Introduction

This study looks at how speakers of English and German structure space when describing entities such as the layout of a town or village or when giving instructions on how to assemble parts of an object. The cross-linguistic comparison focuses on the types of spatial concepts used to structure space in complex tasks of this kind and how they differ across languages.

We assume that with the definition of a specific communicative task such as a description or instruction, the information to be expressed is not mapped directly from memory into linguistic form (see also Garrod & Sanford 1988, Nuyts 1992). In language production, speakers generate a temporary conceptual structure which focuses a specific set of pragmatic, semantic, and syntactic options and sets guidelines for the process of mapping information into linguistic form.

This conceptual structure consists of a network of abstract conceptual domains such as space, time, objects, events, modality, etc. which allows speakers to establish a coherent frame when locating entities in space and time, when selecting viewpoints on the events related, when specifying their validity and so on (Stutterheim & Klein 1989). How is this temporary level of representation organized? The body of information expressed in a specific communicative task can be treated as an organized structure which answers a specific question, or quaestio. A task which the speaker views as best resolved by presenting information in a narrative form answers the question What happened to x at time t1 at t2, etc.?