Introduction: Borders of the Mind

The force of the border, such as it is, is not only that it potentially stops migrants from entering, but that it also builds migrants in its image. The border has long ceased to be a mere physical border. Rather, it is a series of technologies deployed to regulate the movement of peoples across space that deeply structure social and legal relations and are internalized by mobile humans. It moves with them, against them, and sometimes for them (Shachar, Reference Shachar2020b). As such, territory is redefined as a series of pockets of immobility that constrain the outflow and inflow of persons. More importantly, the border shapes what I would describe as a series of mobile identities, the identities one needs to endorse in order to pass the border, conceived in increasingly narrow terms (Sajjad, Reference Sajjad2018). In effect, the border not only regulates bodies, but it also produces particular kinds of subjecthood and subjectivity. These relate, for example, to claims, often reductive and simplistic, about the conditions in the country one left (Kelley-Widmer & McKee, Reference Kelley-Widmer and McKee2024) as well as the particular reasons one feels it necessary to invoke to justify one’s mobility.

The harshest effect of the border is surely on “bare migrants” who cannot claim to be refugees, most notably those “forced migrants” fleeing poverty, ecological degradation, and crime – but not specifically persecution in the sense of the 1951 Convention. If refugees have the right to asylum, then some otherwise deserving migrants will want to qualify as refugees, at some cost to themselves and to the system. Quite aside from populist rants about “fake refugees,” it is inevitable that some migrants will be tempted to claim asylum, especially in a context where what constitutes persecution may be ambiguous and where the narrowness of asylum is all too obvious. This includes migrants pretending to be children when they are not (Gibb, Reference Gibb2018); pretending to be part of a family unit when they are not (Burnett, Reference Burnett2019); pretending to be of a different nationality than they are (e.g., Syrian rather than Moroccan or Turkish) (George, Reference George2015); and criminals or terrorists posing as refugees (Reuters, 2016).

By contrast, those who can claim persecution are a priori and paradoxically in a privileged position (Price, Reference Price2006) in that they at least formally have a right to asylum protected under international law. The relative privilege, for the purposes of cross-border mobility, of refugees has of course become somewhat theoretical as states have deployed considerable resources to make it difficult for even would-be refugees to seek asylum (Gammeltoft-Hansen, Reference Gammeltoft-Hansen, Chetail and Bauloz2014). This has been in part because of a suspicion that many asylum seekers are, in fact, “migrants in disguise,” but also quite simply because of states’ lack of enthusiasm to comply with their 1951 obligations even in relation to bona fide refugees. Still, asylum seekers are historically and legally privileged subjects when it comes to cross-border mobility in a context where states maintain a strict upper hand over who crosses their borders.

Nonetheless, even asylum seekers feel the particular pressures of the border. Whereas the border forces some migrants to claim asylum, this chapter is interested in the converse phenomenon, namely how asylum seekers are under pressure to tone down the extent to which they are also, in some respects, migrants, lest this compromise their bona fides as refugees. In seeking to uphold a category of asylum seekers fleeing persecution as entirely distinct from that of migrants, the international legal regime of cross-border mobility does violence to the more complex reality of mobile humans’ trajectories. Specifically, it confronts them with the choice of one of two stereotypical roles: on the one hand, the bare human fleeing the immediate threat of persecution at the expense of all else for mere protection; on the other hand, the fully agentic but unprivileged migrant.

It thus forces refugees into a bind: either seek protection in the closest or first “safe” territory that they can access by prioritizing their most immediate need for protection, but at the cost of frustrating their life plans; or “migrate” to the place of their choosing to seek asylum but then risk reinforcing the suspicion that they are, indeed, migrants. This tendency is reinforced by an evolution of the conception of refugeehood away from a more political emphasis on human dignity toward a more humanitarian vision, and away from an understanding of asylum as a right toward an understanding of asylum as a favor (Owen, Reference Owen, Oliver, Madura and Ahmed2019) that imposes its own pressure to conform. Even the “privilege” of refugeehood, it turns out, is paid at the price of the voluntary suppression of one’s agency, including as it might otherwise manifest itself as a project of full and long-term resettlement in the asylum state far beyond the immediate needs of protection.

