Navigating whiteness from the margins: Finnish, Somali, and Arabic speakers ’ experiences of racialization, (in)visibility, and (im)mobility in Gothenburg, Sweden

: This paper examines the relationship between language, (in)visibility, and (im)mobility in racialized spaces, focusing on Finnish, Somali, and Arabic speakers in Sweden. Using a theoretical framework based on hegemonic whiteness and intersectionality, the study explores how multilingual practices and sub-jectivities intersect with race, religion, gender, and class to shape social visibility and mobility. The research draws on linguistic ethnographic data, including interviews, linguistic landscape documentation, and an analysis of the media discourse. The study ﬁ nds that while Finnish speakers have become invisible due to assimilation policies, Somali and Arabic speakers are hypervisible in Swedish public spaces and discourse, although Arabic speakers are sometimes, and in relation to other migrants, nearing Swedish whiteness. However, all three languages and their speakers are constrained by a white normativity that reproduces inequality. The paper challenges simplistic notions of mobility/immobility and visibility/invisibility in the context of a changing racial order in Sweden, where whiteness serves as a binary sorting mechanism that perpetuates inequality. Overall, this research sheds light on the complex entanglement of language, visibility, and mobility in white


Introduction
During the last half-century, Sweden has often been hailed as a migration-friendly welfare state (see e.g.Jensen et al. 2017;Koopmans and Statham 1999).Moreover, Sweden has been viewed as an example of "exceptionalism" vis-à-vis the rest of the world, i.e., as an ideal society without the problems ensuing from colonialism, racism, and xenophobia.However, as Loftsdóttir and Jensen (2012) have argued, "the interwoven racial, gendered and nationalistic ideologies associated with the colonial project form part of contemporary Nordic identities".Moreover, the accounts of migrants who came to Gothenburg, where this study is conducted, in the 1970-2000s reveal that it has become tougher to feel at home in Swedish society (Hanoush et al. 2023).While, in the last decades of the 20 th century, migrants felt that Swedish individuals and institutions tried to understand their situation and were willing to offer some help, over the last few years overt expressions of racism and everyday discrimination have effectively undermined the idea of exceptionalism.The current centre-right-wing government coalition explicitly argues that integration policies should be "demand-driven" on the one hand, and "emphasize the individual's own responsibility," on the other (Regeringen 2023).This means that migrants are to a large extent held responsible for their own shortcomings, and they are subjected to increasing demands to adapt to "Swedish" ways of being, speaking, and behaving.But what counts as "Swedish"?According to Hübinette and Lundström (2014: 246), "the central core and the master signifier of Swedishness" is constituted by "whiteness".The dominant understanding is that "a Swede is a white person, and a non-white person is therefore not, and cannot fully become a Swede" (ibid.).
In this paper, we draw upon Hübinette andLundström's (2014, 2020) account of hegemonic whiteness in Sweden, i.e. that white perspectives and privileges bind white people together regardless of e.g.social class and political affiliation (Hübinette and Lundström 2020), as well as Ahmed's (2007) phenomenology of whiteness, to explore how racialized speakers of Finnish, Somali, and Arabic navigate and negotiate whiteness in Sweden.We examine how the speakers' multilingual practices and subjectivities are related to social (in)visibility and (im)mobility, as well as to the intersections of race, religion, gender, and class.For this purpose, we align with Bourdieu's early argument that in linguistic exchange, "what speaks is not the utterance, the language, but the whole social person " (1977: 653).Moreover, we take a diachronic and spatial perspective, analysing how these experiences of (in)visibility and (im)mobility are related to changes in social and physical space: the national public sphere (media representations and political regimes), on the one hand, and the local public space (private and public signage), on the other.Finally, in order to contribute to the understanding of the interrelationship between race and language across time and space (see Flores and Rosa 2017), we compare the experiences of three groups: speakers of Finnish, Somali, and Arabic.We focus on these three groups since the speakers of these languages have all migrated to Sweden, but they have done it during different periods of hegemonic whiteness (Hübinette andLundström 2014, 2020; see more in next section).As a result, these three groups clearly illustrate how the boundaries of whiteness are negotiated, produced, and reproduced through time and space.Put bluntly, while the Finnish speakers in Sweden have become 'white', over time, and the Somali speakers experience constant racialization, the Arabic speakers aspire to whiteness, but do not succeed in such an attempt.Partly for these reasons, the three groups also present different patterns of social visibility, which we examine in this article.The comparative spatial-diachronic approach therefore allows us to explore the ideological development of hegemonic whiteness and the ways in which differently minorized and racialized speakers navigate white "Swedish" space.Importantly, Arabic speakers and Somali speakers are not mutually exclusive categories, with many Somali speakers claiming knowledge of Arabic while Arabic speakers often take a distanced stance towards Somalis based on past colonial relations.This makes their comparison especially relevant for disentangling the complex interrelationships of language and race.However, for the purposes of this paper, 'Arabic speakers' means people with a heritage in a region where Arabic is the main language.The intersectional gaze that we deploy sees 'Arabic speakers', and all other speaker categories, as social categorisations that in social interaction can sometimes be seen as referring to language narrowly and sometimes more broadly to class and race.
Against this backdrop, the paper asks the following research questions: What is the degree of (in)visibility of Arabic, Finnish, and Somali and their speakers in Swedish mainstream media and public signage in Gothenburg?How do Finnish, Somali, and Arabic speakers in Gothenburg navigate and negotiate whiteness?In what ways does it affect their language use, visibility, and mobility?What are the differences and similarities between the three speaker-groups' experiences and how can the differences be explained?To contextualize these questions, we first describe the ideological changes in the Swedish migration and integration regimes.Next, we present our conceptual framework, centred around the notions of whiteness, visibility, mobility, and the margins.The fourth section describes our data and methods.The analysis is presented in two sections, the first focusing on signage in public space and media representations of Finnish, Somali, and Arabic, and the second on the speakers' everyday experiences of navigating white space.
2 Background: three ideological frameworks for Swedish migration policy The territory currently known as Sweden has always been characterized by ethnic and linguistic diversity.Due to the presence of Indigenous minoritiesthe Sámiand various migrant groups who have been moving to Sweden since the 16 th century, Sweden has a long history of linguistic diversity.It lies beyond the scope of this article to give a detailed overview of the history of migration to Sweden (see Boguslaw 2012).
Here, we focus on the ideological frameworks that have characterized Swedish migration policy since World War II.Based on an overview of policy documents and political debates about Sweden since the 1950s, it is possible to single out three ideological phases.
The first phasebetween 1945 and the end of the 1960scovers a period characterized by "labour immigration" (Östberg and Andersson 2013).The Swedish post-war economy flourished and was in need of migrants to handle the fast growth (Hinnfors et al. 2012).In these decades, most migrants came to Sweden from Finland, Greece, Yugoslavia, Italy and Hungary.During this period, migrants who settled in Sweden were expected to abandon their cultural traditions and embrace Swedish "norms".Discursively, such ideology was manifested in the word "assimilation".
Since the beginning of the 1970s, however, labour immigration began to decline due to the decreasing economy.At the same time, Sweden received refugees fleeing their countries because of wars or for political motivations (Abiri 2000;Schuster 2000).Most migrants in the 1970s and 1980s came from Chile, Iran, Lebanon, Poland, and Turkey, and since the 1990s, from the former Yugoslavia, Afghanistan, Iraq, Somalia, and Syria.
