An outline of Miller's approach

Miller's carefully crafted account is framed by three methodological commitments. First, a dual aspect approach to human beings as subjects of justice which present us not simply as ‘needy and vulnerable creatures who cannot live decent, let alone flourishing, lives unless they are given at least a minimal bundle of freedoms, opportunities, and resources’ (emphasis added) but also as ‘choosing agents who must take responsibility for their own lives’ (pp. 5–6). This insistence on viewing others under both aspects is an expression of respect: ‘Our relationship [to others] becomes a more equal one to the extent that we consider not only their needs but also their capacities for choice and responsibility.’ (p. 7). Second, a contextualist approach to principles of justice: ‘the principles that tell us what counts as a just distribution of some good are specific to the context in which the distribution is taking place […] the relevant principle will depend on what is being distributed, by whom and among whom: especially on the kind of relationship that exists between the people among whom the distribution is occurring.’ (pp. 13–4). Note that it is perfectly possible for an egalitarian theorist to endorse this kind of contextualist stance concerning distributive justice and Simon Caney does so when he accepts that the existence of a sense of nationality of the kind Miller defends could function as establishing a relevant disanalogy between domestic and global realms as far as principles of justice are concerned.Footnote ⁵ Thus, the methodological contextualism at work in Miller's argument is compatible with both egalitarian and non-egalitarian positions; it simply specifies a set of terms on which this substantive argument about distributive principles is to be played out. The third and final of Miller's methodological principles is his adoption of the late Rawlsian notion of realistic utopianism as specifying the level or kind of theorising with which he is engaged. This does, as Miller recognises, open the theorist to attack from two directions in that their normative claims can be criticised as either too weak (bound by too limited an account of what is practically possible) or too strong (not limited enough by its account of what is practically possible). Interestingly, Miller only considers (to set aside) such responses as global responses (for example, by IR realists) and does not draw attention to the more interesting case of a global endorsement of this approach combined with local rejections of some parts (perhaps all) of Miller's normative arguments as reliant on, for example, economic and/or sociological assumptions that can't be substantiated.

Against this methodological backdrop, the structure of Miller's substantive argumentation falls into two parts: first, a core anti-egalitarian argument and, second, two elaborations of the implications of this anti-egalitarian view for global justice with respect to migration and global poverty respectively. The core argument is offered through four steps: (1) establishing the distinction between weak (equal moral concern) and strong (equal treatment in some substantive sense) forms of moral cosmopolitanism; (2) making plausible the claim that membership of a nation can be a source of special obligations; (3) offering a critique of global egalitarianism; and (4) showing that members can be liable in respect of the actions of their nations. Granting (1) for the sake of argument, I will focus on (2), (3) and (4).

Nationality and special obligations

Isn't one's nationality simply a morally arbitrary feature that should be discounted for purposes of justice? Such is the position of Joseph Carens', for instance, in his classic moral cosmopolitanism argument for open borders.Footnote ⁶ To defuse the force of this intuition, Miller offers a conceptual argument and a substantive argument. The former distinguishes two senses of ‘morally arbitrary’, the descriptive sense that some feature is a product of unchosen luck and the normative sense that a feature should not count in determining distributions. Miller argues the second does not track the first by noting disabilities are products of unchosen luck yet most people would acknowledge that they should effect distributions (pp. 32–3). The latter argues that in the same way that family relationship may give rise to special obligations that should be recognised in any account of justice, so too may relationships between co-nationals. On what grounds? Miller's argument runs thus: Nations can (at least in principle) meet three conditions that are jointly sufficient for this claim: (a) that nationality (like friendship) is an intrinsically valuable relationship that gives rise to special obligations; (b) that these special obligations are internally related to the good of the relationship; and (c) that this relationship does not entail injustice with respect to those excluded from it (what Samuel Scheffler calls ‘the distributive objection’,Footnote ⁷ although distributive injustice is hardly the only form of injustice that exclusion might involve). Of course, even if true, this argument only shows that nations can be a legitimate source of special obligations to co-nationals, not that any existing nations meet this standard and thus Miller asserts that we ‘have obligations to our compatriots to the extent that our nation meets the [specified] conditions’ (p. 37, fn. 21). This argument raises three significant issues for the plausibility of Miller's account and, given their importance, it is worth attending to each in a little detail.

