Anti-immigrant backlash: the Democratic Dilemma for immigration policy

Abstract

Anti-immigrant backlash has emerged in recent years as a considerable threat to democracy. In many countries illiberal and anti-democratic political forces driven in part by anti-immigrant sentiment have damaged or threaten to damage the proper functioning of democratic institutions. Should policymakers try to avert this threat by implementing more restrictive immigration policy? If they do so, they may expose immigrants to unjust exclusion. If they do not, they may risk democratic dysfunction, even democratic failure. I will call this the Democratic Dilemma for immigration policy. In this paper I argue that this is a hard ethical dilemma that does not lend itself to a straightforward resolution. I propose an analytic and evaluative framework for assessing possible policy responses to the Dilemma to aid policymakers’ as well as the public’s ethical judgement.

Anti-immigrant sentiment has played a major role in the recent advance of illiberal and anti-democratic politics, e.g., far-right populism. As immigration has become an increasingly salient political issue in many contemporary democracies, political forces with illiberal and anti-democratic tendencies have increasingly mobilised around it (Hadj-Abdou, 2021). Alternative für Deutschland (AfD) in Germany, Fratelli d'Italia in Italy, and the radical wing of the Republican Party in the US, including the Trump movement, have exploited anti-immigrant sentiment in pursuit of political programmes which are widely recognized as undermining democratic stability. Michael Blake argues that this can lead to an ethical dilemma:

“democracies can be undermined by a simple refusal, on the part of a sufficient number of citizens, to do the jobs required by democratic life—to sacrifice for the common good, to voluntarily pay taxes, to take the law as authoritative, and so on. What should we do if, as it turns out, there is a limit to the amount of linguistic, cultural, or racial diversity that is acceptable to the current membership of a given society? The Constitution, it is said, is not a suicide pact, and neither is liberal democracy; there is no good in holding fast in the name of diversity if it causes the annihilation of democratic practice. But neither is it palatable to limit immigration simply because those who are already here are willing to act upon malign preferences. There is, I think, no particular good answer to this dilemma—and it may become rather pressing for many countries in the coming years, as countries sending migrants abroad may be culturally and ethnically unlike the countries receiving them.” (Blake, 2014, 535).

Consider, for example, Germany’s response to the so-called “refugee crisis” in 2015. Angela Merkel’s CDU/CSU government initially took a welcoming stance, admitting over one million refugees. This policy, however, also fuelled anti-immigrant sentiment, and – many argue – played an important role in the subsequent advance of AfD (Joppke, 2021, 137). In the following years, CDU/CSU changed its approach to immigration policy supporting the infamous “EU-Turkey deal” to reduce the number of refugees in Europe. This is partly explained by CSU’s opposition to a permissive asylum policy, but also to a considerable extent, by the perceived threat of AfD’s rise (Hertner, 2022, 475–76). AfD is also seen by many as a threat to democracy; numerous voices called upon the Federal Office for the Protection of the Constitution (BfV) to monitor AfD as a potential threat to the constitutional order. If anti-immigrant backlash gives rise to democracy-threatening political forces, could this not justify adopting a more restrictive immigration policy to halt its advance? On the other hand, both the EU-Turkey deal as well as Europe’s turn to more restrictive immigration policy after 2015 have been criticised as causing and enabling serious and unjust harm to refugees (Hillier-Smith, 2020). Should democracy be defended at such a price?

Like Blake, I believe that these kinds of cases present a dilemma, and under some conditions, even a hard ethical dilemma (Bauböck et al., 2022) which I will call the Democratic Dilemma for immigration policy. Unlike Blake, however, I believe that there are at least some good answers to it. In this paper I propose an analytic and evaluative framework for understanding the nature of the Democratic Dilemma and evaluating policy measures that may be taken to prevent or respond to it. This paper contributes to a growing literature in immigration ethics on the challenge of illiberal anti-immigrant attitudes, a problem hitherto largely neglected in the field (Amantini, 2022),¹ by offering a systematic analysis of the Dilemma, drawing on research not only in immigration ethics, but also democratic theory, and empirical social and political science, and by identifying the policy responses which are best suited to enhance justice overall even under the non-ideal consideration under discussion. The structure of the paper is as follows. In the first section, I introduce the Democratic Dilemma. Secondly, I argue that it is best understood as a hard ethical dilemma for immigration policy. In the third and fourth sections, I turn to possible policy measures against the Dilemma. First, I discuss options for preventing the Dilemma from arising in the first place, and second, I discuss policy responses in cases when the Dilemma as already arisen. The last section concludes.

The Democratic Dilemma

In this section I begin by clarifying what I mean by the Democratic Dilemma. The Dilemma occurs when unjust immigration policy has a reasonable chance to counteract democracy-threatening anti-immigrant backlash. In such cases, a choice is to be made between two bad outcomes, i.e., subjecting immigrants and would-be immigrants to injustice to try to halt the advance of democracy-threatening forces, and upholding immigration justice by putting democracy, with all its benefits, in jeopardy. In other words, in such cases, two equally important goals of immigration policy cannot be satisfied at the same time: immigration justice and democracy. I will elaborate momentarily. But first, some explanation is due about my focus on immigration policy.

