David Miller, ‘Selecting
Immigrants’1
David Miller (Nuffield College, Oxford)
CSSJ Working Papers Series, SJ034
April 2015
Centre for the Study of Social Justice Department of Politics and International Relations
University of Oxford Manor Road, Oxford OX1 3UQ
United Kingdom Tel: +44 1865 278703 Fax: +44 1865 278725 http://social-justice.politics.ox.ac.uk
1 Paper prepared for the conference on ‘The Lampedusa Dilemma: Global Flows and Closed Borders. What should Europe do?’ organised by the EUI Forum on Migration, Demography and Citizenship, European University Institute, Florence, 17-18 November 2014.
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
2
In the developed liberal democracies today, the immigration issue has become intractable as a result
of three conflicting pressures. The first is the increasing number of people from developing
countries who wish to enter, whether to escape poverty or civil war or simply to improve their
material prospects. Polling by Gallup, for example, suggests that 38% of those living in Sub-Saharan
Africa and 21% of those living in the Middle East and North Africa would prefer to migrate
permanently.2 The second is the increasing reluctance of citizens within those societies to accept
large numbers of incoming migrants. In the UK, for example, an opinion poll in late 2013 found that
80% of those who were asked thought that current levels of net inward migration were too high,
85% thought that immigration was putting too much pressure on public services such as schools,
hospitals and housing, and 64% thought that over the last decade immigration had not been good
for British society as a whole.3 Broadly the same picture holds across Europe.4 The third is the
diminished capacity of governments to control immigration flows by means that are judged
acceptable by international law and opinion. Even setting aside the special case of the EU with its
principle of internal free movement, the prevailing human rights culture stays the hand of
governments who seek peremptorily either to prevent unwanted immigrants from entering or to
deport then once they have gained a foothold inside.
Under these circumstances, developing a defensible policy for selecting which immigrants to admit,
and on what terms, becomes a priority. In sketching the outlines of such a policy, I make three
assumptions, which I shall not defend here (though I have done so elsewhere5). The first is that
there is no human right to immigrate: the simple fact of being refused entry by a state does not, in
itself, violate anyone’s human rights.6 The second is that democratic states can legitimately shape
their immigration policies in the light of their overall national goals and priorities, whether these are
2 N. Esipova, J. Ray, and R. Srinivasan, The World’s Potential Migrants: Who They Are, Where They Want to Go, and Why It Matters (Gallup Inc., 2010-2011). Overall, 40% of those living in the poorest quartile of countries have expressed a wish to migrate – see P. Collier, Exodus: Immigration and Multiculturalism in the 21st Century (London: Allen Lane, 2013) p. 167. 3 See http://www.harrisinteractive.com/vault/HI_UK_News_Daily_Mail_Poll-Nov13.pdf. 4 See E. Iversflaten, ‘Threatened by Diversity: Why restrictive asylum and immigration policies appeal to western Europeans’, Journal of Elections, Public Opinion and Parties, 15 (2005), 21-45. 5 See D. Miller, ‘Immigration: the case for limits’ in A. Cohen and C. Wellman (eds.), Contemporary Debates in Applied Ethics (Oxford: Blackwell, 2005); D. Miller, ‘Is there a Human Right to Immigrate?’ in S. Fine and L. Ypi (eds.), Migration in Political Theory: The Ethics of Movement and Membership (Oxford: Oxford University Press, forthcoming). I am currently completing the manuscript of Strangers in our Midst: the political philosophy of immigration, to be published by Harvard University Press.
