‘We value your food but not your language’: Education systems and nation-building processes in Flanders

Abstract

Education systems are crucial social and cultural apparatuses. They are designed to homogenize at least to a large extent the discourses and praxis of the citizens of a nation by channelling them as much as possible through a unified educational system. However, in ethnically and culturally diversified societies, these homogenizing social institutions can become counterproductive as they are primarily designed by and for the dominant ethnic group. This issue is particularly important in nation-building processes, something that is still explicitly ongoing in Flanders (Belgium), as these institutions put forward a unified nation with one culture and language. In this article we study how these macro-level processes can be related to the interaction processes in and between families and schools. The central research question this article tries to answer is: is the idea of a culturally and linguistically unified representation of the Flemish nation relevant when studying teacher-pupil-parent relations? We aim to build upon the existing literature by relating meso-level processes of institutional racialization and culturalization to micro-level interactions in educational settings and to macro-level processes of nation-building in Flanders. To answer these questions, we rely upon data from two qualitative research projects: (1) one study that focuses on the family socialization processes in ethnic majority and minority families and collected 42 interviews with parents; and (2) a second study that focuses on the educational trajectories of students, with specific attention on the school environment, and collected 114 interviews with students and 57 interviews with school actors. The findings show clear similarities between the macro-level processes and the interactions between families and schools in such a way that the article discusses if ongoing and explicit nation-building processes are capable of appreciating the existing diversity in a nation.

Introduction: The role of education in nation-building processes

Education systems are crucial social and cultural apparatuses, especially in nation-building processes (Gellner, 1983). They are designed to homogenize at least to a large extent the discourses and praxis of the citizens of a nation by channelling them as much as possible through a unified educational system, from their early years onwards (Eriksen, 2010; Hannerz, 1992). This is done primarily to create a level of certainty and predictability in the interaction processes between members of the same ‘imagined community’ and was deemed necessary following the uprooting of many individuals during the period of industrialization and also afterwards in times of social upheaval (Anderson, 1991). Educational systems are perceived as primary structural sites for fostering social cohesion, unity and homogenization. However, they come with a toll and are contested social contexts (Levinson and Pollock, 2011). Constructing a national community (or any other kind of social group) implies a simplification of social reality and the categorization of individuals into in- and outgroups (Jenkins, 2008; Turner and Reynolds, 2001). Moreover, what is crucial in these processes and the construction of cultural apparatuses such as the educational system, is that they are designed and implemented by those of the ruling class (Bourdieu and Passeron, 1977; Hannerz, 1992). As many scholars have shown, crucial in nation-building processes is often the construction of (the imagination of) ‘one language and culture’ that binds all (Anderson, 1991; Eriksen, 2010; Gellner, 1983). This language and culture is to a large extent transmitted through nationally established educational systems, and they reflect primarily the linguistic and cultural background of the dominant ethnic group (Wimmer, 2006). Therefore, and for the purposes of this article, we refer to this ‘ruling elite’ managing and controlling education, as the dominant ethnic majority.

The crucial role this ethnic majority plays in these processes is certainly true for the Flemish educational system – the focus of this article – that was established by the northern and Dutch speaking part of Belgium. Education became a full-blown Flemish community competence in 1988 when it was transferred from the federal Belgian level to the Flemish regional level. This was done based upon the (implicit and explicit) idea that the Flemish and Walloon communities of Belgium are fundamentally different when it comes to the language and the culture they want to transmit in their educational system. In this process, education is represented as a person- and culture-related competence that is specifically based upon the needs and goals of the people of the Flemish community. As such, this has already highlighted the specific role of the education system in the representation of the ‘Flemish nation’. This is particularly so at a time when the discussion of Flemish independence has reached a new height with the construction of a federal Belgian and a regional Flemish government wherein the separatist N-VA, the New-Flemish Alliance, is the largest political party on both levels. One of the consequences of these Flemish nation-building processes is the writing of a ‘Charter for Flanders’, by the Flemish government in 2012, a more symbolic constitution which cannot legally or judicially be ratified by the Flemish community and its representatives as they do not have that competence. ¹ Nevertheless, Flanders, its institutions and population are described in this 41-page document, and its preamble offers some interesting insights into how Flanders is perceived by its ‘ruling elite’. On page one it is argued that:

Flanders has developed itself through a long historical process and due to the efforts of different generations, to a State of the Belgian Kingdom and to a part of the European Union. It is a democratic and social constitutional state and forms a nation with its own language and culture (Vlaams Parlement 2012: 1).

Despite the development of this document at a time when Flanders is rapidly diversifying and many of its larger cities such as Antwerp, Ghent, Mechelen and Genk are, or will be in the near future, majority-minority cities (Kasinitz et al., 2002), the Flemish language and culture are put forward as central to the Flemish nation. The Charter does not deny the existing ethnic, linguistic, religious and cultural diversity; on the contrary, it proclaims that all languages and cultures are welcome. However, the cultural and linguistic foundation of the Flemish nation remains unified in one specific language and culture and is based upon the work and cultural transmission of many generations of Flemish. In times of social unrest with respect to, e.g. the consequences of migration and diversity, nationalist claims receive more support (Eriksen, 2010). Indeed, in several nations in the past decade similar counter-reactions have emerged, declaring multiculturalism as problematic or even a failure, and national identities are put forward as a solution to this ‘anomaly’ (Alba and Foner, 2015; Lates, 2011). Recent events of the so-called radicalization of ‘Muslim youth’ or the large influx of Syrian refugees in Europe have generated similar reactions and discourses in Flanders, and also across Europe. (Imaginary) lines are drawn of who is with us and who is not; who belongs to the nation, and who does not; who adheres to ‘our’ values and culture and who does not. Consequently, these processes are at the same time inclusive as well as exclusive: All that want to adhere to or assimilate into the dominant group are welcome, but the nation is ultimately founded upon one language and culture. Indeed, diversity management often seems to take the form of excluding those ethno-cultural elements represented as foreign and/or problematic (Wimmer, 2006). Nevertheless, the ‘Charter for Flanders’ was developed by leftwing as well as rightwing political parties and is represented as a document with broad political support. As such, this document can be perceived as the overall basis of Flemish political policy and consequently also as the background for designing and implementing educational policy, one of Flanders’ most important social institutions, of which it has almost full control. In this article I try to relate these broader social processes of a seemingly unified Flemish nation with more meso-level and micro-level phenomena occurring in everyday classrooms and families in majority-minority cities.

