Keeping out extremists: Refugees, would‐be immigrants, and ideological exclusion

: Many people want to live in liberal democracies because they are liberal and democratic. Yet it would be mistaken, indeed naïve, to assume that this applies to all would-be residents. Just as some inhabitants of liberal democracies oppose one or more fundamental liberal-democratic values and principles (e.g. the rule of law, freedoms of conscience and speech, rights to private property and to political participation), so there are foreign would-be residents who do so, who might include individuals with e.g. Jihadist, Neo-Nazi, and radical anarchist views. Proceeding on the assumption that there exists no unconditional moral right to immigrate, this article asks whether it is ever morally permissible for liberal democracies to deny residence to nonnationals based on evidence that they personally hold extremist views. I argue that this is sometimes the case. Specifically, my contention is that even if we adopt a cosmopolitan perspective on which states are not allowed to prioritise the interests of their own citizens and residents over those of foreign nonresidents, there are two conditions under which such exclusions are justified even when refugees are being refused admission (although, as I suggest, states might be morally required in such cases to admit other refugees instead). ABSTRACT Many people want to live in liberal democracies because they are liberal and democratic. Yet it would be mistaken, indeed na € ı ve, to assume that this applies to all would-be residents. Just as some inhabitants of liberal democracies oppose one or more fundamental liberal-democratic values and principles (e.g. the rule of law, freedoms of conscience and speech, rights to private property and to political participation), so there are foreign would-be residents who do so, who might include individuals with e.g. Jihadist, Neo-Nazi, and radical anarchist views. Proceeding on the assumption that there exists no unconditional moral right to immigrate, this article asks whether it is ever morally permissible for liberal democracies to deny residence to nonnationals based on evidence that they personally hold extremist views. I argue that this is sometimes the case. Speci ﬁ cally, my contention is that even if we adopt a cosmopolitan perspective on which states are not allowed to prioritise the interests of their own citizens and residents over those of foreign nonresidents, there are two conditions under which such exclusions are justi ﬁ ed even when refugees are being refused admission (although, as I suggest, states might be morally required in such cases to admit other refugees instead).


Introduction
Many people want to live in liberal democracies because they are liberal and democratic. Those fleeing persecution based on their religion, ethnicity, sexual orientation, and/or political views are often drawn to these societies precisely because they offer extensive sets of (equal) rights and liberties. 1 Yet it would be mistaken, indeed na€ ıve, to assume that this applies to all would-be residents. Just as some inhabitants of liberal democracies oppose one or more fundamental liberal-democratic values and principles (e.g. the rule of law, civic equality, freedoms of conscience, speech and association, the right to private property, the right vote and run for political office), so there are would-be residents who do so. Examples may be found among refugees and regular would-be immigrants with e.g. Jihadist, Neo-Nazi, and radical anarchist views. For the sake of brevity, I shall refer to such individuals as 'would-be residents with extremist views' or simply as 'extremists'. (Exactly where the boundary between extremist views and nonextremist views ought to be drawn is a difficult issue that I cannot delve into here; however, we do not need to settle it to be able to identify views that are unequivocally extremist on which I focus within this article, such as ones that existing governments ought to be overthrown in order to establish a caliphate, fascist state, or anarchist society and ones that specific groups within society ought to be persecuted, such as Jews, Muslims, and/or homosexuals.) My aim in what follows is to consider whether it is ever morally justifiable for liberal-democratic states to deny both short-term and long-term residence to nonnationals based on their suspected extremist views and, if so, when. I say 'suspected extremist views' as opposed to 'extremist views' simpliciter in order to highlight that as the state and its officials cannot directly look inside people's heads, they can never have a 100% certainty as to what people's political convictions are. Even when they have strong, and possibly decisive, reasons for believing that someone holds extremist views, there always remains the possibility that the person in question has recently undergone a political conversion or that he or she has been feigning to have extremist sympathies for some reason.
In asking whether it is ever morally justifiable to refuse residence to nonnationals based on their suspected extremist views, my focus will be on foreign would-be residents who satisfy the following criteria: -They are not currently present on the territory of the would-be host society.
