ABSTRACT

Observations about the increasing volume and diversity of textual material available for humanities and social science research are at the tipping point of cliché but nevertheless point to exciting and challenging opportunities. The focus in this chapter is on the use of automated techniques to analyze the content of textual data sets, known as corpora, related to climate change-i.e., to provide manageable overviews of the content, to highlight interesting linguistic patterning, and to automatically annotate, or code, the material. It argues that the scale, complexity, variety, and dynamic nature of climate change discourses mean that data-driven approaches in particular have an important role in analyzing large text corpora. The chapter presents a brief review of the text analysis techniques that have been used to investigate climate change discourse, especially in news media and social media-i.e., text classification, topic modeling, and various corpus linguistic techniques that have been used for corpus-based discourse analysis.