The Enlarged Kyd Canon: Comments on a Recent Study
MacDonald P. Jackson · Notes and Queries · 2020
Thomas Kyd’s reputation as a playwright rests chiefly on The Spanish Tragedy, but also on Soliman and Perseda, and Cornelia, the last a translation from the French of Robert Garnier. But Brian Vickers has a contract with Boydell and Brewer to produce, with the help of a team of scholars, an edition of Kyd’s works that will include three further plays, claimed as his alone, and two to which he is alleged to have contributed the lion’s share.1 It seems important to scrutinize the methods resulting in new attributions that, if accepted, must revolutionize historical accounts of Elizabethan drama. In a recent article, Albert C. Yang attempts to ‘validate’ the enlarged Kyd canon.2 He believes he has succeeded. I want to explain why I do not. My concern here is solely with that one publication, not with other arguments in favour of augmenting the number of Kyd’s plays, as listed in Yang’s ‘Works cited’. Yang employs ‘an information categorization method that considers rank-order frequency of all the words in texts’, though the method necessarily requires that only the frequencies of words that are shared by pairs of texts to be compared are ranked.3 The technique, which affords a measure of the likeness between texts, has previously achieved some success,4 but in this case, Yang applies it in a way that fails to demonstrate that any one of the candidates for admission to the Kyd canon—King Leir, Fair Em, Arden of Faversham, and the bulk of 1 Henry VI and Edward III—is indeed Kyd’s. The inadequacy lies in the limited range of plays used. These are the dramatic texts of the extended Kyd canon and the thirty-six Shakespeare First Folio plays plus Pericles. Cluster analysis that includes only the six whole plays accepted as by Kyd or conjecturally attributed to him places these on a different branch of a ‘phylogenetic tree’ from Shakespeare’s plays. Yang’s Figure 3 displays this tree. Supposing it gives an accurate picture of the closeness of the relationships between plays—though there are reasons for granting no such supposition, as we shall see—all it shows is that Shakespeare’s plays can be separated from the others. It does not tell us that King Leir, Fair Em, and Arden of Faversham are by Kyd. To provide evidence for that conclusion, Yang would need to have included plays by Marlowe, Greene, Peele, and others, along with (say) the eight earliest Shakespeare plays whose sole authorship is undoubted, and shown that the plays putatively Kyd’s form a separate cluster with The Spanish Tragedy, Soliman and Perseda, and Cornelia.