When, if ever, denying residence to those already present is morally permissible raises thorny questions about the morality of deportations that would take us too far afield. -They are not known to have any plans to engage in seditious or terrorist activities within the would-be host society as states might have discovered through tapped phone calls, intercepted emails, witness testimonies, and so on. When people are known to have such plans, it is more plausible to say that they are being refused residence on the basis of these plans rather than on the basis of their extremist views as such, and it is the latter type of exclusion that I am interested in within this article. -They are personally suspected of holding extremist views. As such, I will not focus on what has become known as President Trump's 'Muslim-ban', which barred citizens from eight different countriesof which six have Muslim-majoritiesfrom entering the United States in 2017. Even if this ban was a genuine attempt to prevent extremists from entering the country, it did not involve any assessments of people's personal political beliefs. 2 Attempts to exclude would-be residents with extremist views can be found among various liberal democracies. In the United Kingdom, for instance, one might be denied a residence permit for having expressed views during one's life that 'foment, justify or glorify terrorist violence in furtherance of particular beliefs; seek to provoke others to terrorist acts; or foster hatred which might lead to inter-community violence in the UK'. 3 Similarly, under the REAL ID act of 2005, the United States has rendered inadmissible any alien who has espoused terrorist activity or encouraged others to do so, 4 which fits into a long history of ideological exclusion (in 1903, the US congress passed the Alien Immigration Act which sought to exclude anarchists, whilst in 1952, it enacted the McCarran-Walter Act which was meant to prevent communists from entering country). The Dutch refugee policy is another case in point. According to some reports, 5 The Netherlands rejects circa 20% of refugees who are recommended for placement by the United Nation's refugee organisation (UNCHR) under the 2016 EU-Turkey refugee agreement because of 'extremist sympathies or overly conservative views'. The fact that it is not uncommon for liberal democracies to deny residence based on people's suspected extremist views renders it remarkable that normative theorists have paid scant attention to this practice. Whilst there is an extensive literature on how states should treat extremists who are already residing within the country, including on whether their extremist views ought to be tolerated 6 and on whether these views ought to disqualify any foreign nationals among them from becoming citizens, 7 the question of when, if ever, it is morally justifiable to refuse both permanent and temporary residence to nonnationals based on their suspected extremist views remains to be explored in detail. 8 In helping to fill this lacuna, I proceed on the assumption that there exists no unconditional moral right to immigrate. 9 Whereas a defence of this assumption falls outside of this article's remit, it should be noted that it is widely accepted among contemporary normative theorists and has been defended in depth by authors such as David Miller, 10 Christopher Wellman, 11 and Michael Blake. 12 At the same time, I assume that states lack an unconditional moral right to exclude would-be residents. Such a right would fail to respect the latter's moral status given that many members of this group have strong interests in settling within (specific) foreign countries, whether temporarily or permanently. Other things being equal, these interests are strongest for refugees; since refugees' own states are unwilling and/or unable to protect their human rights, refusing (temporary) residence permits to these individuals will often mean that their most fundamental needse.g. to shelter, physical safety, adequate nutrition, and basic medical carewill go unmet unless some other country is willing to grant them residence in which these needs are likely to be fulfilled. However, being denied residence might impose substantial costs on regular would-be immigrants as well, including on those who have significantly better employment opportunities within the would-be host society and on those who seek to be reunited with relatives and/or friends who are living there already.
In the remainder of this article, I will argue that it is sometimes morally permissible to deny residence to nonnationals based on evidence that they personally hold extremist views. Specifically, my contention will be that even if we accept a cosmopolitan moral theory on which states are not allowed to prioritise the interests of their own citizens and residents over those of foreign nonresidents, there are at least two conditions under which such ideological exclusions are justified even when it is refugees who are being excluded. (That said, I will suggest that excluding refugees might impose a moral duty upon states to admit other refugees who are not suspected of entertaining extremist beliefs.) The reason for adopting such a theory is that when nonnationals are rightfully denied residence from a cosmopolitan perspective, they will normally be rightfully denied residence on (more) statist or nationalist theories as well. To see this, it ought to be observed that such theories assign less weight to the interests of foreign would-be residentshenceforth 'would-be residents'-compared to the interests of their own citizens and residents, as well as less weight to the interests that other members of the would-be sending society might have in the would-be residents being admitted (which might be financial when they are likely to receive remittances from them, but also political when the would-be residents are likely to hinder the creation or maintenance of liberal-democratic institutions within the would-be sending society if they remain, or simply likely to cause greater threats to these institutions than if they were to move abroad). Adopting a cosmopolitan perspective thus allows us to identify conditions under which denying admission to would-be residents based on their suspected extremist views is always likely to be morally permissible, which is what I am interested in for the purposes of this article.
Before looking at the two conditions, a note on how states might gather information about the political convictions of would-be residents is in order. Without attempting to provide an exhaustive list, one important way involves consular officers interviewing would-be residents under oath. 13 Another consists of vetting these individuals for past expressions of extremist views, as well as for convictions for extremist acts (e.g. terrorist activities, the propagation of hate speech) and for participation within extremist movements and political parties. At the same time, states should look for signs that those seeking residence might positively endorse liberal-democratic values and principles. One may think of membership of proliberal and prodemocratic groups, as well as of recent endorsements of civil and political rights and of rejections of extremist ideologies (e.g. the McCarran-Walter act of 1952 contained a provision that allowed for the admission of those who had openly opposed their previously held extremist ideology for at least two years). 14

Threats to Liberal-Democratic Orders
One condition under which even from a cosmopolitan perspective, nonnationals will often be rightfully denied residence based on their suspected extremist views is when their admission poses a substantial threat to the survival of the would-be host society's liberal-democratic order. This condition is predicated on an assumption that the survival of liberal democracies depends on whether enough members of society respect the values and principles undergirding them, such as the rule of law, freedoms of speech, conscience and association, and rights to political participation. 15 The best-known example of a country in which this requirement was not satisfied is, of course, Weimar Germany, which transformed into a fascist state during the late 1920s and early 1930s.
