Claims to a nation, dressing the part and other boundary making strategies by skilled migrants in response to ethnic categorization

ABSTRACT This article is about self-defined social identities, other people’s perceptions of us and the potentially conflictual relationship between these two. Building on a Barthian focus on group boundaries, the article takes the interplay between external categorizations and internal group definitions as its point of departure to examine how individuals negotiate the boundaries of their social identities. Based on a case study of skilled migrants with racialized ethnicities in Finland, I look at how they express their self-defined identity as well-to-do, skilled professionals in the face of contradicting categorizations of them as un-skilled, lower-class migrant subjects. I identify two types of complementary approaches employed by the skilled migrants in boundary making strategies to their identity negotiations: those de-emphasizing ethnicity (or its importance), and those emphasizing class status. These approaches are two sides of the same coin; coming from different perspectives, they both aim at a more positively viewed identity, and for individuals to be seen as well-to-do, educated, working professionals, rather than as ethnic migrant subjects. As such, the article also highlights the interconnection of class and ethnicity for the social identities of skilled migrants in Finland.


Introduction
I am sitting in a café in Helsinki with an interviewee, discussing the ins and outs of life in Finland as a skilled migrant. On asking him how he feels the Finns see him, just walking down the street or meeting him for the first time, he says: 'They think I'm a refugee living off their social system'. My interviewee is of Indian ethnic heritage. He has lived in several countries in his life before coming to Finland 12 years ago to study for his master's degree. He runs his own healthcare practice in the centre of Helsinki. He recently acquired Finnish citizenship and speaks fluent Finnish. Slightly taken aback by his answer, I urge him to elaborate: Here the thing is, that in Finland you can not make out … there are not so many foreigners here who are really highly educated and working in well-up places so they don't have a … It's really that you belong to this group and we're all foreigners and that they think that I might be also a refugee.
He lets out a little laugh. I ask him how, in an ideal world, he would like to be seen by Finns.
That's a very nice question. I think he should see me as a person who is working here and paying the taxes and doing a normal life and that I'm doing everything in a normal society that a person can do. That is, working, paying taxes and living a normal life. But that perception is not there, but it is what I would like them to see, just a basic working professional or normal person working and paying his tax. That would be nice. This interviewee's views are not exceptional; my data contain many similar examples of skilled migrants with racialized ethnicities who feel that they are assumed to be refugees or asylum seekers and that they are categorized as uneducated, unemployed individuals on the basis of their somatic features. They are perceived through the racialized image of 'the migrant', who is portrayed as 'a non-privileged, non-white, non-western (refugee) subject in search of a better future in "our" country' (Lundström, 2017, p. 79). This image is in conflict with how they see themselves as well-to-do professionals (Koskela, 2019).
This article is about boundary making strategies against categorizations that are perceived as conflicting with one's self-defined identity. Boundary making in this context refers to the formation of boundaries between groups as a result of intergroup negotiations (Barth, 1969;Jenkins, 1994Jenkins, , 2000. These boundaries, and the identities they enclose, are not predetermined or immutable; they are seen as invented, socialized and constructed through interaction. The malleability of boundaries becomes 'particularly explicit during migration' (La Barbera, 2015, p. 3); in the structures of a new host country, migrants have to (re-)negotiate their identities within new categorizations, new social locations and in terms of new Others (ibid; Chryssochoou, 2004).
Based on a case study of skilled migrants in Finland, in this article I look specifically at the experiences of those skilled migrants who have ethnicities that are racialized and valued negatively in the Finnish context. In general, skilled migrants with high educational qualifications, and relevant occupational skills and earning levels are seen as the most desired type of migrants in Finland (Avonius & Kestilä-Kekkonen, 2018;Jaakkola, 2009). Still, many of them feel primarily judged by their somatic features rather than any educational or work-based merits (Koskela, 2014(Koskela, , 2019. Ethnicities that are viewed negatively in Finland include most sub-Saharan Africans, especially Somalians (FRA, 2017;Keskinen et al., 2018). A recent study of 12 European countries showed that migrants from sub-Saharan Africa are most likely to encounter racist harassment specifically in Finland (FRA, 2018). Moroccans and anyone considered 'Arab', and to a lesser extent, Asian immigrants are also valued relatively low in studies of Finns' 'ethnic hierarchies' (Jaakkola, 2005(Jaakkola, , 2009. These are all groups that stand out due to their somatic features (e.g. skin colour, hair, facial features). Other groups that are discriminated against include Russians and the Roma people (FRA, 2017;Jaakkola, 2009), more for historically specific reasons and other forms of visibility (such as language and ways of dressing, Leinonen, 2012a), than racialized somatic difference.
