ABSTRACT
After three decades of research on the expression of space in language, the time has come to take stock of the analyses and proposals that have emerged to explain the functioning of static spatial prepositions and, more generally, adpositions (including cases). 2 This is the goal we pursue in this chapter, chiefly examining the semantic aspects of spatial adpositions. We begin with an overview of the most significant studies in this field, highlighting two main ways of tackling the semantics of locative markers: one is essentially geometrical, and usually involves the notion of region, whereas the other lends greater importance to the functions of entities. In the section on prepositions/adpositions as relations, we then demonstrate that the arguments of spatial prepositions are not given the same status in these two frameworks, which therefore differ in their more or less relational character. Drawing chiefly on typological and crosslinguistic data, the section on typological issues and the ensuing discussion suggest that an approach combining function and geometry (in particular, inclusion within a region) is more accurate than a point of view that only takes one of these two aspects into account. As we will see, this type of approach allows us to gain a much more subtle and contrasting picture of static spatial adpositions in language.