In this chapter, I argue that the iron discipline of the border requires asylum seekers to play the part of the “good” refugee (a refugee defined by pure protection needs) (Szczepanik, Reference Szczepanik2016). This not only creates a constant culture of bad faith, but it also risks the constant erasure of their agency, notably as it expresses itself in strongly entertained preferences to seek asylum in some countries rather than others. The insistent focus on the protection of refugees, however well-intentioned it may be, ends up imprisoning refugees in a unidimensional identity, that of the performance of their humanitarian frailty. It makes incomprehensible why they would take considerable risks to seek asylum in some countries that are removed from the closest “safe” country. It also makes their specific needs in terms of the eventual location of their asylum “unsayable” for activist groups supporting asylum seekers’ rights who may be wary of blurring the distinction with migrants. Eventually, this plays into the hands of governments vying to make asylum seeking difficult, and ordinary migration even more so. In short, refugees are called upon to symbolically reinforce the very border that they seek to traverse – enlisted in their own oppression, as it were, and in breaking solidarity with their fellow “migrant” travelers who cannot rely on a compelling narrative of persecution.

The chapter begins by highlighting why the agency of asylum seekers in relation to the country where they wish to emigrate is impossible to entertain within the categories entertained by the border. This is reinforced by a politics of defending asylum seekers that constantly reinscribes their difference from migrants – and only manages to protect the former at the expense of the latter. The chapter goes on to show how the preferences of asylum seekers cannot be addressed within the law, which at best offers a very limited and contrived defense of their ability to request asylum “in the country where they find themselves.” Contra this invisibility, the next section emphasizes the obvious sociolegal point that asylum seekers are more than just asylum seekers; namely that, woven into their demand for protection from persecution, are multiple aspirations to gain asylum in particular countries that are also so many normative claims against the border and the obligation it imposes to be either a refugee or a migrant. In the conclusion, the chapter speculates about the legacy of a border ideology that severely constrains asylum seekers’ ability to articulate their mobility as a political claim.

1 The Invisibility of Refugees as Migrants

Consider the explosive debate that started in 2022 following the UK government’s project to send asylum seekers who make it to Britain across the Channel to Rwanda. The project has been described by Human Rights Watch as “cruelty itself” (Ahmed & McDonnell, Reference Ahmed and McDonnell2022). But what exactly is wrong with that plan? Many arguments that it will be expensive or inefficient in deterring smuggling are not particularly principled or based on what would be best for refugees. It may be that the British government is seen as skirting its asylum responsibilities, but in a way it is clearly deploying considerable resources to allow another (apparently willing) state to do so. But the prevailing discourse suggests something more complex, namely a genuine reluctance to send asylum seekers to countries other than the one where they have already secured a foothold as a result of their own, often considerable efforts. What is specifically “cruel,” then, about sending asylum seekers to another country than the one they have managed to reach?

That deportation to Rwanda is not per se the problem is shown by the fact that it is more generally third safe state agreements that have been heavily criticized (Linden-Retek, Chapter 3; Schmalz, Chapter 4). For example, activists have long railed against Australia’s arrangements with Nauru and Papua New Guinea to screen asylum seekers on the Manus Islands. Similarly, the EU’s project to ask the government of Niger to establish frontier zones on African soil has been roundly condemned. Again, though, it is not clear from a protection standpoint what is wrong with being resettled in Nauru (as distinct from being detained in Nauru in squalid conditions, evidently), for example, rather than Australia.Footnote ¹ Nauru may not be as wealthy or welcoming for refugees as Australia (although hasn’t Australia amply shown on some level how unwelcoming it in fact is?) but maybe it is “safe enough” for the purposes of bare protection compared with daily bombardments in Aleppo or Taliban repression?

To be clear, it remains possible of course to challenge safe third country discourse on protection grounds alone, by pointing out that some such countries are, in fact, not safe, or at least not quite as safe as they appear to be. Third countries might turn out to be less than ideal from the point of view of asylum itself. It may be that such schemes are part of an ill-thought-out plan to simply “dump” asylum seekers in countries from which they will promptly leave anew and therefore evidence both a dereliction of duty from host states and ultimately unworkable policies. Israel’s infamous practice of sending Eritreans and Somalis to Uganda and Rwanda with $3,500 cash in their pocket but no official status led many to eventually leave again (Gidron, Reference Gidron2018).

