Article body

Introduction

Shortly after 9/11 I was approached by a former and deeply worried teacher-student. Some pupils in her secondary school-class had – in a situation where shock and disbelief were predominant – expressed feelings of understanding and support for the attacks. The teacher asked me what to do: How could she reach these pupils? My advice was rather vague; point out and discuss the evils of political violence in general and especially when it is used against innocent people. A week later a relieved teacher reappeared. Her pupils had changed their views completely. The attacks in New York and Washington were now seen as terrorist and totally repugnant. From the view of a political scientist and teacher trainer the change was comforting. Unfortunately it had nothing to do with the arguments I had suggested. Instead the transformation was linked to the pupil’s background in Iraqi Kurdistan. During the period between my two talks with the teacher a radical Islamist movement – with alleged links to Al Qaida – had entered Iraq across the border from Iran and tried to establish itself in the territory. Resistance from Kurdish armed forces had led to clashes, with a number of casualties among traditional Kurdish parties. Upon hearing this the pupils in Sweden simply draw the conclusion that if a group with links to Al Qaida attacks their homeland the same sort of organization must be condemned when it attacks someone else. For the pupils, born and raised in Sweden, what happened in Kurdistan was more important for their world view than what happened in the country where they live and go to school. This incident serves as starting point for reflections on prerequisites for the development of belonging and integration among young immigrants to their new polity.

As suggested below, an inclusive national identity can serve as a means for the development of belonging and integration. National identities will be discussed from the perspectives of young immigrants and young people within the majority population. Important questions to be considered are: What causes young immigrants to identify with the country in which they reside? and: To what extent is the majority population prepared to let them do that?

International migration challenges classical analyses of national identity (Cesarani & Fulbrook, 1996; Joppke, 1998). Classical analyses view national identity pertaining to a nation-state as the result of a merger between political commitment and national pride. National commitment and pride connected both to countries of origin and countries of arrival are potential essential components that shape new national identities. The identity-shaping taking place can be seen as processes of negotiation between the nation-state and representatives for immigrant groups (Kastoryano, 2002). Kastoryano uses ’negotiation’ in a semi-formal sense, as an ongoing process between formal and distinct parties, the state and immigrants’ voluntary organizations. Here the concept of negotiated identities will be used to discuss ’the bargaining offers’ of two informal parties, young immigrants and young people within the majority population. Their views on criteria for ’Swedishness’ will be seen as everyday expressions – in a context of secondary school experiences – of what constitutes a national identity (cf. Karner, 2005, p. 426). Special attention will be paid to express feelings of commitment to the country where you live and to criteria of ethnic inclusion and exclusion. Choosing to focus on young people creates the opportunity to explore attitudes on national inclusion and exclusion among individuals within the so called critical period (Sears, 2002). This period (approximately from 12 to 25 years of age) is of considerable importance in an individual’s life in forming views on politics and society (Pennebaker & Banasik, 1997). Attitudes of young people on issues of inclusion and exclusion thus provide an indication of future possibilities of upholding inclusive national identities. An inclusive national identity connects the principle of territory with criteria for citizenship that an individual can choose to attach her to, emphasizing the ability to use the majority language as a key criterion. Such an identity presumably facilitates societal integration of immigrants.

Data from a study (Lödén, 2005), in which 1000 secondary school students identify themselves ethnically and give their views on, inter alia, ’Swedishness’ and feelings of belonging to Sweden, will be used to discuss similarities and differences concerning national identity and identification, as well as inclusion and exclusion between those identifying themselves as ’Swedes’ and those who see themselves as something else. The empirical evidence presented shows encouraging results as to the readiness of Swedish youth to adopt inclusive criteria of ’Swedishness’, thus aiding those interested in identifying themselves as ’Swedish’. At the same time the evidence shows some cause for concern regarding the stability of this readiness and to the sense of belonging to Sweden among ’non-Swedish’ students.

