The purpose of this dissertation is to examine the communicative goals that are accomplished by eight types of figurative language: hyperbole, idioms, indirect requests, irony, meiosis, metaphor, rhetorical questions, and simile. Two experiments that address this issue are described. In Experiment 1, a goal taxonomy that includes each of the eight figures of speech was developed. This goal taxonomy indicated that unique figures of speech are used to accomplish unique communicative goals. This finding can be used to broaden our understanding of the functional and theoretical differences that exist between the eight figures. The goal taxonomy was converted into an MDS solution, and was compared to an MDS solution that had been derived from experts. Once these unique goals were established, a second experiment was conducted to determine if readers are deriving these goals from the context that surrounds the use of the figures. In Experiment 2, subjects produced the various figures of speech in contexts that were originally intended for other figures. It was predicted that the similarity of function from Experiment 1 would determine the "interchangeability" of the figures in Experiment 2. This prediction was not supported by the data. However, the figures that were produced in Experiment 2 could also be analyzed in order to determine how often subjects combined the different tropes. A matrix of this "co-occurrence" is presented. The results of these two experiments are discussed in terms of the theories and assumptions that surround figurative language research. (Dissertation Abstracts)