Perfecting Imperfect Duties: The Institutionalisation of a Universal Right to Asylum

SOME philosophers have argued that citizens of all states have a universal moral right to be provided asylum in other states under certain circumstances of deprivation.¹ In its generic form, the universal moral right to asylum may be understood as a welfare entitlement of all persons, and it is directed against no agent in particular.² But if the right to asylum is against no agent in particular, can it be considered a genuine right in the first place? Some critics have claimed that universal rights to assistance do not constitute genuine rights, as their generic form does not identify the specific agents responsible for their fulfilment. If this criticism of universal rights to assistance is correct, it also undermines the possibility of a universal right to asylum. What should we make of this line of argument? In the current article, I will focus on examining this issue more closely in the context of a universal moral right to asylum. I will leave aside the more specific formulations of the right to asylum, that is, whether the scope of persons with a claim to asylum extends beyond those seeking shelter from persecution.³ As well, I will stop short of providing a justification for a universal moral right to asylum. The article focuses strictly on showing why the fact that in its generic form a universal right to asylum is against the international community at large does not prove to be detrimental to it.

The article is divided into three main sections. In the first section I will focus on the distinction between imperfect and perfect duties, as it serves an important function in the criticism of universal welfare rights. In the second section, I will turn to Onora O'Neill's argument against universal welfare rights. In that section I will suggest three alternative ways in which the idea of a universal right to asylum can be vindicated from O'Neill's criticism. The final section of the article is dedicated to closer enquiry into practical efforts to institutionalise a universal welfare right to asylum. I will outline the general characteristics of three specific institutional schemes aiming to perfect imperfect duties, and examine some of the challenges these schemes face.

I. PERFECT AND IMPERFECT DUTIES

Let us start with the distinction between perfect and imperfect duties. The distinction emerged in the writings of early modern natural law theorists, and the first known theorist to refer to it is Hugo Grotius.⁴ In the following centuries many philosophers, including John Stuart Mill and Immanuel Kant, used the distinction to explain how there could be duties that differ from each other. In the last chapter of Utilitarianism, Mill argues that ‘duties of perfect obligation are those duties in virtue of which a correlative right resides in some person or persons; duties of imperfect obligation are those moral obligations which do not give birth to any right’.⁵ While Mill's account clarifies how the distinction is often used in moral enquiries, it does not tell the whole story. Many philosophers have set forth more as well as less stringent overlapping accounts of perfect and imperfect duties. To exemplify the plurality of interpretations, George Rainbolt outlines eight different formulations of the distinction, including Kant's, Pufendorf's, and Mill's. He notes that Kant alone articulates three formulations of the distinction, and Mill two formulations.⁶ But while there is no general consensus on what exactly the distinction is supposed to represent, there are nevertheless some distinguishable core characteristics that may be argued to constitute the general structure of each category.

A. Perfect Duties : A Preliminary Account

Two core characteristics may be recognised in perfect duties, the first being specified performance. Perfect duties include a specification of a performance, either action or forbearance, which is required to successfully discharge the duty. A common example of a universal perfect duty is the duty not to interfere with others' physical integrity. The duty not to interfere with others' physical integrity identifies a specified performance required from each duty-bearer: non-interference. The universal duty not to interfere with others' physical integrity also specifies the time when the performance should take place: the duty binds regardless of time and place.⁷ Apart from universal perfect duties there can also be special perfect duties. By their form, special perfect duties may be either negative duties or positive duties. Keeping a promise is often given as an example of a special perfect duty. The act of promising, which follows the universal maxim ‘one should always keep one's promises’, constitutes a special relationship between specified agents. In the case of special perfect duties, a specified performance is required to discharge the duty, and apart from specifying the performance the content of the duty also specifies when and where the performance is required.

The second characteristic of perfect duties is the identification of the recipient(s). Perfect duties not only require specified performances from a duty-bearer, but also that the duties are discharged by the duty-bearer to specified recipients. A perfect duty identifies the agents to whom the duty-bearers are obligated to discharge their duties: the duties of non-interference are discharged by each duty-bearer simultaneously to all existing recipients, and the duties of keeping a promise are discharged to the particular recipients specified in the content of the promise. In short, the following preliminary formulation of the generic structure of universal perfect duties may be set forth: universal perfect duties clarify the performance that is required from the duty-bearers, and all recipients can simultaneously enjoy the output when a particular agent discharges his or her universal perfect duties. In turn, special perfect duties come into existence with special relationships, they bind all agents in existing special relationships, and the content of the duty specifies how, to whom, when, and where the duty should be discharged.

B. Imperfect Duties : A Preliminary Account

Imperfect duties are more complex in their generic form than perfect duties are in either of their two senses. Imperfect duties are often considered as universal duties, although some theorists note that there can also be imperfect special duties.⁸ I will leave this category of duties unexamined, as it is not relevant for the purposes of the present enquiry. There are many formulations of universal imperfect duties, and it is not evident that all principles of imperfect duty have the same core characteristics. Kant, for example, recognises imperfect duties of self-improvement towards oneself, and duties of love, respect, and beneficence towards others.⁹ Here, I will focus solely on imperfect duties to assist others.

Imperfect duties constitute a moral category that is separable from ‘non-duties’ in that these duties require agents to make binding moral commitments. Imperfect duties are, after all, duties and therefore stand apart from supererogatory acts. Unlike supererogation, imperfect duties do not require that we behave like saints and heroes by engaging in abnormally risky meritorious actions. Supererogatory acts may be considered to ‘extend beyond duties’, and a failure to engage in supererogatory acts does not result in moral condemnation.¹⁰ But a failure to discharge imperfect duties does bring about moral condemnation. While imperfect duties differ from supererogatory acts, they are also separable from perfect duties in that intrinsically their binding is looser in many senses than the binding in perfect duties. While a universal perfect duty requires individuals to comply with the duty everywhere and at all times, the intrinsic requirements in universal imperfect duties are not specific to the same degree. As the word ‘imperfect’ suggests, by their form these duties are somehow ‘incomplete’.

While there is no general consensus among philosophers on the idea of incompleteness in the context of imperfect duties, the idea of ‘latitude’ may nevertheless be said to be a close approximation of it. Latitude in imperfect duties can be understood to consist of (1) indeterminate action, (2) indeterminate occasion, and (3) indeterminate recipient. One of the ways in which imperfect duties are incomplete is that they give leeway to the duty-bearer regarding the actions with which the duty is eventually discharged. Contrary to the universal duty of non-interference, which specifies the exact required act, the general duty to assist others does not intrinsically specify the particular actions that ought to be performed to successfully discharge the duty. There is a broad range of ways to satisfy the requirements of positive imperfect duties to aid the needy.

