Crossbred Language Practices and Identity Construction Among the Youth from the Second Generation of Congolese Migration in Belgium: School Issues

1 PhD Researcher, Université Catholique de Louvain, Institut Langage et Communication (ILC), Louvain-LaNeuve, Belgium, kasongo.priscilla@gmail.com Abstract: Transitions phenomena such as migration can result, among individuals involved, in identity and language instabilities. To regulate those, migrants can build strategies. These are the strategies that we focus on for our doctoral research. In this study, we analyse language practices of the youth from the second generation of Congolese migration in Belgium, in order to identify the identity strategies they set up, and to construct profiles corresponding to those strategies. Considering a dynamic conception of identity, this research extends those on identity strategies, and is part of the theory of social identity and self-categorization. The methodology is qualitative, inspired by the Grounded Theory Method, and proceeds through individual and semi-structured interviews. Our initial hypothesis suggested a three levels system of choice (extreme / radical, situational moderation, or crossbreeding). On this basis, we achieved a first exploratory field that allowed us to specify the topic of language practices. In the second field, we studied language thematic in context. This article is published after a presentation made during the 19th Congress of the World Association for Educational Research (AMSE-AMCE-WEAR), which took place in Succeava (Romania), from the 4th to 8th June 2018. In the following pages, we present the general context of our research, the first field results, a preliminary discussion, and the link with school issues.

Mobility experiences such as migration are also likely to lead to changing disruptions, of practices, representations and language directory of individuals (Van den Avenne, 2005). Thus, the migration process would result in the disturbances of migrant identification, representation and language.
To manage all these disturbances and adapt to new spaces and realities, the individual will implement strategies, "(...) conducts deployed in the conscious or unconscious purpose of avoiding, or at least moderate subjective tensions" (Camilleri, Costa-Lascoux, Hily, & Vermes, 2000: 31) caused by encountered difficulties.
If newcomer migrants are inevitably affected by the phenomena of tension and identity strategies, their children cases should also be considered. These young, otherwise called 'second generation', are "(...) children of immigrants, that is to say, people born in a country other than that of their parents (or who joined the immigration countries of their parents at a very young age, before compulsory schooling) " (Manço, 2012: 3). For them, identity tensions are even more complex that they operate in an environment with different values and traditions, sometimes diametrically opposed to their original culture, which leads to a worsening of tensions (Manço, 2002). Moreover, daily immersed in multicultural situations, they are more exposed to the development of intercultural skills, allowing them "(...) to conciliate culture of origin and host culture, without denying one or reject the other (...) one of the major difficulties (...) [being] precisely to master values and behaviours valued in the two types of societies (...) without falling into blocking contradictions" (Manço, 2002: 14-15). In other words, to achieve a cultural 'crossbreeding' 2 (Laplantine & Nouss, 1997). Between Belgian and Congolese culture, the youth of the second generation of Congolese migration in Belgium are no exception.
However, scientific research on migration is often oriented towards migrant's integration, especially by language, and often, in francophone countries, by French (Noyau, 1984). The inclusion of the native country languages in the integration process, especially among youth from migration is still uncommon (Billiez, 1985). As for works on Congolese immigration in Belgium, they focus more on migratory flows (Flahaux & Shoumaker, 2013) and very often in a statistical and demographic perspective (Schoumaker, Flahaux, & Schoumaker, 2012). And regarding studies on Congolese communities currently established in Belgium (Vause, Schoumaker, & Verhoeven, 2011) or the youth from this immigration, few address the communicational and linguistic approach.
Thus, our research aims to provide qualitative data on language practices of the youth of the second generation of Congolese migration in Belgium. Thereby, it attempts to answer the following two key questions: "Can languages be mobilized as identification and belonging factor? ", and arguing so," how language use is likely to reveal a speech about oneself, speech which can lead to identify strategies and establish identity construction profiles? ".

