Republicanism and the legitimacy of state border controls

ABSTRACT

A number of recent articles have invoked the republican ideal of non-domination to justify either open borders, and/or the reduction of states’ discretionary powers to unilaterally determine immigration policy. In this paper, I show that such arguments are one-sided, as they fail to fully account for the deep ambiguity of the very ideal which they invoke. In fact, non-domination lends just as powerful support to maintaining state border controls as it does to dismantling them. There are only two exceptions to the rule. It is well established that promoting non-domination demands, one the one hand, that refugees be admitted, and second, that all migrants have a right to contest decisions concerning their own admission. But aside from these things, the policy implications of the ideal are unclear. In itself, therefore, it is insufficient to justify either open borders, or the reduction of states said discretionary powers. Such arguments will have to rely on other, additional moral criteria.

KEYWORDS:

• republicanism
• non-domination
• freedom
• immigration
• ethics of migration
• free state

A number of recent articles have invoked the republican ideal of non-domination to justify either open borders (Costa 2020), and/or the reduction of states’ discretionary powers to unilaterally determine immigration policy (Benton 2014; Honohan 2014; Hoye 2019; Owen 2014; Rochel 2021; Sager 2014). The essence of these arguments is the following. Every person has an elementary and legitimate claim for freedom as non-domination. State border controls exercised unilaterally violate this claim for migrants and are therefore definitely immoral. In this paper, I show that such arguments are one-sided, as they fail to fully account for the deep ambiguity of the very ideal which they invoke. In fact, non-domination lends just as powerful support to maintaining state border controls as it does to dismantling them.

To be sure, this is partially acknowledged by some of the articles cited above. To back up their overall assessment, they offer a number of arguments to show that despite the ambiguity, non-domination lends much more powerful support to their claim. The chief contribution of my paper is to show that none of these arguments is successful. There are only two exceptions to the rule. It is well established that promoting non-domination demands, on the one hand, that refugees be admitted, and second, that all migrants have a right to contest decisions concerning their own admission. But aside from these things, the policy implications of the ideal are unclear. In itself, therefore, it is insufficient to justify either open borders, or the reduction of states said discretionary powers. Such arguments will have to rely on other, additional moral criteria.

The paper is structured as follows. In the first part I set out the claim according to which republican theory offers a solution to the dilemma concerning state border controls. In the second I identify general reasons that explain the limitation of this claim. In the third I consider the arguments offered to back up the claim according to which republican theory does offer an unambiguous solution. As we shall see, there is indeed reason to believe that the existence of state border controls constitutes domination. In the fourth part I briefly examine the policy implications that follow from the normative arguments previously set out. These are the policies that aim to weaken the discretionary powers of the state. In the fifth part, however, I show that such policies may also be contrary to the non-domination criterion, the implications of which are therefore highly ambiguous. In the final part of the paper, I consider possible arguments for resolving the dilemma. None of them proves to be convincing.

A republican approach?

There is great controversy regarding the legitimacy of state border controls. Arguments in favour of maintaining such controls primarily invoke rights for self-determination. They also tend to point out the (alleged) negative impact mass immigration has in receiving societies to show that self-determining communities have good reason to want to control settlement to their territories. (see Alesina and La Ferrara 2002; Delhey and Newton 2005; Dinesen, Schaeffer, and Mannemar Sønderskov 2020; Putnam 2007; Soroka et al. 2018). On the other hand, arguments questioning the legitimacy of state border controls primarily rely on the idea that state border controls violate basic claims of migrants such as their freedom of movement (Sager 2020), or their interest in escaping poverty (Carens 2013).

A prominent view in recent discourse has been that republican theory offers a distinct and more or less unambiguous approach to the issue, one that is derived from its freedom ideal. Republicans believe that freedom demands immunity from domination – both public and private. The relevant issue for them is not primarily the extent to which some people (or groups) actually interfere with others, but whether there are agents in society who have the power to do so arbitrarily. In fact, they contend that for the most part only arbitrary interference is contrary to freedom. By contrast, interference that is somehow checked is not (Pettit 2014). Of course, there is some disagreement among republicans as to what all this precisely means. Perhaps most common is the definition according to which such freedom demands being in control of one’s own fate (Pettit 2012).

At any rate, this is the definition that is applied either exclusively (Fine 2014, 12; Hoye 2019, 2) or in combination with alternative definitions (Benton 2014, 52; Honohan 2014, 34–6; Rochel 2021, 560–1; Sager 2014, 3–4) by most of the papers invoking the ideal of non-domination in the debate on the legitimacy of state border controls.¹ The alternatives are on the one hand the procedural, and on the other the interest-based accounts. I will return to discussing these in the final section of the paper. We shall see in Costa’s case that there may be ways to argue that the entire puzzle arises from the fact that various conceptualizations of domination (and arbitrariness) have different implications regarding the question of the legitimacy of state border controls (Costa 2020). I will show, however, that such arguments are not convincing. Fleshing out the precise meaning of non-arbitrariness is beside the point here. All the conceptualizations, without exception, are highly ambiguous with regards to the legitimacy of state border controls. (And as a result, my central claim stands: non-domination is in itself insufficient to justify either open borders, or the reduction of states said discretionary powers.)

