The making and unmaking of ethnic boundaries in the public sphere: The case of Norway

Introduction

A growing body of scholarly literature is concerned with the conditions for free speech in democratic, multi-ethnic societies (e.g. Bleich, 2011; Gelber, 2011; Maitra and McGowan, 2012; Maussen and Grillo, 2013). One strand of this literature employs a legal-normative perspective by addressing the constitutional and legal limits to public hate speech and the extent to which such limits are justifiable in light of principles of liberal democracy (e.g. Bader, 2013; Bleich, 2011, 2013; Herz and Molnar, 2012; Waldron, 2012). Another strand of research, often building on some version of postcolonial theory or the notion of Islamophobia, is more concerned with how minorities' experience participation in the public sphere in practice, regularly demonstrating that ethnic and religious minorities are targets of racist or anti-Muslim sentiments, harassment and even verbal and physical threats (e.g. Bangstad, 2013, 2015; Eide, 2010).

This article positions itself within the latter strand of the literature, but departs from previous studies by applying theories of ethnic boundary-making (e.g. Alba, 2005; Wimmer, 2008, 2013) to this field of research. Drawing on in-depth interviews with active, media-experienced individuals with various ethnic and religious minority backgrounds regularly participating in public debates, I explore when and under what conditions ethnicity and religious background become salient in determining minorities’ experiences with the public sphere. The interviews demonstrate a range of experiences and possible positions, and suggest that important processes of change are taking place. Several individuals, particularly those of the ‘second generation’, are able to strategically cross existing ethnic boundaries by transcending the ‘minority box’ and inhabiting a space in the public sphere defined by individual merits rather than ethnic or religious background.

The theoretical implication of this finding is that researchers in this field need to be cautious in assuming that race, ethnicity or religion are principal determinants of opportunities and rather highlight the situational character of ethnic identity by exploring the individual ability to change a given situation (Tajfel, 1978; Wimmer, 2008, 2013). This does not imply that every individual of ethnic or religious minority background experience similar opportunities of boundary-crossing, or that the blurring of ethnic boundaries on a societal level will occur without conflict or negative reactions from parts of the majority society. Rather, it suggests that one has to give careful attention to the variety of boundary-making strategies, as Wimmer (2013: 42) has pointed to, as well as to occurring processes of change. By focusing empirically on the making and unmaking of ethnic boundaries in the mainstream public sphere in Norway, the article fills a gap in the growing literature on free speech in multicultural societies, which too often has been concerned with either normative ideas of how the public sphere works or fixed conceptions of the all-encompassing significance of race, ethnicity or religion.

Theories of ethnic boundaries

The importance of social boundaries in defining processes of inclusion and exclusion has received increasing attention in recent years (e.g. Lamont and Molnár, 2002). The specific branch of this literature which focuses on ethnic boundaries usually takes as its point of departure the work of the Norwegian social anthropologist Fredrik Barth. In his introduction to Ethnic Groups and Boundaries (1969), Barth famously departs from the dominant Herderian conception of ethnic groups as equal to groups who have a shared identity, a common history and a traditional cultural heritage. Instead he introduces a constructivist conception of ethnicity by defining ethnic identities as boundaries made in the interplay between self-identification and the attribution of characteristics by others. Accordingly, ethnicity or ethnic groups are not fixed entities given once and for all, but rather constructions of everyday interaction created through boundary-making processes internally and externally, and which may change over time.

Much scholarly work has later built on, discussed and developed Barth's relational definition of ethnicity (e.g. Eriksen, 2002: Chapter 3; Handelman, 1977; Jenkins, 1997). One strand of this literature has developed the concept of ethnic boundaries into a relevant lens for studying how immigrants and their descendants fare in European societies. Writing from the perspective of immigrant integration Bauböck (1994) and Zolberg and Long (1999), for example, distinguishes between ‘boundary crossing’ and ‘boundary blurring’. According to Bauböck, boundary crossing occurs when individual migrants succeed in transcending the distinction between insiders and outsiders while leaving the existing structure of ethnic relations in the receiving society unaffected. Boundary blurring, on the other hand, takes place when the structure of the receiving society changes in important aspects. Zolberg and Long (1999) add a new dimension to the processes of boundary change; what they coin ‘boundary shifting’. According to these authors, boundary shifting involves a reconstruction of a group's identity in which the differentiating line between members and non-members are relocated.

