Skip to main content

Why liberal nationalism does not resolve the progressive’s trilemma: Comment on Will Kymlicka’s article: “Solidarity in Diverse Societies”

An Erratum to this article was published on 08 July 2016

Why liberal nationalism does not resolve the progressive’s trilemma

Solidarity is an elusive ideal, but it is worth chasing it nevertheless. In his essay, Will Kymlicka invokes solidarity within the nation, which has a progressive potential when it includes culturally diverse groups into a national We and when it can be mobilized against class privilege. Kymlicka is concerned that these two goals may conflict with each other and hopes that national solidarity can mitigate the conflict.

A trilemma, not a dilemma

The first thing to note is that progressive appeals for solidarity have often targeted outsiders as well as insiders. “Long live international solidarity” was the most popular slogan at the left wing street manifestations of my student years. I agree with Kymlicka that it is worth retrieving the old progressive ideal of solidarity when thinking about how to respond to cultural diversity and increasing social inequality. But maybe we should keep it open for both versions of solidarity: that among insiders and that with outsiders?

When we apply this idea to the progressive’s dilemma, it turns into a trilemma, because solidarity is then also at stake when considering whom to admit. The trilemma is between openness for immigration, multicultural inclusion and social redistribution: How is it possible to achieve social solidarity in culturally diverse states with fairly open borders? Openness for immigration is a traditional progressive ideal just as much as the other two and especially the admission of refugees has often been advocated in the language of solidarity. Of course, relatively open borders can also be supported on other grounds, including utilitarian ones (because sending states, immigration states and migrants will be better off) or libertarian ones (because individuals should be free to cross borders unless states need to restrict immigration to protect freedom itself). And of course, pleading for the admission of immigrants on grounds of solidarity often leads to selective openness for immigrants who are co-ethnics of a national majority or refugees who are persecuted because they share our beliefs and values. Kymlicka is probably right that solidarity stretches rarely to those outside some circle of affinity. But it is still worth considering how solidarity could be mobilized also towards those whose claims to admission are based on needs rather than belonging. As a progressive liberal, I feel committed to relatively open borders on all three grounds: freedom, utility and solidarity with outsiders.

It seems plausible that there is not only a trade-off between multicultural recognition of diversity and social solidarity, but also between openness for new immigration and solidarity and, finally, even between openness and multicultural integration. The latter may seem intuitively less plausible, but what Ruhs and Martin (2008) have called the numbers vs. rights dilemma can be interpreted in this way: Liberal states could admit more migrants if they were less committed to offering them multicultural rights and access to citizenship.

Moral dilemmas keep normative theorists going. But unlike moral dilemmas, which can be analysed without any knowledge of real world contexts, public policy trade-offs build on empirical assumptions that are open to be tested or refuted through studying actual cases. Kymlicka and Banting (2006) did a great job in challenging some of the empirical assumptions underlying the original version of the progressive’s dilemma: the idea that multicultural policies undermine public support and institutional sustainability of welfare states. Carefully designed comparative studies may also be able to refute the idea that openness for immigration necessarily conflicts with open access to citizenshipFootnote 1 or with sustainable welfare states. To challenge the assumed trade-offs, it seems enough to find cases where all the three good things go together and others where regressive policies on one dimension do not strengthen progressive policies on the others, but produce instead backlashes on these too. Canada and Sweden are often mentioned as states that are comparatively open for immigration, have more resilient policies of multicultural recognition and rather stable welfare states. The negative version of this refutation of the trilemma also seems to apply widely. Welfare state retrenchment has usually gone hand in hand with conservative backlashes against immigration as well as multiculturalism. And – as the example of the US in the period after World War I demonstrates – closing the doors to new immigration is likely to generate assimilationist responses to cultural diversity (or the other way round). One can go further and postulate even positive feedback loops between the horns of the trilemma. For example, demographers have argued for a long time that maintaining welfare states in rapidly ageing European and Japanese societies requires more openness for immigration.

All this does not mean, however, that progressives can comfort themselves that they do not have to face the trilemma after all. While it may not be based on hard-wired trade-offs, it is certainly strongly present in democratic politics. This is a core message of Kymlicka’s essay with which I agree entirely: in order to win democratic support for their agenda, progressives must be able to convince citizens that their immigration and multicultural policies do not kill their egalitarian social policies. What I disagree with is Kymlicka’s proposal that liberal nationalism provides an answer to this problem.

From liberal nation-building to populist ressentiment

Kymlicka is right that in the past liberal and social-democratic nation-building has often successfully brought about what he calls ‘convergence of preferences on the unit of decision-making‘ with regard to social policies, minority rights and – I would add – immigrant admission. This has made it possible to promote democratic deliberation, democratic compromises between conflicting goals and – most importantly – a sense of belonging to a political community in which citizens share rights and responsibilities.