The second characteristic of latitude is the indeterminate occasion on which each duty-bearer is obligated to discharge his or her duties of assistance. Imperfect duties do not intrinsically specify whether I should discharge my duties tomorrow or next week. Kant's account of the duty of beneficence provides an example. On Kant's understanding it is our duty to promote the happiness of others sometimes, but one may avoid doing so at any time that one feels inclined.¹¹ The latitude of occasion should not, however, be understood as releasing the duty-bearer fully from discharging the duty. As was noted already, unlike supererogation, imperfect duties morally bind individuals to discharge their duties towards the needy. If I never give aid to the needy, I am surely violating my imperfect duties.

Finally, imperfect duties do not include a specification of the exact agents to whom the duties should be discharged. While universal perfect duties can be discharged simultaneously to all agents, this is not the case with imperfect duties. There can be more needy persons than a duty-bearer has capacities to assist, and for a duty-bearer to discharge imperfect duties of assistance it is only a requirement that the person assists some unidentified needy. Discharging imperfect duties is constrained by limited capacities of positive action and the physical reality of being in one place at a time, and there is nothing in the structure of the imperfect duty that identifies a special relationship between a particular duty-bearer and a particular recipient. In other words, the internal structure of imperfect duties leaves the following question unanswered: to which particular recipient should each duty-bearer discharge her duties? To sum up, the generic structure of imperfect duties to assist suggests that duty-bearers ought to discharge their duties in accordance with the following maxim: assist some needy, somehow, and sometimes.

II. ONORA O'NEILL ON THE GENUINENESS OF MORAL RIGHTS

Now, let us turn to O'Neill's criticism that, if valid, may be thought to undermine the idea of universal right to asylum. O'Neill argues that there exists an asymmetry between ‘universal welfare rights’ and ‘universal liberty rights’. On her view, universal welfare rights are rights to assistance, and universal liberty rights are rights that require only omission from duty-bearers. O'Neill claims that there exists an asymmetry between the two categories of rights because only universal liberty rights are ‘claimable’. She suggests that there are two central conditions for claimability: (1) it is possible to identify the responsible duty-bearers when a right is violated or is not fulfilled, and (2) the right must constitute a claim for a particular determinate performance. If these two conditions are not met, the right in question may not be considered claimable.

O'Neill argues that by their form universal liberty rights correspond to perfect duties; they satisfy the conditions of claimability whether or not specific institutions for protecting rights have been established.¹² The duties corresponding to a universal liberty right can be discharged by a duty-bearer simultaneously to all right-bearers, as there are no positive actions of assistance required from the duty-bearer. For example, a duty-bearer can at least in principle abstain from interfering simultaneously with all bearers of the right not to be subjected to bodily harm. Then, O'Neill argues, if a duty-bearer fails to discharge his or her duties to a bearer of a universal liberty right, the right-bearer has a clear understanding of the specific duty-bearer against whom the right constitutes a claim and of the particular performance required for the right-bearer to have access to the substance of the right.

O'Neill also argues that universal welfare rights do not intrinsically include components that specify the particular duty-bearers against whom these rights constitute claims. Put differently, by their form universal welfare rights may be said to correspond to imperfect duties. O'Neill contends that when we talk about universal welfare rights ‘it may be possible to state what ought to be provided or delivered, but it will be impossible to state who ought to do the providing or delivering [. . .] unless there are established institutions and well-defined special relationships’.¹³ In other words, unless there are institutions in place that form special relationships between right-bearers and duty-bearers, the duties corresponding to a welfare right are imperfect duties instead of special perfect duties. If these perfecting mechanisms exist then a welfare right is a claimable ‘special welfare right’ instead of an ‘unclaimable’ general welfare right. O'Neill argues that claimability is a necessary existence condition for a genuine right, and suggests that unless ‘obligation-bearers are identifiable by right-holders, claims to have rights amount only to rhetoric: nothing can be claimed, waived or enforced if it is indeterminate where the claim should be lodged, for whom it may be waived or on whom it could be enforced’.¹⁴ She concludes that while universal liberty rights and special welfare rights can satisfy the necessary conditions of claimability, universal welfare rights amount only to ‘manifesto rights against unspecified others’, and they seem ‘bitter mockery to the poor and needy, for whom these rights matter most’.¹⁵

O'Neill's criticism of universal welfare rights is applicable to the universal right to asylum in all its more specific forms. There would be no need to vindicate the right to asylum from O'Neill's criticism if it was considered to be a special welfare right. But this would require the existence of global institutions with similar mechanisms of allocating responsibilities between duty-bearers as there are for special welfare rights in welfare states. While there currently exist institutions dedicated to the protection of persons who are fleeing their home countries, these institutions do not set any binding positive obligations of protection. While the United Nations High Commissioner for Refugees (UNHCR) and NGOs actively assist in the evacuation of persons from their home countries and provide temporary shelter and food to thousands, these organisations do not have the authority to grant asylum to any persons. The final decision to provide asylum remains with individual states. In international law the right of states not to grant asylum takes precedence over the right of individuals to receive it.¹⁶ Apart from certain requirements regarding the treatment of asylum seekers, international law recognises only that each state has a binding non-refoulement obligation to persons fleeing their home countries. The non-refoulement obligation is founded on the moral imperative ‘one should not actively cause harm to others’, and it recognises minimum standards of justice in the expulsion of non-citizens from the territory of a state.¹⁷ In other words, we are far from a world where the international mediating institutions for the protection of persons seeking asylum would be comparable to mediating institutions in the welfare state. Then the question becomes, how damaging is O'Neill's criticism to the idea of a universal right to asylum? Next, I will turn to consider three objections to O'Neill's criticism, each of which suggests that O'Neill's claims are not detrimental to the idea of a universal right to asylum.

A. Objection 1

One way to respond to O'Neill's argument is by showing that her claim that only universal welfare rights correspond to imperfect duties is not valid. An attempt to vindicate a universal right to asylum in this way may be considered as a ‘negative strategy’. If it is concluded that universal liberty rights do not correspond strictly to perfect duties, then both categories of rights face the same dilemma regarding claimability: these rights are against unspecified others and for unspecified performances.

