Racial profiling in the racial welfare state: Examining the order of policing in the Nordic region

This article builds on two interview studies on racial profiling conducted in Finland and Sweden. It examines policing practices in order to elaborate on the understanding of what we define as the ‘racial welfare state’. The analysis draws attention to the ways that bordering practices reproduce racial orders, within and beyond the nation-state. The embeddedness of the Nordic region in the western sphere, with its colonial legacies, is highlighted through the empirical material that focuses on the consequences of internal and external migration controls, as well as more general police stop-and-search practices. The study underlines the need to investigate racial profiling as a practice that enforces an imagined community based not on whiteness in general, but on Nordic whiteness in particular as the norm against which the bodies of ‘others’ are measured.


Introduction
The first wave of research on racial profiling was concentrated on encounters with law enforcement while driving. The developing research field has since examined racial profiling from a range of perspectives, such as the War on Drugs, the War on Terror, border and migration controls, as well as policies that target stigmatized urban areas and their populations (for recent contributions, see Gaston, 2019;Parmar, 2020;Welch, 2019). These studies have convincingly shown that racial bias regulates police work in different national and regional contexts.
In this study, we fill a gap by examining the Nordic welfare state context, and ask: what social order is reproduced through practices of racial profiling? Despite the acknowledged need for context-specific studies, racial profiling has scarcely been documented, or analytically explored, by Nordic researchers. This is a shortcoming because, first, the prevalent notions of Nordic exceptionalism give the impression that the region is characterized by less repressive policies; and second, policing in the countries of the region has specific characteristics that deserve detailed attention (Barker, 2018).
The overall aim of the article is to untangle the power relations of which racial profiling is an expression through an understanding of what we call the 'racial welfare state'. We develop the idea of the 'racial state' (Goldberg, 2002) to fit into and take into account the particular traits of the Nordic racial formation (Omi and Winant, 2014). Our point of departure is that race has been a core organizing principle throughout history, pivotal in shaping practices of inclusion and exclusion, even in countries that have perceived themselves as outsiders to colonialism (Jensen and Loftsdóttir, 2012;Keskinen et al., 2009). More specifically, we emphasize the entanglement of race in the idea and institutions of the nation-state, embedded in distinct colonial relations and histories (Balibar, 1991;Komlosy et al., 2016).
The concept of the racial welfare state also refers to the intertwinement of racialized social control and ordering that operates through policing and welfare state institutions. We build on scholarship that argues for the key role of the police in enforcing the current social order (Fassin, 2013), but seek to highlight their interconnectedness to welfare institutions and ideologies (Høigård, 2011), as well as the effects of the neoliberalization trends in recent decades (Amar, 2010). In the effort to highlight the transnational elements of the racial welfare state, we also draw on recent theorizing on the criminalization of migration (Bosworth et al., 2018) and underline the role of the 'crimmigration control system' (Bowling and Westenra, 2018).
Empirically, this analysis expands on our previous, more descriptive oriented research, which is based on interviews with the police and people with experiences of racial profiling in Sweden and Finland. We have shown that to be singled out from the crowd by the police-especially for visible minorities such as Roma and youth racialized as nonwhite-is a common, everyday experience (Keskinen et al., 2018;Schclarek Mulinari, 2019a). In this study, we advance a more theoretical discussion, arguing that to emanate from material generated in two different countries enables an increased possibility of generalization of the interpretations.
The article is organized as follows: first, we address the role of racial profiling in the Nordic context. Second, we elaborate a theoretical approach to the racial welfare state and the role of the police in it. Third, we introduce the empirical material and concretize our analytical perspective. Fourth, we examine the ways in which the social order is enforced through racial profiling, and its connection to central elements of the Nordic welfare state. Fifth, the concluding discussion spotlights the new knowledge our analysis produces.
Before moving on, a clarification is in order on the core concept of the study. We understand 'racial profiling' as an umbrella term that refers to selective and racializing police practices that (re)produce societal hierarchies by targeting specific groups based on ideas of race, ethnicity, religion or national identity (for a similar understanding, see Glover, 2009).

Racial profiling in the Nordic context
Within the mainstream of Nordic criminology, the relationship between policing, race and racism has only superficially been discussed. The trust in the Nordic welfare state model has been strong, generating extensive discussions of the exceptionality of the penal order in the region (Pratt, 2008). Tellingly for the prevalence of the understanding of the model's unique character is that similar debates have preoccupied other disciplines: early feminist researchers, for example, conceptualized the Nordic model as women-friendly, emphasizing the support given by the welfare state to women's employment and childcare services (Hernes, 1987).