But in many cases third countries will indeed be relatively safe, including for asylum seekers. Notably, the reasons for individual persecution in refugees’ state of origin will often no longer be present. For example, although it has been pointed out quite correctly that Rwanda does not have a perfect human rights record (although nor does the UK, to be clear), the gravest violations it has been accused of are against long-time dissidents and activists heavily involved in the contestation of Kagame’s stranglehold on power. It is not particularly plausible that the Somalis, Yemenis, or Iranians who might end up living in Rwanda would be on the receiving end of that kind of dissident-oriented repression. There is, maybe, just a touch of racism involved, as if an African country could not possibly be a safe haven. Indeed, on protection grounds alone, all of these destinations (Rwanda, Nauru, etc.) may be as good as any. Offshore processing and safe third-country agreements in principle at least ensure that asylum claims are examined, as opposed to practices of pushing back refugees so that they may not even seek asylum (Gammeltoft-Hansen, Reference Gammeltoft-Hansen, Chetail and Bauloz2014).

Finally, the issue is not only with poor countries acting as safe third states. Famously, in Calais, hundreds of migrants have crossed the whole of Europe only to then risk their lives daily to cross the channel to the UK. Calais has often created puzzlement. Why would one risk one’s life as an asylum seeker to reach the UK? Surely refugees are no more likely to be persecuted in France than on the other side of the Channel and are, effectively, quite safe by the time they have made it that far. France may be a country with many flaws, but it is a stable and rich democracy with a well-developed rule of law and rights protections. Do not those individuals believe their claim to being asylum seekers by seeking to enter the UK illegally despite already being in a country where they are objectively safe?

This does not mean, to be clear, that it is a good idea to send Somalis, Yeminis, or Iranians to Rwanda – but not necessarily for protection-focused reasons. It is almost as if governments had chosen to take international refugee law at its word by insisting that they are taking protection needs – albeit only protection needs – seriously. Troublingly, then, it is the one-dimensionality of protection discourse itself that paves the way for governments to promote first safe country logics. After all, if basic protection is ensured, then any safe third state is equal to any other. Safe destination countries are interchangeable. It is that element of the border discourse that makes it incomprehensible and even suspicious that Iranians or Kurds, having made it all the way to Germany or France, would risk their lives on a dinghy to cross the channel. And while there is an obligation under the Refugee Convention to process asylum claims on one’s own territory, it is hard to see how this should be incompatible, in principle, with a distribution, including a fairly authoritarian one, of the burden of processing and welcoming refugees (after all, it is not particularly fair that refugees end up predominantly seeking asylum in some countries rather than others, for example through geographic accident).

2 The Politics of Protecting Asylum Seekers

Instead, what these examples suggest is the inability of conventional asylum seeker discourse to provide a strong account of what is wrong in such situations and, in particular, to ascribe any particular normative value to refugee choice of country. The elephant in the room, however, is the suspicion that asylum seekers also express strong preferences to seek asylum in certain countries rather than others (as explored in more detail in Section 3) – that is, that they want to both seek asylum and immigrate. That agency, however, is what cannot be expressed under the conditions set by the border because of the way it might undermine the notion that refugees are a special kind of migrant defined almost exclusively by their flight from persecution. This could then dangerously feed into state discourses all too willing to consider that refugees have all along been, in fact, would-be migrants.

This means that asylum seekers themselves need to tone down their own agency and essentially play the part of the “good” refugee (a refugee defined by their pure protection needs) (Szczepanik, Reference Szczepanik2016). Asylum seekers’ portrayal by sympathetic constituencies, it turns out, increasingly relies on “perceptions of helplessness, suffering and ‘deservingness’” (Neikirk, Reference Neikirk2017: 63) that objectively marginalize refugee agency. This toning down of agency is also evident in the long-standing construction of the asylum seeker as a victim of circumstances, as clearly emphasized in the slogans of the United Nations High Commissioner for Refugees (UNHCR) that “No one choses to become a refugee”, and “People become refugees not by choice but by circumstance” (United Nations High Commissioner for Refugees [UNHCR], 2022c).

Such statements are problematic given the absence of incompatibility between being propelled to move by circumstances and the significance of choice: Refugees, needless to say, are not automatons but living subjects who ponder the very difficult decision to depart and then exercise multiple forms of agency on the way including, as it turns out, about their hoped for final destination. Having minimized their agency at the point of departure, however, it should come as no surprise that the refugee paradigm does not elaborate on that continued agency en route to the final destination. This is consonant, in fact, with a deeper aversion to representing refugees as historical and political agents as a result of a “totalizing narratives of humanitarianism” (Sigona, Reference Sigona, Fiddian-Qasmiyeh, Loescher, Long and Sigona2014: 378).