There are two different ways in which the validity of O'Neill's claim that universal liberty rights correspond strictly to perfect duties might be questioned. The first argument appeals to the distinction between ‘having’ a right and ‘enjoying’ secure access to the substance of a right. O'Neill's argument relies on a utopian pre-institutional understanding of having rights, and she is not concerned with the range of performances required from others for a right-bearer to enjoy sufficiently secure access to the substance of a right. Only in utopian circumstances where we are not concerned with the protection of rights do liberty rights correspond strictly to perfect duties. In the non-ideal world, however, it is to be expected that not everyone will respect each others' rights. Thus, a meaningful opportunity to have access to the substance of liberty rights is attached not only to perfect duties.

As Henry Shue argues, to have rights in any meaningful sense requires the establishment of protective institutions.¹⁸ Once we move our focus from the utopian pre-institutional circumstances of having rights to the more realistic circumstances of enjoying rights and the establishment of protective institutions, the distinction between welfare rights and liberty rights is no longer obvious. Jeremy Waldron makes this point eloquently. He contends that when we focus on the protection of rights, each right regardless of its form might be ‘best thought of not correlative to one particular duty [. . .] but as generating waves of duties, some of them duties of omission, some of them duties of commission, some of them too complicated to fit easily under either heading’.¹⁹ When the question is about the protection of rights, it is not directly evident which specific duty-bearers should do what and towards whom. On these grounds it may be argued that universal liberty rights, too, correspond to imperfect duties.

There is, however, an obvious response available to this line of argument: we should not be concerned with the institutional protection of rights when analysing the genuineness of rights. Is the question regarding the existence of a right not prior to the question regarding the institutionalisation of a right and its protection? The response of the critic should not be accepted without further explanation.²⁰ In our current world millions have come to enjoy the substance of the universal right to asylum that is outlined in the Universal Declaration of Human Rights (UDHR) Article 14(1) with the help of the UNHCR, active national governments, and many NGOs. At the end of 2009, more than 26 million people—10.4 million Convention refugees and 15.6 million internally displaced persons—were receiving protection or assistance from the UNHCR.²¹ During 2009 alone, 19 countries reported the admission of a total of 112,400 resettled Convention refugees.²² Also, the UNHCR estimates that during 2008–2009 around 2 million stateless persons were granted citizenship in a state other than the state of their origin.²³ For each of these persons, the UDHR right to asylum may be said to amount to something more than ‘mockery’ and ‘rhetoric’.

Conversely, many of the so-called universal liberty rights may be said to constitute mockery and amount only to rhetoric for those millions around the world who are not currently enjoying the substances of these rights. Places like Iraq, Iran, North Korea, Saudi-Arabia, Afghanistan, Sudan, and Somalia are not well known for their respect for such universal liberty rights as freedom of speech, freedom of assembly, and the right not to be subjected to bodily harm. For those millions of persons who are not enjoying the substance of these liberty rights, it is not much of a consolation that these rights may be claimed in principle from particular identifiable duty-bearers. It can easily be forgotten that what is at stake with rights is not simply abstract normative claims, but rather genuine human needs and meaningful opportunities for their fulfilment. The strict focus on pre-institutional circumstances in the evaluation of genuine rights has to be explained by the theorist wishing to defend this position, and it is not enough to conveniently insist that only pre-institutional circumstances matter.

But there is no need to rely on the success of the previous argument. Even if we examine universal liberty rights in pre-institutional circumstances, it is still not the case that they correspond always and only to perfect duties. Elizabeth Ashford outlines this line of argument in her response to O'Neill. Ashford rightly argues that O'Neill's claim essentially refers to relationships between a small group of agents, and it no longer applies when we broaden our focus and attempt to accommodate a more complex set of human relations and social institutions (the aim of which is not the protection of rights).²⁴ In our current world, systems of social interaction are extremely complex, and it has become extremely hard not only to avoid harming distant others, but also to trace moral responsibility for specific violations of liberty rights back to particular identifiable agents.²⁵ Even when moral responsibility may be traced back to particular agents, the group of agents can be extremely large, for example, all persons in Western countries. When the question is about actively caused harms that are the result of complex social systems, the picture of claimability of universal liberty rights seems to be more in line with universal welfare rights: there is a large group of duty-bearers who are morally responsible for the right-bearer not having access to the substance of the right.

At this point it might still be argued that the content of the duties corresponding to universal liberty rights nevertheless specify the particular performances required from each duty-bearer even when the question is about large groups of duty-bearers: each duty-bearer should stop actively harming the right-bearer. Therefore, it is still the case that universal liberty rights correspond strictly to perfect duties. But this response is unsatisfactory. It may also be the case that participation in an unjust social institution actively harming a bearer of a universal liberty right is not reasonably avoidable to a duty-bearer. In these circumstances the primary obligation of the duty-bearer is to reform the unjust institution that is actively harming the right-bearer. This requires positive action, and it is not directly clear what specific acts are required from the duty-bearer.²⁶ What can further complicate the situation is that the collective organisation of duty-bearers may be a necessary condition for any meaningful reformation of the social institution that is actively harming some persons' universal liberty rights. When respecting a universal liberty right requires positive action from duty-bearers, there can remain the same issue of (un)claimability as with universal welfare rights: duty-bearers are required to perform unspecific positive acts. If we follow O'Neill's argument to its logical conclusion, it means that there cannot be genuine liberty rights not to be harmed by unjust social institutions in which particular duty-bearers cannot be reasonably expected to fully avoid participating. If this is correct, there are strong reasons to be sceptical of O'Neill's argument that a right has to correspond strictly to perfect duties in pre-institutional circumstances for it to constitute a genuine right.

B. Objection 2

The second reason why we should reject the conclusion that a universal right to asylum cannot be a genuine right is that, contrary to what O'Neill argues, the claimability of a universal right to asylum does not require the existence of global mediating institutions comparable to those existing in welfare states. This is essentially because the establishment of global mediating institutions is not a necessary condition for the imperfect duties corresponding to a universal right to asylum to become perfected. By its structure an imperfect duty refers essentially to indeterminateness regarding time, place, specific action, and particular recipient. The welfare state perfects imperfect duties by functioning as a ‘clearinghouse’.²⁷ It takes from each bearer of imperfect duties an amount of resources which is sufficient for discharging the duty-bearers' obligations but not necessarily enough to cover all the needy. The welfare state then distributes the resources to the needy, therefore constituting a mediating link, that is, a special relationship, between bearers of imperfect duties and the recipients. Within the modern welfare states, a common way of perfecting imperfect duties has been the establishment of a system of taxation.²⁸

But while the existence of a welfare state imposing a system of taxation is one way in which imperfect duties may be perfected, it is not the only way. Other circumstances can constitute special relationships that perfect imperfect duties. Special relationships that occur in the absence of mediating institutions can specify which particular duty-bearer should discharge duties to which particular recipient, how, and when. One common example of a scenario in which imperfect duties can become perfected by other circumstances is a situation of easy rescue. Circumstances may single out one particular duty-bearer to rescue a needy person because there are no other duty-bearers who are in a similarly situated position to assist.

