Language Use and Language Contact in Brussels

Jeanine Treffers-Daller Faculty of Languages and European Studies, University of the West of England, Frenchay Campus, Coldharbour Lane, Bristol BS16 1QY, UK

Brusselsoccupies a veryspecial position on theLinguistic Frontier, because the 19 communitiesthat form Brussels-Capital are anautonomous region within theFederal Stateof Belgium.The articlefirst gives a shortoverview of thehistorical development of variousaspects of thesituation of theRegion, asthese are essential for an under- standing of theinstitutional and constitutionalframework within which Brussels functionscurrently. The mainfocus of thearticle is on knowledge and useof thevari- etiesof Frenchand Dutch inBrussels-Capital,and on theeducational system, which receiveda lotof attentionin recentresearch. Finally, some attention is given toattitudes towardsthe languages and languagevarieties and to linguisticaspects of language contact.

Introduction Brusselsoccupies a very special,if notunique positionon the Linguistic Frontier,because the 19municipalitieswhich form the Brusselsmetropolis are an autonomousregion in whatis now the federal stateof Belgium. Becauseof its specialstatus, in thisspecial issue aseparatepaper isreserved forthe description of the historicalbackground, the educationalsystem, the language varieties used, the attitudesof the speakerstowards these language varieties,and linguistic aspects of language contact in Brussels.

Language Knowledge and Language Use: Historical Perspectives Brusselsis situated on the Germanicside ofthe LinguisticFrontier, in aterri- torythat used tobe entirely Dutch-speaking. 1 All observersagree thatBrussels is originallya Dutch-speaking city,but the emphasisneeds tobe onthe word‘ origi- nally’(Deprez et al.,1981:94). Until the middle ofthe 18thcentury itwas still almosthomogeneously Dutch-speaking, except fora smallFrench-speaking elite (Van Velthoven, 1987:21). After Belgium became independent in 1830,Dutch continuouslylost ground in Brussels,as a resultof a processof language shift towardsFrench, generally knownunder the termFrenchification. Many factors areresponsible forthis phenomenon. Firstof all, it is important to realisethat French wasthe only officiallanguage ofthe countryuntil 1898(McRae,1986: 25). In the secondplace, the internationalprestige ofFrench alsoplayed arolein the processof Frenchification.The prestige ofFrench asaninternationallanguage contrastedsharply with the lackof prestige attachedto the localvariety of Dutch spoken by the majorityof the inhabitantsof Brussels (De Vriendt &Willemyns, 1987:202). The shortperiod during which the Northernand Southern provinces were united under King WilliamI (1814–1830) was insufficient tostrengthen the positionof StandardDutch. This is not only due tothe factthat Dutch rule was

50 Brussels 51

preceded by aperiod of20yearsof French rule (1795–1814) in which Dutch was excluded fromofficial use. The causesof the delay in the development and spreadof StandardDutch in Flandersgo back to the politicalsplit of the Dutch-speaking territoryin the 16thcentury, when the Northernpart of the Low Countriesrevolted against the Roman-CatholicHabsburg monarchy,but the Southern partremained under Habsburg domination(Van de Craen& Willemyns, 1988).The South lostmany members ofitsintellectual elite tothe North.Although the southernelite contributedto the standardisationof Dutch, the speakersin the South were cutoff fromthis process. As a result,standardisa- tionstagnated in the South andFrench tookon mostof the functionsStandard Dutch obtained in the North. After Belgian Independence, the Flemish provinceswere onceagain cut off fromthe Netherlands.Knowledge ofStandardDutch wasfarless widespread in Flandersthan in the Netherlandsat thattime. According to Willemyns (1984;in De Vriendt &Willemyns, 1987:224), ‘ StandardDutch ishowever used consider- ably morein Brusselsand surroundings than in the remaining partof Flanders’. Apartfrom the factorsmentioned above,the schoolsystem contributed to language shift,as until the end of the 19thcentury, the language of primary educationin Brusselswas French. 2 In particularwith the introductionof compul- soryprimary education in 1914,the schoolplayed amajorrole in the processof Frenchificationin Brussels.It was not until afterWorld War I thata Dutch-speaking schoolsystem was gradually built up (De Vriendt, 1984).As is well known,Dutch-speaking university educationwas not available anywhere in the countryuntil 1930,when the Universityof Gentbecame Dutch-speaking, as a result of pressure of the Flemish Movement. Finally, economicfactors contributed to enhancing the prestige ofFrench. In the 19thcentury, the economicposition of Walloniawas much strongerthan that ofFlanders, due tolarge-scale capital investment in heavy industry(Van Velthoven, 1987:17).Flanders, on the contrary,sank into poverty, which made it even easierto associateDutch withpoverty and backwardness. After the World WarII, however,heavy industrywas confronted with a fundamentalcrisis throughoutEurope, andWallonia was no exception (Witte &Van Velthoven, 1999:200). Due toindustrial development in Flanders,the economicpower rela- tionswere reversed,with Flanders overtaking Wallonia for the firsttime in the 1960s.

Language Censuses The processof language shiftin Brusselsis reflected in the language censuses, but their interpretationremains very difficult formany reasons, one ofthem being thateach time different questionswere asked.The firstofficial language censusof 1846showsthat the percentage ofthe populationthat used French most frequently wasmuch higher in Brussels(37%) than in Gent(5%) or in Antwerp (1.9%).Even if the Brusselsfigures maybe exaggerated, 3 they showthat Frenchificationwas much fasterin Brusselsthan elsewhere in the country.This wasin partdue toWalloonimmigrants, who formed the majorityof the immigrantsin the firsthalf ofthe 19thcentury (De Metsenaere, 1987).Between the language censusof 1866 and the lastlanguage censusin 1947,the number of 52 Journal of Multilingual and Multicultural Development