State of the art
Considering the research question, we set 'language' and 'identity' as key concepts of this study, which oriented the research theoretical choices. Regarding the state of the art, starting from a dynamic and constructivist conception of identity, our research is part of the studies on identity strategies. Regarding language, it is anchored in sociolinguistics.

Dynamic identity
Our research is part of the studies on identity strategies. However, we cannot address the issue of identity strategies without addressing the dynamic identity one. Accepting that social actors are able to set up identity strategies in order to adapt to their environment, is recognizing beforehand that they build their identity according to their social interactions and experienced contexts.
However, the dynamic concept of identity has not always been unanimous. Indeed, it has evolved from a static conception to an increasingly dynamic and complex one. In the sociological tradition, the collective consciousness was systematically separated from the individual one, the second being determined by the first (holistic-deterministic line) (Benedict & Raphael, 1950;Bourdieu, 1979;Parsons, 2013). The hyper socialized individual had little flexibility. Later, the social actor was more considered to be involved in his personality determination (individualist paradigm) (Boudon & Bourricaud, 1982;Crozier & Friedberg, 1977). Cultural variations, although very utilitarian, were therefore possible within common social behaviors [concept of 'basic personality' (Benedict, 1973;Kardiner, Linton, Prigent & Lefort, 1969)]. As for the constructivist approach (Garfinkel & Quéré, 2007;Goffman, 1973;Verhoeven, 1992), it presents an autonomous social actor, with critical, post-conventional and reflective capacities, allowing him not only a conformity and integration to the social system, but also a detachment of its conventional forms.
Once this dynamic and constructivist conception of identity has emerged, it opened the way for further researches on the subject, including those on the identity strategies.

Sociolinguistic
Besides the construction of identity, our subject is grounded in linguistic studies. Given the importance of the socio-cultural aspect in the study, we opted for sociolinguistics (Hymes, 2003). Sometimes called the sociology of language, "sociolinguistics is the study of language varieties characteristics, their functions characteristics and their interlocutors characteristics, considering that these three factors constantly influence each other, change, and modify mutually within a linguistic community" (Fishman, 1971: 20).
This scientific field seems indeed the most appropriate since it allows us to make the link between our informants' language practices (varieties), and their main social feature, which is to be youth from the second generation of Congolese migration in Belgium.
Note that in this study we will talk more of 'language' and not of 'linguistic', to highlight the discourse analysis of our informants on their language use (macro-sociolinguistic perspective) (Boyer & Bayo, 1996).

Hypothesis
At the very beginning of our research, we hypothesized that the youth of the second generation of Congolese migration in Belgium, facing their two cultural reference communities (Congolese and Belgian), mobilize, depending on the context, a three levels identity choices system: the extreme and radical choice, the choice of situational moderation, and the crossbreeding choice.
The results of the first exploratory field allowed us to enrich and to extend our hypothesis by considering:  The youth different levels of culture (Hofstede, 1994) (Congolese, African, Belgian, European and / or international / multicultural) and their proximity to each;  The particular cultural and symbolic proximity of Belgian and Congolese levels, and their variation depending on the primary socialization, life experiences, time, etc.;  The coexistence of two levels of identity: a 'hard' dominant levelconcept of 'saliency' 3 (Licata, 2007) -, fairly steady, little progressive in time, and influenced by the primary socialization; And the flexible dynamic level, daily adaptive, and more scalable;  The ability of Congolese ethnic cultures to generate a 'hard' identity level, culturally more open and mixed (non-EU multicultural sociability);  And the contextual dimension of identity choices. All positions are likely to be invoked, therefore, these young people are 'de facto' culturally mixed.
Note, however, that the hypothesis of the second field (and that we consider today for the entire study) concerns not only the identity positioning strategies, but especially their (possible) link with language practices. Thus, it would matter to verify and to demonstrate not only the relationship between language use and identity construction, but also to study its modalities.
Another precision is important here. We believe that the use of hypotheses is neither mandatory nor necessary when using a qualitative method such as GTM. Indeed, GTM doesn't classically work with hypotheses (Sandu, 2018). In fact, when approaching the field, the researcher is not supposed to have a prejudice on phenomena taking place, and that he'll discover later, throughout the different fields.
However for us, the method has been more of a reference point, rather than a strict framework to be followed. Therefore, at the very beginning of our research, we formulated our presumptions in the 'hypothesis' section. During our research, this hypothesis evolved and served us as the first explanatory analysis for the phenomena studied in the field. This type of use corresponds to what Lejeune calls the 'medium range hypothesis' (Lejeune, 2014 : 15).
Presently, at the end of our research, we've decided to keep the hypothesis, not to confirm or refute the initial presumptions, but rather to show our reasoning evolution.