But even if we expanded the picture from an isolated easy rescue to the assistance of persons who are bearers of the universal right to asylum, there seems to be no principled reasons why the imperfect duties could not be perfected by circumstances. Duties to assist are imperfect if there are more right-bearers than a duty-bearer may be reasonably expected to assist, or when within a large group of potential duty-bearers it is not directly clear who should do what. But as Zofia Stemplowska rightly points out, from this indeterminacy it does not directly follow that the rights corresponding to imperfect duties are not claimable. We could just as well allow all those whose universal welfare rights generate imperfect duties to claim assistance from anyone ‘who has not taken reasonable steps to fulfil her or his duty (to whomever and in whatever reasonable way she or he might choose)’.²⁹ In this type of claimability system, a special relationship comes into existence when a right-bearer singles out a duty-bearer and the duty-bearer has not already taken steps to commit to assist some other right-bearers. The choice of a particular action, the choice of the particular recipient, and the choice of the particular time has in this type of scenario been decided by the right-bearer on the duty-bearer's behalf.

As was suggested earlier, there currently exist no systematic global mechanisms to allocate responsibilities to fulfil the right to asylum. But the absence of systematic mediating institutions at the global level does not mean that special relationships cannot be formed at all. Special relationships can be said to come into existence when a bearer of the right to asylum enters the jurisdictional territory of a particular state capable of offering protection. In fact, when a person currently seeks asylum in a particular state, he or she is generally said to ‘claim asylum’. For a person to claim asylum is to single out a state comprising bearers of imperfect duties to assist her. Then, it may reasonably be asked whether a state, which has not already taken reasonable steps to discharge its members' imperfect duties to bearers of the right to asylum, may reject the claim of a right-bearer entering the state's jurisdictional territory by appealing to the idea of imperfect duties. When a state argues that it wishes to deport a particular right-bearer claiming asylum in order to assist some other non-identified group of right-bearers on some later occasion, the right-bearer who is already claiming asylum may rightly respond with moral indignation.

In the current world, the protection of the UDHR Article 14 follows fairly closely this idea of claimability. Therefore, the system of claimability for the right to asylum sketched above is not far-fetched. Currently, the binding non-refoulement obligation entails that states can find themselves under a de facto obligation to provide asylum to persons that have entered their jurisdiction and claimed asylum. First-asylum states may not find other states willing to share the burdens, and they cannot deport the asylum seekers back to their home countries.³⁰ The current system could be restructured into a form that recognises for each state a positive obligation backed with legal sanctions to provide asylum to right-bearers entering the state to claim asylum. The central reason why the non-refoulement obligation has not been revised into a form of positive obligation is centrally because there is a lack of political will to do so, not because to do so would be excessively problematic.³¹

C. Objection 3

The final objection focuses on O'Neill's account of claimability as an existence condition of a right. It may be reasonably asked why it is exactly that claimability in the sense O'Neill formulates it constitutes a necessary condition for the existence of a right. As has been noted already, claimability in the sense that O'Neill refers to it requires that particular duty-bearers responsible for the violation or the unfulfilment of a person's rights are identifiable. While the exact identification of the particular agent responsible is surely a central condition for establishing blameworthiness, it may be reasonably asked why we should hold the identification of the particular responsible duty-bearers to be a necessary condition for the existence of a right.³²

Here, O'Neill's own argument on prior questions regarding rights may be turned against her. As was noted earlier, O'Neill argues that we should focus on the pre-institutional circumstances of rights violations. But we may just as well ask her why we should be focusing on the violations or the unfilfilment of rights instead of the prior question regarding their existence. It might be argued that the allocation of duties is a further question of strategy that follows the moral question regarding the existence of a right, and that the deontic implications of a right are essentially dynamic. This approach to the concept of right differs noticeably from O'Neill's approach.³³ It starts by examining what exactly can constitute a sufficiently weighty reason to hold others under a duty of justice.³⁴ On this view, rights function as a justificatory foundation for obligations to make the corresponding imperfect duties perfect.³⁵ Following Jeremy Waldron, it could be argued that a universal right to asylum may be considered ‘a normative resource base from which a whole array of moral requirements can be developed’.³⁶ Then, the vindication of a universal right to asylum is not dependent on its pre-institutional claimability, but instead primarily on whether or not the need for asylum under certain circumstances of deprivation is sufficiently weighty to ground obligations of justice in others.

On this approach, a universal right to asylum may be understood as generating binding moral obligations for the creation of circumstances under which it becomes a realisable right to all right-bearers.³⁷ In other words, there are ‘indirect duties’ corresponding to a universal right to asylum to create, maintain, and enhance global mediating institutions that directly fulfil the right.³⁸ Joel Feinberg argues that universal welfare rights that correspond to imperfect duties constitute ‘permanent possibilities of right, the natural seed from which rights grow’.³⁹ But if it is the case that the need for asylum constitutes a sufficiently weighty reason to ground duties of justice in others, it may be considered something more than simply a ‘possibility of right’. It is then a possibility with the weight of justice attached to it.

