Philosophies of integration? Elite views on citizenship policies in Scandinavia

Research on the ‘civic turn’ in European citizenship policies suggests that concerns over immigrant integration have fueled the restrictive development of citizenship policies in recent decades. However, few efforts have been made to explore the normative ideas underlying this development. Departing from Favell’s (1998) influential concept of ‘philosophies of integration’, this article draws on elite-interviews with top-level bureaucrats, politicians and citizenship experts in the Scandinavian countries and explores how ideas about nationhood and integration have influenced the divergence in citizenship policies in the region, and what overall purpose the policy-changes reflect. We find that, especially in Denmark, the gradual introduction of ever more demanding civic integration requirements appears to be a part of the broader aim of controlling the inflow of migrants and not to enhance the integration of those already present in the country. Conversely, the Swedish liberal approach to citizenship, which regularly has been analyzed as a steppingstone to societal integration, might in reality have been a reflection of benign neglect. In Norway, immigrant integration has clearly been part of the underlying rationale for changes in naturalization requirements, yet the control dimension have played an increasingly important role. These findings suggests that, although the citizenship institution remains important in nation states’ efforts to solve their ‘ethnic dilemmas’, the underlying rationale behind policy-change is more multifaceted than previously understood.


Introduction
Since the turn of the century, citizenship policies have been revitalized across Europe (Goodman, 2014;Koopmans et al., 2012;Vink and Baub€ ock, 2013). Widespread concerns of inadequate integration of immigrants underlie this development (Bloemraad, 2017;Brochmann and Hagelund, 2012;Joppke, 2007;Mouritsen, 2015) and has led many states to introduce civic integration requirements, such as employment, language tests and demanding requirements of knowledge about society and culture, as part of their rules for acquiring citizenship (Goodman, 2014;Joppke, 2017). The trend toward including integration requirements as conditions for naturalization is referred to as the 'civic turn' in the citizenship literature and has received massive attention in recent years (e.g. Borevi et al., 2017;Goodman, 2012Goodman, , 2014Jensen et al., 2017;Joppke, 2007Joppke, , 2017; Mouritsen et al., 2019). However, few efforts have been made to explore the rationale behind the changes: Does the introduction of civic integration requirements reflect a wish to improve the inclusion of newcomers? Or might restrictive changes in citizenship policies, in some cases, rather be understood as a new means to bolster immigration control mechanisms in the hands of nation-states increasingly restrained by internationalization?
This article focuses on the Scandinavian nation-states which, since the early 2000s, have developed distinctly different approaches to citizenship acquisition, with Norway being positioned between liberal Sweden and restrictive Denmark (Brochmann and Seland, 2010;Jensen et al., 2017;Midtbøen, 2015). By analyzing the underlying rationale of the three emerging citizenship regimes, we reexamine prior assumptions on the close relationship between citizenship and integration in the region.
Theoretically, the article builds on Favell's (1998) classical notion of 'philosophies of integration'. According to Favell, citizenship policies do not primarily reflect historical notions of nationhood (cf. Brubaker, 1992), but rather more recent, consensual ideas about the pragmatic political solutions to ethnic pluralism within a given nation state. Favell (1998: 25) claims that the dilemmas of largescale immigration tend to fuel processes of institutional change, in which national elites use 'the language of citizenship, integration and national identity' to form a new society-wide agreement on the terms of naturalization. This new consensus, he suggests, may be achieved either by looking forward, reaching agreement on what the national identity should look like in the future, or by looking backward in time, 'via the (re)invention of a historical myth of the nation as the founding source of these new institutions ' (25).
Established analyses on the divergent citizenship policies within Scandinavia have indicated deeply entrenched ideological positions as explanation for the current differences between the countries, but that enhancement of immigrant integration has been the main driver of this development (e.g. Brochmann and Seland, 2010;Jensen et al., 2017;Midtbøen, 2015). However, few studies have actually delved into the rationale of policy-making using an actor-centric approach, in which the elite actors who shape and define policies are studied directly. Building on Favell (1998) as well as insights from the literature on historical institutionalism and its theory of constraint (e.g. Campbell, 1998;Rothstein and Uslaner, 2005;Schmidt, 2008), the article analyzes elite-interviews with top-level bureaucrats, politicians and citizenship experts in the Scandinavian countries and explores how ideas about nationhood and integration have influenced the divergence in citizenship policies in the region, and what overall purpose the policy changes reflect. The informants in the respective countries, all relate to -and to a certain extent reflect -their state's 'philosophies': Sweden is portrayed as rightsoriented, modern and forward-looking, Denmark as nationally oriented, restrictive and backward-looking, while Norway combines immigrant integration and immigration control in its less ideologically anchored approach. However, our analyses suggest that, while the aim of citizenship policies in all three countries remains being 'political solutions to ethnic pluralism' (cf. Favell, 1998), immigrant integration does not appear to be the main goal of citizenship policies in either Sweden or Denmark.