But whichever way they conceptualize non-domination, many authors propose that promoting it will support either of two ideas. One is that border controls as such, another is that border controls exercised by states unilaterally are immoral. This is a question of legitimacy. What the common republican conclusion suggests is that states do not have a legitimate claim to restrict migration and settlement to their territories. This is not because state border controls interfere with movement or because they retard efforts to equalize opportunities globally. Instead, the argument is that such border controls are immoral because they dominate migrants. Their functioning results in a situation where migrants are frequently exposed to the arbitrary will of receiving state’s authorities and even their citizens, placing them in a severely unequal position of power, significantly stripping them from control of their lives.

The current state of the debate is as follows. The majority of the articles that have invoked the republican ideal of non-domination to justify reducing states’ powers to unilaterally determine immigration policy either do not engage with the potential ambiguities of the ideal they invoke (Benton 2014; Costa 2020; Hoye 2019; Owen 2014; Sager 2014), or do so only in brief passages (Honohan 2014; Rochel 2021). Few alternative approaches are available. One is that of Philip Pettit who argues that state border controls in themselves do not constitute domination in any way (Pettit 2012, 162). Another interesting example is that of Matthew Hoye who although himself shares the view that the implications of the non-domination ideal are unambiguous, is less convinced about the unambiguous nature of republican theory taken more holistically (Hoye 2016, 2018). Finally, Sarah Fine offers an argument to show the ambiguities of the republican ideal (2014).

The two reasons behind the ambiguity

In this paper, I depart from all the aforementioned strands within the discourse. I show that while state border controls are genuine forms of domination, this does not settle the dilemma at hand. And while I share Fine’s overall assessment regarding the ambiguity of the republican ideal, I also do not believe that her argument fully grasps the reasons behind this. I offer two key arguments to back up my claims.² In the sections below, I elaborate on them further.

The first argument is a purely conceptual one. It is that standard definitions of non-domination involve a concern for egalitarian aims as well as for democratic self-determination. As we have seen, these are some of the key considerations around which the entire debate regarding the legitimacy of state border controls tends to revolve. So, on the one hand, a standard argument against state border controls is that they contribute to the maintaining of global inequalities. Dismantling them would provide large numbers of people in the ‘global south’ with the opportunity to migrate to wealthier areas of the world, thereby helping them escape poverty. Clearly, this argument is closely analogous with republican concerns over the domination of migrants, as being dominated precisely means a sort of inequality – namely (severe) inequality of power. The reason some republicans propose the dismantling of state border controls is precisely that they believe this will result in power relations becoming more egalitarian between migrants and citizens of receiving states.

On the other hand, a standard argument for state border controls is that they are instruments for the exercise of democratic self-determination. This argument is also analogous with republican concerns over domination. According to the standard definition, non-domination demands that agents be in control of – or in other words, have the ability to determine – their own fates. If applied to citizens of receiving states rather than to migrants, such control may entail that citizens be able to implement border restrictions through controlling the representative institutions of the state. There is, therefore, a conceptual reason behind the ambiguity of the non-domination ideal.

The second reason behind the ambiguity has to do with the complicated relation between republican and cosmopolitan ideals. The complication is due to the fact that republican theory contains three desiderata which often pull in different directions. One is reducing informal power – say, the power of multinational corporations – on a global scale. This pulls in the direction of establishing institutions which regulate actors ‘currently operating in a rule-free zone’ (Ronzoni 2017, 204). At the same time, the desiderata of resisting ‘global despotism’ exercised by great power-centres and also that of fostering civic virtue, pulls away from this commitment. As a matter of fact, existing supranational governance is increasingly ‘exposed to the risk of technocratic distortions and poor accountability’ (Ronzoni 2017, 204) and, for reasons elaborated on below, is also less likely than the state is to achieve success at virtue cultivation. So, the challenge is how to balance such divergent aims.

This puzzle applies to the case of state border controls because republican arguments in favour of dismantling such controls usually entail support for either strengthening existing cosmopolitan institutions or establishing new ones which could take over the relevant competences – either partially, or entirely. It is widely believed that if decision-making regarding border controls were made accountable to all the relevant parties – migrants included – rather than to citizens of the receiving state alone, then overall domination could significantly be reduced. But clearly, while the establishment of an international body for the supervision of border controls could reduce domination exerted by states, it would at the same time increase the dominating potential of the international bodies in question as well as retard efforts aimed at virtue cultivation.

It is not easy to balance the issue. On the one hand, should concerns with informal power asymmetries be afforded priority and cosmopolitan institutions strengthened accordingly, that would simply bring us back to the challenge posed by supranational governance being increasingly exposed to the risk of technocratic distortions and poor accountability. There clearly is a limit to the extent to which republican aims can be reconciled with cosmopolitan ones. On the other hand, should international regulation remain limited – as suggested by Ronzoni (2017, 207) – that would leave the dominating power of states over migrants largely intact. The only real way to bypass the dilemma is weakening states powers to unilaterally determine immigration policy without at the same time delegating too great an authority to cosmopolitan institutions. This could be done by delegating a degree of power directly to those involved, namely migrants, granting them some form of say in admissions policies. As we shall see in the final section, however, such rights will have to be rather limited in scope if they are to be reconciled with the other republican considerations.