At this point it might be responded that the argument overlooks the fact that imperfect duties are unenforceable. The argument regarding the unenforceability of imperfect duties may be understood either as a claim that imperfect duties cannot be enforced or as a claim that they should not be enforced. But there is nothing in the concept of imperfect duty to support either of these objections. The central element of the concept is latitude, which has already been suggested to be perfectible with the occurrence of circumstances that constitute special relationships and with the establishment of mediating institutions. There is nothing in the concept about normative justification against enforcement.⁴⁰ While clearly enforceability requires that duties are specifiable, this constitutes only a practical epistemic condition for their enforcement. It does not undermine the justification for the enforcement of duties of beneficence, and it carries no moral weight in the argument. To simply claim that imperfect duties should not be enforced because they are imperfect is to beg the question regarding the legitimacy of enforcement altogether. Before it can be concluded that imperfect duties should not be enforced it needs to be shown that a right corresponding to imperfect duties is not sufficiently weighty to function as a justification for the enforcement of duties.⁴¹

To sum up the current section, one reply to the criticism that the right to asylum amounts to a ‘manifesto right’ is: ‘so what?’ Whether we in the end call it a ‘metaright’, ‘manifesto right’, ‘possibility of a right’, ‘claim’, or a ‘genuine right’ is a matter of indifference if we follow the account of rights outlined in the current section. As such it can set binding moral demands for the structuring of all territorial political communities and the international domain. To give up on a universal right to asylum before genuine attempts to establish global mediating institutions have been made is to overlook the possible moral weight of the claim and the fact that there are perfectible duties corresponding to it. The excuse of inadequate institutions that is used to defend the abandonment of a universal right to asylum is, in the words of Shue, ‘not simply misguided, but perverse. The absence of the means for fulfilling a goal is being used as the excuse for not fulfilling it, when, if we seriously had the goal, we would be working on the creation of the necessary bridges from here to there.’⁴²

III. THREE POLICY PROPOSALS FOR PERFECTING STATES' IMPERFECT DUTIES

So far I have considered alternative ways to vindicate a universal moral right to asylum from O'Neill's criticism. In the current section, I will focus more closely on efforts to perfect the imperfect duties corresponding to a universal right to asylum. I suggested earlier that the concept of imperfect duty refers essentially to indeterminateness regarding time, place, specific action, and particular recipient. I also suggested that one mediating institution that can perfect the imperfect duties corresponding a universal right to asylum is the state. The state may be understood to constitute a ‘clearinghouse’ that can help to perfect the imperfect duties of its citizens by—among other things—taxing them and by using the taxes to make collective provisions for asylum seeking right-bearers. But it is important to recognise that the issue of perfecting imperfect duties does not occur strictly at the individual level. In addition to the individual level there is a need to address the issue also at the interstate level. Unless there are some ‘perfecting arrangements’ between states, each particular state is practically in an identical situation as an individual duty-bearer whose duties are not being perfected by mediating institutions: it remains unclear which state should do what in the protection of a universal right to asylum.

Then the question becomes: how exactly should the imperfect duties corresponding a universal right to asylum be perfected at the interstate level? In the current section I will turn to examine more closely three practical policy proposals: ‘the compensatory scheme’, ‘the physical burden-sharing scheme’, and ‘the tradable quota scheme’. The examined policy proposals suggest collective frameworks of action for states. Put differently, the policy proposals specify what exactly each particular state comprising bearers of imperfect duties should do vis-à-vis other states in the protection of a universal right to asylum. The following enquiry is limited in the sense that I will stop short of offering a conclusive ranking of the introduced stand-alone models. My aim here is strictly to offer a roadmap for future moral enquiries focusing on the protection of a universal right to asylum. I will outline the central characteristics of each proposal, look at the real-world circumstances in which they have been implemented in preliminary forms, and raise some challenges the schemes face as stand-alone models.

A. The Compensatory Scheme

One policy option for perfecting imperfect duties is ‘the compensatory scheme’. This proposal follows the model of taxation systems in welfare states, and adapts it to international circumstances of refugee protection. In the compensatory scheme all duty-bearers contribute (through their states) to a ‘global refugee protection fund’ an amount of resources which is sufficient for discharging their obligations, and those states facing floods of right-bearing asylum seekers are compensated for their protection efforts from the existing fund. As a policy proposal, the compensatory scheme is not far-fetched. In the current world, the UNHCR, which receives its annual budget ($2.78 billion for the year 2011) from the UN member states, may be argued to constitute a compensatory arrangement that is perfecting imperfect duties. The UNHCR budget is dedicated to the global assistance of ‘persons of concern’, which include persons who qualify as refugees under the 1951 UN Convention Relating to the Status of Refugees, persons who have been forced to leave their countries as a result of conflict or events seriously disturbing public order, returnees, stateless persons, and, in some situations, internally displaced persons.⁴³ In addition, the compensatory scheme has been implemented in a preliminary regional form. The European Refugee Fund (ERF), which has the budget of 628 million Euros for the period 2008–2013, distributes funds among EU member states on the basis of ‘objective criteria relating to the number of asylum seekers’.⁴⁴ Following the ERF and the UNHCR as templates, a comprehensive global scheme could be established that discharges imperfect duties of citizens of all states in the protection of a universal right to asylum.

There are, however, several issues of concern that need to be addressed satisfactorily in the implementation of a global compensatory scheme. Let us consider two challenges that the compensatory scheme faces as a stand-alone scheme. Firstly, there are questions regarding reward incentives for non-entrée policies that seem to remain in place in the compensatory scheme. In the scheme, the non-refoulement principle constitutes the fundamental principle on the grounds of which it is determined which particular states physically accommodate the right-bearers. As asylum claims filed on the territory of one state effectively release other states from physical accommodation, states know that with the implementation of strategic policies they may unilaterally influence the amount of ‘physical burden’ they will eventually have to bear. Considering these factors, states can engage in self-interested calculation and end up applying evasive tactics in order to avoid asylum seekers filing asylum claims on their territories. This can undermine the effectiveness of protecting the right to asylum and lead to asylum seekers' prolonged search for a state that is not implementing non-entrée policies.

Secondly, it needs to be explained why the physical burdens of the protection of a general right to asylum should not be distributed in a more equitable way between states with capacities to accommodate right-bearers. The adoption of the compensatory scheme as a global stand-alone scheme entails that some states can be ‘unlucky’ and have obligations to physically accommodate a large number of asylum seeking right-bearers simply due to the fact that they share borders with countries facing humanitarian crises. Is it genuinely the case that there are no moral requirements for the establishment of institutions that guarantee a more equitable distribution of physical burdens in the protection of a right to asylum than the non-refoulement principle? Put differently, a satisfactory answer needs to be offered to citizens of states bearing the most physical burdens of protecting a right to asylum when they ask: ‘why should we accommodate all the right-bearers entering our territory from the neighbouring countries? After all, as in many cases bearers of the right to asylum can migrate to other states from the first-asylum state, it is not inevitable that the burdens of physical accommodation fall on the first-asylum state.’