Distinguishing between ‘bright’ and ‘blurred’ boundaries, Alba (2005, 2009) has further advanced theories of the changing potential of ethnic boundaries. When ethnic boundaries are bright, for example due to differences in formal status or language skills, who belongs to the mainstream majority group and who are members of ethnic minority groups is indisputable. Blurred boundaries, on the contrary, refers to a situation in which the boundary between the majority and ethnic minority groups is more ambiguous. This is typically the case for the second generation who usually are citizens of their parents’ destination country, speaks the majority language fluently and often have acquired the dominant cultural codes through education and general socialisation, and thereby are part of the majority community in a way that their parents often did not achieve. Still, the second generation often maintains ties to their own ethnic or religious group, for example through bilingualism, marriage practices, remittances, preservation of the family's cultural traditions and inter-generational, informal interaction and may thus identify with an ethnic or religious minority group. Further, even if they wish to achieve full membership in majority society, exclusion on the basis of ethnic or religious background may still contribute to maintaining – or clarifying – the ethnic boundary. This underscores the idea that ethnicity is created in the interplay between self-identification and attribution of identity from the outside, as Barth (1969) observed, and as Jenkins (1997) has further developed. And it shows how ‘the nature of the boundary affects fundamentally the processes by which individuals gain access to the opportunities afforded the majority’, as Alba (2005: 22) puts it.

Drawing on the work of among others Barth (1969), Bourdieu (1991), Brubaker (2004) and Alba (2005), Wimmer (2008, 2013) highlights the need for a disentangling of the ethnic and non-ethnic mechanisms leading to the outcome of interest and to employ a ‘de-ethnicized’ analytical lens. According to Wimmer, much research on ethnicity implicitly employs a Herderian concept of groups, which takes for granted that ‘dividing society into ethnic groups is analytically and empirically meaningful’ (Wimmer, 2013: 10). Instead, processes of ethnic boundary-making should be viewed in light of how actors strategically involve in struggles over what boundaries are relevant, for example by challenging hierarchical orders of ethnic categories, by changing one's own position by individual boundary crossing or simply by insisting on alternative forms of belonging.

Applying theories of ethnic boundaries to studies of the public sphere

For empirical research, theories of ethnic boundary-making indicate the need for paying particular attention to the variety of strategies used by actors to manage, challenge or overcome ethnic boundaries. As a locus of ethnic boundary-making processes, the public sphere is a particularly powerful site. How individuals of minority backgrounds manage strategically, how they are categorised by others and the extent to which they are ‘locked’ into ethnic categories when engaging in public debates have a strong signalling effect. For various ethnic communities the experiences of ‘co-ethnics’, who they might view as representatives or spoke-persons, may serve as a proxy for social inclusion and determine whether they will engage in public matters in the future. For the mainstream society, the presence of individuals of ethnic or religious minority background that succeed in claiming influential positions in the public sphere may contribute to making stereotyped images of ‘immigrants’ or particular ethnic or religious groups harder to maintain. However, such an effect will depend on the way these individuals manoeuvre in the public sphere and the extent to which they are viewed as exceptional ‘outliers’ or actual representatives of minority groups.