Dutch monolingualsdropped sharply,from 46.2% to 9.5%, whereas the number ofFrench monolingualsrose from 19.3% to 37%. In 194724.4% of the Brussels populationwas registered asusing Dutch exclusively orprimarily,and 70.6%as using French only ormainly (McRae, 1986: 295). All these figures should howeverbe seen asindicatingtendencies ratherthan exact proportions, because the resultsof all language censusesare heavily contestedfor methodological reasons.For the lastthree censuses,an additionalproblem wasthat the climate wasvery hostileto the Flemish, because ofFlemish collaborationwith the Germanoccupant during the twoworldwars (see Gubin, 1978fora detailedcriti- cism of the census data). ForBrussels and its suburbs inparticular,the censusresults had direct polit- icalimplications. Since the language lawof 1932,municipalities in the area surroundingBrussels had to adopt external bilingualism (in contactswith the public) if they countedmore than 30% Francophones.As soon as thisfigure rose to50%,French alsobecame the internallanguage ofthe municipal administration,in additionto Dutch.Clearly, the expansionof Brussels– which continued toannex moreand more Flemish municipalities– formeda majorthreat to the Flemish cause.In 1963,the boundariesof the Brusselsagglomeration were offi- ciallyestablished and confined to19municipalitiesthat formed part of itso far. Six municipalitieson the borderof the agglomeration,situated in the Dutch-speaking partof the province ofBrabant,obtained a specialstatus with so-calledfacilities for Francophones: Drogenbos, Kraainem, Linkebeek, Sint-Genesius-Rode, Wemmel andWezembeek-Oppem (Witte,1993: 12; see also Willemyns in thisissue). Fromthe Flemish perspective, the delimitationof Brusselswas avery importantmilestone, which limited the dangerof the French ‘oilstain’(olievlek) spreading overthe Flemish country-side.From a Francophoneperspective, onthe otherhand, this delimitation meant the imposi- tion of an artificial ‘collar’ (carcan) on the natural growth of the capital. Itis impossible to give anyofficial figures ofthe numbers ofspeakers of each language group after1947, because language censuseshave been abolished,due tothe tensionsthey created.A number ofsurveysof language knowledge and use were carriedout from the 1960sonwards, each of which came up witha different result.In these surveys,estimates of the percentage of speakersof Dutch range from14% to27%(see Treffers-Daller, 1994for a detaileddiscussion). Janssens(2001) shows how difficult itis to obtain reliable dataon knowledge and use ofDutch in Brussels.In asurvey of2500Brusselers, Janssens shows that 10% claimsto come from families in whichonly Dutch (ora varietyof Dutch) is spoken, but 70%of the samesample claims to havea reasonableknowledge of Dutch.One ofthe problems withthe survey is,of course,that data are basedon reportedlanguage behaviour,rather than observed language behaviour,and it remains unclear to what extent the data represent actual knowledge and use. Anotherproblem withmany surveys is that they implicitly orexplicitly assumespeakers are either Dutch-speaking orFrench-speaking, andexclude the possibilityfor speakers to be classifiedas being bilingual. Clearly,figures about the number ofbilinguals arevery difficult togive, because speakersmay have very different opinionsabout the meaning ofthe notion‘ bilingual’. An extreme example ofthe problemsinvolved in assessingthe number of bilinguals in Brusselscan be found in asurvey in Le Soir (1985),which claimedthere were only Brussels 53

1.8%‘ absolutebilinguals’ in Brussels.The figure isprobably solow because manypeople hesitatedto say thatthey were absolutelybilingual, despite the fact thatthey use French andDutch onadailybasis. In addition,it is probably true to saythat both language groupshave negative attitudestowards bilingualism. Francophonesused toconsider bilingualism tobe useful forthe Flemish, but not forthe Walloons(Destré e, 1923),whereas the Flemish feared bilingualism was nothing morethan a transitionto French monolingualism,which was no doubt the casefor many indigenous inhabitantsand Flemish immigrantsin Brussels.It isnot surprising therefore thatin reforming the Belgian stateindividual bilin- gualismhas not become amodelgoal (Witte &BaetensBeardsmore, 1987: 8). The priorityhas rather been oncreating bilingual structures,based on individual monolingualism.In practicalterms, this means that it is now possible forthe individualin Brusselsto use only French oronly Dutch in politicaland administrative matters, education and cultural activities.

