What was in Jean Cantineau’s Questionnaire for His Geolinguistic Study of Algerian Arabic?

International audience In the eld of Arabic dialectology, as it pertains to Maghrebi dialects, Jean Cantineau (1899-1956) is well known for his attempt at exploring the linguistic variation within Algerian Arabic, while he taught and then held a chair at the Faculty of Letters of Algiers from 1933 to 1946. As shown by the first paper — entitled “Géographie linguistique des parlers arabes algériens” — in the series he was to publish on the topic, he viewed this enterprise as a foray into linguistic geography. To summarize his findings made on the basis of the data he, and in some locations his collaborators, collected between 1935 and 1939, he envisioned a linguistic atlas loosely modeled after the one by Bergsträßer for Syria and Palestine and to be published by 1941. It was in fact his atlas on the Arabic dialects in Ħōrân that got published in 1940. As can be pieced together from different sources, an illustrated study on Maghrebi Arabic that included 150 maps was completed by Cantineau, but never published. We had the opportunity to access a privately-owned remnant of Cantineau’s notebooks from his eldwork in Algeria. The direct result of this is that we can now present the full content of Cantineau’s two questionnaires (one being an abridged version of the other). Cantineau’s preliminary remarks and maps on Algerian Arabic serve as a reference point in time to understand part of the intricate history of a dialect that remains, to this day, insufficiently documented. It is therefore not a trivial matter to understand what the scope of his observations was. Through the study of 9 partially filled-in survey notebooks, this presentation explores the fieldwork done in the vicinity of Medea in 1936. We used the data from the Archives Nationales d’Outre-Mer to geolocate each survey point. Without failing to pay close attention to the regularity with which the various entries are treated, we attempt at generating a picture of the linguistic variation this small collection of data exhibits. The results are then used to assess how much information was (necessarily) left out of Cantineau’s series of papers.