The emergence of three national citizenship regimes
The citizenship policy divergence in Scandinavia is a fairly new phenomenon. In the long time span from 1880 to 1979, the three states cooperated closely in the field and, though not successful, even aimed at establishing a common Scandinavian citizenship in the mid-20th century (Brochmann, 2007;Midtbøen et al., 2018). Since the late 1970s, however, Norway, Sweden and Denmark have revised their laws partly or more extensively, leading to distinct differences among the three countries. Sweden started the latest major revision in 1997, passing a new, liberal bill in 2001, whereas Norway soon followed suit, implementing a new, semistrict law in 2006. Denmark has instigated a number of restrictive alterations in the naturalization requirements since the early 2000s, resulting in a highly strict citizenship policy (Midtbøen et al., 2018).
It is plausible to interpret this individualization of citizenship law-making as an upgrading of the issues at stake, politically speaking. National interests have become too important in this field to be subordinated to a common Scandinavian norm, suggesting that national sovereignty has intentionally been strengthened (Brochmann, 2007;Midtbøen et al., 2018). The significance of migration as a political and societal phenomenon lingers behind this change of atmosphere, as it underlies the politicization of the citizenship institution elsewhere in Europe (Bloemraad, 2017;Goodman, 2014;Joppke, 2017).
With increased migration -both immigration and emigration -the citizenship legislation has gradually gained complexity over the years. This article concentrates on three issues that tend to draw attention and contention in public debates, which also constitute significant dividing lines between the three countries: The required time of residence (waiting period), the civic integration requirements (language skills, knowledge about society, culture and polity, and economic selfsupport); and the question of dual citizenship. These criteria are conducive to the purpose of analyzing the open or hidden assumptions in the legislation on questions of nationhood, cultural/value requirements for membership and underlying ideas of integration. The expert interviews also centered on these themes.
As to waiting time, Danish rules require legal residence without substantial gaps for the last nine years. Sweden has the lowest residence requirement, five years, whereas Norway requires seven out of the previous 10 years. All three countries apply shorter waiting period for spouses of citizens, Nordic citizens, former citizens as well as stateless persons (Midtbøen et al., 2018). The extent of civic integration measures represents another dividing line: Again, Sweden is placed on the extreme liberal side, with no language skill requirement, no citizenship test, nor any requirement as to economic self-support. In practice, an applicant can be granted Swedish citizenship if s/he can prove identity, has reached the age of 18, holds a permanent residence permit, has lived in Sweden for the last five years and has shown 'good conduct', which basically alludes to criminality. Norway is next on the ladder, in providing citizenship (by right) if a person is 18 years old with an established identity, has seven years of residency, has passed (fairly easy) tests in the Norwegian language and in knowledge of society, and has shown good conduct. Denmark is the hardliner in the group, with stricter requirements on all dimensions as compared to the other two. Claims as to waiting period, language skills, societal and cultural knowledge, economic self-support and conduct (criminal record), are significantly higher than among its neighbors.
Sweden, as the first Nordic country, accepted dual citizenship in 2001. When Denmark followed suit in 2015, the decision was seemingly at odds with the consistent strictness in the country's approach to citizenship. However, the two main arguments in favor of dual citizenship were that Danish emigrants would be allowed to keep or retain their citizenship, and beside this, allowing for revocation of citizenship held by dual citizens if they engaged in terror actions. Consequently, the reform was hardly an act of liberalization (Midtbøen, 2019). Norway was the last state in the region, allowing dual citizenship from January 2020. As in Denmark, the arguments were mixed and the issue of revocation figured prominently (Midtbøen, 2019).
The increasing divergence in Scandinavian naturalization policies is striking. Why would institutionally and culturally similar countries with a longestablished cooperation in the field of citizenship choose such different paths?
Previous research has suggested that changes in citizenship policy-making has been spurred in order to enhance -by different means -immigrant integration (Brochmann and Seland, 2010;Midtbøen, 2015), yet no previous study has investigated the underlying policy rationale by exploring how elite actors in the three countries make sense of the divergent development since the turn of this century.