Republican arguments against state border controls

But before exploring that question, let us review the arguments republicans offer to show that state border controls dominate migrants! For now, there will be no need to differentiate between the various forms of movement. Arguably, all migrants face some instances of domination – public, or private. First of all, developed countries often cooperate with their less developed neighbours in order to stop migrants before they reach their borders. Think of the EU-Turkey deal. Since migrants have no control over the making of such agreements, they are arguably dominated by them as well (Costa 2020, 3).

Second, migrants who attempt entry legally place themselves at the mercy of the receiving bureaucracy (Costa 2020, 7–8; Owen 2014; Rochel 2021). For this to count as domination, bureaucratic power does not even need to be totally unchecked. Take the case of the EU, for instance. Here, bureaucratic power is limited by internal supervision, as well as NGO-s and political parties which monitor breaches of the law (see Rochel 2021). The reason why the relation may be counted as dominating is that migrants themselves do not have any say in whether they are allowed entry or not. Such decisions are made unilaterally.

Thirdly, there is the case of those who were permitted legal entry. Such people may be subject to domination for a number of reasons. For instance, they may be at risk of deportation; in most cases they do not have equal political rights with natives, and as a result their claims are usually not afforded equal consideration by political parties (Costa 2020, 4–5). Additionally, acquiring a visa is often conditional on being employed. This pushes immigrants into accepting unjust, even exploitative conditions at the workplace (Benton 2014, 57).

A lot depends on the concrete situation, though. Benton argues that the degree to which a legal immigrant is dominated varies depending on how costly it is for them to exit their situation – either by emigrating once more or acquiring citizenship. Possible costs involved are manifold. Some of them are of a personal character, others of an administrative one. A personal cost is the loss of friends, family, the social network, and the workplace already acquired in the country where the migrant has settled. From an administrative point of view the relevant question is how difficult it is to become a naturalized citizen. More general indicators for the assessment of the degree of domination include the length of time the migrant has already spent in given country, whether he or she speaks the language of the local majority, what sort of legal status he or she has, how high his or her income is, and of course, what the conditions are like in the country of his or her origin – the country therefore where he or she might be sent back to (Benton 2014).

Fourth and finally, there is the case of illegal immigrants. Arguably, these people are subjected to the most severe forms of domination. By crossing the border, they breach national law. This will give them an immediate reason to avoid authorities, so as to make sure their situation remains concealed, and no great danger of persecution or deportation arises. This opens possibilities for enhanced forms of private domination, of course. Illegal immigrants will be increasingly exposed to their employers, landlords, and even their neighbours, who can simply report them to the authorities if any sort of conflict arises. In fact, if they run into trouble, illegal immigrants will have good reason to be weary of going to the police as this would also risk exposure to immigration authorities, and perhaps deportation too. Put bluntly, then, illegal immigrants are not equally protected by the law (Costa 2020, 4).

Policy suggestions derived from republican arguments against state border controls

As for the question of what follows for policy, there are two main strands among the republican critics of state border controls. While some argue that border controls as such are immoral and that an open borders policy is to be preferred (Costa 2020), the majority of the articles cited suggests that only state border controls exercised unilaterally are immoral. To explain this, one need only remember that according to the republican theory of freedom, interferences do not necessarily count as restrictions of the relevant kind. Only if they entail domination also. And if there is a way to make border controls non-arbitrary, then they need not be removed either.

Perhaps there is a way to achieve this. In fact, there are two possibilities on offer. One solution is to provide migrants with a say in the decision-making process concerning their own entry and possibly concerning the relevant policymaking as well (Honohan 2014, 34–42). Another solution is the establishment of an international body for the supervision of state border policies. In such ways, state sovereignty could be weakened significantly enough to reduce the domination migrants are exposed to from the side of state bureaucracies and the rest (Fine 2014, 24–5; Honohan 2014; Costa 2020, 8).

On the other hand, contemporary republican scholarship is more or less unanimous in suggesting that migrants who have already settled in the receiving state ought to be provided with a legal status that is similar to citizenship. They argue that this status ought to be provided automatically after migrants settle in the country (Benton 2014, 34–42; Owen 2014), even, perhaps, if entry was done illegally (Sager 2014). The aim is to include them in the democratic decision-making process, providing them with a degree of control over their own fate (Benton 2014; Fine 2014, 21–22). Of course, other provisions are available. It might be necessary to set up a number of ‘sanctuary cities’ where the laws requiring the deportation of illegal aliens are not enforced (Hoye 2019). The laws concerning naturalization ought to be relaxed and the level of welfare provisions increased (Benton 2014).