Methodology
To test our hypothesis, we use the qualitative method. Indeed, as soon as one favors a constructivist approach of the social and of the identity, there is no question "[...] to seek for social identities fixed in substantial types, but for identity formation processes, and for ways to build ones social identity through a positioning in relation to objects, situations, experiences and other" (Verhoeven, 1992: 96).
In this sense, we are inspired by the qualitative Grounded Theory Method (Lejeune, 2014) (GTM) to stay as close as possible to our informants' subjective experience. We say 'inspired' to point out that, as you'll see in the discussion section, our methodological and empirical approach is rather heterogeneous.
On the field, we question the youth of the second generation of Congolese migration in Belgium, with the technique of individual and semistructured interviews.
And to avoid considering chromatic identity issues (Smeralda-Amon, 2002), we do not take into account the biological metis (young people from a Belgian-Congolese white-black couple).
The young people concerned inhabit, study or work in Saint-Gilles, Ixelles, Etterbeek, Brussels and Schaerbeek municipalities.
Brussels city was chosen because it is one of the great places of concentration for the Congolese immigrant population in Belgium (Schoonvaere, 2010). And among Brussels municipalities, those cited above have a significant interest in terms of cultural diversity (with high foreign population and / or mixity rates) (DGSIE, 2012).
Management of field results is carried out, as part of GTM, by content analysis (De Bonville, 2006;Robert & Bouillaguet, 1997), and through the three types of coding proposed by the method (open, axial and selective 4 ).

Theoretical framework
As stated above, in the state-of-the-art section, the two key concepts of 'identity' and 'language' guided the study theoretical choices. As well, the theoretical framework of this research has been configured to make a space for each of these concepts.

Intercultural communication and language practices
Regarding language, the study is immersed in intercultural communication (Abdallah-Pretceille & Porcher, 2001;Gudykunst, 1983;Kerzil & Vinsonneau, 2004;Ladmiral & Lipiansky, 1989;Marmoz & Derrij, 2001), more specifically in the literature on 'intercultural skills' (Manço, 2002;Prêteur & De Leonardis, 1995;Pry, Guillain, & Foxonet, 1997). Indeed, considering both the Belgian nationality and living environment of our informants on the one hand, and their Congolese cultural backgrounds one the other hand, we put these two cultures under discussion in the personal identity situations of our informants. In addition, when these young people move from one social environment to another (i.e. in the strategies they put in place), they develop specific communication skills, of understanding and adapting to different contexts, in short, skills in cross-cultural situations.

Identity strategies
For theorizing our research, concerning identity strategies, we rely on four authors: Tap (1980) and Manço (1999) reveal three identity strategies involved in the psychological development and social integration process: The strategies of "coping", identification, and projection.
In Camilleri et al. work (2002), the identity strategies are mobilized in case of balance rupture between the value and the fact identities: The strategies of identity conflicts avoidance by simple consistency, by complex consistency, and of code conflicts moderation. Verhoeven (2006) is interested in young immigrant in schools and identifies three identity strategies: The communitarian conception of society; the assimilationist strategy; and complex identity strategies (crossbreeding).
Also concerned by young immigrants, Manço (1999) identifies four identity postures: assimilation postures and of conformant differentiation (simple, passive and defensive), and also assimilation and individuating differentiation strategies (complex, active and offensive).