It is only recently that studies focusing exclusively on racial profiling have been published (for exceptions, see Hydén and Lundberg, 2004;Pettersson, 2005). Scholars throughout the Nordic region have, however, repeatedly, and for some time, highlighted tensions between the police and ethnic minorities (Egharevba, 2011;Haller et al., 2019;Saarikkomäki, 2017;Sollund, 2006). These findings are at odds with prevailing discourses on multiculturalism, tolerance and equal rights within the police in the Nordic countries (Wieslander, 2014).
While these studies have been important in underlining the conflictual relationship between the police and groups racialized as non-white, they suffer from a theoretical shortcoming by conceptualizing racial profiling as a phenomenon that appeared in the aftermath of migration from non-European countries in the 1960s, and the multicultural society that this is seen to produce.
In contrast, we emphasize the continuum of state surveillance that has targeted the poor in general and ethnic minorities in particular, who have been subjected to forced removals, labour camps, repressive child protection practices, sterilization programmes and assimilatory policies. The Nordic states also actively developed scientific racism, for example through the State Institute for Racial Biology that operated in Sweden from 1922 until 1958. Moreover, migration controls have been a salient feature, with bans on, for example, Roma from entering the Nordic countries during extensive periods. The role of the police in such state actions has been to implement political decisions, but also to create statistics and register information about minorities (Ds, 2014;Ericsson, 2016;Isaksson, 2001;Pulma, 2006).
It can be stated that the Nordic societies, rather than being characterized by a moderate penal order, have been built on repressive and assimilative policies. This feature has made sociologist Vanessa Barker (2013) describe the Nordic welfare states as Janusfaced regimes: generous towards those defined as deserving and belonging, but interventionist towards those seen as deviant 'others'.
We maintain that key elements in this regard are notions of whiteness, which have been a cornerstone for the state-building projects in the region (Jensen and Loftsdóttir, 2012;Keskinen et al., 2009). Although some groups racialized as non-white have been included in the nation, they have not been granted a place among those with self-evident belonging. Significantly, the groups racialized as non-white have changed over time: until the post-war era, when scientific racism lost momentum, Finns in Sweden were actually regarded as a race of lower standing, excluded from Nordic whiteness. Also, as labour migrants entering Sweden in large numbers in the 1960s and 1970s, the Finns were subjected to criminalizing discourses and repression (Laskar, 2017;Lundström and Teitelbaum, 2017).
In this regard, it is important to account for the restructuring of the Nordic model during recent decades. The erosion of the public sector and the rapid growth of social inequalities in the wake of neoliberal policies have profoundly affected racial relations (Neergaard et al., 2017). Also important to account for are the new migration regimes in the Nordic countries: the efforts to reduce 'levels' of asylum seekers have been followed by a move towards welfare chauvinism in the sense that state provision is increasingly based on ethno-nationalist and often racializing criteria (Keskinen, 2016). It can be concluded that, within the sphere of law and order, authoritarian and nationalist forms of policing have been accentuated throughout the region (Barker, 2018).

The racial welfare state and the police
To investigate the Nordic countries as racial welfare state formations means that we emphasize the continuum of white hegemony and mistrust towards ethnic minorities and groups defined as 'others' (Keskinen et al., 2009;Mulinari and Neergaard, 2017). As in other parts of Europe, policies that criminalize the mobility of the racialized poor have been a marked feature of the Nordic model (Aas, 2011;De los Reyes, 2006). We argue that race has been an essential element in the imagining of community (Anderson, 2016(Anderson, [1983), fundamental in producing the Nordic region as part of the white, and economically prosperous, West. Rather than being outsiders to the colonial project, the region has been deeply engaged in the European civilizing mission (Hansen and Jonsson, 2017;Jensen and Loftsdóttir, 2012). More specifically, critical race scholar David Theo Goldberg's (2002: 104) understanding of the racial state inspires us: States are racial more deeply because of the structural position they occupy in producing and reproducing, constituting and effecting racially shaped spaces and places, groups and events, life worlds and possibilities, accesses and restrictions, inclusions and exclusions, conceptions and modes of representation.
Accordingly, we understand that the institutions of the modern nation-states are racial, but also racializing; that is, they (re)produce race and racism (Omi and Winant, 2014). States do not, however, all have the same power, neither do the inhabitants have the same rights; instead, these are related to their position in the geographically and hierarchically arranged world economy (Balibar, 1991;Komlosy et al., 2016). The production of the European subject as essentially white is a result of historical processes that build on notions of the inferiority of the 'other' and the superiority of the West, in varying and flexible ways, as well as persisting global inequalities of power and wealth (Gutiérrez Rodríguez, 2018;Hall, 1992).
