While a corpus linguistic technique has been applied to various studies in text and discourse analysis, it has not been much adopted in stylistic analysis of literary texts. The present study, therefore, applies a corpus-driven approach to Jane Austen’s six major novels, in order to see how well this new method works with literary texts, compared with what has been observed in previous studies of Jane Austen’s language. It has been found that the corpus-driven approach can provide quite a few results that are useful in supporting and refining literary scholars’ intuitive observations on the author’s works. Some of the linguistic patterns derived from the comparative corpus-driven method have not been remarked on before in any previous studies and hence can serve as new textual evidence in the study of Jane Austen’s writing style. Despite such great potential for the study of style in literary works, it is suggested that the analyst’s knowledge and understanding of the text(s) under study is crucial in interpreting and evaluating those results because the corpus-driven approach to literary texts relies heavily on quantitative data.