Something like the Democratic Dilemma may present itself in various forms to different actors. Activists and NGOs advocating for just immigration policy in a society ridden with anti-immigrant sentiment, voters deliberating whether to cast a vote for a moderately anti-immigrant candidate to avoid an illiberal populist takeover, politicians considering what immigration policy to endorse in their campaign against anti-immigrant populist opponents may all face an ethical dilemma in one form or another. However, policymakers, i.e., elected or appointed decision-makers charged with creating binding rules governing border control, visas, work permits, residence permits, asylum, and so on, play a particularly central role in shaping immigration in a given country, thus a focus on policymaking seems appropriate.

Still, one may find this objectionable. Alison Jaggar rightly critiques immigration ethicists when they “implicitly situate themselves as policy makers within wealthy states facing a problem of uninvited intruders” (Jaggar, 2020, 106) thereby reinforcing an unwarranted Westphalian and neo-colonialist approach to immigration. This critique certainly has merit, however, considerations for the ethical evaluation of immigration policy may not only, or even primarily, be addressed to policymakers as advice, but also, and perhaps more importantly, to the democratic public in whose name policymakers act and who hold them to account. Ethical considerations about immigration policy are important for us as citizens to evaluate the policies our own state enacts in our name, and to decide whether we need to advocate for change. Thus, the following argument is offered to aid not only policymaking, but also citizens’ judgement.

Turning to the Democratic Dilemma, I claimed that it occurs when democracy-threatening anti-immigrant backlash brings immigration justice into conflict with democracy. What does this mean? In immigration ethics, there are numerous views about immigration justice, from the open borders view (Carens, 2013; Sager, 2020) to views emphasizing self-determination and the right to exclude (Miller, 2016; Song, 2019), each defining the range of just immigration policies differently. In the following argument, I do not assume any particular theory of immigration justice. I only make one stipulation, i.e., that immigrants and would-be immigrants have at least some non-trivial moral claims which immigration justice requires receiving states to honour.² How can immigration justice, however one understands it, come into conflict with democracy? Note that I do not claim that immigration justice and democracy inherently conflict with each other. Rather, the conflict arises from very specific circumstances, i.e., democracy-threatening anti-immigrant backlash.

Although the concept of backlash is widely used in scholarship immigration and beyond,³ there is no general consensus on its precise definition (Kustov, 2023, 1185). Here I define democracy-threatening anti-immigrant backlash, or simply backlash for short, as engagement in or support for anti-democratic politics within certain parts of the population, partly or wholly as a reaction to perceived excessive, or otherwise undesirable, immigration. Think of the aforementioned examples of far-right populist parties whose support, according to empirical evidence, is partially driven by anti-immigrant sentiment (Berman, 2021, 76; Mudde, 2019; 101; Norris & Inglehart, 2019, 205). I will say more about this evidence later. For now, let me clarify the sense in which such backlash is democracy-threatening.

The claim that backlash is democracy-threatening might seem like an overstatement. Only in the most extreme cases does it threaten democratic institutions and procedures, e.g., elections, directly; think of the attack on the US Capitol on January 6, 2021, executed by parts of a movement centred on a staunchly anti-immigrant populist politician. In most cases, far-right populists, for example, compete in elections and, at least nominally, respect democratic and constitutional formalities. Yet most democratic theorists would agree that there is more to properly democratic politics than adherence to merely formal rules and procedures. Even so-called agonist theories of democracy, emphasizing the conflictual nature of democratic politics, argue that democracy requires a shared commitment to “domesticating political conflict” and treating political opponents as adversaries rather than enemies (Mouffe, 2000). Others emphasize commitment to pluralism, the political equality of opponents, and basic norms of civility in political interaction (Christiano, 2008; Dahl, 1989; Müller, 2021; Talisse, 2021; Wilson, 2019) as requirements of properly democratic politics, while yet others offer even richer conceptions of the substance of democratic self-government (Brettschneider, 2007).

As with immigration justice, I do not commit to any particular theory of democracy here. For whatever theory one adopts, backlash does threaten properly democratic politics. This is not to say that anti-immigrant sentiment is inherently anti-democratic. Political actors may fully respect the requirements of properly democratic politics while opposing immigration. But backlash often manifests as populist and authoritarian politics which can undermine pluralism (Müller, 2016) and civility (Lenard, 2022) and often involves racist and discriminatory politics incompatible with a commitment to political equality (Hajnal & Abrajano, 2015). Of course, democracy can exhibit a certain level of resilience (Volacu & Aligica, 2023); democratic norms, let alone mere democratic formalities, e.g., elections, can survive such onslaughts. But the protracted undermining of the requirements of properly democratic politics can leave democratic institutions hollowed out and vulnerable to autocratic state capture (Levitsky & Ziblatt, 2018; Przeworski, 2019; Wolkenstein, 2023).