It is true that the Weimar Republic did not collapse because of the admission of foreign extremiststhough Hitler was Austrian, he had already lived in Germany before the interwar period and fought for this country during the First World War. In fact, there are few, if any, real-world examples of countries where an influx of foreign extremists caused the existing liberal-democratic order to break down. Nonetheless, we can certainly conceive of societies where such an influx would make a significant and potentially decisive contribution to a liberal democracy's demise. One might think of deeply divided societies 16 with large extremist populations where only slightly tilting the ratio between extremists and moderates in favour of the former would be fatal for the existing liberal-democratic institutions, especially when these institutions are relatively new and therefore not as resilient as they might be elsewhere. 17 Other examples may be found among societies where an extremist group already has enough members for seizing political power but where it has so-far failed to do so because of infighting.
Here the admission of foreigners with the same ideology might prove fatal for the host society's liberal-democratic institutions to the extent that the foreigners manage to unite the domestic extremists, something which they may be able to do because their outsider status gives them legitimacy in the eyes of the warrying sides. Still other examples may be found among societies who wish to offer permanent residence to a large number of refugees or foreign workers. To the extent that the refugees or foreign workers come from a region with a high concentration of extremists, then insofar as their numbers are large relative to those of the host populationsas they will quickly be in countries with small populations such as Liechtenstein, Luxemburg, Iceland, and Malta, which all have fewer than 700,000 inhabitants (and Liechtenstein fewer than 40,000) -failing to select on the basis of ideology might pose (long-term) threat to the host society's liberal-democratic institutions. None of these scenarios seem farfetched, and so there is reason for taking seriously the possibility that, at some point in the future, there will be liberal democracies for which the admission of even small numbers of would-be residents with extremist views poses an existential threat.
In saying that admitting would-be residents with extremist views can imperil the survival of liberal democracies under certain conditions, I do not mean to suggest that these individuals will invariably act upon their extremist views once admitted. Some will have other priorities, such as providing for their families. Others will come to endorse liberal-democratic values and principles as a result of living within the wouldbe host society. Whilst all of this is correct, the fact that those who engage in extremist behaviours are usually motivated by extremist ideologies means that individuals with extremist views remain more likely to undermine liberal-democratic institutions than those who lack such views, all other things being equal. Accordingly, when states are able to assess the political beliefs of would-be residents with reasonable accuracy, excluding those with suspected extremist views is likely to help to protect those institutions alongside various other protective measures that they might take (e.g. censoring extremist speech, banning extremist social movements and political parties, investing in civic education and civic integration programs). 18 What I want to suggest here is that even if one accepts a version of cosmopolitanism that does not allow states to prioritise the interests of their own citizens and residents over those of foreign nonresidents, there are circumstances under which it is morally permissible for states to exclude would-be residents based on their suspected extremist views in order to protect their liberal-democratic institutions. To vindicate this claim, it should be observed that, compared to other types of regimes, liberal democracies have the best track record when it comes to respecting, protecting, and securing the fundamental rights of their citizens and residents, 19 and, in so doing, looking after the latter's vital interests understood as 'those conditions that are necessary for the realisation of a minimally decent human life'. 20 Beyond that, they are the only regimes capable of offering their citizens opportunities to live lives that are not just minimally decent but also free and equal, by which I mean that citizens enjoy extensive liberties in (broadly) equal sets, including freedoms of speech, religion, and association, as well as rights to private property and to political participation. Even when these liberties are always imperfectly realised in practice, liberal democracies still manage to realise them to a greater degree than other politico-legal systems. 21 A third reason why preserving liberal democracies is of great importance is that liberal-democratic regimes are relatively successful in negotiating and complying with international treaties that help to achieve justice among states. 22 Though a lot more could be said about this topic, these cursory observations suggest that the collapse of a liberal democracy comes at a huge human cost, one that makes it reasonable to think that when admitting would-be residents with suspected extremist views imperils the liberal-democratic order of the would-be host society, refusing admission will often be justified not only from a statist or nationalist perspective, but also from a cosmopolitan one. This is true, I believe, even when the excluded would-be residents are refugees as opposed to regulare.g. economicimmigrants. What makes the ideological exclusion of this group justifiable in some cases is that, whilst the human rights of refugees are not adequately protected by their own states, for the host society's liberal-democratic institutions to collapse will ordinarily have the implication that the human rights of a much greater number of individuals are left without adequate protection compared to a situation where the refugees with suspected extremist views are refused residence but the liberal-democratic structures of the would-be host society survive. (Notice that that these individuals need not just include existing citizens and residents, but they might also include the newly admitted refugees with suspected extremist views themselves, as well as future citizens and residents of the would-be host society and citizens and residents of other societies who might have received (development) aid from the old liberal-democratic regime and who in some cases might even be attacked by any new authoritarian regime that is established in the same way that Nazi Germany attacked Poland and various other countries after the demise of the Weimar Republic.) To see why the breakdown of a liberal-democratic order will often be worse from a human rights perspective, it should be noted that apart from the violence and bloodshed that may and often does accompany the breakdown of politico-legal institutions, any authoritarian order that emerges from the ashes is likely to be less capable and/or willing to secure people's most fundamental rights (e.g. to bodily integrity, security, adequate nutrition) than the old liberal-democratic regime based on the track records of these different regime types. Furthermore, there are at least some civil and political liberties that nonliberal-democratic regimes by definition do not (de facto) recognise, such as rights to political participation, freedom of speech, and freedom of religion.