B. The Physical Burden-sharing Scheme

Another option for perfecting imperfect duties at the interstate level is ‘the physical burden-sharing scheme’. This policy proposal suggests that physical accommodation quotas should be established for each state on the grounds of their citizens' imperfect duties. The adoption of a physical burden-sharing scheme entails the establishment of a collectivised status-determination process. Contrary to the compensatory scheme, in the physical burden-sharing scheme states no longer determine whether asylum seekers arriving on their territory are entitled to be physically accommodated in the state. When an asylum seeker arrives in a particular country and enters the collectivised status-determination process, she is claiming to be recognised in one of the eventually distributable quotas.⁴⁵ In other words, each state determines the status of asylum applicants on behalf of the collective of states they represent. The implementation of the collectivised status-determination process and the establishment of proportional quotas of physical burden for each state entails that whatever the outcome of status-determination in a particular state, no state is required to physically accommodate all the bona fide right-bearers it has the capacity to accommodate.

As is the case with the compensatory scheme, there currently exists a preliminary physical burden-sharing scheme. A range of UN member states have committed themselves to accommodating annual quotas of UNHCR refugees. It might be argued that by binding themselves to annual quota obligations these states have moved to perfect their citizens' imperfect duties in the global protection of the right to asylum. Every year with the help of UNHCR a number of vulnerable Convention refugees are resettled in a range of ‘quota countries’. Since 2008, the number of quota countries has doubled from 12 to 24, and in 2010 the 24 resettlement countries provided nearly 80,000 places for UNHCR resettlement submissions.⁴⁶ The UNHCR resettlement programme constitutes an institutionalised system of physical burden-sharing which provides a template for a more comprehensive scheme. The programme could be redesigned in such a way that it perfects imperfect duties in the international domain and offers protection to those who bear a universal right to asylum.

The physical burden-sharing scheme avoids one incentive states have for implementing non-entrée policies which exists in the compensatory scheme. The non-refoulement obligation still applies to each state in the physical burden-sharing scheme, but it is not the primary mechanism of distributing right-bearers to their final place of asylum. In addition, there is another benefit that the physical burden-sharing scheme has over the compensatory scheme. The collectivised status-determination process can deter false asylum applications in a way that is not available to the compensatory scheme. Those who are not in actual need of protection may attempt to exploit the global asylum system to gain entry to a particular country. But if the applicants are not certain that they will eventually be accommodated into the country where they have applied for asylum, those without genuine need may be more hesitant to claim asylum.⁴⁷

But, in the end, the physical burden-sharing scheme is not without challenges as a stand-alone scheme. Firstly, there are questions related to the morality of family reunification: is it morally acceptable to reject family reunification in order to maintain a system that distributes right-bearers proportionally between countries with the capacity to accommodate them? If the correct answer to this question is ‘no’, this entails that the physical burden-sharing scheme may constitute an inadequate stand-alone model for perfecting imperfect duties in the protection of the right to asylum. If family reunification is a consideration that overrides the aim of proportional physical burden distribution, and if efforts toward family reunification lead to disproportional physical accommodation of right-bearers, the physical burden-sharing scheme may have to be complemented with the compensatory scheme.

Secondly, there are questions related to circumstances where the creation of asylum in the right-bearers' home states constitutes the most effective form of protection. In general terms, a moral right to asylum may be understood as a right that entitles persons to access a delineated protective zone in which they receive shelter from certain harms and threats of harm to which they would be vulnerable outside the zone. In a world divided into territorial states there are two centrally different ways in which a right to asylum may be protected. Right-bearers are either provided access to a territorial state other than their home state, or a protective zone is created in the right-bearers' home state. Then, the following questions emerge: should we hold on to the physical burden-sharing scheme even at the cost of effective protection of the right to asylum? Or should it be accepted that the physical burden-sharing scheme as a stand-alone model constitutes an inadequate arrangement for perfecting imperfect duties? If sometimes the second strategy of protection can more effectively provide access to the substance of a right to asylum than the first strategy, are there sufficiently strong reasons to conclude that global mediating institutions should be established in such way that they take into consideration both of these strategy options?

C. The Tradable Quota Scheme

It is also possible to combine the physical burden-sharing scheme and the compensatory scheme into a single stand-alone scheme. This synthesis is called the ‘tradable quota scheme’.⁴⁸ The starting-point of the tradable quota scheme is the physical burden-sharing scheme. It recognises the collectivised status-determination process and the establishment of physical quotas for each state with capacities to accommodate right-bearers. But the tradable quota scheme also expands beyond the physical burden-sharing scheme, as it allows states to trade the initially allocated physical quotas. In other words, the scheme is less rigid than the physical burden-sharing scheme with regard to the form in which the burdens of protecting the right to asylum are borne by duty-bearers. It entails the establishment of a market forum in which states can deliberate on the final form of discharging their citizens' duties. In cases where there are no agreements on compensation, the quotas are accommodated as in the physical burden-sharing scheme: the state to which a quota is initially allocated accommodates its non-traded quota. But if states come into bilateral agreements on compensation, the physical quotas may be accommodated in states other than the states to which they were initially allocated.

While the tradable quota scheme constitutes probably the most radical proposal of the three models examined here, it is not without a precedent. A preliminary market scheme existed to resettle persons displaced by the Vietnam War. After the communist victories in Vietnam, Cambodia, and Laos in 1975, over two million people fled to the neighbouring countries. These people received fairly little international assistance until the principles of a comprehensive resettlement program where laid down at a Geneva conference in 1979. The established resettlement program was based on explicit recognition of responsibility-sharing, and the availability of first asylum in the Southeast Asian region was made contingent on resettlement in other countries.⁴⁹ From 1979 until 1989 over 1.7 million Indochinese refugees were resettled through this preliminary market system shaped by immigration criteria, humanitarian concerns, and political factors.⁵⁰

As the physical burden-sharing scheme constitutes the starting-point for the tradable quota scheme, it should be obvious that the scheme faces identical challenges. In the implementation of a tradable quota scheme the issue of family reunification needs to be addressed satisfactorily, as well as the circumstances where the more effective strategy in the protection of persons' rights to asylum is the creation of safe zones in their home countries. In addition, there are further moral challenges that need to be examined more closely. These issues concern the morality of the trading practices: does the scheme humiliate right-bearers when it incorporates them into tradable quotas? Does the scheme allow discriminatory treatment of the right-bearers? Does the scheme allow affluent states to exploit less affluent states in the marketplace? What are the forms of compensation, if any, that are morally acceptable? In a nutshell, the appeal of the tradable quota scheme as an arrangement for perfecting imperfect duties depends essentially on how we rank it in relation to the physical burden-sharing scheme, that is, the question is about whether the tradable quota scheme constitutes a morally desirable extension that gives freedom to states to choose how to discharge their citizens' duties, or whether it introduces morally problematic trading practices.