But nevertheless, as Sager points out, democratic rights do not have a real substitute here. They will be of primary importance. First, judicial review in the United States is deficient because courts can guarantee equal treatment under existing law only. Political rights, on the other hand, provide migrants with the opportunity to change the law as well, to introduce new concerns and agendas etc. Second, labour rights cannot fully replace political rights either because existing labour unions often have conflicting interests with migrant workers, as the arrival of the latter may retard some of the hard-won victories in the battle for higher wages. Such unions cannot reasonably be expected to protect immigrant interests. A third candidate is movement politics, or the ‘agonistic’ forms of political struggle. These are deficient for the simple reason that they cannot provide the kinds of (legal) guarantees formal political rights can. Fourth and finally, one might rely on sending countries to give voice to their diaspora. Sager is again sceptical, pointing to the fact that sending countries have no real power outside their borders (Sager 2014, 11–2). This, however, is a rash generalization. Should power asymmetry favour the sender, the provision of voice can be successful. The government of Bulgaria does not have the resources to give voice to its citizens in the United States. But the reverse is not true. The U.S. government is usually quite successful in providing for its citizens abroad. In another sense, though, Sager does have a point. Relying on sender countries is something that may work in a very limited number of cases only. Usually, this option is not available.³

Republican arguments in favour of state border controls

None of this means, however, that the dilemma is settled. In fact, the non-domination ideal lends powerful support for maintaining state border controls, the same way it does for dismantling them. First, there is the conceptual issue I point out in section two of this paper. Non-domination entails a strong concern for the exposure of migrants as well as for democratic self-determination. It has both an individual and a collective dimension – and democratic self-determination has a lot to do with the latter (Fine 2014). When national sovereignty is restricted either by an international organization or a group of migrants, the receiving political community loses part of its ability to control its own fate – say, determine the policies at home, or create law. This ability is partially ceded to foreigners. If it is up to an international body to determine what number of migrants this or that state is required to allow entry to the citizens of the given country have a reduced ability to organize political life at home according to their own collective will (Fine 2014, 22–3). And this is very much a question of power and domination. If a democratically elected government cannot decide on a basic issue such as border controls, that does not only reduce sovereign rights, but also democratic control within the country (Fine 2014, 24; see also Laborde and Ronzoni 2016; Skinner 2010, 98–9).

Secondly, there is an issue which even Fine does not consider. She sees potential for domination in the relations between international organizations and states. But the relation between states and their own citizens may become just as dominating. This happens when citizens have no democratic control over the relevant policies at home. Obviously, governments may be ‘biased’ in either of two ways. Sometimes the majority favours open borders, and yet government decides to restrict immigration. But the reverse is also possible. There are plenty of cases when the majority favours further restrictions on immigration, but government disregards this and allows for the inflow of large numbers of people. If there is a measurable majority preference, formed within democratic procedures, then if government consistently disregards this that may arguably be considered domination too. What we would be dealing with is a typical case of non-democratic, and thus, by republican standards, unfree governance. People would lose control to a significant degree over managing the population levels, social circumstances, cultural character, and other segments of their own communities.

Thirdly, it is not only the non-domination criterion itself that may support the case for state border controls. The same is true about some of the things which republicans consider key instruments for upholding non-domination. One such instrument is civic virtue. With a normative twist, republicans define it as readiness to engage in public life in order to secure the common good of a non-dominating government where all members are treated as equals (Pettit 2012, 228). Importantly, however, they also believe that the success of policies promoting the cultivation of such virtue largely depends on the degree to which people identify with the polity in question (Honohan 2002, 171–4; Laborde 2008, 247; Pettit 1997, 126; Viroli 2002, 86). Only if such civic (patriotic) ties are strong will citizens be likely to regard virtuous behaviour worth the effort. And success is absolutely vital, since republicans believe that for the stability of any free polity, citizens will have to be virtuous to a relatively high degree.

This poses a challenge for proponents of the view according to which non-domination, and republican theory in general, supports the case for dismantling state border controls. Nowadays and in the foreseeable future, reasonably strong ties of the mentioned sort can be found in the state alone, the defence of which is necessitated, it would seem, for promoting virtue cultivation (Ronzoni 2017, 204). Perhaps things will change with time. Cosmopolitan identity might become more prevalent globally, thereby making it suitable to support cosmopolitan institutions. But this will not happen in the foreseeable future. And that calls for caution because disregarding the importance of the institutions and political identities that currently uphold republican government might eventually lead to its weakening (Laborde 2002, 604).⁴

Indeed, there is some empirical evidence to suggest that mass immigration may erode social cohesion, civility, and trust – all of which are part and parcel of the civic ties republicans customarily deem necessary for the cultivation of the relevant civic virtues (Dinesen, Schaeffer, and Mannemar Sønderskov 2020). This is something Pettit stresses in particular. In his view, developed countries do not have the capacity to permit entry to all of those in the ‘global south’ who are in need of assistance. Should they open their borders, sooner or later immigration to their territories will become so massive as to become unsustainable. This is a disaster scenario: sudden and excessive overpopulation may lead to the weakening of civic ties and in the long run to the disintegration of the very institutions that uphold freedom and the rule of law (Pettit 2012; see Costa 2020, p- 6).

