CAIRN-INT.INFO : International Edition

In its counts of the population living in France, the French administration has long differentiated inhabitants according to their place of birth and their nationality. At the end of the colonial period, even though they were French, the Algerians living in metropolitan France were labelled as ‘French Muslims of Algeria’ and enumerated separately in the census. Angéline Escafré-Dublet, Lionel Kesztenbaum, and Patrick Simon explain how the government managed to identify them in the census, while pretending not to do so.

1From the end of the 19th century, the census bulletin in metropolitan France (mainland France and Corsica) included questions on place of birth and nationality presented in a relatively stable format. From 1891, a distinction was made between French citizens by birth, naturalized citizens, and foreigners. This triple categorization reflects the French administration’s approach to qualifying the diversity of origins [1]. Unlike the United States, for example, where inhabitants were qualified by ‘race’ and ‘ethnicity’ under a system that evolved over the 20th century, the French census applied a set of legal categories, or rather ‘almost’ legal, given that distinguishing naturalized French citizens was not justified by any particular status, but served to put a label on ‘French people of foreign origin’.

2This structure was not applied identically in the former French colonial empire, however, where ‘natives’ did not have the same citizenship rights as Europeans, be they from metropolitan France or elsewhere [2]. Colonial statistics recorded ethnic and religious groups in ways that varied from one territory to another. But how were colonial subjects identified when they came to live in metropolitan France?

Counting the colonials

3Before the Second World War, the distinction made in the French colonies between French citizens (mainly colonial settlers) and French subjects (indigenous populations) was reflected in the metropolitan census, with the former categorized as ‘French by birth’ and the latter grouped with ‘foreigners’. The situation changed from 1946, when, in an effort to quell the emerging independence movements, all inhabitants of the French colonies became ‘citizens of the French Union’. As a major consequence of this reform, the colonial populations living in metropolitan France became legally indistinguishable from the native-born French. This was especially true for the metropolitan census, in which both groups were identified as French.

4The situation of Algerians was especially salient, given the scale of population flows and the status of Algeria in the French empire. Until 1946, Algerians’ mobility was controlled and restricted, both within Algeria (all moves to a different administrative area [département] had to be reported) and towards metropolitan France. The new status of colonial populations under the 1946 constitution had profound implications, since citizens of the French Union now had freedom to travel, notably to metropolitan France, and enjoyed formal equality of rights on French soil. Consequently, large numbers of Algerians migrated to France in response to growing demand for workers for post-war reconstruction. This mobility was especially worrying to the French authorities because they had very few statistical or other means to assess the numbers of new entrants and their characteristics.

Box 1. Categories

The categories and labels reflect both the desire to identify distinct populations of Algeria and the problematic status of this distinction. While by no means official, the term ‘French Muslims of Algeria’ was widely used in administrative documents and elsewhere. The INSEE publications reveal a certain hesitancy: the 1954 census tables use the term ‘Muslims of Algerian Origin’ (Musulmans originaires d’Algérie), those of 1962 refer to ‘Algerian Muslims’ (Musulmans algériens), while many internal INSEE documents use the label ‘North Africans’.

5To control these particular migrant flows, the French authorities repeatedly attempted to enumerate this so-called ‘floating’ population. A census of ‘Muslim families’ was organized in 1951 by the Ministry of Public Health and Population, and a ‘Social survey on the situation of Algerian Muslims living in metropolitan France’ was conducted in 1953 by the Ministry of the Interior. Yet neither of these sources gave a precise count of the number of Algerians in France. It was the census scheduled for 1954 that offered the best potential solution for enumerating this population.

6But herein lies the problem: given that the French and colonial populations were officially identical in every way, the administration could not be seen—at least openly—to treat colonial migrants differently, so it was impossible to include a question such as ‘Are you a French Muslim from Algeria?’ in the metropolitan census bulletins. A simple and practical solution would have been to use place of birth, but the case of Algeria, with its numerous colonial settlers, raised the problem of how to distinguish between those now labelled as ‘French Muslims of Algeria’ (Français musulmans d’Algérie; see Box 1) and the Algerian French (Français d’Algérie).