In 1968, the Swedish government clearly stated in a directive that "migrants shall have the opportunity to have the same standard of living as the host population" (cited in Prop 1968: 142).Such a programmatic statement was based on the realization on the part of Swedish politicians that assimilation had not been an effective equalizer of differences.As a result, the government appointed a parliamentary investigationthe so-called Immigrant Investigationwith the aim "to chart migrants' and minorities' situation and propose societal measures with a view to facilitating their opportunities to 'adapt' to life in Sweden" (cited in Prop 1968: 142).Based on the committee's report, the parliament decided on the principles that should guide Swedish immigrant policy: equality, freedom of choice, and partnership.This meant (i) an attempt to achieve equality of opportunity for both immigrants and Swedes; (ii) freedom of choice for immigrants to decide to what degree they wanted to preserve their cultural/linguistic traditions; and (iii) collaboration between immigrants and Swedes.Assimilation was replaced with integration as a two-way process involving both migrants and Swedes.A new keyword entered political discourse: multiculturalism, later replaced with diversity (see Invandrarpolitiska kommittén 1996: 55).
This ideological and discursive regime has been shifting dramatically since 2010, the year that marked the entry into parliament of Sverigedemokraterna ('the Swedish Democrats'), a nationalist and right-wing populist party with Nazi roots and a clear anti-immigration agenda.Until 2019, the mainstream political parties distanced themselves from the Swedish Democrats and their negative view on immigration.Since then, their rhetoric on the issue has become increasingly similar (Krzyżanowski 2018;Milani et al. 2021).
While the expression "Swedish values" was nearly unheard of within the context of the explicitly multicultural politics of the 1980s, it has since become a buzzword employed by almost all parties across the political spectrum (Bauer et al. 2024).The period from the 2010s to the ongoing 2020s has been characterized by an ideological and discursive tension between the relic of multiculturalism and diversity and the resurgence of arguments in support of assimilation, that is, for the acquisition of a set of Swedish norms and values (whatever these may be).
The three ideological regimes described above coincide partly with what Hübinette andLundström (2014, 2020) have described as the three phases of hegemonic whiteness: "white purity" , "white solidarity" (1968-2001), and "white melancholy" (2001 onwards).By describing Sweden's contemporary history in the light of race and whiteness, Hübinette and Lundström capture the following trajectory: Sweden went from being a leader in racial biology research with explicit racial thinking at the beginning of the 20 th century to being one of the most generous immigration countries in the Western world.The phase of "white solidarity" is also a period characterized by a "colour blind" anti-racism at the official level.However, this phase ended abruptly at the beginning of the 21 st century.One aspect of Swedish "colour blindness" highlighted by Hübinette and Lundström (2014) is that the word race has been abolished from Swedish legislation, today often replaced by the terminology of ethnicity and culture.This prevents racial discrimination from being prosecuted and makes racist crimes more invisible.Furthermore, "colour blindness" prevents an understanding of the race-obsessed Swedish history and constrains the public debate on today's racialization processes.In their analysis, Hübinette and Lundström show how race intersects with categories such as class and gender.Departing from the assumption that Swedishness is primarily defined by whiteness, they show how it operates as "a structuring principle [that] reproduces itself through different means over these apparently contrasting periods" (Hübinette and Lundström 2014: 425).
Today, almost 30 % of the Swedish population is constituted by Swedish citizens who have immigrated from other countries or are growing up with parents with a non-Swedish background (SCB 2022).This means that the whiteness norm seriously challenges national integration and social cohesion.
Regarding the three languages in focus in our study, Finnish has been spoken in Sweden for centuries.Finland was under Swedish rule from the late 13 th century until 1809, and people moved frequently between Finland and Sweden.After the Second World War, a large-scale migration began, becoming particularly extensive in the 1960s-70s, in response to demands in Swedish industry.People from poorer areas of Finland moved to wealthier parts of Sweden, where industrial labour was needed.This inequality has long characterized the power relation between the two languages and their speakers.The relationship is usually described in terms of Swedish supremacy and Finnish subordination (Snellmann 2006).Today, there are about 30 000 Finnish speakers in Gothenburg.
In 2015, Finnish was replaced by Arabic as the second most spoken language in Sweden (Parkvall 2019).However, Arabic-speaking people have been immigrating to Sweden since the 1980s, primarily from Iraq, Lebanon, and Syria.During the socalled European refugee crisis in 2015, the number of Arabic speakers in Sweden increased (Swedish Migration Agency 2023).In 2019, over 5 % of the Gothenburgers were born in Arabic-dominated countries (Göteborgsbladet 2022).
In the late 1980s, an increasing number of Somali-speaking people migrated to Sweden, and in recent years , 7 % of the people who sought asylum in Sweden came from Somalia (Swedish Migration Agency 2023).Today, about 15-20,000 Somali speakers live in Gothenburg (Göteborgsbladet 2022).
3 Looking at other languages from a position of whiteness: data, method, ethical considerations In this paper, we are addressing social visibility (Anthias 2008;Yekani and Nowicka 2022: 21) as the ways of seeing Finnish, Somali, and Arabic speakers that circulate in Swedish society.We assume that the social visibility of these speakers and their associated languages depends on a complex web of historical, geopolitical, economic, racial, and ideological processes.It further affects the mobility of these languages and speakers, both physically through the city and symbolically through society, i.e., along scales of socio-economic class.To demonstrate this complex interplay, the analysis covers three sources of social visibility: visual materializations of language in public space; mentions and representations in the media; and perceptions by the speakers and others (who see them, react to, and comment on their visual appearance and linguistic performance) elicited through interviews.
The data was collected within a larger multi-sited and mobile linguistic ethnography taking place in four Gothenburg neighbourhoods over the course of 2017-2022.1A photo documentation of public and private signage in four Gothenburg neighbourhoods served as a starting point for the discussion of the visibility of languages.All signs and texts on all establishments in the central streets of each neighbourhood were photographed by two of the authors between May and September 2019, summing up 1885 photos, and then classified according to which languages were used (for a detailed analysis, see Rosendahl et al. 2023).The data for the analysis of how the media frames Finnish, Somali, and Arabic consists of articles published between 01/01/1980 and 12/31/2022.It was compiled through Retriever, a digital archive of Swedish newspapers and periodicals.The corpus was built using the search terms språk* AND Göteborg [language* AND Gothenburg].In total, the corpus comprises 63 083 articles (58 506 367 words).In the analyses, we used the freeware corpus toolbox #Lancsbox (Brezina et al. 2021) to study the frequency of collocates, that is, words that frequently co-occur when Finnish, Somali, and Arabic are mentioned in the Swedish press.
Moreover, we interviewed around 25 speakers of each of Finnish, Somali, and Arabic during the period 2017-2022.The interviewees were mainly selected according to the snowball principle.This means that the sample is not statistically representative, and the data presents the voices and perceptions of a limited number of individuals.The voices are sometimes contradictory and, as in all interview studies, the interviewees have their own motives and agendas.A small majority of the interviewees are women; the ages of the interviewees vary between 20 and 80.
