This paper is concerned with the linguistic landscape in urban Gambia. It reviews recent work done on linguistic landscapes and explores the relation between The Gambia's social and ethnolinguistic diversity and visible linguistic phenomena in the public space from an ethnographic and social semiotic perspective. It is argued that the occasional use of local languages in an otherwise English-only environment (as, e.g.,
1The linguistic landscapeIn the past decade or so, linguists and other social scientists have turned their interest to visible linguistic phenomena in the public space. The object of these studies can be identified as the linguistic landscape. Most authors concerned with this field of research acknowledge that the concept of 'linguistic landscape' was coined by Landry and Bourhis (1997) in a psycholinguistic study of ethnolinguistic vitality: 'The language of public road signs, advertising billboards, street names, place names, commercial shop signs, and public signs on government buildings combines to form the linguistic landscape of a given territory, region, or urban agglomeration ' (1997:25). The linguistic landscape here is a psychological factor among other correlatives influencing language attitudes and the (perceived) ethnolinguistic vitality of one language in the presence of another. Although Landry and Bourhis make clear that the background for their study is the situation of French-English bilingualism in Québec, they give no description of an actual linguistic landscape. Their work is less sociolinguistic than it is social psychological. This makes this early work on linguistic landscape of limited interest for an ethnographic sociolinguistic project. More interesting in this respect are the articles in Gorter (2006) and Backhaus (2007) in which the concept is further developed and coupled with a descriptive ambition. For instance, detailed quantitative descriptions are given of the relative presence of Hebrew, Arabic and English in the streets of Israel (Ben-Rafael et al., 2006), of the relative public visibility of minority languages Basque and Frisian alongside national (Spanish and Dutch) and international languages (English) in the Basque country and Friesland respectively (Cenoz and Gorter, 2006), and of the uses of multiple scripts and languages in Bangkok (Huebner, 2006) and Tokyo (Backhaus, 2006).These studies indeed open a 'new approach to multilingualism' (Gorter, 2006) and introduce several interesting concepts (e.g. the distinction between government-issued 'top-down' signs and local, often commercial 'bottom-up' signage). However, they theoretically remain somewhat unsophisticated and 'positivistic' in the sense that they are primarily concerned with counting the occurrences of different languages in a multilingual ecology in order to measure linguistic diversity or evaluate ethnolinguistic vitality -a clear legacy from the field of social psychology from where the term was borrowed. Beyond statistical assertions of the kind, 'In neighbourhood X, n % of signbo...