Doctoral Dissertation Research: Investigating the linguistic expression of motion

Humans can move through space, detect the movements of other entities, and use language to describe these movements. Linguistic descriptions of motion serve to provide directions, to talk about a trip, and to identify moving objects with respect to their surroundings, including landmarks like forests or rivers. All languages have words for expressing motion, but languages vary widely in their use of additional resources, such as prefixes or suffixes that add further information about motion to a given word. Meanings relating to the expression of motion are also variable across languages and can include quite precise designations – such as motion on an inclined surface or in relation to a river’s flow. Cross-linguistic differences in this domain shed light on the connections between language, cognition, way of life, and environment. This doctoral project trains a graduate student and expands a scientific understanding of the domain of motion expression through language documentation and linguistic experiments relating to motion descriptions. Other benefits to society include the development of a database of annotated recordings that will be archived for others to use, the training of research assistants in aspects of language science research, and the elaboration of materials exploring how motion is represented in language. Moreover, this project supports the advancement of artificial intelligence (AI) technologies through the creation of linguistically grounded language documentation and other machine-readable infrastructure. This doctoral dissertation project addresses the linguistic study of motion events as expressed by a range of lexical and grammatical resources. It does so via study of a language in which the expression of motion is particularly complex and intricate. This research includes attention to how the linguistic resources for encoding motion intersect with other lexical and grammatical resources for representing such notions as posture and path, environmental features such as rivers, and demonstratives as indicators of goals of motion. The methodology includes the recording of spontaneous speech in monologues, conversations, and other speech events that take place in a variety of contexts, including when speakers are themselves on the move, together with recorded experiments involving visual stimuli of motion descriptions. Recordings are made in situ in the language community and are transcribed, translated, and annotated, using software and other AI technologies. All materials are deposited in an archive, where they are accessible as a digital corpus for use by the broader public and researchers in a variety of scientific fields. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.