In this study of how individuals of ethnic and religious minority background fare when participating in the Norwegian public sphere, I follow Wimmer and employ a ‘de-ethnicized’ analytical lens, exploring in what ways the informants’ experiences are shaped by ethnicity or religion, or rather by other dimensions like their gender, ideological position and political stance in controversial issues. Of course, it is difficult, if not impossible, to determine the exact impact and relative importance of these various dimensions of social identity, which may also overlap (Jenkins, 1997). This difficulty does however not imply that one should refrain from analytically separate ethnicity or religion from other ‘grounds for difference’ (Brubaker, 2015), which may prove more important for individual experiences in concrete social situations. Rather, it implies a careful consideration of the strategies used by these individuals when they make their ways in a mediated public sphere structured by editors who ultimately decide who are granted access to the public sphere, and on what terms this access is granted. In employing theories of ethnic boundary-making to a study of the dynamics of the Norwegian public sphere, I highlight whether ethnicity and religion constitute bright boundaries defining the terms of participation, whether single individuals experience that they have the power to address and criticise the existence of ethnic hierarchies and the extent to which they are able to cross existing boundaries and be accepted as full-fledged members of majority society. In the end of the article, I return to the question of whether experiences of boundary-crossing at the individual level are testament of broader processes of boundary-blurring at the societal level.

The Norwegian context

Historically a country of emigration, Norway did not become a net immigration country until 1967, when the country experienced a substantial influx of labour migrants from Pakistan, Turkey, Morocco and India. Still, it has a long history of immigration, dating back to the integration of the Norwegian state around the year 900 (Brochmann and Kjeldstadli, 2008). In addition to the indigenous Sami population who resided in the territory before the majority arrived, five groups – Jews, Kvens, Forest Finns, Rom and Romani – have status as national minorities in Norway today, and they have their particular history of migration – for some groups dating back hundreds of years. ¹

In 1975, Norway introduced a moratorium on labour migration in tandem with the rest of North and West Europe. Still, the number of immigrants continued to grow, to some extent because particularly highly skilled immigrants from outside Europe were exempted from this rule particularly and recruited to fill positions in the booming oil industry, but primarily due to increasing humanitarian and family migration (Brochmann and Kjeldstadli, 2008). After the EU enlargement in 2004 and 2007, the number of labour migrants, primarily from Eastern Europe, has moreover grown rapidly. By 1 January 2015, immigrants and their children made up 15.6% of the Norwegian population. Polish labour migrants constitute the largest immigrant group in Norway, followed by immigrants from Sweden, Lithuania and Somalia. Among the second generation, individuals with Pakistani origin make up the largest group, followed by descendants of Somali and Iraqi immigrants (Egge-Hoveid and Sandnes, 2015).

In terms of how immigrants and their children fare in Norwegian society, the evidence is somewhat mixed. On the one hand, the overall labour market outcomes of immigrants in Norway are relatively favourable when compared to other countries (OECD, 2012). However, there are large differences between different immigrant groups in which some are employed at higher rates than the native majority while others have very low employment rates. In sum, there is a considerable, overall employment gap between natives and the immigrant population that seem to persist over time. For the second generation, the situation is more optimistic. On average, Norwegian-born children of immigrants are enrolled in higher education to a larger extent than majority peers at the same age level and they have higher employment rates than the first generation, although somewhat lower than for the majority population at the same age (Olsen, 2016). On the other hand, both immigrants and their children face problems in Norwegian society, not least in terms of discrimination. A survey on perceived discrimination among immigrants from 10 different countries showed that more than 50% reported having experienced discrimination in at least one area of social life (Tronstad, 2009). Recent field experiments have confirmed that individuals with Pakistani or Arabic-sounding names indeed are discriminated against when trying to access the labour market or the rental housing market compared to ethnic Norwegians with equal credentials (Andersson, Jakobsson et al., 2012; Birkelund et al., 2014; Midtbøen, 2016; Midtbøen and Rogstad, 2012).

Throughout the 2000s and particularly since the Muhammed cartoon controversy in 2006, a recurring debate in Norway – as elsewhere in Europe (Klausen, 2009; Lindekilde et al., 2009) – has involved the notion of free speech and its limits. In this debate, the conditions under which individuals of ethnic and religious minority background can access the public sphere, and the role racism and Islamophobia play in structuring these conditions, have been central issues (e.g. Bangstad, 2015; Midtbøen and Steen-Johnsen, 2016). In parallel with this debate, a number of new voices, often, but far from exclusively, young Muslims, have entered the Norwegian public debate. These young individuals tend to contrast their experiences to former ‘spokespersons’ from their parents’ generation who are seldom viewed as representative for the new generation (Andersson, Jacobsen et al., 2012).