IV. CONCLUSION

In this article I have examined the universal entitlement to asylum and its protection. I claimed that O'Neill's criticism of universal welfare rights does not prove to be detrimental to the possibility of a universal right to asylum. I outlined three separate lines of argument against O'Neill, each of which supports the same conclusion from a different perspective. I also introduced three policy proposals for perfecting the imperfect duties of the protection of a universal right to asylum. The policy proposals specify (in different ways) what exactly each particular state should do with respect to asylum seeking right-bearers. Although the policy proposals offer guidelines for the perfection of imperfect duties at the interstate level, it is important to note that they are directly linked to efforts to perfect imperfect duties at the individual level. In fact, the perfection of duties at the interstate level may be understood as an essential step in further efforts to perfect individual duty-bearers' imperfect duties. This is centrally because states cannot satisfactorily perfect their citizens' duties unless it is clear what citizens of each state should collectively do in the global protection of a universal right to asylum. Put differently, clarifying collective action at the interstate level helps also to solve unclarities regarding action at the individual level. Once it is clearer what a state should do for particular right-bearers, the state can move to implement specific intra-state institutions that perfect the imperfect duties of its citizens. In more general terms, the three policy proposals I have introduced can be understood to offer a roadmap for future moral enquiries focusing on the protection of a universal right to asylum. If it is concluded that a universal moral right to asylum in fact exists, it is worthwhile to consider the moral dimensions of the three proposals more closely.

Footnotes

• 1

  On the idea of a universal moral right to asylum see Joseph Carens, ‘Who should get in? The ethics of immigration admissions’, Ethics & International Affairs, 17 (2003), pp. 95–110; Michael Dummett, On Migration and Refugees (London: Routledge, 2001), p. 37; David Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2007), pp. 224–7.

• 2

  Put differently, it may be considered as something akin to an international version of the UDHR Article 8 stating that ‘everyone has the right to an effective remedy by the competent national tribunals for acts violating the fundamental rights granted him by the constitution or by law’.

• 3

  The UDHR Article 14 (1), which outlines one formulation of the right to asylum, recognises persecution as a necessary condition for being eligible for asylum. When Article 14 was drafted, many states rejected the position that there is a moral obligation to grant asylum to any persons, let alone that any possible moral obligation should formulated into a legal obligation. The original draft of Article 14 included the language ‘be granted asylum’, but it finally settled into the form ‘seek and to enjoy asylum’. Put differently, apart from outlining a controversial account of the scope of persons with the right to asylum, in its current form the UDHR right to asylum is effectively a ‘right of states to grant asylum’ rather than a right of persons to be granted asylum. See Jane MacAdam, ‘Asylum and the Universal Declaration of Human Rights’, Refugee Survey Quarterly, 27 (2008), pp. 3–12.

• 4

  Grotius makes a distinction between ‘perfect right’ and ‘imperfect obligation of charity’. He writes that the imperfect obligation of charity and other virtues of the same kind are not cognisable in a court of justice. Hugo Grotius, On the Law of War and Peace (Whitefish: Kessinger, 2004), II.XXII.xvi, p. 214.

• 5

  John Stuart Mill, Utilitarianism (London: Longmans, Green, Reader, and Dyer, 1867), p. 74.

• 6

  George Rainbolt, ‘Perfect and imperfect obligations’, Philosophical Studies, 98 (2000), 233–56 at p. 233.

• 7

  On how perfect duties specify the required performance and occasion, see Violetta Igneski, ‘Perfect and imperfect duties of aid’, Social Theory and Practice, 32 (2006), 439–66.

• 8

  Onora O'Neill, Towards Justice and Virtue (Cambridge: Cambridge University Press, 1996), p. 152.

• 9

  Immanuel Kant, The Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge University Press, 1991). For an insightful analysis of Kant's account of perfect and imperfect duties, see Thomas E. Hill Jr., ‘Kant on imperfect duty and supererogation’, Kant-Studien, 62 (1971), 55–76. See also Kok-Chor Tan ‘Kantian ethics and global justice’, Social Theory and Practice, 23 (1997), 53–73.

• 10

  See Joel Feinberg, ‘The moral and legal responsibility of the bad Samaritan’, Criminal Justice Ethics, 3 (1984), 57–69 at p. 64.

• 11

  Immanuel Kant, The Metaphysics of Morals, p. 194. See also Thomas E. Hill Jr., ‘Kant on imperfect duty and supererogation’, p. 64; Liam Murphy, Moral Demands in Nonideal Theory (Oxford: Oxford University Press, 2000), pp. 71–2. Murphy points out that if Kant's maxim of beneficence is accepted as valid, this entails that it is forbidden for a person to adopt a maxim of indifference to the welfare of others, but a few minor acts of beneficence from time to time would seem to be enough to re-establish her good name.

• 12

  O'Neill, Towards Justice and Virtue, p. 132.

• 13

  Onora O'Neill, Bounds of Justice (Cambridge: Cambridge University Press, 2000), p. 105.

• 14

  O'Neill, Towards Justice and Virtue, p. 129. This position is close to Mill's argument. Mill argues that ‘no one has a moral right to our generosity or beneficence, because we are not morally bound to practise those virtues towards any given individual’. J. S. Mill, Utilitarianism, p. 75.

• 15

  O'Neill, Towards Justice and Virtue, pp. 132–133; Bounds of Justice, p. 125. The concept ‘manifesto right’ appeared previously in Joel Feinberg, Social Philosophy (New Jersey: Prentice-Hall, 1973), pp. 94–95. According to Feinberg, the universal social rights in UDHR articles 22–27 are ‘rights in an unusual new “manifesto sense”, for, unlike all other claim-rights, they are not necessarily correlated with the duties of any assignable persons’.

• 16

  Emma Haddad, The Refugee in International Society (Cambridge: Cambridge University Press, 2008), p. 79; Satvinder Singh Juss, International Migration and Global Justice (Aldershot: Ashgate, 2006), pp. 47–52.

• 17

  One of the international treaties it is outlined in is the Convention. Article 33 of the Convention states that ‘no contracting state shall expel or return (“refouler”) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened’.

• 18

  Henry Shue, Basic Rights (Princeton: Princeton University Press, 1996), p. 39.

