Multilingual Education in the City of Brussels or How to Implement European Language Policy in Primary Schools.

Author(s): Piet Van de Craen
Institution/Organisation: Vrije Universiteit Brussel (BE)

1. Description of the case

1.1 Scope of the initiative

At the end of the 1990s a decree administered by the then Wallonian Minister of Education, Laurette Onkelinx, made it possible for schools belonging to the French-speaking school network, to start with content and language integrated learning (CLIL) provided they entered a specific demand describing the language pedagogical project they envisaged. As a result a small number of schools started out with what they called immersion programmes. In fact, it is called enseignement de type immersif. Later the number considerably increased. By now they are more than 200 schools involved.

Around the same time the school board of the Dutch-speaking primary schools in Brussels got interested as well. Contacts were laid between the board and the university and it was decided that from the school year 2001-2002 a CLIL-like initiative would start in three schools, namely first in the communes of Schaarbeek and Etterbeek, and soon  Sint-Pieters-Woluwe joined. Now also a primary school in Laken has adhered. It was decided to start with one form per year and build it up from there.

The type of initiative was regional, small scale and involving three kinds of partners (i) the schools and the school board, (ii) the school counselling service and (iii) the university, i.c. the Vrije Universiteit Brussel who would do the follow up of the activity. An ad hoc administration board was founded presided by the university, i.c. the author of these lines. The name given to the initiative was STIMOB which stands for stimulating multilingual education in Brussels, in Dutch stimulerend meertalig onderwijs in Brussel.


1.2 Languages involved

From the very beginning it was clear that the languages involved could not be but Dutch and French. Dutch because the schools all belong to the Dutch-speaking school network and French because, first, it is the lingua franca in Brussels spoken by the majority, second, a substantial number of pupils speak French at home and, third, the most decisive argument of them all, the Belgian language laws of 1962-1963 do not allow any other first foreign language but French.

Despite this rather clear and unavoidable choice of language within the founding group many discussions about the desirability of other languages were held. One member insisted heavily on Spanish while others were more in favour of English. Needless to say French won this battle. One might wonder why such a natural choice is not adopted as a matter of fact. This has to do with the complicated attitudes of Dutch speakers versus French due to historical and political reasons. Explanations for this can be found elsewhere (McRae 1988, Murphy 1988, Van de Craen 2003, Allain et al 2007). Interestingly enough after five years of STIMOB and realising the potential of the approach the school decided enthusiastically to add a number of hours of English to the Dutch-French curriculum.

1.3 Learning outcomes

The learning outcomes of STIMOB have been extensively studied. The researchers decided to focus on six tenets as they call it. At the same time it is an example of how studies of multilingualism can contribute to the body scientific knowledge of multilingual teaching. As such this approach can strengthen language policies and policy-driven initiatives in particular areas.

The six tenets under study are: (1) what is the knowledge of the target language (compared to monolinguals), (2) what is the knowledge of the mother tongue (compared to monolinguals), (3) what is the quality of pupil’s subject matter knowledge, (4) how far have their attitudes and motivation toward language learning been affected, (5) how is their cognitive development affected and (6) how is their brain affected (see for extensive coverage of this Van de Craen et al 2007a, 2007b, 2007c, 2008)

1.4 Practical realisation

In this part attention will be paid to the organisation of the initiative, the sociolinguistic context, the number of hours devoted to the project and the subject matter understudy.

The organisation

It was decided to start in each school with one form at the time. After six years the whole school would be part of the initiative. This careful approach had the advantage that, slowly but steadily, teachers could come to grips with a new reality, namely that in their schools, unlike before, French would be an integral part of daily activities. Teachers were found in the French-speaking school network. The first one was actually a teacher of Dutch in the French school system with an interest in bilingualism and language learning. The idea of teaching in French in a Dutch school greatly appealed to her. Other teachers who joined later were mostly young and very enthusiastic about the initiative. Of course being bilingual is a prerequisite although they only teach in one language.

The sociolinguistic context

It was also decided that within the sociolinguistic context of Brussels it would be meaningless to devote as much as 70% of the curriculum in the other language as is done in a number of Wallonian schools (see Lecocq et al 2004). In Brussels and also in most Dutch-speaking schools the majority is French-speaking. As a result it was considered much more appropriate to devote about 20% of the curriculum to French. This would result in a win-win situation since three kinds of pupils can be distinguished for whom the initiative has a different aim: (i) Dutch speakers are on their way to become multilingual in the best possible way, (ii) French-speakers receive support in their mother tongue whereby their subject matter learning is boosted, (iii) speakers of other languages such as Berber, Moroccan, Spanish and various East European languages receive instruction in the two official languages of the area, which will greatly help in their future professional careers.

