Portal de Periódicos da CAPES

The post‐nationalization of immigrant rights: a theory in search of evidence 1

2012; Wiley; Volume: 63; Issue: 1 Linguagem: Inglês

10.1111/j.1468-4446.2011.01401.x

1468-4446

Ruud Koopmans,

Historical and Contemporary Political Dynamics

In her contribution to this issue, Yasemin Soysal makes a brave stand for her theory of post-national rights, even in the face of changed circumstances. Originally (Soysal 1994), the theory was designed to account for the extension of all kinds of civic and social rights to immigrants in Western Europe, in spite of the fact that most immigrants had not naturalized to become citizens, not least because European immigration countries were not always keen on making immigrants into nationals. Extending the world-cultural institutional perspective of John Meyer and his associates (e.g., 1997) to the realm of immigrant rights, Soysal placed this apparent anomaly in the context of the spread of supranational human rights discourses in the postwar period, pushed forward by the mobilization of non-governmental organizations, and backed by international treaties and supranational forms of governance. Through a combination of self-commitment to these international human rights norms, pressure by national and supranational immigrant and human rights organizations, and if necessary binding rulings of supranational courts, Soysal argued, nation-states had come to embrace, sometimes willingly, sometimes grudgingly, a highly universalist and individualized notion of rights, leading to a strong decoupling of rights from national citizenship, which, in her words, had become 'in terms of its translation into rights and privileges . . . no longer a significant construction' (Soysal 1998: 208). Empirically, Soysal's primary point of reference has always been the European Union (EU), which is hardly surprising given that the EU is the world's most advanced system of supranational and intergovernmental governance, and therefore the most likely case to observe a post-nationalization of rights. In her contribution to this issue, she again makes a case for the influence of European integration on the rights of immigrants. However, similar to her original presentation in Limits of Citizenship she falls far short of providing a systematic empirical underpinning of her argument. If the argument about post-national rights and the impact of European integration is to have any value it needs to lead to some empirically testable predictions. Unfortunately, Soysal's argument mostly stays on a level of abstraction that makes it difficult to recognize testable hypotheses. At any rate she doesn't specify any herself. Let me therefore try to derive some empirical corollaries from her argument. First, we should not expect to see great differences in the rights that member states extend to immigrants. Because the answer to this question may be in the eye of the beholder (some may find the differences impressive, others the commonalities), we need to further specify it in a comparative framework, both cross-nationally and diachronically. The second, cross-national, corollary is therefore that EU countries should be more similar to other member states than to non-EU countries. The third, diachronic, corollary is that, due to the common pressures exerted by supranational human rights regimes and discourses, countries in general, and, because of the added pressures exerted by common EU regulations and discourses, EU member states in particular should converge to more similar policies regarding immigrant rights over the course of time. It would certainly be unfair to argue that there is no place for cross-national differences in Soysal's argument. She has always emphasized – and reiterates this in her contribution to this issue – that the implementation of universal rights remains largely situated on the national level and therefore 'dependant upon particular institutions and their social and historical contexts.' This is perfectly sensible, but if the post-national argument has anything to offer beyond theories that emphasize these national institutions and contexts, it should make predictions that go beyond those derived from national particularisms. And such predictions need to be of the type just specified: cross-national differences must be relatively limited, smaller within the EU than outside it, and declining over time, especially within the EU. Absent empirical support for these corollaries, post-nationalism cannot be accepted as a valid account of the current state of affairs regarding immigrant rights in Europe. At the time when Soysal wrote Limits of Citizenship, the large majority of immigrants in Europe were still foreign nationals. By now this has changed, and the majority of them have become citizens through naturalization or birthright acquisition. Since the Maastricht Treaty of 1992, citizenship in one of the member states automatically confers European citizenship. However, even for EU citizens the post-nationalization of rights is much more limited than is often thought. Yes, they can travel freely to other member states and take up work there. But depending on the country where they live, they will enjoy very weak or very strong protection against dismissal, will fall back in income a bit or a lot if they become unemployed, may wear their headscarves to school and work or not, can send their child to a denominational school or not, have their child enjoy state-funded classes in the language of their country of origin or not, be allowed to slaughter animals according to their religious rite or not, and finally at the end of their lives have the right