Of course, the human cost of a liberal democracy's demise will need to be discounted by the probability that admitting would-be residents with suspected extremist views will (eventually) prove fatal for its politico-legal institutions. When the risk posed by admitting these individuals is small, denying residence to themor simply to any refugees among thembased on their suspected extremist views might not be morally permissible, at least not on grounds that doing so helps to protect the liberal-democratic order. (As I will argue within Section 4, there might still be another justification for excluding them in such cases.) Yet as the risk becomes larger, there comes a point where even from a cosmopolitan perspective, the interests of would-be residents in being admitted are overridden, including those of refugees. Such cases may be likened to one where a group of shipwrecked individuals managed to embark a lifeboat that is close to capsizing under their joint weight; insofar as allowing one more individual to enter the lifeboat comes with a high risk that it will sink, it would seem justifiable, ceteris paribus, for those on board to refuse to help a drowning person get into the boat even when her life should be assigned as much weight as the lives of each of them individually. Similarly, when a liberal-democratic order threatens to go under when would-be residents with suspected extremist views are admitted, as it does in the three scenarios mentioned above (i.e. the one with the critical extremists-moderates ratio, the one where foreign extremists threaten to unite an internally divided group of extremists, and the one where large groups of refugees or foreign workers are to be admitted from a region with a high concentration of extremists), it looks like ideological exclusions will often be a proportionate response even when it is refugees who are being refused admission. (Without trying to put a specific percentage on how large this threat must be, which might be impossible to do, my sense is that a 20% or 30% risk increase would certainly be enough.) Before moving on, two comments are in order. One was already alluded to in the introduction, namely that when states refuse admission to refugees based on the latter's suspected extremist views, then especially from a cosmopolitan perspective, they will often have a moral responsibility to admit other refugees to replace them who are not personally suspected of entertaining extremist beliefs. This will be the case when they are morally and possibly legally required to admit a fixed number of refugees.I n fact, states whose liberal-democratic institutions are embattled might even have moral duties to admit a more specific group of refugees in order to bolster these institutions, namely refugees who positively endorse liberal-democratic values.
The other comment is to do with the fact that, since I am adopting a cosmopolitan perspective within this article for argumentative purposes, I am committed to giving the same amount of (per capita) weight to the interests of members of the would-be host society as I give to the interests of members of the would-be sending society. This has a noteworthy implication, namely that even when admitting certain would-be residents with suspected extremist views would endanger the liberal-democratic institutions of the would-be host society, refusing admission to these individuals might still be morally impermissible when doing so is likely to impose even greater costs upon the would-be sending society, perhaps because the individuals involved are likely to cause even greater damage to its liberal-democratic institutions or because they are likely to significantly hinder the establishment of such institutions.
Though this is correct, my focus in this section and the next is on cases where the expected costs of admitting would-be residents with suspected extremist views are greater for the would-be host society than the costs of refusing admission are for the would-be sending society. The reason is that because the demise of a liberal-democratic order tends to be so costly (see my earlier comments in this section), cases where (a) the would-be host society's liberal democracy would be jeopardised by the admission of suspected foreign extremists yet where (b) the expected costs of refusing admission to these individuals remain higher for the would-be sending society than the expected costs of granting admission are for the would-be host society are significantly less likely to arise than ones where condition (a) obtains but (b) does not.

Objections
The not-the-least-restrictive-means objection There are several objections that might be raised against my qualified defence of ideological residence refusals so far. The first states that protecting liberal-democratic institutions from threats posed by would-be residents with extremist views does not require that these individuals be denied residence. 23 On one version of this objection, it suffices that they be denied citizenship as this would bar them from participating within national elections. Since this approach allows them to settle within the wouldbe host society, it may be said to be preferable on grounds of being less restrictive.
I believe that this proposal is plagued by two important objections. One maintains that a democratic deficit is created when people who have been resident within a country for a long time are refused the opportunity to naturalise, as it was just mentioned that only citizens enjoy full political rights within most countries. 24 These individuals would be denied a say over the laws and institutions under which they have lived for an extensive period and that will have usually had a large impact on their lives. However, since some critics might respond that the fact that extremists hold antidemocratic views means that they cannot reasonably complain about being excluded from democratic decision-making, and since a discussion of this topic would take us too far afield (for rejoinders, one might consider the work of Alexander Kirshner who argues that because even extremists will typically have nonextremist views on a range of public issues that affect their lives, they have legitimate claims to political participation), 25 I do not want to put too much weight on the current objection here.
