Language census: Part of the population census

Background

When Belgium held its first population census, in 1846, the opportunity was taken to make an assessment of the language situation. The census officers the municipalities sent out to undertake the survey asked general questions about the socio-economic situation of the households. However, they were also required to investigate language activity. The questions varied from census to census. In 1946 the interviewees had a choice between 'French or Walloon', 'Flemish or Dutch', German, English, or 'other languages'.

Starting in 1866 the interviewees were able to choose between three national languages (French, 'Flemish' and German), 4 combinations (French-Flemish, French-German, Flemish-German and French-Flemish-German) or the option 'none of the three'. It was unclear whether the question concerned the mother tongue, language knowledge or language activity.

The 'deaf and dumb' option was also available in 1866.

In 1880 children under two years of age were a separate category, 'regarded as not speaking'.

In 1910 an assessment was also made of the language bilingual people most frequently used: French, Flemish or German.

Another category to be added, in 1930, contained bilingual people who 'had not indicated the language they most frequently used'. From that time on censuses featured percentages in addition to absolute figures.

During the last language census, in 1947, the reference was no longer to 'Flemish' but to 'Dutch'. The number of unilingual and bilingual people who expressed a preference for a particular national language was also added up.

Political implications

To start with these censuses had no political implications: pursuant to the constitution citizens in the new Belgium were free to use the language they wanted to but nearly all official operations were carried out in French. In the light of language laws at the end of the 19th century, and particularly those from 1921 and 1932, the prevailing monolingual French character of the Belgian state came under increasing pressure. The impact of introducing the general single voting right (1921) and the growing economic clout of the Flemish was primarily responsible for the status of Dutch being upgraded in the central institutions and in Brussels. Concern that public life would become bilingual across-the-board prompted the Walloon Movement to call for monolingual language areas to be demarcated. As part of the language area demarcation process, a language border had to be determined and the language censuses represented a key political instrument towards this end (apart from urban development and other arguments).

For private purposes, people could still use whatever language they chose, but one language law after another stated that judicial, administrative and educational activities should mainly be carried out in the local language: Dutch in Flanders and French in Wallonia. The Brussels agglomeration was allowed a separate language status: officially bilingual but in practice mainly French. Almost all local authorities in Brussels opted for French as the internal official language.

In the light of the language census findings, Brussels was seen to be systematically expanding. The growing population and the urbanisation process meant the main Dutch-speaking municipalities on the periphery of Brussels were becoming systematically Frenchified – the so-called oil stain phenomenon.

An increasing number of the surrounding municipalities were officially become bilingual. In the wake of the 1920 language census, the Brussels agglomeration was extended from 15 to 17 municipalities: Sint-Pieters-Woluwe and Sint-Stevens-Woluwe were assigned a Brussels language status. Laken, Haren and Neder-over-Heembeek were also annexed by the city of Brussels.

The 1932 language law for administrative affairs even specifically linked the language status of language border municipalities to the language census findings. As soon as 30% of the population declared speaking a language other than the official language of the municipality, the municipality was required to attend to these inhabitants in their own language (a kind of facility before its time). As soon as 50% of the population declared speaking another language, the local authority had to change the official language and thus shift the language border. The language censuses therefore had direct political implications. One outcome was that Sint-Stevens-Woluwe was taken back out of the Brussels agglomeration because less than 30% of the population there were French speakers.

Right after the next census (owing to the Second World War this was organised only in 1947) more and more voices were raised in protest on the Flemish side. Owing to this controversy the 1947 findings were published only in 1954. What it revealed was that Brussels had once again grown. Ganshoren, Evere and Sint-Agatha-Berchem had more than 50% of French speakers and were assigned a Brussels language status. Drogenbos, Wemmel, Kraainem and Linkebeek had more than 30% of French speakers and were beholden to a system based on external bilingualism. The objectivity of the censuses had already been called into question but the Flemish Movement mobilised on a massive scale as a result of three Flemish peripheral municipalities being re-annexed to Brussels. Flemish associations, such as the Davidsfonds, the Vermeylenfonds and the Willemsfonds, said the censuses had been manipulated and called for them to be abolished and for the language border to be permanently established.

The Flemish Action Committee for Brussels and the Language Border (VABT) was launched. More than 500 Flemish local authorities opposed any new language census. The government thereupon postponed the 1957 census and on 24 July1961 language censuses were abolished in Belgium by law. In 1962 the language border was established as part of a larger community compromise.

No official information about language knowledge or language activity has been issued since then. Elections results are sometimes used as indicators for the language composition of Brussels and of the Flemish peripheral communities or estimates are made on the basis of the language in which identity cards or driving licences are issued but as with the earlier language census, such analyses provide an incomplete and rough and ready picture of the language situation.

A study by the Free University of Brussels (VUB) called 'Taalfaciliteiten in de rand' features a survey on language activity, showing that the highly diversified language situation is partly influenced by migration flows and other social developments. Kind en Gezin (Child and Family) publishes municipal children reports featuring various kinds of information including information about the language spoken at home.

Our database contains many more documents. Most of them are only available in Dutch. However, you can enter English search terms.