Democracy may not spectacularly collapse overnight due to backlash, but anti-democratic politics, fuelled partly by anti-immigrant sentiment, can seriously damage it, to an extent which, many argue, is cause for alarm. Not only because democracy is a good in itself, necessary for the normative legitimacy of state power (Buchanan, 2002; Christiano, 2008), the political equality of citizens (Schemmel, 2021; Wilson, 2019), and so on,⁴ but also because it is a prerequisite of many societal goods, including immigration justice. Anti-democratic actors, e.g., far-right populists, in power can and do introduce unjust anti-immigrant policies; think of the US’s 2017 Muslim ban, or the restrictive and often inhumane asylum policy of Hungary after 2015 (Hungarian Helsinki Committee, 2018). One may object that well-functioning democracies are perfectly capable of treating immigrants in abhorrent and inhumane ways too; consider Australia’s widely criticised refugee policy. However, as Sen (1999) notes, the chances of successfully counteracting such practices are greatest in a well-functioning democracy where popular demands, including the demands of immigrants, can exert pressure on public decision-making.

Thus, if unjust immigration policy has a reasonable chance of counteracting democracy-threatening anti-immigrant backlash, e.g., by taking some of the wind out of the sail of backlash and dissuading at least some people form supporting anti-democratic politics, then policymakers may face a hard choice between upholding immigration justice or democracy. This is, of course, a big if. Sometimes immigration policy may have no bearing on whether backlash occurs or how it develops (Kustov, 2023). Indeed, Italy’s 2017 decision to make a controversial agreement with Libya to reduce the influx of refugees did little to halt the advance first of Salvini’s Lega, and then Meloni’s Fratelli d’Italia, for example. But the picture is more complex. As van Spanje and de Graaf (2018) show, mainstream parties can reduce anti-immigration parties’ support by adopting their policy positions, while simultaneously ostracizing them (cf. Spoon & Klüver, 2020).⁵ Of course, uncertainty reigns in politics, and it is often very hard to determine if a strategy in a particular context has a decent chance of succeeding. But suppose that in a given situation, policymakers do have good reasons to think that unjust immigration policy can counteract democracy-threatening anti-immigrant backlash. Does this really constitute a hard ethical dilemma for immigration policy as I claim? I will turn to this question in the next section.

A hard ethical dilemma

In this section I argue that the Democratic Dilemma is a hard ethical dilemma for immigration policy. I borrow this term from Bauböck et al. (2022) who define it as a conflict between comparably worthy moral values without a straightforward resolution. Such dilemmas are grounded in specific empirical circumstances and embedded into particular political institutions. Hard ethical dilemmas, as understood here, are not universal or unavoidable. Thus, to say that the Democratic Dilemma is a hard ethical dilemma is not to say that immigration justice and democracy always conflict or are theoretically incompatible. As I noted earlier, I do not believe that any such theoretical conflict exists. Rather, the Democratic Dilemma is a hard ethical dilemma precisely because it emerges in the specific context of politicized immigration and ascendant anti-democratic politics driven, in part, by anti-immigrant sentiment.

This also means that one can acknowledge the Democratic Dilemma as a hard ethical dilemma even if one has strong cosmopolitan commitments. Indeed, even Joseph Carens, perhaps the most influential defender of open borders, who holds that his view about immigration justice derives from the basic commitments of a liberal democratic society (Carens, 2013), acknowledges that “[f]rom the perspective of a realistic morality, the likelihood of a backlash is one possible consequence that should be taken into account in assessing the desirability of adopting a particular policy.” (Carens, 1996, 160) As I argued in the previous section, these consequences may indeed be dire. That is, even if one holds that, in the abstract, democracy requires quite open and permissive immigration policy (Bauböck, 2020), if not open borders (Carens, 2013), one can still acknowledge that under the specific circumstances of backlash, the two values can conflict.

Still, the mere existence of democracy-threatening anti-immigrant backlash is not sufficient for generating a hard ethical dilemma. Let me elaborate by considering a series of hypothetical examples.⁶ First:

Low risk. Backlash occurs in a state resulting in a fringe anti-immigrant party. If nothing is done, there is a low chance that they successfully stage a violent coup and overthrow the democratic order. Immigrants entering the state are mainly refugees fleeing persecution. Restricting asylum policy would expose them to a high risk of human rights violations while reducing the threat to democracy from low to very low.

In a case like Low risk, there is no dilemma at all. Refugees face persecution, and their human rights are at serious risk; their claims to entry seem very strong on almost all theories of immigration justice. At the same time, the chance that the fringe anti-democrats can seriously harm democracy is low. Policymakers do not face a hard choice, let alone a dilemma: refugees clearly should not be exposed to the high risk of serious human rights violations to avoid a low risk of democratic dysfunction. This suggests that the relevant value conflict only constitutes a hard ethical dilemma if the risks to democracy and immigration justice are comparable. Consider:

High risk. Backlash occurs in a state resulting in a large and powerful xenophobic populist movement posing a great and immediate threat to democratic stability. If nothing is done, there is a high chance that they soon take power and seriously damage both immigration justice and democracy. Immigrants entering the state are mainly refugees fleeing persecution. Restricting asylum policy would expose them to a high risk of serious human rights violations, while reducing the threat of backlash to democracy from high to low.

