A bridge between statistics and literature: the graphs of Oscar Wilde's literary genres

The availability of computing devices and the proliferation of electronic texts (the so-called 'e-texts') in centres for literary and linguistic computing in major universities have encouraged non-traditional applications of statistics. With the drudgery of computation and text encoding diminished, research in the field of computational stylistics is accelerating. In this paper, it is shown how projections onto the Cartesian plane of 25-dimensional vectors related to the frequency of occurrence of 25 prepositions can distinguish between Oscar Wilde's plays and essays. Such an application illustrates that it is possible to find unusual and intriguing examples of how statistics can impinge on unexpected territory.