Multi-dimensional analysis of register variation in Spanish

Northern Arizona University, Flagstaff AZ

Investigators

Abstract

With National Science Foundation support, Drs. Douglas Biber and Mark Davies will conduct two years of research on the linguistic characteristics of spoken and written registers in Spanish (like conversation, lectures, essays, fiction, and academic writing). The research will use corpus-based methods that depend on computer programs to analyze the language characteristics of a large text collection. Corpus analyses describe how speakers and writers actually use language. While previous research has focused on building Spanish corpora, this project is the first large-scale investigation of registers in Spanish corpora. Researchers will use computer programs to analyze linguistic features in a multi-million word corpus and statistical analyses to determine how linguistic features are distributed across the texts in the corpus. The project's main goal is to describe and compare the language characteristics of registers in Spanish. The first step is to describe the characteristics of contemporary registers, and then the project will describe linguistic change in Spanish registers over the past 400 years. Finally, the project will compare the patterns of register variation in Spanish to patterns found in other languages. This research is important for several reasons. One immediate application is to language teaching. At present, most language instruction does not take into account the specific goals of the student (e.g., for conversation vs. for academic study). But because registers differ in their linguistic characteristics, language instruction can be more effective if targeted to focus on the most salient features in a register. The linguistic descriptions resulting from this research will provide information that is relevant for teaching applications of this type. The project will also result in computer programs and a grammatically coded (or 'tagged') corpus of Spanish. The computer program and tagged corpus will be made available for on-site research projects at the NAU Corpus Linguistic Research Program. The tagged corpus will also be available as part of an existing NEH-sponsored web site. This corpus will be the first large tagged corpus of Spanish (or any of the Romance languages) that is freely available online for use by other researchers.