The case and the setting

Sweden is generally not regarded as a classic country of immigration compared to, for example, the US, the UK, France, Canada or Australia. However, during the last 40 years the ethnic composition of Swedish society has changed in ways familiar to other European countries; Sweden has become a country of immigration. The percentage of the population born in a foreign country (12,1 per cent) is comparable to those of several traditional countries of immigration, such as the UK (9,1), France (10,7), Spain (11,1), the Netherlands (10,1) and the US (12,9). Canada (18,9) and Australia (20,3) still have a considerably larger portion of their populations born in foreign countries (United Nations 2006).

In the 1960s Sweden still was ethnically a rather homogenous country. Work force immigration from the 1950s through the 1970s and asylum immigration from the 1980s radically changed this. Work force immigrants came mostly from Finland, Greece and Yugoslavia. Former Yugoslavia, Iraq, former Soviet Union and Somalia has been the main emigrant countries for asylum-migrants coming to Sweden. Today approximately 20 per cent of a total population of nine million are of foreign origin (Statistiska centralbyrån, 2004). Thus, the Swedish experience offers a case of attitudes towards national identity in a rather recent country of immigration. This experience Sweden shares with several countries in Europe.

While earlier immigrants arrived to an expanding labour-market those arriving today often face unemployment. Rates of unemployment among immigrants are three to four times the size of unemployment levels for the population as a whole (Arbetsmarknadsstyrelsen, 2002). Immigrants are also faring worse than Swedes according to several socio-economic indicators. They often live in socially and economically declining areas and are to a higher degree dependent on social benefits (Socialstyrelsen, 2006). Those born outside Sweden receive disability benefits at considerably higher rates than Swedes (Svenska Dagbladet 15 December, 2005).

Perhaps even more alarming is the fact that immigrants are faring considerably worse than the population at large according to political and democratic indicators. Immigrants are showing lower, and declining, turnouts in both the general elections and in the municipal council elections. In the general elections in 2006 the total turnout was 82 per cent but only 67 per cent among foreign born. In the municipal council elections the same year the difference was even bigger; 78 per cent for Swedish citizens as compared to 37 per cent for foreign citizens. The latter figure in addition represents a significant decrease from 60 per cent in 1976 when the right for foreign citizens to vote in municipal elections was introduced (Statistiska centralbyrån, 2006). Foreign born citizens are to a lesser degree members of a political party (4,8 per cent as compared to 6,4 per cent for the population as a whole) and they have access to a daily newspaper to a far lesser degree; 72,6 per cent as compared to 51,3 per cent (Statistiska centralbyrån, 2005a & 2005b).

Several important indicators point in the same direction; immigrants seem to take part in the political life of their new countries to a far lesser degree than the majority population. If this pattern remains it will have implications for the democratic state.

The problem

This kind of situation is by no means unique to Sweden. Many European countries face questions of how to make immigrants, often coming from non-democratic countries with strong authoritarian traditions, feel like members and active participants in their countries of immigration. Riots among immigrant youth in France, Denmark and Sweden in recent years can be seen as dramatic expressions of more general and unresolved problems of integration and belonging well-known to both classic and new countries of immigration.

Integration is still mainly, despite globalisation and europeanisation, an issue for the nation-state. This is so for two reasons. The immigrants obviously have to live somewhere and that ’somewhere’ is always situated within a nation-state. Secondly, democratic political processes are and will, for the foreseeable future, be organized mainly within the context of the nation-state. In order to function satisfactorily the democratic state needs democratically minded citizens. It needs citizens who are willing to solve problems together, who are able to communicate in order to find solutions acceptable to as many as possible, and who are willing to follow the decisions being made. This, in turn, requires that enough people find it meaningful to identify with a nation-state like this. It also requires that those already identifying with this state are prepared to let those willing to identify, do so.

If, instead a large, and growing, portion of the population does not identify with the democratic state, it will have negative implications for the representativeness and legitimacy of this state; the state will become less representative of the population inhabiting its territory and, accordingly, seen as less legitimate by its inhabitants. Developments like this present a challenge, perhaps even a threat, to the continued maintenance of the democratic state. Such processes of alienation might be further aggravated if identification with the nation-state is made dependent upon criteria which are very difficult or even impossible for immigrants to meet. An interesting aspect of this is that individual experiences of fairness seem to be of importance in enhancing identification with the democratic state.