The first issue involves drawing attention to the point that at various points Miller's account mobilises two distinct senses of nationality: nationality as membership of a national community and nationality as the external legal face of citizenship. Although Miller's ‘official’ position is that he is addressing nations and not states, this position seems to slip on more than one occasion. Thus, for example, in initially framing his argument Miller writes in relation to the circumstances of global justice that ‘if we consider how people relate to one another at that level, one very important mode is as citizens of independent national communities’ (p. 17) and again ‘I have drawn attention to particularly to citizenship in nation-states as a key factor that differentiates people's relationships within political communities from their relationships at [a] global level.’ (p. 17) and again: ‘The particular question we are examining is whether the circumstances of global justice should be taken to include the existence of separate states whose members belong to different national cultures, and who therefore value their capacity to be politically self-determining.’ (p. 19). While it is true that Miller's discussion of national responsibility in chapters five and six does restrict itself to a focus on the nation, the discussion of immigration in chapter 8 is more or less entirely state-focused. In general, his default position seems to be one of addressing nation-states rather than nations as such. This matters since, as Miller acknowledges,Footnote ⁸ one can construct accounts of the intrinsic value of citizenship and of special obligations to compatriots that makes no appeal whatsoever to the notion of national belonging. Both Jurgen Habermas' conception of constitutional patriotism and Andrew Mason's conception of belonging to a polity denote clear (and distinct) theoretical alternatives to Miller's favoured liberal nationalist account.Footnote ⁹ Moreover, whether or not we have reasons in justice to prefer these civic models, if they are feasible practical alternatives, we certainly have moral reasons to prefer them since they impose lesser costs and burdens on culturally diverse citizens as well as being less restrictive with respect to resident aliens seeking to become members of the polity. Miller must, then, be committed to the claim that the models proposed by Habermas and Mason are not feasible practical alternatives to the liberal nationalist model – yet, as far as I can tell, Miller offers no compelling empirical evidence for this claim either in On Nationality,Footnote ¹⁰ where the empirical evidence considered would apply equally to all three modelsFootnote ¹¹ nor in this latest work. This is not entirely Miller's fault since direct evidence is hard to come by and, in a more recent article, Miller has tried to support his position through the use of indirect evidence; however, it remains the case, given the centrality of the claim concerning national belonging to Miller's whole account, that what grounds this account – as well as the accounts of his rivals – is currently a wager on what the empirical conditions of maintaining social democratic states turn out to be.

The second topic concerns Miller's claim that we ‘have obligations to our compatriots to the extent that our nation meets the [specified] conditions' (p. 37 fn. 21). Two problems arise here. First, it is not exactly clear what this claim means. Does this mean, for example, that as the degree of injustice that arises as a by-product of our valuing national community increases or declines, then so, as a matter of inverse proportionality, do our obligations to compatriots? It is unclear why we should think this and Miller's own account of human rights addressed to generic basic needs would appear to cut against such a view. Second, there are further problems concerning the determination of the scope and extent of co-nationals’ special obligations to each other. On the one hand, a problem arises concerning scope since the class of persons enjoying the formal legal status of nationality need not coincide with the class of persons who appropriately recognise each other as having special obligations to each other as co-nationals. On the other hand, since, in contrast to the case of friendship, certain requirements of mutual recognition of special obligations to co-nationals are given coercive institutional expression through law, the fact that our co-nationals act as if they recognise the value of national community does not entail that they do so (unless we have good reason to believe that institutional recognition reliably tracks interpersonal recognition).Footnote ¹² In terms of Miller's own analogies of friendship and family, we may ask if I owe special obligations to individuals who act as if friends or family but for reasons (prudential or moral) not based in an acknowledgment of the intrinsic value of friendship or family. If the answer is ‘no’, then this finding stands in real tension with Miller's implicit strategy of treating the legal status of nationality and the performance of the duties of citizenship as proxies for these issues of scope and extent precisely because this strategy puts rather too much emphasis on the institutional rather than the interpersonal, where the latter (and not the former) is the key justificatory ground for special obligations between co-nationals. If the answer is ‘yes’, then it is not clear why we should be focusing on national identity rather than simply on citizenship. (None of which is to say that one cannot specify special obligations in terms of institutional structures that are both coercive and cooperative structures with respect to citizens, simply that Miller does not take this route.)

