Act, L 226, Folketinget 2020-1 [Den.]. ↑

126. Addis Getachew, Pan African Body Slams Denmark over Aliens Act, Anadolu Agency (Aug. 3, 2021), https://www.aa.com.tr/en/africa/pan-african-body-slams-denmark-over-aliens-act/2323262#! [https://perma.cc/AB8B-MV2Y]. ↑

127. The proposed European New Pact on Asylum and Migration, for example, emphasizes rapid border procedures, not offshoring. See also U.K. plan to fly asylum-seekers to Rwanda draws outrage, NPR (Apr. 15, 2022), https://www.npr.org/2022/04/15/1093006759/uk-plan-to-fly-asylum-seekers-to-rwanda-draws-outrage [https://perma.cc/6WW4-4WB9]. ↑

128. . UK to send asylum seekers to Rwanda under controversial new deal, Aljazeera (Apr. 14, 2022), https://www.aljazeera.com/news/2022/4/14/uk-to-sign-deal-to-send-male-channel-refugees-to-rwanda-reports [https://perma.cc/U5E8-UTFG]. ↑

129. . Id. ↑

130. . Id. ↑

131. . Id. ↑

132. For further elaboration, see, for example, the contributions by Linos & Chachko, supra note 30, and Michael Doyle, Janine Prantl & Mark J. Wood, Principles for Responsibility-Sharing: Proximity, Culpability, Moral Accountability and Capability, 110 Calif. L. Rev. 935 (2022), in this Symposium. ↑

133. Transcript of Donald Trump’s Immigration Speech, N.Y. Times, Sept. 1, 2016, https://www.nytimes.com/2016/09/02/us/politics/transcript-trump-immigration-speech.html [https://perma.cc/Q5PV-767A]. ↑

134. Am. Immigr. Laws. Assoc., Deconstructing the Invisible Wall: How Policy Changes by the Trump Administration Are Slowing and Restricting Legal Migration (Mar. 19, 2018). For more on the asylum system, see Schoenholtz et al., supra note 52. ↑

135. On the use of time to delay access, see Elizabeth F. Cohen, The Political Value of Time: Citizenship, Duration, and Democratic Justice (2018). ↑

136. Public Health Service (PHS) Act, 42 U.S.C. § 265. ↑

137. Peter Margulies, The Courts Restore ‘Remain in Mexico’ Program: An End to Judicial Deference?, Lawfare (Sept. 3, 2021), https://www.lawfareblog.com/courts-restore-remain-mexico-program-end-judicial-deference [https://perma.cc/A946-J5QK]. ↑

138. See Emily J. Johanson, The Migrant Protection Protocols: A Death Knell for Asylum, 11 UC Irvine L. Rev. 873 (2021). ↑

139. Press Release, U.S. Dep’t of Homeland Sec., Secretary Nielsen Signs Historic Regional Compact with Central America to Stem Irregular Migration at Source, Confront U.S. Border Crisis (Mar. 28, 2019), https://www.dhs.gov/news/2019/03/28/secretary-nielsen-signs-historic-regional-compact-central-america-stem-irregular [https://perma.cc/SP8Z-EGXA]. ↑

140. Sofia Menchu, After Guatemala, U.S. Seeks Migration Deals with Honduras, El Salvador, Reuters (Aug. 1, 2019), https://www.reuters.com/article/uk-usa-immigration-guatemala-mcaleenan-idUKKCN1UR5G5 [https://perma.cc/KZD5-WA6D]. The United States signed such an agreement for “cooperation in the examination of protection claims” with El Salvador on September 20, 2019. ↑

141. The agreement with Guatemala went into effect for a brief period before the three ACAs were suspended and terminated by the Biden administration. ↑

142. 84 Fed. Reg. 33829. ↑

143. E. Bay Sanctuary Covenant v. Barr, 519 F. Supp. 3d 663, 664 (N.D. Cal. 2021). The rule was challenged before the courts in several cases. ↑

144. Luke Henriques-Gomes, Donald Trump Says ‘Much Can Be Learned’ from Australia’s Hardline Asylum Seeker Policies, Guardian (June 26, 2019), https://www.theguardian.com/us-news/2019/jun/27/donald-trump-says-much-can-be-learned-from-australias-hardline-asylum-seeker-policies [https://perma.cc/J6UB-KWXC]. ↑

145. T. Alexander Aleinikoff & David Owen, Refugee Protection: “Here” or “There”? 2 (Robert Schuman Ctr. for Advanced Stud., Working Paper No. 2021/64, 2021). ↑

146. Id. The reliance of wealthier countries on partner countries requires significant coordination, bilaterally or multilaterally. This can be achieved through negotiated agreements that are expected to provide benefits to the different parties involved (allowing “gatekeepers” in strategic locations to gain leverage vis-à-vis more powerful destination states, challenging unequally distributed gains from centuries of extraction capitalism and colonialism), or via coercive politicking that involves threats, domination, or even violence. On this distinction, see Gerasimos Tsourapas, Migration Diplomacy in the Global South: Cooperation, Coercion and Issue Linkage in Gaddafi’s Libya, 38 Third World Q. 2367 (2017). ↑

147. Dogachan Dagi, The EU-Turkey Migration Deal: Performance and Prospects, Eur. Foreign Affrs. Rev. 197, 210–11 (July 2020). ↑