to be buried without a coffin or not. For non-naturalized nationals of third countries, rights and duties change even more when they live in one EU country or another. To obtain entry, residence and citizenship rights, they may or may not be required to pass language and citizenship tests of varying content and difficulty, they may or may not keep dual nationality if they naturalize, it will be relatively easy or very difficult for them to get a marriage partner over from a non-EU country, they can or cannot be expelled for dependence on social welfare and minor criminal offences, and they can work in virtually all public-sector jobs or in virtually none. That is the reality of 'post-national' Europe anno 2012. To be sure, mostly these rights are specified – though not necessarily applied – in a non-discriminatory way. For instance, the Netherlands have not denied Muslims the possibility to set up state-funded religious schools, which was a long-established right for Christians. France has banned the headscarf from schools, but the same legislation also banned the Jewish kippah and 'large'– there the equality already becomes quite relative – Christian crosses. This adherence to principles of equality before the law regardless of religious background would probably be interpreted by Soysal as evidence in favour of the penetration of supranational discourses of individualized human rights. However, we can just as easily explain these examples as following from Dutch and French commitment to their own constitutional principles of equality –liberté, egalité, fraternité was not a postwar import – and to discourses of 'sovereignty within one's own circle' (one of the founding principles of Dutch pillarization) or 'laïcité' (state secularism). Moreover, the non-discriminatory application of rights that we find in these examples is far from universal. Immigrants in many parts of the globe – e.g., Nigeria, the Gulf states, or Malaysia – can testify that the practical implications of supranational human rights discourses are often remote to say the least. But even in Europe we find examples of blatant unequal treatment of religious groups, who are allowed to build church towers, but not minarets in Switzerland, or can teach in monastic habit in cross-decorated class rooms, but not in a headscarf in the South of Germany. Moreover, there are many cases in which the principle of equal treatment is suspended because there is no Christian precedent, allowing slaughtering according to the Muslim rite to be banned referring to animal rights – I guess with some good will one can interpret that as the consequence of a powerful supranational discourse, too – or Muslim burials referring to hygiene standards. But, as I have indicated above, one must not necessarily be impressed by these cross-national differences and may argue that the commonalities are far greater and that it is these commonalities that are shaped by supranational rights regimes and discourses. While cross-national commonalities may derive from common subjection to supranational influences, this is not the only possible explanation. As among others Christian Joppke (e.g., 2007) has argued, liberal democracies also share internal constitutive principles – such as the freedom to exercise one's religion or the rule-of-law principle that equal cases should be treated equally – that imply that the granting of rights to individuals and groups will be more similar across democracies than it will be between them and non-democracies. If we find however that within the set of liberal democracies, those that belong to the world's most advanced supranational project, the EU, are more similar to each other than to liberal democracies outside the EU, this would be compelling evidence in favour of the influence of post-national constellations. The latest edition of the MIPEX indicator system of immigrant rights (MIPEX 2010) offers an opportunity to test this hypothesis. For the 2010 edition, the MIPEX team have gathered information on immigrant policies and rights in 33 countries – the 27 EU member states, two non-EU European countries (Norway and Switzerland), and four non-European OECD countries (the USA, Canada, Australia, and Japan) – in the domains of access to nationality, long-term residence, labour market mobility, family reunion, education, political participation, and anti-discrimination. Looking at the results of this exercise, one is hard-pressed to discover any particular pattern of rights that distinguishes EU countries from the rest. In fact, both the country that is most generous in extending rights – Sweden, with a score of 83 on the scale from 0 to 100 – and the most restrictive country – Latvia with a score of 31 – are EU countries. That makes Sweden quite similar to Canada (score 72) and Latvia to Japan (score 38), but neither of the two typical for some kind of EU regime of immigrant rights. We find the same pattern if we look at the seven domains of immigrant rights that MIPEX distinguishes separately: the EU always emerges as an extremely heterogeneous bunch with even more variation in immigrant rights than there is between self-proclaimed 'multicultural' classical immigration countries such as Canada and Australia, on the one hand, and classically ethnic-exclusivist Japan, on the other. Even if we only focus on European countries and thus control for the cultural and institutional differences between the classical immigration countries, Japan, and Europe – which anyway following Soysal's argument amounts to controlling for something that is no longer very relevant when it comes to rights – we find that the two non-EU member states do