The third of the three conditions that Miller specifies – that the special obligations between co-nationals should not entail injustice to outsiders – leads directly to the third issue I wish to consider. Miller's key move in relation to the distributive objection to special obligations is to point out that answering the question of whether special obligations entail injustice hangs on how one specifies the general obligations that obtain between the persons or groups in question. If these general obligations are understood in egalitarian fashion, special obligations are likely to be judged unjust unless they serve as effective delivery mechanisms for general obligations specified; by contrast, if the general obligations are construed in a non-egalitarian fashion, say, as a basic needs account, then there is likely to be more scope for viewing special obligations that do not serve the delivery of general obligations as not unjust. Thus, the distributive objection cannot be used to ground commitment to global egalitarianism (or, for that matter, Miller's global non-egalitarianism), rather independent arguments concerning the nature of the general obligations we owe to each other are required. Notice though that implicit in this view – or at least Miller's elaboration of it – is the claim that the only general obligations at stake are those that we owe to each other as human beings qua human being; but this claim should, I think, be rejected since we may also reasonably view each other as individuals and groups (including nations) who are all subject to a regime of global governance comprised in significant degree by the international society of states as well as the regional and transnational institutions of governance which have emerged within and from it – and, consequently, as owing each other general and reciprocal justification of the political structures and power dynamics of inclusion and exclusion currently characteristic of this regime of global governance. Thus, to take an issue that one would think of direct relevance to Miller's nationalist concerns, we might think that such general and reciprocal justification is required for the fact that the capacities of ‘nations’ as agents in the global political arena and their vulnerabilities to other actors are significantly shaped by, for example, their recognition or non-recognition within international law: the comparison between the non-recognition of ‘First Nations’ (indigenous peoples) as persons in international law and the contrasting recognition of nations that have the political form of a sovereign state is an obvious and poignant example. In this context, I incline to Rainer Forst's view that the most basic principle of justice – at any level – is a right to justification.Footnote ¹³ The main point here is that, despite Miller's welcome emphasis on issues of agency, he pays relatively little attention to the political question of how particular forms of individual and collective agency are constituted through power relations (nor, more specifically, in how the actions of powerful agents can shape the landscape of international law).Footnote ¹⁴ As we will see, this issue returns in respect of his discussion of national responsibility.

Against global egalitarianism

Miller's argument against global egalitarianism involves specifying two problems with this position: the metric problem (global distributive egalitarians cannot specify what substantive equality of treatment would mean) and the dynamic problem (the problem posed for global equality by the existence of self-governing political communities whose decisions structure the current and future availability of resources and opportunities). I shall leave aside the first issue in order to focus on the second. This is in part because the argument offered in relation to the dynamic problem helps itself to the assumption – for the purposes of argument – that there is some metric in terms of which a principle of global equality can be couched and thus we can reasonably take Miller to be offering the dynamic problem to be the fundamental issue at stake in that it persists regardless of the validity of argument concerning the metric problem (it is also in part because Miller's consideration of metrics is limited to equality of resources and equality of opportunity which hardly exhausts the range of metrics available).

Miller imagines two pairs of nation-states each starting from an equal resource base: (a) Affluenza which decides to use up its resource base rapidly to sustain high consumption and Ecologia which chooses to conserve resources by engaging in sustainable development; and (b) Procreatia which encourages large families and population growth, and Condominium which adopts a strictly enforced family planning policy to ensure a stable population. Miller then assumes that level of advantage are determined solely by domestic policy and, other things being equal, per capita resource levels will become greater in Ecologia than in Affluenza and greater in Condominium than in Procreatia (pp. 68–90).Footnote ¹⁵ The point of these examples is to undermine the attractiveness of an egalitarian policy by showing that it is either too weak if limited to the notion of equal starting points or creates perverse incentives if recast as preserving equal access to advantage over time and unfairly penalises the citizens of Ecologia and Condominium (pp. 70–2). Miller acknowledges that a possibility ‘would be to deny nations rights to self-determination in all those areas of policy that have an impact of levels of advantage’ but insists that ‘this is tantamount to doing away with self-determination altogether’. He further claims this position cannot be saved by requiring each generation to have an equal starting point since generations are fictions in the relevant sense so allowing nations to decide autonomously for a generation and then applying an international tax-and-transfer regime would not solve the problem (p. 73).