The social identity theory
Finally, we want to emphasize the importance of social identity and self-categorization theory (Licata, 2007) in our research. Indeed, this theory considers the dynamic concept of identity by focusing on the plurality and context concepts in the social construction of identity. It considers that in a multicultural situation, social actors mobilize different types of 'directories' to adapt to different contexts (De Singly, 2003), developing a multidimensional identity (Lahire, 2005).
This theory joins our study because it studies both relations between groups, as the relationship between individuals and their group. Moreover, the theory also emphasizes a dynamic conception of identity, central point of our research. Finally, through the concepts of strategies (individual and group) and identity 'saliency', the theory of social identity and selfcategorization joined all three main postulates of our first hypothesis. In this way, the social identity theory comes to unify our multidisciplinary and somewhat eclectic theoretical framework.

Empirical part
In this second part, we'll present more specifically the two fields we've achieved so far, and we'll try to link the two.

First field: Exploratory
We conducted the first field in March 2015. The aim was exploratory: we wanted to make a contact with ground realities to see the process at work within the theoretical framework defined. The results were supposed to help us define a more concrete research context, especially through the second field. Therefore, the first field consisted in 3 interviews with a total of 4 hours of recording, for 121 pages of transcript (word-for-word transcript). The interviews lasted an average of 1 hour 23 minutes. The longest interview lasted 1 hour and 59 minutes. And the shortest interview lasted 1 hour 4 minutes. Regarding the transcript, we treated 40 pages per interview on average, the longest was 55 pages, and the shorter was 29 pages.
We interviewed two women and one man between 23 and 37 years old. According to our initial sample, they are Belgian of Congolese origin, and attended school in Brussels and Wallonia. Their level of education varies between Bachelor and Master. The level of education of their parents varies between the professional and the academic.
In terms of analysis, the recurring element of this first field is the concept of 'crossbreeding'. Indeed, we have found that informants' statements were not unequivocal as to their identity positioning, but always nuanced. The profiles identified certainly allowed to emphasize the leadings, but some factors highlight the variations necessary to understand the phenomenon: the cultural conditions of primary socialization had induced some type of crossbreeding, expressed in particular linguistic practices and friendly sociability with, as a result, some conception of self, and some apprehension and the corresponding reactions towards racism / rejection experiments.
Based on this analysis, we decided to explore, in the second field, the degree and manner of crossbreeding among the young people from the Congolese immigration in Belgium, connecting the elements identified as significant during this first exploratory field (cultural conditions of primary socialization, language practice, friendly sociability, civic design itself, apprehension and / reaction to racism / rejection).

Second field: Language practice in context
We achieved our second field from December 2015 to March 2016. The theme was: The linguistic practice of the youth from the second generation of Congolese migration, related to their primary mode of socialization, their sociability, their language practice and their citizenship.
This field consisted of 12 interviews (6 women and 6 men), for a total of 18 hours 15 minutes of recording, and 137 pages of transcripts (thematical transcript). On average, we achieved 1 hour 31 minutes per interview, for about 11 pages for each transcript. The longest interview lasted 2 hours 40 minutes, and the shortest lasted 52 minutes.
Informants were born about the 80s (the youngest was born in 1993, and the oldest was born in 1976). They are mostly either students (including adults returning to education) or in medical sector (doctors and nurses).
To analyse this field which was larger than the previous one, we started with preliminary analyses, followed by the actual analysis. These two ways of entering our datas are also linked with to the GTM coding process. The Preliminary analysis consisted in a global quantitative approach and involved open and axial coding, while the actual analysis refers to a more qualitative approach based more on the selective coding.