Before turning to the results of the study we have to consider, first, what makes identification take place and, second, what demands we should make upon a national identity that can be supportive to processes of identification with a democratic state.[1]

When does identification take place?

Identity needs objects of identification to come into existence. Objects of identification are used by the individual as symbols in order to understand the complex realities of society. For an identification to be made, ’the symbols have to be appropriate as a mode of behaviour and attitude for a particular and real experience’ (Bloom, 1990, p. 51).

’The nation’ can be such a symbolic object of identification. Thus, national identity is, according to Bloom, a condition ’in which a mass of people have made the same identification with national symbols – have internalised the symbols of the nation – so that they may act as one psychological group when there is a threat to, or the possibility of enhancement of, these symbols of national identity’ (Bloom, 1990, p. 52).

National identity comes into existence when the nation is considered a relevant object for human experience. Such experiences can be of the most divergent kinds, e.g. war or democracy. Accordingly, national identities can be very different. An identity taking common destinies, real or imagined, as its point of reference tends to emphasize lineage and ethnicity, while an identity taking common values as its point of reference tends to emphasize the principle of territory. The individual simultaneously can have multiple identities, personal and social. The identities can be more or less overlapping, with different emphasis depending upon the situation (Deaux, 1996). Here we are concerned with one of many possible social identities, national identity.

The creation of national identity is a constantly ongoing process. In this process the individual develops a sense of belonging to the nation. New generations are socialized into a changing but continuing, and often perceived as invariable, national identity. The possible achievement – or non-achievement – of such an identity among large groups of immigrants might be of crucial importance for the workings – even survival – of the democratic state. Different views of what constitutes the national identity can be seen as bargaining offers in a negotiation concerning the substance of such an identity. The views might be constructed out of experiences of, e.g., the importance of language, religion or ethnicity from countries of emigration and immigration. Experiences brought from the former can be seen as specific contributions of the bargaining offer of immigrant youth concerning the new national identity. The combined offers will, through continuous processes of negotiation, form views, more or less stable, on ’Swedishness’ and possible belongings to this ’–ness’. The achievement of a superordinate national identity is, I suggest, facilitated to the extent it is perceived as inclusive.

National identity as superordinate identity

National identity is the superordinate identity to focus on if we are interested in processes of identification with the nation-state. Superordinate national identities might also, as we have seen, be of very different kinds. What do we know about relationships between super- and subordinate identities? And what do we know about the effects of inclusion and exclusion related to an superordinate identity?

Creation of a superordinate identity, incorporating two or more subgroups, is a way of decreasing intergroup conflict, according to Spinner-Halev and Theiss-Morse (Spinner-Halev & Theiss-Morse, 2003, p. 524). Their claim is based on a review of the psychological literature on self-esteem and group behaviour. A superordinate identity has, at least, two important effects. It can minimize ’the differences people see between the ingroup and the outgroup’, and it can ’reduce competitiveness between groups by encouraging members to be less concerned about the relative gains of the ingroup versus the outgroup’. Spinner-Halev’s and Theiss-Morse’s conclusion is that ’[p]eople who share a superordinate identity tend to be more concerned with procedural justice than with distributive outcomes. If the process is fair for all superordinate group members, then members do not focus on subgroup identity’ (Spinner-Halev & Theiss-Morse, 2003, p. 524).

’Fairness’ seems to be a crucial aspect of the superordinate identity if it should work as a rallying point for different subgroups. This observation is supported by evidence presented by Citrin, Wong and Duff. Based on survey data regarding views of different ethnic groups living in the United States on ’Americanness’, they conclude that ’… the evidence… points to the advantages of a common sense of American identity founded on the realization of equal citizenship’ (Citrin, Wong & Duff, 2001, p. 97).