This argument does not strike me as overly convincing since it constructs too stark a contrast between autonomous decision-making and lack of self-determination. Autonomous decision-making does not mean being able to choose anything, it means being able to choose within a structure of just rules. If it is a rule of justice that we have obligations to future generations to try to ensure that they are no worse off in terms of access to advantage than we are, it would be perfectly reasonable to have an international tax regime that applied to all nations and that taxed activities on the basis of our best current knowledge of their probable implications, other things being equal, for the access to advantage of the future generations. Such a tax regime would no more undermine the right of nations to determine their own policies than a just tax regime within a state undermines the capacity of its members to engage in autonomous decision-making; in both cases, all that has happened is that the costs of various choices are regulated by more general concerns of justice (consider how the example of a global carbon tax in conjunction with the possibility of trading carbon tax credits allows states to pursue different paths). Moreover, as in the state context, such a tax regime could apply continuously as could transfers in the form of a resource allocation at birth to each individual. There is not, I think, anything particularly difficult in principle about this scenario.

On the basis of Miller's argument, I can find two objections to such a proposal. The first is that, under the scenario he envisages, the resource-poorer children of Affluenza and Procreatia should direct their complaints towards their predecessors and don't have a claim on the citizens of Ecologia and Condominium (p. 72). But this objection is addressed by the fact that, under my proposal, it is their predecessors who are more highly taxed for their profligate ways and by the further point that, in the absence of such tax revenue, these children do have a claim against the international society of states (and, hence, against Ecologia and Condominium as members of this international community) for not ensuring that such resources are available. The second objection is Miller's rejection of the principle of ensuring that future generations have an equal resource entitlement as a principle of justice; he comments thus:

But this objection presupposes the validity of the non-egalitarian principle of justice that Miller is proposing in the course of presenting the very arguments that are meant to lead us to it. Since, on my view, a consistent global egalitarianism will hold the principle of ensuring that future generations are not worse off in terms of access to advantage than the current generation, Miller can't simply assume away this fact in order to facilitate his objections to egalitarianism.

Even if these responses by Miller are unconvincing, however, he does have two further fallback positions available to him. The first would be to argue that any such international tax-and-transfer regime would require a global state on the grounds that people would only comply with such a system given the global equivalent of a shared sense of national belonging. While this response is not at all compelling in the context of the stylised example constructed by Miller, it would be foolish to underestimate its force in real world circumstances in which nation-states of inherited privilege confront nation-states of inherited deprivation on the scale of privilege and deprivation that currently prevails in our world. The second fallback position is the argument concerning the metric problem. Miller may not wish to fallback onto this position since presumptively it does less work for his argument than would be accomplished than the substantive critique of global egalitarianism and because it is unclear that this argument would itself hold up under pressure. Arguably, Miller makes the argument plausible by demanding more fine-grained comparative judgments than may be possible, but it is not an objection to the possibility of measurement that our available instruments of measurement can only yield more coarse-grained judgments than we would ideally like; one should be a realistic utopian about measurement as well as principles.

National responsibility

Bracketing off the fact that Miller's critique of global egalitarianism does not strike me as fully persuasive, what of his final move – that of establishing that members can be liable for the responsibilities of their nation? There are three elements to Miller's argument. First, he offers an account of collective responsibility in general that shows how individual members of such collectives can be held liable for the outcomes of collective actions. Second, he tries to show that this analysis covers nations as contemporaneous groups of people. Third, he attempts to extend this argument to cover liability for the past actions of nations (pp. 113–4).

Miller offers two ideal-typical models of collective responsibility: the like-mindedness model and the cooperative practice model. In the first model, exemplified by the figure of the mob, the focus is on that fact that the collective brings about some outcome (even if the individual members do not intend to do so) that is a product of their collective behaviour where this behaviour is characterised by common-mindedness. In the second model, exemplified by the figure of the employee-controlled firm, the focus is on the responsibility of members of a cooperative practice for the damage caused by that practice, where the presumption is that the costs of repairing the damage be borne by all members (including those who dissent with respect to continuing the damaging aspects of the practice but benefit from its continuance). The distinction between the two models is, first, that the former requires the notion of a common identity, whereas the latter does not and, second, that the latter imposes fairness requirements that the former need not (pp. 114–9). What, then, do members have to do to avoid acquiring liability for the actions of the collective? Miller writes: ‘Unfortunately, it is difficult to say anything more precise than that he or she must take all reasonable steps to prevent the outcome occurring [that are compatible with not incurring greater costs than people in general could be reasonably expected to bear].’ (p. 121, my insertion – cf. p. 122). In the case of the like-minded model, this seems reasonable but in the case of the cooperative practice model, we can say more. Thus, Miller himself notes that if we change ‘our example in such a way that decisions […] are taken by a small clique who keep the rest of the workforce in the dark about the whole issue, or skew the distribution of rewards in such a way that one section of the workforce could reasonably claim to be working on exploitative terms, and collective responsibility no longer extends to all members but at most to the decision-makers or the leading beneficiaries of the practice.’ (p. 119). We might also add that, even on the unadjusted example, an individual who dissented, contributed the additional benefit that he or she received through the persistence of the damage-generating element of the practice towards repairing the damage incurred, and encouraged others to do likewise would seem to be relieved of responsibility.

