Psychological verbs in Spanish: Structure and acquisition

The structure of psychological verbs has been a controversial issue because of the apparent mismatch between their semantic structures and their syntactic configuration. Most proposals have tried to explain the relationship between semantics and syntax by positing a semantic structure for the verbs based on a list of semantic roles.; In this dissertation I demonstrate the inadequacy of semantic roles to deal with the facts of psychological verbs. Following Pinker (1989), I propose a semantic structure based on a limited number of semantic primitives that make up what is called the Lexical Conceptual Structure (LCS) of the verb. From there, a linking rule maps the arguments of the verb into syntactic positions. These linking rules are sensitive to the position of the argument in the LCS and not to particular semantic roles.; Based on this model I distinguish three classes of psychological verbs in Spanish represented by the verbs temer, 'to fear', preocupar 'to worry', and gustar 'to like'. I show how the LCS captures the semantic similarities among the three classes while at the same time it explains and predicts the different syntactic behaviors. This model explains, for example, why a verb like preocupar can enter into a syntactic variation between an accusative object (la molesta) and a dative one (le molesta ), while a verb like gustar cannot. It also predicts when the speakers will choose to use each of the two possibilities.; The acquisition of some psychological-verb structures is also problematic for learners of Spanish as a second language. The model proposed in this dissertation predicts what structures will pose more problems for acquisition and explains those facts based on the relative markedness of each LCS. The final part of the dissertation includes three studies that compare the effectiveness of two different ways of teaching these structures. I prove that a teaching approach such as Input Processing, proposed by VanPatten (1996), that focuses on the interpretation of the input is more beneficial than the traditional approach that emphasizes production.