Number of hours

Pupils receive two hours a week instruction in French in a CLIL like approach. This means that in the first two forms part of the subject matter is repeated by a native speaker of French. This repetitive approach does gradually change. The third form does not practice it any more. In fact, the repetitive approach is a result of the language law. The law stipulates that, in Brussels, in the first two forms two hours can be devoted to repetition of subject matter without stipulating the language in which this is done. Since we wanted to abide by the law we used this loophole to teach in French. However, after a few years it dawned on the ad hoc group and the researchers that this approach is a powerful approach within the method used since repetition avoids any misunderstanding that might arise through teaching in a foreign language. In fact, it is a guarantee that messages come across effortlessly.

Subject matter

Schools could also choose which subject matter they would use. A choice was given between arithmetic, environment studies or the arts such as drawing and other related activities. As it happened one school opted for arithmetic, another for environment and still another for drawing. From the third form onwards the schools were more inclined to use environmental studies as their favourite CLIL activity.

2. Background to the initiative

2.1. Context

The sociolinguistic school population of Brussels is highly mixed. As has been mentioned before this means that the Dutch-speaking school system is populated with three kinds of pupils: Dutch-speakers, who are a minority, French-speakers, most of the time they form the majority, and speakers of other languages, i.e. traditional migrant’s languages, such as Berber, Moroccan and Turkish but also Kurdish, Serbian, Polish, Russian, Spanish and various African languages. It is to be noted that because of the dominant position of French in the streets a number of these pupils have some street French. We call them faux francophones (false speakers of French) because often this French is not standard and full of mistakes.

In Brussels’ primary schools language learning does not start before grade three, i.e. when pupils are 8 years old. This is two years earlier than in the rest of the country where it is only allowed from grade five. A decree of 2004 on language initiation and sensitization makes it possible to start earlier but few schools tend to do this. It is also to be noted that the language law stipulates explicitly that teaching should be done in Dutch for Dutch-speaking schools and in French for French-speaking schools. The law does not permit to teach in any other language.

However, a number of initiatives have undermined the law since 1982. First, the private Brussels’ organization Foyer got permission to teach part of the primary school curriculum in Turkish, Spanish and Italian. This was an attempt to import some of the Canadian immersion techniques originating in the late sixties and early seventies. The organization is still going strong today. In 1998 the aforementioned decree Onkelinx saw the light, while in 2001 STIMOB started. Finally in January 2008 the Flemish Ministry of Education itself took the initiative to carry out a CLIL experiment in nine secondary schools not located in the Brussels area or in its periphery.

A study of the underlying political discourse of the Flemish administration carried out by Allain (see Allain et al 2007) reveals a highly ambivalent attitude towards early language teaching. While foreign language knowledge is not rejected as such official documents tend to emphasize knowledge of Dutch as the best chance for building a career, multilingualism is seen as a danger to the mother tongue and the European language policies should be adapted (but to what is unclear). Finally, it is suggested that multilingual areas such as Brussels and the Brussels periphery should not get any form of multilingual education.

Most Dutch-speaking politicians, contrary to their French-speaking counterparts, have a very narrow minded view on language and language teaching. Their discourse is imbued by a hardly hidden fear and distrust for the French language. This is undoubtedly a remnant of the activities of the Flemish Movement seeking equal chances for Dutch-speakers in the 19th century. But this battle has been fought and won a long time ago although the discourse has not changed for those seeking more than linguistic equality, namely for those seeking an independent Flanders Region. While before, these ideas were considered extreme and associated with right wing politics, in 2008 these right wing political organisations have succeeded in impregnating traditional right and even some left wing groups with these ideas. The succession of political crises in Belgium since the election of June 2007 and continuing through 2008 is ample proof of this.

Despite the unfavourable political climate the language law has over the years been undermined by a number of initiatives favouring European language policies and, in particular, early multilingual education.

As a result of this climate individual schoolteachers show a variety of reactions to the project not always in favour of European ideas. Most teachers working in Dutch-speaking primary schools do not live in the city itself. They are commuters coming to Brussels to work but returning in the evening to the Dutch-speaking villages surrounding Brussels. While the school population shows all aspects of urbanity the teachers are not always prepared to cope with this. Often they bring with them old-fashioned preconceived ideas about the dominantly French-speaking city. It is, therefore, not easy to convince them of the value of a multilingual teaching approach.

2.2. Strategic goals of the initiative

At least five reasons can be identified with respect to the introduction of Stimob in the Brussels’ Dutch-speaking schools.

In Brussels’ Dutch-speaking primary schools the majority of the pupils do not speak Dutch as their mother tongue. In fact most pupils are submersed in Dutch when they come in since no special care is taken for them, i.e. they are considered as speakers of Dutch. Needless to say that this is an undesirable approach. What the Stimob initiative was and still is trying to do in the first place is to build bridges between the language of the home and that of the school not by submersion but by creating an environment where immersion rules.

A second reason to implement a new language teaching approach is the following. In a ‘normal’ Dutch-speaking school in Brussels teachers would still behave as if they were in a ‘normal’ Dutch-speaking school in Flanders, i.e. disregarding completely the home languages of the pupils. In some schools one can see signs that actually forbid pupils to speak French. Needless to say that such an attitude is not creating a feeling of well being. By implementing a learning environment in French respect for the language and its speakers is automatically engendered.

A third reason is the fact that research has shown that learning in another language than the traditional school language or just early language learning has a positive effect on the cognitive development of young learners (cf. Wilburn Robinson 1992).

A fourth reason has to be with attitudes and motivation. It is well known that positive attitudes toward the language enhance motivation. It was felt that learning through a language enhances both attitudes and motivation cf. also Merisuo-Storm 2007).

A fifth reason is European language policy. The recommendation in the white paper (cf. White paper 1995) says that 18 year olds are supposed to have 2 + 1 languages. It is not mentioned in this document how this can be achieved but it is clear that it only can be achieved through some form of bilingual education of which CLIL is one of the most powerful and researched approaches.

3. Success indicators

Success indicators can be divided in societal and individual ones. Since societal ones will show later it was decided to study the individual ones. As was indicated in 1.3. six aspects were studied:

  1. knowledge of the target language (compared to monolinguals),
  2. knowledge of the mother tongue (compared to monolinguals),
  3. the quality of pupil’s subject matter knowledge,
  4. how far have their attitudes and motivation toward language learning been affected?
  5. how is their cognitive development affected?
  6. how is their brain affected?

Below is a summary of Van de Craen et al 2007a, 2007b, 2007c and 2008. Knowledge of the target language is better than compared with the results of traditional language teaching methods. Interestingly enough the knowledge of the mother tongue did not decline, quite the contrary could be observed in many instances. Subject matter knowledge remains the same or is better while qualitative research showed attitudes and motivation greatly improved. The most startling results, however, were obtained in the results of cognitive tests: pupils showed better results in administering calibrated maths’ test even if mathematics was not the subject matter under study in the second language (see also Jäppinen 2005). When it comes to the brain it was shown that multilingual brains are different than monolingual brains and that bilingual school brains evolve in the same way than bilingual brains. Some scholars have called such brains better brains (Blakemore & Frith 2005) and were this indeed the case then an additional argument has been found to foster multilingual education in this sense.

4. Success factors

The main success factors that have made the initiative a success can be summarized as follows.

Thorough analysis of the situation

It was felt important to have a good analysis of the situation. The term definition of the situation refers to an analysis from within rather than from without (cf Thomas 1923; Janowitz 1966). An analysis from within has the merits of being recognized by the protagonists in this case this refers mainly to the teachers. They have to believe in what they are doing and especially changing since change educational practices of teachers is one of the hardest things to do.

Strong institutional support for the language initiative

Stimob is convinced that without strong institutional support the initiative would have failed. School authorities are the main protagonists when it comes to change. Their importance can hardly be overestimated. By the same token this urges reflection on the education of school authorities.

Languages being offered in another way than the traditional one

The fact that languages are taught in an implicit way and that less attention is paid to grammar is certainly an important aspect on the primary school level. However, in some cases this may lead to neglecting reading and writing and this should be strongly avoided. The fact that implicit learning is promoted should not mean that all aspects of cognitive learning should be abandoned (see also Lyster 2007).

The continuity of the initiative

When it comes to learning few things are as important as continuity. It is surprising that this should be emphasized. However, teachers and parents alike have a tendency to forget this simple rule of thumb. While it is accepted that pupils do not become mathematical wizards over night it is often not accepted that it takes years to master a language. As a result many strange ideas surround language learning. Emphasizing continuity by teachers and school authorities is the way forward. The project under study has taken this into account and it is surely one of the success factors.

5. Lessons to be Learned

Out of  the Stimob experience numerous lessons can be drawn.