Additionally, in some cases, mass immigration undermines the consensus concerning key political norms in society – an issue I return to in the final section of the paper. Hoye draws attention to the fact that the language, religion, customs, and political views of immigrant communities are often very different from those of the receiving society. This is a problem because promoting virtue cultivation – in other words: a perfectionist policy –, will demand a certain consensus concerning the relevant norms. Incidentally, many (but obviously not all) immigrants bring with them a number of moral and political views that are contrary to the desired consensus. So, a possible way to protect the (tentative) consensus within states will be restricting immigration (Hoye 2016, 3).⁵ Otherwise, it will be at constant risk of disintegration.

Finally, mass immigration poses a potential threat to another thing that most republicans consider an important guarantee of non-domination, namely the welfare state. The potential harm is limited insofar as it does not significantly influence the welfare state as a whole but merely specific programmes that migrants are more likely to use. The typical example is the reduction of spending for unemployment benefits (Soroka et al. 2018). Such results are what is in common parlance called ‘welfare chauvinism’. On the other hand, the harm takes time to unravel. This is because immigration usually happens gradually and because popular grievances over the process are usually not quick to receive political articulation either. But even so, there is a potential threat. For all these reasons, citizens of receiving states will have an incentive to want to retain control over the levels of immigration – and state border controls are a means to this end.

Let me underline that with regards to worries about civic virtue and the welfare state, we are merely dealing with incentives. Therefore, the argument here is different from either one of the two before. Unlike those, this one does not lend clear support for state border controls. This is because what ought to matter for republicans is that immigration be kept at a level that is consistent with promoting civic virtue and defending the welfare state. In principle, it is possible to imagine a system in which immigration policy is (in part) determined by an international body (or is determined by states and migrants in concert) but does not allow for levels of immigration to (developed) states that would pose too big of a challenge. However, even so, retaining border controls does have an important advantage from the state’s point of view. This way, the state has far more autonomy in managing population levels and determining what exactly the preferred level of immigration should be.

Towards resolving the dilemma

In what follows, I explore the possibility of resolving the puzzle at hand. In doing so, I engage with arguments offered by previous articles which aim to show that republican theory lends more powerful support to the case against state border controls. In addition, I offer possible arguments of my own. None of them will prove to be successful.

Balancing various considerations

It would obviously settle the dilemma if it could be shown that the domination receiving societies potentially face is miniscule compared to the domination which threatens migrants. But more often than not, such comparing is very difficult to do. We are dealing with apples and oranges, so to speak. And this is where the relevance of the conceptual problem introduced in the second section really emerges. To recall, the complicated nature of the non-domination ideal is due to the fact that it entails a powerful concern for both migrants’ plea for being treated as equals to members of receiving societies, and for these societies’ claims for self-determination as well. These are very different considerations, both relevant, but difficult to compare.

In fact, by appealing to non-domination, republicans merely import the terms of the well-known debate between supporters of open borders on the one hand, and supporters of closed borders on the other. If anything, they make it even more difficult to balance insofar as they force the divergent considerations into a single, thereby deeply ambiguous concept. Following republican terms, then, the issue of the legitimacy of state border controls ceases to be one of balancing the relevance of different values defined by different concepts – such as equality and self-determination – and instead becomes one of interpreting the implications of a single concept.

The only real exception to the rule is the case of refugees fleeing wars, persecution, or starvation (Fine 2014, 14–5). Thus far, I have been speaking in general terms, not differentiating between the various types of migration. But refugees are exposed to arbitrary interferences to such a high degree that the reduction of self-determining rights in receiving societies is indeed a comparatively marginal form of domination. If not permitted entry, refugees face severe suffering, often death. But there is nothing controversial about their case either. Even those advocating a closed borders policy accept that refugees ought to be allowed entry (Fine 2014, 17; Miller 2016; Pettit 1997, 152). If the non-domination criterion solves the dilemma only in such uncontroversial cases, then it does not seem all that useful after all. Things would be different if it could offer solve the dilemma for controversial cases too, namely economic migrants. This, however, it cannot offer.

The potential challenge posed by mass immigration

Rochel concedes that there is ambiguity to the republican ideal, but then goes on to argue that the domination produced by the weakening of sovereign rights is not too great a price to pay for the assistance of those in need (Rochel 2021, 570–1). He draws on evidence from history to support this claim. While migrants frequently are in great need of assistance, there is no evidence of mass immigration causing the disintegration of institutions upholding freedom as non-domination. Western Europe has been a destination for such migration for over six decades, and yet none of its countries have descended into autocratic rule.

But while this is true, Rochel seems to overlook part of the issue here. First of all, one reason the ‘disaster scenario’ has not yet come to pass is arguably the fact that the open borders policy has not been practiced in Western Europe. Therefore, the empirical evidence Rochel appeals to possibly derives from the very practices, which he presumably wishes gone. Secondly, even if the ‘instruments’ upholding non-domination have not been dismantled, there are nevertheless a number of negative tendencies present. I have already alluded to the empirical evidence suggesting that mass immigration erodes civic ties, social cohesion, and the rest. These findings are not set in stone. But to disregard them completely and to claim that there is nothing to suggest that mass immigration poses a challenge to these institutions is an overstatement.