A historical narrative of Finland as a homogeneous, white nation is constructed by processes of Othering against these minorities (Keskinen, 2019). The majority is typically considered 'above ethnicity', and whiteness hence becomes the norm against which somebody can be considered 'visibly ethnic' and excluded from the image of who belongs to the nation. This article takes the stance, that despite their privileged socioeconomic position, skilled migrants are subjected to similar racializing discourses in Finland as other migrants. Hierarchies based on nationality, language and ethnicity intersect with skilled migrants' class status afforded by their educational and professional standing (Leinonen, 2012b). Especially skilled migrants with ethnicities that are racialized in the context of their host societies exist between two contradictory discourses, and two migrant experiences (see also Mozetic, 2018;Yanasmayan, 2016): on the one hand, they are middleand upper-class individuals with high earning and educational statuses, living their lives at the level of a global economy, a cosmopolitan culture, and positive discourses on internationalization and globalization (Fechter & Walsh, 2012). On the other hand, they enter the realm of judgement on the basis of their racialized somatic features as soon as they step outside of their professional environment (e.g, Jaskulowski & Pawlak, 2019;Koskela, 2019).
This article consists of four parts. In the first, I introduce the theoretical context of the study, and review previous research on boundary making strategies, focusing on minorityhost society relations. The second part introduces my research data on ethnic skilled migrants living in Finland. In the third part, I discuss my findings on how skilled migrants with racialized ethnicities respond to being categorized according to their somatic features and present examples of the strategies of identity negotiations that emerge from the data, which I divide into the two complementary approaches that I see the skilled migrants use. I conclude with a discussion of the main arguments of the article.

Theoretical context
Drawing on interactionist identity theories (Jenkins, 1994(Jenkins, , 2000Okamura, 1981;Wimmer, 2008aWimmer, , 2008b, this article shares a Barthian focus on group boundaries and the processes of social negotiations that construct them, rather than seeing the groups themselves as immutable entities (Barth, 1969). Jenkins' distinction between 'groups' and 'categories' (1994) is central to my argument. According to this distinction, a group identity is our own internal definition, whereas a category is a definition imposed on us by others, for example by a dominant national group on minority groups. In a new setting, such as after migration, one becomes subject to new categorizations (external definition) by the dominant group. 'Social Identity is never unilateral', declares Jenkins, continuing: 'individuals have some control over how they are perceived in the interaction order 1 , but their categorization by others is always moot ' (2000, p. 8). Therefore, they must renegotiate their group identity (internal definition) in this new environment, which is inevitably also shaped by the experience of being categorized. The form that these renegotiations take depends on how consensual or conflictual the external definition is perceived as being, and on the power relations that define the opportunities for resistance against the categorization. For skilled migrants with racialized ethnicities, the categorizations imposed on them are often perceived as conflictual; their self-defined identity is rooted in a class-based understanding of themselves as desired migrants with high educational and occupational status (Koskela, 2019). Yet they feel they are (mistakenly) categorized as lower-class, lower-skilled humanitarian migrants based on their racialized somatic features (ibid). This conflict between internal and external definitions leads to various identity and boundary-making strategies being employed in order to negotiate a more positive, 'truthful' identity for oneself.
Jenkins himself introduces five possible responses to imposed categorization. These range from internalization and acceptance to resisting or even denying the truthfulness of the categorization in regards to one's self-defined identity (2000, p. 21). There are many other examples of theories listing responses against imposed categorization (see e.g. Alba, 2005;Shelton et al., 2006;Tilly, 2004). However, it is Andreas Wimmer's vast work on the subject of ethnic boundary making (2005, 2008a, 2008b, 2009, 2013) that makes the most ambitious claim of being a comprehensive typology of all possible variations of boundary making strategies. Wimmer's typology is divided into five main strategies. The first two are about moving or redrawing group boundaries, either by 'expanding' (1 2 ) or 'contracting' (2) the range of people included within the boundary. These are collective strategies that affect whole groups; for example, an ethnic group resisting the dominant national order. They typically include acceptance by, or at the least involvement of an authoritative power, as in processes of nation building (2008a, p. 1031). The last three are more individual strategies. They concern modifying the meaning or implications of the boundaries in different ways. One can aim to challenge the ethnic hierarchy by 'transvaluation' (3), either by means of 'normative inversion' (a) (reversing the order) or 'equalization' (b) (aiming for moral and political equality). Another type of modifying strategies are the 'positional moves' (4), aiming to change one's own categorical membership of an otherwise accepted hierarchy. These consist of collective 'repositioning' (a) and individual 'crossing' (b) (by 'passing' (c) or 'assimilating' (d)) strategies in order to claim another membership. Lastly, 'boundary blurring' (5) strategies can be employed to emphasize other possible identity categories (2008a, p. 1044, 2008b).
Even though Wimmer's theories are specifically about ethnic boundary making, I feel they adapt well to an analyses of what constitutes a multi-ethnic, multi-national group. Essentially the conflict revolves around the skilled migrants in question having visible, racialized ethnicities and being highly-educated professionals of higher socio-economic class status. Ethnicity and class, or any other social categorization, cannot be understood separately; they are intersecting social locations which affect identity and belonging, both through external expectations and internal importance of ethnicity and class in any given context (Anthias, 2013;Christensen, 2009;Koskela, 2019). In the very idea of 'the migrant', racialized and classed understandings intersect (Guðjónsdóttir, 2014;Lundström, 2017), leading to assumptions about class status merely based on ethnic visibility, as is the case with skilled migrants with racialized ethnicities (Koskela, 2019). While ethnicity and class intersect at this level of categorizations, stereotypes and assumptions, this intersectionality can also be employed for identity claims by individuals (Anthias, 2013). Anthias urges us to also look at 'the functioning of the categories separately as salient aspects of discourse and practise' (ibid, p. 8). It is indeed clear from the case study that both ethnicity and class can be used in boundary making strategies, and that they also appear in the form of narrative and discourse as well as symbolic capital. Morosanu and Fox (2013) agree: 'ethnicised stigma does not always lead to ethnicised strategies for dealing with that stigma. To the contrary, it can also encourage migrants to develop non-ethnicised responses' (ibid, p. 448). Class is employed by the individuals in my data as such a 'non-ethnicised response' to ethnicized categorizations.

Previous research
Many previous case studies discuss migrants' strategies of (ethnic) boundary making against being categorized by others. Much of this research concerns lower-skilled minority ethnic groups trying to distinguish themselves from, for example, 'the standard label of Latino immigrant' (Boccagni, 2014, on Ecuadorians in Italy), or from the stereotyped Arab immigrant (Lamont & Askartova, 2002, on North Africans in France adopting a position as the 'good Arab'. See also Koskinen, 2015, on coping mechanisms against racialization by international adoptees in Finland, and Day, 1998, on factory workers in Sweden). Although offering parallels with the myriad of strategies that are used by ethnic minorities for negotiating more favourable identities for themselves, these studies differ from the case of skilled migrants because their participants lack the same higher class status and all the symbolic capital this brings. In Kennedy's (2007) words, skilled migrants possess a 'middle class tool kit' consisting of economic and social capital from their employment situation, an individualistic position not tied by kin, and a familiarity with global occupational cultures that allows them to navigate their lives in their host country with relative ease (p. 357; see also Leinonen, 2012a). Therefore, their processes of identity negotiations will also be different to those of migrants from lower social classes.
Other case studies take the class status of migrants into consideration but exclude the stigma of racialized ethnicities. The viability of 'passing' as a member of the majority (Lulle & Balode, 2014, on Latvian women in Finland; Lan & Wu, 2016, on Taiwanese students in China) or benefiting from positively viewed difference (Guðjónsdóttir, 2014, on Icelandics in Norway) are strategies that are not available to migrants with racialized ethnicities. However, some studies have dealt directly with skilled migrants with racialized ethnicities. They provide an insight into how economic contribution (Jaskulowski & Pawlak, 2019, on 'visible' skilled migrants in Poland), middle-classness (Chang, 2014, on Taiwanese women in Finland), career status and cosmopolitan identity (Habti, 2012, on Arabs in Finland), ways of dressing (Hirvi, 2013, on Sikhs in Finland and US), or simply emphasizing middle-class tastes and lifestyles (Valenta, 2008(Valenta, , 2009, on Iraqi and Bosnian migrants in Norway) can be employed to deflect from the negative effects of racialized ethnicities (see also Yijälä & Nyman, 2017, on skilled Iraqi refugees in Finland; Mozetic, 2018, on refugee doctors in Sweden). Racializing the 'migrant Other' in order to emphasize whiteness and to be seen as a qualified skilled professional has also been identified as strategy among Russian migrants in Finland (Krivonos, 2017). Another whole group of relevant studies deal with class-based strategies of separation from a stigmatized 'internal "other"' (Morosanu & Fox, 2013, p. 443), for example, Romanian skilled migrants in the UK 'emphasising achieved rather than ascribed characteristics in varied ways' (ibid, p. 452) to escape being mistaken as Roma people (see also Yanasmayan, 2016, on Turkish migrants in several locations; Genova, 2017, on Bulgarian skilled migrants in the UK; Lan, 2012, on migrants from Hong Kong in the US; Nagel, 2005, on Arab skilled migrants in the UK).
The article follows in the footsteps of these case studies. Deploying Wimmer's typology, it adds to the field of research by examining the specificities of skilled migrants' identity negotiations, especially as they relate to being racialized as un-educated, assumedly humanitarian migrants in the context of the normative whiteness pertaining in many Western societies. By focusing on the individual micro-level experiences of the skilled migrants themselves, the article points to the possibilities for individual agency among structural constraints. It offers new insights into identity negotiation strategies by individuals who have symbolic power afforded by their class status and therefore offers a perspective on these negotiations in a situation of less power differentiation between those categorizing and those claiming their own group identity. By introducing examples of class-based responses to ethnic categorization, the article also highlights the myriad of strategies that can be used side-by-side in pursuit of projecting a more desirable or truthful identity in face of contradicting categorizations.

Data and methods
There are about 387,000 foreign-born residents living in Finland, which makes up 7% of the population (Statistics Finland, 2018). Numbers for skilled migration are difficult to deduct from available data but 40% of those with a slightly wider definition of 'foreign background' (themselves or both parents were born abroad) between the ages of 25-54 have a higher education degree (Sutela & Larja, 2015). This number is slightly higher in the greater Helsinki region (45%) and among foreign nationals of Western countries (58%). The data used in this article are part of wider research project on skilled migrants' lives in Finland. I conducted ethnographic fieldwork intermittently between 2008 and 2012 in the activities of various social organizations in Helsinki aimed at an international crowd, attracting many skilled migrants living in the city. For the purposes of the research, 'skilled migrants' were defined as those holding higher-level degrees and/or comparable work experience in various fields of business, technology, finance, education, and so on. Furthermore, a self-identification as a 'skilled migrant' was also considered important. The data include skilled migrants from various countries, ethnicities, professions, ages, and genders.
In this article, I focus on a sub-set of these data that can help us understand the experiences of skilled migrants with negatively viewed, racialized ethnicities. As well as ethnographic material that includes interactions with Finns and other skilled migrants, this sub-set also contains interviews with racialized skilled migrants. These interviews include 3 unstructured ethnographic interviews conducted during fieldwork and 8 semi-structured, in-depth interviews with individuals from the following self-defined ethnic origins: China (2), Ghana, India (3), Iran, Kenya, Panama, Togo, and Vietnam. The interviewees identified these as (one of) their ethnicities, even if some of their nationalities were different: three have dual nationalities, and one considers themselves 'mixed-raced'. The variation of nationalities and ethnicities of research participants included through the ethnographic fieldwork is even wider. Any divergences between self-defined ethnicity, country of origin and current citizenships are mentioned after quoted excerpts where relevant to the argument. Regardless of these distinctions, in this article I approach ethnicity as a social construct whose significance can only be defined situationally and by the subjects themselves. Therefore, the focus of the data collection has been on the perspective of the skilled migrants themselves: all of the participants included in the sub-set of data have racialized somatic features which the informants themselves recognize as being 'visible' in the Finnish context and therefore relevant to their experience as skilled migrants in Finland.
The interviewees were recruited from the fieldwork. All have university-level education, ranging from Bachelor's to Doctoral degrees. They represent many professional fields varying from administrative staff to IT, sales and research. All but one, who was unemployed at the time of the interview, were working in employment related to their education and most could be described as middle-class wage earners. Only one of the interviewees is female. This is not representative, but slightly indicative: the majority of female skilled migrants seem to be from Western countries. The interviewees had been in Finland between 1 and 12 years. Their ages range from early twenties to early forties. Their Finnish-language skills vary from non-existent to fluent. The majority were single. Two stated a religious affiliation. Both the ethnographic and the semi-structured interviews were conducted in familiar settings in different cafés frequented by many of the skilled migrants in their free time. The latter were between one and two hours long and they were recorded and subscribed in their entirety.

Findings: strategies for identity renegotiations and boundary making
I will now discuss what the data revealed about how skilled migrants with racialized ethnicities respond to being categorized according to their somatic features. I have focused on the strategies of boundary making as stipulated by Wimmer and identified two complementary approaches: those de-emphasizing ethnicity (or its importance), and those emphasizing class status.

Strategies de-emphasizing (the importance of) ethnicity
4.1.1. Claims to a nation 'Race', ethnicity, and nationality are interrelated categorizations (Wimmer, 2008b, p. 973). Although ethnicity can be negotiated in a multitude of ways (e.g. Jenkins, 2000;Wimmer, 2008a), nationality has been described as 'a sort of trump card in the game of identity' (Pickering, 2001, p. 90 3 ). It is an unwavering identification in that it cannot be denied, whatever one's skin colour may suggest. As such, it can be a source of symbolic capital. For example, Leinonen discusses how 'the fact of citizenship' positions an African American migrant hierarchically higher than a black migrant from the African continent, even if both may experience racial discrimination (Leinonen, 2012b, p. 249). I have seen such 'fact of citizenship' used as a strategy for deflecting from a negatively valued ethnicity by, for example, migrants who have acquired another, more positively viewed (Western) citizenship at some point in their migration history. Bringing up another citizenship can 'neutralise' the initial negative valuation based on somatic features, such as in the case of this skilled migrant, whose parents emigrated to Canada with him when he was a small child: According to Wimmer's typology, this example falls under 'boundary blurring' (5), whereby 'other, nonethnic principles are promoted and the legitimacy of ethnic, national, or ethnosomatic boundaries undermined' (2013, p. 61). Okamura (1981), also discussing ways of negotiating identities, states that 'individuals have the option of asserting either their primary ethnic identity or other social identities, such as those derivative of class or occupation, that they legitimately hold' (p. 460, italics added). However, some related strategies in the data can be seen as being derived from the borderlines of 'that they legitimately hold'. For example, when an informant introduces themselves as an American, when in fact they are a Turkish national with no other link to the US than having studied at a university there, I interpret that they are doing so in order to gain what they perceive as a more favourable identity or group membership through an affiliation to another nationality (also found by Morosanu & Fox, 2013). This is a 'positional move' of 'passing' (4c) as a member of another, more positively viewed category. Positional moves according to Wimmer are not concerned with affecting the hierarchy itself, they are more about repositioning oneself within the hierarchy in a more positive location. In another example of claiming a national belonging as a 'passing' (4c) strategy, a Ghanaian skilled migrant (IT, male. 3 years in Finland) told me he was 'half-British' as an explanation as to why he was supporting England in the World Cup game we were watching. Surprised that I had not known this despite having known him for over a year, I enquired further. It turned out that because he counted English as his other mother tongue, loved football and had a cultural appreciation for all things British, he 'felt British', adding: 'I never think about the fact that I am black'. He had never visited the UK.
Apart from fluid understandings of national belonging such as these, migration history and life experiences seem to be called upon to help in identity negotiations using other 'boundary blurring' (5) or 'positional move' (4) strategies: I think wherever you live, you … you're taking that culture as part of you. It's just like who your friends are. You're influenced by your environment, so … I have become Finnish. So, I would consider myself part Vietnamese, part Canadian, part Colombian and part Finnish. (Canadian citizen with Vietnamese ethnic heritage, economics, male. Lived in Colombia for 1 year and now in Finland for 10 years) For me, having no [national] identity is a good thing. So, I'm not stereotyped, I'm not a stereotype. So basically, I can make fun of everybody, and nobody can make fun of me! (Indian ethnic heritage, electrical engineering, male. Brought up around the world following father's work in an international company, acquired Dutch citizenship as an adult, 4 years in Finland) Indeed, the migration histories of many skilled migrants' families or themselves are such that they are quite justified in responding to questions about where they are from with an evasive 'It's complicated' (Born in India to Indian parents, grew up in West-Africa and is now a Finnish citizen). Cosmopolitan identities not defined by any one nationality can be used as 'boundary blurring' (5) to 'de-emphasize ethnic, racial or national boundaries' (Wimmer, 2008a(Wimmer, , p. 1042, as well as to 'reposition' (4a) oneself in a better position on (or even outside of) the hierarchy, while not contesting the boundaries or even the hierarchization of ethnic categorizations themselves (ibid).

Deflecting from racism
Those skilled migrants who had negatively valued racialized somatic features seemed to constantly downplay and trivialize the racial prejudice that they had encountered. When asked if they had encountered racism in Finland, the interviewees often used belittling strategies to deflect from their experiences: Well of course you can notice sometimes, I like, sit on the bus, people don't really want to sit next to one … something like this. But it's nothing really like … not in an offensive way, they just … don't. ( (2008a, p. 1037). It accepts the existence of ethnic boundaries, but 'reverses the existing rank order' so that the excluded become 'morally, intellectually, and culturally superior to the dominant group' (ibid). In the examples above, both of the Indian informants place themselves, or at least their own views of themselves, above members of the majority Finnish group whose views they portray as dismissible and indifferent to their lives.
Humour is also used for the same goal of deflecting from racism and ultimately the unwanted importance that ethnicity has in the lives of racialized skilled migrants. My data contain much laughing and joking: things assumed to be negative or difficult because of one's ethnicity or nationality are just laughed about (e.g. Indian guys finding it hard to get dates with Finnish women, or an Asian researcher being mistaken for someone who works in a Chinese restaurant, or an African IT-professional being assumed to be a refugee from Somalia). However, although these are also related to 'transvaluation' (3) strategies, I find that humour as a strategy does not aim at moral superiority; it is rather a plea for equality. In all the examples from my ethnographic data mentioned above, humour is used to draw attention away from the implications of inequality in these situations by disguising them as funny misunderstandings, rather than anything to do with the real character of the person. As such, they are more like what Wimmer refers to as the 'equalization' (3b) type of 'transvaluation ' (2008a, p. 1031). The objective of such strategies is to establish a moral and political equality in regard to the dominant group (ibid). However, I suspect that underneath the humour, downplaying and denial lay more serious issues. At the very least, the use of humour implies an acute awareness of being seen more negatively because of one's racialized ethnicity, despite being a supposedly valued migrant. I see these strategies as also being more about denying the importance bestowed on ethnicity, than trying to deny or distract from one's ethnicity itself. 4.1.3. The company you keep Some strategies are more directly about claiming another, more positively valued social identity. One of the most direct ways of negotiating ethnic boundaries is 'positional moves' (4) in relation to an existing boundary, by claiming membership of another group (Wimmer, 2008b). Here, a Kenyan skilled migrant talks about attending social meetings of international student groups: Actually, I fit in pretty well, because when I go there, those people don't look at me as a Somali or like that, or a Kenyan guy. They say: 'Oh, he must be an exchange student', and I could be from France, who knows, or Belgian […] you fit in pretty well without having to say anything, without having to introduce yourself. And if I go to X [a social club meeting], I look more of an expatriate, because it's more expats that are hanging out there. If I go to work … ok, work, of course, they think I'm an expat, if I go to Nokia, they'll say: 'Oh, he's just an expat'. (Kenyan,IT,male. 9 years in Finland) In this example, the new position is achieved quite directly by choosing with whom one associates socially in order to signal belonging to that group. This excerpt contains all the elements of 'positional moves' of 'repositioning' (4a), 'crossing' (4b) and 'passing' (4c). Conversely, these same strategies can also be employed for disassociation from a negatively categorized group (rather than association with a positive group). Strategies aiming at disassociation are typically directed towards one's co-ethnics (e.g. Boccagni, 2014;Genova, 2017). As racialized somatic features link people together by looks alone, a co-ethnic's behaviour can have a direct negative effect on an unrelated person, resulting in what can be called 'collective shame' (Shelton et al., 2006, p. 328). For example, an Indian skilled migrant told me a story about being on a train, with Indian or Pakistani immigrants in the same carriage who were happily chatting out loud and even singing, as is culturally customary to them. The Indian informant had been in Finland long enough to know this would cause negative reactions from the Finns (which it did, dirty looks were given), so he hid behind his (Finnish) newspaper so that he would not be recognized as Indian and therefore associated with this type of behaviour.
Boccagni (2014) discusses a related strategy that he has coined 'selective disalignment' in his case study of Ecuadorians in Italy. He sees this strategy employed especially 'when systematic alignment with co-nationals is perceived as detrimental for one's reputation in the host society, or for the chances of "moving forward"' (ibid, p. 65). Sometimes such disalignment is not directed at one's co-nationals; instead it takes the form of disassociation from any immigrants who are, for example, seen as ignorant of Finnish cultural norms. In the following instance, as in the example on the train, this ignorance is linked to the belief that Finns customarily appreciate quietness: I'm quite loud [laughing]. So, every time I realize that I … I say to myself 'ok, calm down, be a little more quiet so you're not bothering people so much'. So, I still do that, and I'm trying to change. But I'm also bothered by that when people do that [laughing]. It's like, I was in my Finnish class and two women were sitting like three meters from each other and talking. And I was just like 'why don't they sit beside each other and talk to each other?', rather than like, talk loudly, across the room, while 10 other students were in the room, being quiet. (Canadian citizen with Vietnamese ethnic heritage, economics, male. 10 years in Finland) Wimmer also names 'assimilation' (4d) as one of his 'positional moves' (2008a). Indeed, adopting supposed Finnish cultural characteristics, such as being quiet and non-social in public places, are something that a 'good' migrant should take note of and respect. 'I try not to disturb people', said a Togolese skilled migrant as an explanation to why she does not feel free to talk in her mother tongue in what could be considered a loud voice by Finnish standards in public (administration, female. 1 year in Finland). 'Assimilation' practices place one at least a little closer to the dominant majority group, even if they do not quite allow one to cross the ethnic boundary completely (ibid).

Strategies emphasizing class status
Strategies emphasizing class status are a complementary approach to deflecting from the focus on ethnicity. According to Wimmer's typology, these are all strategies of 'boundary blurring' (5), whereby 'other, non-ethnic principles are promoted and thus the legitimacy of ethnic, national or ethno-somatic boundaries undermined' (2008a, p. 1041).

Speaking and dressing the part
During my fieldwork, I observed countless situations in which a skilled migrant with racialized ethnicity drew attention to or emphasized their class status by diverting conversation to issues related to work, their profession or other class markers (see also Habti, 2012;Morosanu &Fox, 2013 andValenta, 2008 for similar findings). This approach was most often employed through direct articulations to the audience; for example, concerning their occupation, the fact that they worked full time, earned comfortable amounts of money, paid taxes to Finland, and even that they had come to the country on a working visa (rather than as refugees or asylum seekers). In addition to such direct emphasis on class through discourse, class markers can be communicated through 'the ways of dressing, forms of non-verbal communication or "typical" consumption and leisure activities' (Boccagni, 2014, p. 6). Many of the Indian informants remarked that they made an effort to 'dress Western'. The same applies to wearing business attire to assert one's status as a professional: And my dad, who used to wear t-shirts and jeans before, once he moved to Holland, or Muscat, even in Oman, he always used to wear formals. Just so that it shows [that he's a professional person]. (Indian ethnic heritage, brought up around the world, acquired Dutch citizenship. Electrical engineering,male. 4 years in Finland) Although in Wimmer's terms, any use of class references is primarily considered a 'boundary blurring' (5) strategy, I have seen examples of when class markers, such as ways of dressing, may also be used more like 'positional moves' (4). As already discussed above, in regard to strategies de-emphasizing ethnicity, 'repositioning' (4a) or crossing (4b) can be achieved by disassociating from one's co-ethnics. In my data, strategies involving ways of dressing were also mostly about disassociation, about not wanting to be confused with other types of migrants, especially humanitarian migrants, and specifically with lower-class members of one's own ethnic or national group. Below, for example, a Kenyan skilled migrant talks about not wanting to be classed as a hip-hop-clothes-wearing, second generation African immigrant: I would like to say I would like people to see me based on … how I'm dressed more like … you know, if you're dressed like a decent person, then you're a decent person. If you dress like a hooligan, then you're a hooligan. If you dress like a responsible person, then … as I said.
Because we're all humans, and first impressions say everything about you, most … well 50% before you speak. You can tell if somebody, he might be Somali or whatever, but if he's dressed like casual decent person … […] if somebody is dressed up appropriately, treat him like any other person. (Kenyan, IT, male. 9 years in Finland) Hirvi (2013) has discussed how Sikhs in Finland also negotiate identities through dress. Regarding the decision to remove the Sikh turban, she states that: They do so in order to avoid trouble, harassment, racist attacks or just the gaze and questions of other people. Adopting such situational dressing practices can be further seen as a means of obtaining respect from others by communicating a willingness to adapt to the prevailing cultural norms of a particular context. (2013, p. 85) Albeit this leans toward 'assimilation' (4d), and although the Sikh turban might be a more visible sign of (religious) identity, skilled migrants, even without such obvious symbols of Otherness in Finland, consider what their way of dressing might portray to others in terms of their social status, class, and ultimately their social identity.

'Us' versus 'them'
In trying to define their social identities, 'immigrants strategically try to adopt cultural markers that signify full membership and distance themselves from stigmatised others through boundary work' (Wimmer, 2009, p. 245). Turning our attention back to the boundary focus of interactionalist theories, much of the class rhetoric involves drawing a boundary between 'us' and 'them'. Being able to benefit from a higher class status relies on a comparison to be made with lower-class, less-educated migrants. In my data, this is articulated through references to earning one's own money and not relying on the Finnish government for anything, and how this should separate skilled migrants from the problems of humanitarian migration. I also witnessed direct anti-immigration sentiments from skilled migrants. This tendency is more pronounced among those who themselves have negatively viewed ethnicities. For example, after an 'immigration-critical' populist party first appeared on the Finnish political scene in 2010, an Indian skilled migrant who had already lived in Finland for 10 years told me he voted for them in the local elections because immigrants should not be molly-coddled and 'handed stuff for free' (Indian, solo-entrepreneur, male).
As well as erecting a boundary between skilled migrants and other types of migrants, this 'good' versus 'bad' migrant rhetoric is also used to lessen the boundary between skilled migrants and Finns. This is achieved through the morality of the 'transvaluation' strategy of 'equalization' (3b), such as in the quote from the Indian health-care professional in the introduction: by emphasizing oneself as a working migrant, who is 'living a normal life', paying taxes to Finland and who can afford a lifestyle that is on par with that of Finns. Class as a socio-cultural construct includes shared subjectivities and practices and does not consist of purely economic considerations (Fresnoza-Flot & Shinozaki, 2017). Therefore, it is possible to further use such 'class-related cultural capital, lifestyle and taste as tools for separation and class identification' (Valenta, 2008, p. 7). By emphasizing their shared socio-economic class with Finns through speech, dress and even consumer choices, the skilled migrants are further strengthening their separation from the image of 'the migrant' as a non-privileged subject (Lundström, 2017).

Conclusions
Skilled migrants with racialized ethnicities feel negatively and incorrectly categorized as lower-skilled ethnic migrants by the majority society in Finland. This is in conflict with their self-defined group identity as educated, working, well-to-do individuals. They respond to this conflict with identity claims that take various forms of boundary making strategies, either de-emphasizing (the importance of) one's ethnicity or emphasizing class status, which can be seen as two sides of the same coin. However, while the strategies involving ethnicity mainly concern emphasizing one's individual differences in comparison to a stereotype, class-based strategies are often based on comparisons to another category of lower-class, un-skilled migrants. Yet the aim is the same; to correct what is felt to be a mistaken categorization and to be seen in a better, more positive light.
I will conclude by summarizing the three main arguments of this article. First of all, even though the imposed categorizations that the skilled migrants object to concern ethnicity, the strategies used in boundary making are based on both ethnicity and class status. This is due to the close intersectionality of ethnicity and class, especially as it applies to skilled migrants with racialized ethnicities: their experiences fall somewhere in between highclass, 'elite' migrants and ethnic immigrant minorities (Mozetic, 2018;Yanasmayan, 2016). Furthermore, the 'middle-class tool kit' (Kennedy, 2007) they possess allows them symbolic means with which to strategically use class in negotiating their group boundaries. Ethnic based strategies more clearly and directly concern escaping ethnic stigma but I see strategies emphasizing class-status as essentially having the same aim; when emphasizing class, the skilled migrants are also de-emphasizing the importance and meaning bestowed on ethnicity. Furthermore, they are challenging the legitimacy of ethnicity as a standard of categorization in the first place (Wimmer, 2013).
Secondly, many of the examples given concern boundary making strategies that define 'us' in opposition to 'them'. This is especially true of class-based strategies, which depict skilled migrants as 'good', valued individuals, compared to 'bad', un-skilled ethnic migrants with little to contribute to the Finnish economy or culture (Koskela, 2010). Paradoxically then, while trying to fight being categorized as such, they uphold and even strengthen this image of 'the migrant' as the 'nonprivileged, non-white, non-western' subject (Lundström, 2017, p. 79), so that it can serve as a comparison, as a description of something that the skilled migrants themselves are not, and in definition as their Other. Yanasmayan (2016) also remarks on this paradox: 'new migrants do not necessarily seek to eradicate established prejudices and improve "ethnic hierarchies". On the contrary, by reproducing the "ethnic hierarchies", new migrants contribute to keeping them alive' (ibid, p. 2055). Maintaining the stereotype is necessary for asserting how one individually differs from the categorizations of the hierarchy. Indeed, the most used strategies in my data were 'positional moves' (4), which do not contest the boundaries of ethnic categories or even the hierarchy itself, rather only one's own position in it. Therefore, 'status change through boundary crossing or repositioning reproduces the overall hierarchy by reinforcing its empirical significance and normative legitimacy' (Wimmer, 2013, p. 59).
This leads me to my third argument, which is that even though the observed strategies concern social identities, they are individual. Having deployed Wimmer's typology to explore which strategies the skilled migrants use, the analyses also pointed to the type of strategies that were absent in the data: namely 'expanding' (1) and 'contracting' (2) strategies. These are the two most resolutely communal types, which aim to affect the categorization of a whole group. First of all, skilled migrants are not a sufficiently cohesive or homogeneous group for collective boundary making strategies. This applies to both how they are categorized by others and how they see themselves (Koskela, 2019). Furthermore, the strategies are about improving their own image rather than about solidarity with one's own ethnic group, or with others who are categorized negatively. Shelton et al. (2006) state that due to the potential of being viewed in a stereotypical (and inaccurate) way, individuals focus on 'how they personally are being seen through the lens of their group membership' (ibid, p. 324). These strategies are hence essentially communally worded individual motivations.
Wimmer agrees; these strategies are used by individuals 'as a means to justify or at least make plausible his or her claims to honor, power, and wealth ' (2005, p. 56). They are about reclaiming the right to define oneself, about claiming the identities that are important to oneself and which one perceives as valued. Individuals negotiate the categories imposed on them 'in such a way as to give legitimacy to their own claims to moral worth and social standing and to place themselves at the top of the prestige pyramid' (Wimmer, 2013, p. 94). Thus through all these strategies, ethnic skilled migrants who feel mistakenly categorized by others are stepping up, asserting their agency, fighting prevalent categorizations, and claiming their right to define themselves. Notes 1. Jenkins is here referring to Goffman (1983). 2. For ease of reading, the strategies will be referred to numerically when mentioned later in the article. For a comprehensive chart of the types and their subtypes, see (Wimmer, 2008a(Wimmer, , p. 1044. 3. Pickering is here referring to Calhoun (1997).

Disclosure statement
No potential conflict of interest was reported by the author(s).

Funding
This work was supported by Helsingin Yliopisto; Suomen Kulttuurirahasto.