Preliminary analysis
Preliminary analysis of the second field allowed us to have a general overview of the field.
We mainly discussed the general themes as identified in the interview guide.
Our interview guide consisted of five themes: knowledge and vision of Congo, socialization and sociability, language practice, identity and citizenship, and apprehension / reaction to racism / rejection. To start, we analysed them transversely (situation among all informants for a given theme).
For the knowledge and vision of Congo theme, we found that the country experience is qualitatively negative, due to a lack of inclusion (language and identification) and of environment (culture) understanding ; Congo's vision is changing from positive to negative from childhood to adulthood, or from prior to after the trip (family and media influence); and economic aspects predominate in the construction of the country (positive and negative) vision, while cultural aspects act as brakes on permanent installation.
The socialization and sociability are mainly Congolese (family) and multicultural, with a tendency to Belgian-Congolese. However, the partner choice is more oriented towards the Belgian (white and European) culture.
Considering linguistic practice, it is very diverse. We were able to identify 17 languages, including 2 Belgian, the 4 Congolese national languages, and several Europeans, besides atypical ones (sign language), and from outside Europe (Russian).
In terms of identity and civic participation, we noted that the symbolic link 6 is more related to Congo (connection with the origins, "do not forget"), despite experiences of rejection and language discrimination (Lingala 7 ) in the Congolese community. The link to Belgium is also positive, but more practical, with a rather negative trend. The Belgian symbolic link doesn't survive after childhood, especially because of racism experiences. This explains the lack of interest of informants for the concept of 'integration'. Moreover, the feeling of inclusion is quite low in both communities, although pretty lower in the Belgian community.
Finally, regarding racism and rejection, apprehension and reactions are more vivid in adulthood. Indeed, in childhood, youth are very few prepared by family for diversity and racism issues, so the reactions are more violent. In adulthood, while parents are providing more advices (strategic, pacifists and intellectualists), young have either more passive reactions (not complaining too much, adjusting, bowing down) or a lack of interest in the subject.

Intermediate results: between language and identity
The main objective of our study is to build identity construction profiles, linked to language practices. We had already identified three types of profiles at the end of the first field (Belgian, Congolese and Belgian-Congolese), and the second field analyses allowed us to go beyond this level. First, we analysed languages and language categories, and then we analysed identity profiles, each incorporating the relevant languages and language categories.

Language analysis
For the analysis of languages, we identified sixteen, of which the most mentioned are, listed in order of importance: French, Lingala, English and Dutch. In addition, we've also gathered these different languages in language categories which are, also in order of importance: the category of Indo-European languages, the international languages one, that of Congolese Bantu languages, and the category of Belgian languages. We also noted general and transversal themes such as linguistic discrimination (glottophobia 8 ), and the relationship to language.
What emerges overall is an obvious multilingualism, considering the languages or language categories. For languages, French and Lingala are predominant, and for language categories, those of international and of Congolese Bantu languages dominate.
We considered a significant and multilingual use of French, although quite practical; for Lingala, a strong symbolic link, complete with a small practice; and for English and Dutch, a strong practical aspect, and a weak symbolic link.
Thus, Indo-European languages corresponds to everyday and friendly use, but not intimate; international languages are marked by multilingualism and code-mixing; Bantu languages dominate in the symbolic but are weakened by the low use; and Belgian languages are mostly marked by the low symbolic link.
And for the general transversal themes of linguistic discrimination and relationship to language, we noted a strong symbolic link to community, family and language, and a large opening (to different worlds and conceptions).
Therefore, the linguistic practice of our informants is generally francophone and multilingual, dominated by the French-Lingala association. The French practice reveals the importance of living areas crossed into Belgian society, but also the family transmission (and therefore the practice of parents and of Congo) and the international opening.
The multilingual practice involves a large number of languages and language categories, but also of multilingual language use, of which the predominance of the code-mixing shows important intercultural skills. Not to mention the overwhelmingly positive mention of linguistic discrimination and symbolic conception of language.
Finally, the French-Lingala multilingual association enables the construction of the multiculturally mixed Belgo-Congolese identity.
These practices thus reveal a Belgian-Congolese multicultural identity among our informants, with a strong Congolese symbolic link, multiple intercultural skills, and a practical and daily Belgian and European anchoring.

