How Ideology is reflected in The Time Machine: A Corpus-based Approach

Much has been accumulated in the research on science fiction, corpus method to literary works, and critical discourse analysis on literary works, while research concerning the combination of these three elements is just beginning. The present study is a case study for examining how a corpus-based approach can combine with CDA and contribute to research on literary works. Specifically, Lancaster Semantic Analysis System (USAS) is firstly used to perform semantic encoding for the text of H.G. Wells’ science fiction The Time Machine. Then the encoded text is imported into Sketch Engine, the ultimate tool to explore how language works. Second, the word list and the keyword program are used for word filtering. The filtered words are then divided into 3 categories, namely, character, environment, and psychology, according to different descriptive aspects. Third, the distribution and collocation of object words in different categories are tested by the sketch engine programs or USAS. Finally, CDA is carried out on these data in combination with the time of the text. Findings from the study have shown that language in The Time Machine is ideology-loaded, characterized by the distinctive modification of different characters, the vagueness of the psychological process, and the diversity of narrative perspectives. In response to scepticism of quantitative stylistics from literary critics, this paper serves to reinforce the literary value of simple quantitative text and corpus data. At the theoretical level, this study tries to explain how micro textual resources can interface with macro discourse, such as ideology and social cognition. At the methodological level, this study promotes the application of the combination of corpus linguistics and critical discourse analysis in stylistics.