Doctoral Dissertation Research: On Space in Conversation: Coordinating Perspectives in Spatial Reference

In order to successfully communicate, speakers use a set of tools to assess what their interlocutors know. Previous research on these tools suggests that speakers may assume that hearers have knowledge and abilities similar to their own. They may also do what they can to make the conversation easy to follow for all of the participants involved. This is particularly important in conversations about where objects are located. Understanding a description of a location often requires inferring the perspective from which it has been given. It has been found that speakers often use their hearer's perspective when giving a spatial description, for example the hearer's 'right' and 'left'. However, there are languages and linguistic communities that do not use terms such as 'right' and 'left', at least not in the same ways that English speakers do. Instead, speakers of such languages prefer to use descriptions with a perspective derived from the environment, for example from local mountains. These descriptions require calculations to take place in the minds of both the speaker and the hearer, possibly making them more difficult, but they have the benefit of being independent of the interlocutors' perspectives, evading such questions as whether 'right' refers to the speaker's right or the hearer's right. In an increasingly globalized world, differences in how linguistic communities describe the location of objects and make inferences about the perspectives from which descriptions are given can lead to miscommunications. This dissertation will examine for the first time how preferences for using environment-centered descriptions interact with how speakers assess the knowledge of hearers in conversation. To do this, research will be conducted on an indigenous language of Mexico called Tseltal Maya, in which speakers prefer to use the local landscape to describe the locations of objects. For comparison, data will also be collected from English speakers, who have been shown to prefer speaker-centered or hearer-centered descriptions. Both Tseltal Maya and English speakers will participate in a series of tasks in two participants work together to rebuild a scene using objects that have been provided for them. The tasks are designed to generate linguistic data that is rich in spatial descriptions. The aim of this project is to better understand how community preferences interact with human desires and abilities to facilitate understanding of spatial descriptions in conversation.