Becoming tolerable: subject constitution of Roma mediators in Finnish schools

ABSTRACT Although Finnish politics relating to the Roma tend to be perceived internationally as fairly successful, several obstacles exist for the Roma in education and the labour market. Training of Roma mediators has been actively promoted in Finland to improve the school performance and equality of Roma pupils. This article, based on ethnographic research, focuses on exploring how the current discursive terrain around the topics of tolerance and prejudice functions in the everyday work of mediators. It is argued that the present discourses in school expose the mediators to unequal power relations of tolerance. The terms for being tolerated are set by the potential tolerating actors, the school community. The mediators aim to supply knowledge about the Roma and try to address prejudices as representatives of the Roma. The study identified three different strategies that the mediators used when encountering prejudice: making sure one does not seem too different, parody and feigning naivety. The analysis suggests that the present discursive terrain creates obstacles to addressing inequalities, discrimination and racism in educational contexts. The responsibility for tackling discrimination is placed on the shoulders of individual Roma – not the whole school community.


Introduction
Since the 2000s, both the Council of Europe (CoE) and the European Union have emphasised that the Roma experience discrimination across Europe (e.g. CoE 2000;ECRI 2011;EU Council 2011;CoE 2012). This has led to a multitude of projects, as well as to the rise of a field of research and reports about the Roma and Roma integration (Gobbo 2015;Araújo 2016). However, European and national frameworks for multicultural and minority politics and policies have been criticised for their inability to sufficiently tackle inequality, discrimination and racism (Gobbo 2015(Gobbo , 2010Miskovic 2009;Araújo 2016). Araújo (2016, 1) argues that the absence of analysis of the production of racism and difference leads to policy frameworks in which Roma minorities are seen to 'fit modern institutions' inadequately, rather than addressing the processes by which racism and differences in institutions and policies are produced (see also Miskovic 2009).
One example of a tool to promote the equality of Roma is the use of Roma mediators. Individuals with a Roma background, or with a good knowledge of Roma issues, are trained to act as mediators between the Roma and public institutions. Both the European Union and the Council of Europe have promoted the use of Roma mediators in all European countries with Roma minorities (CoE 2012;European Commission 2012;Kyuchukov 2012). This article focuses on Roma mediators working in Finnish schools, where the mediators are usually professional teaching assistants with a Roma background.
Mediation, as a measure designed to tackle the inequalities that Finnish Roma face, is fairly new in Finland, but is expanding rapidly. Along with schools and day-care centres, workers with a Finnish Roma background are being recruited into the social services, elderly services and health services to act as mediators between public institutions and Finnish Roma (The Ministry for Foreign Affairs of Finland [MFAF] 2011; Ministry of Social Affairs and Health [MSAH] 2009[MSAH] , 2018. In the school context, the mediators' work is generally perceived as a manner to facilitate communication between the school and Roma communities, as well as between teachers, other school staff and the parents of Roma children (Rus and Zatreanu 2009). They are also expected to act as role models for Roma pupils, support the identity of Roma children and make pupils feel secure MSAH 2009;FNBE 2010;MFAF 2011;Rajala et al. 2011). In addition, Roma mediators are supposed to educate the school community about Roma culture and promote 'tolerance' (Rajala et al. 2011).
Tolerance, as a concept, has long and contradictory roots (for more, see e.g. Hage 2000;Goldberg 2004;Brown 2006). According to Brown (2006, 2) tolerance talk has experienced a 'global renaissance' since the 1980s. The concept returned to use when multiculturalism came to be considered the fundamental problem of liberal democratic citizenship (Brown 2006). The concept of tolerance surfaced in Finnish educational discussion in the 1990s, when migration to Finland greatly increased (Lappalainen 2006;Holm and Londen 2010), and is used e.g. in the latest government programme (Finnish Government 2015). However, the discourse on tolerance and practices of tolerance have been found to produce asymmetrical power relations, in which the terms for toleration are established through the tolerating agent's position of power (Hage 2000;Goldberg 2004Goldberg , 2006. As Goldberg (2004) states, those tolerated 'have to demonstrate their deserving to be tolerated'. Thus, tolerance discourses produce and position subjects (see also Brown 2006). In this article, we focus on how Roma mediators discuss tolerance to make sense of their work and how they become posited through the tolerance discourses in the school context.

Ethnography informed by post-structural feminist theories
Our ethnographic study was informed by post-structural feminist theories. We analysed how the current discursive terrain surrounding the topics of 'tolerance' and 'prejudice' function in the everyday work of Roma mediators in the Finnish school context. To accomplish this, the analysis draws on Judith Butler's work on subject constitution (Butler 1995(Butler , 1997 and the ways the use of the concept has been developed in the context of education and ethnography (St. Pierre 2000;Laws and Davies 2000;Davies 2006;Youdell 2006aYoudell , 2006b. The process of subjectification is an articulation and form of power (Foucault 1980;Butler 1997), and the process of subject constitution arises through discourses. 1 In our analysis, we use this concept to analyse how discourses and power work in the contexts studied.
In the analysis, subjectification is perceived as a process by which the individual is rendered as a subject and is subjected to relations of power through discourse (Davies 2006;Youdell 2006a;Phoenix 2009). Power provides the conditions of the existence of the subject (Foucault 1980(Foucault , 1982Butler 1997). Subjectification is an ambivalent process in which submission and mastery paradoxically take place simultaneously in the same acts (Davies 2006). This does not mean that individuals do not have agency, but that the agency of an individual is initiated and sustained by the discourses that render the subject. Agency emerges in the junctures of the discourses when discourses are renewed (Butler 1995, 135-136). However, one cannot choose the discourses that one is dependent on for one's existence and agency (Butler 1997). To investigate subjectification in the working life of Roma mediators, we drew on ethnographic methodology. Ethnography as a methodology is able to take the various ways a subject is produced in its contexts into account (Skeggs 2001, 433) and to see how discourses operate (Youdell 2006b;Butler 2006).

The Finnish context for the work of Roma mediators
The term Roma is usually used as an umbrella term for various Roma groups. The Finnish Roma (also called Kale) is one of many such groups. In 1997, Finland ratified the Council of Europe treaty number 157 'Framework Convention for the Protection of National Minorities', defining the Finnish Roma as one of its national minorities. It is estimated that there are approximately 10,000-12,000 Finnish Roma in Finland today. 2 Housing policies since the 1970s, unlike in many European countries, provided permanent residences for the Finnish Roma and there are no segregated residential areas for Roma people (Tervonen 2012). The mother tongue of most Finnish Roma is Finnish and sometimes Swedish. Romani has a minority language status, but very few Roma speak it nowadays, especially as a mother tongue (Hedman 2009).
Until the late 1960s, Roma politics were dominated by non-Roma. In the late 1960s, however, they came to own their political organisations and movements, such as the Suomen mustalaisyhdistys (Finnish Gypsy Association). 3 The new movements and organisations criticised the earlier Roma policies for affirming assimilation and exclusion. At the end of the 1960s, international and national Roma politics came to emphasise minority rights and later human rights as well (Pulma 2005;Friman-Korpela 2012;Tervonen 2012;Friman-Korpela 2014).
Nowadays, the state has organised permanent structures for Roma affairs such as the Finnish Advisory Board on Romani Issues (RONK), which promotes co-operation between officials and Roma communities. RONK has been commended internationally and acknowledged as being a good model in organising Roma affairs (Friman-Korpela 2012). Additionally, there are also local advisory boards on Romani issues. There is a Roma Education Group in the Finnish National Agency for Education (FNAE), which promotes the education of the Roma. Since the 1990s, Finland has been lauded within Europe as a state that effectively protects the human rights of the Roma minority (Friman-Korpela 2014).
Improvement of education is one of the main aims of Finnish national Roma policy (MSAH 2009(MSAH , 2018. It has been estimated that in the late 1960s or early 1970s the Roma started to attend school regularly for the first time (Syrjä and Valtakari 2008, 44-45). However, two national surveys show that the school performance of Roma pupils differs from the national average. Pupils with a Roma background tend to more often learn in segregated educational environments, such as special education classes or home schooling, than the average Finnish pupil (FNBE 2004;Rajala et al. 2011). Additionally, drop out among Roma youth, and the number of pupils not applying to secondary education is higher than average among the Roma (Rajala et al. 2011;75-81; see also Niemi, Mietola, and Helakorpi 2010;84-86). Prejudice, racism and bullying of Roma pupils have been found to be common in Finnish schools (Junkala and Tawah 2009;Rajala et al. 2011;41-46). Problems in education are often seen as leading to problems in the labour market. 4

The ethnographic study
The present study is based on ethnographic research conducted in four municipalities in various parts of Finland. Jenni, the first author of the article, spent 16 days with Roma mediators; approximately one hundred hours in the field. Additionally, two mediators were interviewed without observing their work. The participants were initially located through the networks of the FNAE Roma education group. The FNAE Roma education group has been active in municipalities in networking on Roma education issues. Municipalities which were known as being active on Roma issues were approached to find out whether there was a Roma mediator working there. Jenni sent the mediators e-mails explaining her project and asking whether that person would be interested in participating in some way in the research. The details concerning possible observation and interviews were decided with the participants. When the person had decided to participate, Jenni contacted the school principal and the municipality to obtain research permissions.
During participant observation, Jenni accompanied the Roma mediators for the whole working day. In addition, interviews were scheduled either before or after the school day. The participants were asked about the events of the day and their work in general. The days spent together made it possible to hear about the ways in which the mediators talked about their work and made sense of it. The discussions took place in various locations: in cars, cafes, school canteens, teachers' lounges, classes, school yards, libraries and meeting rooms. Hanging out together made it possible to ask the participants to interpret episodes that were experienced together (Heyl 2001). There were 14 h of recorded discussions and interviews and 150 notebook pages of handwritten notes.
How the work of Roma mediators was organised in schools varied. Some worked as general teaching assistants with all the children in their school, whereas some worked in all the schools in the municipality that had Roma pupils. Most of the participants also worked as Romani language teachers in their municipality. They organised clubs, camps and happenings for Roma pupils and families, as well as culture days, lectures and exhibitions about Roma culture for the school community, and even for all the teaching staff of the municipality.
The analysis process following the participant observations and interviews included three phases of reading the interviews and observation data. In the first phase, the data were read with the following questions in mind: 'when does Roma-ness become explicitly significant in their talk and in everyday life?'; 'what kinds of meanings are given to Roma-ness?'; 'when is Roma-ness not mentioned?' We used the term 'Roma-ness' as an umbrella term for the notions that are in the data explicitly connected with Roma, with Roma identity or with becoming identified as a Roma in the education or work of the mediators. Thus, we asked what kind of 'shape' 'Roma-ness' obtains and what kind of dimensions of 'Roma-ness' become significant in the education and work of the Roma mediators. In the second phase, the extracts and episodes where Roma-ness had become significant (i.e. mentioned/taken up) were read and listened to again. Attention was paid to the issues associated with Roma-ness and to the tenor and nuances of the episodes and their talking. Attention was also devoted to contradictions within the talk of the participants, and to contradictions between what was said and what was observed. In the third phase, examples of the data were chosen to conceptualise, understand and represent the material analytically (Coffey and Atkinson 1996;Youdell 2006b). The concept of subjectification began to manifest itself throughout these analytic phases. Specifically, the analysis began highlighting those pieces of data in which the participants' talked about tolerance, indicating the strategies they employed at work in dealing with prejudice.
Becoming tolerable: a condition for successful working The accounts of many of the interviewees included statements that their Roma-ness made it possible for them to promote 'tolerance'. In the following example, Riku, who was the only Roma to graduate from his class in a teaching-assistant training course, was asked about the meaning of Romaness during the training 5 . He started by explaining the opportunity to talk about Roma culture and to promote tolerance in the teaching assistant training. He ended by reporting how the other students had been suspicious of him: Jenni: in the teaching assistant training did you feel that your Roma background had some special meaning? Riku: mm. . . Jenni: in the training kind of or. . . Riku: well quite a lot, like in the training period I had a chance to give to others from my own culture. So I kind of thought that the other students learned something about Roma culture. Jenni: yeah.
Riku: a lot. So I could share that kind of tolerance and then, like, promote it there. Jenni: yeah.
Riku: and in the end I got the kind of feedback on the kind of barriers that were broken down. At the beginning some had been totally like 'we'll see'.
In Riku's account, tolerance was promoted by educating people about Roma-ness and Roma culture. This is in line with the policy expectations (Finnish National Board of Education [FNBE] 2004; Immonen-Oikkonen 2010; FNBE 2010). However, since Riku educated the other teaching assistants about Roma culture, he also described how the students had been suspicious of him at the beginning of the teaching-assistant training. Riku explained that the other students had reported afterwards that barriers had been removed. To succeed in promoting tolerance, Riku had to become tolerable himself. Riku's narrative showed him in the position of the 'possibly tolerated' and his co-students in the positions of 'possibly tolerating' actors. 'The possibly tolerating' actors created the conditions for tolerance in terms of 'we'll see' (cf. Goldberg 2004). Riku had to take his place under their suspicious gaze and he did not mention feelings of discomfort about this in his account. He viewed their perspective as mundane and commonplace. Thus, the excerpt shows how Riku's actions in the discursive terrain of tolerance and its promotion rendered him as a subject, and subjected him to the relations of power in his course programme. Similar to Riku, all the participants in the study talked about prejudices and attitudes towards the Roma and themselves as being self-evident, common remarks. For example, one participant explained that he had thanked a worker at the school, who had remarked that he was doing some good PR-work for the Roma. When the participants were asked what it was like to work in their school community, many of them stated that everything went well because people were accustomed to the Roma. Both the idea of doing some good PRwork for the Roma and the idea that the Roma are people that one needs to get accustomed to, rests on the assumption that the Roma have a dubious status. Many participants reported that to ensure that they would be recognised as good workers they had to do twice as much work as the otherssince they were Roma. Thus, the tolerance discourse seemed to function in such a way that becoming tolerable themselves turned into a condition for doing their work. They wanted their own presence to have the function of 'breaking down barriers' and promoting tolerance. To do so, they needed to seem tolerable so that they could become accepted as good workers.
The participants expressed the opinion that they understood the prejudiced individuals and could explain their acts. In the next example, Allan reported on an event where a teacher had (verbally) attacked him in front of a student. Allan had been working as a teaching assistant for several years and had completed his formal teaching assistant training a few years prior to the interview. During his earlier practical training, a teacher had launched into him very aggressively. Allan talked about the event with apparent anger and also described how he had argued with the teacher and left the classroom. Then he continued: Allan: The student was looking all astonished (gives a laugh) you know. That the teacher started when the student was there, to attack me there. Jenni: (clucks) Allan: Well, the teacher was a bit, a quite special character. Jenni: Yeah, phew.
In the end, Allan laughed and gave reasons for the teacher's behaviour by stating that he was 'a special character'. Thus, Allan finally explained the incident by referring to the teacher's personality. One can find similar anecdotes throughout the data, expressed in fairly similar waysa bit of laughter and an explanation at the end. The participants also described such individuals as 'fools' or 'weird', lacking education or feeling afraid. The 'perpetrators' were excused and stripped of responsibility for their discriminatory acts. They were described as being uninformed, weird or fearful. References to discrimination or racism, due to fear and ignorance, are prevalent in Finland, as well as other countries (see, e.g. Lentin 2008a;Souto 2011;Alemanji 2016). The policy solution is often to educate people about 'the other culture' (Riitaoja 2013;Helakorpi, Lappalainen, and Mietola 2018). Our informants also wanted to tackle discrimination through providing knowledge of Roma culture and Roma-ness: Mertsi: Plus of course we have an obligation to provide information about Roma culture. Knowledge, it removes fear when one becomes acquainted with the Roma culture. And then the truth is also that one person with a Roma background in one's circle of acquaintances is someone they can ask and who they dare to ask and with whom one can laugh and talk freely.
Mertsi saw the responsibility for dispelling prejudices as residing with the Roma themselves, who needed to inform others about Roma culture. When Mertsi was talking about tackling prejudices in his account, he stated that somebody with a Roma background needed to enter into the social circles of non-Roma, to be available as a kind of informant. Furthermore, when the study participants tried to explain why people might be prejudiced or might discriminate, they sometimes blamed other Roma: Mirjami: And I feel that it's good for everybody among the majority children as well that they learn to accept and get to know. There are many kinds of us in every culture and also in our culture there are people who don't know how to behave. But when they get to know me they find out that not everyone is the same. Jenni: Mm Mirjami: If they have come across a person who couldn't behave, then they hopefully get a better experience from me.

Mirjami's account makes a distinction between herself and poorly behaved
Roma. This reasoningattributing prejudice to people having negative experiences with the Romawas commonplace among the respondents. Mirjami also explained that when children got to know her, they would discover that not all Roma are the same; hence that not all Roma behave badly. To change people's attitudes, Mirjami indicated that she needed to act in a manner that would not be interpreted as 'bad behaviour'. This implies that when discrimination and prejudice can be attributed to fear and ignorance, the Roma are forced to take the initiative and the ultimate responsibility to fight against inequality. They need to explain Roma-ness and Roma culture; they should be available to non-Roma as friendly informants, and they need to appear to be tolerable. They are constantly at risk of reinforcing existing prejudices against the Roma if they are seen as acting in a manner that might not be considered 'good behaviour'. The tolerating agents, the non-Roma majority, are not expected to assume any responsibility for the inequalities. Furthermore, inequality becomes a question of individual Roma acting 'in the right way', not an issue of the structural, historically produced inequalities and racism that needs to be tackled and analysed politically.

Navigating and questioning: strategies to encounter prejudices
The participants talked about the strategies they employed to deal with the prejudice they faced in schools. We perceive these strategies as negotiations within the existing discourses of the schools. We identified three strategies that were described as being employed to manage everyday life in the schools: making sure one does not seem too different, parody and feigning naivety.

Finding commonality
In one of the interviews, the informant Armas described his work in a school where many had strong negative biases against the Roma. Armas worked in all the schools that had Roma pupils in the municipality. He only worked with Roma pupils and families. In the excerpt below, Armas talked about his encounters in a teachers' lounge, in a school with strong prejudices against Roma.
Jenni: Yeah, right, so how did it feel, for instance, to go to the teachers' lounge or to the teachers? Armas: Well I just went there and, you see, the world championships were going on at the time. And I knew, I've been doing sports. So it was the funniest thing. You see, it was funny that when we started to talk about a hockey match and then I talked about my own teams and I said it like this. And they got so excited that let's start betting [. . .] (laughs). Jenni: (laughs) Armas: In other words, they saw that this is not just that a Roma can't do it, understand anything else. How surprised were they by that? And this was probably it. Often here in [the locality] they are surprised that I know so much about general things; sport, Finnish history, teams, ice hockey. Jenni: yeah Armas: this is what feels nice to them. When I know how to talk about those things and I can name the teams. They feel like oh this is like us. This person knows the same kinds of things that we do. And this is the key. They think that this person is not from our own culture. This person doesn't have eyes and ears open, they think. And this has worked also in [one other locality]. That I jumped in the middle of them and I went along with their world. And they saw that we also have the same kind of world.
Throughout the excerpt, Armas can be seen acting in-line with mainstream attitudes and interests, constructed through mutual interest in sport and particularly the favourite national sport, ice hockey. He claimed that it is important that the school staff understand that he and other Roma are part of the same culture and that they shared 'the same kind of world'. In the interview, Armas saw it as a problem that the non-Roma staff did not think he shared the same culture, a difference that needed to be overcome in order for him to get along with them. One of the participants explained that it was eye opening for the teachers to realise that Roma were not 'that special' and 'mystical'. She mentioned that it was important for the staff to realise that her life was not strikingly different. Comments like 'do I cook potatoes or pasta today?' or 'I should buy my children shoes for the winter' resonated among the non-Roma. One of the participants mentioned that he told school staff that Roma do things that are 'completely normal'. Thus, one strategy in engaging with bias in the everyday life of the school was to make sure that you could come across as not being too different.

Parody
In the interviews and observations, the interviewees repeatedly parodied Roma-ness. One example of this is when Jenni asked Roope, a general teaching assistant, who worked with all the children in the school, why one of the teachers told her that Roope referred to himself as an exceptional Roma, since he was employed. The teacher had told Jenni that Roope was a great worker and an important male role model. One stereotypical representation of Roma in Finland is that they do not work, and Roope played with that stereotype. He explained that 'these kinds of jokes' worked as good 'ice-breakers'. Roope's joke was multi-layered, since while it focused on, and racialised Roma, at the same time the joke 'laughed back' at the non-Roma who took this stereotype seriously. One other participant said that he used 'Roma humour' to show that 'we're not straight-laced'. The use of parody and humour were described as ways to get along with others in the school community. Thus, such humour and parody were communication strategies with non-Roma so that Roma did not appear too 'straight laced' or that the 'ice could be broken'.

Feigning naivety
The ability to ignore prejudice was also described as an important qualification for a Roma mediator. In one of the interviews, the informant Ramona talked about a municipality that she worked for. While there had been previous attempts to promote the equality of Roma in education, they had failed before Ramona started working. Ramona was asked why this was so. She explained that she refused to give up, even though she often hit a brick wall. One needs to move beyond the prejudices and negative attitudes to succeed. Ramona's interpretation was that the teaching assistants before her failed because they could not get beyond the prejudices. Ramona's account illustrates how confrontations regarding 'race'/ethnicity were difficult in their work. It seems that teaching assistants needed to stay positive at all times (cf. Ahmed 2009Ahmed , 2012.

Concluding remarks: becoming subjects in a challenging landscape
This article focuses on the dynamics of tolerance and prejudice in the everyday work of Roma mediators in Finnish schools. This ethnographic study shows how promoting tolerance involves a process in which the Roma school mediators are rendered as subjects and are subjected to power relations (see e.g. Butler 1997;Youdell 2006a;Phoenix 2009). The analysis suggests that at present Roma mediators need to 'become tolerable' in uneven power relations to succeed. It is the 'tolerating actors' who decide what is acceptable and the Roma mediators 'have to demonstrate that they deserve to be tolerated' (cf. Goldberg 2004, 38). The mediators sometimes viewed existing prejudices and lack of tolerance as the result of ignorance or negative experiences, caused by the Roma themselves. Also, discrimination and racism were seen as the result of fear and ignorance among the non-Roma. The Roma mediators aimed to solve this by imparting knowledge about the Roma and by trying to dispel prejudices by being good representatives of Roma. Tolerance promotion was expected to happen in a positive and upbeat, even humorous, climate. Open confrontations regarding 'race/ethnicity' were deemed inappropriate (cf. Ahmed 2009Ahmed , 2012. Roma mediators in the study encountered prejudices by actively negotiating within the discourses available to them. They had different strategies to navigate these discourses such as finding commonality, parodying Roma-ness and feigning naïvety. The theory of subjectification suggests that the agency of the subject emerges in the junctures when discourses are renewed (Butler 1995, 135-136). The capacity of mediators to submit to, and master the discourses, enables them to have an impact in the school and allows them to function successfully as mediators.
Tolerance is typically claimed to constitute a political polarity for racism and bias (Hage 2000). Nonetheless, the concept of tolerance and the discourse surrounding the concept has led to considerable critical reflection. As described at the outset, European multiculturalism and minority politics have been criticised for lacking any analysis of the production of difference and racism. One of these critiques has been aimed at the tolerance discourse. This discourse has been criticised for facilitating and being facilitated by depoliticising processes that constructs inequalities and racism as personal, individual and natural occurrences rather than inviting a political analysis and moving towards political solutions (Alemanji 2016;Goldberg 2006;Brown 2006;Lentin 2008aLentin , 2008bLentin and Titley 2012).
Educating Roma mediators is seen as an important tool to tackle discrimination against the Roma. It is expected that their presence in schools and their knowledge about the Roma will lead to a decrease in prejudice and negative attitudes towards the Roma (e.g. FNBE 2010). Even though bullying and negative attitudes against the Roma in education have been recognised, the measures used to tackle inequalities are mostly about guiding Roma parents and helping Roma pupils with homework and planning their educational paths (Helakorpi, Lappalainen, and Mietola 2018). Increasing the readiness of the teachers to support Roma pupils, supporting the teaching of the Roma language and producing materials for students and teachers about the history, culture and present situations of the Finnish Roma are also promoted (such as in MSAH 2009MSAH , 2014MSAH , 2018. The pervasive nature of prejudice towards the Roma has been highlighted in multiple reports, as is the expectation that knowledge about Roma culture and individuals with a Roma background working in schools will lead to less prejudice. Roma mediators are expected to promote tolerance in their work by being in contact with the school community and by providing knowledge about the Roma (FNBE 2010). However, our analysis suggests that the discursive playing field places obstacles in the way of truly addressing inequalities, discrimination and racism in educational contexts. The responsibility for tackling discrimination is placed on the Roma themselvesnot on the whole school community. The mediator practice would greatly profit from an analysis of how power relations function in schools, as well as analysis of how racism and difference is produced in schools. Otherwise, there is a significant risk that the current discriminatory structures and various political processes propelling them will become invisible and will be reduced to matters of individual attitudes and relations. Notes 1. The research literature also uses 'subjectivation' or 'subjection'. 2. The number is an estimate. In Finland, there are no statistics based on ethnicity.
Additionally, there are estimated to be about 3000 Finnish Roma in Sweden. 3. Nowadays, it is called the Suomen romaniyhdistys (Finnish Roma Association). 4. A national report on the employment of Roma was conducted in 2008. It was estimated then that about 20-30% of Roma adults were unemployed (Syrjä and Valtakari 2008, 130), when the overall number of unemployment in Finland among 15-64-year-olds was 6.4% (Statistics Finland 2008). 5. The names of the interviewees have been changed. The small number of teaching assistants with a Roma background has meant that all the information that is not viable for the analysis has been changed.

Notes on contributors
Jenni Helakorpi is a doctoral student in the Nordic Centre of Excellence Justice through Education, at the Faculty of Educational Sciences, University of Helsinki. In her PhD-project, she studies the policies and practices which aim to promote the basic education of Roma and Traveller minorities in Finland, Norway and Sweden.
Sirpa Lappalainen is an Associate Professor in sociology at University of Eastern Finland. Her research interests are in sociology of education, cultural and feminist studies in education and ethnographic methodologies.
Fritjof Sahlström is a professor in education at Åbo Akademi University in Finland. His research has focused on the organisation of interaction in educational settings, on developing ways of documenting and analysing classroom interaction relying on Conversation Analysis, and on the relationships between interaction, participation and social reproduction.