As Kymlicka is ready to concede, liberal nationalism does not guarantee this happy outcome if there is a fundamental disagreement about the unit question. On the one hand, as we have seen at different times in Belgium, Britain, Canada and Spain, even fairly liberal forms of nationalism can undermine social solidarity, create political gridlocks or break up otherwise functioning democratic states where a sense of shared belonging to a nation is absent or has faded away. On the other hand, the fact that these plurinational states have survived for so long with high levels of immigration and levels of redistribution that are not lower than those of comparable political economies, suggests that convergence of preferences on the unit question may be less important for social solidarity than Kymlicka thinks.

Let us consider first the preconditions under which liberal nationalism could historically resolve the trilemma by creating a sufficiently strong progressive consensus on the three goals. In some cases, partial reconciliation has been achieved by imagining the nation itself primarily as a welfare community (as in the Swedish peoples’ home) or as an immigrant society (as in the US, Canada or Australia). These two images of the nation have rarely been fully combined, which suggests that nation-building may in the past have been able to reconcile two of the three goals, but not all three. The question is whether even the partial solutions are still viable today.

The steep rise of the Sweden Democrats and similar pro-welfare but anti-immigrant parties in other parts of Europe suggests this has become very difficult. At least in Europe, those who pose as defenders of the nation-state are without exception opposed to immigration and multiculturalism. Only Catalan and Scottish secessionists combine today nationalism with pro-immigrant rhetoric. I would, however, take no bet that this rhetoric would survive for very long in case these regions became independent. Moreover, the goal of separation itself involves a desire to break the bonds of social solidarity within the existing British and Spanish states and seems thus deeply at odds with a progressive agenda of building inclusive welfare states.

Liberals should also not idealize the historic conditions under which welfare nations and immigrant nations had been built in the past. Sweden’s peoples’ home was closely associated with eugenicist and sometimes openly racist ideas about the people. And the immigrant nations of the US, Canada and Australia were built as white settler societies that took the land from indigenous peoples and were for a long time very selective with regard to immigrants they saw as fitting the national mould. Fortunately, these ideas are no longer acceptable and the contexts for such nation-building projects no longer exist. In Europe, with the partial exception of 19th century France, immigration was never seen as a resource for nation-building and today it is less likely than ever to be perceived in this way. Paradoxically, the national case for immigration on grounds of demographic and economic needs is much stronger today in Europe than in North America, but this argument has been spectacularly unsuccessful with hardly any resonance in public opinion and very little impact on public policies.

Kymlicka is aware of these conditions that make it difficult to reconcile nationalist ideologies with progressive attitudes to immigration and cultural diversity. He writes that “nationhood works best when it is deep in the background, as a taken-for-granted presupposition of social life.” Where national identities are foregrounded, they tend to prime anti-foreigner attitudes. But this diagnosis seems at odds with the idea that it is nationhood that provides the necessary condition for inclusive solidarity in diverse societies. Nations are after all imagined communities rather than background facts of social life that exist independently of people’s subjective sense of belonging and identity. If citizens are less tolerant of diversity when they imagine themselves as belonging to a nation, how can the promotion of national identities then help to reconcile social solidarity with recognition of diversity and openness for immigration?

Rescaling democratic solidarity

I suggest therefore that progressives need to look beyond the nation. Beyond does not mean after. Kymlicka’s critique of postnational responses is to some extent justified: they often “assume […] converging preferences, but offer no explanation for them … and … they typically smuggle back in the very assumptions of nationhood that they purport to reject.” What looking “beyond the nation” means instead is searching for political community “across”, “above” or also “below” the nation-state, without denying its existence and relevance as the crucial level in relation to which all other forms of political community define themselves.

The premise for this exploration is that democracy needs convergence between decision-makers and decision-takers but no convergence of preferences on the unit of decision making for all decisions. There are two contrasts between nationalism and democracy in this regard. First, nationalists imagine the world as one of distinct nations with separate territories and memberships. Democracy, however, can bind together different units of decision making that are territorially nested within each other – as in federal states or unions of independent states. And it can connect separate units of decision-making that share overlapping memberships – as happens when migrants are simultaneously citizens of sending and receiving states. Second, nationalists assume that nation-states (or autonomous national territories within states) should control all decisions that affect their citizens. By contrast, stable democracy requires that the units of decision-making should be functionally adequate for the decisions that they take.

The nationalist ideal has a powerful grip on most citizens’ – and theorists’ – minds, but seems to be increasingly out-of-synch with political and economic developments in the present world. Nation-states are less and less capable of controlling their own political agendas while they exercise at the same time far too much control over the agendas of substate polities as well as supranational unions of states that are better suited to address problems that nation-states cannot deal with adequately. The new nationalism of the populist right is a backward looking mobilisation of political ressentiments rather than a feasible project of strengthening the nation-state.