Migrants are no dominators

For a political community to be free by republican standards it needs to be immune from domination both within and without (Laborde and Ronzoni 2016; Pettit 2014). This is why republicans are weary of the power held by large states, or the influence of multinational corporations. But according to Honohan, immigration has nothing to do with these things. She suggests that state border controls are illegitimate because immigration restriction is not analogous with territorial defence and is therefore not necessary for a state to secure external domination. For unlike large states and corporations, migrants lack the resources to count as dominating agents. In fact, more often than not they are the ones to have fewer resources than the receiving community (Honohan 2014, 43–4).

Honohan has a point here. Nevertheless, implementing her insight will face a couple of challenges. First of all, Honohan thinks that immigration policies should not be determined by states but rather by an international body. But this just returns us to the dilemma regarding the relation between republican and cosmopolitan ideals. Migrants may not be dominators themselves, but in Honohan’s model reducing their domination does entail strengthening global institutions which by republican standards do (or may) indeed count as dominating.

Honohan’s insight may, however, be implemented without running into this familiar challenge. Namely, the whole problem can be bypassed simply by weakening states’ powers to unilaterally determine immigration policy without at the same time delegating too great an authority to cosmopolitan institutions. As we have seen, the way to do this is by delegating some power directly to migrants – an idea Honohan contemplates but does not elaborate on unfortunately (Honohan 2014, 42). So, the question is what the scope of migrants’ say will be in all this. Veto rights for migrants would not make much sense. This way, state authorities would lose all their real power in this respect. So, evidently, migrants can only be given limited influence in the decision-making process, say, a right to contest.

But the question is who gets a say in what. Should migrants be granted influence only in the decision-making process concerning their own entry, or should they be able to also participate in the process of determining immigration policy? Honohan answers both questions in the affirmative (Honohan 2014, 42). On the other hand, is it only migrants who have come into contact with the receiving state’s authorities who should be granted a say, or potential immigrants too? These are difficult questions. Granting influence in the decision-making process concerning migrants’ own entry is both reasonable and feasible. It would give control to them in an important issue concerning their own lives without placing too large a burden on the receiving state. Granting influence in policy making is a whole other matter entirely. For as long as immigration policy is determined by national parliament (or perhaps referenda), this would involve giving political rights to potentially large numbers of people, millions or more. Aside from facing practical difficulties, such a move would lead to a severe loss of collective self-determination rights for receiving states, thereby returning us to the dilemma once more (see also Abizadeh 2008).

What arbitrary power means

Thus far, I have employed the more or less standard definition of domination. I assumed that an agent is dominated to the extent that he or she is not in control of his or her own fate. However, this not the only definition on offer in contemporary republican discourse. There are two alternatives to it. Costa compares them and explores their respective implications for immigration policy (2020). We have already seen that non-domination defined as being ‘in control’ of one’s fate would count as a highly ambiguous ideal for present purposes. This is something Costa does not acknowledge, which is a shortcoming of her argument.

But what of the other definitions? The second definition is that there is domination a) when there is social engagement between two agents; b) one party depends more on the engagement then the other; c) there is power asymmetry; and d) if the asymmetry is so severe as to allow the stronger party to interfere with the weaker one arbitrarily (Lovett 2012). Non-domination requires that power be restricted by procedural means, such as law, rules, institutions, and the rest. According to Costa, this second definition does allow for ambiguity insofar as its scope is relatively narrow. Namely, the procedural definition does not capture all instances of domination which would Costa count as such intuitively. The example of migrants who have not yet crossed the border is a case in point. As Costa argues, the reason these people are so exposed to domination is precisely that they are not yet engaged in any sort of social interaction with the receiving society. The problem with Lovett’s definition is that because of its emphasis on social interaction and dependence it cannot identify this latter sort of harm. A second and unrelated problem is that the procedural criterion in itself says nothing of the desired content of the procedures. Racist discrimination in admission: is it good or bad? This is a question that simply cannot be answered by appealing to the procedural criterion alone – or so goes Costa’s argument (Costa 2020, 11–3).

The third definition of arbitrariness is the one Costa finds most promising. This is the interest-based account, developed also by Pettit in his early work (Pettit 1997). On this account, interference is arbitrary if it does not serve the common avowable (or objective) interest of those who are subjected to it. And if there are interferences that are contrary to the objective interest of economic migrants, then state border controls are certainly one of those. Such migrants have left their homes precisely in order to find better life opportunities for themselves. State border controls stop them from doing so and can therefore be regarded as arbitrary interferences (Costa 2020, 14–6).