Based on this theoretical framework, we address racism within law enforcement not as an anomaly, but as part of the interconnection of local and global power relations (Owusu-Bempah, 2017;Parmar, 2020) that structure the relation between race, policing and space (Dikeç, 2007;Fassin, 2013;Rios, 2011). We argue that criminalization is an embedded aspect of the social order upheld through racializing policing practices (Armenta, 2017). These enforce boundaries of belonging (Bosworth et al., 2018;Yuval-Davis et al., 2018). In particular we focus on the crimmigration control system that binds together local and national policing to European and global systems of surveillance, punishment and exclusion (Aas, 2011;Bowling and Westenra, 2018). From our perspective, the War on Drugs, the War on Terror, border and migration controls, as well as policies that target the population in stigmatized residential areas, are parts of a complex pattern of racial missions of policing (Amar, 2010;Golash-Boza, 2016).
A central feature in the current era is neoliberalism, which has reconfigured racial power relations and, by extension, the struggle about how humanity should be ordered. Scholars have argued that, rather than focusing on the well-being of citizens, state authorities nowadays are preoccupied with the production of (in)security from racially constructed and perceived threats (Fekete, 2009;Wacquant, 2009). This has paved the way for the racial state to develop into an increasingly invasive-and in parts privatized-repressive apparatus invested in the criminalization of migration in particular, and groups racialized as non-white in general (Goldberg, 2009;Kalra et al., 2013).

Material and method
The studies that this article builds on have used individual and group interview data as their main material. 1 In the Swedish study (Schclarek Mulinari, 2019a), eight individual interviews and five focus group interviews, involving 28 participants, were organized to discuss personal experiences of racial profiling. The participants identified as Muslims, Afro-Swedes, Roma and/or young people who had grown up in territorially stigmatized urban areas in larger Swedish cities. Moreover, seven representatives of the police were interviewed. The Finnish study (Keskinen et al., 2018), on the other hand, includes a larger span of interview material: 26 police officers were interviewed about their work practices, and 145 people were interviewed individually or in groups about their experiences of racial profiling. Even though these informants were racialized as non-white by Finnish society, they identified in various ways regarding ethnicity and nationality, including Finnish Roma and Roma from Central and Eastern Europe as well as those having transnational roots in countries in the Middle East, Latin America, Asia and Africa. A specificity of the Finnish study is that it addressed experiences of racial profiling not only by the police, but also by private policing by security guards.
Despite the difference in size and to some extent composition, the interview methodology of the two studies largely corresponds: both studies included informants with experiences of racial profiling varying in regards to age, gender and background. The studies also incorporated interviews with police officers who were almost exclusively whitemajority citizens. We contend that the two materials complement each other, enabling the identification of central characteristics of racial profiling in the Nordic context. Significant in this regard is that, from a global perspective, the five neighbouring countries that constitute the region can be viewed as very similar. From a regionally embedded perspective, however, it could be argued that Finland and Sweden differ largely, for example regarding demography. Partly because Finland has historically had a more restrictive migration policy, a greater number of people racialized as non-white today live in Sweden. Rather than focusing on the distinctive qualities, the objective of this article is to produce a more general understanding and, thus, we focus on common features.
As for how the analysis of the empirical material has been undertaken, we are inspired by attempts to expand the criminological gaze by engaging with critical race methodology (Glover, 2019). This is an approach that underlines the experiences of those exposed to racism (Solórzano and Yosso, 2002). Within this methodology, thematic analysis is used to identify macro-forces in their intersection with everyday experiences (Lawless and Chen, 2019). Concretely, the coding procedure departed from a comparison of the two data sets focusing on the central and recurring themes related to experiences of racial profiling, such as how it was understood (how, where and when did it take place?) and analytically explained (why did it happen?). In a second stage, the intersections between crime/criminalization and inclusion/exclusion were explored based on our understanding of the police as enforcers of: (a) the social order that they are rooted in; (b) an imagined community that the racial state is built on; and (c) the bordering practices embedded in the power of the crimmigration control system.
The empirical material is presented following the idea of the relative continuity and the constant evolution of racial relations (Hall, 1992). Based on this line of thought, and our attempt to address central features of the racial welfare state, we focus on: (1) the historical continuity of racial profiling in the Nordic context; (2) the complexity that the phenomenon encompasses; and (3) the central changes.
In order to secure the readability of the text, and to uphold transparency, the gender and citizenship of the informants are indicated in the analysis, as well as ethnic/regional background. In the effort to safeguard the anonymity of our informants, the age of each is given an approximate value. For the sake of argument, we focus exclusively in the first section on experiences of our Roma informants: given that this group's trajectory in the Nordic region goes several centuries back, it underlines the importance of historically situating practices of racial profiling. In the following two sections, we turn to experiences from a broader selection of informants, thereby adding nuances to our discussion.