Miller then proceeds to argue that both models can be applied to nations since (i) they are characterised by like-mindedness expressed in and through the shared national culture and (ii) they are engaged in a cooperative practice, not least insofar as they are concerned to reproduce their shared national culture. The key issue that I want to draw attention to here concerns the relationship between nation and the political self-determination to which it aspires; this matters on Miller's account because, roughly, the greater the degree of collective self-determination, the more applicable the claim that individual members are liable for the outcomes produced by national actions.

Miller picks out three possibilities that he takes as particularly important: a nation subject to external imperial rule that, consequently, lacks self-determination at all; a nation which possesses its own state but is subject to authoritarian rule; and a democratic nation-state. Now these are certainly real possibilities and working through them does reveal much about the conditions under which ascriptions of responsibility to the members of the nation or some subset of members can be justified; however, it seems strange not to have considered a fourth case, the nation within a multinational federal democratic state that possesses real but limited self-determination rights. Of course Miller may have good reason to avoid this case since it would seem to lend support to the view expressed in my criticism of his argument against global egalitarianism that self-determination can be real even when significantly constrained by wider principles of justice – and, in this case, on Miller's own view, principles of social justice that are egalitarian in character. Be that as it may, the interesting point about this case in the present context is that while it may make sense to hold members of nations liable for their actions, the issue of responsibility for acts that are articulated through the apparatus of the state no longer seems to be tied to the sub-state national communities. One could address this by construing the acts of the state as products of a cooperative practice between either its constituent nations or its citizens or some differentiated combination of the two – but which of these options would be pertinent would depend on the specific political structure of the state, that is, the arrangements that specify and formally constitute the agents who participate (directly or indirectly via representatives) in decision-making. The upshot of this observation is twofold. First, it points out that states can be construed as collective agents on the cooperative practice model and, more importantly, second, it highlights the point that in addressing issues of responsibility, one also needs to attend to relations of power involved in the constitution of forms of agency and the justifiability of those power relations.

This point returns once more in Miller's discussion of national responsibility for the past in which he offers a thoroughly sane and sensible discussion of the principles that should guide reflection on four kinds of issues:

What is conspicuously missing from this list, however, is the kind of issue that concerns many ‘First Nations’, namely, the historic wrong of being deprived of their status as agents who are recognised as nations in international law and can engage in activities such a treaty-making, foreign relations, etc.Footnote ¹⁶ The nature of this wrong is, we may think, more fundamental (which need not always mean more serious) than the kinds of wrong to which Miller draws attention since it pertains to the character of the agency of a people itself.

Migration and refugees

At this stage, I want to turn to address a more specific thematic focus of Miller's argument, namely, his reflections on immigration. To do so, however, we need to note that whereas our focus thus far has been on outcome responsibility (that is on responsibility for the causal outcomes of one's acts and omissions), Miller offers a needs-based account of human rights (as the anti-egalitarian view of the general obligations owed to human beings) as triggering remedial responsibility (that is responsibility considered from the standpoint of capacity to relieve the harm). Such human rights are understood as related to basic generic human needs and are intended to secure the minimal conditions of a decent life. As we'll see this account plays a central role in his reflections on migration. We should also recall a point that I made earlier here, namely, that in his discussion of migration, Miller gives up talk of nations in favour of talk of states. In one way, this is perfectly understandable since, in our current world, states have (only slightly qualified) sovereign authority over their borders. However, it is a pity since it entails that Miller's discussion leaves aside complexities of the issue of free movement within states that might be raised by multinational states and bear on the topics that he is engaging. Consequently, it may be worth beginning with some reflections on this issue.