In High risk, there does seem to be a hard ethical dilemma. In this case, honouring refugees’ valid claims to entry, and thereby upholding immigration justice, requires sacrificing the proper functioning of democracy, including democracy’s ability to support immigration justice in the future, while preserving democracy requires sacrificing immigration justice. Something like High risk may be thought of as the ideal type of the Democratic Dilemma, although such a clear case is unlikely to occur in real life.

But one may doubt that even in a case like High risk there exists a hard ethical dilemma, for a genuine dilemma would involve a hard choice between two options, none of which is obviously preferable to the other. Here, however, it may be argued, one of the choices is obviously preferable; for making immigration policy more restrictive to avoid backlash is a non-starter. As Ryan Pevnick recently put it, “it is audacious, even indecent, for existing citizens to (a) insist that they will, because of outgroup hostility, undermine democratic institutions if too many immigrants arrive and then (b) appeal to the resulting bad consequences to justify restrictions. They are, one might say, taking democracy hostage to win a policy battle over immigration.” (Pevnick, 2024, 336–37) It is categorically unacceptable, one may continue, for policymakers to aid taking democracy hostage in this way. As Blake (2020) notes, this would, in effect, give “the bigot a veto” over immigration policy. The choice, therefore, is obvious; even if anti-immigrant backlash threatens democracy, implementing unjustly restrictive immigration policy should never be the answer.

I agree that citizens’ objectionable attitudes should not dictate immigration policy. But this hardly means that policymakers can simply ignore these attitudes. Some have argued, for example, that popular attitudes about immigration should count as feasibility constraints on immigration policy (Del Savio, 2020; Ruhs, 2022). However, I am unsure if backlash renders just immigration policy unfeasible in a meaningful sense of the term; rather, I believe that under backlash such immigration policy, while feasible, may be very costly for democracy – a value comparable to that of immigration justice. If one agrees with Carens that such costs “should be taken into account in assessing the desirability of adopting a particular policy” (Carens, 1996, 160), then one may well face a dilemma where upholding immigration justice and democracy each have intolerable costs. The Democratic Dilemma poses a problem precisely because under some specific circumstances doing what immigration justice dictates and keeping democracy in good working order at the same time may not be simultaneously achievable objectives. To say that there is a hard dilemma here is not to deny but to affirm that making immigration policy more restrictive would be deeply objectionable.

Of course, once again, the role of uncertainty in politics needs to be emphasized. Given the many variables and uncertainties surrounding any real-life situation, it may rarely, if ever, be possible to clearly and unambiguously determine whether the Democratic Dilemma has in fact arisen, e.g., whether the boundary between Low risk and High risk has been crossed. I called High risk an ideal type of the Democratic Dilemma, and the description is fitting. Particular real-life cases may approximate clear cases like High risk to a greater or a lesser extent, and they constitute a hard ethical dilemma only to this extent.

Preventing the dilemma

Having argued, in the previous section, that the Democratic Dilemma is in fact a hard ethical dilemma for immigration policy, in this and the next section I turn to possible policy measures that may be taken to respond to the Dilemma or prevent it from arising in the first place. Let me begin with prevention. As I noted in the previous section, the Democratic Dilemma arises only under specific empirical conditions. When wealthy Scandinavians migrate between Norway and Sweden, no tension between immigration and democracy seems to arise, for example. One course of action policymakers may take, then, is to try to eliminate the preconditions of the Dilemma, thus preventing it from arising in the first place.

The Democratic Dilemma has two preconditions. First, there need to be immigrants or would-be immigrants with sufficiently weighty valid claims on the receiving states which they must honour under immigration justice, e.g., refugees claiming asylum, and second, democracy-threatening anti-immigrant backlash. Accordingly, there are two possible preventative measures, i.e., pre-empting immigrants’ valid claims or preventing backlash. Let me discuss each in turn. By pre-empting immigrants’ claims I mean making it the case, through permissible means, that immigrants do not have valid claims on receiving states. For example, to avoid a case like High risk, policymakers may try to reduce the number of refugees globally, e.g., by pursuing a foreign policy aimed at ensuring that fewer states persecute their citizens and that human rights are better protected worldwide.

Of course, claims pre-emption may be exceptionally difficult or impossible depending on one’s theory of immigration justice. Theories endorsing open borders may allow only for few or no cases of claims pre-emption. But even if immigration justice allows for claims pre-emption, it may be a practically difficult matter. Not only do immigrants’ claims differ in terms of their strength and basis—some immigrants are refugees, others are labour migrants, yet others seek family reunion—such that pre-empting all or most of them may be an extraordinarily complicated task, but the success of pre-emptive efforts is rarely up to policymakers in a single state alone. For example, strengthening human rights worldwide to reduce refugee numbers requires cooperation between states as well as non-state actors. For these reasons, pre-emptive strategies may only be of limited utility to policymakers.

Is preventing backlash a more promising strategy? In many countries, of course, the opportunity for preventing backlash has already passed. Anti-immigrant populists not only have appeared, but already won office in many places, e.g., the US and Italy. Still, these preventative measures are worth discussing, first, because in some countries there may still be a chance of prevention, and second, because many preventative measures can also help mitigate backlash once it occurs. But before asking how to prevent democracy-threatening anti-immigrant backlash, we should ask what causes it. This is the subject of a large and ongoing debate in political and social science which I cannot hope to settle here. Still, considering some of the empirical literature and their possible policy implications may be instructive.