Enforcing order: Continuity
Police practices that seek to control the mobility of those defined as the deviant 'others' are a central part of the history of the modern state in the Nordic region. In our material, this is especially visible in the stories told by the Roma interviewees, who have a long trajectory of being exposed to systematic state repression: Now, you said that we should tell our own stories. However, this is about my grandfather. He and grandma check in to a hotel. A few hours later, the hotel manager knocks on the door. 'You have to leave quickly, there's a whole horde of people outside.' This was, of course, a long time ago, maybe 50-60 years ago. [. . .] Grandpa pulled the door shut with his belt, and then they ran for their lives. Jumped on a train, bought the tickets there. This is what I can say: no police officers were on their side in that situation. They were part of the crowd.
(Roma/Sweden, 25, male) The interviewee describes a dramatic incident that has lived on in the family's memory for over half a century. He relates a story about his grandparents having to escape to save their lives, while the police officers not only did not protect them but, rather, were 'part of the crowd'. Even though this type of incident is not part of the everyday of our informants today, it is a key point of departure for understanding on whose 'side' the police have been throughout history. Our interpretation is that the informant's statement is an example of the ways that exclusionary and racially ordering state practices have been part of the Nordic welfare model all along (Jensen and Loftsdóttir, 2012). We deduce also that this embodied family memory highlights the power relations of the racial welfare state, and stands in sharp contrast to the ideas of the Nordic model as exceptional and inclusive. How can we make such a claim?
First, the story told has resonance in well-documented events. The best-known was in 1948 in the Swedish town of Jönköping where a mob, without the police intervening, physically abused and chased members of the Roma community out of the city (Ds, 2014: 8). Second, in line with the argument of critical race scholars, counter-stories such as this, that disrupt dominant frames of understandings, are crucial as a point of departure when analysing racism because they allow for an engagement with history from the perspective of those traditionally silenced (Solórzano and Yosso, 2002). Third, even though the informant knows that it is asked of him and the others to 'tell our own stories', he uses the event to situate his own experiences of racial profiling-underlining that the past is interlinked to the present (Glover, 2019: 379). By challenging the premise of the interview, he reveals a deficit in the line of questioning: the narrow timespan that follows from asking about personal experiences. The experiences shared by several of the informants underlined the importance of understanding the past through previous and early encounters with the police: When we drove through Sweden, as we often did when I was younger, there were cities such as Örebro, where you knew that the police were stopping Roma. They dragged the whole family out of the car in order to intimidate us. They harassed and frightened the children, so that you would take a detour.
(Roma/Sweden, 50, woman) While the first citation illustrated a form of racial profiling laissez-faire in the sense that interactions between parties were free from governmental intervention, the quote above represents the opposite tendency: racial profiling as an exaggerated use of force. It is important in this regard to understand racial profiling as part of the paradox of overpolicing/under-policing of minorities and groups racialized as non-white (Rios, 2011: 54). More specifically, the interviewee refers to a cross-generational experience in which a whole city turns into a no-go zone. The situation can be interpreted in relation to anthropologist Didier Fassin's (2013: 92) description of how police stops are a form of physical education, where 'the habit of humiliation is designed to produce the habitus of humility'. In this case, the informant describes how the police would 'intimidate us', and target specifically the children, making whole families 'take a detour'. The practice keeps an ethnic minority group out from a public place, and thus 'in their place'. Pivotal is that what can be referred to as Driving While Roma is not a phenomenon from the past, but characterizes everyday experiences of our interviewees today: The awareness created by historical continuities of disproportionate police actions is present in the quotation, where the interviewee tells about being treated as a suspect due to a claimed 'familiar' look, which we understand to be an outcome of a stereotypical and racializing police gaze. While the informant is hailed as a suspicious subject, her response is not submission but contestation (Glover, 2009). 'I said: "Really?"' Her questioning of the police in the situation is a way to resist the fact that she is not treated as a citizen with equal rights, even though she has a 'social security number'-a modern technique of the state to administrate, classify and distinguish those who are perceived as non-belonging from those who have rightful access to residence. As such, racial profiling is part of a racial order that reduces rights to something held not by humans per se, nor by citizens, but by people who embody the imagined characteristics of the national territory (Anderson, 2016(Anderson, [1983).