6 In some cases, however, it might lead to a violation of that person’s human rights: the distinction is important.
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
3
economic, cultural, environmental, humanitarian etc. An important aspect of national self-
determination is deciding who is going to form part of the ‘self’ in future. The third is that this right
of self-determination is nevertheless limited by what I call ‘the weak cosmopolitan premise’,
according to which all human beings are entitled to equal moral consideration when agents
(whether states or individual people) decide how to act towards them. This means in particular that
a prospective migrant seeking to enter must have her claim considered, and if it is denied she must
be presented with reasonable grounds for refusal. 7
Selecting between potential immigrants is justifiable, therefore, but how should it be done? There
are two dimensions that we need to be consider: the first is the inherent nature of the claim to enter
that the immigrant is making; the second is the nature of the connection (if any) that already exists
between the immigrant and the receiving state. On one dimension, we have the familiar distinction
between refugees and economic migrants, where refugees are those whose claim is based on the
threat to their human rights created by remaining in their current state of residence, and economic
migrants are all those who have an interest in moving to a new society, whether to study, to find
work, or to pursue some personal project, but who cannot cite a threat to their human rights as
grounds for admission. On the second dimension, there are those who qualify as what I call
‘particularity claimants’ and those who do not. Particularity claimants are people who assert that
one particular state owes them admission by virtue of what has happened in the past. A clear case
would be one in which a group of people have been led to believe that they had a right to immigrate
should their circumstances require it.8 Another example would be people who have performed
some service for the state, and claim now that being allowed to immigrate is the appropriate form of
recompense.9 Particularity claimants might also be refugees or economic migrants, but what
distinguishes them (and justifies the rather awkward label I am applying to them) is that their claim
7 Someone might ask why, if there is no human right to immigrate, states have to justify their refusal to those they exclude. But compare applicants for a job: no-one has a right to that job, but they are nonetheless entitled to be selected by a fair procedure, and to be given reasons for why they were not chosen. 8 For example the Ugandan Asians who held British passports but whose right to immigrate was abruptly removed by the Immigration Act of 1971. When Idi Amin came to power and threatened to expel them at short notice, the British Government recognized its obligation and allowed them to enter. The episode is described in R. Winder, Bloody Foreigners: the Story of Immigration to Britain (London: Little Brown, 2004), ch. 22. 9 Consider the case of the Nepalese Ghurkhas who, after serving in the British Army, have sought the right to reside in Britain after retiring. This right was granted to them by a High Court decision in 2008. According to the actress Joanna Lumley who spearheaded their campaign, ‘The whole campaign has been based on the belief that those who have fought and been prepared to die for our country should have the right to live in our country’ (http://www.gurkhajustice.org.uk/).
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
4
is held against one particular state, whereas refugees and economic migrants, although they have
chosen to apply in one place, might in many cases find that their rights or interests were equally well
served by being admitted elsewhere.
The distinctions I have drawn suggest two priority rules that states should follow in selecting
immigrants: 1. Refugees as a category should have priority over economic migrants; 2. Within each
category particularity claimants should have priority over others. The rationale for the first rule is
that states have an obligation to admit refugees (the nature and extent of which will be explored
shortly) whereas they have no such obligation to admit economic migrants. The rationale for the
second rule is that a state has more reason to acknowledge a claim that stems from an existing
relationship with the immigrant than one that is general in nature. This, however, does not yet
settle whether a refugee without a particularity claim should always get precedence over an
economic migrant who has one. Consider the following case: suppose the U.K. Border Agency has
(for some reason) to make a choice between two applicants for admission: a refugee from South
Sudan, who can credibly show that her life is in danger because she has been an outspoken critic of
the regime, but who has no previous connection to the UK, and a young man from Iraq who worked
as a translator for the British Army during the Gulf War, but who can no longer find work (so he is
poor but not yet in desperate straits). Who should be taken first? Well, perhaps the Sudanese, since
time is of the essence and she needs immediate help. But maybe she can claim less than the Iraqi
eventually: if the Agency has made an arrangement for refugees from Sudan to be accommodated in
neighbouring Kenya, that may offer sufficient protection for her human rights. The Iraqi man, on the
other hand, may have a desert claim that can only be redeemed if he is provided with the
opportunities that come with being allowed into Britain.