The institutional embeddedness of citizenship thinking
For a long time, research on citizenship was dominated by national models thinking, which aimed to explain how national differences in the legislation were the result of national traditions, institutional path-dependency and underlying ideas of nationhood (e.g. Brochmann and Seland, 2010;Brubaker, 1989Brubaker, , 1992Favell, 1998;Joppke, 1999;Midtbøen, 2015;Mouritsen, 2013). The 'civic turn' in citizenship studies (cf. Mouritsen et al., 2019) represented a break with this tradition, claiming instead that national policies were converging (Joppke, 2007). A heated debate about the extent of cross-national convergence has been ongoing ever since. Some argue for a continuation of distinct national models (e.g. Goodman, 2012Goodman, , 2014Mouritsen, 2013;Mouritsen et al., 2019), while others claim convergence, and that the basic rationale of reforms across countries is 'about furthering the integration of immigrants into mainstream institutions (policy goal), by means of mandatory language and civic knowledge courses and tests (policy instrument)' (Joppke, 2017(Joppke, : 1156. Taking as our point of departure the de facto policy differences that exist among the Scandinavian countries, we aim at digging deeper into the normative ideas underpinning citizenship policy-making. In doing this, we build on Favell's (1998) notion that national citizenship models may be viewed as reflecting different philosophies of integration. These philosophies are 'based on contrasting understandings of core concepts such as citizenship, nationality, pluralism, autonomy, equality, public order and tolerance ' (1998: 2) and constitute a set of more or less consensual ideas about the political solutions to 'ethnic dilemmas'. Favell's approach stands in contrast to the more static conceptions of nationhood well known from Brubaker's (1992) early work, yet it also differs from most contributions in the civic integration literature, as it focuses strongly on the defining role of elite actors. Indeed, Favell (1998: 25) claims that, in a situation of perceived political crisis related to immigration, national elites use 'the language of citizenship, integration and national identity' to form a new agreement on the formal terms in which newcomers are granted access to the nation, which later may provide a 'stable frame for dealing with immigration problems'. Rather than simply reflecting historical ideas about nationhood, citizenship policies may be interpreted as pragmatic solutions to perceived political problems that are anchored in ideas about integration.
Favell's notion of applied philosophies of integration serves as a bridge between the sphere of ideas and the institutional context on the ground. It also combines well with historical institutionalism and its theory of constraint. This theory provides tools for analyzing how ideas and institutions serve to limit the options policymakers take into consideration in their dealings with policy problems (Campbell, 1998). Tacit normative premises form the opportunity structures for elites; i.e. which ideas that are considered possible to convert into policy-makingthe 'logic of appropriateness' in the terminology of March and Olsen (2002). In practice, this means that, in order to have new policies, elites have to speak to prevailing norms and values of a society, which makes framing and timing pertinent. The institutions are carriers of ideas or 'collective memories' -changeable over time (Rothstein andUslaner, 2005, referred in Schmidt, 2008: 318). Campbell (1998: 381) extends the traditional historical institutionalism by emphasizing how ideas and institutions provide 'conceptual repertoires' that political actors can utilize to convince the public or other politicians through symbols and discursive schemas. These schemas pander to the values and sentiments that are prevalent in the public. Policy change may come about through deliberate modifications and re-combinations of existing institutional elements in new and socially acceptable ways.
So, what are the predominating 'collective memories' and the 'shared purpose' that bureaucratic elites, politicians and experts in the Scandinavian countries utilize when making sense of new citizenship policies after the turn of the century? What 'philosophies of integration' can be traced from the citizenship policymaking processes in Scandinavia -if any?

Data and method
The analyses presented in this article are based on in-depth expert interviews with top-level bureaucrats, politicians and citizenship experts in the three countries. The interviews took place during the fall of 2017 and winter 2018 in the three capital cities, Oslo, Copenhagen and Stockholm. Altogether, in-depth interviews with 22 individuals were carried out, divided between senior bureaucrats from the respective state administrations, politicians who were part of the processes around the turn of the century, as well as legal scholars and social scientists with expertise in citizenship issues. The interviews were semi-structured, lasting usually for about one hour and the respondents were promised anonymity in order to speak as freely as possible. This was particularly important for the state employees, who would otherwise hesitate to detail their views on policy processes in the past.
The interviews were structured by an interview guide with a set of questions standardized across all three cities, focusing on certain aspects of the historical trajectories of the three cases involved. The interview guide covered a range of topics related to citizenship, as well as detailed questions about each informant's role -if any -in the development of citizenship policies in the country of interest.
We asked specific questions about the processes leading up to the changes in each country's naturalization policy in the 2000s, including details about the key features of the legislation -dual citizenship, waiting time and civic integration requirements (knowledge and language tests, employment, etc.). We also asked the informants to reflect more generally about the meaning and status of citizenship in their country, and whether and how citizenship reflects more underlying ideas of nationhood and integration. Finally, the informants were encouraged to reflect on more recent changes in citizenship policies, including debates about citizenship revocation as a measure in the fight against terrorism.
Elite interviews differ from ordinary in-depth interviews in the sense that informants are recruited, not randomly and anonymously, but consciously for who they are and which position they occupy (Hochschild, 2009). In our interviews, we used an 'ideational' approach (cf. Jensen, 2019), focusing on the informants' views of the normative ideas and policy goals underlying the policies. We expected to find quite different narratives among the three types of informants recruited -politicians, bureaucrats and citizenship experts -yet the differences in accounts were mainly between countries rather than the type of informant. Even though citizenship experts hold a very different role than do politicians and bureaucrats, their analyses of the development were strikingly similar to accounts coming from the other two groups, indicating a fairly entrenched evaluation of the ideational foundation within the respective nation-states.