Identifying French Muslims from Algeria

7The French administration thus sought to identify the French Muslims from Algeria without ever giving the impression that they were being treated differently [3]. The census administration explicitly demanded these populations be treated exactly like all other French citizens when interviewed by the census agent—the risk of arousing their hostility was too great. Moreover, including a distinction tinged with colonialism in the census of metropolitan France would have been an overly explicit reference to their inequality of status. The distinction could only be made once census taking was completed (see Figure 1).

8During meetings to prepare the census, politicians and senior INSEE officials foresaw the difficulties of enumerating this population owing to their precarious living conditions (‘those without employment or who live in hostels, either legally or illegally’), frequent illiteracy, and poor understanding of French. In the instructions they received, census agents were invited to ‘naturally fill in the census questionnaire themselves’. For these reasons, it was recommended that the agents assigned to areas with large North African populations be ‘among the best’ (Instruction relative to the enumeration of North Africans, 20 April 1954). A later memo even suggested that ‘these population categories be enumerated by census agents who know them well (for example, North African labour force inspectors) and who speak Arabic’ and that ‘the assistance of welfare organizations working with North Africans be solicited’ (Instruction to mayors, 1954 census, CAEF H1207).

Figure 1

Figure 1

Figure 1

Figure 2

Figure 2

Figure 2

9In practice, the census implementation reports revealed the administration’s prejudices against Algerians (one INSEE general director believed them to be ‘a census category far too inclined to slip through the net’) and the porous frontier between census taking and policing, as in Nancy, where ‘a series of police operations were conducted on 20 May [1954], in which 3,000 individuals were enumerated and identified, and a certain number arrested’ (reports of regional divisions, CAEF H1214).

Coding French Muslims from Algeria

10As no specific questions were included in the census, the colonial migrants from Algeria living in metropolitan France in 1954 and 1962 were identified behind closed doors, in the offices of the INSEE regional divisions, when the census bulletins were processed.

11A box on the right-hand side of the individual bulletin completed by the census agent was used to code nationality in response to the question ‘If you are a Foreigner, specify your nationality’ (see Figure 2). For foreigners, a two-digit country code, higher than 10, was recorded. For French citizens, the first digit was 0 and the second distinguished those born in metropolitan France (code 00) from the others. It was at this stage that people born in Algeria were distinguished as either ‘Muslim’ or ‘non-Muslim’. For the censuses of 1954 and 1962, the coding personnel at the INSEE regional divisions were instructed to write 01 when the individual bulletin met the three following criteria: place of birth is Algeria; enumerated person reports French nationality; and ‘he/she has an Arab- or Berber-sounding name and forename’ (Coding instructions, 28 January 1963, CAEF B58372; see Box 2). The coding was based on a list of ‘the most common Muslim forenames found in the civil records in Algeria’ distributed at the time of coding in an appendix to the country codes. Those with a ‘Christian or Jewish’ forename, on the other hand, were coded as ‘French by birth, born in Algeria’.

12This ex post coding stage was simplified by the gradual mechanization of the census in the years following the Second World War. While the challenge—already immense—for the 1946 census was simply to enumerate a population still shifting in response to the aftermath of war (soldiers, displaced populations, refugees), those of 1954 and 1962 aimed to establish a detailed economic portrait of the country in a context of reconstruction and economic planning. To this end, the collected data were coded and centralized to provide tables and maps to the INSEE headquarters in Paris. The specific coding of the French Muslims from Algeria formed part of this ambition and was conducted in violation of the legal imperative not to distinguish populations by origin or religion. However, as the population count thus obtained was always lower than the Interior Ministry’s estimates, the data were largely ignored (memo of 12 May 1958, CAEF H1207).