In presenting this case as a test of our moral intuitions, I have already assumed that the obligation to
accept refugees is not unlimited, and therefore that there may justifiably be selection among those
who are claiming refugee status. First, it is a responsibility that falls upon all states able to provide
the necessary refuge, and each state, therefore, is only required to discharge its fair share of that
responsibility. Ideally this would be done by entering into an international scheme for placing
refugees according to each state’s capacity to absorb them.10 In the absence of such a scheme, it is
permissible for states to enter into bilateral or multilateral agreements whereby states who receive
more asylum applications that they are obliged to accept can pass asylum-seekers on to other places
willing to take them into, provided always that their human rights will be adequately safeguarded in
10 There is a substantial literature on refugee burden-sharing schemes, and the criteria that might be used to judge each state’s quota. For a helpful review, see T. Kritzman-Amir, ‘Not in My Backyard: On the Morality of Responsibility Sharing in Refugee Law’, Brooklyn Journal of International Law, 34 (2009), Part III.
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
5
those places. Second the obligation is an obligation to provide temporary sanctuary, and it only
becomes an obligation to grant permanent residence in cases where it becomes clear that the
refugee has no realistic prospect of returning safely to her own society within a reasonable space of
time.
On these assumptions, receiving states may have to select, among those can prove their claim to
refugee status, people they will take in themselves and others who they will pass on under one of
the arrangements outlined above. So what grounds for selection are permissible? Consider four
possibilities: 1) The refugee’s need for permanent settlement; 2) The causal role played by the
receiving state in creating the situation from which the refugee is escaping; 3) The likely economic
contribution of the refugee to the receiving society; 4) The degree of cultural affinity between
refugee and host political community.
1) This seems a relevant consideration. Although the places to which refugees are transferred must
be human rights compliant, and this means that they provide all the opportunities that are needed
to live a decent human life and not just food, shelter and the other immediate necessities, under the
kind of arrangement envisaged (realistically one in which rich developed states pass on a proportion
of those who apply for asylum to less developed countries) there will inevitably be less assurance
that the same opportunities will continue to be available far into the future. This matters less if the
stay is only going to be temporary.
2) Consider next situations in which the state to which the asylum-seeker applies is at least in part
responsible for making her into a refugee. These will typically be cases in which it has intervened in
her country of origin, creating conflict between national or ethnic groups which expose her to
threats of persecution – for example the position of some Iraqi Kurds after the Iraq war. The
granting of asylum may then be viewed as a form of reparation.11 This makes the refugee into
particularity claimant, and provides grounds why she should be admitted to the state in question
rather than to some other place – her reparative claim is a claim against that state in particular, and
may not be satisfied by a promise of refuge somewhere else (this will depend on the extent of her
loss). As Souter argues, refugees’ choices about where to claim asylum gain additional significance
in these circumstances: ‘after causing or contributing to their displacement, heeding
11 See J. Souter, ‘Towards a Theory of Asylum as Reparation for Past Injustice’, Political Studies, 62 (2014), 326-42 who provides a detailed analysis of the conditions under which asylum claims of this kind are valid
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
6
refugees’ wishes is the least that responsible states can do’.12 Indeed they may be able to claim not
just temporary asylum but permanent residence on reparative grounds.
3) Many states choose which immigrants to accept by examining whether they bring special skills
that will contribute to the economy. But can this criterion also be used, legitimately, when deciding
which asylum-seekers to admit? Keeping in mind that the refugee’s claim is based on the threat to
his human rights, not on his potential contribution, it might seem arbitrary to give him any kind of
priority on this basis. Certainly it would be unacceptable if the asylum claim itself were to be
assessed more generously in the case of those who were seen as having valuable skills. But assume
that the claim is assessed strictly on the grounds of the seriousness of the threat to the asylum-
seeker’s human rights, could productive skill nevertheless count at the second stage, when deciding
whether asylum is offered in the state of first entry or somewhere else? I believe this would be
legitimate only in cases where the state is offering something more than asylum to the refugee –
when it is offering permanent resettlement to someone who does not automatically qualify for it.