Rather strangely Miller starts by considering the right to freedom of movement as the right to move in physical space (pp. 205–6) rather than the right to move within and across public jurisdictions which is the issue actually at stake (pointing out that we can't invade someone's private property, except under special conditions, and that our movement in relations to public spaces is regulated in various ways is rather besides the point in this context except as a rhetorical device). But setting this aside, we may reasonably raise the question of why a sub-state nation should not have the right to restrict movement of non-national citizens into its territory? Given Miller's concerns with the preservation of national cultures, the assimilation of incomers into the nation, etc., would it not seem reasonable that sub-state nations could restrict entry on the same kind of grounds that allow nations that are also states to restrict entry? It seems likely that Miller will resist this suggestion on the grounds that states (even when not nation-states?) are appropriate sites of social justice and, hence, that it would be unjust to restrict the freedom of one's fellow citizens to move within and across sub-state public jurisdictions in pursuing their interests. If this is Miller's response, then his argument for restricting entry ends up hanging entirely on his independent critique of global egalitarianism. However, if Miller accepts that in principle it is acceptable that within a multinational state with no overarching (supra-)national identity, there could be restrictions on freedom of movement that are grounded the interests of the discrete national communities, then one of the fundamental tenets of liberal political thought is undermined. Neither prospect looks especially inviting.

In practice, Miller's strategy is to argue that a right of freedom of movement across state jurisdictions is not required to provide the (basic) human rights that comprise our general obligations to one another except in circumstances where one's own state does not protect one's human rights. So let us consider this argument.

Miller's position with respect to so-called ‘voluntary’ migration is this:

We may perhaps agree with Miller that such cases don't ground an unlimited human right of freedom of movement, but it is not clear contra Miller that they can't ground a circumscribed human right to move to another state. Let us then consider the case of Fred who is engaged to his fellow-citizen Annabel but on a final short bachelor holiday in another state meets Claire and falls head over heels in love; consequently, Fred breaks off his engagement with Annabel and he and Claire decide to marry. The questions that arise for Miller's view are whether Fred or Claire have: (i) a right to marry; (ii) a right to live in the same state; and (iii) a right to determine which state to live in? I raise the first question because Miller's argument appears to be that insofar as your own state provides you with an adequate range of marriage partners, then you don't have a right to travel to other states to find other potential marriage partners. Fair enough, but given that you have found one either by tourist travel (which I assume Miller would allow) or, if you prefer, on the internet, what then? I think that we can reasonably assume that being able to live with the person that you love (and one may generalise this to family members among who on Miller's view, and mine, there are agent-specific special obligations of care) is a generic human need insofar as that person is a singular non-substitutable source of value for you and, hence, you would not be ‘frustrated’ but ‘devastated’ to be denied the right to live together – perhaps no longer seeing any compelling reason to go on with life. Moreover, there seems to me to be no compelling reason why the (let us assume) happily married couple should not be allowed to choose in which of the two states of which one is national to reside.

If one grants this case, however, it is no longer clear whether one's interest in joining a religious community not represented in one's home state should not also ground a right to move to another state, at least insofar as practising that religion requires that you participate in such a religious community in a way that rules out remote participation. (To confess, one requires a confessor.) For the religious believer, it is just a mistake to characterise her relationship to her faith as the option ‘she most prefers’ because her faith is not a preference within a range of options, it is a necessity governing the fate of her soul. In this case, the Christian believer would object that not only is Miller undermining her chance of a minimally decent life in this world but also in the next! We can extend this line of argument (albeit without the additional issue of immortality) to question whether access to the practice of opera is so easily denied to Miller's would-be opera singer since it is also a feature of human beings that they can come to love certain practices – say, opera – where, as in the case of marriage and religion, this means simply that the practice becomes a singular non-substitutable source of value for them. Thus, if the opera lover is in state A and across the border in state B, there is a widespread practice of opera. I see little reason to deny a right to cross the border for opera-training.