The situation related by the informant can also be interpreted through Goldberg's (2002: 105-106) description about the exercise of racial rule as deeply ambivalent and caught 'in the struggle between subjection and citizenship', where this latter status is 'never quite [to be] reached for the racially immature'. We connect this to the idea of the racial welfare state by emphasizing the connections of policing to other welfare institutions, such as schools: The police make school visits: they come and speak in front of the class. My son was in fifth grade. [. . .] When he gets home, he tells me that the police officer had extended his hand and knew his name. He doesn't get it, of course; he's in fifth grade. But to me, it's clear that it was about a surveying of our boys. So the next day, I go to his school and say to his teacher: 'How strange. Do these cops know all the children by their first names? Because my boy has never even met the police. And I'm not a criminal, so there's no reason. Except that we're Roma.' Then they said: 'No, no, but there was a situation in which we talked about various boys at the school. His name probably just came up.' (Roma/Sweden,40,woman) Many Roma informants emphasized the role of other state institutions than the police when recapping stories about exclusion and criminalization, even though this was not the focus of the interviews. The reaction from the teacher in the story is illuminating in this context: 'His name probably just came up.' The lack of critical attitude towards the police, and state authorities in general, is a characteristic feature of the Nordic model and, we argue, a central feature of the racial welfare state. In order to understand the informant's reaction, one needs to take into account the two-faced character of the racial welfare state: those who can claim unquestioned belonging would probably not react as strongly as the informant does. Perhaps they would even perceive the outstretched hand as a nice gesture, as an expression of the community-building police work that evidences the trustworthiness of the welfare model and its institutions. In contrast, this informant sees it as 'a surveying of our boys'. From her position, the (allegedly) well-meaning and supportive gestures of the state agents have a totally different meaning. What fuels her aversion is the long tradition within the police of generating discriminatory ethnic registers: the 'Roma register' kept by the police in southern Sweden, revealed in 2013 and that resulted in the conviction of the Swedish state for discrimination, is but the latest example. Therefore, to grasp racial profiling as a contemporary phenomenon, it has to be examined through the lens of continuity and perceptions of the past: through the history of the racial welfare state.

Enforcing community: Complexity
In the previous section, the focus on historical continuity brought racial profiling of the Roma community in Finland and Sweden under the spotlight. In this section, we elaborate the complexity of who gets targeted in the contemporary conjuncture by engaging with a broader range of experiences. In addition to Driving While Roma, racial profiling also takes the form of Driving While Black: I've been stopped many times when driving, totally unnecessarily. [The latest time] the police turned their car around and drove after us. [. . .] They never told us the reason; there was no real reason. Everything was in order, as always, but they wanted to see who was there and inspect us.

(East African background/Finland, 30, man)
What the informant highlights is the tension between legitimate controls and stops without 'real reason'. Even though he classifies the controls as 'totally unnecessarily', they are not seen as illogical: the aim of the police is to 'inspect us'. The emphasis on having everything 'in order, as always', is illuminating. What the informant points towards is the way that the police situate race in a structure of dominance, making controls and stops a far too common experience. However, the police officers interviewed in the two studies we build on usually denied racial profiling as part of police practices. They tended to regard their work as race-neutral and based on facts, building on a colourblind ideology typical for the Nordic welfare state context (Lundström and Teitelbaum, 2017). Beyond this idealization, the interviews with police officers also opened up for more complex understandings of the role that race has in their work: If, for example, we are carrying out immigration control, the fact is that we both probably understand what is meant when we say a man looks Finnish, Swedish or Norwegian, for example. So we probably think that you are a white person. When one is thinking quickly. So when thinking about it like this, certainly fewer white people, for example, than dark-skinned people are checked in immigration control. I think that is pretty obvious.
(Police officer /Finland,35,man) In the studied countries, the law does not allow police controls based solely on how a person 'looks'. They need some ground for suspicion. Thus, the statement contradicts legal arrangements but also prevailing discourses within the police in the Nordic countries that emphasize equal rights and non-discrimination (Wieslander, 2014). Nevertheless, in the interview, this informant openly explains how migration and border controls are interwoven with notions of national and regional belonging, at least when 'thinking quickly'. The logic of the informant sheds light on the fact that the institutions of the modern nation-state produce race through racializing practices (Omi and Winant, 2014), and more concretely that the order reproduced is based on perceptions of whiteness visa-vis who is perceived as 'dark-skinned': We use skin colour as a selection criterion in certain situations. If you're carrying out an internal immigration control, as the police are obliged to do since we joined the EU, then [. . .] if you see two individuals in the subway, one a towhead and the other like you, well, it's not really surprising that someone looking for foreigners automatically looks at you and not at the towhead.