States are surely permitted to do this, in the same way that they can offer resettlement to refugees
who have been granted asylum elsewhere, and when they do so it is reasonable to take account of
the refugee’s prospective contribution.13
4) Can states select in favour of their cultural kin when deciding who to admit as refugees? The
rationale for this is set out clearly by Carens, though it is not so clear whether he accepts it himself:
As an empirical matter, it is almost certainly the case that a state’s willingness to take in
refugees will depend in part on the extent to which the current population identifies with the
refugees and their plight. Moreover, other things being equal, it will be easier for the
refugees themselves to adapt to the new society and for the receiving society to include them,
the more the refugees resemble the existing population with respect to language, culture,
religion, history, and so on.14
To take a concrete example, the wars in Syria and Iraq in 2014 led to calls in some quarters for
traditionally Christian countries such as the UK to give priority to Christian refugees escaping from
12 Souter, ‘Asylum as Reparation for Past Injustice’, pp. 335-6. 13 Could those who are moved elsewhere under a burden-sharing arrangement complain about the unequal treatment they are receiving? I do not think so. The important point is that they are treated equally at the point at which their claim to asylum is assessed, and thereafter in ways that respect their human rights. That the state does more for some refugees than it is obliged to do is not an injustice to the others. 14 J. Carens, The Ethics of Immigration (New York: Oxford University Press, 2013), p. 214.
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
7
these countries. This was in justified in part on the grounds that Christian families were undergoing
particularly severe persecution, but also on the grounds that Christian states had special obligations
to people who shared their national religion. The first part of this claim is clearly relevant, but what
about the second?
Such an argument from common culture seems hard to defend, unless it can be presented as a way
of dividing responsibilities between states. In the Iraq/Syria case it was claimed that Muslim
refugees would be more likely to be offered sanctuary by neighbouring Islamic states such as Jordan.
Assuming this is true, and that states more generally are inclined to give precedence to those who
share their citizens’ cultural or religious values, then it would be justifiable for each state to take this
into account. But without such a background, and considering the nature of the obligation towards
refugees, cultural selection does not seem defensible.
I turn now to selection criteria for immigrants who make no claim to refugee status – ‘economic
migrants’ in the broad sense. Since states have discretion over whether to admit such immigrants in
the first place, it might seem that they have carte blanche as to whom they select, even if this means
choosing on grounds of race or national origin (such as in the notorious ‘White Australia’ policy of
the 1920s and 1930s). How can we show that this does not follow? It might appear to be ruled out
by the human right against discrimination. But on closer inspection this turns out to be too weak an
instrument, since there are contexts in which it seems perfectly permissible to discriminate on
grounds, for example, of gender, language or religion. It isn’t a breach of human rights if a political
party decides to draw up an all-women short list to select its candidate in a particular constituency, if
a public broadcaster chooses only among those able to read the news in Welsh, or a church confines
membership to those who belong to its own faith. . So the human right against discrimination must
be interpreted as prohibiting discrimination on grounds that are irrelevant to the right or benefit
being allocated. Those who in the past defended selecting immigrants by race or national origin
thought that they could justify using these criteria by appeal to the ‘character’ or ‘moral health’ of
their societies. To defeat these arguments requires showing that such claims are either false, or
irrelevant, for substantive reasons.
An initially more promising avenue is to argue that selecting immigrants on grounds such as race or
religion is an injustice to some existing citizens, namely those who belong to the group or groups
that the immigration policy disfavours.15 By discriminating in this way, the state appears to be
15 It is followed in J. Carens, ‘Who Should Get In? The Ethics of Immigration Admissions’, Ethics and International Affairs, 17 (2003), 95-110, and at greater length in M. Blake ‘Discretionary Immigration’,
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
8
labelling these people as second-class citizens. As Michael Blake has put the point, ‘the state making
a statement of racial preference in immigration necessarily makes a statement of racial preference
domestically as well’.16 This will often provide states with strong reasons not to pursue
discriminatory admissions policies, but a limitation of this approach is that it would not apply to a
state that was already religiously or ethnically homogeneous and whose members wished it to
remain so.17 Notice also that the argument hinges upon the injustice that is done to existing citizens
whose status is lowered by the discriminatory policy, not on any wrong that is done specifically to
the excluded candidates for admission, and this seems to put the emphasis in the wrong place.