The interviews were semi-structured, i.e., instead of starting from a fixed set of questions, the starting point was the visibility and status of each language in Gothenburg, with the focus then moving to how the interviewee experienced speaking Finnish, Somali, or Arabic in Gothenburg, and how it affected their patterns and possibilities of mobility (both physical and social).Inspired by the research field of oral history (Thor Tureby and Johansson 2020), the interviewees themselves had to highlight what they considered important to convey, and the interviews therefore varied to some extent. 2ll interviews were conducted without an interpreter, most in Swedish and a few in English.This affected the selection of participants and the nature of the conversations, as the interviewees generally were not speaking their first language.However, there are advantages to not conducting interpreter-mediated interviews.The participation of an interpreter conveys an additional dimension of interpretation and sometimes gives participants a sense of insecurity (Winlund 2021).Interviews can always be said to be unequal in different forms and degrees; here, it is particularly important to consider that, being white middle class, we as interviewers also represent the majority society and thus automatically become, in the interview situation, "the white listening subject" (Flores and Rosa 2017), which means that we (consciously or unconsciously) embody a subject position that reproduces white privilege and norms.
Overall, while the main focus of this article is on the interview data, we believe that it is impossible to fully understand what the participants in the study said without taking into account the broader language ideological environment of Gothenburg.This is something that can be revealed by mapping the presence/ absence of languages in both media and the built environment.

Whiteness and the politics of mobility and visibility
As Ahmed (2007) explains, racialization is a prerequisite for whiteness, i.e., the perception and categorization of people as "white" or "non-white", "coloured" or "black", is based on prior racialization.However, the Swedish public discourse has been dominated by what has been called "post-racism" (Goldberg 2015), or "the belief that racist actions and utteranceswhen they occur (which they do with alarming frequency)are individual and occasional rather than systemic and systematical" (Stroud 2017: 239).This belief leads to a structural invisibilization of racism, which also tends to be made imperceptible.In Swedish research and activism, this aspect is often referred to as the "color blindness" of hegemonic whiteness (Hübinette andLundström 2014, 2020).
The visibility of a person or an object depends on the gaze of the viewer.The gaze is a cultural and social construct emerging out of the viewer's engagement with the world, through which s/he has learned to see and interpret what s/he is seeing."Visibility is thus an outcome of a complex social process which locates people in particular relational positions" (Yekani and Nowicka 2022: 21).In Sweden, this resonates in the ways in which migrants are differentiated and separated from the (national) Swedish population or included in it.While the separation of migrants as a non-Swedish or not-fully-Swedish category (Hübinette and Lundström 2014) makes them extra visible, their inclusion tends to make them invisible.
At the same time, race is a social categorization strongly relying on essentialized ideas of visible aspects (Yekani and Nowicka 2022: 22).This implies that racialized bodies and voices that "stand out" from white spaces can become publicly accessible sites for discrimination, surveillance, and control (Ahmed 2007).In this sense, they become hypervisible in a Swedish context: marked and easily subjected to stigmatization and marginalization.Simultaneously, within the same white spaces, white subjects, bodies, and voices are rendered invisible in a positive and privileged way.Hence, white hegemony is a system of privilege and restriction, a binary sorting mechanism that includes and favours some while excluding and disfavouring others.
Critical mobility studies (Sheller 2011: 1) show how mobility is always linked to immobility and encompasses embodied practices, ideologies, and meanings.Building on Sheller, Ahmed (2007) writes about the "politics of mobility" that regulates who can move unhindered across the lines that divide different spaces, as well as which bodies can inhabit different spaces.A good and relevant example for our analysis is a person's name.According to Ahmed (2007: 132−133), "to inherit a Muslim name in the West, is to inherit the impossibility of a body that can 'trail behind', or even to inherit the impossibility of extending the body's reach".A similar pattern has Navigating whiteness from the margins been revealed in Sweden by Carlsson and Rooth (2006), who show how job applicants with Swedish-sounding names were called to an interview 50 % more often than applicants with Arabic-sounding names.In these examples, mobility is limited only by a foreign name; it constitutes a disturbance in the space-met, a potential threat.As will be illustrated in the analysis below, language and skin colour interact with names in this politics of mobility.
The contemporary working of this politics of mobility and its effects on racialized speakers relies, "even in Sweden" (Pred 2000), on colonial history.As Stroud (2007: 26) explains, "under colonialism, linguistic descriptions created 'sociolinguistic hierarchies within and between languages which were projected onto categories such as class, gender, ethnicity, and which legitimized the exploitation of certain groups in labour'".In the 19 th century and the first half of the 20 th century, Swedish researchers used scientific disciplines to describe Finnish speakers as a different race, i.e. as not white, and the results were used to legitimize colonialism in the east (Laskar 2017).Swedish Finns then continued to be racialized, along with new immigrant groups (Mulinari and Neergaard 2017: 91).However, as Mulinari and Neergaard (ibid.)point out, racialization is an ongoing process and several of the immigrant groups that first came to Sweden have since been partially "de-racialized as the 'migrant other' and re-racialized as White".Instead, new migrant groups fill the position of the racialized Other.Such reflections resonate with Wynter's (2003) theorization of the ways in which whiteness operates by positioning itself as "human" while devaluing the racialized Other as "non-human".
Stroud's quote above further shows the importance of taking an intersectional perspective, which takes into account the mutual constitution of several categorization processes in order to understand the linguistic hierarchy operating in a particular time and place.When we look at an individual language and its speakers, we are trying to describe the complexity within such a group, or what McCall (2005) calls "intracategorical complexity".In Nash's (2008) words, such an approach considers the dangers of categorization without rejecting the categories themselves.Following McCall (2005McCall ( : 1774)), intracategorical complexity takes marginalized intersectional identities as an analytical starting point "in order to reveal the complexity of lived experience within such groups".Our interview data gives ample proof of such complexity and shows how social categorizations and the systems that create the power orders give rise to often contradictory voices.
Finally, taking the margins as "a heuristic lens through which to interrogate the production of knowledge about particular socio-linguistic arrangements" (Milani 2014: 9), we see on the one hand how social mobility from the margins towards the centre always occurs at a cost, be it a language, an identity, or a name.Hegemonic whiteness constantly requires new social categories, groups, and languages to inhabit the margins, to embody and perform marginality.On the other hand, "what counts as 'marginal' […] is the object of continual negotiations" (ibid.):marginalized subjects and speakers are also making resistance and claiming space and recognition for their beliefs, bodies, and voices.
5 Assessing the visibility of Finnish, Somali, and Arabic via media representations and signage in public space Regarding the visibility of Finnish, Somali, and Arabic in the Swedish press, the corpus analysis shows that Finnish is comparatively often mentioned in newspapers and magazines, while Arabic and especially Somali are rarely written about.Not unexpectedly, Swedish is the language, nationality, and ethnicity term that is by far the most frequently mentioned in texts about Gothenburg.From 1980 to 2022, svensk* 'Swed*' occurs 144 603 times in the corpus, followed by engelsk* 'English*' (20 720), fransk* 'French*' (14 205), tysk* 'German*' (13 809), and dansk* 'Danish*' (10,598).Of the three languages in focus here, Finnish comes in fifth place (8 176), Arabic in twelfth place (4 493) and Somali in 19 th place (1 473).
The results in Table 1 show that the occurrence of all three language, nationality, and ethnicity terms was low until the beginning of the 1990s, mainly because the number of articles accessible at that time was negligible.As the following collocation analysis shows, the noticeable peaks for the three terms can often be explained by specific events that resulted in higher media coverage.
Finnish is often named in Swedish newspapers in enumerations with other official national minority languages (e.g., Sami, Meänkieli) and other Nordic languages (e.g., Danish, Norwegian).However, Finnish is rarely related to other migrant languages, such as Arabic.In this way, media representations contribute to making Finnish 'float' between minority and migrant status.Other frequent collocates are words that associate Finnish with cultural activities and expressions (e.g., literature, radio) and the Swedish language law.
Most of the peaks in the media coverage are related to reporting on the development of the Swedish language law.The peak at the beginning of the 21 st century is primarily a result of articles that discuss the declaration of Finnish as a minority language by the Minority Language Committee of Sweden.In 2009, Finnish was classified by law as one of five national minority languages, which was the subject of numerous articles in 2009 and 2010.
The media's framing of Arabic and Somali differs from that of Finnish.Both are seldom mentioned with European languages but frequently in enumerations of other migrant languages (e.g., Kurdish, Persian).Other words that typically co-occur with arabisk* 'Arabic*' and somalisk* 'Somali*' connect them explicitly to migration (e.g., immigration, migrant, refugee).In addition, the press connects Arabic to religion, for instance, through the frequent use of Islam, Muslim, and Allah.Somali, on the contrary, is frequently linked to family contexts such as parenthood and upbringing (e.g., family, parents, children).Somali speakers are also often referred to as a collective.
Media texts mentioning Arabic were particularly frequent during 2015 and 2016, when many Syrians migrated to Sweden.The minor peaks in the reporting on Somali are mainly due to often cited comments by politicians, teachers, and others regarding the willingness of Somali-speaking persons to learn Swedish.
While Finnish is most often mentioned in press texts about Gothenburg (see Table 1), Arabic is most visible in Gothenburg's urban space.Arabic is represented in 5 % of the signs in our photo documentation, while Somali is represented in 4 % and Finnish in 1 % of the signs.As in the press, Swedish and English are also the most visible languages in urban space (see more in Rosendahl et al. 2023).
Arabic is concentrated in a few areas of Gothenburg, mainly outside the city centre.Arabic texts are found mostly on commercial signs belonging to smaller businesses and signs connected to religious activities.Previous studies have proven that Arabic also serves a status-enhancing function in multilingual areas in Gothenburg, even in businesses not run by Arabic speakers (see Aharon and Löfdahl 2022).Somali texts are also concentrated in a few areas and are mainly used on commercial signs.The visual presence of Finnish in Gothenburg is low, and signs using Finnish are, according to our photo-documented data, only found in the satellite district of Gårdsten.Finnish is almost exclusively used on the signs of cultural associations and on nameplates on residential buildings.
To sum up, the visibility of Finnish, Somali, and Arabic is relatively low compared to other languages.This applies to the press texts as well as to the photo-documented data.However, the two analyses show a relatedness between Arabic and Somali.Finnish is treated differently, both regarding media coverage and the active use of the language in public space.Arabic and Somali are somewhat more frequently used in public space, but compared to Finnish are less often mentioned in the press.Both Arabic and Somali are primarily framed in the press through the lens of migration.The languages and their speakers are portrayed as "new" to Sweden; they are linked to religion, and the speakers' willingness to learn Swedish becomes the focal point of media interest.Thus, the two languages and their speakers are made visible through their otherness.Finnish, in contrast, is hardly visible in public space and the press texts deal primarily with the legal status of the Finnish language in Sweden.
6 Finnish speakers' experiences 6.1 The colonial production and inheritance of shame Feelings of inferiority is a theme that recurs in the interviews with older Finnish speakers.In particular, they reflect on how the colonial past of Swedish-Finnish relations conditions the present and materializes in feelings of inferiority and shame.An elderly Swedish-Finnish woman says: "jag tror det finns i den finska identiteten, det finns ju någon form av mindervärdeskomplex rent historiskt gentemot Sverige" [I think it is in the Finnish identity, there is some kind of inferiority complex historically towards Sweden].Several interviewees report on such feelings being transmitted over generations, with parents educating their children into compliant and overachieving Selves, asking them to do more and be better than their Swedish schoolmates and colleagues.A Finnish and Arabic speaking Swedish female in her 40s says: Jag fick med mig hemifrån så här väldigt tidigt, tror jag, att: 'Du måste-' Alltså, hon var supersträng med mig i skolan liksom.<skrattar> Hon var aldrig nöjd.
[I got from home very early, I think, that: 'You have to-' I mean, she was super strict with me in school.<laughs> She was never satisfied.] Navigating whiteness from the margins By doing better and adapting to the white normativity, which in turn implies an invisibility, it is possible to renegotiate one's position in the margins and facilitate social mobility.
Yet, in the interviews, shame about people's Finnish origins is often expressed.In particular, second-generation Swedish Finns express shame about their parents, their language and professions and, as one interviewee says, "their imposed background".This shame has remained; older interviewees tell us how they were silenced on the bus in order not to stand out, and younger ones relate how their parents insisted on speaking Swedish, a language they had not really mastered, instead of the mother tongue at home.At the same time, the children were made hyper-visible through institutional policies like Finnish-speaking classes, which led to the segregation of the Finnish-speaking school children from their Swedish-speaking peers (see Ågren 2006).

Producing ambivalence by policing language
Since 2000 and the official recognition of Finnish as a Swedish national minority language in the so-called Language Law (SFS nr 2009: 600), Finnish is subject to explicit language policy, giving the speakers "the opportunity to learn, develop and use" their language, which thus implicitly enhances the speakers' social visibility (see Table 1) and facilitates their social mobility.Since 2022, Finnish as a national minority language has, through the establishment of language centres, been given additional opportunities for revitalization, making the group more visible, at least at the political and administrative level.Despite this, most of our interviewees describe a situation of intergenerational language loss and the Swedishization of their names.In other words, the language policy seems to have only had limited effects so far.The status of Finnish as a national minority language puts the Swedish-Finnish group in an ambivalent position between immigrantship and minority status, which was also seen in the analysis of the Swedish media above.While the minority status is based on a long historical presence in the country, with the Minority Act aiming to protect the minority from linguistic and cultural assimilation, immigranthood rather requires adaptation and thus an expectation of invisibility.This makes the Swedish-Finnish group in Sweden special and ambivalent.Moreover, the fact that the group is given rights while at the same time the Finnish language is socially devalued, contributes to double invisibility, i.e. the inherent oppression is made invisible.
We argue that this ambivalence can partly be understood through an intersectional gaze in which hegemonic whiteness intersects closely with social class.Whiteness is unstable; it is a process that is constantly negotiated and renegotiated in relation to the surrounding society, and through such floating whiteness, the position in the margins (Milani 2014) becomes constantly negotiable.In the past Finns were not considered to be White, but to belong to another "race".This, combined with the fact that the young migrants came from poor parts of Finland (Ågren 2006;Snellman 2006), has probably contributed to the low status of the Swedish-Finnish minority in Gothenburg, and to the perpetuation of a classifying and racializing gaze directed by the majority towards the Finnish language and its speakers.