Earlier I quoted Blake who suggests that backlash stems from some citizens’ limited tolerance for “linguistic, cultural, or racial diversity” induced by immigration. Diversity has, in fact, increased in recent decades in many democracies due to immigration (Norris & Inglehart, 2019, 179–180). Although immigration policy in Western democracies has generally become more restrictive, especially for low-skilled migrants (Baycan-Herzog, 2021; Joppke, 2021), the number of migrants has increased; in Europe from around 20 million in 1990 to over 40 million in 2020, and in North America from over 20 million in 1990 to over 50 million in 2020, the majority of migrants coming from the Global South (McAuliffe & Triandafyllidou, 2021). During the same time, support for anti-democratic political forces with a strong anti-immigrant agenda, has also dramatically increased (Mudde, 2019, 21). However, many other things happened in these decades, and anti-democratic parties today are usually not single-issue parties focused exclusively on immigration. Is there evidence that support for anti-democratic politics is in fact driven by anti-immigrant sentiment?

The causes and drivers of anti-democratic politics in contemporary democracies are complex and certainly not reducible to immigration alone (Berman, 2021; Norris & Inglehart, 2019; Przeworski, 2019). Still, there is considerable evidence that immigration and immigration-attitudes play a non-negligible role. Anti-immigrant attitudes have been consistently found to be strong predictors of support for far-right and populist parties in Europe (Arzheimer, 2009; Ivarsflaten, 2008; Kehrberg, 2015) and they were also strong predictors of support for Trump in the 2016 election (Hooghe & Dassonneville, 2018). The share of immigrants within the population, especially non-Western immigrants, have been shown to strongly correlate with support for anti-immigrant and far-right parties (Dustmann et al., 2019; Halla et al., 2017; Otto & Steinhardt, 2014; see Pevnick, 2024). Anti-immigrant attitudes have been found to strongly correlate with authoritarian attitudes and low support for democracy (Bartels, 2020; Norris & Inglehart, 2019, 195). This suggests that many of those who support anti-democratic, e.g., far-right populist, politics in contemporary democracies, do so, to a non-negligible extent, out of anti-immigrant sentiment. But do these backlashers really object to increased diversity as Blake suggests?

A prominent line of argument in immigration ethics supports the diversity-based explanation. Drawing on Robert Putnam’s (2007) influential study, scholars like David Miller and Sarah Song argue that immigration-induced diversity tends to generate anti-democratic, or at least democracy-incompatible, attitudes because it undermines social trust and solidarity which are necessary for the proper functioning of democracy (Miller, 2016, 64; Song, 2019, 68). Although Song and Miller do not invoke this argument to explain backlash, Patti Lenard (2012) argues that the lack of social trust in diverse societies can pose a serious threat of democratic dysfunction and may lead such societies down a path to become “severely divided” along ethno-cultural lines. However, as Holtug (2021) persuasively argues, the overall empirical evidence about immigration-induced diversity reducing trust is at best mixed. Furthermore, the effect of diversity appears to be conditioned by a number of factors, including inequality (Larsen, 2013) and residential segregation (Uslaner, 2010). When diversity reduces trust and solidarity, this is often because diverse groups occupy unequal and disadvantaged social positions and have sparse interaction with the rest of society enabling their stigmatization as parasitic or dangerous, i.e., untrustworthy Others.

Furthermore, the argument from diversity and social trust also fails to explain what may be called the “distribution” of backlash. Anti-immigrant sentiment and support for anti-immigration parties is higher among those experiencing relative economic deprivation and relative prosperity (Jetten et al., 2015). In more recent works, Blake too notes that the rise of democracy-threatening backlash in the US coincides with a marked decline in the living standards of white working-class Americans (Blake, 2020, 137; cf. McQuarrie, 2017). Part of what gives rise to backlash, he argues, is the loss of both stable economic prospects as well as social status for such groups; a kind of social humiliation which manifests as radical nativist politics (Blake, 2021). Jetten (2019) argues that a similar sense of status threat, i.e., anxieties about the prospect of losing status and wealth to groups perceived as socially inferior, underlies anti-immigrant sentiment among affluent populations. Kustov (2023) also presents evidence that backlash is best explained by status threat, rather than mere intolerance of diversity or cultural difference. Of course, status anxiety, due to worsening economic and social prospects, frequently manifests as racialized prejudice and ethnic antagonism toward culturally different populations (Mudde, 2019, 101). That is, democracy-threatening anti-immigrant backlash often takes shape as characteristically “white backlash” (Hajnal & Abrajano, 2015).

Finally, we should keep in mind the “supply-side” explanations of backlash. Individuals do not form anti-democratic anti-immigrant attitudes independently of their social and political environment. Research suggests that democracy-threatening backlash is not best explained by growing rates of immigration and diversity alone, but crucially, by the salience of immigration as a public issue and its public perception (Dennison & Geddes, 2019; Norris & Inglehart, 2019, 194), which is shaped both by media messaging and political communication (Dennison, 2019). Thus, thinking of the causes of backlash, one must not lose sight of the fact that backlash is often deliberately promoted by political actors, e.g., populist leaders and right-wing parties, for political gain (Hajnal, 2021).

This suggests that backlash is the result of multiple factors, including the rise of immigration-induced diversity, socio-economic changes inducing status threat within certain parts of the population, and media messaging and deliberate political attempts to channel this sense of threat through anti-immigrant sentiment to anti-democratic politics (Bonikowski, 2017). This indicates several possible policy interventions for the sake of preventing backlash. Reducing diversity-inducing immigration may be one option; studies show that high levels of immigration do correspond to heightened anti-immigrant sentiment, at least in the short term (Claassen & McLaren, 2022; Kustov, 2023). However, this would likely involve discriminatory immigration policy, e.g., the targeted exclusion of would-be immigrants based on culture, religion, or race. Thus, given that other options are also available, this strategy should generally be avoided.

Another strategy may target the public perception of immigrants, focusing on integration policies aimed at the transformation of both immigrants’ and citizens’ attitudes and practices to enable their participation in social cooperation as equals (Klarenbeek, 2019). Fostering optimal contact between citizens and diverse immigrant groups may play a crucial role in this process, as the so-called Intergroup Contact Theory holds (Pettigrew & Tropp, 2006). However, it may be equally important to reduce levels of inequality and strengthen distributive justice, thus insuring a greater portion of the population against precarity and status loss. Levels of inequality seem to be linked with the prevalence of anti-immigrant sentiment (Esses et al., 2001; Jetten et al., 2015), and as Holtug (2021) argues, following Larsen (2013), lower levels of inequality and a well-functioning welfare state can contribute to a less hostile public perception of both low-skilled, low-income immigrants and the domestic poor. Finally, to safeguard against anti-democratic actors deliberately promoting backlash policymakers may try to strengthen the rule of law, ensuring that even if anti-democratic actors win office, they cannot use their power to the detriment of democracy. Some even argue that if such actors are “unreasonable” in the sense of openly rejecting and posing a serious threat to fundamental liberal democratic values, a liberal democratic state may legitimately infringe upon their political liberties to contain and neutralize them (Müller, 2016; Quong, 2004).

This short discussion could provide neither a conclusive explanation of backlash nor an exhaustive list of possible preventative measures. Still, it illustrates the potential complexity of the relevant causal mechanisms and the wide range of policy tools available for preventing backlash. Many of these tools, e.g., good integration, reducing inequality, and strengthening the rule of law, may also help mitigate backlash once it has already occurred. Still, as with claims pre-emption, these policy tools are limited. First, there must be a limit to how much policymakers may legitimately try to shape citizens’ attitudes and preferences without objectionably infringing on the democratic process. Policymakers authorized by and working on behalf of the democratic public have no unlimited mandate to tailor citizens’ preferences to their own liking. Second, any attempt to prevent backlash may suddenly be undermined through external shocks, such as a sudden influx of refugees or a global economic crisis heightening economic uncertainty and, with it, popular anxieties about immigrants. In such cases, despite policymakers’ efforts to prevent it, the Democratic Dilemma may nonetheless arise.

Responding to the dilemma

In the previous section, I examined the possible policy measures for preventing the Democratic Dilemma from arising in the first place. In this section I discuss the possible policy responses that may be taken once the Dilemma has already arisen, and how these responses may be ethically evaluated. Before I begin, let me reiterate that whether the Democratic Dilemma has arisen is itself usually a matter of uncertainty. Clear cases, such as High risk are rarely forthcoming in real life. The wide-ranging complexities that need to be considered in the evaluation of any particular policy decision in any given situation cannot be accommodated within a general theoretical discussion such as this one. Thus, the following argument is bound to be abstract and schematic. Still, I believe it is instructive; examining clear, although stylized cases of hard ethical dilemmas can help us gain insight about more complex real-life scenarios.

The Democratic Dilemma occurs when backlash brings immigration justice and democracy into conflict. This means that once the Dilemma has arisen, any policy response involves considerable moral cost, either in terms of immigration injustice, or risk to democracy. This is clearly illustrated by High risk. However, for the sake of a more nuanced discussion, it is worth considering a more complex example first:

Complex risk. Backlash occurs in a state resulting in a large and powerful xenophobic populist movement posing a great and immediate threat to democratic stability. If nothing is done, there is a high chance that they soon take power and seriously damage both immigration justice and democracy. Immigrants entering the state are mainly refugees fleeing persecution and labour migrants seeking better jobs. Restricting asylum policy would expose refugees to a high risk of serious human rights violations, while reducing the threat to democracy from high to low. Restricting labour migration would expose labour migrants to unjust but moderate economic hardship, while reducing the threat to democracy from high to moderate.

In Complex risk, policymakers have essentially two options: restricting asylum or restricting labour migration. The first involves a great cost to immigration justice, but also a great benefit to democracy, while the second a moderate cost to immigration justice and a moderate benefit to democracy. How should we ethically evaluate the policy choice made in such a situation?