Why should we strive for unit convergence when trying to overcome the progressive trilemma? Isn’t it a hopeless endeavour to try and squeeze into a single unit all the powers required for achieving progressive goals? Wouldn’t it be better to ask which unit is best suited to deal with social welfare, which one can best accommodate cultural diversity and which one should regulate immigrant admission? Wouldn’t it be plausible to assign primary responsibility for “diversity management” to cities, for social welfare entitlements to states, and for migration policies to the European Union or similar unions of states in other global regions, such a South America?

Kymlicka may object that doing so would make it impossible to achieve any of the three goals. If a supranational unit decides who is admitted and can freely move within its territory, how can nation-states then maintain their distinct welfare regimes? And if local governments manage cultural diversity issues, doesn’t this undermine the shared sense of belonging that citizens must feel to support a national welfare regime? The answer to this objection is that convergence of preferences is indeed needed, but about a division of tasks and powers in an integrated multilevel polity rather than about a single unit of decision-making. In other words, a multilevel and transnational answer to the progressive trilemma can only work if individuals come to regard themselves as multiple citizens across all levels of the polity and accept that those who move across borders are transnational citizens who belong simultaneously to territorially distinct polities.

Rescaling solidarity in this way is certainly not an easy task, but it seems to me a more realist utopia than liberal nationalism, in the sense of a normative solution to a major problem that meets basic conditions of feasibility in the current world. This approach could even help with addressing the challenge that Kymlicka raises in the concluding paragraphs of his essay. As he acknowledges there, liberal nationalism has been able to partially realize the ideal of inclusive solidarity in diverse societies where “immigrants have traditionally been seen as permanent residents and future citizens”. He concedes that “[i]t is far less clear that this model works for temporary migrants.” By contrast, in a multilevel and transnational conception of democracy, temporary migrants can be accepted as citizens at the local level by virtue of their residence and as citizens of a supranational union if they are also citizens of one of its member states. Kymlicka is right to worry about the sources of solidarity in “a world of liquid mobility”. But this is to a significant degree already our world and progressives should think hard how their ideals can still be realized within this world. The answer that seems most plausible to me is to adjust the units of political decision-making in such a way that natives, settled immigrants and temporary migrants can still see themselves as sharing membership in democratic communities. Just as nation-building had initially the effect of assigning shared citizenship to sedentary populations and uprooted peasants turned into mobile industrial workers, so building supranational unions of states can today capture a lot of geographic mobility and temporary migration within a larger unit and turn it into internal freedom of movement linked to a shared citizenship.

For the same reasons that applied historically to nation-building, it would also be necessary to strengthen the competences for immigrant and refugee admission at supranational level in order to keep borders fairly open for the admission or immigrants and refugees from outside the union. This conclusion seems obvious when watching the nasty games of border closure and passing on of refugees that are currently being played by the member states of the European Union. Yet supranational unions need not replicate the patterns of nation-building on a larger scale. None of the currently existing ones in Europe and South America is designed to do so. Nation-states will remain the basic building blocks of the international political system. Supranational unions that adopt progressive agendas could provide an economic environment that shelters welfare states against the forces of globalized neoliberal capitalism. But I share Kymlicka’s view that they are not the units within which a lot of social solidarity can be mobilized to support progressive policies of redistribution.

While nation-states may remain the right units for promoting comprehensive conceptions of social justice, they are increasingly inadequate for coping with cultural diversity emerging from temporary migration rather than settlement migration or territorial minority nationalisms. Cities provide alternative political spaces that are open for newcomers and in which solidarity among strangers who come from elsewhere and are going to move elsewhere could be mobilized through a conception of citizenship derived from residence rather than birth.

In short: if progressives face today a trilemma how to reconcile their commitments to fairly open borders, recognition of cultural diversity and social redistribution, they should think harder which political units are best suited to achieve each of these goals.


  1. The classic statement of this dilemma is Walzer (1983), chap. 2.


  • Kymlicka, W., & Banting, K. (Eds.). (2006). Multiculturalism and the Welfare State. Oxford: Oxford University Press.

    Google Scholar 

  • Ruhs, M., & Martin, P. L. (2008). Numbers vs. Rights: Trade-Offs and Guest Worker Programs. International Migration Review, 42(1), 249–265. doi:10.1111/j.1747-7379.2007.00120.x.

  • Walzer, M. (1983). Spheres of justice. A defense of pluralism and equality. New York: Basic Books.

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Rainer Bauböck.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bauböck, R. Why liberal nationalism does not resolve the progressive’s trilemma: Comment on Will Kymlicka’s article: “Solidarity in Diverse Societies”. CMS 4, 10 (2016).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: