The Invention of Style

Abstract

Style is a rare word in Middle English. It has long been assumed that the concept of a writer's individual or personal style only developed during the early modern period. This paper traces the emergence of style in English writing from the fourteenth to the sixteenth century, performing textual analysis at both the macro and the micro level by using computer software (Voyant; Stylo for R; AntConc) in tandem with traditional close reading. Two major databases were deployed: first, a collection of 279 Middle English digital texts assembled by my research assistant, Imogen Forbes-Macphail, and second, the Middle English Glossarial Database created by Professor Larry Benson, which includes lemmatized texts of Chaucer and Gower's English corpora. The results show that the literary sense of "style" is introduced to the English literary tradition by Chaucer, by way of Petrarch, and then more fully explored by Lydgate, especially in his Fall of Princes. Using stylometry software (Stylo for R by M. Eder), the essay shows in a series of graphs how Chaucer and Gower's styles are distinct from one another, using principal components analysis, cluster analysis, a bootstrap consensus tree, and network analysis; these graphs also show a clear distinction between Chaucer's verse and his prose. I suggest that the difference between Chaucer and Gower is related to these writers' explicit gestures toward the "high style" (Chaucer) and the "plain style" (Gower). The final section of the paper returns to the larger question of style as the word appears in the Forbes-Macphail database, investigating more closely the large spike in usage that appears in the fifteenth century in Lydgate's poetry. By examining his use of personal pronouns and adjectives in conjunction with style, I show that Lydgate pioneered the notion of a writer's personal style, in contradistinction to the rhetorical levels of style (high and low) to which Chaucer and Gower refer.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 3 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.