not stand out. Norway (66) is halfway between Sweden (83) and Denmark (53) and very close to Finland (69), and Switzerland (43) very close to Austria (42). That doesn't sound like post-nationalism, but more like the persistence of shared institutional and cultural patterns among kin nations, regardless of EU membership. Granted that cross-national differences in immigrant rights are still substantial, even within the EU, one may still argue that under the influence of the penetration of supranational human rights these differences have become smaller, and that this convergence has been particularly pronounced in the EU. Here, the MIPEX data are not very helpful, because they only cover two measurement points that are very close in time, 2010 and 2007 (the 2004 predecessor was based on a different scoring system and can therefore not be directly compared to the later measurements). Together with my colleagues Ines Michalowski and Stine Waibel, I have analysed the development of immigrant rights in ten West European immigration countries across the period 1980–2008 (see Koopmans, Michalowski and Waibel 2012). Where a direct comparison is possible, our data correlate strongly with the MIPEX indicators, as well as with other studies of immigrant rights (Banting and Kymlicka 2004; Howard 2009; Janoski 2010; Waldrauch and Hofinger 1997). We find that in most areas of immigrant rights – the starkest exception being marriage migration rights – and in most countries – except France and Denmark – immigrant rights expanded over time. However, this was not a linear trend, as policies in most countries and most areas became more restrictive again since 2001. Contrary to the prediction of the post-national rights perspective, cross-national differences became larger rather than smaller over the period 1980–2008. We found evidence of policy convergence (as indicated by a comparison of standard deviations) in only two of the eight areas of immigrant rights that we analysed (antidiscrimination and protection against expulsion), but strong divergence in four areas (access to public service employment, cultural rights in education, other cultural and religious rights, and marriage migration rights) and stable cross-national variance in the remaining two areas (political rights and access to nationality). In addition, we investigated which explanatory mechanisms can account for these observed patterns. Like MIPEX, our study includes two countries that are not part of the EU – Norway and Switzerland. In addition, we could investigate for two countries – Sweden and Austria – whether their accession to the EU in 1995 had an impact on their attribution of rights to immigrants. We find that EU membership cannot account for any of the findings. Although two EU countries that originally had rather exclusive citizenship policies, Germany and Belgium, show strong rights expansion, this is also true for Switzerland. Moreover, the two countries that deviate from the trend towards greater inclusiveness over the period 1980–2008 (Denmark and France) are EU members. For Austria and Sweden, we do not observe any impact of their accession to the EU. Sweden already had the most inclusive policies on immigrant rights in 1980, long before it joined the EU. Austria belonged to the most restrictive countries in 1980, but after its accession to the EU dropped to the last place in the ranking of our ten countries, being overtaken by Germany and Switzerland, respectively an EU and a non-EU member. Whatever accounts for the trends we find in immigrant rights, it doesn't seem to have much to do with EU membership, or, judging from the predominance of temporal divergence rather than convergence, with any kind of common supranational driving force. Instead, we find strong evidence of national path dependence as countries' positions are strongly predicted by where they stood in 1980. Moreover, changes towards greater inclusiveness or restrictiveness are best explained by domestic factors. Strong right-wing populist parties tended to keep countries on restrictive paths or to reverse liberalization trends, especially after 2001. Liberalization of immigrant rights, by contrast, occurred where, through a combination of high levels of immigration and inclusive naturalization laws, immigrants came to make up a large part of the electorate and thereby a significant domestic political force. The predominant divergence trend that we observe across countries is related to these mechanisms. Countries that had restrictive policies in 1980 were more likely to also be the countries where right-wing populist parties were subsequently successful, and this kept these countries on restrictive paths. Conversely, countries with comparatively inclusive immigrant rights in 1980 were more likely to see immigrants become a significant electoral force in the following decades, which pushed these countries further along inclusive paths. In her contribution, Soysal pays special attention to one example of recent changes regarding immigration rights, namely the introduction of citizenship and integration tests as a precondition for naturalization, permanent residence and marriage migration in several European countries. Referring to these tests as evidence for the penetration of supranational human rights norms, as Soysal does, is courageous because they clearly limit immigrants' access to rights. In Germany, the increased demands that are made on candidates' German language abilities has done away with much of the positive effects that the 2000 liberalization of nationality law initially had on naturalization rates. In the Netherlands, naturalization rates have