The objection that I wish to focus on instead maintains that the degree to which refusing citizenship to extremists can help to protect liberal democracies is highly limited and often insufficient. To see this, it should be noted that, whilst most countries only allow citizens to vote and run for office within general elections (New Zealand is an exception), there are various other ways in which would-be residents with extremist views can exert political influence that do not require them to be enfranchised on either a national or local level, or even to be granted a permanent residence permit as opposed to a temporary one. For example, they might do so though lobbying, spying, bribing, issuing threats, using violence, organising marches and protests, distributing pamphlets and books, and using their knowledge and skills to support extremist parties and movements within the host society. Even when some of these activities can be undertaken from abroad as well, such as bribing and supporting extremist parties within the would-be host society, engaging them will in many cases be easier for individuals when they are residing within the would-be host society. 26 More indirectly, residents with extremist views might pose a threat to the liberal-democratic order by transmitting their extremist values to any children that they might have, who will usually also be residents of the society and in some cases citizens.
In response, it might be said that the kinds of subversive activities just mentioned can be prevented by closely monitoring those with suspected extremist views and by interfering with them before they have the chance to engage in actions that will harm the liberal-democratic order. Whilst being monitored by the state imposes significant costs upon people, these costs will often be smaller than the ones that immigrants would have incurred had they been refused the right to settle within the society, especially as far as refugees are concerned. 27 I think that the force of this version of the not-the-least-restrictive-means objection is limited as well. Apart from the fact that the costs of such monitoring and preventive interference to liberal-democratic values might be excessive, particularly when it comes to the value of privacy, there will in many cases not be enough resources for states to closely monitor all immigrants with suspected extremist views, let alone all inhabitants with suspected extremist views simpliciter. Furthermore, even when people are being monitored, this does not guarantee that the state will find out about their planned extremist activities; for example, no less than 61% of Jihadist terrorist attacks that were committed in France between 2012 and 2019 where committed by individuals who were on a French terror watch-list. 28 Still another problem is that to the extent that states do discover plans for extremist activities, there will not always be enough time left to intervene successfully, which in any case can go wrong irrespective of how much time remains.
What follows from this, I believe, is that although monitoring extremists and trying to thwart their subversive activities might help to protect liberal democracies, it will not always take away the need to try to prevent foreign extremists from settling within the country.

The objection from negligible individual contributions
At this point, a critic may object that the admission of any single would-be resident with extremist views is bound to have a negligible impact upon the survival of a liberaldemocratic order even when said individual is disposed to act upon these views. 29 From this, our critic may infer that refusing residence to people is never justified when this is done in order to protect liberal democracies, as to do so fails to respect their individuality by making them responsible for the behaviour of a group over which they lack meaningful control.
One problem with this argument is that it understates the political influence that single individuals sometimes have, particularly those who are rich, charismatic, and/or politically well-connected. Just think of Hitler's role in the demise of the Weimer Republic, or, more recently, of the authoritarian backsliding of Hungary, Russia, and Turkey under the leaderships of Orban, Putin, and Erdogan respectively. Especially when foreign extremists come from authoritarian countries, moreover, there is a risk that they will bring knowledge and skills in creating and sustaining oppressive regimes that are not or only scarcely available within the host society.
Another problem lies in the assumption that people's individuality is not respected when states legally restrict behaviours that become harmful only when two or more persons engage in them. Compare laws that prohibit people from driving old diesel cars within city centres. Whereas any single individual driving such a polluting car has a negligible impact on the city's air quality, the fact that the pollution produced by hundreds if not thousands of people driving such cars does have a significant impact explains why we do not ordinarily think that such restrictions fail to respect people as individuals. But if the fact that a harm is jointly produced does not render it morally impermissible to restrict individual behaviours in such cases, then it is unclear why things would be different when it comes to any jointly significant but individually negligible threats that would-be residents with suspected extremist views might pose to the survival of liberal democracies.

The generalisation objection
Still another objection against denying residence to people based on their suspected extremist views maintains that even if states can reliably ascertain whether someone entertains extremist beliefs and even when such beliefs are correlated with extremist behaviours, excluding them based on the mere probability that they will act upon these beliefs is morally problematic. On this view, respecting people's individuality precludes states from engaging in such statistical discrimination.
The first thing to note here is that there are certainly cases where statistical discrimination looks morally problematic. For example, even if we assume that someone who is Afro-American is statistically more likely to engage in criminal activities than members of other ethnic groups, the practice of many US police officers to subject Black people to racial profiling would remain objectionable. 30 Suppose that this is correct. Even then, it does not follow that refusing people residence based on their suspected extremist views must be objectionable as well, as a comparison of these forms of statistical discrimination suggests that there are morally relevant differences.