Reorganizing the local language hierarchy
Gradually, new groups have moved to Sweden and been sorted into the local racial hierarchy.As an elderly Finnish-speaking woman said: "now there are new scapegoats", i.e., new residents who are less white than the Finnish speakers and are positioned on a lower stage in the racial hierarchy.Hegemonic whiteness and inequality thus contribute to the production of competition and stratification between different migrant groups.Reflecting on the expectations directed towards migrants from the majority society, a middle-aged Arabic and Finnish-speaking female describes "a political need for distinguishing and separating" Us from Them, where in contemporary Sweden "Them stands for Somalis and Arabs", rather than Finns.A woman who migrated from Finland to Sweden in the 1960s believes that "newly arrived immigrant children" are not ashamed of their language but speak it unhindered in a way that the Finnish children have not done.Inequality is thus manifested in that higher demands for adaptation are placed on less privileged groups, which is an expression of racialization.The woman's description testifies both to envy towards what she calls "newly arrived immigrant children" and irritation that they do not easily fall into the local language hierarchy.

Becoming white and invisible: the price of leaving the margins
The majority's expectation of adaptation and inclusion into hegemonic whiteness can also be seen in other interviews.The interviewees talk about how they abandoned their language and their Finnish-sounding names to be part of Swedish whiteness, thus illustrating how groups that were not previously considered white can now be included in the Swedish whiteness.However, the interviews bear witness to the fact that leaving one's position on the margins is not uncomplicated.As illustrated by the middle-aged man in the quote below, the social invisibility that Navigating whiteness from the margins comes with becoming Swedish and white is not entirely positive.In a way, he thinks that being Finnish speaking in Sweden could have been easier if they had had additional differentiating physical and cultural traits: Jag tror att problemet med oss är att vi är så pass försvenskade, så att det enda som egentligen … liksom stort skiljer oss från majoritetsbefolkningen är ju att vi har det här andra språket.Så hade vi, hade vi haft ett mer särskiljande kulturellt uttryck i något yttre så där kanske, så hade det kanske varit lättare då.
[I think that the problem with us is that we are so heavily Swedishized that the only thing that really … sort of distinguishes us from the majority population is that we have this other language.So if we, if we had a more distinctive cultural expression in something external like that, it might have been easier then.] However, the socio-economic perspective is central to understanding the social (im)mobility of the Finnish group in Sweden and it is also linked to visibility.Ågren (2006: 193) describes how identification with the working class and the suburbs was perceived as more important for the social visibility of the Swedish Finns than the language itself.This illustrates how whiteness in Sweden is constructed through class-based traits, and sometimes, as in the case with the Finnish speakers, even more so than through the language.Many of Ågren's interviewees describe feelings of shame regarding their parents' class affiliation and immobility, while the image of the Finns as hard workers was cemented early on in the Swedish media.Today, the Finnish speakers have made a collective upward class movement, but as illustrated by the man below, this has had negative consequences for their linguistic practices: As Ahmed (2007: 150) has put it, "Whiteness could be described as an ongoing and unfinished history, which orientates bodies in specific directions, affecting how they 'take up' space."Today, the Swedish-Finnish group has undergone upward social mobility and left the areas where they previously lived.As noted by the young man in the quote above, the journey from the margins into the centre has a price: leaving the language and the naming traditions behind to be included in the Swedish whiteness.As we will see in the analysis of Somali and Arabic speakers below, social class always matters in racialization and whitening processes, but it plays out differently under different migration and integration regimes and is mediated and negotiated in different ways in public discourse for different language groups.

Racializing space and mobility
Almost all the Somali-speaking interviewees live in north-eastern Gothenburg.Here, people get by well in Somali.Several of the Somali speakers interviewed believe that language affects their mobility, where they choose to live, where they feel at home and safe.But, as one woman stresses, it is not really the Somali language that poses problems.Above all, it is the Somali body that is limiting.The interviews show how people are made aware daily that the colour of their skin stands out from the Swedish norm.The racialized body and its language can only renegotiate its position on the margins and move towards the centre with difficulty.
According to Ahmed (2007), public spaces are shaped by the bodies that reside in them, and the spaces are defined and colored by these bodies.In white spaces, white bodies "are made invisible" at the same time as non-white bodies "become hypervisible when they do not pass, which means they 'stand out' and 'stand apart'" (Ahmed 2007: 159).One Somali-speaking woman describes the feelings of unease that arise when she enters into white spaces in Gothenburg: Jag är aldrig mig själv, det är den här (den somaliska kvinnan) folk ser när man kommer in i ett rum.När jag kommer till Angeredpuh, vad skönt.
[I am never myself, it is this (the Somali woman) people see when you enter a room.When I come to Angered [a migrant-dense Gothenburg neighbourhood]phew, that's nice.]While she is racialized by the white spaces in the city centre, the feelings of relief and comfort that she links to the neighbourhood of Angered contribute to the hegemonic reification of this space as racialized Other (cf.Järlehed 2022).

Gendered embodiment of "all the bad"
Although racialized space in Sweden constrains the mobility of all non-white people, our interviews show the existence of a gender hierarchy.As one Somali-speaking woman says: "A black Muslim woman stands at the bottom of society."This hierarchy is also present in the quote of this middle-aged Somali-speaking man: "somaliska tjejer har sjal, och hijab, och liknande … massa kläder som kan orsaka att de är mer synliga än alla andra människor."[Somali girls wear headscarves, and hijabs, and similar … lots of clothes that can cause them to be more visible than other people.]The veiled black woman and her language becomes hyper-visible in the white spaces.One woman says that in her physical appearance she represents everything "non-Swedish".In response to the question of what non-Swedish means, she replies: "it's all that's terrible, all the bad".Her statement illustrates how Swedishness as a discursive construction is not only highly valued but also closely linked to whiteness.Other interviewees joke about skin whitening and surgery to become Swedish, to leave their position in the margin.Such imagined remaking of the self is no less radical than the abandonment of language and name changes reported by the Finnish speakers above: they are all prices one has to pay for leaving the margins.
However, the picture is not unambiguous.While the interviewees describe the stigma of being a black woman, it is emphasized by some that young women, in particular, find it easier to integrate into Swedish society; the feelings of alienation and combativeness seem to go hand in hand: Det går bra för de somaliska tjejerna.Det går inte lika bra, tyvärr, för våra pojkar, men för tjejerna går det bra.[…] Som somalisk kvinna så känner jag, jag har inga begränsningar.Det spelar ingen roll om jag är beslöjad eller inte.Jag kan göra vad jag vill.
[The Somali girls are doing well.It's not as good, unfortunately, for our boys, but the girls are doing well.[…] As a Somali woman, I feel I have no limitations.It doesn't matter if I am veiled or not.I can do whatever I want] 7.3 Racializing the migrant voice and internalizing racialization Milani (2007Milani ( , 2008) ) has shown how the Swedish language plays a central role in constructing "Swedishness".In our media data, the learning of the Swedish language is consistently framed as crucial for becoming 'integrated' as a migrant in Sweden.As can be seen in the following quote from a letter to the editor in Gothenburg's newspaper, the responsibility for learning Swedish is typically placed at the individual level: Att försöka tillgodogöra sig det nya landets språk är en förutsättning för vidare utbildning och arbete.Detta är helt upp till den enskilda individen.Att anklaga den Svenska arbetsmarknaden och flyktingpolitiken för att vara rasistiska är att frånsvära sig allt personligt ansvar -språk först, utbildning och arbete sedan!(Göteborgs-Posten, 2 May 1997) [To attempt to acquire the language of the new country is a prerequisite for further education and employment.This is entirely up to the individual.Accusing the Swedish job market and refugee policy of being racist is to absolve oneself of all personal responsibility -language first, education and employment later!]None of the Somali interviewees feel that the majority society appreciates their language.A young woman describes her feelings when she, as a child, arrived in Sweden and encountered the expectations of the dominant society: "Du måste lära dig svenska för att humaniseras" [You have to learn Swedish to be humanized].Her feeling illustrates how the contemporary language ideology operates within a postcolonial context that has sorted people into human (white) and not fully human (racialized) (Flores 2021).
