On the “Whats” and “Hows” of “Where”: The Role of Salience in Spatial Descriptions

Abstract

According to Clark [1] language is a joint activity between speaker and listener, undertaken to accomplish a shared goal. In the case of spatial descriptions, one such goal is for a speaker to assist a listener in finding a sought-for object. For example, imagine misplacing your keys on a cluttered desktop, and asking your friend if s/he knows where they are. In response, there are a variety of spatial descriptions that your friend can select that vary in complexity, ranging from a simple deictic expression such as “there” (and typically accompanied by a pointing gesture), to a much more complicated description such as “its on the desk, under the shelf, to the left of the book and in front of the phone.” Between these two extremes are descriptions of the form “The keys are by the book”, consisting of three parts: the located object that is being sought (i.e., the keys); the reference object from which the location of the located object is specified (i.e., the book) and the spatial term that conveys the spatial relation between these two objects (i.e., by). For inquiries of this type (“where are my keys?”), the located object is pre-specified, but the speaker needs to select an appropriate spatial term and an appropriate reference object. My research focuses on the representations and processes by which a speaker selects these spatial terms and reference objects, and the representations and processes by which a listener comprehends these ensuing descriptions.