'Philosophies of integration' in Scandinavia?
Decades of cooperation between the Nordic countries in the field of citizenship has reflected common ground on the very basic perceptions of the institution; fundamental rights, ideas of belonging and ideas of nationhood relative to institutionalized structures of meaning. The baseline of the citizenship thinking in the region was part of the collective memories and shared purpose of the Nordic polity (Brochmann and Seland, 2010;Midtbøen et al., 2018;Wickstr€ om, 2017). Comprehensive immigration has, however, induced and crystalized major changes in this basic historical legacy. Political interests, interacting with institutional specificities, have induced a new dynamic within the region -in different ways and to different degrees -thus changing the ways in which citizenship policies are utilized as tools within broader political endeavors.

Sweden
When Sweden adopted its new Citizenship Act (which replaced the existing act from 1950), in 2001, it was based on an explicit wish to modernize the legislation and to implement international conventions on citizenship (Bernitz and Bernitz, 2006;Midtbøen, 2015). Most importantly, Sweden was first in the region to accept dual citizenship, but the new law also strengthened individuals' and particularly children's rights, by allowing citizenship by declaration to more groups (Bernitz, 2012: 19). Since 2001, only minor legal changes have been made. One of them was the introduction of a citizenship ceremony in 2006, which in 2014 became mandatory for municipalities to organize. Besides this, a new Article 1 in the law, defining 'the meaning of citizenship' (Medborgarskapets betydelse) was instigated (SOU2013: 29). These two provisions can be interpreted as an attempt to upgrade the Swedish citizenship institution by symbolical means (Midtbøen et al., 2018).
Consistent with the official image of Sweden as liberal and immigrant-friendly, the informants conveyed that citizenship explicitly connected to ideas about the nation have not been on the public agenda over the past decades. Indeed, across our interviews, the informants made statements such as 'citizenship has not been a matter of concern in Sweden' and 'nationality is not a big issue here'. Furthermore, Sweden's approach to citizenship issues was characterized as 'remarkably stable' in the period from the early 1980s to 2015, with an 'almost complete consensus' on the liberal approach across the political spectrum.
This liberal approach to citizenship is underscored by the strong emphasis on the rights dimension. As one of the bureaucrats said, 'everyone who wants to be a Swedish citizen should have the opportunity'. Integration requirements such as language skills, basic knowledge about culture or society or economic selfsufficiency were not discussed in the process leading to the new citizenship act in 2001. Consequently, the core of the Swedishness-matter is seen as rights, equality and togetherness across groups of the population. Citizenship revocation is not allowed on any grounds, illustrating the strong rights-based approach (Midtbøen et al., 2018).
Interestingly, while citizenship policies in many countries were politicized via immigration in the 1990s and early 2000s (Goodman, 20014), Sweden was less affected by this trend. Indeed, as questions related to national identity and integration re-entered the political scene in the late 1990s (and not least in the aftermath of 9/11, 2001), Sweden's national identity remained strongly anchored in ideas of 'cosmopolitan internationalism'. As one of the citizenship experts put it, 'Sweden's national identity is modern, forward-looking'. Or as another expert said: We've seen ourselves as the world's conscience, and this is rooted in each and every one of us. . . Rather than focusing on citizenship and national belonging, people have been gathered around the idea of Sweden's role in the worldthat's the identity.
One of our expert informants even explained that the resilience against restrictions, and particularly the unaffectedness by 9/11 on the national discourse, was caused by a 'hyper-investment in the idea about the Swedish regime of goodness': 'Enlightenment has been bottom line in Sweden [. . .]. The authorities have had a program for population molding -to socialize people into good ways of thinking.' However, times might be changing in Sweden. In the past few years, several Swedish (conservative) politicians have argued for a need to enhance the status of citizenship. Language tests and upgraded ceremonies have been suggested, in order to create a sense of cohesion (samh€ orighet). The central backdrop for these propositions is the migration crisis in 2015, which caused an extraordinary influx of refugees to Sweden, and the rise of the right-wing Sweden Democrats as a prominent political force. These two, probably interrelated, phenomena have brought citizenship into the political debate after decades in hibernation. As one of the bureaucrats pointed out: 'Now most politicians agree that it's necessary to foster some kind of common Swedish identity [. . .]. The Swedish society is quite divided into segments that have little contact, and large groups do not feel as participants in the society in which they live.' For two decades, the underlying ideas of Swedish citizenship policy wasaccording to our informants -characterized by a striking, liberal elite consensus that citizenship in the country should be hardly conditional. A central part of this 'philosophy' is that Swedish national identity, culture or traditions are illegitimate bases of citizenship, and that civic integration requirements are irrelevant measures for citizenship acquisition. Since 2015, the situation seems to have changed somewhat -the long-lasting refusal to discuss the emotional and cultural elements of citizenship, integration and belonging has gradually been giving in -yet the impact of this development is not clear. Whether a more open public debate on migrationrelated problems will result in a changed approach to citizenship and naturalization -a new 'elite formulation' (cf. Favell, 1998) -remains to be seen.