[Uh, and I mean, Somali language is a little bit higher level of the voice on the voice … when speaking is a little bit high.It sounds like it's a little bit loud.That can it … also cause, because we are people who speak a little high … loud.] In this quote, visibility is connected to perceived audibility, i.e., how people in different contexts perceive how others talk affects the visibility of the latter and how racialized bodies are positioned as linguistically deficient.This racialization of migrant voices is often internalized, and sometimes reproduced in the migrants' gaze upon their "own" group: Eh, det kanske skiljer sig lite mer kulturellt sett.Jag menar, när jag flyttade till Göteborg så tyckte jag själv att de somalierna som bor här i Göteborg var mer högljudda.
[Well, maybe it differs a bit more culturally.I mean, when I moved to Gothenburg, I myself thought that the Somalis living here in Gothenburg were more vocal.]This middle-aged Somali-speaking man spent his first years in Sweden in a small village surrounded by Swedish speakers and adapted his speech volume to them.When he later moved to Gothenburg and started socializing with other Somali speakers, he found them to be "loud".

Speaking back to media
In the analysis of the media data above, we saw how the ethnic marker somalisk* 'Somali*' was generally connected to what in Sweden is perceived as a low-status language and to politically contested semantic fields such as migration.Somali speakers seem to refer to and relate to the often negative media image of them to a Navigating whiteness from the margins greater extent than the other groups: "du vet, det här med media.Vi är väldigt påverkade av media tyvärr."[you know, the media.We are very influenced by the media, unfortunately] (Somali-speaking mother).
Previous studies of Swedish media representations of migration show that migrants often are depicted negatively as welfare recipients (Krzyżanowski 2018;Wojahn 2023).This is particularly true for the Somali group.When reporting on Somali groups in Sweden, the Swedish press extensively focuses on topics related to unemployment.Based on our analysis of the media data, this is prominent during the period spanning from the 1990s to the 2010s.In many articles, the Somali community is highlighted as the migrant group with the highest unemployment rate: "only a few percent of Somalis in, for example, Gothenburg have work" (Göteborgs-Posten, 25 April 1997); "the Somali group's unemployment is the highest, over 90 percent, of all refugee groups" (Göteborgs-Posten, 2 May 1997).
[A while ago, the media wrote a lot about Somalis being dependent on social benefits and sitting at home, but today many of them work in the service sector; driving public transport, working in healthcare or the cleaning industry.I don't know anyone here in Gothenburg who is unemployed, who doesn't do anything at home or who is on social benefits, compared to how it was before.]As exemplified by these quotes, Somali speakers often have negative experiences of the media, which they find gives an exaggerated negative picture of their group.One woman further highlights the problem that her children internalize the dominant racializing discourse, resulting in complex feelings and behaviors of shame and denial about their Somali language and heritage.
Despite, or in response to, this stigmatizing media portrayal, several Somalispeaking interviewees present the intention of "flipping the script": that is, "in place of failure, one emphasizes pedagogies of hope, possibility and success" (Ibrahim 2011: 631), as in this quote from a young Somali-speaking youth worker: Yes, det finns de … de största … rapparna nu i Sverige tror jag, … åtta eller sju av de tio som är störst i Sverige nu, så är det somalier.… deras album heter något somaliska namn eller liknande, eller med somaliska slangord som … som får i gång de här ungdomarna.
[Yes, there are the … the biggest … rappers now in Sweden I think, … uh, eight or seven of the ten biggest in Sweden now, they are Somalis.… their album has a Somali name or something like that, or with Somali slang words like … that gets these young people going.]Stories of success like the one performed by the Somali rappers can thus serve as a strategy to renegotiate the position on the margins.

Reactive racialization
Despite the negative attitude from the majority towards the Somali language, Somali children participate to a large extent in mother tongue classes (Palm et al. 2019).In our interviews with Somali speakers, we see how their feelings of abandonment and alienation lead them to cling to and emphasise 'their' Somali identity traits, thus breaking off even more from Swedishness and whiteness.This reactive racialization thus strengthens and reproduces whiteness as a binary sorting mechanism: you either fit in, or you don't.At the same time, language becomes something to hold on to when the racialized body is denied access to the white spaces, as this Somali-speaking woman states: Så här är det, ju mer utanför samhället man känner sig, desto mer måste man hålla sitt språk och modersmål och allt det där levande.Ju mer integrerad man är och ju mer accepterad man känner sig, desto mindre viktigt blir det.
[Here's the thing, the more outside of society you feel, the more you have to keep your language and mother tongue and all that alive.The more integrated you are and the more accepted you feel, the less important it becomes.]This contrasts with how the Finnish-speaking male above reflected on the negative aspects of becoming white and Swedish (see 6.4).The contrast shows how racial orders are historically and geopolitically situated and how minorized speakers in differently racialized bodies meet different affordances in different spaces.A good example is presented by one of the Somali interviewees, a Somali-speaking woman, who reflects on the different affordances regarding integration for Somali immigrants in Sweden and in English-speaking countries: alltså USA, England, Australien … Där är det inte många som kan, eh … somaliska.Utan det är väldigt mycket engelska … eh.Där blir … Men de tänker jag också, de är ju mer integrerade, mer accepterade.Där är den amerikanska identiteten alltså jämbördig med … många känner sig som afro-eller somaliska amerikaner.
[so USA, England, Australia … There are not many who know, uh … Somali.Instead, there is a lot of English … eh.There … But I also think they are more integrated, more accepted.So there the American identity is equal to … many feel like Afro-or Somali Americans.]Admittedly, the U.S., the U.K. and Australia have had high levels of immigration due to their imperialist pasts, but the interviewee's words should also be seen against the backdrop of the Swedish colour blindness and the idea of Swedish exceptionalism that invisibilizes racialization.In addition, these countries are English-speaking and the Somalis who have moved there have tended to be English-speaking, which is largely a matter of social class.Additionally, they have larger black populations, which makes Somalis less visually salient in public spaces.
The interviews with Finnish and Somali speakers showed how they are subjected to and experience processes of marginalization and stigmatization.The interviewees described constraints on their physical and social mobility and their effects on how self-identification is expressed, as well as on attitudes to the languages and naming traditions: while some (the second-generation immigrants) of the Finnish-speaking Gothenburgers have decided to leave their language and Finnish-sounding names behind, the majority of Somali speakers tend to hold on to their language and naming traditions because without those attributes they feel like "no one".
In the interviews with Arabic speakers, we observe a different attitude and note how they are often "gazing back at the eyes of power and declaring themselves 'subjects'" (Ibrahim 2011: 619), which could indicate that the Arabic-speaking population in Gothenburg is less affected by stigmatization and marginalization than speakers of Finnish and Somali.Few of them express experiences of any mental, social, or spatial barrier to speaking Arabic.Some Arabic speakers think that Arabic will develop in Sweden and that it will even affect Swedish.The following quote is from a teacher of Arabic as mother tongue: Det arabiska tar mycket plats nu.Arabiska är väldigt stort-det andra största i Sverige.Många vill lära sig arabiska […] Det finns överallt; affärer, apoteket, biblioteket, vårdcentralen.