The irony, then, is that even as some activists may be wary on a deeper principled level of the refugee/migrant distinction, they are often driven to reinstate it on a strategic and political level to comply with the sprawling exigencies of the border. Most are committed to the notion that we should at least protect refugees and that it would be dramatic if asylum seeker protections were watered down even more significantly than they already are. This then makes it difficult to speak about asylum seekers’ agency in a context where anything that might draw attention to their simultaneous “migrancy” could compromise the imagined purity of their asylum seeking or even endanger the “asylum seeking” category as such (“bogus refugees”). Even “pro-asylum” organizations thus tend to reproduce a discourse of the refugee as a universal dehistoricized category of humanity (Malkki, Reference Malkki1995).

In turn, this has two self-fulfilling if unintended consequences. First, it further plays into government efforts to strongly draw a line between refugees and migrants since the two are constantly being pitted against each other, in large part and ironically of course already to comply with the border’s dichotomous language. Second, asylum seekers themselves, having been so typecast, may internalize the border’s exigencies and foreground their protection needs at the expense of any sense of their life strategies. The refugee label becomes “a master status that defines a person above and beyond any other form of identity” (Marlowe, Reference Marlowe2010). This makes their otherwise expressed agency about choice of country difficult to explain since it cannot be invoked in its own terms, even as it transfers the production of knowledge about the refugee to Western humanitarian “experts” (Rajaram, Reference Rajaram2002).

3 Refugee Choice of Asylum Countries under International Law: Neither Prohibited nor Allowed?

This invisibility of asylum seeker agency is, crucially, not just the result of policies but inscribed in the law itself, so that attempts to seek clarity in the international refugee legal regime tend to be inconclusive at best: The law is part of the border, just as the border is part of the law.

It is true that the Refugee Convention protects individuals who have secured a foothold in one state from being deported to another without consideration of their claim. Nor is there anything that suggests that asylum seekers should as a matter of international law claim asylum in the first safe country. That is clearly a late development that is not borne out by the 1951 Convention itself. A refugee may seek asylum in whichever country she happens to be able to do so. As a British court put it in Ex parte Adimi, therefore, “some element of choice is indeed open to refugees as to where they may properly claim asylum,”Footnote ² in at least the very bare sense that the Convention is silent on the issue. An asylum seeker may thus cross a safe country and nonetheless decide not to make an asylum claim there. In a sense, then, by default at least, the Refugee Convention protects asylum seekers’ de facto choice of a particular country to hear their claim.

But nor does the Refugee Convention particularly facilitate the ability to choose the country in which one seeks asylum. It merely seems to conceive of individual as, at best, “popping up” in countries where they then proceed to claim asylum and have a right to have that claim heard in situ. The normative value of such arrangements seems largely pragmatic and functional. Indeed, although deportation tends to have a bad name (not least because of its historical associations), this is not so much because of deportation itself as because of where one is being deported to. Consider, for example, the possibility that the deportation regime between the UK and Rwanda emerged the other way round, namely that asylum seekers found in Rwanda were, following an agreement with the UK, to have their asylum claims processed there. Leave aside for the moment the geopolitical improbability of such an arrangement: Would we still insist that Rwanda process the asylum applications at the expense of the UK because of a fetishization of the place wherein a claim was first made? Would we do so, moreover, even in the face of evidence that asylum seekers were, in fact, delighted that their applications would be processed in the UK rather than Rwanda? In short, our problem seems to be less with deportations of asylum seekers in and of themselves than the fact that deportation may in some cases run against the better wishes of asylum seekers, in ways that are not accounted for by the dominant model.

As to the argument that refugees may demand asylum in the country of their choosing, although it is a proper construction under international law, it may sound normatively quite thin. Refugee law might still be understood to militate against excessive risk-taking in seeking asylum in that there is something contradictory about fleeing great persecution risk only to then further “needlessly” endanger one’s life in seeking to make an asylum claim in one country rather than another. Moreover, although there is clearly preference in Article 31 of the Refugee Convention for not penalizing asylum seekers for entering the country illegally to claim asylum, this is only if they are “coming directly from a territory where their life or freedom was threatened” and provided they “present themselves without delay to the authorities and show good cause for their illegal entry or presence.” This suggests an emphasis on asylum as a form of immediate protection that is not a particularly good description of asylum seekers who have crossed the whole of Europe to, for example, then seek asylum in the UK.