But here again, Costa’s argument is one-sided. She does not seem to acknowledge that the interest-based approach, like the other ones, may just as easily be employed to defend state border controls. It is one thing that such controls do not serve the objective interests of migrants. But they may serve the objective interests of others. Take the economy as an example! Mass migration is often contrary to the interests of the receiving society insofar as it does sometimes place a burden on the welfare system and leads to the arrival of unneeded labour force. Sure, this is not always the case. When there is shortage of workforce, immigration is vital for increasing productivity. But sometimes it nevertheless is the case that economic calculation dictates restrictions on immigration.⁶ And there are other things to take into account too. One is the worry concerning the weakening of trust in society. Another is the reduction of certain welfare provisions. Drawing on the interest-based approach as advocated by Costa, state immigration controls will not be the only interferences to count as dominating. The same will be true about an international body’s decision to coerce a state into accepting more migrants than it is necessary for the needs of the local economy. In fact, for the receiving state the interest-based criterion might even justify discrimination in allowing entry on the basis of education, age, supposed economic ‘gain’ or other factors.

Are border controls objects of sovereign rights?

Honohan further suggests that another possible solution to the dilemma is to establish that border controls are not objects of sovereign rights, or collective self-determination rights either. Sovereignty is about the state’s authority vis-à-vis its subjects and other states. From this angle it is irrelevant who crosses the border and when. State sovereignty is not reduced by ceding authority over border controls to an international body. Similarly, although all societies have a right for self-determination, this does not entail a right to control movement. Political associations are not like clubs and therefore their membership cannot be restricted to a select few (Honohan 2014, 43–44).

Perhaps this is so. But Honohan provides no arguments in favour of her claim. And this leads to an impasse. Honohan’s claim makes the whole issue into a matter of definition. Something is definitely lost for states and political communities when border controls are supervised by an international body. It is of course up to debate what this something is. But Honohan does not even enter the debate. To be clear, I do not wish to suggest that states have a legitimate claim to control movement on their borders, or that this claim supersedes migrants’ claims for entry. The question is not this. The question is whether or not border controls can even be interpreted either as objects of sovereign or of self-determination rights. Honohan provides no arguments to back up her claim according to which they cannot be interpreted this way.

Mass immigration may in fact help cultivate civic virtues

All in all, then, the non-domination ideal provides no solution to the dilemma at hand. But perhaps republican theory, considered more holistically, does. To show this, Hoye returns to the issue of virtue cultivation. He does this by taking a fresh look at the historical sources. One of his references is Machiavelli, who famously believed that conquering states will do good to settle the population of the newly acquired territory in the capital. The underlying reason is of course that a populous city is stronger than a small one, as it will have more soldiers to recruit and more traders to pay taxes. But Hoye notices another important element in Machiavelli’s argument.

As already mentioned, immigrants often hold moral and political views that significantly differ from the (tentative) local consensus in receiving states.⁷ Hoye believes that one possible response to this challenge is maintaining state border controls and restricting immigration. But this is not the only possibility. On Hoye’s reading, Machiavelli supplies an alternative. This is that the occasional weakening of the consensus might be useful in the long run. Norms are augmented by forcing them to prove their value. Virtue will stay strong only if tested from time to time. And if this is true, that might even make immigration the very ‘engine’ of maintaining non-domination in the receiving country (Hoye 2016, 8).

Of course, it is hard to say whether testing the stability of a political consensus is really more efficient than protecting it by restricting the inflow of people who might challenge it. The latter is definitely the safer option. Firstly, the more people reject republican values, the more difficult it will be to promote them in society. Here it might be useful to repeat what this refers to, precisely. There is a question of how much shared values the ‘consensus’ demands. For instance, immigrants are often just as concerned like receiving societies are about environmental, or public health issues for instance. But this is beside the point. Republican virtue means readiness to engage in public life in order to secure the common good of a non-dominating government where all members are treated as equals. So, the stability of the consensus demands that large segments of society value freedom itself. Secondly, mass immigration does not pose a challenge to the ‘republican consensus’ only because of (some) immigrants’ views. It has also led to the emergence of a xenophobic far-right in Europe, the presence of which also undermines the promotion of non-domination.

But Hoye has other historical references besides Machiavelli. These are from 18^th and 19^th century American discourse. Hoye shows that republicans of the time did not see immigration as a threat. Rather, they took it as an opportunity. They believed that those fleeing oppression will in their new patria become the most ardent defenders of liberty, more ardent that those who were born into it and have become ‘lazy’, taking their good situation for granted (Hoye 2018, 18–22).

There are a number of issues with this argument. Firstly, it applies to a specific type of migrant only – namely the refugee. It does not apply to economic migrants. No doubt, this is an unintended conclusion. Hoye’s argument is meant to address the issues of migration and border controls in a general fashion. If, however, the criterion for admission is fleeing from oppression, then surely, the logical conclusion is that only refugees ought to be allowed entry.

Secondly, just like the previous argument, this also rests on dubious empirical assumptions – this time concerning motives. People leaving oppressive societies do not all mean to avoid the oppression itself. Many migrate for economic reasons instead. As a consequence, they will most certainly not all become ardent supporters of liberty and rights in their new patria. Some of them will be supporters of the oppressive views they hold. This is of course a problem that Hoye himself is familiar with, as his previous, ‘Machiavellian’ argument rested on the assumption that it is real.