Denmark
Since the turn of the century, Denmark has implemented a range of changes in the requirements for naturalization, all of them pressing newcomers to take an active part in the integration process (Adamo, 2012). The restrictive trend started in 2002, in which the required time of residence increased, conduct conditions were tightened, and demands regarding Danish language proficiency and familiarity with Danish society, history, and culture were introduced. The trend continued in 2006, when the level of language skills was raised and new requirements concerning financial self-sufficiency and the passing of a citizenship test (indfødsretsprøve) were adopted. The election of a new center-left government in 2011 brought the restrictive trend in Danish citizenship policies to a halt (Ersbøll, 2015). The requirements of language and civic knowledge and self-sufficiency were relaxed, yet these relaxations were quickly annulled by the succeeding conservative government in 2015 (Midtbøen et al., 2018).
Across our Danish informants, there is consensus on the interpretation of the restrictive changes in previous years: Citizenship is seen as a reward, as the end station of a completed integration process. As one of the Danish experts put it: In Denmark, we don't talk about citizenship as a means of integration. We shall not have rules that spur the integration process. Enhedslisten [The Radical Left Party] would perhaps argue along those lines, but the vast majority in Parliament talks about citizenship as a gift, something to strive for, something you have to make an effort to get. In that sense, a Danish citizenship is a gift from the Danish people; from the Parliament.
The relationship between integration and national belonging is interesting in the Danish citizenship context. According to one Danish politician, the rationale of the strict approach to citizenship in Denmark is 'the creation of well-functioning liberal-democratic members of society, as well as preserving the nation'. He goes on to explain that 'it's not a goal to have as many as possibly to naturalize, what counts is knowledge and competence to be able to vote and participate. It's no hurry!' Similarly, an informant from the Danish state administration claimed that, in contrast to Sweden and Norway, . . . there has never been declared that Denmark wants people to get a Danish citizenship. [ . . . ] The politicians don't want to replace the majority population. It is and shall be difficult to acquire a Danish citizenship. If you can't speak Danish, you're not Danish. You're not Danish by heart and then you can't participate in the democracy.
The notion of being 'Danish by heart' inevitably invokes the question of the significance of the nation in Danish citizenship thinking. While the Swedish informants referred to Sweden's national identity as 'forward-looking', 'modern' and characterized by 'cosmopolitan internationalism', the Danish informants emphasize the importance of Denmark being 'a small and vulnerable nation', 'a ministate' historically subordinated to Germany. Particularly the Danish People's Party [DF] is described as viewing the 'citizenship as the core of the nation'. According to one informant in the state administration, 'The Danish People's Party talks about citizenship as membership in the nation [ . . . ] For DF, I think there's a concern for a dilution of Danishness.' For other parties, the nation is not as important as the focus on making individuals full-fledged members of Danish society, but proficiency in the Danish language seems to matter for all.
Although the question of citizenship in Denmark is described by our informants as extremely politicized -with new rules being issued almost annually and with changes in the naturalization rules when there is a change of government -a striking common interpretation is observed as to the value of the citizenship institution. It should take an effort to become formal part of Denmark, and the efforts are primarily to be made by prospective citizens through various means of integration. In striking contrast to the Swedish informants, the Danish experts all highlighted the importance of national identity, culture and tradition as underlying considerations in Denmark's demanding citizenship regime. Moreover, and somewhat paradoxically, although Denmark is among the countries in Europe with the toughest civic integration requirements, immigrant integration does not appear to be the main goal of Danish naturalization policies.

Norway
Norway adopted its new citizenship act in 2006, replacing the existing law from 1950, after a long-lasting process, involving a preparatory document (NOU, 2000: 32) with dissenting notes on, among other issues, the question of dual citizenship. The resulting new law signaled a mixed approach. On one hand, the principle of singular citizenship was maintained, naturalization was conditional on some level of language proficiency, and an oath of allegiance was introduced as an element in a (voluntary) ceremony for new citizens. On the other hand, all those who fulfill the conditions for naturalization were given the right to Norwegian citizenship, thus curtailing the discretionary power of the authorities (Midtbøen, 2015). Since 2017, it is required that applicants for citizenship must have passed tests in the Norwegian language and in knowledge about Norwegian history, culture and society. Furthermore, in 2018, the Parliament voted in favor of accepting dual citizenship, as in Denmark, partly to enable citizenship revocation of individuals who have been engaged in acts of terror (Midtbøen, 2019).
