‘The Limitations of Stylometry’: Unconscious Versus Conscious Use of Frequent Words

Abstract

A PREVIOUSLY published note, ‘Titus Andronicus: Peele, Shakespeare, Peele’ shows that the final scene of the play was composed by the same author as the first scene, George Peele.1 Table 1 of the note shows that the first 5,000 words and the last 4,854 words of Titus Andronicus have similar relative frequencies of and. These two sections of the play are sorted in the table between the two whole plays, Peele’s David and Bethsabe and The Battle of Alcazar, that have the highest percentages of and out of the forty tabulated plays. While the first 5,000 words of Titus are now accepted as Peele’s, the last 4,854 have been in the past ascribed to Shakespeare along with the ten thousand words of the intervening body of the play. Macdonald Jackson chose and as a criterion for determining authorship in Titus Andronicus, later endorsed by Brian Vickers.2 Also among Hugh Craig and Arthur Kinney’s fifty-five best marker words for distinguishing Shakespeare and Peele, the words and and thy were at the top of the list ‘of the words that Peele uses more than Shakespeare’.3