148. For elaboration on the concept of the border spectacle, see, for example, The Borders of “Europe”: Autonomy of Migration, Tactics of Bordering (Nicholas de Genova ed., 2017). ↑

149. To provide one example, the Belarusian Embassy in Erbil outsourced visa applications to a number of travel agencies in Baghdad. See Rod Mudge, The Route from Iraq to Belarus: How Are Migrants Getting to Europe?, DW (Nov. 9, 2021), https://www.dw.com/en/the-route-from-iraq-to-belarus-how-are-migrants-getting-to-europe/a-59636629 [https://perma.cc/T36R-L4LF]. ↑

150. Marc Santora, Monika Pronczuck & Andrew Higgins, Accusing Belarus of Trying to Create a Migrant Crisis, the E.U. Sides with Poland in a Border Standoff, N.Y. Times, Nov. 10, 2021, https://www.nytimes.com/live/2021/11/10/world/poland-belarus-border-migrants/europe-shows-support-for-poland-as-crisis-at-the-border-deepens [https://perma.cc/EK96-WHRY] (citing European Council president, Charles Michel). ↑

151. The Turkish government used similar techniques of taking footage of the Greek border guards preventing entry through the Pazarkule gate in its campaign to “shame” the EU. ↑

152. Ayelet Shachar, The Birthright Lottery: Citizenship and Global Inequality (2009). ↑

153. Manifesting classic shifting border thinking, the European Commissioner for Home Affairs, Ylva Johansson, stated that the EU was stepping up “outreach with partner countries” to prevent migrants from reaching Belarus in the first place. “Our urgent priority is to turn off the supply coming into Minsk airport,” she tweeted. Ylva Johansson (@YlvaJohansson), Twitter (Nov. 8, 2021, 11:24 AM), https://twitter.com/ylvajohansson/status/1457791267480342528 [https://perma.cc/SM7N-D3EG]. ↑

154. Operational Data Portal: Ukraine Refugee Situation, UNHCR: UN Refugee Agency (2022), https://data2.unhcr.org/en/situations/ukraine [https://perma.cc/7KT8-VUYY]. ↑

155. Michelle Langrand, UNHCR: Escalation of War in Ukraine Could Drive 4 Million Refugees, Geneva Sols. (Feb. 25, 2022), https://genevasolutions.news/peace-humanitarian/unhcr-escalation-of-war-in-ukraine-could-drive-4-million-refugees [https://perma.cc/H63G-9BJF]. ↑

156. William Spindler, 2015: The Year of the European’s Refugee Crisis, UNHCR: UN Refugee Agency (Dec. 8, 2015), https://www.unhcr.org/news/stories/2015/12/56ec1ebde/2015-year-europes-refugee-crisis.html [https://perma.cc/5D6D-LSQ5]. ↑

157. Phillip Connor, Pew Rsch. Ctr., Number of Refugees in Europe Surges to Record 1.3 Million in 2015, at 6 n.1 (Aug. 2, 2016), https://www.pewresearch.org/global/2016/08/02/number-of-refugees-to-europe-surges-to-record-1-3-million-in-2015/ [https://perma.cc/9TW4-XTB8]. ↑

158. Flüchtlingskrise 2015: Wir schaffen das, Zweites Deutsches Fern sehen (Sept. 15, 2021), https://www.zdf.de/politik/unsere-merkel-jahre/videos/video-wir-schaffen-das-100.html [https://perma.cc/FL59-Q246] (videorecording of the press conference on Aug. 31, 2015, during which Angela Merkel spoke to the press and made the “Wir schaffen das” statement). ↑

159. Council Implementing Decision 2022/382 of 4 March 2022 Establishing the Existence of a Mass Influx of Displaced Persons from Ukraine Within the Meaning of Article 5 of Directive 2001/55/EC, and Having the Effect of Introducing Temporary Protection, 2022 O.J. (L 71) 1 (EU). ↑

160. Temporary Protection, Eur. Comm’n: Migration & Home Affs. (2022), https://ec.europa.eu/home-affairs/policies/migration-and-asylum/common-european-asylum-system/temporary-protection_en [https://perma.cc/5XU6-B5XZ]. ↑

161. The UNHCR provided guidelines for temporary protection in 1994, following the crisis in the former Yugoslavia. This crisis was also the catalyzer for the EU’s adoption of the Temporary Protection Directive in 2001, following several years of negotiation within the bloc. See U.N. High Comm’r for Refugees, Note on Temporary Protection in a Broader Context, Refworld (Jan. 1, 1994), https://www.refworld.org/docid/3ae6b32514.html [https://perma.cc/2XBA-PSYN]. ↑

162. Ukraine: EU Agrees Plan to Aid Refugee Resettlement, DW (Mar. 28, 2022), https://www.dw.com/en/ukraine-eu-agrees-plan-to-aid-refugee-resettlement/a-61277944 [https://perma.cc/QXC3-42J3]. There is a risk that schemes that rely on temporary protection will provide short term relief but erode the long-term protection offered by asylum. In the EU scheme, the option of applying for asylum remains open to those who seek, or need, it. ↑

163. U.N. High Comm’r for Refugees, supra note 32. ↑

164. Id. ↑

165. Refugee Data Finder, supra note 40. ↑

166. Id. ↑