91

Although statistical surveys involving questions on languages are not common, they have nonetheless developed in recent decades. This article seeks to take stock of the data currently available and reflect on the way “language” as an object is measured by considering what are taken as equivalents and assessed in these surveys. The aim is to understand how “languages” are identified and for what purpose. The goals of such measurements vary from discipline to discipline (statistics, linguistics, demography, sociology, and so on), and they may also diverge from public policies, which tend to favor a method of counting languages (only immigrant languages, for example, with the use of French and/or other languages seen as an indicator of migrant integration). This overview aims to examine the now-available data from all angles, and to make a number of recommendations for future surveys that would provide a fuller picture of linguistic diversity and better understanding of the relationships between languages and adults resident in France. In particular, these include highlighting the importance of drawing on ethnographic work when developing future questionnaires with sections on languages.