First, there are general ethical constraints on policy choice which apply to all cases. Non-compliance with these constraints clearly condemns a policy choice as beyond the moral pale. For example, policymakers cannot implement unnecessarily and disproportionately cruel or unjust policies, e.g., needlessly separating migrant children from their parents, denying sea rescue to migrants in unseaworthy vessels, or introducing wrongfully discriminatory immigration policies, e.g., “Muslim bans,” even if, due to some strange circumstances, these were sure to secure democracy. Second, there are also special constraints which apply only in some cases. For example, duties of reparation for historic injustice may generate special obligations for perpetrator states, e.g., former colonizers, against immigrants from states where the historic injustice occurred (cf. Souter, 2022). Immigrants’ claims themselves may be sources of special constraints. In Complex risk, for example, asylum seekers may argue that they have a special claim on the state to protect their human rights which should enjoy absolute priority compared to the claims of labour migrants. Therefore, restricting asylum instead of restricting labour migration is unconscionable.⁷ Whether this is so, and what special constraints apply more generally, of course, depends on one’s theory of immigration justice, on which I remain neutral here.

Let us suppose that neither asylum restriction nor labour restriction violates any general or special ethical constraints on policy choice. If this is so, which one is the preferrable choice from an ethical point of view? There are three possible answers: first, restricting asylum is preferable because it minimizes risk to democracy; second, restricting labour migration is preferable because it minimizes immigration injustice; third, the two are on a par, no ethical consideration recommend either choice and policymakers are free to make up their mind. Which answer is correct? This is far from obvious; there are arguments worth considering for each one.

In defence of restricting asylum for the sake of minimizing risk to democracy, one may emphasize that, as I discussed above, democracy is not only a good in itself, but also a prerequisite for a range of societal goods including immigration justice. The prospects of immigration justice and justice in general are worse if authoritarian ethno-nationalists grab power. But democracy can only function if citizens, including those with anti-immigrant sentiment, cooperate. If nothing else, sheer prudence dictates not alienating these citizens further, and seeking some sort of compromise with them, if only to preserve the prospect of restoring democratic stability in the future (Knobloch & Mieth, 2021; Macedo, 2020). Therefore, securing democracy should be our first priority, even if this means greater (immediate) damage to immigration justice.

In defence of restricting labour migration for the sake of minimizing immigration injustice, one may point out that immigration injustice is likely to disproportionately harm vulnerable immigrant groups, e.g., refugees or immigrants from the Global South. Although democratic dysfunction may also harm them, this harm is likely to be dwarfed compared to the harm of their exclusion from rights, opportunities, and even the state’s territory. Refugees fleeing persecution and labour migrants fleeing life-threatening poverty may rightly prefer the risk of democratic dysfunction in their destination states to the disadvantages they face in their state of origin.

Furthermore, just as democratic dysfunction can harm immigration justice, immigration injustice can harm democracy. Preserving democracy through unjust immigration policy creates moral hazard by encouraging the pursuit of political goals by anti-democratic means. Moreover, the very value of democracy may be undermined by deep injustices. For example, apartheid South Africa might have had formally democratic institutions, at least at some points in its existence, but those institutions were used to uphold an oppressive social system. On no reasonable view does such “Herrenvolk democracy,” i.e., a system which is “democratic for the master race but tyrannical for the subordinate groups” (van den Berghe, 1967, 19), instantiate the values, e.g., normative legitimacy or political equality, that make democracy worthwhile. The same must apply to excessive injustice committed by democracy beyond its borders. A democracy engaged in genocidal colonialism, for example, is not a democracy worth having, however one thinks of the value of democracy. Thus, if immigration injustice reaches sufficiently high levels, then trying to save democracy through unjust immigration policy becomes a self-undermining pursuit, for it undermines the very value that makes democracy worth having.

Finally, in defence of the third option, i.e., the ethical parity of restricting asylum and restricting labour migration, one may argue that neither argument above is decisive, although they both offer considerations worth taking into account. Faced with such complex and momentous choices as the ones presented by the Democratic Dilemma, it must be up to policymakers’ reason and conscience to decide the right course of action. Ethical theory reaches its limits here and must give way to practical wisdom and political prudence. And perhaps even if there is an ethically correct choice, given the especially high stakes, the numerous variables, and great levels of uncertainty, the wrong choice may at least be excusable insofar as it involves a bona fide weighing of all relevant considerations.

Which of these arguments is most convincing? As this short discussion already indicates, a full answer would require a much more thorough engagement with a number of foundational questions both in immigration ethics and democratic theory. Here I can only venture to provide a tentative answer which is as follows. Under contemporary global conditions, deep immigration injustice is likely to affect the fundamental and immediate interests of immigrants and to undermine democracy’s value. Contemporary democracies implement immigration policy against the background of deep and arguably unjust global inequalities. Immigrants, especially from the Global South, often suffer not only from great material disadvantage, but also from status harm and disempowerment within global political and economic systems (Akhtar, 2022). Under these conditions, restrictive immigration policy is particularly prone to entrenching injustice and suffering (Sager, 2020), undermining immigrants’ liberty (Kukathas, 2021), and their equal standing both within and beyond the state’s borders (cf. Sharp, 2022). That is, restrictive immigration policy, especially in powerful states of the Global North, upholds an international system which Carens famously compared to feudal class privilege (Carens, 2013, 226). A democracy which rests on such foundations comes dangerously close to being a type of “Herrenvolk democracy,” a democracy not worth having.

