Firing the canon through the computer: Lexical bundles, literary genre and the translation of a best-seller

This paper compares “The Da Vinci Code” and its translation in Portuguese against the language of canonical and non-canonical literary works. Based on the notion of literariness, as forwarded by the Russian Formalists, the study looks at the lexical density of these works. Two corpora are compiled and formulaicity is checked with the help of computers. Classification of lexical bundles is then carried out and seen in the light of socio-cultural differences. The results show that formulaic expressions are used differently in these works and that there are slight changes in the types of lexical bundles in both corpora. This paper contributes to the characterization of the language of literary texts by using empirical observations rather than hermeneutic argumentation. In addition, it brings out the advantages of corpus analysis for the study of literature.