Although an economic migrant cannot assert a right to be admitted, she does typically have a strong
claim, based on how her interests will be advanced by moving – for example through working in a
different kind of job, or for a much higher wage, than she could hope to obtain at home. According
to the weak cosmopolitan premise stipulated above, to turn down such a claim without giving
relevant reasons for the refusal is to show disrespect for the person making it. It is to treat her as
though she were of no moral significance. This extends also to the selection of immigrants from the
pool of applicants. It is not sufficient merely to put forward the general reasons in favour of
immigration controls. If John is going to be granted entry while Jaime is turned away, the latter
must be offered relevant reasons for his unequal treatment.
This appeal to weak cosmopolitanism explains why the state is not entitled to use merely arbitrary
methods in choosing which immigrants to admit, but it does not yet settle which reasons should
count in making the selection, and so far, therefore, does not explain what is wrong about using
race, ethnicity, and other such criteria. One way to narrow down the list is to say that the reasons
must be ones that the immigrants themselves can accept. It is obvious enough that no immigrant
will regard her own skin colour as legitimate grounds for exclusion. But a problem then arises in
cases where the receiving state and the prospective immigrant hold different views about what
should count as relevant. Suppose, for example, that a state decides to admit only high-skilled
immigrants on the grounds that it has a greater economic need for these than for low-skilled
workers. An immigrant without the relevant skills might reject this on the grounds that he (and
others like him) deserves a chance to improve his condition. So it is asking too much to say that the
Philosophical Topics 30 (2002), 273-89 and M. Blake ‘Immigration’ in R. Frey and C. Wellman (eds.) A Companion to Applied Ethics (Oxford: Blackwell, 2003). I also used the argument in an earlier discussion: D. Miller, National Responsibility and Global Justice (Oxford: Oxford University Press, 2007), ch. 8. 16 M. Blake ‘Discretionary Immigration’, p. 284. 17 This is conceded by Blake in Blake, ‘Discretionary Immigration’, p. 285. See also M. Walzer, Spheres of Justice (Oxford: Martin Robertson, 1983, pp. 35-51 and the discussion in Blake, ‘Immigration’.
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
9
reasons the state gives must also be ones that the immigrants can accept (if ‘can accept’ means ‘will
in fact accept once these reasons are explained’). Instead the relevant condition is that the reasons
the state gives for its selective admissions policy must be good reasons, reasons that the immigrants
ought to accept given the general aims of the policy.18
A more difficult question is whether it can be justifiable to select in favour of those who already have
the political or cultural attributes that will enable them to fit more easily into the society they are
joining. Consider political attributes first: can liberal democracies choose immigrants who have
already demonstrated their democratic credentials as opposed to those who espouse other political
values, assuming that this can be reliably established? Most commentators, including strong liberals
such as Carens, agree that states may exclude people who pose a threat to national security by
virtue of the beliefs that they hold, such as those liable to engage in terrorist acts.19 But in such
cases it is the disposition to act, rather than the beliefs themselves, that forms the reason for
exclusion. What about those whose political beliefs are such that they do not acknowledge the
authority of the state they wish to join, even though they have no intention of sabotaging it by
violent or other means? All states, not least liberal states, depend on their members complying
voluntarily with their laws most of the time, and presumably a belief in the state’s legitimacy is one
of the main sources of compliance. Someone who lacks that belief may keep the law for other
reasons (prudence, respect for the rights of others) but is likely to be less reliable in carrying out her
duties as a citizen. So there is some reason for favouring committed democrats when choosing
immigrants. On the other hand, liberal democracies do not require all of their existing citizens to
sign up personally to their founding principles: they are prepared to tolerate anarchists, fascists and
others, leaving them free to express their beliefs and to attempt to persuade others of their
correctness within the limits of the law. On balance, then, selection on political grounds would be
justifiable only in cases where illiberal or undemocratic immigrants were applying in sufficient
numbers that their presence would create social conflicts or disrupt the working of democratic
institutions.20
The argument that can be made for cultural selection raises different questions. We are
contemplating here immigrant groups whose cultural commitments are different from those of the
18 A rather similar position is taken in Blake, ‘Immigration and Political Equality’, where it is formulated in the language of ‘reasons that immigrants could not reasonably reject’ (p. 971). 19 Carens, The Ethics of Immigration, ch. 9. 20 As Carens puts it ‘the problem is not with any single immigrant’s views, but with the collective effect of ideas hostile to democracy’. (Carens, The Ethics of Immigration, p. 176).
