Abstract

Conventionally, the judgment of the degree of linguistic complexity in a text relies much on the subjective interpretation of the readers. Besides being less generalizable across readers, the subjective approach has to remain limited to only a limited number of texts that human eyes can scan through. This study investigated the issue of linguistic simplicity of the poetical works of William Wordsworth over the ones of Alexander Pope by means of some computational tools for text analysis. In trying to investigate if Wordsworth’s language was simpler, this study took a corpus consisting of a larger chunk of text than usually taken for subjective analysis; it consisted of the poems from the Lyrical Ballads and The Prelude by Wordsworth, and a collection of poetical works by Pope. The works of Wordsworth and Pope were then compared using a number of computational measures: the extent of overlapping vocabulary, type-token ratio, word frequency, hapax-token ratio, word recycling rates, and so on. Results indicated that the language of Wordsworth was not significantly simpler than that of Pope.

Key words: Romantic poetry, text analysis, computational measures, digital humanities, linguistic complexity.