In fact, even if asylum seekers are not violating international law by choosing where they seek asylum, this does not mean that they cannot otherwise be removed to a third safe country. The Refugee Convention, in particular, does not necessarily prohibit administrative practices such as the European Dublin rules, which allow states to send back asylum claimants who have already been fingerprinted in another EU country to that country. Nongovernmental organizations sensing that the “safe third country” conceit is just an excuse for states to opt out of their asylum obligations – but wary of producing any discourse that might hint at refugees being migrants – have sometimes tackled the practice merely on protection grounds. For example, Pro Asyl argues that: “The very idea that a country could be deemed safe is absurd. Simply taking the European Union as an example, who could seriously argue that some minorities do not currently suffer from discrimination, accompanied by sometimes deadly violence, within its borders?” (Pro Asyl, 2016). That argument, however, seems confused. If there are no safe countries at all, then what is the point of the refugee regime? Clearly, many and probably all (European) states have problems of discrimination, which are nonetheless not sufficient to automatically make them unsafe for the purposes of asylum. The argument seems to claim too much and to miss an opportunity to cast the wrong of expulsion for what it is.

In short, the Convention certainly does not prohibit choice of country where one seeks asylum but nor does it facilitate it, except at best through its silences. Refugee law, one senses, has no strong theory of why it might be appropriate to allow asylum seekers to choose the country where they apply, except through the relatively narrow route of stopping states from going on a slippery slope that leads to the avoidance of their refugee Convention obligations. This ultimately leaves where one manages to seek asylum to chance, including geographical proximity and accessibility, financial means and connections, networks, age/gender/race, and so on. It also encourages risk-taking and trafficking networks. Most importantly, it is normatively impervious to the very real agency that asylum seekers express when they seek to obtain asylum in some countries rather than others.

4 And yet They Chose: Refugees as Migrants

The singular focus on protection as the only goal of asylum seekers flies in the face of more complex understandings of refugee motivations and trajectories that simultaneously challenge the border’s hold on imaginations. Refugees are not solely defined by the fact that they seek protection. They are complex agents with complex backgrounds and ambitions (Barsky, Reference Barsky2017), for whom fleeing conditions of persecution, obtaining protection, and securing prospects for a decent life are all intertwined as part of processes of “mixed migration” (Kane & Peterson, Reference Kane and Peterson1995). Refugees may in some cases have wanted to migrate all along, and persecution merely have been a decisive factor in triggering departure; or they may have never thought of emigrating before encountering persecution but been immediately thrust through their flight into a web of complex life decisions about where to ultimately seek asylum. But migrate they do, in a context where refugee flight “represent[s] one variant of the more general phenomenon of migration” and “involves many of the considerations other migrants go through when physically disconnecting from home regions or countries” (Essed, Frerks, & Schrijvers, Reference Essed, Frerks, Schrijvers, Essed, Frerks and Schrijvers2004: 7).

The point seems to be that, like ordinary migration, the decision to seek asylum is driven by push and pull factors and is, ultimately, a real decision, albeit one evidently constrained by circumstances: One would not, for the sake of argument, seek asylum at any cost anymore, by contrast, than one would do so for merely futile reasons. Clearly, if one’s expected eventual situation abroad were worse than the dangers to which one is exposed at home, then seeking asylum would not make sense. Of course, the problem is compounded by the fact that the decision to flee involves a complex form of risk management in conditions where one never knows fully the exact risk of harm at home, the likelihood of being able to seek and obtain asylum abroad, or the kind of life that one may eventually have as a refugee. The element of rationality involved, however, is not nullified by the fact that in practice the choice of destination is heavily constrained by misinformation, opportunity, and criminal exploitation (Gilbert & Koser, Reference Gilbert and Koser2006).