Identity construction analysis
Considering identity construction, we identified four main orientations. They are, in order of importance: the Congolese orientation, the multicultural one (including the Belgian-Congolese), and the Belgian orientation. We also identified general themes, especially the one concerning racism and / or rejection (apprehension and reaction), but also the general conception of identity, cultural and sports activities, the partner choice, etc.
For the Congolese identity orientation, our informant's identity oscillates between, on one hand, community strategies of cultural transmission (Congolese), and secondly, personal identity strategies of development and social inclusion (Belgian).
Regarding the multicultural identity orientation, the central problem is the constant tension between, on one side, the Belgian-Congolese multiculturalism / crossbreeding (local and qualitative), and on the other side, the international / global multiculturalism / crossbreeding (overall and quantitative). Both kinds may change over time (from local to global).
Finally, concerning the Belgian identity orientation, analysis of themes and different integration conceptions allowed us to distinguish two main profiles: the Belgian-Congolese profile and the Belgian profile. The difference between the two lies in the intensity of diaspora experience. Here, it's all about the way Congo was extended in migration in Belgium. Diasporic experiences (their existence, quality and intensity) will determine either a uniquely Belgian attachment, by dissociation with the cultural and Congolese migration issues; or a more Belgo-Congolese identity positioning. In this sense, the Belgian-Congolese profile will allow the consideration of several types of issues but will also generate different attachments to Belgium. If the Belgian-Congolese profile is the most potentially dangerous of the two, for the sense of belonging to Belgium, it is also the one that is most likely to produce a mixed Belgo-Congolese identification, built in a reconciliation attempt of the two basic cultural references, and therefore the most harmonious one on the identity point of view.
In conclusion for identity profiles, we can say that: crossbreeding is central to building the identity of informants (predominance of the Belgo-Congolese identity profile, with different perspectives, according to the preferred orientation); there is a connection between language and identity construction among our informants (predominance of language issues in identity orientations and importance of linguistic and territorial anchor in the identity orientations deployment); the symbolic and meaningful signification of the Congolese and the Belgian identity orientations is marked in their ability to bring out specific issues (migration for the Congolese orientation, and integration for the Belgian one), but also in their ability to produce quite radical identity profiles.

Preliminary discussion
Before addressing the school issues analysis, we would like to briefly discuss a few points: our methodological-empirical approach and our provisional theoretical model.
The first point highlight an important characteristic of our research : the crossing of methodological boundaries.
Firstly, throughout the research, our overall methodological positioning was essentially qualitative, (understanding modalities VS explaining causality), but sometimes we approached realities in a quantitative way when relevant and necessary (considering the number of occurrences or causality link). In this sense, we consider that our general empirical approach was mainly qualitative, associated with another, quantitative, which was complementary to it.
Secondly, to build our subject, at first, we constructed a hypothesis that was not yet connected to field experiments. This way of proceeding could reveal a hypothetico-deductive posture, but that was not the case. Indeed, as mentionned, our first hypothesis served above all as a necessary starting point for expressing our first research intuitions. We knew that it would be modified in the field depending on the results, and that's exactly what happened: the initial hypothesis was neither confirmed nor reversed, but rather reworked and completed. Our approach is therefore more empirico-deductive.
By the way, as a researcher, we argue for mixed methods, especially in social sciences.
Then, concerning the theoretical model, although we cannot yet present its details, the model construction work being in progress, we expose here its structure and general functioning.
Thus, at this stage of our analysis, we can identify two main parts in the global identity construction model generated : the crossbred identity conctruction with its factors, and the continuum of mixed identity construction, built around identity orientations, and including language practices, as well as different conceptions of integration.
So the identity construction of our informants is mixed, generated by their primary socialization, expressed through their sociability and their language practice, and implying a certain civic identity and a particular stance on racism / rejection.
When specifically considering its language expression, we can see that this identity crossbreeding develops on a continuum with three identity orientations: a Congolese, multicultural and Belgian orientation. The identity orientations are constructed in such a way that the first and the third represent the continuum extremities, that each one can be positively and negatively oriented, and represents a specific conception of integration. On the other hand, there could be, among informants, a predominant and rigid orientation, and a secondary and more flexible one, which may manifest itself according to circumstances.
The general analysis of our thesis is under construction, will take into account these elements, and will develop them.

School issues
School issues cross the entire second field, both in terms of the predominant languages and in identity orientations. These issues are particularly related to Dutch and to the Belgian identity orientation. Indeed, in Belgium, Dutch is a key language in the professional environment, and it is essentially in school that the young francophones learn it. And among the youth from the Congolese immigration, this language also plays an important role in their sense of belonging to the Belgium. Let's explore this more in detail.

Language practices
As stated previously, the school language practice is mainly linked with Dutch. In the corpus, very few labels concern other languages. However, we were able to find some associated with French, English and Lingala.
Dutch is thus the language which school practice is the most mentioned by our informants. A dozen labels in the corpus mention learning in schools. The language is also spoken in academic and professional environment, code-switching with French and English.
Therefore, Dutch becomes increasingly important over time, for young from the Congolese immigration, particularly through school (learning) and professional environment (requirement for employment). In this sense, while Dutch can be connected to the desire of integration into Belgian society, in the Congolese community mainly French-speaking, it will sometimes participate in the exclusion of the young Dutch speakers, functioning as the opposition mark between the two cultural communities (Congolese and Belgian).
As for French, the school practice is mostly friendly. It concerns especially close friends, mostly Congolese, and in combination with Lingala. If one also considers the quite important parents influence for learning and using French, we clearly see the link between language and the intimate of community (childhood, close friends, Congolese, family). Under these conditions, the school system will more deeply strengthen and anchor the relation to French.
Like for Dutch, the school practice of English is linked to learning situations. English is not one of the Belgian national languages, however, it is highly demanded on the international job market, demand tenfold with the presence of European institutions in Brussels. However, English does not come out of this utilitarian aspect, and does not represent anything really symbolic for young people from the Congolese immigration in Belgium.
Lingala is the least language marked by school. Indeed, the practice of Lingala is less spoken in the friendly than in the family atmosphere. In this context, it is mostly spoken with parents and embodies the language of authority. When used with friends, at school, it's mostly Congolese close friends, and in association with the French. And Lingala is considered by our informants as an integrating element in the Congolese community of Belgium, playing the role of identity marker (link to the origins), with a strong symbolic link. Thus, Lingala is primarily a community language, that's why it doesn't find any social relay into Belgian social structures, such as the school system, except in the informality of community friendships.
In conclusion, we can see that among young people from the Congolese immigration in Belgium, the school language practice is predominated by Belgian (Dutch and French) and Indo-European language categories (Italian, Spanish, Latin, Greek etc.).
Thus, the school language practice allows Belgian and multicultural identity orientations to take root among our informants, from their learning and the friendly relations that they allow to build, for a better integration in their life environment (Belgian, academic and professional), but also for their personal development (friendly), extra-communitarian, and more expanded.

Identity orientations
The school practice is valid for the three identity orientations. However, it's more marked in the Belgian identity orientation, a little less in the multicultural one, and lower in the Congolese identity orientation.