The Constitutional and Institutional Framework Asfor its constitutional and institutional status, Brussels occupies a very specialposition on the LinguisticFrontier, because Brussels-Capitalis an autonomousregion in the federal stateof Belgium. In the courseof a 25-yearlong processof reformof the state,in whichfinding asolutionfor ‘ the Brusselsprob- lem’was the mostdifficult issue,Brussels became one ofthe three autonomous regionsof the new Federal State(see alsoWillemyns andthe bibliography tothat contributionin thisissue). In 1980,the politicalinstitutions of the Walloon Region andthe Flemish Region were put in place,but ittook until 1988for the Brussels-CapitalRegion toobtain its own political institutions. 4 The firstelec- tions for the Brussels Council took place in 1989. The Flemish hadlong opposedthe institutionof Brusselsas aseparateregion, because they feared thatBrussels and Wallonia together could form a front againstFlanders (Detant, 1995: 19). In addition,the Flemish felt thatthe bestway toprotect the interestsof the Flemish populationin Brusselswas to maintain as closea link aspossible between Flandersand Brussels. In the compromise reachedduring the secondconstitutional reform, the Francophoneshad to make asacrificetoo in thatthey wereunable toobtainthe breakthrough ofthe ‘carcan’: the limitsof the Brussels-CapitalRegion remained unaltered.The Flemish, onthe otherhand, tried to maintain the links withBrussels as clearly as possible by choosingBrussels as itscapital (Witte, 1998: 14). The officially bilingual Region of Brussels-Capitalis situated between twoofficially monolingualregions: a monolingual Dutch-speaking Region,Flanders (Vlaanderen) in the North,and a monolingualFrench-speaking Region,Wallonia (Wallonie) in the South. 5 The Regionsare to a large extent autonomous,which means that they havetheir own regional governments and their own parliaments, known as the Councils. The divisionof labourbetween the different levels ofgovernmentis a very complexaffair in Belgium. In the firstplace, there isthe Federal Governmentand the Federal Parliament,which remain responsible, amongother things, for finances (in order,for example, toguarantee monetary union), the army,impor- tantparts of civil,commercial and criminal law, social security, foreign affairs, relationshipswith other members ofthe European Union andNATO, animpor- 54 Journal of Multilingual and Multicultural Development tantpart of healthcare and major aspects of the country’s internalaffairs. The responsibility forother matters is distributed between the three Regions (Brussels,Flanders and Wallonia) and the three Language Communities(the Flemish community,the French communityand the Germancommunity). The Regionsare responsible formatters relating to their respective territories,such as economicpolicy, employment, regionaldevelopment, agriculture,housing, transport,energy andso forth. Matters relating to culture andeducation, however,are the responsibility ofthe three Communities.The Flemish andthe French Communitiestherefore eachexercise their competencies inculturaland educationalmatters in their respective Regions,as well asin Brussels. 6 The complexitiesof the divisionof labourbetween the different governmental levels in Belgium become very clearif one realisesthat there aresix different governments(including the Governmentof the Germancommunity), each of which has a share of the administrative and legislative power. 7 The Councilof Brussels-Capitalis allowed to voteregional laws, called ordi- nances,in mattersrelating to its competencies, and can thus develop agenuine regionallegal system,valid only in the Region ofBrussels-Capital, within the limitsof the boundariesset by Federal Lawand European legislation.As conflict between the twolanguage communitiesin the Councilcannot be ruled out,even afterfederalisation, a so-called alarmbell procedure hasbeen put inplace.This meansthat in casetwo-thirds of the members ofalinguisticgroup considera particularproposal to have a negative impacton the interestsof thatgroup, the Regional Government can intervene. The inhabitantsof Brussels-Capitalelect their deputies forthe Council directlyby compulsoryvoting. As the politicalparties are split along linguistic lines, the compositionof the BrusselsCouncil gives aninteresting perspective on the relativeimportance of bothlanguage communitiesin the metropolis. Currently (elections of1999),64 (or82%) of the 75Councildeputies belong to French-speaking partiesand 11 (or 17%) to Dutch-speaking parties.It would, however,be amistaketo cometoanyconclusions regarding knowledge oruse of the twolanguagesin the Capitalon the basisof votingbehaviour. As there areno bilingual parties,bilinguals haveto choose either aFrench-speaking partyor a Dutch-speaking partyand they maywell chooseon the basisof issuesentirely unrelated to language matters.

Languages Varieties in Brussels Adescriptionof the language varietiesthat are being used inBrusselsis more complexthan in manyareas around the linguistic frontier,because ofthe wide range ofvarietiesthat are being spoken, andbecause ofthe terminologicalconfu- sionaround these varieties.De Vriendt andWillemyns (1987)distinguish four different varietiesof Dutch,and four different varietiesof French. Aswehave seen above,traditionally, the Brusselspopulation used tospeak alocalvariety of Dutch,sometimes entitled ‘BrusselsDutch’ (De Vriendt &Willemyns, 1987; Treffers-Daller, 1994)or ‘BrusselsFlemish’ (De Vriendt &Goyvaerts,1989). This dialectbelongs tothe group ofBrabantic dialects spoken in the centralpart of the Dutch-speaking territoryof Belgium. Localinhabitants often refer toit as ‘Brussels’or ‘ Flemish’, atermwhich linguistsreserve forthe Flemish dialectsthat Brussels 55

arespoken in the Westof Flandersonly. Nowadays very few speakersare monolingual usersof BrusselsDutch, as dialectusage isgoing downeverywhere in the country,but particularlyin Brussels(see Willemyns, 1979on reported knowledge ofdialects by Flemish studentsfrom various Flemish provinces).According toDe Vriendt andGoyvaerts (1989) the everyday use ofthisdialect nearly always implies thatthe speaker alsohas an activecommand of French, whichis, among otherthings, related to the tendency ofthe Brusselspopulation to visit French ratherthan Dutch schools:traditionally, therefore, the inhabitantsof Brussels tended tochoose French ratherthan Dutch astheir codefor formal and written purposes. The use ofstandard languages is another very complexmatter. Two varieties of StandardDutch arebeing used in Brussels:Belgian Dutch andStandard Dutch.De Vriendt andWillemyns (1987:204) define Belgian Dutch as‘ a supra-regionallanguage which ismore or lessstandardised and may, for those whospeak it,function asastandardlanguage’ . Itdiffers fromStandard Dutch, one ofthe three officiallanguages of the country,in thatit is influenced by Brabanticdialects and because itdisplays many archaisms, dialecticisms, galli- cisms, purisms, etc. Asforthe French varieties,it is important to see thatthere isno such thing as anindigenous BrusselsFrench variety,because French isan ‘imported’language in Brussels(De Vriendt &Willemyns, 1987:205). Yet, French asspoken bythe inhabitantsof Brusselsmay possess sufficient commoncharacteristics to considerit a separatevariety (Baetens Beardsmore, 1971). According to De Vriendt andWillemyns (1987:206) Brussels French isa kind ofBelgian French thatis isolated from other varieties because ofits particular status in Brussels. StandardFrench isthe secondofficial language ofthe city.It is generally used in the media and in other formal domains.

Attitudes In thissection I will tryto summarisesome findings in relationto attitudes towardsBrussels, its inhabitants and its languages, as they canbe found inthe literature,in particularin McRae(1986). Clearly, these attitudesmust be seen in the contextof the problematicintergroup relationsbetween the Dutch-speaking andthe French-speaking populationsin Belgian societyas a whole.Brussels does,however, occupy aspecialposition in Belgium, andthere iscertainly evidence thatthe inhabitantsof Brussels,‘ Brusselers’, aswewill callthem, see themselves as distinct from the Flemish and the Walloons. Afirstimportant statement about the Brusselerscomes from Jules Destrée (1912:11, mytranslation)in afamousopen letterto the King. In hisopinion, there areno Belgians in Belgium: only Walloonsand Flemish. Hegives the following, extremely negative evaluation of the Brusselers: They seem tohave combined the shortcomingsof the tworaces 8 and to havesacrificed their goodqualities. Their meansof communicationis an awful gibberish, thathas been popularised by the Beulemans andthe Kaeckebroek families, 9 whounwittingly stressthe comicalaspects of this language use. They areignorant and sceptical. Their idealis a kind of comfortablemediocrity. They don’t believe in anything,and are incapable 56 Journal of Multilingual and Multicultural Development