Norway's naturalization requirements are more restrictive than Sweden's, but more liberal than Denmark's. According to one of our informants in the state administration, the Norwegian middle way was intentionally established when the new citizenship act was discussed in the early 2000s: This was a point in time where Norway generally was in the middle, and we wanted to stay in the middle. [ . . . ] There was agreement that the Danish rhetoric went too far. And there were no close connections to the Swedish government. The Norwegian government did not want to make big changes; it wanted to stick to the status quo.
The internal divergence in the NOU committee largely reflected the main opposing views in citizenship debates both nationally and internationally. It appears from the interviews with members of the committee that the majority of the committee was convinced that their mandate was to propose a liberalization of the citizenship law, including dual citizenship: We were certain that Erna Solberg [the responsible Minister for citizenship issues at the time] was positive about dual citizenship, and she was, but we handed in the report only months before 9/11, and our interpretation was that this was the time when she turned around on this question. [ . . . ] 9/11 was detrimental for the outcome of the process. Suddenly, it was about showing loyalty to the country.
The committee majority had not prepared a very principled argument for their line of 'modernization' and liberalization. Their justifications came out with formulations strikingly similar to the equivalent Swedish Committee Report of 1999 (SOU1999: 34), stressing the inevitability of liberalization in an internationalized world. The committee minority, on the other hand, delivered a meticulous historical review, emphasizing the welfare state's significance for the building of trust and traditions of equal treatment. The citizenship institute was seen as part of this tradition in polity, also holding an important role in terms of signaling expectations for new citizens. Furthermore, dual citizenship was refuted by arguing that it would grant rights to immigrants that native citizens could not claim, thereby weakening the principle of equality.
The interviews with different committee members underlined, in retrospect, the likely importance of 9/11 as an explanation for the fact that the Government chose to lean on the argument of the minority position: 'Nobody saw citizenship as a means for integration at the time; it was part of the immigration policies. And to that end, one did not want to signal a liberalization by allowing dual citizenship.' Gradually since the turn of the century, stricter immigration control and integration requirements have become successively more central in Norway, however. According to one of our informants in the state administration: . . . the new law of 2006 had as a premise that people should become citizens when residents on a permanent basis. With the current conservative coalition government, this has changed into a more demanding attitude: Citizenship is a privilege and one should deserve it. This is a very profound change.
In recent years, Norway's naturalization requirements have gradually moved in a more restrictive direction (Midtbøen et al., 2018). Similar to Denmark, there are also indications of a gradual politicization of the issue in Norway. One of the top bureaucrats worried that too strong a focus on the control side of the citizenship complex might threat the necessary social cohesion: 'People living in the country permanently must be treated in a way that bolsters trust in the system. [. . .] That's the foundational force for creating cohesion in society.' There is a fear that more testing means 'elitization of the citizenship institute which again spells exclusion.' Or as another bureaucrat said: 'There is a risk that by not including the most marginalized, the problems to be eradicated will be reinforced.' Compared to Sweden and Denmark, where citizenship policies appear to have been driven by a stronger ideological impulse and more consensus about the respective main lines, Norway's approach does not seem to rest on a set of overarching ideas about the relationship between citizenship, nation and integration. Rather, Norway's pragmatic 'philosophy' aims at preserving the legitimacy of the institution both among the majority, through immigration control and moderate integration requirements, and by the minorities, by signaling that uptake in the national community is within reach.

Discussion
Contention over immigration has moved citizenship policies from the shadows of Scandinavian polity and into the public battleground. Worries over immigrant integration combined with difficulties in implementing effective border control has brought citizenship policies in as 'the last gate' approach, and both Denmark and Norway have introduced civic integration requirements to increase the exclusiveness of the citizenship institution (cf. Goodman, 2014;Mouritsen et al., 2019). Citizenship policies have the advantage -seen from the nationstate perspective -of still being subject to national sovereignty. Thus, as international law successively constrains the number of control parameters at the European states' disposal, e.g. through intra-EU mobility and humanitarian conventions, citizenship policies have gained importance in Scandinavia, as elsewhere in Europe (Goodman, 2014;Koopmans et al., 2012;Vink and Baub€ ock, 2013).
Building on Favell's (1998) concept of 'philosophies of integration', this article has dug deeper into the rationale behind these policy developments, by exploring how national elites in the three Scandinavian countries -top bureaucrats, politicians and citizenship experts -make sense of the particular development of citizenship policies in their respective countries since the turn of the century. Indeed, the interviews undertaken in this study reveal strikingly different ideas of the role that citizenship is meant to play in the three Scandinavian countries, and whether and how it is linked to broader concerns of integration. The interview data clearly suggests that an 'ideational' perspective on national models of citizenship remains relevant (cf. Jensen, 2019): The three Scandinavian countries build on different 'philosophies' in their citizenship policies (cf. Favell, 1998), both regarding the aims of the policy-making and the means to get there. The countries expose different national narratives about common purpose and collective memories; i.e. there are different cultural repertoires to utilize for politicians -a different logic of appropriateness (cf. March and Olsen, 2002).