How these individuals fare when engaging in public debates is, however, less known. Previous research has suggested that Norwegian media tend to favour Muslim voices who serve as potential ‘modernisers’ and ‘reformers’ of Islam, forming a hierarchy which privileges certain positions and points of view over others (Bangstad, 2013, 2015; Gullestad, 2006). Whether or not this is the case, how the situation for individuals of Muslim background resembles or differs from other religious groups, and whether the distinction between visible and non-visible minorities is relevant in determining the access to public positions and terms of participation, are empirical questions. By exploring the factors defining the terms for participation among a selection of highly experienced individuals with various ethnic and religious minority backgrounds, this article sets out to examine the current conditions for participation in the Norway's mainstream public sphere.

Methods and data

The empirical starting point of the article is data generated from a large-scale research project called The Status of Freedom of Speech in Norway (Enjolras et al., 2014). One of the project's goals was to study the conditions of free speech in a multi-ethnic and multireligious perspective. The project included a comprehensive survey among a representative sample of the Norwegian population as well as a subsample of immigrants and children of immigrants from Asia, Africa and Eastern Europe, asking detailed questions about the access to and the experiences with the public sphere in Norway.

The survey results, published elsewhere (Midtbøen and Steen-Johnsen, 2014, 2016), showed that immigrants and their children participate on par with the majority population, but also – somewhat surprisingly – that there are no significant differences in the share reporting experiences of negative or harassing comments. ² The survey demonstrated, however, that there are systematic differences between groups in terms of the nature and consequences of these comments. While individuals with majority background mainly are experiencing negative comments related to the content of what they write and their political standpoint, immigrants and their children report far more often that negative comments are related to religion, ethnicity, national origin or skin colour, and, when having such experiences, that they are far more likely to be hesitant towards public participation in the future.

In addition to the survey, in-depth interviews with 17 individuals of ethnic or religious minority background who are active participants in public debates were conducted. The in-depth interviews, which are analysed in this article, provide an opportunity to study the experiences of elite individuals with minority background who have succeeded in gaining a prominent position in the mainstream public sphere. I define them as ‘elite individuals’ because they are far from representative for ‘immigrants’, ‘ethnic minorities’ or any religious group in Norway, but rather occupy powerful positions in the public sphere by virtue of their visibility, resources and influence (see Khan, 2012: for a discussion of elite defintions).

The informants were selected on the basis of their active participation in public debates in Norway over the past years, often involving controversial questions related to immigration and integration. They had backgrounds from Bosnia, Hungary, India, Iraq, Iran, Norway, Pakistan, Somalia and Sri Lanka, and include individuals who self-identify as Muslims, Hindus, Sikhs, Jews and Christians, as well as secular and atheist. The recruitment of informants with a variety of ethnic and religious backgrounds reflects a deliberate choice not to implicitly engage in ‘methodological Islamism’ (Brubaker, 2013: 13), which simply conflates minorities with Muslims. Further, in an effort to disentangle ethnicity and religion from other aspects of social identity, the sample includes both males and females (7 women and 10 men), about half of the informants are themselves immigrants and half of them are born in Norway, and they represent different political and ideological positions; illustrated by their different takes on controversial issues related to, for example, gender equality and conservative religious practices.

Some of the informants are active in politics or organisational life, while others are independent writers and public intellectuals. As all informants are prominent actors in the Norwegian public sphere, the standard anonymization offered to informants in qualitative studies is somewhat more difficult to achieve in this study. I have still chosen to not use their full names, but as their positions in controversial issues often will be well known their identities may be revealed for readers familiar with the Norwegian context. This possibility, however, is acknowledged by all informants and my use of the interview material is made in agreement with all of them.