Not to mention that motives may be problematic on both sides. Interestingly, Hoye draws a conclusion from 18^th century American discourse without taking seriously some of its important aspects. Most of all, he glosses over the fact that republicans of the time were not expecting the arrival of either Catholic, Jewish, or Muslim refugees, but rather that of Protestants from Western and Central Europe. These communities were culturally very similar to the majority in the States. Their acceptance might even have been motivated by racist motivation.

Finally, the psychological assumption Hoye builds on is similarly questionable. It is certainly possible that a society which grows used to the benefits of non-domination becomes to a degree ‘lazy’ and takes said benefits for granted. But the reverse is just as possible. In fact, societies with long democratic traditions are much more likely to foster the civic virtues necessary for maintaining a non-dominating government than societies that are largely lacking such traditions. The recent history of Eastern European countries is a case in point.

From the non-ideal to the ideal

Another possible resolution would be to suggest that what we are dealing here is a matter of feasibility. Perhaps republican ideals are ambiguous only insofar as they entail different implications for ideal and non-ideal settings respectively. In Rawlsian terms, ideal theory refers to the effort of setting down principles of justice that are meant to regulate society’s basic institutions. Essentially, it is an account of the sort of society we ought to be aiming for. Importantly, feasibility is not an issue here. It is simply assumed that people will comply with the principles in question (Rawls 1971, 245). The point is therefore simply to define long term goals for which any society ought to be aiming for. By contrast, non-ideal theory deals with devising principles for actual, real-world circumstances, where strict compliance is difficult, sometimes even impossible to obtain. Such theorizing is about how said ideals might be achieved if compliance is not taken for granted.

Republican arguments for state border controls are often rooted in contingent historical fact – for instance, that supranational institutions do not command powerful enough attachment and that they are increasingly exposed to the risk of technocratic distortions and poor accountability. Additionally, the argument in question is highly contingent on republican speculation concerning the importance of civic patriotism being correct. Which is of course far from obvious. For instance, there is plenty of room to argue that rather than civic bonds, it is self-interest from the side of citizens that is demanded for a flourishing democracy (Levy 2017).

Ideally, however, the non-domination ideal, if applied in an egalitarian fashion, would seem to demand that everybody be in control of their own fate – migrants and members of receiving societies alike. For if it were possible to create democratically accountable cosmopolitan institutions – as perhaps in the long run it might turn out to be – and then to use these effectively to foster civic virtues, surely the arguments for state border controls would lose much of their potency.

But this is no resolution to the ambiguity either. In fact, delegating power to international bodies and migrants is not the only arrangement that may, in principle, count as non-dominating. For instance, one can imagine a world where all states are democratic, no state dominates any other, and movement is regulated by bilateral treaties between states. Surely, in such a scenario states’ interference with movement would be controlled by all those affected by it, insofar as the treaties would be made on their terms. A farfetched idea, for sure. But not more so than the ideal scenario which supports the case for republican cosmopolitanism.

Conclusion

Let me underline something: I did not argue in favour of any policy. Not for closed borders, and not for open ones; not for the weakening of state sovereignty, and not for its maintaining either. All I wanted to show is that the non-domination criterion provides no clear guidelines for deciding which policy to favour. Neither does republican theory in general. In themselves, therefore, they are insufficient to justify any of the policies mentioned. Arguments aimed at justifying these things will have to rely on other, additional moral criteria.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

The work was supported by the Nemzeti Kutatási Fejlesztési és Innovációs Hivatal [PD-143603]; Nemzeti Kutatási Fejlesztési és Innovációs Hivatal [FK-138367].

Notes

¹ The combining is sometimes done implicitly (at least in part). But even where Pettit’s control-based approach is not invoked explicitly, the idea that for an interference to count as non-arbitrary it needs to be controlled by those affected by it is nevertheless explicitly there.

² I identify two root causes behind the indeterminacy of the non-domination ideal. However, I do not rule out that there may be further possible causes.

³ Of course, issues regarding border-situations on the one hand and integration on the other need to be distinguished. For instance, even if it were established that the non-domination ideal supports state border controls, it still might be the case that it also supports the integration policies discussed.

⁴ Not all republicans need face this particular challenge. Arguments based on the value of non-domination alone do not for instance. Only arguments that embrace the republican political outlook more holistically to include the concern for virtue cultivation also. However, it is an open question whether support for non-domination qualifies an argument as republican or if the label applies only if support for non-domination is coupled with additional elements of the political outlook historically associated with republicanism.

⁵ This is not a policy Hoye supports. He merely states that it is a possible solution that follows from the empirical assumptions shared in the literature he examines.

⁶ This is a theoretical argument. Indeed, in most developed countries, migration has had a positive effect on the economy (d’Albis, Boubtane, and Coulibaly 2018). Notable exceptions to this rule are the extensive welfare states of Scandinavia (Nannestad 2007, p. 527–28).

⁷ The evidence regarding the extent of the differences in values held by natives and non-Western immigrants in Europe is mixed (see Alexander and Welzel 2011; Fuchs, Fan, and von Scheve 2020; Norris and Inglehart 2012; Tausch 2016).