One could expect, perhaps naively, that countries would be interested in those who know and use several languages, and would make available statistics that reflect the bi- or multilingualism of their population. As we will see below, this is far from the case. In fact, finding out how many bilinguals there are in countries such as Canada, the United Kingdom, France, Switzerland and the United States, among many others, is a very real challenge.
Some nations simply do not have language questions in their censuses. Belgium and France are two examples of this. In Belgium, the last census that contained such questions was in 1947. Up to that date, the results had been used to either attend to inhabitants in their own language (if 30% of the population declared speaking a language other than the official language in the area in question) or to change the official language of the municipalities (if 50% of the population declared speaking it). But under the pressure of certain groups who did not want to accept language shifts in areas around Brussels, language censuses were abolished in Belgium in 1961. No official information about language knowledge and use has been obtained since then via the census. It is only by going to other information sources, such as European surveys of languages, that one can get an idea of the level of bilingualism in Belgium, as we will see in the next part.
France does not have language questions in their census either. The word “langue” (language) does not appear in the 2020 census form, nor in the preceding ones. When I wrote in 2013 to the French National Institute of Statistics and Economic Studies (INSEE) which is in charge of the census, the curt reply I received was that the information I requested did not fall within what it offers. This is due to the fact that, officially, the French Republic wants to give equal treatment to everyone by interacting directly with the individual, without going through minorities, communities, or groups, be they religious, regional, linguistic, etc. A less official reason is that it also prefers not to put too strong a light on the fact that France contains many minority languages, including immigrant languages, spoken by literally millions of people. This said, regional languages were finally recognized as belonging to the French national heritage in an amendment to the French Constitution in 2008, and various language surveys do exist, some sponsored or cosponsored by the state. It will be one of these surveys that we will refer to in the next part.
There are many other countries that do not have questions on languages in their censuses, and even less on bilingualism. Thus, Christopher (2011) states that in the Commonwealth, only 37 of the 71 census authorities included language questions in recent censuses, leaving many without such questions.
Other national censuses do have language questions but for very specific reasons which do not really have to do with bilingualism. The examples we will mention here are those of England and the United States. In England, before 2011, no question about language had been asked in its censuses, unlike in Wales, Scotland and Northern Ireland where one question had been asked. The Office of National Statistics finally decided to ask two questions in 2011 for a number of reasons, according to Sebba (2017). Among these we find: enabling the government bodies to meet their duties under legislation governing race relations and disability discrimination; allowing local and central government to allocate resources for teaching English as a second language, and for translation services within public services; and supporting regional or minority languages like Cornish and British Sign Language.
Question 18 of the 2011 Census was, “What is your main language?” and Question 19 (asked only of those who had indicated a language other than English to the previous question), “How well can you speak English?” The four possible answers for the latter were: “Very well,” “Well,” “Not well” and “Not at all.” The results obtained only allow us to get an idea of the degree of bilingualism, based on language knowledge and not language use, for a very small proportion of the population for whom English is not their main language, a mere 7.7%. As Sebba (2017) writes, bilingualism is statistically visible only for those who have a “main language” other than English. We know nothing about the bilingualism status of 92.3% of the population! Once again, only European surveys can help us get at the state of bilingualism in England and, more generally, in the United Kingdom, as we will see below.
Another country that has language questions, but with very specific reasons, is the United States. The three questions that were developed in 1980, and that have been used since 2000 in the yearly American Community Survey (ACS), came about in an effort to respond to the necessity to know more about those with limited English language proficiency (Siegel, Martin, and Bruno 2001). So as to implement the Civil Rights Act, The Bilingual Education Act, and the Voting Rights Act, there was a need to accommodate people who had difficulties communicating in English. The questions are: “Does this person speak a language other than English at home?,” “What is this language?,” “How well does this person speak English? Very well, Well, Not well, and Not at all.” There are definite limits to how far this assessment goes. Thus, children under five are not covered, people who use a second or third language in their everyday lives, but only English at home, are not counted, English proficiency is self assessed and may be influenced by a person’s original culture, and so on. However, as we will see in the next part, the results to these questions do allow us to get a far better picture of the status of bilingualism in the United States than in England.
Finally, there are those countries that have the appropriate language questions in their censuses to work out the proportion of bilinguals but that put the emphasis on specific classes of bilinguals. Here we will consider Canada and Switzerland. The questions and answers of the census that Canada conducts every five years are a dream for anyone interested in languages and bi-/multilingualism in a country. The 2016 version contains five language questions, the last two in the long form of the questionnaire only:
7. Can this person speak English or French well enough to conduct a conversation?
8a. What language does this person speak most often at home, and 8b. Does this person speak any other languages on a regular basis at home?
9. What is the language that this person first learned at home in childhood and still understands?
16. What language(s), other than English or French, can this person speak well enough to conduct a conversation?
45a. In this job, what language did this person use most often? and 45b. Did this person use any other languages on a regular basis in this job?
A quick glance at these questions, and at past censuses, shows that the census agency is primarily interested in the country’s two official languages, English and French. Statistics Canada states this clearly in its documentation as well as on its many web pages dedicated to the results. Thus there is a page, “Statistics on official languages in Canada,”7 where the two languages are lauded with statements such as, “French and English are the languages of inclusion,” and “Official languages and bilingualism are at the heart of Canadian identity.” This said, non-official languages (immigrant languages and aboriginal languages) are increasingly being analyzed and reported.
With the tables made available by Statistics Canada, and with a bit of help from agency statisticians, one can get a good overview of bilingualism in Canada based on language knowledge (see Questions 7 and 16 above) and on language use (Questions 8a and 8b, as well as Questions 45a and 45b). The one result that is not easily available is the proportion of the population that uses two or more languages in daily life whatever the activity (work, home, social activities, etc.) and without counting a person twice.
Switzerland also has a national census that asks the right sort of questions, although fewer than Canada, and that allows one to get at the proportion of bilinguals in the country if done correctly. But here problems occur with how Swiss Statistics define bilingualism. I will concentrate on the first language question as it is the data obtained with this question that was used by the agency in 2012 to estimate the number of bi- or multilinguals in the country.8 Here is a translation of the question: “What is your main language, i.e. the language in which you think and that you know the best? If you think in several languages and know them very well, then name these languages.” This was followed by a number of language