The interviews lasted between one and two and a half hours, were recorded on tape and transcribed in full-length. The conversations were semi-structured using an interview guide. In all interviews, different aspects of the informants' experiences of public participation were highlighted, including the terms for participation, the concrete experiences of each informant and what consequences the sum of conditions and experiences have on any future public engagement. Analysing the interview material, I draw on theories of ethnic boundaries to explore how the informants manoeuvre strategically in the public sphere to submit to, challenge or overcome existing ethnic boundaries. In the end of the article, I discuss the implications of these analyses for future altering of ethnic boundaries at a societal level.

Representation and self-censorship

As all of the informants in this study already posit positions in the mainstream public sphere in Norway, the question is not so much whether they have access, but whether they can participate as individuals or rather are recruited to represent specific ethnic or religious groups. The issue of representativeness is relevant for all the informants, but particularly strong among those who hold political positions; two young politicians with background from Sri Lanka and India, respectively. One of them says that the party she belongs to actively uses her and other minority politicians to mobilise ‘their’ ethnic communities: ‘When you are brown and in politics, then the party likes to give you tasks that make it so you appeal to immigrant women’, as she puts it. The other politician, born in Norway to Indian parents, explicitly addresses the challenges of representation when being asked what topics she engages in. Although she is generally interested in issues related to immigration and integration, she refrains from commenting or debating these issues in the public because she is afraid to be ‘lumped’ into a minority category and that the media will only be interested in her opinions as an ‘immigrant politician’, as she has observed has happened with other politicians:

I've certainly seen this happening to other politicians with minority backgrounds […] I don't doubt their qualifications, but when they get asked questions it always has something to do with immigration, you know, even though they might be focused on entirely different things… It was like this is in the beginning for me too, but I was very keen to say that I had no duties or anything that makes me able to comment on that issue. And then I've referred them to white people instead.

That Norwegian media is mostly interested in minorities engaging in specific topics – typically questions that pertain to religion, integration or immigration – is also mentioned by other informants in this study. One of them, a young Muslim woman who came to Norway before adolescence, finds, for example, that it is far easier to get published if one confirms majority stereotypes and adopts the role of an ‘internal critic’ of one's own community:

There's supposed to be a bit of a wow factor, you know. For the media, if you are a young woman and want to talk about the patriarchal society in your country of origin, you know – then it's like go ahead! […] I feel that if you are a minority who just confirms what the majority wanted you to say, but cannot state themselves, then it's very easy. […] In a way, you are only interesting as long as you are a critic of your own community. A puppet, a minority who says what fits. If you have independent opinions then it's sort of like that, no, then there isn't the same amount of space, you know.

Second, media's wish for minorities to speak about minority-related issues is but one factor determining what topics the informants in this study choose to engage in. Another powerful factor is the alleged attitudes towards controversial issues in the minority communities themselves. Illustrative of this point is the reflections made by two of the informants when they are asked if there are topics they strategically avoid. The first informant, a young lawyer with Muslim background who is a well-known critic of anti-immigration sentiments in the majority population, answers the following to this question:

Yes, I have, among other things, not written anything about homosexuality. […] I don't know if it's a conscious choice, but it might be really controversial that I, as a Muslim, would comment on it, that perhaps I avoid it. That I might receive more reactions that I can handle, especially from my own people, you know, from the Somali group.

Another way in which alleged attitudes in minority communities may influence on what topics the informants in this study choose to engage is gathered from an interview with a male, highly educated informant with Pakistani descent who also serve as a regular columnist in a national newspaper. He says that he generally refrains from addressing questions about religion because he fears severe reactions from the Muslim community in Norway about the fact that he is no longer a believer:

I try to avoid writing about religion. […] I was of course born in a Muslim family and raised as a Muslim, but somewhere in my life I discovered that there was no God… so I feel like it has been difficult to participate in that debate. I'm afraid to express an opinion, actually. […] And that's because I would be perceived as an apostate by a lot of people, and some believe that that should be punished by death and that sort of thing… […] It restrains my freedom of speech, actually. That's how I feel.

The significance of position

The variety of sources for self-censorship mentioned above seem to have parallels to the variety of sources that the informants in this study perceive to threaten their right to speak freely – if any at all. As Bangstad (2015: 225) points out, there is a tendency in the hate speech literature to assume that harassment or threats are directed primarily at minorities and that the senders primarily are individuals or groups of majority origin. However, it is important both to be aware that individuals of majority background may be targets of harassment (e.g. Hagen, 2015) and that individuals of ethnic and religious minority background not necessarily have negative experiences.

That being said, several of the informants in this study do report serious attacks on their freedom of speech. Two of the female informants report having experienced concrete threats and been subjected to physical violence on the street. Many have experienced receiving unpleasant and derogatory comments on the phone, or through text messages or social media, and several of them have reported the incident to the police to make visible the fact that harassment, threats and even hate crime do take place. Still, the picture is not one-dimensional. Some informants have not experienced anything negative when engaging in public debates and are surprised about the focus on the harsh debate climate in Norway, which they themselves have not seen anything of. One of the informants, a journalist with South Asian background, directly states that he believes the focus on this phenomenon is exaggerated:

I've received messages stating that I should just go back to where I came from. Unpleasant messages, that sort of thing. But I've never experienced it as… it is uncomfortable, but it is a random person who reacts to it, so what? I have never received a death threat. […] But some of the emails … some people could possibly be afraid of those kinds of emails. But I think this problem is given too much attention.

The analysis of the interviews conducted in this study point to some possible mediating factors determining who are more likely to be recipients of harassment and from whom threats are sent or verbal or physical assaults executed. First, it seems clear that the particular topics the informants engage in matter a great deal. Taking a strong stance on controversial issues like gender equality, immigration or the role of Islam in Norwegian society includes running the risk for negative response. Second, and relatedly, the specific position taken in debates over these issues seems to define who the informants believe threatens their freedom of speech. Among the informants in this study are Muslims who actively defend the right to wear religious symbols and are generally preoccupied with traditional religious practices being accepted, while others are among the most eloquent critics of conservative Islam. One of the informants from the former group, a well-known critic of Norwegian integration debates, started her career as a columnist in a national newspaper by addressing how debates over female genital mutilation are inflected by Islamophobic tendencies. She states that this topic resulted in a range of negative reactions:

These reactions came from forces in society that are hostile towards Islam. They wrote about me and criticised me for not seeing what is best for the child. I criticised how the debate had developed, that the crude generalizations created the impression that all Somalis in Norway circumcise their children, which is simply not true.

Linking the experiences of the different informants with their stance on controversial issues suggests that the positions they inhabit in the public debate are crucial: The ‘internal critics’ report striking inclusion in the Norwegian majority society, but they often face harassment from Islamist groups, and even opposition from individual researchers and politicians on the left. Those who engage in questions of discriminatory and racist attitudes in majority society report on the contrary of support from the minority communities, but have often received threats from Islamophobic or extreme right groups. A third group have intermediate positions in public debates, or engage in less controversial issues, and these often have not experienced very negative situations at all. This variation highlights the need for careful examination of the consequences of public engagement and points to the danger of jumping too quickly to conclusions regarding how ethnicity and religion shape the experiences of individuals.

Ethnic boundary-crossing and processes of change

Pointing to the different sources of self-censorship and the significance of position in controversial debates demonstrates the importance of employing a de-ethnicized analytical lens in studies of how minority individuals fare in the public sphere. However, most of the examples used so far still underscore the fact that ethnic and religious background matter – in the sense that the informants strategically manage their background one way or the other. But are there also signs of ethnic boundary-crossing, in the sense that some individuals have been able to transcend the existing ethnic boundary separating ‘insiders’ from ‘outsiders’ to the extent that they can participate in the public sphere as individuals regardless of their background, fully accepted as members of majority society?