ofgenerosity orenthusiasm.(… )These inhabitantsof the capital(… )are not a separate people at all; they are a collection of half-breeds. Thisnegative view ofthe inhabitantsof Brussels also emerges fromrecent studiesinto group imagesand attitudes, a very detailedoverview of whichcan be found in McRae(1986). According to McRae(1986: 92), the sharplypejorative imagescan be explained onthe basisof frustration and perhaps envy onthe part ofthe moredisadvantaged periphery in relationto a morefavoured centre. On the otherhand, it may well be the casethat the Flemish andthe Walloongroups in the conflict,which sawa complete separationof allpublic servicesin Brussels intoa French- anda Dutch-speaking networkas the only solution,have little sympathyfor those who have found apragmaticsolution to the problems,and use bothlanguages in their everyday lives,seemingly withoutencountering conflict.Just like bilinguals anywhere in the world,Brusselers can be heard switchinglanguages according to interlocutorand topic. For the aboveWalloon observer,this pragmatic attitude is evidence forthe factthat Brusselers ‘ donot believe in anything’and that they are‘ halfbreeds’. Such negative attitudes prevail in much morerecent literature,also on the Flemish side,where the factof alternativelyusing tolanguages is seen asafailure tobe consistent,and those whodemonstrate such behaviour are classified as ‘ambivalent’(Louckx, 1978). In addition,from a Flemish perspective, the indigenous inhabitantsof Brussels leave the impressionof being ‘very Frenchified’(Louckx, 1978: 58). Their behav- iourcontrasts sharply with that of the ‘bewuste Vlamingen’, whospeak Dutch consciouslyand deliberately in Brusselsunder all(or almost all) circumstances (Deprez et al.,1981).There issome evidence, however,that indigenous Brusselers(born andbred in one ofthe 19municipalities) donot consider them- selvesto be either Walloonor Flemish, but aseparatecategory (Louckx, 1978; Treffers-Daller, 1994).Many ofthe informantsin the lattertwo studiesexpressed pro-Belgian andanti-federalist views, as well asa markedantipathy towards regionalpolitical parties and Flemish andWalloon nationalism. In mainstream Belgian politics,it would be hardto find anydefendants ofthe idealof abilingual Brusselsin which French andDutch wouldlive togetherharmoniously nowadays,but in the 19century thisview wasprevalent amongthe so-called ‘Flamingantsde Bruxelles’(Gubin, 1978).They consideredthe hybrid character ofthe cityto be adistinctasset. Bringing twolanguages together in one townwas consideredto be stimulatingand enriching, aprivilege unfortunately denied to monolingual nations. Deprez et al.(1981,1986) present afascinatinganalysis of the language attitudesof secondaryschool pupils in Dutch-speaking andFrench-speaking schoolsin Brussels.The methodologyand results of the matchedguise studies they carriedout cannot be discussedhere in anydetail, but someimportant findings shouldbe mentioned.Deprez et al.found strikingdifferences between the attitudesof French-speaking pupils andDutch-speaking pupils towardsa number ofstimuli they were presented with:a setof eight different types of speakersin Brussels,ranging froma personwho always speaks Dutch ora varietyof Dutch, to a personwho speaks only French in Brussels.In various subpartsof the studythe datawere collectedby researcherswho presented themselvesin different guises,which again are typicalfor the Brusselssituation: Brussels 57

aresearcherfrom Antwerp whospoke Dutch,a researcherfrom Brussels who presented himself asaFrenchified Fleming (speaking either French orDutch), anda researcherwho presented himself asaWalloonunable tospeak Dutch.As the Dutch-speaking pupils belong tothe minorityin Brussels,they feel threat- ened by the presence ofaresearcherwhom they perceive tobe amember ofthe oppressing majority:the researcherwho presents himself in their schoolin a French guise orin the guise ofa frenchified Fleming. The French-speaking pupils, onthe otherhand, react less negatively tothe presence ofa researcher whopresents himself inDutch attheir school.For these French-speaking pupils, whobelong tothe majorityin Brussels,there isno real threat in the presence ofa minority language speaker in their midst.