What does characterize refugees is the fact that they have departed a state because of a well-founded fear of persecution and therefore in conditions that leave less time for the deliberate planning of emigration – yet that does not defeat the fundamentally migratory nature of asylum seeking. In all cases and in a context in which flight and asylum seeking do not simply “happen” but typically require complex organization, courage, and drive, the decision about where to ultimately seek asylum is at least as important as the decision to flee and seek asylum at all. Moreover, that decision-making process is rendered even more pointed by the many efforts deployed by states to make it difficult to seek asylum: This means that refugees can leave very little, in fact, to chance. In particular, the effort to enforce a first-safe-country rule creates an incentive for asylum seekers to make sure that they are not “stuck” in a country which is less than ideal for their purposes, and therefore to redouble their efforts to target a particular country for asylum.

In practice, the decision to seek asylum in some countries rather than others is framed by a series of considerations. The primary one is of course seeking a place where one will be free and safe from persecution or harassment, either from the long arm of the state of origin and also at the hands of the host state. Considerations of relative ease of access and likelihood of obtaining asylum will also feature prominently. But factors more characteristic of any migration process will also determine the choice of country. These include prospects for gainful employment and social integration. Clearly, not all safe states are equally attractive when it comes to opportunities (in that respect the UK may well prove more attractive than Rwanda for many asylum seekers), and even generously granted asylum may lead to a life of de facto economic and social misery. The job market is a factor, as is the possibility of having one’s studies recognized, the existence of welfare services, and the availability of low-skilled work for the unqualified. Cultural factors may also have a role, notably language. To flee the danger of persecution to find oneself in a perpetual state of economic and social alienation is not an attractive prospect.

Finally, asylum seekers have tended to be drawn to countries where, quite simply, they already have personal links and where they may be reunited with family, relatives, community, or more broadly a diaspora for sentimental but also practical and economic reasons. A key for understanding refugee choice, then, is that it is a form of inchoate family reunification, albeit operating wholly aside from the legal category of family reunification. More generally, asylum choice reflects patterns of “chain migration” that depend on kinship networks. Although diasporic studies are sometimes dissociated from the study of immigration, in practice the “pull” factor of already established diasporic communities is a strong one. One European study found that the “network effects” involved in some countries already “having become popular destinations for some time” are the single biggest predictor of choice of asylum country (Neumayer, Reference Neumayer2004, 176). In short, refugees seek to both obtain asylum and to make sense of their life trajectories in ways that may only seem difficult to scrutinize as a result of the disciplinary biases of the law-as-border.

5 Arguing for Refugees as Migrants

One would not think less of an asylum claimant’s claim to be a refugee simply because that claimant was not content with seeking asylum in the first safe country, given the obviousness of the ties that bind to some places rather than others. In taking risks to reach certain countries for protection not only on protection grounds, refugees fully express, de facto, their dual identity as asylum seekers and migrants. They also express, evidently, legitimate human aspirations – just not the ones that the international refugee regime is trained to have its sights on. In that respect, asylum seekers arguably do not merely seek the border: They also call it to account as a particular political-legal choice. The claims that are expressed in the process of crossing borders can be understood as not just strategic but also normative.Footnote ³ They are a challenge to the border itself, and its tendency to prevent asylum seekers from choosing a country that fulfills their aspirations as migrants.

To ignore or deny refugee agency is, at the most basic level, to fail to treat refugees seriously as moral agents seeking the best lives for themselves. States that deny that agency add insult to injury in a sense, compounding an initial persecution in the state of origin by denying asylum seekers the ability to remedy it to the best of their abilities in another state. As such, they fail in honoring a deeper cosmopolitan obligation of hospitality that is also a test of their moral integrity. By highlighting that part of refugees’ trajectory is based on a series of voluntary choices, conversely, one allays fears that to give effect to these choices is to perversely do the persecuting state’s bidding, while tapping into a rich register of liberal reasons why one would want to recognize mobile humans’ “life plans.”

Moreover, as David Owen has argued, there may be deeper sui generis grounds for acknowledging “refugee choices […] as having normative significance, namely that such recognition of refugee as autonomous sources of claims to justification is the normative basis of the political institution of refugeehood.” In such a situation, “[t]o exclude or ignore refugee choices is to fail to exhibit the moral respect for persons expressed in the claim to equal political standing (as effective membership of a state) of persons that the institution of refugeehood is designed to uphold in the face of state failures” (Owen, Reference Owen, Oliver, Madura and Ahmed2019: 36–37).