Answering this question makes relevant the important distinction between the first and second generation, and as such the importance of time and birth place, when discussing the attachment to a given country and sense of entitlement to particular positions (Erdal and Ezzati, 2015). Several of the informants in this study are themselves immigrants who have served as prominent participants in public debates for decades. These informants share a conception of the Norwegian public sphere as an arena of bright ethnic boundaries, in which minorities cannot access, where the media are uninterested in their perspectives, and where the general knowledge of issues pertaining to ethnic diversity are scarce. The younger informants, on the other hand, tend to have other experiences, both in terms of access and their ability to challenge or even cross ethnic boundaries. Several of these young informants report being contacted directly by journalists and editors and encouraged to write feature articles, become regular columnists or participate in debates. For many of them, this was the beginning of a public involvement. One of these individuals, a women of Pakistani descent born in Norway, says she and several of the other, active children of immigrants have been ‘made’ as public figures by editors who have been very conscious in their efforts to recruit new voices:

A generational shift has taken place. […] Previously we always had to be on the defensive. The agenda of the new generation is that we want to set the premises ourselves, you know? There are completely different needs too, because you feel that you are Norwegian. So that is a very clear development.

There are a series of similar statements from the younger informants, which also are indicative of ethnic boundary-crossing. One example is a highly educated man of Pakistani origin born and raised in Norway, who is an active writer in national media. He has experienced a change over time, from representing what he calls ‘the Norwegian-Pakistani’ or ‘Norwegian-Muslim’ community in Norway, to gradually adopting a more individual position. When asked if he feel that he represents himself or a broader community, he answers:

That has been a bit in transition. Previously, what I wrote had to do with minorities, but this is something I've deliberately chosen to move away from. I don't want to view myself as a minority. Why should I be a minority? I was born and raised in Norway and see myself as Norwegian and that I belong to the majority. And then people tend to say to me, oh, you're so clever, you must be a role model for the minority. Why can't I be a model for the majority?

Based on the interviews in this study, it seems clear that ethnic or religious boundary-crossing at the individual level is both possible and occurring. Single individuals are able to carve out a space for participation based on individual merits and preferences rather than being conditioned on engagement in minority issues or sanctioned on the basis of ethnic or religious background. The question is whether this development is evident of broader, societal processes of what Alba (2005) calls boundary-blurring, indicating that the ethnic boundaries between ‘insiders’ and ‘outsiders’ become ambiguous and less clear-cut.

Of course, the interview material is too small to make any firm conclusions about this issue, but there are indications pointing in this direction. Particularly the informants who are born and raised in Norway experience extensive access to the public sphere. Many have few or no negative experiences with such participation, but on the contrary feel that they have access to positions on equal footing as their majority peers. This is in line with the striking social mobility characterizing the second generation in Norway, who despite of their parents' often disadvantaged positions in Norwegian society outperform the majority in terms of enrolment in higher education and have taken giant steps in closing the employment gap vis-à-vis the majority, even among women (Egge-Hoveid and Sandnes, 2015; Hermansen, 2013; Nadim, 2014). This is important to get across in a study of ethnic and religious minorities' experience of participating in the Norwegian public sphere, because it underscores that the boundaries between groups are open for negotiation and changes over time and across generations.

Still, although all of the informants in this study have access to the public sphere and many of them have been able to gain a position as an individual choosing what topics to engage in, ‘having a voice and being provided access to the public sphere is not synonymous with being heard nor being heard in the precise way one would like to be heard’, as Bangstad (2015: 23) puts it. Several informants are expected to represent specific ethnic or religious groups, they may find that editors and journalists want them to solely write about minority-related issues, and they may experience negative comments, harassment and threats that are clearly related to their minority background. These are examples of bright ethnic or religious boundaries experienced even by children of immigrants born and bred in Norway, indicating that these are processes of change and not change fulfilled.