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
10
majority of existing citizens – though we should also distinguish between cases where the existing
state is already multicultural and has enacted multicultural policies (e.g. Canada), and cases in which
it is more culturally homogeneous (e.g. Japan): the issue becomes more pressing in the latter
circumstances. Immigrants who speak a different language, practise a different religion, or have a
different lifestyle from the majority may pose two kinds of problem. The first is simply the cost of
incorporating them into the host society on terms of equality. There will, for example, be the cost
of translating public documents into a new language or of providing translators in courts and social
service agencies; or if religion is the source of the division, the cost of accommodating religious
practices where these impose different requirements on believers outside of the mainstream. Some
of these costs can be passed to the immigrants themselves, but others will be borne by the state,
and indirectly, therefore, by citizens at large.
There are of course likely to be compensating benefits that come with increasing cultural diversity.
The point is simply that if we look at (economic) immigration as a practice that is governed by the
logic of mutual advantage, both costs and benefits need to be factored in when considering
selection policy. Some of the costs may only be apparent with hindsight, as it becomes clear what
the equal treatment of minority cultures actually requires. This also applies to the second potential
problem. Culture is not only a matter of belief or of practice, but also of identity. This raises a
concern about the way in which culture can come to constitute a line of fracture within a political
community, possibly leading to the formation of ‘parallel societies’, whose members have very little
contact with those beyond their own community; and also a concern about the effects of cultural
diversity on social trust, and through that on people’s willingness to support welfare states and
other instruments of social justice. These are by no means inevitable consequences of admitting
immigrants with cultural backgrounds different from those of the majority, but they are possible
consequences, and avoiding them may again prove to be somewhat costly, this time in the form of
support for active integration programmes. This is the point at which the state’s existing cultural
character becomes important: a state that is already well-equipped with multicultural policies can
more easily tackle these problems than one that is not. There is, however, no independent
requirement that a state should embrace multiculturalism prior to deciding upon its admission
policy. Democracies are entitled to decide how far they wish to protect their inherited national
cultures, and how far to encourage cultural diversity within their borders.
To sum up, selective immigration requires that states give reasons for the policies they apply, and
these reasons must relate to the legitimate purposes of the state itself, as manifested in its other
policy decisions. Selection on economic grounds is the least controversial example, but other forms
CSSJ Working Paper SJ034 April 2015 __________________________________________________________________________________
11
of positive discrimination cannot be ruled out: if a society wants to enhance its sporting reputation,
for example, I cannot see why it should not seek to attract immigrants who will later qualify for the
national teams. Giving reasons of this kind shows sufficient respect for those who are refused entry,
disappointed though they may be. Recall that the later part of the discussion relates only to
economic migrants. Where refugees are concerned, there is much less scope for selecting on
grounds other than the refugee’s own need for sanctuary and the opportunity for a decent life.