Educational Issues Much tothe surpriseof many outside observers, there areno French-Dutch bilingual schoolsin Belgium. Schoolsin Brussels(as well asin anyother area of the country)use only one language ofinstruction,except forforeign language classes.This can only be understoodfrom a historicalperspective. Asspacedoes notallow us to describe the educationin Brusselsbefore 1830,the readeris referred toBehling andDe Metsenaere (1979)for details about education in Brusselsduring the French period (1795–1814) and to Behling andDe Metsenaere (1982)for the period ofunification with the Northernprovinces (1814–1830). According to McRae (1986), for half acentury afterBelgian Inde- pendence, mostteaching in Brusselswas offered in French togenerations of Dutch-speaking pupils, mostof whomunderstood little or nothingof whatwas saidin the classroom.In addition,many of the teachersdid not understand Dutch (Van Velthoven, 1987:34). It is not until 1881that Karel Buls,then mayorof Brussels,replaced the existing structureby asystemof transmutation.Children were allocatedto French orDutch classes,according to the language they spoke. Itwas the schoolprincipal andnot the parentswho decided aboutthe child’s language. In three steps,children were prepared forFrench asthe medium of instruction,as thatwas still considered the bestsolution for Brussels. The system failed, formany reasons, among other things, because ofthe lackof qualified personnel andprejudice onthe partof the Dutch-speaking parents.The transmutationclasses quickly became very unpopular andfinally only the poorestchil- dren attendedthe Dutch classes.Dutch educationthen became synonymous withpoverty and backwardness, and it was only throughthe medium ofFrench thatone couldbreak outof thissystem. The Dutch classeswere emptied in 1914 when the citycouncil introduced the freedom of the headof the family for primary education (Van Velthoven, 1987: 36). The language lawsof 1932laid down the principle ofterritorialityin language matters.For education this meant the language of instructionin Flanders, Walloniaand the Germancantons was tobe the language ofthe region (McRae, 1986:220). The language lawsof 1932 formed an important watershed in the country’s educationalhistory, as they madethe formationof aDutch-speaking middle classin Flanderspossible. 10 In Brusselsand in bilingual communesalong the linguistic frontier,the language ofinstructionwould be ‘the mothertongue of the child orthe child’s usuallanguage’ (McRae, 1986: 220). The freedom ofthe 58 Journal of Multilingual and Multicultural Development headof the family wasabolished(Witte &Van Velthoven, 1999:138). The headof the family hadto officially declarewhat the mothertongue ofthe child was,and there were controlmechanisms which aimed to check thisdeclaration. Despite the factthat the headof the family wasnot free tochoose the language ofthe schoolfor his children, the declarationwas open tomanipulationby parentsor schoolprincipals who believed thatknowledge ofFrench wasa prerequisite to upwardmobility. In practicetherefore, accordingto Deprez et al.(1981:105), the Frenchificationof the Brusselspopulation through education continued unabated.The language lawsof 1963 abolished the transmutationclasses and the number ofDutch schoolswas increased. In 1970the ‘libertédu père defamille’ wasreintroduced (Witte &Van Velthoven, 1999:212). In compensationfor the reintroductionof the freedom ofthe headof the family,the Flemish obtained someconcessions: a considerablenumber ofDutch-speaking day-carecentres were tobe createdand in Dutch-speaking primaryschools the normsfor splitting up classes were lowered significantly (Deprez & Wynants, 1989: 31). Todaythe Brusselspopulation is free tochoose a French-speaking ora Dutch-speaking schooland both networks are well developed, even though there aremore French-speaking primaryand secondary schools. There are Dutch-speaking andFrench-speaking primaryschools in all19 municipalities, but sixmunicipalities do not have a Dutch-speaking secondaryschool (Elsene/Ixelles, St Gilles,St Joost-ten-Node/St Josse-ten-Noode,St Lambrechts-Woluwe/Woluwe-St Lambert,Vorst/ Forestand Watermaal- Bosvoorde/Watermael-Boitsfort).A comparisonof datafrom the Vlaamse Gemeenschapscommissieand the CommunautéFranç aise de Belgique shows that36% of the primaryschools (including infantsand junior departments)in Brussels-Capitalare Dutch-speaking and64% areFrench-speaking. The propor- tionof Dutch-speaking schoolsis slightly lowerat the secondarylevel: 29%and 71% (see Table 1 for more details). The VlaamseGemeenschapscommissie (VGC) alsoprovides numbers of pupils currently attendingDutch-speaking schools.The figures showthat numbers in kindergartensand primary schools increased by 2%and3% respec- tively between 1999and 2000, whereas there wasa slightdecrease (0.05%) of pupils in Dutch-speaking secondaryschools (see Table 2).It is interesting tonote thatthe Dutch-speaking schoolslose some pupils every yearto French-speaking schools:in 1999–2000 7% of the children in the kindergartensleft the Dutch-speaking schoolfor a French-speaking primaryschool, whereas 6% ofthe children in the primaryschool went to a French-speaking secondaryschool. It maywell be the casethat these arechildren fromhomogeneous French-speaking families(see below). D’Hondt(1999) notes that the number of pupils in Dutch-speaking secondaryschools dropped since 1990,the yearin whichthe firstcounts of pupils in secondaryeducation were made(Gatz, 1996). There is alsoevidence fora significantexchange withFlanders: on the one hand,11% of the Dutch-speaking primaryschool population left Brusselsfor a secondary schoolin Flanders.On the otherhand, the dataof the VGCshowthat 892 (42%)of the 2103pupils in the firstyear of the Dutch-speaking secondaryschools come from the Flemish provinces. Acomparisonof the figures in Table 2withthose provided by Deprez and Wynants(1989) shows that the Dutch educationalsystem has clearly progressed Brussels 59

Table 1 Dutch-speaking and French-speaking schools in Brussels Capital (all networks)* Municipalities Dutch-speaking schoolsFrench-speaking schools Primary Secondary French Secondary infants + maternelle + junior schools primaire Anderlecht 31 9 28 11 Brussel/Bruxelles** 25 11 67 14 Elsene/Ixelles 2 – 15 9 Etterbeek 5 1 12 8 Evere 6 1 6 1 Ganshoren 5 1 4 3 Jette 12 3 12 3 Koekelberg 3 2 3 1 Ouderghem/Auderghem 6 1 7 2 Schaarbeek/Schaerbeek 10 2 26 11 StAgatha–Berchem/ 7 1 3 – Berchem–Ste Agathe St Gilles 3 – 8 6 St Jans Molenbeek/ 14 2 17 3 Molenbeek–St Jean St Joost-ten Node/ 2 – 8 2 St Josse-ten-Noode St Lambrechts Woluwe/ 6 – 15 5 Woluwe-St Lambert St Pieters 74 11 5 Woluwe/Woluwe St Pi- erre Ukkel/Uccle 8 2 21 7 Vorst Forest 7 – 14 3 Watermaal–Bosvoorde/ 3 – 7 3 Watermael-Boitsfort Total 162 40 284 97 Notes: * There are three main networks: the official subsidised network, the free subsidised network (catholic and neutral) and the network of the Communauté franç aise/ Vlaamse Gemeenschapscommissie. The figures were last updated on last updated 13/02/2001 (primary school) and on 08/03/2000 (secondary school). ** The figures for Brussels refer to Brussels 1000, Haren/Haeren, Neder-over-Heembeek and Laken/Laeken. Data source : Brussels Onderwijspunt, website of the Vlaamse Gemeenschapscommissie: http://bop.vgc.be/scholen/index.html and the site interréseaux de l’Administration Géné rale de l’Enseignement de la Recherche Scientifique de la Communauté Franç aise de Belgique: http://www.agers.cfwb.be/index.htm

since the 1980s.The numbers in Dutch-speaking primaryschools had been drop- ping since the end ofthe 1960s,when 15,150children were registered,and reachedan all-timelow in 1983–1984 when 8268pupils were registered.In the Francophoneschools numbers fell too,however. The reasonsfor these falling numbers areto be soughtin the dropof the birth rateand an urban exodus 60 Journal of Multilingual and Multicultural Development

Table 2 Pupils inDutch-speaking kindergartens, primaryschools and secondary schools Kindergartens Primary schools Secondary schools 1999 2000 1999 2000 1999 2000 8666 8860 11,451 11,808 11,347 11,341 Data source : The website of the Brussels Onderwijspunt (Vlaamse Gemeenschapscommissie), September 2000. (Deprez &Wynants,1989). In the 1980sthe Dutch-speaking schoolpopulation formedlittle more than 10% of the totalBrussels school population, with the remaining 90% of pupils attending Francophone schools. Itis interesting tonote that, despite the factthat bilingual educationdoes not officially existin Brussels,Dutch schoolsare not only attendedby children from homogeneousDutch-speaking families(i.e. familiesin whichboth parents are Dutch-speaking, cf.De Bleyser et al.,2001;Deprez et al.,1981;Deprez &Wynants, 1989;Gielen &Louckx,1984). Table 3gives anoverview of the development in Dutch-speaking primaryschools, and shows that the percentage ofchildren fromhomogeneous Dutch-speaking backgroundsin primaryschools decreased steadilybetween the early1980s andthe academicyear 1999– 2000 (data from the Vlaamse Gemeenschapscommissie). Table 3 Language background ofpupils in Dutch-speaking primaryschools in Brussels School Homogenous Homogeneous Homogeneous Mixed (Dutch and year Dutch-speaking French-speaking other language another language) (neither Dutch nor French) 80–81 7225 85.1% 205 2.4% 171 2% 890 10.5% 99–00 2852 24.9% 3173 27.7% 2305 20.2% 3109 27.2% Data source : Vlaamse Gemeenschapscommissie, in De Bleyser et al. (2001: 364).

Thisphenomenon, forwhich De Bleyser et al.(2001)coined the term‘ wild immersion’is generally attributedto the followingfactors: the growingimpor- tanceof bilingualism in Brussels,the factthat there arefewer immigrantchildren inDutch-speaking schooland the qualityof the education.It is not known how manyhomogeneous Dutch-speaking familiessend their children to Francophoneschools nowadays. According to a MarketSegmentation Enquiry carriedout on behalf ofthe NCCin 1979(Deprez &Wynants,1989), more children fromhomogeneous Dutch-speaking familiesattend Francophone schools thanDutch-speaking schools(7000 versus 5600). According to this enquiry, the majorityof children ofmixed backgrounds(Dutch/ French) attendFrancophone schools:3200 children in Dutch-speaking schoolsand 20,000 children in Francophoneschools. More recent informationregarding the language background ofthe children in Francophoneschools is however necessary to help clarifywhether educationcontinues to play arolein the Frenchificationprocess in Brussels. Brussels 61 Language Contact Language mixing ofallkinds (borrowing, code-switching and interference) havea very badpress in Brussels,but thisis the casein manybilingual communities,according to Poplack(1980) whocalls it astigmatisedsociolinguistic marker. In asituationof conflictbetween language communities,it is not surprising that thisnegative attitudetowards language mixing ispronounced very clearlyby variousobservers, who see itas evidence ofthe speakersbeing ‘semi-bilingual’ orunable tospeak anylanguage properly. Destrée (1912:11) qualifies this mixtureas anawful gibberish, andVan Velthoven (1987:36) sees it ascultural impoverishment.Wilmars (1971: 80, my translation),a Flemish observer,gives the following characterisation of language use among Brusselers: When acommonFlemish speaker begins tospeak French in Brussels,he quickly discoversthat he will never be able tospeak like agentleman. And ashe isunwilling to‘ murder’the beautiful French language, he triesto overcomehis language problems by simply chatteringaway, mixing French andFlemish. The resultis the awful language usage thatis ridiculed in Pourquoi Pas . Treffers-Daller (1992)shows that code-switching is no longer currentpractice amongthe younger generationsof indigenous inhabitantsof Brussels.Among thoseinformants who have been toDutch-speaking schools,code-switching and borrowingwas found significantlyless than among those who went to French-speaking schools.In additionto the factorsmentioned above,the polaris- ationbetween the twolanguage communitiesin Brusselsmay be responsible for the factthat code-switching is not currently commonpractice. As Myers-Scotton (1993:128) puts it: code-switching as an unmarked choiceis not predicted to occurin ‘communitieswhere the maincandidates for such switching are also symbolsof present intergroup competitionor conflict’ . Finally, strongpurist traditions on the French and the Dutch sides may well have played a role here. Treffers-Daller (1999)shows that borrowing, the incorporationof featuresof one language intothe other,on the otherhand, is a frequent phenomenon in the localvarieties spoken in Brussels.Brussels Dutch mainlyborrows lexical items fromFrench, whereasstructural borrowing from French islimited. For Brussels French, the oppositeis true.Lexical borrowing from Brussels Dutch isless impor- tantin BrusselsFrench, but structuralinfluences fromBrussels Dutch canbe found morefrequently. The overallpicture thusreveals basic asymmetries between the influences in bothdirections. These asymmetriescan be predicted andexplained withthe help ofThomason and Kaufman’ s frameworkfor contact-inducedchange. It is well knownthat many speakers of BrusselsDutch haveexperienced language shiftin the directionof the prestige language French. Asamatterof fact,there arefew monolingual speakersofBrussels Dutch (De Vriendt &Willemyns, 1987:217). As a resultof the processof language shift, French asspoken in Brusselsis typically marked by substrate(and adstrate) influence of the Germanicvarieties. This influence becomesapparent in phonology andsyntax rather than in the lexicon,as predicted by Thomasonand Kaufman. 62 Journal of Multilingual and Multicultural Development

