As in many Western countries, in the Netherlands and in Belgium there is a growing concern about the educational achievement of ethnic minority children, which instigated a body of educational research (e.g. Karsten, e.a, 2002; Driessen, e.a., 2003; Duquet, e.a., 2006; Lacante, e.a. 2007). However, underlying these studies lies the assumption that educational underachievement of certain students (e.g. those of ethnic minority, working class background) can be explained by their individual deficiencies. This point of view can be called ‘deficit thinking’ (Valencia, 1997, pp. 1-8). Our contribution deals with the very nature and the consequences of the deficit thinking. Firstly, we expound our theoretical critique of the deficit thinking. Based on the theory of Pierre Bourdieu (1977; 1998) we describe how ‘deficit thinking’ constitutes a one-sided interpretation of some Bourdieusian concepts, such as the notion of cultural capital, as it claims that minority children do not have ‘enough’ cultural capital. Furthermore, we argue that deficit thinking lacks understanding of the Bourdieusian notion of ‘field’. Secondly, we reason that the deficit thinking is commonly translated into linguistic-deficit thinking. In other words, language deficiency of ethnic minority students is put forward as an explanatory factor for their educational performances. Specifically, in the Netherlands and in Belgium there is a strong emphasis on the knowledge of Dutch, in the field of both educational research and practice. Beside educational issues, most present-day problems of ethnic minorities have been linked to a supposed linguistic deficit (Blommaert & Van Avermaet, 2006). Our aim is to identify processes by which the importance of Dutch is emphasized in the school context and to grasp in which ways this emphasis affects the students’ linguistic behavior. Viewed from our theoretical perspective, we call this overemphasis on the importance of Dutch the `doxa of Dutch’. Following Bourdieu (1977), the concept of ‘doxa’ denotes a particular point of view of the dominant group, which presents and imposes itself as a universal point of view. Accordingly, following hypotheses are formulated. Firstly, we expect that people from both the ethnic dominant group, namely the Belgian majority members, and the ethnic dominated group, namely the Belgian-Turkish minority, ‘believes’ in the doxa (of Dutch). Though, secondly, while the former can be expected to act orthodox, i.e. acting in defense of the doxa of Dutch, the latter can be expected to act heterodox, i.e. acting opposite to the doxa of Dutch (Bourdieu, 1977, pp.163-167).

In order to test our hypotheses we use in-depth interviews that are conducted with 19 Turkish and Belgian students from Belgian secondary education. These interviews are transcribed and coded. To meet the requirements of descriptive inference, the selection of respondents is done on the independent variables (King e.a., 1994). These variables are the ethnic background (Belgian / Belgian-Turkish) and parental education (high educated / low educated).

Supporting the hypothesis of ‘doxa of Dutch’, we found that almost all actors at school ‘believed’ in the importance of acquisition of Dutch language. However, the responses of ethnic dominant (Belgian) and ethnic dominated (Belgian-Turkish) actors varied when it comes to the importance of Dutch in private sphere. While the Belgian students claimed the overall importance of Dutch, the Turkish-Belgian students refused the value of speaking Dutch at home, in private sphere. Supporting our second hypothesis, we found that believing in the importance of the Dutch, did not necessarily have an effect on the linguistic behavior. In a concrete manner, we stated that ethnic dominant (Belgian) students acted orthodox, i.e they acted in defense of the doxa of Dutch as they spoke Dutch and as they expected that everybody have to speak Dutch, while the ethnic dominated (Belgian-Turkish) students acted heterodox, i.e. they resisted the doxa as they spoke another language.