The difference on which I want to focus here lies in the fact that, although those with suspected extremist views might never engage in extremist behaviours once admitted, for them to be suspected of having extremist views means that they must have already acted in ways that are morally problematic. This is because unless they have openly endorsed or spread such views or engaged in other types of anti-liberaldemocratic behaviour, there is no reason for assuming that they personally hold extremist views (see my comments about this in the introduction). By contrast, even if there is a statistically significant correlation between being Afro-American and having a proclivity to engage in criminal behaviour, and even if one believes that Afro-Americans who have committed criminal acts are (partially) to blame for being racially profiled by police officers (which is false in my view, but a defence of this claim is beyond this article's scope), most Afro-Americans would still lack moral responsibility for the fact that they are subjected to such profiling as they have never committed any crimes.
In short, because those who are personally suspected of harbouring extremist views will bear both causal and moral responsibility for the existence of such suspicions, it is far from clear whether their use as a proxy for future extremist behaviour is objectionable. I should hasten to add that even when there are some moral problems with it, and even when we add these costs to the social, psychological, and financial costs that being refused residence might impose on would-be residents as well as their friends and relatives, it remains unlikely that when admitting would-be residents with suspected extremist views imperils the liberal-democratic order of the would-be host society, these costs will trump the enormous costs that come with the break-down of such orders. To see this, it should be recalled that the break-down of a liberal democracy will normally deprive thousands if not millions of people of the enjoyment of wide range of human rights. 31 The deportation objection A fourth objection maintains that insofar as there are cases where it is morally justified to deny residence to foreign would-be residents based on their suspected extremist views, then it must also sometimes be justified to deport long-term residents based on their suspected extremist views. Since the latter is unjustifiable even when the longterm residents lack citizenship, the objection continues, the former must be unjustifiable as well. 32 I believe that this analogy is spurious. One thing it overlooks is that states will often bear a greater degree of moral responsibility for the fact that their long-term residents hold extremist views than the countries to which these individuals are deported or than indeed any other country, which has the effect that such deportations are likely to create unfairness between states. As I argue in more detail elsewhere, 33 this is so especially when people have spent a greater part of their lives within their country of residence than anywhere elsewhere; when they have grown up there; and when their radicalisation occurred predominantly within this society.
Another problem with the proposed analogy is that it elides over the fact that, generally speaking, forcing individuals to leave a country in which they have been living for an extensive period imposes much greater costs upon their autonomy and well-being than denying them the right to immigrate to a specific country (which is not to deny that the costs of being refused residence can be considerable for people; e.g. when they are refugees or when they have relatives or friends within the would-be host society with whom they seek to be reunited, they often will be). Not only might being deported mean that people lose their job and/or have their education disrupted, they will have usually developed emotional ties to at least some individuals within their country of residence 34 and/or to the territory itself. 35 In addition to this, the mere knowledge that one risks being deported can, and frequently will, take a heavy psychological toll of people.

Threats to the Enjoyment of Basic Rights and Liberties
The previous section has suggested that denying residence to nonnationals based on their suspected extremist views will often be morally justified when their admission poses a substantial threat to the survival of the would-be host society's liberal-democratic order. This was found to be true even if we adopt a cosmopolitan perspective on which states are not allowed to prioritise the interests of their own citizens and residents over those of foreign nonresidents and even if the excluded individuals are refugees. The aim of this section is to propose another condition under which such ideological exclusions will sometimes be justified even from a cosmopolitan perspective and even when it is refugees who are being excluded (although here too, states might have moral duties to admit other refugees in their stead who are not suspected of entertaining extremist beliefs). This condition obtains when admitting would-be residents with suspected extremist views poses a substantial threat to the ability of existing citizens and residents of the would-be host society to enjoy their basic rights and liberties, including their rights to life, bodily integrity, and security. Since people can violate others' basic rights and liberties without endangering a country's liberal-democratic order more broadly (more on this below) and in many cases without (much) help from others, we should not be surprised to find that the current risk is more common than the one discussed within the previous section.
For examples of extremist behaviours that undermine the enjoyment of basic rights and liberties without imperilling wider liberal-democratic structures, one might think of the Jihadist terrorist attacks that were committed across the Western world during the 2000s and 2010s, including within the United States, the United Kingdom, France, Germany, Spain, and Belgium. Given how little political clout Muslim extremists had within these countries at the time of the attacks and continue to have until this day, it would be a stretch to say that the politico-legal institutions of these countries were endangered by said attacks. (Indeed, the biggest challenge to their liberal democracies might not have come from the attacks themselves, but from the ways in which their respective governments responded to them if the oft-made criticism that many of them unduly restricted civil liberties in their wake is correct.) Nonetheless, it is beyond dispute that the attacks in question violated people's basic rights and liberties, including their rights to life and to bodily integrity.
Another example can be found among the activities of right-wing extremists within Germany. Whilst these individuals lack the wherewithal to overthrow Germany's liberal-democratic institutions (this is so even if the rise of the far-right Alternative for Germany (AFD) has exerted some pressure on these institutions in recent years), there have been many violent and, in some cases, lethal attacks by German neo-Nazis over the past decades. One might think of the string of deadly attacks that the National Socialist Underground (NSU) committed on immigrants during the 2000s, as well as of the murdering of Christian Democratic Union of Germany (CDU) politician Walter L€ ubcke in 2019 36 who was killed for his prorefuge stance.