In contrast, while democracy-threatening anti-immigrant backlash is a serious problem, it rarely affects the immediate interests of either citizens or immigrants; the Dilemma usually does not involve the threat of democracy’s immediate catastrophic collapse. Democratic dysfunction manifests through complex mechanisms over time, usually leaving policymakers with numerous options to tackle it without making immigration policy gravely unjust. This suggests that in most cases policymakers should try to minimise immigration injustice. Given the levels of disadvantage and vulnerability most immigrants suffer from today, policymakers in contemporary democracies should generally prioritize preserving immigration justice and seek ways to counteract backlash and democratic backsliding in other ways, e.g., the continued implementation of some preventative measures, such as good integration, reducing inequality, and strengthening the rule of law, even if this means a greater risk to democracy. That is, in a case like Complex risk, they should choose restricting labour migration, rather than restricting asylum, even though this choice too has a high moral cost, as it leaves democracy in considerable risk.

But what about a case like High risk, where the only option to avert great and immediate threat to democracy is restricting asylum? Here I am not as sure about the preferability of minimising immigration injustice over minimising risk to democracy. Serious and long-term damage to democracy can ultimately preclude the pursuit not only of immigration justice, but justice more generally. Does this mean that in High risk saving democracy should be prioritized? Perhaps under reasonably just global conditions it would have that implication; if great numbers of immigrants did not suffer from systemic disadvantage, deprivation, vulnerability, and violence, which unjust immigration policy entrenches and amplifies, then perhaps democracy, as an important good in itself and a prerequisite of many societal goods, should be favoured. However, existing global conditions are not reasonably just; whatever choice policymakers end up making, they risk deep and enduring injustice, whether effected by democratic dysfunction or unjust immigration policy. Therefore, I am inclined to think that in extreme cases, such as High risk, ethical considerations do not recommend either restricting asylum or restricting labour migration. Unlike in less extreme cases, policymakers can, but are not required to, prioritize democracy.

To summarize, if the Democratic Dilemma arises, policymakers face a choice between sacrificing either some immigration justice for the sake of greater security for democracy or vice versa. Although either choice involves, potentially grave, moral costs, this does not mean that they are on a par or, as Blake put it, that there is “no particular good answer to this dilemma” (Blake, 2014, 535). I argued, albeit tentatively, that given the background conditions of unjust global inequalities, policy choices responding to the Democratic Dilemma should generally aim at minimising immigration injustice, rather than minimising risk to democracy. Cases like High risk potentially constitute exceptions, but, as noted earlier, they are rarely, if ever, occur in real life.

Conclusion

In this paper I discussed a question about the ethics of immigration: if unjust immigration policy has a reasonable chance to avert or counteract democracy-threatening anti-immigrant backlash, should it be implemented? In other words, should immigration justice ever be traded off for democratic stability? I argued that, in some cases, this constitutes a hard ethical dilemma for immigration policy involving a conflict between two comparably important values, i.e., immigration justice and democracy. Policymakers may try to prevent the Dilemma from arising in first place, e.g., through claims pre-emption, good integration, reducing inequality, and strengthening the rule of law, but there are both practical and ethical limits to these preventative measures.

Once the Democratic Dilemma arises, a choice needs to be made between greater immigration injustice or greater risk to democracy. I argued that in most cases, immigration justice should be prioritized. Not because it matters more than democracy – there is a dilemma precisely because they matter the same – but because immigration injustice likely affects immigrants’ most immediate interests and entrench and amplify their systemic disadvantage, deprivation, and vulnerability. In addition, excessive immigration injustice can undermine the very value of democracy itself. A democracy that regularly commits horrible atrocities either against its own citizens or against outsiders may not be a democracy worth having. Indeed, I believe that democratic citizens of the countries of the Global North should consider whether subjecting mases of immigrants to forced encampment, family separation, or lethal danger at sea, as it is often done in the US, Europe, and Australia, for example, do not already count as such horrible and regular atrocities. That said, I also believe that in the most extreme cases, when the threat to democracy is great and immediate, and possible policy responses as limited, e.g., in High risk, the ethical picture seems less clear and may allow either choice.

However, such extreme cases are rarely, if ever, forthcoming in real life. Indeed, real-life cases only ever approximate clear instances of the Democratic Dilemma to a greater or lesser extent, and the considerations discussed here are relevant to them only to this extent. Democracy-threatening political forces, such as far-right populism, are rarely single-issue anti-immigrant movements, and restricting immigration is probably never sufficient to diffuse them. Still, given the central role of the politicisation of immigration in illiberal and anti-democratic movements today, these considerations do have considerable significance to immigration governance. For this reason, I believe, this paper contributes not only to ongoing debates in immigration ethics but may also aid the ethical judgement of both policymakers and the public to whom these policymakers are accountable.