| Given striking crosslinguistic differences in spatial semantics (e.g., Talmy, 1983, 1985), and maturational effects in other closed-class linguistic systems such as morphosyntax and phonology, this dissertation considers the roles of maturation and input in acquisition of L2 spatial semantics. Participants were 60 non-native English speakers, who had spent at least 5 years in the U.S. and had attended college or graduate school in the U.S. They were divided evenly by age of immersion in English (0--7, 8--13, or 14+), and L1 (Spanish or Korean). The task was to view spatial relationships (support, containment, vertical displacement), describe each in English, then rate predetermined descriptions (in, on, over, or under). Results indicated a decline in spatial semantic competence with increasing age of immersion, offering support for maturational changes: Regardless of L1, later learners had more difficulty than earlier learners, particularly with relationships involving functional information (i.e., information based on the functions that objects perform rather than pure geometric relationships between objects). The possibility of a critical period for spatial semantics is discussed and alternative explanations are reviewed, suggesting ways in which spatial semantics resembles other aspects of linguistic structure. |