The central research question this article tries to answer is: does the idea of a culturally and linguistically unified representation of the Flemish nation have relevance when studying teacher-pupil-parent relations? Related to this, the following subquestions are studied: which narratives emerge in these cities, schools and families with a very ethno-culturally diverse population, when discussing the educational situation of minority students? Is the centrality of Flemish culture and language stressed or are minority languages and cultural elements also valued parts of these narratives? It aims to argue that in a societal context with explicit ongoing nation-building processes, as is the case in Flanders, an educational system will probably focus more on cultural and linguistic homogenization than on explicitly situating diversity as one if its defining features. In doing so, I want to argue that the micro-level interactions between pupils, parents and school staff can be seen as reflections and elaborations of higher level processes – the meso-level of the school and the macro-level of the structural, socio-economic inequalities in broader society – as discussed by institutional racism theory (Phillips, 2011).

In the final section we discuss the potential implications of a highly culturalized educational system for the educational position of minorities and the role of ‘nationalist agendas’.

The situation and structure of the educational field in Flanders

One way of presenting the importance of education in Flanders is by taking a look at the budget accorded to education by the Flemish government. In comparison to other governmental agencies, education receives the highest budget of all domains for which Flanders has the political competence (Turtelboom, 2014). Moreover, research shows the Flemish educational system is one of the most highly trusted social institutions in Flanders: 76% of the ‘Flemings’ had a lot of trust in education (Studiedienst van de Vlaamse Regering, 2012: 35). Indeed, Flanders seems to deserve this huge financial input and social trust since in international comparative educational research such as the PISA-assessments, Flanders finds itself at the top in comparison to similar countries or regions when it comes to performance in mathematics, reading and science subjects (De Meyer and Warlop, 2010). However, what the PISA data also reveal, and what is substantiated by various other research findings, is that strong social inequalities are present in the Flemish education system (Clycq et al., 2014; Danhier et al., 2014; Van Landeghem et al., 2013). These inequalities are primarily related to so-called background variables of students: their socio-economic status, their immigration background and their home language. The meritocratic ideal of education, already argued by Plato as essential for a democratic society (Gordon and Lawton, 2004), is far away. Grade retention rates are much higher for ethnic minority than for ethnic majority students – this is already the case in primary education and the gap only increases in secondary education (Vanduynslager et al., 2013). Based upon nationality: 14% of the Belgian students and 42% of the foreign students have at least one year of grade retention. Based upon home language, the rates are respectively 12.5% for those with Dutch as their home language versus 35% with another home language. In secondary education, this grade retention gap only increases: 27.5% of Belgian national students experience grade retention and 65.5% of foreign national students (based upon home language, the rates are 26% versus 60%). Given this overview on grade retention and the strong differences between social groups, it is to be expected that PISA findings come to similar conclusions. While Flanders is at the top with respect to the maths performance of its students (for language and science, similar results are found), it also displays the highest gap between the best and the worst performing pupils (Jacobs et al., 2009). Moreover, Flanders is also at the top with respect to the variance of performance explained by the socio-economic status of the students: almost 20% is attributed to the SES of students (Danhier et al., 2014). Thus, it is to be expected that specific students such as those with an immigration background, those with a home language other than Dutch and those with a lower socio-economic status are overrepresented in early school leaving rates. To put it into numbers: in Flanders, 40% of the early school leavers (those leaving education without any upper secondary education, i.e. ISCED 3 qualification) are boys with a lower educated mother, who do not have Dutch as a home language and who have an immigration background while this is the case for only 4% of girls with a higher educated mother, Dutch as a home language and without an immigration background (Van Landeghem et al., 2013).