The Danish informants' interpretations reveal a rather clear-cut approach to the citizenry, which is viewed in terms of cultural homogeneity. Danish nationhood values are liberal values, but newcomers will have to adjust in order to become full members of the citizenry. This may take a long time, and whether the molding process ends in naturalization is an open question. The Danish approach does not entail a close connection between naturalization and integration. Consequently, the citizenship policy can be restrictive -serving primarily a control function -as the adjustment process takes place independent of citizenship status. According to our informants, Denmark's citizenship policies are explicitly anchored in nationoriented rhetoric, and there is a near consensus across the political spectrum that prospective citizens must prove affinity with Danish culture and tradition before they are granted full access to the national community. In Denmark, the public philosophy builds on a conviction that integration takes time, as the goal of becoming 'Danish by heart' is ambitious and thorny, even arguably untenable in one generation. Thus, a plausible link is established to the short-term control logic in the immigration policy realm: The fewer newcomers the better, and the citizenship policy should signal this attitude through the longest possible waiting time, and demanding requirements. This hard-liner approach may contribute to the general necessity for immigrants in Denmark to stretch as much as possible to become well-functioning citizens. Yet in terms of integration, the empathizing is one-sidedly with the majority. Thus, the gist of the policy-making is in the dual communication; to the native majority population: 'we are preserving your ground values', and to the minorities: 'you have to do your utmost to become Danish, yet you may not succeed'. Since Denmark no longer is a homogenous society, this lopsided attitude may prove inadequate in generating trust and participation among new members of society. Social cohesion is a moving target, and the basis for generating trust changes with the increasing complexity of society (Laegaard, 2010).
The Swedish informants highlighted a very different approach to the relationship between naturalization and integration. Citizenship thinking in Sweden was coined before immigration was seen as a problem, and all informants agreed that it had remained at the outskirts of the immigration and integration discourse up until today. The policies have been guided by a liberal impulse and hardly any significant voices have opposed this baseline. Swedishness, in terms of ethnic or traditional culture, is consequently disregarded as relevant to citizenship acquisition. Integration is conceptualized as fostered 'organically' through welfare state policies, naturalization only being a buttressing mechanism on the road to full membership. The logic of citizenship policy-making is thus to extend substantial rights to newcomers at an early stage in this process. In Sweden, the majority needs to stretch as much as the newcomers to accommodate a new multicultural era. Integration may take time, but the implicit message has been that this is not a problem. The absence of civic integration requirements is a logical part of this thinking, the same goes for the short waiting period and Sweden's early acceptance of dual citizenship. In Sweden, the main communication is geared towards the majority, but the content of the message appears different from Denmark's: 'Behave and do take responsibility for the inclusion of newcomers.' After 2015, however, this clear-cut situation might be cracking. The informants disclose a slowly emerging change, in which Swedish politicians are increasingly troubled by the fact that their country is a liberal outlier in the European context.
In Norway, the informants reflected on the public policies in a more pragmatic way. What has characterized Norway's citizenship policies through the 2000s is weak and muddled communication on nationhood issues, yet increasingly emphasizing the need for a 'common glue', which from the governmental side has tended to cherish a vague concept of cultural tradition. The duality of the naturalization mechanism is present in their exposure: the perceived necessity of tightening control is countered by reluctance to believe in the conduciveness of stricter demands in order to accomplish integration. However, the informants predominantly underline the change of tide after the turn of the century (post 9/11) towards a more control-oriented naturalization policy. This shift gained momentum with the new government in 2013, with the immigration-critical Progress Party entering office. Previous governments expressed an inclusive attitude by stating that newcomers should naturalize, as a reflection of enhanced integration, yet this message has become more ambiguous with the 2013 Conservative coalition government. A hybrid solution has so far been seen as the appropriate approach. Seven years of waiting as well as moderate civic integration requirements of language and societal knowledge communicate to newcomers that citizenship is within reach. At the same time, the gradually more restrictive Norwegian citizenship policies communicates to the concerned majority population that the authorities have command: 'We take integration seriously!' Consequently, Norwegian citizenship policy reflects a dual communication: Control considerations are gaining more influence and they are increasingly explicit, yet integration considerations remain a basic feature, thus necessitating both comprehensive assistance and tests that are within reach, even for people with meager qualifications.
What do the identification of these highly different rationales for citizenship policy-making in the Scandinavian countries tell us? According to Favell (1998), nation-states' approaches to citizenship and immigration -their 'philosophies of integration' of which citizenship policies are part -are the results of negotiations among national elites who aim at finding solutions to acute policy problems. These solutions are pragmatic and do not necessarily reflect deep-seated notions of nationhood, yet they are shaped by the institutional histories, traditions and cultural repertoires (cf. Campbell, 1998) that are readily available in a given national context. Interestingly, it seems that Sweden and Denmark, more than Norway, have based their policies on an ideologically anchored elite consensus, albeit on very different grounds. The Swedish approach is ideological in the sense that it is consistently modern and forward looking, and based on a form of humanitarian cosmopolitism; yet at the same time, Swedish citizenship policy is stripped of any connotations for national identity, ethnicity or culture. As Ferna´ndez (2019) has pointed out, this does not mean that there is no such thing as Swedish nationalism or cultural traditions, but that such sentiments are detached from the citizenship institution.
By contrast, Denmark bases its citizenship policies on a backward-looking idea of national traditions as well as on a civic leitkultur (cf. Mouritsen, 2013). Despite the introduction of a series of demanding civic integration requirements in the Danish context, immigrant integration through naturalization is not the main goal. Rather, citizenship policies in Denmark appear to be part of a broader set of control mechanisms, which signal to asylum seekers and other humanitarian migrants that they are not welcome. Norway's hybrid model is more pragmatic and still in the making, perhaps moving towards a position of modernized nationalism, in which societal integration in the national welfare state is the key concern. In sum, although Favell's (1998) 'philosophies of integration' framework captures well the key differences in the Scandinavian countries' highly different approaches to citizenship policies, a striking finding emerging from this study is that the master concept which it all hinges on -integration -does not appear to be the driving rationale for citizenship development in either Sweden or Denmark.

Conclusion
This article has shown that the three Scandinavian countries embody quite distinct philosophies of integration and that these philosophies are applied differently in policy-making since the turn of the century. The expert interviews reveal that national specificities as to ideas of inclusion, recognition and participation have been significant for the reform processes of the respective nationality laws in the region. Yet indirectly, the interviews also reveal how the ideological, cultural and institutional legacies are deeply entrenched, thus inducing weighty pathdependencies politicians have to take into account. These legacies, in turn, represent both conducive repertoires and constraints when the political conditions necessitate new orientations or problem solving. At the turn of the century -and particularly after 9/11 (in Denmark and Norway) -new directions in the thinking around rights extension to newcomers were impelled. Citizenship policies were intentionally revaluated in this context, but as this article has shown, the policies were applied differently in the three countries.
Danish politicians had the easiest task in this respect, as the traditionally more nation-oriented philosophy facilitated justifications of a more restrictive turn. Sweden faced the most severe institutional and ideological constraints as a cultural repertoire of civic integration was close to non-existent. The legacy of nation building through democracy and welfare did not provide a political opportunity structure for demands making until -possibly -the migration crisis in 2015. The downside of the Swedish otherwise tolerant and liberal approach can thus be said to hinder reforms when serious societal problems appear. One may argue that newcomers in Sweden have been somewhat callously let alone in their long-term socialization without cultural guidance. The Swedish 'philosophy' -the available narrative of being modern, tolerant and forward-looking -thus seems to have become a constraint for effective reforms. Norway's pragmatic position in citizenship questions may prove to be quite flexible in practice. There are few clear-cut legacies to defend for varying governments, laying the ground for pragmatic adjustments. Citizenship policies can consequently be mobilized as tools in broader political endeavors -as in enhanced control policy -without considerable risks, politically speaking.
Despite the fact that changes in citizenship policies often depart from concerns over lacking immigrant integration, and that changes usually involve the introduction of civic integration requirements (cf. Bloemraad, 2017;Goodman, 2014;Joppke, 2007Joppke, , 2017Mouritsen et al., 2019), a major question arising from the analyses provided in this article is whether integration really should be considered the overall purpose of such policy change. Especially in Denmark, the gradual introduction of ever more demanding civic integration requirements appears to be a vital part of the broader aim of controlling the inflow of migrants and not to enhance the integration of those already present in the country. Conversely, the Swedish liberal approach to citizenship, which has regularly been analyzed as a steppingstone to societal integration, might in reality have been an indication of benign neglect. In Norway, immigrant integration has clearly been part of the underlying rationale for changes in naturalization requirements, yet the control dimension plays an increasingly important role there, too. The citizenship institution undoubtedly still plays a part in nation-states' efforts to solve their 'ethnic dilemmas' (cf. Favell, 1998), yet the underlying rationale appears more multifaceted than previously understood.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This article is part of a larger Scandinavian comparative project, Governing and experiencing citizenship in Multicultural Scandinavia (GOVCIT), funded by the Research Council of Norway (grant number 248007).