[Arabic takes up a lot of space now.Arabic is very bigthe second biggest in Sweden.Many people want to learn Arabic […] It's everywhere; shops, the pharmacy, the library, the health centre.] In the interviews, we talked about the status of Arabic in Gothenburg, and several of the interviewees pointed out that a large number of speakers by default brings high status.They also stress that Arabic, as a global language spoken in many countries, tends to receive high status even in Sweden.How the number of speakers affects attitudes to language (and its speakers) is reflected in the public discourse; especially in the case of Arabic, these numbers are seen as a threat by the majority but lead to a higher status of the language among minorities (see Järlehed 2022;Löfdahl 2022).
An Arabic-speaking woman who founded both a mosque and a school with an Arabic profile says: "Arabic is more exciting than for instance Somali and Kurdish, and it depends on the religion".She refers to Arabic as the language of Islam and says that many people from other minorities want to learn Arabic.She further states that many children study Arabic at mother tongue lessons at school, and several children with Swedish backgrounds also study Arabic at school.She calls Arabic a "strong language" as it is spoken in so many countries, adding: "You are proud to speak Arabic!" One might think that this woman's acts of "flipping the script" (Ibrahim 2011: 631) give an overly positive image of what it is like to speak Arabic in Sweden, but she is not alone in this.Other interviewees give a similar picture.It is interesting to consider why the interviewees (unlike speakers of other minority languages) emphasize the positive aspects of speaking Arabic in Sweden.In the example of the woman who ran a school and a mosque, the explanation is probably that Arabic is part of the business she runs.For her, Arabic has an economic value in Swedish society.The value, however, is not only economic.The way she and other interviewees speak about their mother tongue is part of their self-positioning, where Arabic is used as a resource to renegotiate racialization and the position in the margins, particularly in contrast with other immigrant languages.The interviews show that Arabic speakers (sometimes) refuse to accept the local hierarchy.Instead, the visibility and audibility of Arabic becomes a resource of physical and social mobility.

Mediatized entanglement of language and religion
In contrast to Finnish and Somali, the media analysis depicted Arabic as a language with close ties to religion.In the Swedish press, words like Muslim, Islam, and Koran are frequently mentioned in relation to Arabic (see Section 5).Similar associations are also reflected in almost all interviews.
"If you are Somali, Kurd, or Turk, or anything, you are a Muslim, then you have to read and listen to the Koran.And the Koran is in Arabic."This quote by an Arabicspeaking teacher shows that Arabic has high status among other minorities in Sweden, especially among the Somali speakers.Many of the Somali-speaking interviewees are keen to emphasize their knowledge of Arabic and to use it as symbolic/cultural capital for positioning themselves as good Muslims (cf.Bourdieu 1977).A picture of mutual influence and exchange between Arabic and Islam emerges in the interviews, where Arabic strengthens Islam through high status among Muslims.Islam, in turn, helps to spread Arabic among Muslims and to maintain its high status among them.
In the interviews, language, religion, and culture are intimately entangled and are perceived as impossible to separate.A young Arab man puts it like this in an interview conducted in English: "…we love to learn our children our language, because this is related not only to the language, it's also, if I can say that -I'm not a religion people, but it's related to our religion, it's related to our culture.So, Arabic it's more than language." The link between religion and language also appears in an interview with a young Syrian woman.She distances herself from Arabic and refers to the fact that she is a Christian, not a Muslim.She says that Swedes often think she comes from southern Europe instead of Syria.She assumes it is due to her physical appearance ("brown hair, fair skin") and also because she is unveiled.Through her statement, she positions herself as "white" and distances herself from the racialized Other.According to this reasoning, the veil becomes a marker of Islam, Muslimness, and also Arabic.For Arabic, linguistic vitality (at least partly) depends on religion, which has a unifying function for many language groups.This distinguishes Arabic from most other immigrant minority languages in Sweden.

Adapting to the white listening subject by watching the tongue
Similarly to the Finnish and Somali speakers, Arabic speakers describe their language as being perceived as louder than the surrounding majority language: Ja, alltså, alltså, jag är säker att vissa kopplar det till terrorism eller liknande.Eh, när folk pratar arabiska högt, så vissa bli oroade av det liksom.Eh, … jag, jag förstår det.(young man from Syria) [Yes, well, I'm sure that some people link it to terrorism or the like.Uh, when people speak Arabic out loud, some people get worried about it as well.Uh, … I, I understand that.] As illustrated by this quote, the interviewee experience that some people link Arabic to terrorism, which is part of a wider, global discourse connecting Islam and terrorism (see e.g.Baker et al. 2013).According to Ahmed (2007: 163), being seen as Muslim in the West often "translates into 'could be Terrorist'".For the Arabic speakers, this social disposition has implications for their mobility, and several describe how they adapt to what Flores and Rosa (2017) have termed the "white listening subject", i.e., an imagined listener or listening instance, an "ideological position and mode of perception that shapes our racialized society" (Flores and Rosa 2017: 177).The white listening subject consistently mishears and disqualifies the speech of minoritized bodies as not appropriate and insufficient, thus prompting them to "'watch their tongues' […] through constant self-corrections" (Bourdieu 1977: 656).By describing how he adapts to the white listening subject by shifting to Swedish, a young Arabic-speaking interviewee further illustrates how he internalizes racialization: Finally, linguistic exchanges like this one show, with Bourdieu (1977: 656), how, "by 'watching their tongues', the dominated groups recognize, if not the supervision of the dominant (though they "watch themselves" most closely in their presence), then at least the legitimacy of the dominant language" in practice.

Postcolonial conditioning of inter-racial stratification and domination
While some interviewees express an unambiguous picture of Arabic as a global language with a strong position in Sweden, other voices describe how racialization is experienced, often based on physical appearance.A young woman who was born in Palestine and moved to Sweden when she was two years old says: Du kan vara svensk på riktigt och ändå inte ses riktigt som svensk om du inte har de här utseendemässiga dragen som folk i allmänhet ser som svenska.Det behöver inte bara vara svenskar, det kan vara folk som nyligen kommit hit som inte ser dig som svensk.Alltså, det är inte bara majoritetssamhället.Det är ju också minoriteter som inte ser mig som svensk.
[You can be genuinely Swedish and still not be seen as Swedish if you don't have these visual characteristics that people generally see as Swedish.It doesn't just have to be Swedes, it can be people who have recently arrived here who don't see you as Swedish.So, it's not just the majority society.There are also minorities who don't see me as Swedish.] This quote illustrates how constant racialization occurs even among minorities, reflecting the global racial hierarchy.In the quote below, an Arabic-speaking man born in Tunisia states that the Somali-speaking group is poorly integrated and does not bother to learn Swedish: De [invandrarna] kan aldrig utveckla sig med svenska samhället.Aldrig.De vet inte vad svenskarna är och svenskarna vet inte hur de är, för han behöver inte ens prata svenska.Ja, vi har ju folk här som bor i 30 år och han kan inte gö-sätta en mening ens.Och han säger ju rakt ut: "Varför ska jag lära mig svenska?", "För du bor i Sverige.","Nej", säger han: "Jag umgås, äter och pratar och tittar på somalisk tv och allting.Jag behöver inte det där".