To vindicate the claim that nonnationals are sometimes rightfully denied residence based on the threats they pose to the basic rights and liberties of citizens and residents of the would-be host society even when the excluded are refugees and even when we adopt a cosmopolitan perspective, consider a group of refugees who have lived in ISIS-held territory and who are known to have openly supported this Jihadist organisation. Suppose further that, despite not being known to have engaged in terrorist activities, there is a high probability that they will have actively contributed to ISIS' brutal regime and that some of them will commit terrorist attacks within the would-be host society if granted (temporary) residence. Even when the interests of each of these individuals ought to be given the same amount of weight as the interests of each individual citizen and resident of the would-be host society, there will be a point at which the expected costs that the suspected extremists incur as a result of being refused residence will be outweighed by the expected costs that existing citizens and residents incur as a result of their admission. In order to accept this, we need not buy into some crude form of utilitarianism whereby small reductions in the wellbeing of a large group of individuals can trump the fundamental interests of a small group of individualsfor example, some utilitarian accounts will allow for cases where the refugees' interests in being admitted are overridden by the mild discomfort and fear that thousands of members of the host society will suffer because of the mere knowledge that refugees with suspected extremist views are being granted residence. Since terrorist attacks can, and often do, take lives, besides causing long-term physical and psychological harm, the costs for their victims may, and in many cases will, be at least as high as the ones that refugees with suspected extremist views incur as a result of being denied residence (idem for the costs that the relatives and friends of each of these group might incur). Especially when some of the suspected extremists are likely to have acquired knowledge and skills under ISIS' rule that allow them to cause death and destruction on a large scale, then it is plausible to assume that even on an impartial balancing of interests, the interests of existing citizens and residents will sometimes be weightier.
The same is true, I believe, when those suspected of having extremist views are not necessarily likely to commit terrorist attacks themselves, but they are rather likely to incite or inspire others to do so. Suppose that an imam who is known for his extremist speeches seeks residence within a foreign country. Even when this person faces persecution within his own country, and even when there is no concrete evidence that he will seek to propagate extremist views within the would-be host society once admitted, it seems that when there is a high risk that he will radicalise many local Muslimsas e.g. El Alami Amaouch, a 49-year-old imam and dual Moroccan-Dutch national did in Belgium where was found guilty of recruiting multiple Syria fighters, which led then migration minister Theo Francken to refer to him as a "poison for our Muslim youth" 37 the costs of admitting him will sometimes be too high even from a cosmopolitan perspective. The reason for this lies in the fact that people do not usually become terrorists unless they are radicalised by others, which means that by spreading extremist views and encouraging others to engage in terrorist activities, a lot of harm can be done to people's enjoyment of basic rights and liberties.
To be sure, in order to determine whether it is morally permissible from a cosmopolitan perspective to deny residence to individuals based on the threats they pose to the basic rights and liberties of citizens and residents of the would-be host society, it is also necessary to consider the costs that refusing admission will impose upon other members of the would-be sending society. When would-be residents with suspected extremist views pose a greater threat to the basic rights and liberties of members of their current society if they remain than they would pose to the basic rights and liberties of members of the host society if they were to migrate, then even when the latter threat is substantial, refusing them admission might not be morally justified on an impartial balancing of interests. What is pertinent for us is that, although this is correct, there are various cases where this condition does not obtain. These includes ones where people seek to commit terrorist attacks within foreign countries specifically. They also include ones where people's extremist views are already shared by many members of their current society but only by few members of the would-be host society, which may mean that for them to propagate their extremist views within the latter will allow them to have a larger impact upon people's beliefs, at least when there are local audiences who are receptive to these views. In still other cases, extremists might be so closely monitored by their authoritarian governments that they are able to do much greater evil within a liberal democracy where the relatively extensive freedoms that are availableparticularly when it comes to privacy rightsallow them operate relatively anonymously.

Three familiar objections
There are several objections that might be raised against my claim that refusing residence to nonnationals based on their suspected extremist views will sometimes be justifiable in order to protect the basic rights and liberties of existing citizens and residents of the would-be host society and that this is true even when we adopt a cosmopolitan perspective and even when the excluded individuals are refugees. Three of these objections were already (largely) answered within the previous section.
The first objection maintains that excluding would-be residents with suspected extremist views is never necessary in order to protect the basic rights and liberties of members of the would-be host society. On this view, it suffices that states closely monitor such individuals upon admission and interfere with them before they engage in any extremist activities that would violate the basic rights and liberties of members of the host society. I have mentioned several problems with this type of approach in Section 3.1. One is that frustrating people's attempts to engage in rights-violating extremist activities is often difficult. Another is that the costs of closely monitoring all suspected extremists within society tend to be prohibitive. Still another problem is that such monitoring takes a heavy toll of liberal-democratic values, especially of the value of privacy. What this suggests, I think, is that even if some form of the monitoringcum-preventive-interference approach is necessary for protecting people's basic rights and liberties, there remains a need in some cases for states to refuse residence to would-be residents with suspected extremist views.