(Police officer/Sweden, 60, man) The informant refers to the interviewer, who has Latin-American background, making explicit what might be apparent, and by doing so actually recognizes the white hegemony on which the Nordic racial state formation is based. An interpretation is that this informant, as well as the one quoted previously, is describing how they as professionals are grappling with a structural dilemma produced by the colour-blind ideology of the racial welfare state. On the one hand, the Nordic states do not officially allow discrimination based on racial or national categories. On the other hand, the agents of the states are in charge of enforcing the internal and external borders of the imagined community. As such, their logic can be better understood through philosopher Étienne Balibar's (1991: 47) description of how race and nationalism are 'evacuated within theory' in the modern, liberal nation-state, but constantly re-enter 'by the door of practice'. For our understanding of the racial welfare state, this contradiction is crucial to recognize, because it is from this position that the informant acknowledges the practice of racial profiling, but refers to it as the police being 'obliged' to do so 'since we joined the EU'. This line of argument allows the informant to distance himself and the national authorities from the responsibility, and gives the impression that the police practice described is actually a task not freely chosen.
In our view, by invoking the importance of appearance in relation to practices of migration control, the informant is actually underlining the racial aspects of the transnational crimmigration control system (Bowling and Westenra, 2018). If this interpretation is correct, racial profiling can be conceptualized as an expression of the 'white man's burden', in the sense that racial rule is perceived as a moral responsibility towards the member states of the European Union in the quest for civilization and order (Hansen and Jonsson, 2017). A parallel and not necessarily competing interpretation is that racial profiling is part of the 'global dimensions of penal power and its colonial history' (Bosworth et al., 2018: 43). Racial profiling in the racial welfare state can, from this point of view, be seen as a sign of 'colonial complicity', referring to the region's active participation in, and support for, the civilizing project of the West despite not being among its core empires (Keskinen et al., 2009). In both interpretations, the informant's reference to the 'towhead' is understood as a marker of whiteness and inclusion regulating the boundaries of the national space, as well as of the Western European imagined community.
While the extracts from the interviews above can give the impression that the police depart from a perspective in which the population is divided into citizens/non-citizens based on the dichotomy of white/non-white and ethnic minority/majority, the imagined community reproduced through practices of racial profiling cannot be reduced to such a simplistic binary model. People are read in a more multifaceted interplay of several categories and their markers, which means that the assumptions of who belongs and who does not belong are not always clear-cut, but contextual and shifting: My friend lives in Russia, and he came to visit me. [. . .] So we were just happy that we'd come to the park, and like we were probably visible, because we're quite loud, and we were laughing and like ha-ha-haa and ha-ha. Then two policemen arrived [. . .]. We didn't do anything; I don't think that we disturbed anyone. After all, it was in the evening, and other people were there as well having fun.
(Eastern European background/Finland, 20, man) The example can be interpreted in several ways. On the one hand, it can be taken as a case in point concerning the significance of examining noise and language in relation to police activity: 'we're quite loud', the informant explains. As such, it complicates the discussion of 'looks' for policing. Is it behaviour-and not underlying conceptions of race-that makes the police focus on this particular group? It should be noted that there is a debate concerning the productivity of applying the concept of racialization to understand hierarchies in regions where the western systems of racism based on white supremacy were not adopted during the colonial epoch, such as in Eastern Europe (Komlosy et al., 2016: 311). The informant himself, nevertheless, underlines the importance of appearance: 'we were probably visible'. He also stresses the different treatment by the police towards them compared with all the others who were in the park 'as well having fun'. An interpretation, following the logics of the officer quoted previously, who expressed that there is a shared perception regulating who 'looks Finnish, Swedish or Norwegian', is that the police do not enforce an imagined community based on whiteness in general, but Nordic whiteness in particular: this is the norm against which bodies are measured and certain groups are criminalized. This specific form of whiteness has been constructed in relation to the West, making the informants who are from the eastern parts of Europe targets of the police.
The importance of scrutinizing how the European community is imagined and boundaries of belonging are drawn in different ways is underlined by several examples in our material. The most salient relates to EU citizens from Romania and Bulgaria, many of whom identify as Roma: this is a group particularly vulnerable to police control and discriminatory harassment. As criminologist Katja Franko Aas (2011: 343) has noted when examining the intertwining of crime and migration control: the right to free mobility in the European Union is a privilege extended only to certain groups.
Our argument is that the construction of whiteness, and who counts as the racialized 'other', is a flexible process constantly and contextually defined. Remarkably, the informants themselves offered in-depth analyses of this phenomenon. In the following quote, the interviewer with Latin-American background is again used as a comparison, this time not as an example of a person who does not belong because his body is read as non-white and non-European, but because he is thought of as a temporary visitor in the territorially stigmatized neighbourhood where the interview was conducted: Perhaps the police would have taken you for a customer. [. . .] With your hat, you stick out here. But if you go into town, you'll blend in. We blend in here. But if we go into town like this [pointing to his sweatpants], we'll stand out. There, I dress a little nicer, put on a pair of jeans or a nice sweater. No dark clothes. Hoodies are completely forbidden.
(Middle Eastern background/Sweden, 25, man) The informant's description clarifies what a 'customer' of illegal substances is assumed to look like, pointing towards the importance of exploring how the race-crime nexus is forged, and how it differentiates from racial profiling related to policing practices focused on identifying immigrants due to their status as non-citizens. In a sense, what is indicated is how the internal confines of the racial welfare state are policed based on a racialized mental map of the imagined community. We suggest that the hat worn by the interviewer marks him out as non-local and middle class in the racialized poor suburb. As such, the clothing is a marker of race, class and place, as well as gender and age, that exceeds binary categories such as white/non-white, ethnic minority/majority as well as citizen/non-citizen. The boundaries of belonging upheld through practices of racial profiling should therefore be explored in the intersection of axes of inequality (Parmar, 2020), as well as in the interlinked relation between racialized notions of crime and racialized processes of criminalization (Armenta, 2017). What makes bodies 'stick out' or 'blend in' in the racial welfare state is a complex question, and it is therefore central to explore the multifaceted ways that the police enforce community.

Enforcing borders: Change
According to our interviewees, situations of racial profiling occurred in streets, parks, neighbourhoods, harbours and airports, as well as busy traffic hubs such as railway and metro stations. The material indicates the dispersion of borders from a focus on merely national territory and, at the same time, the requirement to police the European/Schengen borders locally in city spaces and other everyday environments: At the time when we were monitoring immigrants, for example, at the railway station [. . .] we caught some people who had already applied for asylum in some Southern European country years ago and stayed in Europe for years. During [2015] they had been able to travel through Europe quite easily. (Police officer/Finland,30,woman) The informant addresses the role of the police in administrating, in this case, the internal borders of the national territory. For the purposes of our article, it is particularly necessary to underline that the informant is referring to a particular time period. She is talking about how the police reacted to the influx of migrants in 2015, which in mainstream discourse was discussed as a 'refugee crisis'. We argue that periods when the police prioritize 'monitoring immigrants' accentuate the racial mission of this authority (Amar, 2010). The new migration regime enforced in the Nordic countries has deeply impacted the welfare model, paving the way for the increasingly authoritarian ideology to restructure racial relations. The move towards a more chauvinistic racial welfare state (Keskinen, 2016) makes the governing authorities less interested in protecting the rights of immigrants, ensuring that they have decent living conditions, and more attuned to the perceived problem of their being able to 'travel through Europe quite easily'.
Regarding what counts as change for the phenomenon of racial profiling, a central feature concerns those perceived as the racialized 'other'. Today, the figure of the immigrant is closely related to that of the Muslim (Fekete, 2009). We have elsewhere argued for the importance of paying close attention to ways that racial profiling affects this group (Schclarek Mulinari, 2019b). It could be maintained that the large-scale presence of Muslims in the Nordic countries is a phenomenon mainly associated with migration over the last few decades, thereby centring the discussion on what counts as new in demographic change. It could also be argued that the politically viable Islamophobia in the Nordic countries represents a new chapter in the global and local histories of racisms (Kalra et al., 2013). Without discarding these lines of thought, our claim is, however, that there are clear parallels to previous conjunctures: internal and external bordering practices have been a technique through which the Nordic states have channelled out the perceived racial threat for centuries. Consequently, we propose that what counts as new in the way that racial profiling works has to be addressed in relation to the (dis)continuity of the racial welfare state. Or as Nira Yuval-Davis and colleagues (2018: 203) have pointed out: we are witnessing 'de-and re-bordering processes' that 'involve the territorial displacement and relocation of borders and border controls'.
In the previous section we have discussed the expanding relevance of the transnational crimmigrant control system, and argued that racial profiling should be traced in relation to the embeddedness of the Nordic region in the western colonial project. We have in previous sections also indicated the relevance of institutions such as schools in relation to ways that policing is carried out in the racial welfare state. There is certainly a spectrum of situations and institutional settings to explore. An important aspect of change and everyday bordering processes in this regard not yet touched upon is related to the private security guards, whose presence has expanded drastically in recent years (Saarikkomäki, 2017 Shopping while 'you happen to look like an immigrant' is a practice that comes with a risk of being targeted. The prominence in the material of racial profiling by private security guards points to the need to address the neoliberal restructuring of the Nordic welfare states and how it has affected policing. Through the privatization of the monopoly on dealing with violence, the market increasingly regulates racial relations, making race to be to a lesser degree marked by formal state formation (Goldberg, 2009). This is illuminated by the informant, who, rather than identifying as a citizen, categorizes himself as a 'customer'. From this it follows that racial profiling should not only be seen as a bordering practice that produces boundaries between Europe and its 'constitutive outside', but also between those who are perceived to be able to consume legitimate goods and those who cannot.
