First- versus second-language reading comprehension of literary texts.

摘要

This study compared the English (L1) and French (L2) reading comprehension and inferencing skills of undergraduates enrolled in an introductory French literature course at the University of Illinois at Urbana-Champaign. Forty-two participants read opening chapters of two different contes philosophiques by Voltaire, one in French and the other in English. Literal or referential comprehension was measured using a written recall procedure and inferential comprehension was evaluated via a multiple-choice task and a character analysis task.;Two main conditions were created: approximately half of the students read Text A in English and Text B in French whereas the remainder read the same texts in the opposite language. Using quantitative and qualitative methods, the data were analyzed across and within conditions for: (1) number of idea units and kind of information recalled in L1 versus L2; (2) perceptions characters across languages as well as knowledge sources and kinds of reasoning underlying readers' inferences; (3) relationship between L1 and L2 multiple-choice scores and frequency of responses across languages and items. The French proficiency scores of 24 students were analyzed to assess the contribution of L2 linguistic competence to text comprehension. Each participant also completed two questionnaires designed to elicit data on personal and literacy variables.;A statistically significant relationship was found between L1 and L2 recall scores, but there was also a main effect for language. Students recalled more idea units from the English than the French texts. As for content, dramatic and plot-related events were recalled more frequently than descriptive details. Overall, the recalls reflected little sensitivity to underlying irony whether students read in English or French. Across languages, the character profiles written by students generally reflected a noncritical approach to interpretation. Students tended to rely on surface meaning and general knowledge of human psychology to infer meaning; there was little evidence of use of historical, cultural or literary knowledge. Students' character profiles and multiple-choice responses were more similar than different across languages and suggested that critical competence and conceptual knowledge played a greater role than linguistic knowledge in inferential comprehension.