[ The quote resonates with a general negative representation of the Somali-speaking group in Swedish media discourse.This opinion demonstrates how racialization and whiteness as a sorting mechanism exist not only in the majority-minority relationship, but also between minority groups.Here we need to recapitulate how the relationship between Arabic-speaking countries and Somalia has been shaped by colonialism (Löfdahl and Wenner 2018), and how such relations, just like the one between Finland and Sweden, are reproduced in exile.As Tudor (2017: 16) states, racism always exists and operates "in relation to postcolonial social conditions which are historically and geopolitically contextualised".The Arabic-speaking group seeks to move closer to whiteness by positioning another group as less integrated (i.e. less 'white') in the struggle to leave its place on the margins.At the same time, Stoler's (1989) examination of whiteness in relation to Dutch colonialism, shows that colonial societies are not only constructed through race but also through class.This means that race and class mutually constitute each other in determining which groups can be included in whiteness or not.It may even be said that they stand in a dialectical relationship for migrant minorities: you cannot become "white" from a working-class position, and you cannot leave the working class unless you become "white".In the case of the data under investigation here, Arabic speakers and Somali speakers are in an asymmetrical relationship to one another.Race is not the only pivot for such inequality, but levels of education and income also play a key role in dispensing who can approximate whiteness and who cannot.The Finnish group that came to Sweden as labour migrants came from poorer parts of Finland.They were often uneducated (Snellman 2006).Some of the Somalis who came to Sweden have been (or have been considered to be) poorly educated.The Arabic-speaking group has, at least in recent years, been considered more highly educated, although the level of education within the group varies considerably (SCB 2019).
9 Key observations and differences Our examination lays the ground for arguing that, while assimilatory practices have made Finnish invisible, Arabic and Somali have rather been made hypervisible in public spaces and discourse.These differences notwithstanding, the status of these three languages and the social mobility of their speakers have been constrained by the prevailing whiteness of Swedish politics and media.In all three cases, speakers testify to their migrant languages being seen as "too loud", describing how they accommodate to this regimenting reaction from the white environment in different ways and degrees, thus illustrating how racialization often gets internalized and contributes to the production of a governmentality (see Del Percio 2016) that upholds hegemonic whiteness.The racialization of the three groups has to some extent taken place during different periods and in different political spaces.During the period when the majority of Finnish speakers moved to Gothenburg, assimilation and adaptation were the expected ways of dealing with minority status.Although the group has gained recognition in recent years through language policy regulations, the contemporary political discourse has (once again) become dominated by calls for increased adaptation.The interviewees express a feeling of invisibility and subordination to the Swedish majority population.The Swedish-Finnish woman who, in the quote above, reacted to the loud voices of the newly arrived migrant children, expresses both envy and annoyanceenvy that other groups will not be silenced as she herself has been, annoyance at another minority group that does not understand that they "should" fit into the local hierarchy.However, due to their "floating whiteness" and strong adaptation, the group can today be included in the Swedish whiteness.
In relation to whiteness in Sweden, the Swedish-Somali group stands out as hyper-visible.Although a large part of this group came to Sweden during a period when Swedish migration policies were characterized by notions of "multiculturalism" and "diversity", the interviews give a different picture and interviewees describe an increasingly tough social climate with demands for adaptation.However, the racialized body combined with Muslim-coded attributes makes it impossible for the Somali group to "fit in", thus underlining how intersections of race and religion operate in regimenting mobility.Moreover, the Somali speakers' place in the local language hierarchy depends on how they, as racialized persons, are seen in a Swedish context, but also how they as a group are viewed by the former colonial power, i.e., Arabic speakers.As pointed out by Ahmed (2007), the racialized body always breaks away in the white spaces and is therefore not considered as "Swedish".
In the interviews, resistance and resignation are expressed: "We will never be Swedish no matter how hard we try".
In the Arabic interviews, we do not encounter the same strategies of adaptation.Instead, they express their pride in Arabic as a global language and as the language of Islam.Although Arabic is physically invisible in large parts of the city, it has high social visibility and status in certain areas and domains.Only a few interviewees believe Arabic will be weakened in Sweden.Rather, some of them express that the language will strengthen its position.Although some interviewees describe experiences of racialization, most of the interviewees do not.Instead, they describe other minorities (especially Somalis) as the non-integrated Other, thus indicating that the margin is inhabited by other, less white bodies and bringing their own group closer to the white Swedishness.
We argue that whiteness is central to understanding the living conditions and life chances of the three groups in Sweden.However, there is also a class dimension here.On one hand, social mobility is conditioned and restricted by race in several ways.While the Finnish speakers have made a collective upward class movement that facilitated their whitening, the Somali speakers' mobility is constantly blocked through racialization.On the other hand, class is conditioning the whitening process with working class belonging being seen as a hindrance to whiteness.This could be seen in the Arabic speaking interviewees that contrasted their self-declared high education level with the assumedly low education level of Somali speakers.
In sum, through adaptation and invisibility, the Finnish group has been able to move to affluent neighbourhoods and acquire professions that were previously unavailable to them, but at the cost of a lost language.The Somali group, on the other hand, is stopped by their racialized bodies when they attempt physical and social mobility.Although Arabic is mainly linked to certain neighbourhoods, the Arabic interviewees do not express any limitations in terms of physical mobility.Instead, stories of how to get by in Arabic in large parts of the city predominate.At the same time, the use of Arabic provokes aggression and Arabic is often referred to as a "threat", both by individuals and in the public discourse, thus indicating social mobility constraints.

Concluding remarks
Overall, this paper has aimed to complicate facile binaries of mobility/immobility and visibility/invisibility in relation to a changing racial order in Sweden, where whiteness works as a binary sorting mechanism continuously including new members at the expense of racialized others, and thus reproducing inequality.Whiteness is a relative construct, and who is included in the Swedish whiteness varies over time and is essentially a result of both local and global racial orders.Furthermore, the study helps to highlight the fact that visibility and invisibility are not a binary but a continuum, operating at different levels; as shown by the Finnish speakers, visibility at the societal level can lead to invisibility at the individual level.
Finally, this paper shows that how the languages studied survive in Sweden is not only a matter of Swedish integration policy, but that it is equally determined by the global racial hierarchy, with whiteness and, to some extent, class being key components.The colonial history that underpins the global racialization order affects both self-perception and the perception of others, leading to the constant negotiation and renegotiation of positions, which reproduces hierarchical structures and affects the individual's view of their own language.Visibility and audibility are intertwined and, as pointed out by Bourdieu (1977), language itself is only a minor part of what we perceive as a person's speech.Furthermore, the Swedish "colour blindness" makes the oppression of racialized groups invisible, resulting in increased inequality.
Det är lite högt man pratar när man pratar på arabiska[…]när man pratar på arabiska, då känns det att folk får en specifik bild av dig och då blir det liksom svårare att komma in med den personen, liksom.Och det var därför jag också väljer att prata svenskan i stället för arabiskan ibland.[It'sa bit loud when you speak in Arabic […] When you speak in Arabic, then it feels like people get a specific picture of you and then it kind of gets harder to get in (at the pub) with that person, as well.And that's why I also choose to speak Swedish instead of Arabic sometimes.] They [the immigrants] can never develop with Swedish society.Never.They don't know what the Swedes are, and the Swedes don't know what they are like, because he doesn't even have to speak Swedish.Yes, we have people living here for 30 years and he can't even make a sentence.And he says straight out: "Why should I learn Swedish?", "Because you live in Sweden.","No", he says: "I socialize, eat and talk and watch Somali TV and everything.I don't need that [Swedish]".]