14From the late 1940s and, subsequently, when the Algerian war of independence intensified, the Algerian migrants in France became a growing concern for the government, and multiple information-gathering operations were organized.

Box 2. Name-based methods in the census

While origins are commonly identified by scholars and researchers via names and/or forenames to make up for a lack of direct information in administrative files or surveys, this method is rarely applied in the census. One example can be found in the American census, in which Mexican Americans, and Latinos or Hispanics more generally (Spanish-speaking people from Latin America, Cuba, or Puerto Rico), were identified via their native language and, in five south-western states (where they were most numerous) by their name. This method was used until 1970, when a question on ‘Hispanic or Latino origin’ was first included in the census in response to the demands of Mexican American associations. As beneficiaries, alongside African Americans and Asian Americans, of the affirmative action policies in favour of ethnic minorities, they wanted access to more reliable statistics [4].

15Because the standard categories used by the statistical system of metropolitan France (nationality and place of birth) were inapplicable in this context, an identification method based on names and forenames (Box 2) was used to distinguish between colonists and colonial subjects living in France. This method was preferred over other means of identification, such as written or spoken language, reported religion, parentage, or self-identification as a member of an ethnic group. For reasons linked to the context of decolonization, this ethnic characteristic was identified by ex post coding to avoid treating colonials differently during the census. Introduced in 1954, when Algeria was still a French colony, the same method was also applied for both the 1962 census, which took place just before the Évian Accords and Algerian independence, and that of 1968. This method of ethnic identification was not abandoned by the public statistical services until several years after the end of the empire.

16The question of ethnic identification and the need for statistical data emerged once again in the late 1990s at a time of public debate on the integration of immigrants and their descendants, and on discrimination linked to origin or religion. However, the issue with respect to 1954 and 1962 is not ethnic categorization per se, but rather the decision to identify a population ex post while asserting elsewhere that it is no different from the others. The purpose of this identification was not to safeguard the rights of these colonial subjects, but rather to monitor their presence in metropolitan France and on the labour market.


At the end of the Second World War, many Algerians came to live in France, taking advantage of strong labour market demand but also (from 1946) of their newly acquired freedom of movement as citizens of the French Union. This migration worried the French authorities, who introduced a series of targeted and often discriminatory measures. One such measure was the identification of these populations in the census, the main source of statistical data on the population of France. In derogation of the law and the egalitarian principles applied in metropolitan France, INSEE applied in 1954 and 1962 a name-based method to identify Algerians present in France, who were widely categorized as ‘French Muslims of Algeria’ despite having a legal status identical to that of French natives. Statistics were thus compiled based on ethnicity and religion, unlike today’s standard census-taking practices.


  • census
  • country of birth
  • nationality
  • immigrant
  • France
  • Algeria
  • French Muslim of Algeria


    • Online[1] Simon P., 1999, Nationality and origins in French statistics: Ambiguous categories, Population: An English Selection, 11, 193–220.
    • Online[2] Blévis L., 2001, Les avatars de la citoyenneté en Algérie coloniale ou les paradoxes d’une catégorisation, Droit et société, 48(2), 557–581.
    • Online[3] Escafré-Dublet A., Kesztenbaum L., Simon P., 2018, La greffe coloniale en métropole: les Français musulmans dans le recensement de 1954, Sociétés contemporaines, 110(2), 35–59.
    • Online[4] Choldin H. M., 1986, Statistics and politics: The ‘Hispanic issue’ in the 1980 census, Demography, 23(3), 403–418.
Angéline Escafré-Dublet
Université Lumière Lyon 2 and French Institute for Demographic Studies
Lionel Kesztenbaum
French Institute for Demographic Studies
Patrick Simon
French Institute for Demographic Studies
This is the latest publication of the author on cairn.
Uploaded on on 04/11/2020
Distribution électronique pour I.N.E.D © I.N.E.D. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait