Definitional Debates, Mechanisms and Canada: Comment on Will Kymlicka’s article: “Solidarity in Diverse Societies”

Does immigrant-origin diversity undermine states’ willingness to engage in broad social redistribution, from enacting “living” minimum wage laws and progressive taxation, to providing public benefits in order to mitigate economic inequality? In asking this question, Will Kymlicka (2015), one of the staunchest defenders of multiculturalism, takes seriously the question of whether recognition of cultural, ethnic and religious diversity comes at the expense of redistribution. Not surprisingly, Kymlicka makes a strong pitch for a multicultural welfare state, which he distinguishes from neoliberal multiculturalism, exclusionary welfare chauvinism or assimilatory neoliberalism¹ (Kymlicka, 2015).

My own normative view – and my empirical research – is in strong support with Kymlicka’s (2015), position on the value of combining recognition and redistribution. I also agree with his skepticism as to whether existing data reveal any evidence that multicultural policies generate or exacerbate welfare state retrenchment, and in his prescription, namely, the construction of multicultural, liberal nationalism (Bloemraad, 2006; Bloemraad, 2012). Multicultural nationalism is presumably distinct from alternative national solidarities around more exclusionary, homogeneous cultural identities, or a neoliberal, cosmopolitan approach to global membership that urges the erasure of nationalism altogether.

Given my sympathies, this commentary is not so much a challenge to Kymlicka’s (2015), normative argument as a social scientific appraisal of how he does not go far enough in elaborating the mechanisms presumed to produce the progressive’s dilemma. This is problematic, because it means that Kymlicka also does not theorize sufficiently an answer to the question of why multicultural, liberal nationalism might address the mechanisms that erode the welfare state in a context of diversity. I will attempt to sketch out some possible answers and, in doing so, I will quibble with the terminology of “solidarity” as compared to “social membership,” both terms used by Kymlicka in his article. I engage in this definitional debate not to split hairs, but to elucidate some silences in the redistribution/recognition trade-off that need to be addressed moving forward. I finish by speculating how much experience with the Canadian case, the epitome of solidarity through multicultural nationalism in the global North, shapes (and perhaps blinds) the views of both Kymlicka and myself on what is possible elsewhere.

Definitional debates: why solidarity?

What, exactly, is solidarity? And how does it differ from another term that Kymlicka (2015), uses in his article, namely an ethic of “social membership”?

Kymlicka (2015), notes the virtual absence of theorizing on ‘solidarity’ outside of sociology. Indeed, even within sociology, Kymlicka cites social theorist Jeffrey Alexander to argue that solidarity has “disappeared” as a concept and topic. Why then use this term? Kymlicka does not elaborate, beyond a hint that, empirically, contemporary societies function based on a sense of community that goes beyond the coercion of laws or formal institutions and, more explicitly, that a theory of equality that includes economic justice can find the necessary “glue” for a robust welfare state through national solidarity.

My own instinct is to feel uneasy about solidarity, and more comfortable with a language of membership. This reaction stems, I think, from concerns about the strength and primacy of collective obligations inherent in each term, and the openness to multiple and even cross-cutting obligations. According to the Oxford English Dictionary, solidarity is “The fact or quality, on the part of communities, etc., of being perfectly united or at one in some respect, esp. in interests, sympathies, or aspirations.”² Synonyms offered by other dictionaries include unanimity, unity, harmony, cohesion and like-mindedness.³

All of these terms are in tension with Kymlicka’s (2015), starting point: the centrality of democracy and, as he puts it, “facts of pluralism.” It also sits awkwardly, I feel, with a key precept of multiculturalism, namely the recognition, valorization and support of diversity. In contrast, membership refers to being a “constituent element within a social or other organized structure.”⁴ The individual is part of a whole, but this does not entail a ‘perfect unity’ requirement. At the same time, the notion of membership is more than just a transactional or instrumental relationship, like two parties agreeing to a contract. As part of a social group, relationships are implied to extend over some time period, which can generate norms around rules, reciprocity and even engagement in a common enterprise agreed to by members. It is, in a sense, a thinner version of collective action than solidarity, and does not carry the same baggage of like-mindedness.

Who deserves to be part of “us”? And why?

These semantics matter because, as scholars, we must understand the mechanisms behind the construction of the “we” to whom we owe protection from or compensation for market inequalities. Kymlicka (2015), provides a useful starting point in his discussion of deservingness. He lists judgements about an individual’s choice or agency in producing economic disadvantage, the sense of common identity and belonging to a shared society, norms of reciprocity between current and future beneficiaries, and attitudes of civic friendship. Unfortunately, Kymlicka does not then link each of these elements to the progressive’s dilemma he began with, nor the posited solution of liberal, multicultural nationalism on which he ends. Can these links be made?

“Identity” is most clearly a mechanism that those worried about multiculturalism believe reduces support for the welfare state. As more – or certain kinds – of immigrants enter a society, they are presumed to be seen as different by majority group citizens. This can lead to exclusion through welfare chauvinism or, if these strangers are given benefits, to decreased social spending. Empirical support for this mechanism is found in minimal group experiments in social psychology that show people favor in-group over out-group members, and in the political rhetoric of various political parties in the global North. But, as Kymlicka (2015), points out, despite experimental or small-group studies supporting this mechanism – studies done in isolation from other causal processes – social scientists have found no empirical evidence that actual welfare state spending has decreased more in countries embracing multiculturalism compared to those who eschew multiculturalism. Indeed, the negative finding is telling: countries that have taken a hardline stance against multiculturalism, or immigration, have not seen a flowering of their welfare state or dramatic increases in economic equality.

Despite the lack of empirical support, Kymlicka (2015), believes this argument has a certain bite and, perhaps, at the force of being repeated in political rhetoric, it might become a self-fulfilling prophecy, or feed into welfare chauvinism. The logical response, for someone who wants to maximize justice and equality among the largest group, is to make diversity a part of social identity. This, Kymlicka argues is precisely what has happened in Canada, and perhaps Australia and Scotland as well.

It is less clear, however, how embracing liberal, multicultural nationalism addresses the other mechanisms driving logics of deservingness: beliefs about choice, notions of reciprocity or civic ‘attitude’. Critics of multiculturalism sometimes suggest that certain immigrant groups hold cultural views or are encouraged in their cultural isolation such that they ‘choose’ to use social programs rather than enter the labor market (e.g., Koopmans, 2010). These arguments directly link multiculturalism to claims that the welfare state shouldn’t support groups or individuals who choose not to contribute (rather than who face structural problems), people who only ‘take’ rather than give (undermining reciprocity norms), and those who just don’t have the right civic values (thus driving adoption of coercive and more homogenizing civic integration policies). Kymlicka (2015), does not explain how cultivating multicultural nationalism would undermine these critiques.

I do not believe that these other mechanisms, potentially linking recognition to decreased redistribution, have been tested, but they certainly should be subject to empirical scrutiny. Is it the case that residents of countries with stronger multicultural identities are less likely to think that immigrants and their children are part of a reciprocity agreement with longstanding residents, or more likely to think that those of immigrant-origin choose welfare? Clearly articulating the mechanisms presumed to undermine the welfare state, and carefully linking each mechanism to arguments about diversity, will help establish an empirical agenda that allows us to evaluate such theoretical and political claims.

Only in Canada?

Clarifying the posited causal links also raises the question of whether liberal, multiculturalism nationalism is possible in any developed state, or whether it is the product of highly specific historical processes, demographic context, geographic location and political conflicts. Put differently, is such nationalism only possible in Canada, or a few other states?

Let me reiterate that, normatively, I favor a welfare state that produces greater economic equality and a multicultural, liberal nationalism that generates inclusive membership in pluralistic societies. Empirically, I agree that there is no strong evidence for a multiculturalism/ redistribution trade-off, and that Canada, in particular, is a positive example of multicultural nationalism, though perhaps less a prime example of a generous redistributive welfare state. Undoubtedly, however, my views – and those of Will Kymlicka (2015), − have been influenced by living in and studying Canada. Perhaps our understanding of the problem and view of the solution are both biased by these experiences.

In Canada, norms of reciprocity are strongly articulated in the government’s longstanding economic focus in immigrant entry policy. Through the Canadian “point system,” immigrants who do not have family ties to Canadian residents nor have pressing humanitarian reasons to migrate can apply to come to Canada based on their purported ability to fit into the Canadian labor market and society. The proportion of immigrants selected on economic criteria has fluctuated widely over the last 40 years – from just over a third in 1986 to two-thirds in 2010 (Barbieri & Ouellette, 2012) – but the Canadian government has consistently viewed immigration as part of economic growth, whether to populate the West in the 19^th century or drive technological innovation in the 21^st century. The Canadian-born are continuously told that immigrants will help foster economic development, and help pay for the expanding cost of an aging population.⁵ This rhetoric feeds into a discourse of reciprocity, and undermines claims that immigrants’ choose to use social benefits. Debates revolve around the question of whether immigrant unemployment or poverty is due to employers’ failure to recognize foreign credentials, a lack of “Canadian” experience in the labor market, or discrimination. Each argument presumes immigrants do not want to rely on the welfare state, but are forced to do so by forces beyond their control. Given this, Canada’s success might not just be about multiculturalism, but also immigration policy and government discourse on immigrants.

Canadian nationalism and notions of membership also rest on the fact that Canadian policy has long focused on permanent migration. Those who come are immigrants and future Canadian citizens, not sojourners or temporary migrants.⁶ The Canadian government, and most citizens, expects immigrants will stay, and also expect that they will naturalize. The vast majority of immigrants appear to agree, as Canada has arguably the highest rate of immigrant citizenship acquisition in the world (Liebig & Von Haaren, 2011). Thus Canada’s multicultural nationalism is not just about the recognition and accommodation of diversity, but also about political and civic membership, a point that Kymlicka (2015), mentions in passing at the end of his article.

My use of the term membership is deliberate, and brings us back to definitional debates. But it also raises a question about our knowledge of the historical process behind nation-building. The idea of solidarity might well entail a thicker sense of mutual obligation than membership, which could be passive or more instrumental. But solidarity also carries the danger of being more coercive if one needs to subsume one’s individuality to the group’s wishes and ends. Solidarity is, I think, forged in a stronger “we” versus “them” dynamic of conflict. Workers need solidarity in the face of powerful employers. Poles fought in solidarity for democracy against authoritarian elites. The solidarity of English nationalism was forged in multiple wars, and perhaps in colonialism. Empirically, can solidarity be created without a sense of conflictual opposition?

If not, we might need the weaker cohesion of membership, but in doing so, we allow for internal conflict and pluralism.⁷ Indeed, adoption of a multicultural ethos was in part the product of political conflict in Canada, and in the United States (Bloemraad, 2015). Extended to other countries, the path to national inclusion might need to go through civic membership and political conflict that includes immigrant-origin minorities. Perhaps solidarity will arise at the end, but the focus should be on the mechanisms which allow for both recognition and redistribution.