The Belgian identity orientation
In the Belgian identity orientation, school environment can be found in the language practice, in socialization and sociability, and more marginally in identity. In addition, it may call as positive and negative issues.
At the level of Belgian positive orientation, we can find, for language practice, school practice of Dutch and French (Cfr. Previous point). In addition, Belgian / white socialization and sociability is enhanced by their school social relay, thanks to the (parental) argument of academic achievement, resulting from this sociability. These elements allow to relate to the theme of opening. Indeed, the use of these languages allows informants to leave their family and community environments to invest the social Belgian, European and international environments.
Concerning the Belgian negative orientation, school issues are in relation to French. We could note a linguistic discrimination made by our informants, relevant to French, particularly towards Congolese of first generation, or those living in Congo. By their 'good' practice of French, they would be more advantaged. This linguistic discrimination indeed involves a Belgian language and social context but refers to Congolese history and culture. Indeed, the 'good' practice of French, and the superiority it confers, through the important transition into the school system, comes from the Belgian colonial period in the Congo where, to be an 'evolved', native Congolese should, among others, go through school and speak a 'good' French. It is thus likely that this Congolese colonial imaginary has been transmitted more or less unconsciously to informants through their parents.
Similarly, in the data related to the Belgian / white negative sociability, this time to ensure the academic success (synonymous with social success), parents will advise their children to get away from the white sociability that harms by distraction. A distrust that leads, in some cases, to a more and more decreased (Belgian / European / white) sociability of our informants over time.

The multicultural identity orientation
In the multicultural identity orientation, the school issue can be found in positive oriented themes only, in language and sociability issues, the last being predominant.
In terms of language practices, we can find the question in the language practice of code-mixing. Thus, hybrid languages have often been mentioned in the school context, with friends, connecting French, Lingala and Arabic.
Issues related to sociability mainly concerns childhood. Multicultural and even international diversity is mentioned here, to which our informants have been exposed to, in Belgian school environment, encouraging their own openness to diversity, but also expanding their sociability.

The Congolese identity Orientation
In the Congolese identity orientation, school issues are fairly marginal in their positive side, and significantly higher in the negative one. They especially mobilize language and sociability issues.
In the negative Congolese identity orientation, language issues are linked with the relation to ethnic / tribal Congolese languages. When the relation is negative, it refers to their lack of utility and (added) value, and the risk of communitarianism linked to their practice. Lingala is specifically mentioned here.
Indeed, it is pointless to speak a language for which one have very few interlocuters, a language that is not used (or little) in everyday life, because that doesn't have any (or few) areas of use, in different spheres of everyday life (social, professional, school, etc.). If the language is not used in the place where one is considering its future life project (Belgium), its value can be quite low. Finally, regarding communitarization risks, it is interesting to see that here, Lingala is seen as dangerous because, by bringing the youth too intensely closer to the community, it turns them away from the Belgian cultural community, judged here as more legitimate.
These conditions will result, for the young, in low practice of Lingala in childhood, in school, and more and more marginal over time.

Conclusion on school issues
Our research raises the question of identity construction modality, among young of the second generation of Congolese migration in Belgium, from their language practices. School issues light on this question particularly well.
They allow us to understand that, in fact, languages are, for these young, a bridge between their origin culture and their everyday life society. That's how Dutch can meet the social needs of employment in Belgium (otherwise the fact that it carries this problem link it with society), and French carries questions, imaginaries and problems of Congolese and colonial origin.
So even if it is mainly the Belgian identity that benefits from the language advantages (and thus identity) of the educational institution, the other identity orientations will also be able to enjoy it, including the multicultural identity orientation, through hybrid languages practice and cultural diversity (sociability). And while it is true that the Congolese identity orientation takes less advantages than the others on this, because of the lack of social (school) relay of its languages, school is still a privileged space for it. Indeed, it's at school that our informants meet their friends from Congo, with who they enjoy practicing Lingala and assert their Congolese identity part. And with the learning and practice of other languages (French, Dutch and English), this is also where they learn to fully achieve their linguistic and cultural Belgian-Congolese crossbreeding.