Representativeness in corpora of literary texts: introducing the C18P project

Currently there are very few specialised corpora of literary texts that are tailored to the needs of literary critics who are interested in corpus stylistic analyses of prose fiction. Many existing corpora including literary texts were compiled for linguistic research interests and are often unsuitable for corpus stylistic purposes. The paper addresses three of the main problems: the absence of labelling of the texts for literary genre, the use of extracts, and the prevalence of linguistic periodisation schemes. C18P is a corpus of prose fiction designed specifically to address these issues. It traces the early development of the novel from 1700 up until the Victorian era. It can, for instance, be used for an analysis of the characteristic linguistic features of individual literary genres and forms. The following paper introduces the design of the corpus as well as some of its potential uses