A Spanish study shows the importance of a superordinate identity for reducing bias among subgroups sharing this identity (Ros, Huici & Gómez, 2000). In the study on relationships between national identity and regional identities in Spain, the authors conclude that ’… an effective way of reducing ingroup bias is to create a common ingroup identity… [since] among those who still share a common Spanish identity there is a reduction of ingroup bias’ (Ros, Huici & Gómez, 2000, p. 94).

Brewer draws our attention to the salience of the symbols of a superordinate category membership. If this is at hand positive recategorization of outgroup and ingroup into a common superordinate identity can take place (Brewer, 2001, p. 36). Such an identity can, on the other hand, be seen as threats if ’intergroup attitudes and relations have moved into the realm of outgroup hate or overt conflict’ (Ibid).

Brewer’s remark indicates that aspects of, e.g., fairness have to be salient features of a superordinate national identity in order to attract the interest of the potentially excluded. This, in turn, brings us to the question of what characteristics the superordinate identity should have in order to be attractive. The concepts of ’inclusion’ and ’exclusion’ will be useful in doing this.

Inclusion – exclusion

Inclusion and exclusion have been used to characterise what we know as the two main types of citizenship, a political, inclusive one, historically strongly connected to France, and an ethnocultural, exclusive ’German’ type. While these certainly are ideal-types in a Weberian sense and should not be understood as accurate descriptions of actual conditions, they may be used as a point of departure in discussing eligible forms of national identity[2].

A national identity founded on fairness, e.g. the realization of equal citizenship may, as we have seen, be conducive to its power of attraction. An inclusive national identity undoubtedly comes closer to these values than do an exclusive. Such an identity, connecting the principle of territory with certain values, provides the individual with the possibility and choice to attach to it according to criteria the individual himself can, at least in principle, control. Inclusive criteria can be, e.g., to be able to speak the dominant language of the country where you live, respect of the country’s political institutions and laws and to feel as a member of the country where you live[3]. Corresponding exclusive criteria might be to have been born in the country where you live, to have lived in that country for most of your life and to be a follower of the dominant religious faith.

Obviously the first three criteria are possible to reach for almost everyone, although practical obstacles almost impossible to overcome for the individual may of course exist. An illiterate immigrant of 75 years of age certainly will have problems in having a good command of the language of the new country. But there is no principled obstacle that excludes him or her from the possibility of mastering the new language. The opposite applies to the three exclusive criteria. All three describe conditions the individual cannot reasonably influence. The born- and lived in-criteria are obviously of this kind. So, if less obvious, is the criterion of religion since faith can be such an important aspect of the individual’s identity.

Language constitutes a criterion of special interest. A widespread and fairly equally distributed (socially and ethnically) command of the dominant language is, I contend, crucial for the democratic state. A national identity founded on fairness, i.e. an identity oriented towards civic values such as freedom, democracy and equal rights, invites anyone who wants to associate with that identity. But this does not mean that the democratic state should be neutral to different cultures and social identities based upon these cultures within its territory. This argument is developed by Poole (Poole, 1999). He claims that,

’[T]he fact that modern states favour one culture over others is not a mere contingency, but is essential to their practice... [I]t is likely that any State which is appropriate for a modern industrial market society will provide a unified educational and administrative system which will inevitably favour one culture over others. Further, and this is a separate argument, there are good reasons why this should be the case, at least in a liberal and democratic state. Liberalism requires that the State define and protect a range of rights for the citizen; that it provide due legal process through which its members may secure justice; that it conduct its political affairs in ways which are open to scrutiny and criticism; and that it provide an education which allows its citizens to participate in basic social, economic and political activities. In a democratic state, political matters must be open to widespread discussion in which all citizens are able to participate; and citizens must be able to take cognisance of the views and interests of as many other groups as possible.’

(Poole, 1999, p. 121)

As can be seen language, according to this argument, becomes the key to the workings of the democratic state. Citizens must be able to communicate on an equal basis, with each other and with society at large. The ability to use the dominant language of a society is, therefore, not only an important inclusive criterion; it is, in fact, absolutely crucial to the continued up-holding of a democratic state. This, in turn, requires two things. Speakers of the dominant language must allow those who are interested the right to learn this language in order to facilitate their equal participation in society. And, secondly, those who are living on a permanent basis in the country and do not speak this language, must show a manifest intention to learn it.

This brings us back to the study which presents the empirical foundation for the article.

Methodology and data

Data collection for the study was carried out during the autumn semester 2003. A questionnaire of 65 questions was distributed to secondary school students, aged 16-19, in twelve secondary schools in eight municipalities across Sweden. Municipalities and schools were chosen to represent a national average. However there was a slight overrepresentation of larger cities and a corresponding underrepresentation of rural areas in the sample. Considering Swedish housing patterns, this means that there probably is a slight overrepresentation of persons of immigrant origin in the sample. The questionnaire was administered by teachers in classroom settings and was answered by 1034 students. The external drop-out rate was close to nil.

Secondary school students were chosen as objects of the study for two reasons. First, they fall within the so called critical period. Important life events taking place when you are between, approximately, 12 and 25 years of age can be of great importance for your future opinions on society and politics (Pennebaker & Banasik, 1997). One such critical event, taking place during the time spent in secondary school, is participation in general elections. Three out of four students can take part in the elections for the first time (in Sweden you are entitled to do this from your eighteenth birthday). This, in turn, might make people of this age consider their views on society, their own and other persons roles in society and, last but not least, whether they should see themselves as members of that particular society. Second, since 95 per cent of Swedish youth attend secondary school, these students are the citizens of the future. Thus, their opinions have obvious implications for the possibility of maintaining the democratic state.

The study

The survey included, inter alia, questions regarding ethnic self-identification, importance of specified criteria for ’Swedishness’, the sense of belonging to Sweden and preparedness for political engagement. The results on self-identification and criteria for ’Swedishness’ will be presented and discussed before we turn to the other results.

82 per cent of the students identified themselves as ’Swedes’, 12 per cent as something else. The drop-out rate regarding this question was 6 per cent. Within the 12 per cent group one third of the students used a ’mixed label’ as their ethnic description (e.g. Swedish-Finnish or Kurdish-Swedish). The ’mixed label’-group is, as will be shown later, of special interest in discussing the implications of the results as pertaining to feelings of belonging and the possibility of inclusion into the nation-state.

Asked to specify the importance of the earlier discussed, six specified criteria for ’Swedishness’, a majority of the students expressed support for inclusive criteria (see Table 1).  

Table 1. What is important for being a ’real Swede’? Results are shown for inclusive (1-3) and exclusive (4-6) criteria (per cent).

Statement

Agree or not strongly agree

Disagree or strongly disagree

Dot not know

Total

1 To be able to speak Swedish

(N=1030)

93,2

6,5

0,3

100

2. To respect Swedish political institutions and laws

(N = 1028)

88,6

10,5

0,9

100

3 To feel Swedish

(N=1029)

75,8

21,4

2,8

100

4. To have been bord in Sweden

(N = 1030)

50,2

48,8

1,0

100

5 To have lived in Sweden most of one’s life

60,9

38,5

0,6

100

6 To be a Christian

(N=1025)

8,0

89,4

2,6

100

Source: Lödén, 2005, p. 81

As can be seen the importance given to each of the three inclusive criteria is bigger than for any of the three exclusive criteria. Language and respect for political institutions and laws are by far those considered most important. The general support for inclusive criteria is encouraging given the desirability of an inclusive superordinate national identity. Given this perspective opinions on these matters of ’the potentially excluded’ is of special interest. How important do students who identify themselves as something else than Swedes judge the different criteria? In table 2 we can see the results as divided between ’Swedes’ and ’non-Swedes’.

Table 2. What is important for being a ’real Swede’? Results are shown for self-identified Swedes (S) and non-Swedes (nS) for inclusive (1-3) and exclusive (4-6) criteria (per cent).

Statement

Agree or not strongly agree

Disagree or strongly disagree

Dot not know

1 To be able to speak Swedish

(N=1030)

S

93,5

nS

93,6

S

6,2

nS

5,6

S

0,3

nS

0,8

2. To respect Swedish political institutions and laws

(N = 1028)

90,0

83,2

9,2

14,4

0,8

2,4

3 To feel Swedish

(N=1029)

78,2

56,3

18,7

40,4

3,1

3,3

4. To have been bord in Sweden

(N = 1030)

50,8

40,8

48,3

56,8

0,9

2,4

5 To have lived in Sweden most of one’s life

61,2

53,9

38,2

44,4

0,6

1,7

6 To be a Christian

(N=1025)

6,4

17,7

91,0

79,0

2,6

3,3

Source: Lödén, 2005, p. 90

The ability to speak Swedish is still considered, by Swedes as well as non-Swedes, the most important aspect of being Swedish. There are, however, statistically significant differences between the groups concerning the other two inclusive criteria (level of significance 95 per cent, t-test; eta: ethnicity vs. respect for institutions: ,066; ethnicity vs. feel Swedish: ,177). In the first case – respect for Swedish political institutions and laws – the difference is significant but, still, rather small. The difference should not be interpreted as if non-Swedes paying less importance to this criterion were less law-abiding than Swedes. A more plausible interpretation can be made by taking into account findings highlighting the importance of fairness, presented by Kumlin (Kumlin, 2002). He shows that personal welfare state experiences have important effects on the individual’s sense of political trust. Experiences of having been treated ’fairly’, i.e. ’what a person such as me in this situation has the right to expect’, seem to enhance trust in the political system. Personal experiences falling short of fairness expectations, on the other hand, have negative effects on political trust (Kumlin, 2002, p. 307f). The fact that immigrants are faring worse than the majority population on socio-economic and political indicators therefore could be interpreted by individuals of foreign origin as unfair, thus leading to non-Swedish students showing lower levels of agreement to the criterion of law abidance.

The second inclusive criterion where difference between the groups is significant – to feel Swedish – is not surprising. If you define yourself as non-Swedish the inclusiveness of the feel Swedish-criterion is probably less obvious than if you define yourself as Swedish. The enticement to feel Swedish is simply less.

For exclusive criteria there is a significant difference between Swedes and non-Swedes regarding the importance of being Christian (level of significance 95 per cent, t-test; eta: ,140). While the importance attached to this criterion is by far the lowest among both groups, the difference between Swedes and non-Swedes is considerable and interesting. Sweden, in the self-image of Swedes, is considered one of the most secular countries in the world. Holidays, including those with a distinct Christian origin such as Christmas and Easter, often have a pronounced secular character, especially in the private sphere. Still, this does not exclude the possibility that non-Swedes interpret such celebration as ’Christian’ and perceive Sweden as more religious than Swedes do. This, in turn, indicates that Western secularism is culturally embedded to a higher degree than generally perceived.

The great importance non-Swedes attach to language does not automatically mean that they perceive the be-able-to-speak-Swedish criterion as inclusive. Of course they can judge it as both important and exclusive. The results on the students’ feelings of closeness and belonging to Sweden and on political engagement paint, as we will see, a mixed picture concerning possible interpretations of attachment to the language criterion.

Markedly different levels between Swedes and non-Swedes regarding sense of belonging to Sweden point in the direction of interpreting the latter emphasis of the language criterion as a sign of exclusion. The different levels, as shown in the first two rows in table 3, are statistically significant (level of significance 95 per cent, t-test; eta: ,101).  

Table 3. Sense of belonging to Sweden, for self-identified Swedes and non-Swedes (per cent).

Question

“How close do you feel to Sweden ?’

Close or very close

Not close at all or not very close

Can’t choose

Total

Swedes

62,4

33

4,6

100

Non-Swedes

(N=973)

48,3

47,5

4,2

100

Source: Lödén, 2005, p. 73

Non-Swedes’ lower level of a sense of belonging comes as no surprise. If you identify yourself as something else than ’Swedish’ your sense of belonging to Sweden is, for some reason, low. A possible explanation for this could be, in accordance with the argument of Kumlin presented earlier, experiences of ’unfair’ treatment.