In relation to all of these cases, we can recall here a point that Miller admits early in this work, namely, that while membership of a nation is (under certain conditions) of intrinsic value, it is not the only thing that is of intrinsic value – friendship, family and other practices can also be sources of intrinsic value – nor is it necessarily the most important. This being so (and it is!), I find it hard to see why an account of global justice that seeks to take seriously the existence of national communities on the grounds of their intrinsic value, should propose rules of justice concerning freedom of movement that entail the de jure privileging of the value of national community over other sources of intrinsic value. Appealing to adequacy defined in terms of generic human needs is not an adequate answer here since while it may be that for George what is of intrinsic value is music and so an adequate range is fine, for Caroline it may be opera, opera, and only opera. To push this point home, note that if we don't run legal and identity senses of nationality together (as Miller does throughout this argument), we can envisage cases such as Gareth whose national identity is Welsh but whose legal nationality is Argentinean and ask: ‘should Gareth have a human right to move to Wales?’ Given how important and valuable Miller takes national identity to be, one rather suspects that he will want to say ‘yes’ to this question. But on his own account of basic human rights, the answer looks like being ‘no’. Moreover, if Miller does answer ‘yes’ to this question, then it is unclear why the admission would not presumptively apply to other sources of intrinsic value (after all, rights to family re-union migration are well-established in law and political practice – and rightly so).

Finally, let me turn to the case of refugees. We should note two initial points. First, Miller's account of human rights would entail a widening of the current 1951 UN Convention definition of refugees in line with the more expansive definitions adopted by the OAU and the Cartegena Declaration in Latin America – and cogently advanced in intellectual terms in Andrew Schacknove's classic essay ‘Who is a Refugee?’ to which Miller refers (p. 225, fn. 29). Second, Miller argues that: ‘Refugees, then, have a very strong, but not absolute, right to be admitted to place of safety, a right now widely recognized in both law and political practice.’ (p. 227). Why not an absolute right? Miller's argument runs thus:

I confess to finding this passage perplexing on two levels. First, it focuses on whether a given receiving state may legitimately refuse entry to bona fide refugees. But the right of refugees to sanctuary is a claim against the international society of states, not against a particular state, since it is the normative structure of this global governmental order (most particularly the Janus-faced grundnorm of state sovereignty/non-intervention) that grounds the necessity of a right to refuge from one's own state when it breaches the minimal legitimacy requirements of the state-citizen relation. Another way of putting this point is to say that if we adopt Miller's own standard of universal basic human rights, then this standard serves to specify the legitimation criteria for the current regime of global governance (also known as, for the most part, the international society of states) and, consequently, a failure to ensure that bona fide refugees receive the protection of a state which is not their own weakens the claim to legitimacy of this global political order. This suggests, contra Miller, that the refugee's right to sanctuary is an absolute right but that this right does not entail a right to residence in a state of their choice, rather it simply designates a right to sanctuary in some state which is not their own. This argument does, of course, entail the possibility of a mechanism for distributing refugees fairly – that is, as far as possible, both taking into account the wishes, language skills, and suchlike attributes of refugees and the different capacities of states to receive them. What may motivate Miller's confusion is that the current international refugee regime has no effective mechanism for the equitable distribution of refugees – which is why weak states in the proximity of the major refugee producing states tend to bear the brunt of the burden.

This issue bears directly on the second dimension of my perplexity in relation to this passage, namely, Miller's claim that the ‘final judgment must rest with the members of the receiving state, who may decide that they have already done their fair share of refugee resettlement’. Why on earth would anyone think this? One would have thought that to be justified, it would have to be the case not simply that members of the receiving state decide that they have done their fair share but that they actually have done their fair share – and making that judgment requires reference to some internationally endorsable standard of equitable distribution for refugees of the kind that we could reasonably hope that a realistic utopian account might aim to supply. What Miller's discussion fails to acknowledge is that the obligations of any state to refugees are not simply a matter of: (i) not acting in a way such as to produce refugees; and (ii) accepting one's fair share of refugees but also; (iii) acting with other states to ensure the operation of an international refugee regime that (a) provides effective protection to the global refugee population; and (b) distributes the burdens of refugee protection equitably across receiving states; and (iv) supporting the development of international political mechanisms that (a) encourage states to comply with the requirements of human rights; and (b) provide the resources needed for states to fulfil this role. This matters because, for example, to the extent that a state fails to engage in working for – or even obstructs – the development of such a fair scheme of refugee protection, so too its claim to legitimately refuse admission on the basis of ‘having done its fair share’ is weakened. Moreover, as experience painfully attests, without some international norms concerning the equitable distribution of the burdens of refugee protection, ‘doing one's fair share’ tends to collapse into whatever a given state perceives as doing one's fair share – which isn't satisfactory on any grounds. What is most frustrating in this dimension of Miller's account is that his own immensely helpful and subtle discussion of outcome and remedial responsibility would provide much of the intellectual apparatus needed for an adequate treatment of this issue.