Conclusion

Drawing on theories of ethnic boundary-making (Alba, 2005; Barth, 1969; Wimmer, 2013), this article has showed that studies of free speech in multi-ethnic societies must keep an eye on asymmetrical power relations that shape the opportunities of ethnic and religious minorities', without losing awareness of important processes of change that are currently taking place. Recent survey findings demonstrate that immigrants and their children, compared to native Norwegians, often receive comments related to their gender, ethnicity, skin colour and religious background when participating in public debates, and that they are much more likely to refrain from future public engagement if they have had negative experiences (Midtbøen and Steen-Johnsen, 2014, 2016). This suggests that harassment targeted at core identity features is a reality in the Norwegian context. However, when analysing the experiences of elite minority individuals who have succeeded in gaining prominent positions in the Norwegian mainstream public sphere, a striking variety of experiences and strategies are revealed.

First of all, there are different sources leading individuals to engage in self-censorship. Avoiding specific topics can be due to considerations about the danger of being ‘locked’ in a minority category, to avoid confirming stereotypes of Muslims or immigrants in the majority population, or by fear of the reactions in religious communities if taking liberal or secular positions in the public sphere. This variation demonstrates how ethnic and religious boundaries can be drawn both externally and internally – as responses to racism and discrimination in mainstream society or as fear of being ostracised by ‘one's own’ community.

Second, not everyone experience negative comments or harassment when engaging in public debates. Some do, for sure, and these are typically positioned either as ‘internal critics’ of particular religious communities or as critics of racism and Islamophobia in majority society, resulting in negative reactions from extreme groups on either ‘side’ of the debate on immigration and Islam in Norway. However, others have intermediate positions or engage in less controversial issues and these individuals report few, if any, negative experiences. Rather than simply assuming that ethnic or religious background in itself lead to specific reactions, these different experiences suggest that one has to take into account the positions held in the public sphere, which seem to determine whether single individuals become recipients of threats and harassing comments and from whom such comments are sent.

Third, important processes of change seem to be occurring at the present time. The younger informants tend to have more positive experiences of participation, both in terms of access and their ability to challenge ethnic categories and even cross ethnic boundaries. This is consistent with the development in Norwegian public sphere over the past 5 to 10 years, where a significant number of new voices have emerged and many of them claim that there is room for them to express themselves about controversial issues at the core of debates about immigration and religion, but also about topics not related to their ethnic or religious backgrounds at all. These accounts suggest that boundary-crossing at the individual level is possible and currently taking place, which in sum suggests an expansion of the overall space for freedom of speech in Norway, which individualises the debate and makes crude stereotypes about ‘the immigrant’ or ‘the Muslim’ more difficult to maintain.

The important underlying question rising from these findings is whether they are testament of broader processes of boundary-blurring at the macro level or simply exceptional experiences of elite individuals. Suffice it to say, this remains to be seen. On the one hand, there is no doubt that ethnic and religious boundaries still influence participation in the public sphere and many of the informants' accounts of threats and harassment are forceful reminders of the vulnerability that lies in inhabiting a minority position, and that the external categorisation of one's identity is still of major importance in defining ethnic boundaries. The many positive changing trends identified in this article must therefore not cast in doubt the need for continued attention toward the power relations in society that lead to restricting minorities' freedom of speech and belonging to Norway.

On the other hand, I warn against the tendency in the scholarly literature on these matters, which tends to over-emphasise the significance of ethnicity or religion in shaping individuals' access to and experiences with the public sphere. Research in this field should obviously be aware of the challenges faced by ethnic and religious minorities, but paying attention to the variety of experiences, the strategic ways in which ethnic boundaries are challenged and overcome, and the processes of change that are currently taking place are of no less importance. Young adults of immigrant background who are today occupying influential positions in the media, in politics and in civil society serve as powerful representatives of and role models for a new generation currently coming of age. The existence of this diversity of voices is blurring the clear-cut ethnic boundaries, which previously have dominated the Norwegian public sphere. They may contribute to changing our conceptions of what constitutes the Norwegian mainstream society, gradually altering the ethnic boundaries that too often is viewed as stable and fixed in contemporary research.