The policing by private security guards bears witness to a shift in the mode of government in the wake of neoliberal policies, with its sharpening of socio-economic inequalities as well as the increased repression of the racialized urban poor (Wacquant, 2009). A central aspect in this shift is the replacement of redistributive policies to spaces of intervention that have become a focal point for police action (Dikeç, 2007). The spatial reconceptualization of the crime problem is a key feature in new policing strategies developed: Police officers spend their time in public places to prevent crime. We're mainly in areas where people are socially disadvantaged and have a low income. Ethnicity has to do with it because it's mainly immigrants who live in these places. [. . .] Calling it racial profiling is difficult. You can't complain that it's racist. We have to be there. All indicators of where crimes are committed and where people are unsafe show that this is a hotspot. That's where we should be.
(Police officer/Sweden, 50, man) The informant states that it is 'mainly immigrants' who reside in 'low income' areas, offering a valuable clue to understand the racialized geography of law enforcement (Gaston, 2019). The procedure of gathering 'indicators of where crimes are committed' seems objective and neutral, even a necessary way to conduct policing in the attempt to identify the 'hotspot'. However, racial profiling can be understood as part of the historical outcome of race becoming inscribed in space through territorial segregation (Owusu-Bempah, 2017). As such, it is part of the global world-system that situates mostly the racialized poor in places where there are fewer resources (Golash-Boza, 2016). Analytically, what the statement underlines is the importance of the categories of class and place for policing race in the racial welfare state. In a sense, the informant's line of thought is typical for the welfare ideology that has regulated the Nordic states: while obscuring the role of race and racism, social factors and colourblindness are emphasized.
In this regard, it is important to address the difference between the situation in Sweden and that in Finland. While the Swedish interviewees clearly highlighted the consequences of the racialized and classed spatial division of the city, the Finnish informants pointed to the phenomenon in a more geographically dispersed way. This may be due to Swedish residential areas being segregated more along ethnicity and race than in Finland. This fact highlights the need to address not only similarities but also differences in the patterns of racial profiling in the racial welfare state setting, and also shows the need for future analysis.

Conclusions
In this study, we have demonstrated the importance of ground research on racial profiling in perspectives that situate the phenomenon in local, regional and global frames. The analysis has developed the concept of the racial welfare state, a racial formation that we contend is characteristic of the Nordic region. We have underlined the link between policing and the flexible and multilayered role of Nordic whiteness, the geopolitical division of power that structures racial relations locally through patterns of segregation as well as through its interconnectedness to the transnational crimmigration control system. Our overall argument can be read in parallel with the following statement: '[a]s racism extends to the new, it of course builds on the old, so that its articulations and representations overlap as they also reform, becoming ever more complex' (Kalra et al., 2013: 3).
We show the importance of addressing the (dis)continuity-the relation between past and present-for the way that police relations with minorities have been administrated, expounding the understanding of the Nordic model as deeply racial. Our argument is that, even though state policies are based on colour-blind universalism and egalitarianism, the welfare state is-and has been-organized along the lines of racial hierarchies. In this context, the police are engaged in racializing work; that is, they divide people and territory based on racial categories. This, according to us, is not a question of individual shortcomings but a structural dilemma produced by the racial welfare state: while the law prohibits discrimination based on racial or national categories, the agents of the state are tasked to enforce the internal and external borders of both the national and European imagined communities.
The equating of whiteness with nation-state belonging is a key characteristic of the Nordic region. What the informants highlight is the continuum of white hegemony reproduced through law enforcement. While whiteness is more generally connected to Europeanness, the Nordic example also points to the fractures within European whiteness, with certain groups being perceived as more and others less white. In the Nordic context, the division of space and bodies reflects notions of how, for example, Central and Eastern Europeans can be located outside whiteness when national belonging is determined, as well as being subjected to the crimmigration control system. The changing contours of the category of white, wealthy and (Western) European point to the need to understand the transformative nature of how crime and criminals are defined in a process of racialization-and to update the analysis in tandem with political and economic changes. In this context, the role of the police is fundamental. As we have demonstrated, the police in the Nordic countries are pivotal, and have long since been, in the upholding of hierarchies, reproducing boundaries of belonging, and functioning as pillars of the racial welfare state.

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

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/ or publication of this article: Leandro Schclarek Mulinari received funding for the research and publication of the article from Civil Right Defenders in Sweden. Suvi Keskinen received funding for the research and publication of the article from Kone foundation in Finland.

ORCID iD
Leandro Schclarek Mulinari https://orcid.org/0000-0002-8439-1701 Note 1. The Finnish study also included participatory observations and a survey, but these data sets are not discussed in this article.