The factthat Brussels Dutch borrowedextensively fromFrench isalso in line withBloomfield’ s observationthat ‘ borrowinggoes predominantly from the upper language tothe lower’(Bloomfield, 1933:461). Structural borrowing in BrusselsDutch islimited to relativelyminor phenomena. Phonological,morpho- logicaland syntactic influence ismainly visible in the wordsborrowed from French. French phonemes only appearin French borrowings,but notin native words,and derivational suffixes fromFrench arenot attached to Germanicroots. Onthe level ofsyntax,it has been shownthat the specific positionreserved for French adverbsis mainlyaccessible for French borrowings,and hardly for native adverbs.Only in the extensionof the use of van (of) hasthe influence ofFrench become apparentin Dutch structures.This is however a very peripheral phenomenon. The basicsyntax of BrusselsDutch hasremained unaffected by French. Phonologicalinfluence fromthe Germanicvarieties in French, onthe contrary,is not confined tolexical borrowings from Brussels Dutch. Brussels French isundoubtedly markedby Dutch phonologicalrules. On the level of syntaxwe haveseen thatconstituents can be placed in the positionbefore the subject. Thisoccurs in sentences whichdo not contain any lexical item from Dutch.Thus, whereas structural influence in BrusselsDutch isclearlylinked to lexicalborrowing, structural interference inBrusselsFrench isnotconnected to lexicalborrowing at all.This confirms Thomason and Kaufman’ s (1988:114– 115) predictionthat ‘ while borrowedmorphosyntactic structures are more often expressed by actualborrowed morphemes, morphosyntactic interference throughshift more often makesuse ofreinterpreted and/orrestructured target language morphemes’. More detailsabout language contactin Brusselsand about the similaritiesand differences between the Brusselssituation and the contactpatterns at other points along the linguistic frontier can be found in Treffers-Daller (1994, 1999).

Correspondence Any corrspondence shouldbe directedto Dr Jeanine Treffers-Daller,Facultyof Languagesand European Studies, Universit yofthe Westof England, Frenchay Campus,ColdharbourLane,BristolBS161QY,UK([email protected]).

Notes 1.During the lastconstitutional reform, in 1992,the province ofBrabant, in which Brussels is situated, wassplit in two:the province ofFlemish Brabantin the North and the province ofWalloon Brabantin the South.Though situated in the province of Flemish Brabant,Brussels is administratively unrelated toprovince, exceptfor matters ofpublic order, forwhich the province remains responsible (Witte,1998: 30). 2.De Vriendt (1984)however mentions the existence ofDutch-speaking classes in Cath- olicschools and points tothe factthat in towns onthe periphery ofBrussels, such as Molenbeek (now one ofthe nineteen municipalities ofBrussels Capital), Dutch-speaking schools were common. 3.In the census figures of1866, the figure forspeakers ofFrench drops again to20%. This may bean indication thatthe figure of30% given for1846 was exaggerated (Treffers-Daller, 1994: 15) 4.More details canbe obtained from the webpages ofthe Federal Belgian Government (http://belgium.fgov.be/)and the webpages ofthe three regions, in particularthe Region of Brussels-Capital (http://www.brussel.irisnet.be/). Brussels 63

5.The Walloon Region alsocontains the German-speaking cantons,all of which are situated in the Province of Liège. 6.To make it possible foreach Community todevelop policies specifically forBruxelles three particularinstitutions have been created: the Commission Communautaire Française (COCOF),the Commission CommunautaireFlamande (orVlaamse Gemeenschapscommissie VGC)and the Commission CommunautaireCommune (CCC).The COCOFconsists ofthe French-speaking members ofthe Brussels Council and the VGC consist of its Dutch-speaking members. The CCC consists of both. 7.The French Community and the French Region eachhave different governments and different parliaments. Onthe Flemish side, on the other hand, one Government and one Council represent both the Flemish Community and the Flemish Region. 8.Destré e even goes asfaras calling the Flemish and the Walloons twodifferent races: the Germanic and the Romance ‘race’, though Witte and van Velthoven (1999:41) comment that the broad French term ‘race’ is largely synonymous with ‘nation’. 9.Fonson and Wicheler’s (1910)famous play Le mariage de Mlle Beulemans and Léopold Courouble’ s series ofnovels abouta family called ‘the Kaeckebroek owe their popularityto their exaggeration ofthe peculiarities ofBrussels French and their Flemish calques. 10.McRae (1986: 221) points outthat the laws only applied toschools subsidised bythe state. Private unsubsidised schools could still exist outside the law.

References Baetens Beardsmore, H.(1971) Le Français Ré gional de Bruxelles .UniversitéLibre de Bruxelles, Institut de Phonétique, conférences et travaux3. Bruxelles: Presses Universitaires de Bruxelles. Behling, A.and De Metsenaere, M.(1979)De taalsituatiein het lager onderwijs tijdens de Franse Periode. Taal en Sociale Integratie 2, 235–60. Behling, A.and De Metsenaere, M.(1982)De taalsituatiein het lager onderwijs tijdens het Verenigd Koninkrijk, 1814–1830. Taal en Sociale Integratie 6, 199–230. Bloomfield, L. (1933) Language. New York: Henry Holt. De Bleyser, D.,Housen, A.,Mettewie, L.and Pierrard, M.(2001)Taalvaardigheid en attitudes van Nederlandstalige en Franstalige leerlingen in het Nederlandstalige secundair onderwijs in Brussel. InE.Witte and A.Mares (eds) 19keer Brussel. Brusselse Thema’s 7 (pp. 359–74). Brussels: VUB Press. De Metsenaere, M.(1987)The impactof geolinguistic and socialprocesses onthe language situation in Brussels. InE.Witte and H.Baetens-Beardsmore (eds) TheInterdisciplinary Study of Bilingualism in Brussels (pp. 123–166). Clevedon: Multilingual Matters. Deprez, K.,Persoons Y.and Versele, M.(1981)Over het identiteitsgevoel van de Vlamingen in Brussel. Deel 1. Taalen SocialeIntegratie 4(pp.93– 166). Brussels: VUB Press. Deprez, K.,Persoons Y.and Versele, M.(1986)Over het identiteitsgevoel van de Vlamingen in Brussel. Deel 2:Het standpunt van de Franstalige meerderheid. Taal en Sociale Integratie 8 (pp. 333–74). Brussels: VUB Press. Deprez, K.and Wynants, A.(1989)Minority problems: On the progress ofNetherlandic primary educationin Brussels. InK.Deprez (ed.) Languageand Intergroup Relations in Flanders and in the Netherlands (pp. 29–44). Destrée, J.(1912)Lettre auroi sur lasé paration de laWallonie et de laFlandre. Brussels: Weissenbruch. Destrée, J. (1923) Wallons et Flamands, la querelle linguistique en Belgique . Paris: Plon. Detant, A.(1995) DeToepassing van de Taalwetgev ingin deBrusselse Gemeentelij ke Instellingen. Brusselse Thema’s 2 . Brussels: VUB Press. De Vriendt, S.(1984)Le bilinguisme en Flandre et àBruxelles et l’enseignement des langues vivantes. InJ.F.Hamers, J.D.Gendron and R.Vigneault (eds) Du disciplinaire àl’interdisciplinaire dans l’é tude du contactdes langues (pp.137– 152). Qué bec: IRB, B-135. De Vriendt, S.and Goyvaerts, D.(1989)Assimilation and sandhi in Brussels. Leuvense Bijdragen 78 (1), 1–93 64 Journal of Multilingual and Multicultural Development

De Vriendt, S.and Willemyns, R.(1987)Linguistic research on Brussels. InE.Witte and H. Baetens Beardsmore (eds) TheStudy of Urban Bilingualism in Brussels (pp.195– 231). Clevedon: Multilingual Matters. D’Hondt, A-S.(1999) Taalproblematiek in het Brussels onderwÿs. In P.Frantzen (ed.) NederlandstaligeBrusselaars in eenmulticulturele gemeenschap (pp.151– 172). Brussels: VUB Press. Gatz,S. (1996) Taalverwervingin hetNederlandstalig basisonderwijs te Brussel . Brussels Hoofdstedelijk Gewest, Raad Vlaamse Gemeenschapscommissie. Gielen, G.and Louckx,F. (1984)Sociologisch onderzoek naarde herkomst, het taalgedrag en het schoolkeuzegedrag van ouders met kinderen in het Nederlandstalige basisonderwijs van de Brusselse agglomeratie. Taal en Sociale Integratie 7, 161–208. Gubin,E. (1878)La situation des langues àBruxelles auXIXime siècle àlalumiè re d’un examen critique des statistiques. Taalen SocialeIntegratie I,33–79. Brussel: VUB Uitgaven. Janssens, R.(2001)Over Brusselse Vlamingen en het Nederlands in Brussel. In.E. Witte and A.Mares (eds) 19keer Brussel, Brusselse Thema’ s 7 (pp.43– 83). Brussels: VUBPress. Louckx,F. (1978)Linguistic ambivalence ofthe Brussels indigenous population. International Journal of the Sociology of Language 15, 53–60. McRae,K.D. (1986) Conflictand Compromise in MultilingualSocieties. Belgium . Waterloo (Ontario): Wilfried Laurier Press. Myers-Scotton,C. (1993) Socialmotivations for codeswitching:Evidence from Africa . Oxford: Oxford University Press. Poplack,S. (1980)Sometimes I’ll start asentence in Spanish ytermino en español: Toward a typology of code-switching. Linguistics 18, 581–618. Thomason, S.G.and Kaufman,T. (1988) LanguageContact, Creolization, and Genetic Linguistics . Berkeley: University of California Press. Treffers-Daller, J.(1992)French-Dut ch codeswitching in Brussels: Socialfactors explaining its disappearance. Journalof Multilingualand Multicultural Development 13 (1&2), 143–156. Treffers-Daller, J.(1994) MixingTwo Languages:French-Dutch Contact in aComparative Perspective . Berlin: Mouton de Gruyter. Treffers-Daller, J.(1999)Borrowing and shift-induced interference: Contrasting patterns in French-Germanic contactin Brussels and Strasbourg. Bilingualism:Language and Cognition 2, 1–22. Van de Craen, P.and Willemyns, R.(1988)The standardization ofDutch in Flanders. International Journal of the Sociology of Language 73, 45–64. Van Velthoven, H.(1987)The process oflanguage shift in Brussels: Historical background and mechanisms. In E.Witte and H.Baetens Beardsmore (eds) TheStudy of Urban Bilingualism in Brussels (pp. 15–46). Clevedon: Multilingual Matters. Willemyns, R.(1979)Bedenkingen bij het taalgedrag van Vlaamse universiteitsstudenten uit Brussel-Halle-Vilvoorde. Taal en Sociale Integratie II, 141–160. Brussels: VUB. Willemyns, R.(1984)La standardisation linguistique en dehors des centres de gravitéde lalangue: LaFlandre et le Québec. In J.F.Hamers, J.D.Gendron and R.Vigneault (eds) Dudisciplinaireà l’interdisciplinaire dans l’ étudedu contactdes langues (pp.2– 70). Qué bec: IRB, B-135. Wilmars, D. (1971) Diagnose Brussel . Antwerpen: Scriptoria. Witte, E.(1993)Faciliteiten voor taalminderheden in de Brusselse rand. InE.Witte (ed.) De Brusselse Rand. Brusselse Thema’s 1 (pp. 168–209). Brussels: VUB Press. Witte, E.(1998)De taalpolitieke aspecten van het Brusselse gewestvormingsproces (1970–1995). In E. Witte et al. (eds) HetBrussels Hoofdstedelijk Gewest en deTaalwetgeving (pp. 13–34). Brussels: VUB Press. Witte, E.and Baetens-Beardsmore, H.(eds) (1987) TheInterdis ciplinaryStudyof Bilingualism in Brussels . Clevedon: Multilingual Matters. Witte, E.and van Velthoven, H.(1999) Languageand Politics. The Situation in Belgiumin an Historical Perspective . Brussels: VUB Press.