The second objection maintains that refusing residence to people based on their extremist views is morally problematic when there exists no hard evidence that they will act upon these views once admitted. My reply to this objection was that for people to be personally suspected of having extremist viewsand, to reiterate, it is cases where such individual-specific suspicions exist on which I focus within this articlethey must have openly endorsed or spread extremist views in the past or engaged in other types of antiliberal-democratic behaviour (see Section 3.3). Given that they are not being excluded on the basis of proxies for which they lack moral responsibility, such as their race, it is all but clear that they are treated unfairly when would-be host societies do not wish to take a gamble by admitting them. Furthermore, even when there is some unfairness involved, it looks like these moral costsalong with any social, psychological, and material costs that refusing residence might impose on peoplewill be outweighed when the threat that admitting these individuals poses to the enjoyment of basic rights and liberties by citizens and residents of the would-be host society is substantial (as it will be when there is, say, a 30% chance that they will commit terrorist attacks or cause large swathes of the would-be host population to radicalise). 38 The third objection holds that if it is sometimes morally justified to refuse residence to specific individuals based on their suspected extremist views, then there must also be cases where it is justified to deport (noncitizen) long-term residents based on their suspected extremist views, which is understood to be a reductio. In response to this objection, I proposed two reasons for thinking that the proposed analogy is spurious (see Section 3.4). One is that such deportations tend to create unfairness between states, as states will in most cases bear a greater degree of moral responsibility for the fact that their long-term residents hold extremist views than the countries to which these individuals are deported or than indeed any other country. The other is that the costs of such deportationswhether they be social, psychological, financial, or otherwisewill generally be much higher for the deported than the costs of being refused the right to settle within a specific country are for potential immigrants. (The same is true, mutatis mutandis, for the costs that the relatives and friends of each of these groups incur.)

The objection from small personal risks
Even if I am right that none of these objections succeed, there is a further objection that might be raised against the practice of excluding would-be residents with suspected extremist views based on the threats that admitting these individuals might pose to the basic rights and liberties of citizens and residents of the would-be host society. According to this objection, each individual citizen and resident will normally suffer only a small elevated risk of having their basic rights and liberties violated if would-be residents with suspected extremist views are admitted given that this risk is shared with numerous other citizens and residents. Whenever this is the case, such ideological exclusions may be said to impose disproportional costs upon the relevant would-be residents, or at least upon any refugees among them.
One problem with this line of argument is that there are scenarios where admitting would-be residents with suspected extremist views significantly raises people's personal risks of having their basic rights and liberties violated. One might think of a scenario where a given group of would-be residents are likely to assault members of a small religious or ethnic minority once admitted, such as members of a small Jewish or Muslim community. Indeed, in some cases, the elevated risk of having one's basic rights and liberties violated may be significant for all (citizen and noncitizen) inhabitants of a country. Just think of a country with a small and highly urbanised population (e.g. Monaco) where the admission of even a single terrorist can pose a substantial threat to everyone's lives. Other examples include ones where a large number of would-be residents with suspected extremist views is admitted relative to the size of the host population, as well as ones where some of the admitted would-be residents are capable of radicalising large segments of the host population because of their religious credentials, charisma, intellect, and so on. Furthermore, it is at least conceivable that within future societies, even small numbers of people will be able to cause death and destruction across wide areas because of the ease with which new types of biological weapons can be produced that allow them to e.g. contaminate the water supply or spread poisonous gasses.
Butand this brings us to the other problemeven when people's personal risks of having their basic rights and liberties violated by foreign extremists are small, this does not seem to be a decisive objection against denying residence to nonnationals with suspected extremist views as such, at least not without further argument. Consider the high-way bridge collapse in the Italian city of Genoa that took the lives of 43 people in August 2018. Though the personal risks of dying within such accidents are tiny, we still think that it is a real tragedy when people die in them, which is part of the reason why we expect the construction industry to make considerable efforts to prevent them from happening. Or consider cases where there is a high probability that a convicted murder will murder again if he is released from prison. Even when there is no evidence that he will murder any particular individual upon release, meaning that the risk of becoming his victim is small for every single member of society, it may be plausibly argued that this person should not be released as long as the risk that he will murder someone remains substantial. (Notice that in order to accept this, we do not need to deny that, all other things being equal, it is morally worse for people to suffer higher risks of having their basic rights and liberties violated than to suffer lower risks; all that we are committed to is that the probability that some people will have their basic rights and liberties violated has considerable moral weight independently of the individual risks to which people are exposed.) But if this is correct, then it is unclear why we should believe that, no matter how high the probability is that some members of the would-be host society will become the victims of deadly terrorist attacks or of other violations of their basic rights and liberties if would-be-residents with suspected extremist views are granted admission, the interests of the expected victims can never trump the interests of the would-be residents in being admitted when the personal