On the other hand, results concerning non-Swedish students’ political engagement point in the direction of interpreting the language criterion as inclusive. Attitudes toward political engagement were studied in order to find out students’ preparedness to take responsibility for activities necessary to maintain the democratic system. ’Political engagement’ was operationalised into ’political interest’, ’present membership in political organization’ and ’future membership in political organization’.

For two of the three indicators – political interest, future membership – non-Swedes were to a higher, and statistically significant, degree more politically engaged than Swedes. For the third indicator – present political membership – there was no difference between the groups.

To the extent we understand political interest and readiness for membership in political organizations as indicating a general interest in society and its future development, the results concerning non-Swedish youth are encouraging. They seem to be no less interested in the future of Swedish society than Swedish youth are[4]. This preparedness for future political engagement supports interpreting the importance of the language criterion as inclusive. The reason is obvious – in order to take part in political life you must master language. As we can see there is some ground for optimism concerning the possibility of immigrant youth being included in a superordinate national identity.

Still, there are several reasons to be cautious. The first reason concerns the stability among the majority of students as to their readiness to attach to inclusive criteria for national identity. There certainly is a majority opinion for inclusion, but this is by no means overwhelming or self-evident. In terms of percentages the study shows that approximately 55 per cent of the students tend to attach to inclusive criteria, 45 per cent to exclusive. The overriding concern, though, has to do with the factual societal integration of immigrant youth. Success or failure concerning this will have decisive implications on the development of a sense of belonging to society and country.

Conclusions

Two questions were raised initially; ’What makes young immigrants identify with the country where they live?’ and ’To what extent is the majority population prepared to let them do that?’ Returning to the questions we can see that both evidence from previous research and results from the presented study can help in formulating answers.

A superordinate, inclusive national identity is an important vehicle for reducing bias among subgroups within the identity, decreasing intergroup conflict and creating a common sense of belonging (Ros, Huici & Gomez, 2000; Spinner-Halev & Theiss-Morse, 2003; Citrin, Wong & Duff, 2001). Experiences of fair treatment seem to be decisive for feelings of attachment to a superordinate national identity. Inclusive criteria of nationhood can be understood as expressions of fairness. Swedish and non-Swedish students to a large extent express similar attachment to inclusion. But this is not enough in order to make non-Swedes’ feeling of belonging to Sweden reach the same levels as those expressed by Swedes. Such criteria, thus, can be understood as a necessary but not sufficient condition for an identification to take place. In order to be viewed as trustworthy the rhetoric must be complemented by material and substantial evidence from the society at large concerning fair chances to establish oneself as a full member of society.

The majority population is to a considerable extent prepared to let immigrants identify with the country where they live. In the bargaining process concerning the superordinate national identity between young people of Swedish origin and young people of foreign origin, a majority of Swedish secondary school students attach to inclusive criteria for ’Swedishness’. This facilitates the identification as ’Swedish’ for those willing to do so. The observation is not based upon imaginations of any ’naturalness’ of national belonging. It comes from the belief that democracy is and will in the foreseeable future, be organized mainly along the lines of the nation-state. Thus; if we are interested in the maintenance of democracy, young people’s identification with the democratic nation-state is a major concern.

In this perspective the group of students presenting themselves with a mixed label (e.g. Swedish-Kurdish) constitutes an interesting challenge for future research. They do so for two reasons. First, the group represents actual states of national identification – individuals do have mixed origin and migration will make this even more frequent. The mixed label can be seen as the national identity equivalent to dual citizenship within the judicial sphere. Second, it would be of interest to find out what makes an individual decide to define herself as ’Kurdish-Swedish’, not as ’Kurdish’ alone. Are there specific factors that can help us understand why some take while others do not? And how are such factors related to experiences of fair/unfair treatment, senses of belonging and perceptions of the relationship between the individual and the nation-state where she lives? Good answers to questions like these would further our understanding of what makes young people identify or not with the democratic nation-state.