• 19

  Jeremy Waldron, Liberal Rights (Cambridge: Cambridge University Press, 1993), p. 25.

• 20

  As Waldron, ibid., p. 7, rightly asks, why on earth would it be worth fighting for a person's liberty if the person were left in a situation in which the liberty could not in practice be exercised?

• 21

  UNHCR, ‘Global trends 2009’, pp. 4–5, <http://www.unhcr.org/4c11f0be9.html> (accessed 19 July 2011).

• 22

  Ibid., p. 3.

• 23

  UNHCR, Global Report 2009, ‘The Year in Review’, p. 4, <http://www.unhcr.org/4c08f2ee9.pdf> (accessed 19 July 2011).

• 24

  Elizabeth Ashford, ‘Duties imposed by the human right to basic necessities’, Freedom from Poverty as a Human Right, ed. T. Pogge (Oxford: Oxford University Press, 2007), pp. 183–236 at pp. 195–8.

• 25

  See Judith Lichtenberg, ‘Negative duties, positive duties, and the new harms’, Ethics, 120 (2010), pp. 557–578.

• 26

  Elizabeth Ashford, ‘Duties imposed by the human right to basic necessities’, p. 201.

• 27

  This characterisation of the welfare state as a mediating institution is outlined in Waldron, Liberal Rights, p. 17.

• 28

  Feinberg, ‘The moral and legal responsibility of the bad Samaritan’, p. 66.

• 29

  Zofia Stemplowska ‘On the real world duties imposed on us by human rights’, Journal of Social Philosophy, 40 (2009), 466–87 at p. 482.

• 30

  Of course, the system of claimability proposed here would in all likelihood lead to an extremely unfair distribution of burdens between states. But the resulting unfairness in burden-bearing does not directly undermine the current argument. What the argument regarding unfair burden-bearing between duty-bearers suggests is that there are moral reasons to restructure the proposed system of claimability into a form in which the overall burdens in the protection of a universal right to asylum would be distributed more fairly.

• 31

  There are currently no active political efforts to renegotiate the 1951 Convention Relating to the Status of Refugees, which constitutes the central instrument of protection for persons covered by Article 14 of the UDHR. One reason for the lack of renegotiation efforts is the fear that any initiative to modify the Convention in the current political environment could result in lower protection standards.

• 32

  As John Tasioulas rightly points out, it can be asked why the existence of a plurality of mechanisms for securing the right has to undermine the very existence of the right prior to its institutional embodiment. John Tasioulas, ‘The moral reality of human rights’, Freedom from Poverty as a Human Right, ed. Pogge, pp. 75–101 at p. 94.

• 33

  As Waldron points out, it may be the case that the phrase ‘moral right’ does not have a single correct meaning. (Waldron, Liberal Rights, p. 206.)

• 34

  See Joseph Raz, ‘On the nature of rights’, Mind, 93 (1984), 194–214; James W. Nickel, ‘How human rights generate duties to protect and provide’, Human Rights Quarterly, 15 (1993), 77–86.

• 35

  Kok-Chor Tan, Justice without Borders (Cambridge: Cambridge University Press, 2004), p. 52.

• 36

  Waldron, Liberal Rights, p. 213.

• 37

  Amartya Sen argues that there can be a metaright to have policies p(x) that genuinely pursue the objective of making the right to x realisable in ‘Rights and metarights’, The Right to Food, ed. P. Alston and K. Tomasevski (Hague: Nijhoff, 1984), pp. 69–81 at p. 70.

• 38

  The term ‘indirect duties’ appears in Henry Shue, ‘Mediating duties’, Ethics, 98 (1988), 687–704 at p. 696.

• 39

  Joel Feinberg, Rights, Justice, and the Bounds of Liberty (Princeton, NJ: Princeton University Press, 1980), p. 153.

• 40

  See also Waldron, Liberal Rights, p. 17.

• 41

  As Tan, Justice without Borders, p. 52, notes, ‘the imperfect duties corresponding to various rights are not charitable acts but duties, even if it means that we need to first assign and specify them before they can be fully enforced.’

• 42

  Henry Shue, ‘The interdependence of duties’, The Right to Food, ed. Alston and Tomasevski, pp. 83–95 at p. 94.

• 43

  UNHCR, ‘Biennial programme budget 2010–2011 of the Office of the United Nations High Commissioner for Refugees 2009’, p. 30, <http://www.unhcr.org/4abc7cc19.html> (accessed 19 July 2011).

• 44

  European Commission: Home Affairs, ‘The European Refugee Fund 2011’, <http://ec.europa.eu/home-affairs/funding/refugee/funding_refugee_en.htm> (accessed 19 July 2011). See also Eiko Thielemann, ‘Between interests and norms: explaining burden-sharing in the European Union’, Journal of Refugee Studies, 16 (2003), 261–2.

• 45

  On the establishment of a collectivised status-determination process, see Peter Schuck, ‘Refugee burden-sharing: a modest proposal’, Yale Journal of International Law, 22 (1997), 277–9.

• 46

  UNHCR, ‘Projected global resettlement needs 2011’, p. 3, <http://www.unhcr.org/4c31e3716.html> (accessed 19 July 2011).

• 47

  See also Joseph Carens, ‘The philosopher and the policymaker: two perspectives on the ethics of immigration with special attention to the problem of restricting asylum’, Immigration Admissions, ed. M. Hailbronner, D. A. Martin and H. Motomura (New York: Berghahn, 1997), pp. 3–50 at p. 37.

• 48

  An account of the tradable quota scheme was first suggested by a Yale legal scholar Peter Schuck in ‘Refugee burden-sharing’, pp. 270–88. The idea of tradable migration quotas as means to global poverty reduction has been considered in David De la Croix and Axel Gosseries, ‘Procreation, migration, and tradable quotas’, Population Aging, Intergenerational Transfers and the Macroeconomy, ed. R.L. Clark, A. Mason and N. Ogawa (Cheltenham: Edward Elgar, 2007), pp. 227–49; Jesús Fernandez-Huertas Mortaga and Hillel Rapoport, Institute for the Study of Labor Discussion Papers, <http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1867034> (accessed 11 August 2011).

• 49

  Astri Suhrke, ‘Burden-sharing during refugee emergencies: the logic of collective versus national action’, Journal of Refugee Studies, 11 (1998), 405–6.

• 50

  Peter Schuck, ‘Refugee burden-sharing’, pp. 254–9. Astri Suhrke, ‘Burden-sharing during refugee emergencies’, p. 405.