It is important to contextualize these social and ethnic stratification processes within the broader structures of the Flemish educational structure. Individuals’ opportunities are often hampered or increased by these structures and the resources individuals can allocate in their decision-making processes. As the numbers above have shown, strong socio-ethnic inequalities are already present in primary education. However, the structural context of secondary education only seems to widen this gap. Although represented as comprehensive in the first two years of secondary education (when students are around 12–13 years old), categorization of youngsters into the A-track and the B-track is already occurring. This is important, considering that the most vulnerable pupils and those with grade retention in primary education tend to be sorted into the B-track, a track that almost inevitably leads them to the (part-time) vocational track. From the second grade of secondary education (when they are commonly 14–15 years old), students need to choose between the general (leading to tertiary education), technical (in the best case leading to professional bachelor) or the vocational track (leading to the labour market). As is often the case, the general track is perceived and represented as the most prestigious track and the vocational track is located at the bottom of this hierarchy (Van Houtte et al., 2012). Moreover, and given this hierarchy, the Flemish educational system is also known for its frequent downstreaming and the absence of upstreaming students (Stevens and Vermeersch, 2010). Once a student ends up in the vocational track they can almost never climb up this hierarchy and change tracks. This segregation between tracks is at the same time an SES and ethnic segregation that also becomes apparent in the segregation between schools, as many of them find themselves part of a ‘marketization’ of school choice and try to adjust their ‘public image’ accordingly (Mampaey and Zanoni, 2014). At this point it becomes clear how the school and track choices of parents reproduce and increase these segregation processes, given the absolute freedom of parents to enrol their child in the school of their choice (Nouwen and Mahieu, 2012). These findings show how broader social inequalities are influencing educational processes in Flanders. Strong socio-economic and ethnic inequalities endanger the opportunities of those who are not perceived as part of the dominant, Flemish and middle class majority. However, most individuals in society find success in education to be the necessary condition for success in the labour market and in family life (Clycq, Nouwen and Vandenbourcke, 2014). When large subgroups in society remain excluded from this success, social cohesion can be compromised. This brings us to the social theories relating educational processes to the broader concepts of power and identity.

Theorizing home-school congruence and the role of institutional racism

Bourdieu and Passeron (1977) were among the most important scholars pointing out the importance of one’s disposition to perceive and act in social reality, and in particular in the school context. Via, among other important concepts, the concept of habitus, they theorized that the socialization and internalization processes occurring in the family context are constitutive for children’s lives (Jenkins, 2008). However, these family socialization processes have their specific aims often based upon the idea of raising children in such a way that family continuity is ensured (Clycq, 2014; Hughes et al., 2006). Socialization processes occurring in schools often have different aims, as they focus primarily on broader society, are state imposed and are generally designed largely or solely by the dominant ethnic (majority) group (Reay, 2010). National education systems are set up based on the idea to provide an education for youngsters that is complementary and supplementary to the upbringing in the average, middle class, national majority family. It aims to provide education in the ‘language and culture’ of this dominant group in society. This is done to construct an ‘identity safe’ environment for children and parents, and schools become a reflection of the home environment and vice versa. Therefore, often the family and the school socialization processes are relatively congruent and overlap to a large extent. However, this is much less the case for families with a lower SES and/or immigration background and thus there can exist a lot of ‘noise’ in this relationship, and congruence can turn into incongruence and a disconnection. As national education systems are focused primarily on the nationally dominant ethnic and cultural group, by definition the different cultural features – especially language and religion – of minority groups will not be focused on, or only marginally.

Thus, Bourdieu and Passeron (1977) and various others, such as Freire (1970), Giroux (1980), Gillborn (2006), have argued for a long time that more often than not educational systems are most beneficial for those students and families with a dominant ethnic and/or middle or upper class background. Students and families not fitting into the system are often portrayed in deficit terms and are presumed to be lacking the right attitude, values and behaviours (Alba and Holdaway, 2013). They become stigmatized by educators – but also by society in general – as difficult students and are ascribed negative learner identities (Youdell, 2003). In recent decades, these ‘deficits’ have increasingly been described in cultural terms and captured by concepts such as subtractive schooling or subtractive acculturation (Gibson, 1988; Roosens, 1995; Valenzuela, 1999). Specific cultural traits such as language and religion are represented as problematic and an obstacle in the educational success of minorities (Valencia, 2010). At this point, the power differences between social groups become salient as the dominant group has more power and resources to impose their representation of social reality (Bourdieu, 1990; Lukes, 2005).

Equally important is that the dominant groups in society can impose this representation through the channels of society’s main institutions (Sullivan, 2006). Thus, not only do everyday practices of individuals reproduce these imaginations, boundaries and inequalities (Garner, 2007; Lewis, 2004), but also institutions are crucial means to imposing these representations (Bourdieu, 1991). In particular, educational systems reproduce a certain image of what are valuable (future) citizens and what are valuable cultural traits and competencies in society (Lukes, 2005; Yuval-Davis, 2010). So specific ethnic and cultural biases are inbuilt educational systems that are seen to a large extent as promoting a specific cultural outlook. Some therefore discuss this institutional bias as institutional racism or racialization (Gillborn, 2005; Phillips, 2011). However, in the case of Flanders, it seems more accurate to talk about institutional culturalization rather than racism. Ethnic identities are generally welcomed, appreciated and perceived as more or less unproblematic, but it is their respective cultural traits that are identified as problematic. Elsewhere I have discussed this – in line with various other scholars such as Bonilla-Silva (2003) and Lamont and Mizrachi (2012) – as the continuation of othering and stereotyping while at the same avoiding the stigmatization of ‘being racist’, as others’ ethnic identities are no longer problematized, but rather their cultural, religious or linguistic identities (Clycq, 2015)

Institutional culturalization in education policy and in schools

In the case of Flanders as a society, and its educational and school policies in particular, one can certainly argue that there is a subtractive approach towards specific low-valued languages from socio-economic subordinate minorities and migrant groups in particular. Different schools sanction the use of the home or native language on the school premises (Clycq, 2014; Stevens, 2008; Van Praag et al., 2016). This repressive approach resembles the dominant perspective in broader Flemish society and in particular in the context of education and in the labour market. Some policy makers even want to restrict access to social housing based upon one’s willingness or even proficiency in Dutch, the language of the dominant majority and protected in the Belgian constitution (De Standaard, 2014). Dutch is represented as the most important tool for newcomers (and long-term residents) to become successful in society. This view is sustained by many policy makers from the different political parties. One of the ministers of education – traditionally from more leftwing political parties – even argued that multilingualism (interpreted as having a different and less valuable home language than Dutch and placing too much emphasis on it) leads to ‘zerolingualism’ (Blommaert and Van Avermaet, 2008). These statements find a strong support base among educational actors. A recent survey among around 700 Flemish teachers on the issues of diversity showed that 78% of them (fully) agreed with the statement that the most important explanation for the educational difficulties of students with another home language is their lack of proficiency in Dutch (Clycq et al., 2014).

Studies have shown that these discourses and practices on identity, language and culture can be found at the micro-level of student-teacher interactions, as well as at the meso-level of school and macro-level of educational policy and structural inequalities (Pulinx et al., 2015). It has implications for the trust teachers have in students and their ability to be able to teach them (Agirdag et al., 2012) and undermines the upward social mobility and meritocratic agenda of schools (Clycq, Nouwen and Vandenbourcke, 2014). These processes are influenced by the marketization of education and the pressure schools feel when competing with other schools for the best students (often ethnic majority and/or higher SES students) to become a top-ranked school (Mampaey and Zanoni, 2014). However, at the same time, schools and individual educators are wary of being accused of racism and discrimination (Stevens, 2010). They design strategies to show that the problems are not located within the school, and certainly not with the teacher or educational policy, but can be located at the level of the student (Clycq, Nouwen and Vandenbourcke, 2014; Pulinx et al., 2015; Stevens, 2010). As argued before, this is a strategy to avoid the stigma of racism as supposedly neutral deficits are focused upon – limited Dutch proficiency, lack of effort and motivation, making the wrong choices, lack of parental support – rather than speaking of the ‘unintelligent ethnic other’ (Clycq, Nouwen and Vandenbourcke, 2014).

However, not all home languages are considered problematic by the dominant majority. Economic rationales often seem to fuel this categorization of ‘good’ and ‘bad’ multilingualism. One of the previous ministers of education stated that as a society Flanders should value its multilingualism and think about incorporating, e.g. the main languages of the BRIC-countries (Brazil, Russia, India and China), as economic opportunities can open up in the future given their ‘booming economies’ (Smet, 2011). This is a poignant example of additive schooling processes aimed at incorporating a valuable cultural trait – in this case specific home languages – and thereby making the statement that they are useful and relevant for students and for society in general.

Again, this reveals how the dominant group can impose what is good and valuable to pursue, and what is not. When the key positions in society are indeed occupied by a specific ethno-cultural, SES and gender group (Flemish native, Dutch speaking, middle class males), then they have more power to steer the representation of social reality than subordinate groups. As a consequence, they have more power to problematize and pathologize those goals, strategies and practices that they feel are good as well as those that are not. In this article, we try to relate the more macro-level processes of nation-building processes discussed above to the discourses and practices emerging in everyday classrooms and minority families in Flanders. We study which issues and representations of social reality emerge in the narratives on education and diversity, and relate these to the discourses on the Flemish nation. We aim to argue that the discourses and practices to be found in schools and families should be contextualized even more broadly into the dominant imaginations of Flanders. However, before starting with this analysis, we elaborate on the different methods and samples used for this article.

Research method and sample

The discourses of individual parents, students and school actors are central in this article and we try ‘to make sense of, or interpret, phenomena in terms of the meanings people bring to them’ (Denzin and Lincoln, 2005: 3). To understand how people interpret certain phenomena and give meaning to their daily life world, a qualitative research methodology is relevant (Maso and Smaling, 1998). As mentioned, this article is based upon two research projects.

The first project focuses on the socialization narratives as constructed by 42 parents of Belgian-Flemish, Moroccan and Italian origin (from the cities of Antwerp, Genk and Leuven). Semi-structured in-depth interviews were conducted by the researcher, guided by a general topic list containing questions related to identity construction, family life and intergenerational relations. The introductory question was always ‘Can you give me an overview of your family’s history starting from the family you were born in up until your own family now, with special attention to your family composition, residence status, and other life events you find important?’. This proved to be an interesting question (of course, sometimes more subquestions were necessary) and the information collected often gave a good overview of events important to these parents, which could be elaborated on further in the interview. A main interview topic was the idea of what parents felt necessary to transmit to their children, primarily taking into account their ethnic and cultural as well as their gender differences and similarities (for more information, see Clycq, 2012, 2014). The research data are obtained via 27 individual in-depth interviews with fathers and mothers of Belgian-Flemish and Italian origin. Moreover, in most cases, it was possible to interview both parents from the same family, although every respondent was interviewed individually and separately. The respondents have diverse backgrounds, but they all have at least one child and identify themselves as ‘having’ one of the three abovementioned ethnicities. To recruit these respondents, different methods were applied. Public family assistance services and specific organizations were contacted and, in combination with the method of snowball sampling, this allowed for the recruitment of a diverse research sample. In total, 42 respondents were interviewed: 24 women and 18 men. With respect to their ethnic background, 14 Belgian-Flemish individuals (5 men and 9 women), 13 Italians (7 men and 6 women) and 15 Moroccans (6 men and 9 women) participated in the research. Of these respondents, 15 were in a partner relationship while the others were married. Concerning family composition (older and/or younger children), educational background and employment status, variation was pursued and obtained. The Italian and Moroccan parents were mostly born in Belgium, or migrated to Belgium before their children were born. In each of the categories – gender and/or ethnicity – there were respondents who were higher or lower educated, some worked as labourers, others as clerks, teachers, researchers or social workers. The unemployed women were housewives and the two unemployed men were looking for a job. Yet, in this sample, Belgian-Flemish respondents were generally higher educated than Italian and Moroccan respondents. This sample configuration made it possible to study the research topics from a diversity of perspectives. All interviews were executed and transcribed by the researcher, and afterwards coded via a code scheme in Atlas-ti. Codes schemes were printed out and emergent patterns were analysed in more depth. This also allowed for linking seemingly unrelated elements, which encouraged the researcher to look at the data from a new angle.

These insights fuelled the findings of the second research project, which studies primarily the school environment and the educational trajectories of students with an immigrant background. In the latter project, I was the academic coordinator and as such was also involved in the analyses and reporting of the findings. To gain sufficient descriptive and in-depth data on these trajectories, different methodologies were applied. However, in this article, I focus mainly on the qualitative fieldwork executed in nine secondary education schools. The selected schools are situated in one of three Flemish cities: Antwerp, Ghent and Genk, all of which are very ethnically and culturally diverse. We conducted a survey in the school in these three cities, resulting in a database composed of 11,015 unique non-blank records, representing approximately 50% of the foreseen population (3rd and 4th year of secondary education). In this article, only a small amount of the merely descriptive quantitative data are used, as our main argument is built upon the qualitative interviews. In our ethnographic fieldwork, we followed 114 pupils for 2 years and also studied the discourses of 80 teachers, principals and other school actors. For the ethnographic fieldwork, we selected a diversity of schools with respect to their governing body, curriculum, availability of tracks and student composition. In the first half of the second year of the project, we asked a group of around 30 students from diverse backgrounds in each of the 9 schools to come to an information session wherein we explained the purpose of the project and what we planned to do in the next 2 years. Out of this large group, we selected a smaller group of students per school in such a way that a balance was found within and across the schools. Based upon their track and grades, we selected students that were ‘successful’ or not, across ethnic groups and SES backgrounds. The ethnographic fieldwork delivered a wide range of complementary data, all related to the direct environment of at least one of the focus students. To be able to comprehensively address these qualitative data, interviews were fully transcribed, coded and analysed using NVIVO9 software. In the course of this four-year project, around eight researchers (a mix of PhD and junior researchers) were involved and each used the same basic research methodology, topic lists and coding schemes enabling comparison across research sites (the various schools, classes and families) while at the same time leaving room for specific themes to emerge from the data. While a very basic intercoder reliability was executed only a few times (coding the same interviews by different coders), the coding differences were discussed in group manually rather than calculating this difference. As a solution, more elaborate ‘node descriptions’ were designed to harmonize researchers’ interpretations of the different nodes. During this four-year project and two-year ethnographic fieldwork many topics were discussed with the participants. The general underlying questions focused on the discourses of educational success and failure aiming to grasp how ‘educational performance’ is explained. This led us to new topics related to (learner) identity, minority languages, support networks and structural inequalities.

Minority families: Hoping for school support but experiencing exclusion

To start with, a dominant recurring narrative in the discourses of minority families from both research projects – and thus in Chinese, Italian, Moroccan, Polish and Turkish origin families – is the issue of interethnic and intercultural continuity. Parents find it of utmost importance to pass on their home language, their religious affiliation and other important cultural elements. They see it as one of their fundamental parental responsibilities, since mainstream Flemish schools do not take up this role. In all schools, Dutch is the dominant instruction and communication language, and Christian and other cultural elements are omnipresent in these schools. Ethnic majority parents feel comfortable in these institutions insofar as they feel them reflecting their own socialization processes. This is not the case for specific minority languages and religions and, as a consequence, various minority parents feel obliged to focus within their family on the transmission of these elements:

When she [her daughter] was little, I always spoke Italian to her. Just so she could speak our language (Belgian-Italian origin mother, Genk).

Okay, he lives here but still; we are Polish (Belgian-Polish origin mother, Antwerp).

Parents feel this is necessary, not only seeing it as their parental duty but also to ensure intergenerational communication, and thus, ethno-cultural continuity. When, for a variety of reasons, this transmission of, e.g. home language does not happen, specific difficulties arise, as this parent testifies:

In the future I want to pay more attention to this, that they learn at least the basics of Italian. Because it is really important… when I see it now, with my mother, if… My mother only speaks very very little Dutch and then you notice that they [his mother and his children] only very difficultly communicate. And that is a pity, you know. That shouldn’t be… (Belgian-Italian origin father, Genk).

However, this process of balancing the family socialization goals with the more formal goals put forward in the educational system, is not always a stress-free endeavour, as the following parent clarifies. The socialization goals of Moroccan parents not only focus on the home language (similar to Italian or Polish parents) but also on the religious affiliation (similar to Turkish parents), as this is also a salient cultural difference with the dominant majority group.

With Christmas, they say in school ‘a Christmas tree, and Christmas this and Christmas that’, and then at our home [his children ask] ‘Why don’t we have a Christmas tree?’ And then you have to explain it again: ‘Soon we’ll have our Festival of Sacrifice, and then we’ll have our celebrations and new clothes and candy, …’ We try to emphasize these things […] I would be the first to send my children to an Islamic school because, as a parent, you are relieved from that overall responsibility of transmitting all these cultural things. And then you can just concentrate on raising your children, playing with them and doing all sorts of things because the rest is taken care of [in an Islamic school]. But not in our case. You always have to react to these things they take with them from school, like now with Sinterklaas [a traditional Belgian and Dutch festival for children] and Christmas. […] And then you always have to explain yourself. And again, you get frustrated – what should I do with this? Whereas in an Islamic school this would all be self-evident (Belgian-Moroccan origin father, Antwerp).

As this parent poignantly argues, schools can be a strong supportive institution when your ‘habitus’ corresponds with the habitus that is dominant in this field (Bourdieu, 1990). However, when this is not the case, this supportive institution can turn into an ‘oppressive’ institution. This becomes even clearer when the perspectives and discourses of teachers and other school actors, as protagonists of this dominant habitus in the school, are studied. Before coming to this, we should stress that what is implied in all these narratives is that learning and becoming familiar with the cultural elements of the dominant group (Dutch, Christianity and so on) is self-evident for these minority parents and something they do not try to avoid. What is at stake for them is the problematization and oppression of the cultural elements they themselves want to pass on.

Institutional culturalization in Flanders: A subtraction of unwanted cultural elements

The dominant narrative found among school actors is that students’ home language – when it is not Dutch and not a socially valued language such as French or English – is the primary factor for their supposed failure in education. Children in such families are represented as victims of their families and of parental socialization processes in particular.

At home, there is practically no education, leaving aside if they are even followed up [by their parents]. So the school schedules are rarely or never checked… Honestly, I think they’re all, quote unquote, ‘poor souls’… there are some abuses, believe me. It all comes down to what I call language deficiencies. There is no education. The boys don’t even stand a chance, they aren’t educated, they are deserted (native Belgian teacher, Genk).

What is that, a foreign language kid? My foreign language kids are kids that are born in Belgium, second-third generation, and they still can’t speak Dutch. […] no, you have to let them fall flat on their face, sorry, these people need to find this out themselves […] I’m really harsh no, but I’m really angry too […] for me that’s pathetic, I really feel pity for these kids. I’m really angry with these parents, you know! (native Belgian teacher – Antwerp).

School actors relate this representation of dysfunctional family environments and family socialization processes mainly to specific minority families, in particular those of Moroccan and Turkish origin (in Flanders often referred to as ‘Allochtones’).

When we have a student from parents of ‘allochthonous’ background, then that home culture is really different. There they very often say the school has to do what it wants, the school has to make sure that their son or daughter learns […] while you don’t have it like that with the native [Belgian-Flemish] population. They do check their children’s school diaries. I think you have it more with them, that culture of studying is more part of the Western cultures than with the ‘allochthonous’ communities (native Belgian teacher, Genk).

These families are even represented as being deficient in their parenting role, to the extent that teachers need to take up this role too.

There is no [parental] follow-up. We see it when we write something in the diaries [for the parents], there is no reply. Or we ask them to come to the parent meetings, but they don’t come. We almost need to be mother and father for these children (native Belgian teacher, Antwerp).

What is to be expected in educational settings where specific pupils with an immigrant background are represented as problematic and less teachable, is that school actors try to attract those students that seem more valuable or investment worthy.

But it is indeed so… that we try to attract the good pupils and, you know, also the whiter pupils. To guarantee the mix, you know. That is not said explicitly, but that is something we are working on (native Belgian principal, Genk).

However, notwithstanding this negative representation of linguistic and even cultural diversity, school actors do stress the importance of the recognition of this diversity. This creates a seemingly untenable straddle that nevertheless seems to be quite rock steady and non-contradictory in the representation constructed by the dominant group. The following section shows that in the Flemish educational system we see recurrent processes of institutional culturalization rather than of institutional racialization (Phillips, 2011).

Imposing a dominant cultural outlook: Valuing the ‘right’ and excluding the ‘wrong’ kind of diversity

How can ethno-cultural diversity in Flemish schools be valued when school actors perceive this linguistic and even cultural diversity as the main reason for students’ educational failure? The narratives below show the importance of the power of school actors who represent social reality to fit their own perspectives. It shows how processes of institutional culturalization – imposing a dominant cultural outlook and pathologizing and subordinating ‘other’ ethno-cultural identities, outlooks and emblems – are reinforced by micro-level interactions and vice versa (Gillborn 2005; Youdell 2003). Certain meso-level school policy actions – excluding and problematizing minority languages – induce micro-level strategies that translate this exclusionary policy into, e.g. financial sanctioning of students.

As the following teacher explains, he really values (ethnic) diversity, as he feels it is important for minority students to feel support for their identity development; however, at the same time, he feels it necessary to penalize his students monetarily when they speak their home language in school. Moreover, speaking the home language is represented as negative, since racist or insulting language is fined with the same amount as one’s home language. Implied here is that racist language and home language are equally and similarly wrong.

Actually, here in school they are developing a language project. So, hanging up signs in the hallway ‘Here we speak Dutch’. I personally do this already for years. But in a playful manner. So, they have to pay 20 [euro] cent when they do not speak Dutch. When they speak Turkish, Moroccan. But also racist or insulting [language], then they have to pay. I ask them beforehand, if they agree with this. But this way you can stimulate your class in a playful manner to speak Dutch. And then we collect it in an envelope. And when we have enough [money] we go eat kebab [a typical Turkish dish] at lunch break. I feel it important that you go back to their uniqueness with kebab, Belgian-Flemish [students] have to eat fries [a typical Belgian dish] (native Belgian teacher, Genk).

Thus, a clear distinction is made between those cultural elements that are felt as important in one’s identity construction and for which a school can be supportive, and those elements that are perceived as problematic and an obstacle in a student’s educational trajectory. While parents find it of paramount importance to pass on their ‘mother tongue’, school actors predominantly feel this home language is precisely the main cause of all problems and needs to be excluded from the school environment. This belief is very deeply internalized by school actors, even to such an extent that research that indicates home language is not ‘by nature’ problematic and is not the only thing minority parents and students care about, is simply not believed.

When you see students from Turkish origin, when they speak Turkish at home really a lot, when they speak with their friends, at school, when they watch television [in Turkish]… If I’m not mistaken, there are studies saying the opposite, but that seems illogical to me. If you are constantly speaking in Turkish, Moroccan, or it may even be Swahili, if you speak that constantly and only speak Dutch between 8:30 am and 4 pm at school. Then there is no professor that can convince me that that has no influence on your Dutch proficiency. So, actually, that Dutch should be implemented in their home language, their home situation or whatever (native Belgian teacher, Genk).

These narratives show on the one hand that school actors do indeed accept the idea that diversity needs to be valued, but on the other that the specificity of students’ home language has now received such a negative status that it supersedes all other influential elements in one’s educational trajectory. As argued above, certain representations of social reality are so strong that even research findings indicating the opposite are simply not believed and cast aside. Moreover, the survey we conducted among 11,015 pupils, of whom 780 had a Turkish and 862 had a Moroccan immigration background, showed that the majority of the latter two groups indeed watch television in Dutch ‘at least often’ (for more information, see Clycq et al., 2014). An even stronger finding was that at least 80% of these youngsters said they speak Dutch ‘at least regularly’ with their peers and neighbours, while around 46% said they do this with their father or mother, and only 9% said they speak Dutch ‘at least regularly’ with their grandparents. This brings us back to our first findings, where minority parents stressed the importance of transmitting their ‘mother tongue’ to their children to make communication between grandchildren and grandparents possible. However, in the current Flemish educational system, the dominant narrative holds that the ‘mother tongue’ is the root of educational failure and the perspectives and goals of minority families are not appreciated as valuable and reasonable, also given the broader context of nation-building in Flanders.

Discussion: Institutional culturalization as an exponent of Flemish nation-building processes

What we tried to argue in this article is that the internalization of the cultural deficiency of minority languages, parenting practices and educational ambitions into the mindset of many native Belgian teachers cannot solely be explained by looking at the micro-level processes of, e.g. student-teacher interactions. The narrative that these cultural deficiencies are to be found in minority families has become so self-evident that one has to take into account the broader institutional context, at the meso-level of the schools but also at the macro-level of the structural features of society and educational systems in particular. One cannot overlook the importance of the macro-level processes of nation-building in Flanders and the importance of policy makers and educators at all levels to impose a specific dominant cultural outlook on society and its members. The findings show how self-evident and unquestioned certain racist and culturalist practices have become as they are part of a broader dominant public and policy discourse on the Flemish nation and its core institutions.

Therefore, we started this article with a discussion of the role of education systems in nation-building processes (Anderson, 1991; Gellner, 1983). Channelling one’s population through a unified and to a large extent homogenous education system ensures some form of social cohesion as one or a few specific languages and cultural elements are put forward as crucial to being internalized by the nation’s youngsters (Hannerz, 1992). The education system is designed to educate and socialize youngsters in such a way that they can easily be incorporated and participate into mainstream society (Reay, 2010). However, as the goals of these social institutions are designed primarily by and for the dominant ethnic and cultural majority group, these processes always entail the exclusion of specific other cultural elements and can be perceived as oppressive by subordinate groups (Wimmer, 2006). In this article, we wanted to relate the ongoing nation-building process in Flanders – the so-called macro-level – to the domain of education and specifically to the interactions between the home and the school environment – the so-called meso- and micro-levels. An important framework to study our findings is the theory of the institutional racialization or culturalization of the educational system in Flanders (Phillips, 2011). Institutions, in particular those that are concerned with the education and socialization of children into a national imagination, are neither ethnically nor culturally neutral. In this article, we tried to relate the discourses and practices emerging in families and schools to the broader perspective of nation-building studies. The question arises: does the idea of a culturally and linguistically unified representation of the Flemish nation have relevance when studying teacher-pupil-parent relations? More specifically, are the socialization goals in the school and home environment characterized by congruence or incongruence? We studied this by contrasting the goals and strategies put forward by parents on the one hand and school actors on the other with respect to the issue of ethno-cultural diversity, while at the same time considering the broader societal context.

The findings show that at first glance, these two general socialization processes appear to differ and even go in opposite directions. Minority parents seem to emphasize their uniqueness by stressing the importance of their mother tongue and cultural and religious background. School actors stress the importance of adaptation to the culture of the dominant group and try to exclude those cultural elements they feel are negatively influencing the educational trajectories of minority students (Agirdag et al., 2012; Van Praag et al., 2016). However, even in these narratives, the importance of valuing diversity is emphasized. Minority parents find it self-evident that their children learn the dominant language, Dutch, and become acquainted with the other cultural emblems of the ethnic majority group. They do not fear their children not learning Dutch, considering their education is conducted entirely in Dutch; moreover, they see their children watching television in Dutch and using it as the communication tool, mainly with their peers. Minority parents simply want to ‘add’ their mother tongue and other cultural elements to their children’s identity (Agirdag, 2010; Clycq, 2014, Gibson, 1988; Valenzuela, 1999).

Schools actors, too, find it self-evident to value their students’ diverse backgrounds and recognize that this diversity is present in their classrooms and cannot be eradicated (Stevens, 2010). They have been trained and instructed to value this diversity and often attend courses on how to deal with it effectively (Bonilla-Silva, 2003; Lamont and Mizrachi, 2012). However, at the same time, they find it difficult to appreciate the role of ‘mother tongue’ in children’s identity formation and see minority parents’ socialization goals and strategies as obstacles in their children’s educational trajectories (Pulinx et al., 2015). Despite the various trainings on valuing diversity, they do not seem to grasp the importance of one’s mother tongue in identity construction, choosing instead to represent the Flemish educational system as meritocratic and culturally neutral. However, their narratives are oversaturated with so-called culture-specific rules and conditions. Educational success only seems possible if minority families subtract their ‘mother tongue’ from their cultural repertoire and adjust their attitudes and motivations in such a way that they assimilate into the dominant group. When we relate these findings to the broader social processes of nation-building in Flanders, strong similarities become apparent. In the ‘Charter for Flanders’, the dominant political elite represents Flanders as a meritocratic society with a specific culture and language (Vlaams Parlement, 2012). The Charter also specifies that all languages and cultures are valued, but that there is a clear foundation for Flemish society that cannot be denied. This narrative finds its way into the educational system where the Flemish – and now more and more European – ‘culture and values’ and the Dutch language are put forward as the most important cultural emblems minorities need to internalize. As such, the educational system is and remains an important social institution in the nation-building process in Flanders. This becomes very clear when the relation between Dutch and ‘mother tongue’ is discussed, but similar processes can be found when it comes to the relation between majority and minority religion, especially Islam. In secondary education, Islamic teachers and courses are often met with distrust, and in almost all secondary schools in Flanders the wearing of the headscarf is forbidden, while other ‘Christian’ elements remain present. This is certainly the case when the idea of the establishment of ‘Islamic schools’ is uttered, even though there is no public or political debate about the existence of Catholic, Protestant and Jewish schools in Flanders (and Belgium).

In an educational system characterized by strong and even increasing socio-ethnic segregation between schools and educational tracks – with a clear hierarchical structure between the tracks and an overrepresentation of socially vulnerable students (those with an immigration and/or lower SES background) enrolled in the least prestigious tracks, and in grade retention and early school leaving rates (Danhier et al., 2014; De Meyer and Warlop, 2010) – one cannot allow crucial educational actors such as teachers and principals, but also the political elites, to represent the problems in education as a problem of specific minorities. School actors (and politicians) situate educational problems primarily outside of the school and in the home environment. Moreover, although the mantra of valuing diversity is widespread, it is the linguistic and cultural diversity that is identified as the main problem by many educators (Clycq, Nouwen and Vandenbourcke, 2014, Pulinx et al., 2015; Van Praag et al., 2016). Situating part of the problem within the schools implies questioning the status quo, the dominant representation of social reality and especially the privileged position of the dominant national and ethnic majority group (Gillborn, 2005; Phillips, 2011).

Indeed, nation-building processes are primarily concerned with constructing boundaries between an in group and an out group, those who are perceived as valuable citizens who can rely upon the welfare state and its public institutions, and those who cannot (Jenkins, 2008; Wimmer, 2006). However, when these boundaries become too rigid and specific cultural emblems are problematized and even pathologized as a nuisance, then not every citizen is fully respected or has the same opportunities as the majority members of society. Training programmes for school actors can help to change the negative representation of some forms of cultural diversity. However, when the political elite and policy makers continue to represent the Flemish nation as a unified body with one language and one culture, then those training programmes will probably have limited success. The nation-building process of Flanders needs to be more inclusive and more open to the diversity that is already present in current society if it wants to be able to appreciate new forms of diversity that will emerge in the decades to come. The presence of this ethnic, cultural and linguistic diversity should be the starting ground, rather than its representation as exceptional and problematic. The question remains if this is possible in the current nation-building processes of Flanders. The micro-level interactions we studied in this article show that certain exclusionary practices and discourses can thrive in a structural and institutional context wherein it is self-evident to think of cultural, linguistic and religious diversity as the most important problems to becoming successful and a valuable citizen in Flanders. However, sooner rather than later, the dominant ethnic majority in Flanders and society’s institutions will need to approach this question in a nuanced way as over the next few decade(s) this native Flemish majority group will become a minority next to other minorities, in particular in the larger cities. This will slowly change the power relations and dominant discourses on Flanders’ cultural outlook.

However, as various scholars have stressed, the transformations discussed above are not unique to Flanders (Alba and Foner, 2015; Kasinitz et al., 2002). Many metropolitan areas and countries around the world continue to struggle with these issues. In recent years, we have witnessed a relatively widespread public and political negative backlash towards this increasing diversification (Lates, 2011), amplified even further with the recent refugee crisis and so-called radicalization of Muslim youth in Europe. For example, questions are asked about what unifies (European) nations and how an (imagined) European identity and heritage can be preserved (Alba and Foner, 2015). As discussed in this article, one can only wonder at what the repercussions of this supposed need for European ‘cultural’ unity and homogeneity will be for formal education. As Alba and Holdaway (2013) have shown across European countries – but also globally – similar socio-economically vulnerable groups seem to be at risk in education and frequently but not always immigrant and minority groups. In these educational systems, more often than not the so-called cultural difference of the other is indicated to be the most important obstacle and, e.g. the consequences of socio-economic inequalities are neglected (Alba and Holdaway, 2013). Focusing on the specific context of Flanders has shown that explicit ongoing nation-building processes seem to clash with this ethnic and cultural diversification. However, the question remains whether we find similar trends in more ‘established’ or ‘stable’ countries focusing less on developing and strengthening their ‘own’ majority identity.