The denial of refugee agency, moreover, is made even worse when it coincides with long-standing historical legacies of colonialism, discrimination, and war. In such a context, it might be argued that states have particularly onerous obligations to allow asylum seeking if they have historically had a hand in the conditions that have led to persecution in the first place. Such responsibility might be relatively direct and contemporaneous. For example, Iraqi asylum seekers might have a good argument for seeking asylum specifically in the US and the UK, countries that invaded them illegally, precipitating a predictable spiral of violence and resulting persecutions.Footnote ⁴ Afghans could emphasize that they would not have been persecuted by the Taliban had it not been for the US’s sudden and ill-planned pull out.

One could also imagine that asylum seekers would have a particularly strong normative claim to seek asylum in a particular country if that country were a former colonial power whose legacy had created some of the very sources of persecution that led them to flee.Footnote ⁵ Indeed, it may be that persecution in one’s state is linked to longer-term colonial or imperial legacies that have stigmatized certain populations, fostered dictatorship, or made repression endemic. Under that light, for example, Belgium would have a particularly strong responsibility to welcome Rwandan asylum seekers suffering from continued ethnic legacies largely fomented by Belgium colonization (or, which is almost the same, such asylum seekers would have a particularly strong case to make their demands in Belgium).

What agency brings to the table of theories about state responsibilities is the ability to transform the relationship to asylum seekers as one of accountability and not simply of abstract cosmopolitan justice. Within this accountability framework, refugees’ de facto agency acts as a reminder, trigger, and, sometimes, pointed finger. Finally, the argument proposed therein for country choice may be particularly strong in the case of asylum seekers, but that does not nullify its broader viability for all would-be migrants based on a similar analysis of the imbrication of the state of origin and the host state. For example, the movement to theorize reparative justice obligations vis-à-vis climate refugees shows the potential of a remedial approach in further blurring the distinction between conventional refugees and “mere” migrants.

Conclusion: The Border Shapes the Refugee, but Can the Refugee Shape the Border?

In this chapter, I have argued that the division constructed by the border between refugees and migrants has the unfortunate effect, among others, of rendering invisible refugees’ migratory trajectories. This is largely out of a fear that emphasizing such trajectories might actually weaken refugees’ claim to being refugees. This may well be the case in practice, and one may understand the merely tactical need to tone down the extent to which asylum seekers may have strong preferences for seeking asylum in certain countries, lest they be confused with migrants. But I have also stressed that there is no incompatibility between being a refugee and a migrant if one sees asylum seekers’ strategies as complex demands not just for protection in the abstract but for protection in certain countries; and, in fact, not just for protection but, having been forced to leave their states, for the possibility of rich and meaningful lives in particular places in which they came to seek protection.

In effect, refugees may be shaped by the border to act like univocal refugees more than is reasonable or will do justice to their multifaceted lives. What is clear, however, is that refugees also consistently express, most notably through their decisions to leave and travel, visions that one normally associates with migrancy. The problem, however, is that there is no normative narrative to account for such preferences. If choice of country is not to be left to the chance of a quasi-Darwinian struggle (Shachar, Reference Shachar2020b), then under the standard narrative that provides no room for recognizing the agency of refugees, it is at best left to efforts by states to share the burden of asylum seeking in ways that have no logical connection to asylum seekers’ own preferences.

Generally speaking, then, the political cost of the “border within” might be understood as the cost of a culture of bad faith, a culture in which one needs to adopt identities that maximize one’s life prospects for purely instrumental reasons. One result of that culture of hypocrisy is the development of a heavy regulatory and police response to constantly discipline these categories and make sure they are not abused even as the border creates considerable pressures for them to be abused. The border thus acts as a self-fulfilling prophecy, producing the categories that are then internalized by migrants and asylum seekers, and also justifying constant policing to ensure that they continue to act as fix markers of identity. Another political cost of the border might be the erosion of solidarities between “mere migrants” and asylum seekers, even as an increasing body of evidence suggests the proximity of their fates (Scherr, Reference Scherr and Kury2018).

Contra a vision of the international refugee regime being merely and exclusively about protection, an emancipation from the relentless discipline of the border would highlight the case for taking into account asylum seekers’ agency about what country they want to obtain protection in. Indeed, in demanding asylum in this or that country, refugees manifest themselves as political agents of change making pointed demands to particular polities as well as the international system. As such, they also contribute to constantly redrawing and reinventing the border as a porous membrane that cannot prevent the insistent manifestation of political demands to circumvent it.