1. Collaboration between schools and universities

Universities and certainly not language departments usually collaborate with schools. As this project demonstrates it can be done. Universities reaching out to schools might be an interesting approach for spreading European language policies.

2. Creation of an interface

To formalize this collaboration it is advised to create some kind of interface where all the partners can interact and where the decision making related to the project can take place.

3. The follow up

Collaboration between universities and schools is never easy. Schools have different expectations universities have a hard time to fulfil. Universities should communicate as much as they can and therefore the interface is an interesting tool. Results of the follow up should regularly been distributed to the teachers.

4. The success of educational change should never been taken for granted.

Even after a number of years the new ideas might be challenged again. Just as European ideas take times to trickle through new educational ideas take time to get validated by the protagonists. This is especially true in areas where the official language policy is not in line with the ideas and background of the project.

5. Early language learning is a powerful learning tool.

The Stimob project has once again demonstrated what a powerful learning tool CLIL education can be. So far the result surpass by far those of any other approach and outclass a number of well meant but rarely sustained initiatives such as extra-curricular activities, exchange programmes, etc. Additionally it should be pointed out that CLIL initiatives do not cost a lot and that they can easily be implemented. The language pedagogical implications of the Stimob project not only refer to language learning and teaching but also refer to innovative learning practices, change in education and the European language policy.


Allain, L., E. Ceuleers & P. Van de Craen. 2007. Third DYLAN Report. Unpublished report.

Blakemore, S.-J. & U. Frith. 2005. The Learning Brain. Lessons for Education. Oxford: Blackwell.

Jäppinen, A.-K. 2005. ‘Thinking and content learning of mathematics and science as cognitional development in content and language integrated learning (Clil): teaching through a foreign language in Finland’. Language and Education 19, (2), 148:169;

Janowitz, M. 1966. W.I. Thomas. On Social Organization and Social Personality. Chicago: University of Chicago Press.

Lecocq, K., Ph. Mousty, R. Kolinsky, V. Goetry, J. Morais, J. Alegria. 2004. Evaluation des programmes d’immersion en communauté française: une étude longitudinale comparative du développement des compétences linguistiques d’enfants francophones immergés en néerlandais. ULB: Ministère de la Communauté française.

Lyster, R. 2007. Learning and Teaching Through Content. Amsterdam: John Benjamins.

McRae, K. 1986. Belgium. Waterloo, Ont.: Wilfrid Laurier University Press.

Merisuo-Storm, T. 2007. ‘Pupil’s attitudes toward foreign language learning and the development of literacy skills in bilingual education.’ Teaching and Teacher Education 23(2), 226-235.

Murphy, A. 1988. The Regional Dynamics of Language Differentiation in Belgium. Chicago: Committee on Geographical Studies.

Thomas, W. 1923. The Unadjusted Girl. Boston: Little, Brown and Company.

Van de Craen, P. 2003. What, If Anything, Is a Belgian? Yale French Studies 102, Special Issue: Belgian Memories, 24-33;

Van de Craen, P., K. Lochtman, E. Ceuleers, K. Mondt & L. Allain. 2007a. ‘An interdisciplinary approach to CLIL learning in primary schools in Brussels. In: C. Dalton-Puffer & U. Smit (eds) Empirical Perspectives on CLIL Classroom Discourse. Frankfurt am Main: Peter Lang, 253-274.

Van de Craen, P., E. Ceuleers & K. Mondt, 2007b. ‘Cognitive development and bilingualism in primary schools: teaching maths in a CLIL environment’. In: D. Marsh & D. Wolff (eds) Diverse Contexts – Converging Goals. CLIL in Europe. Frankfurt am Main: Peter Lang, 185-200.

Van de Craen, P., K. Mondt, L. Allain & Y. Gao. 2007c. ‘Why and How CLIL Works. An Outline for a CLIL Theory’. Vienna English Working Papers 16(3), 70-78;

Van de Craen, P. E. Ceuleers, K. Mondt & L. Allain. 2008. ‘European multilingual language policies in Belgium and policy-driven research’. In: K. Lauridsen & D. Toudic (eds) Language at Work in Europe. Festschrift in Honour of Wolfgang Mackiewicz. Göttingen: V&R Press, 139-151;

White paper. 1995. Teaching and Learning. Towards a Cognitive Society. Brussels: European Commission.

Wilburn Robinson, D. 1992. ‘The cognitive academic and attitudinal benefits of early language learning’. In: M. Met (ed) Critical Issues in Early Language Learning. New York, White Plains: Longman, 37-43.