significantly declined since the introduction of stricter language requirements. Now one may argue that in an age of post-national rights this matters little, because national citizenship is anyway irrelevant. In view of the fact that in most countries citizenship is a condition for access to core societal positions such as the teaching, police, and military professions that seems difficult to maintain. Moreover, as I have indicated above, access by way of citizenship to the right to vote and be elected is not just important in its own right, but has a reinforcing effect on immigrants' struggle for rights in other domains because of the electoral leverage it provides. In addition, in the Netherlands, one of the countries where language tests have been extended to prospective marriage partners abroad, they have led to a strong decline in the number of transnational marriages (see Wilkinson, Goedvolk and van Doeten 2008). In view of these effects, it is striking that Soysal virtually ignores the language aspect of integration and citizenship tests. In all countries where they have been introduced, immigrants now face significantly higher language barriers to the acquisition of rights than they did in the past. It requires a great deal of imagination to see this as a consequence of the penetration of international human rights standards. If countries had somehow rewarded knowledge of English and other foreign languages, we could have legitimately interpreted this as evidence for a post-nationalization of rights attribution. But now that countries make stronger demands on immigrants' knowledge of the national language, and simultaneously have often downscaled or abolished previous programmes to further immigrant languages, this is apparently somehow not relevant to the post-national argument. Only by ignoring the language component can Soysal conclude that 'current citizenship and integration tests do not reveal anything distinctive about the particularities of a nation'. But even disregarding language, this conclusion is exaggerated. The detailed study by Ines Michalowski (2011) that Soysal refers to indeed shows that the largest category of questions asked in the civic knowledge part of these tests refers to 'politics, history, and geography'. Only a negligible proportion (e.g., seven out of 300 German questions) of these questions refer to the European Union. Some of the typical questions that Michalowski lists indeed refer to general liberal-democratic principles such as 'What is the function of elections in a democracy?' or 'What is the freedom of religion?' But others do refer to national institutions and history. The German candidates are for instance asked when Hitler and the Nazis were in power; their Austrian counterparts what kind of a form of government the country had before 1918; and immigrants to the UK what kind of a constitution the country has, what the Church of England is, and who its head is. These questions certainly refer to national particularisms, albeit not – with the exception of some of the Dutch test questions – to ethno-cultural particularisms. It seems that it is the absence of the latter that Soysal – and similarly Joppke (2007) – interprets as proof that integration requirements are 'not posed as a process of confirming or furthering national collectivity and identity'. That conclusion ignores that civic forms of nationalism have always co-existed with ethnic nationalism and that therefore the absence of elements of the latter does not prove anything per se about the decline of the nation. Moreover, even if one considers the national content of integration and citizenship tests to be weak, it is still considerably stronger than what preceded it, namely no or much more limited requirements regarding knowledge of the national language, institutions, and history. Moreover, the fact that these requirements are imposed only on immigrants clearly runs against the idea of universal, individualized rights. No natives have yet been required to take language tests or to demonstrate their familiarity with national institutions and history. Almost two decades ago, post-nationalism was advanced as an argument to explain the expansion of immigrant rights. Now Soysal's argument seems to be that new requirements imposed on immigrants, and only on them, are also evidence of de-nationalization. The persistence of cross-national differences in important domains of immigrant rights, the absence of any recognizable EU pattern of rights attribution, the introduction of new restrictions on immigrant rights in recent years, and the absence of a trend towards cross-national convergence together demonstrate that, at least in as far as immigrant rights are concerned, there is no empirical basis to support the theory of post-national rights. The most evocative way to illustrate to what extent the theory has become decoupled from reality is to imagine a prospective immigrant who has the choice between several possible destination countries. Soysal asks that immigrant to have faith in the supranational human rights discourse and tells her that in terms of rights it doesn't really matter where she goes, to France or Canada, to Japan or Sweden, to the United Kingdom or the United Emirates. This is what Joe Hill had to say in a famous migrant worker song from 1911: Long-haired preachers come out every night, Try to tell you what's wrong and what's right; But when asked how 'bout something to eat They will answer with voices so sweet: You will eat, bye and bye, In that glorious land above the sky; Work and pray, live on hay, You'll get pie in the sky when you die. (Joe Hill –'The Preacher and the Slave')