Editorial Note: This essay was originally published on the SOF website on February 21, 2011. It has been revised and updated and may be cited as: Shakespeare Oxford Fellowship, “The Stylometric Debate Over Authorship” (2011, rev. 2021), with a link to this page.
Scholars have long disputed what “stylometry” or “stylometrics” may tell us about the authorship of Shakespearean and other writings. These terms refer to the systematic study and statistical analysis of various features of literary style, in order to draw inferences about authorship. It’s a complex and difficult subject. Anyone diving into it may easily get lost in the technical jargon.
The basic appeal of stylometrics for the Shakespeare Authorship Question (SAQ) is obvious. If we could identify sufficient known and undisputed bodies of writing by various authorship candidates, stylometry could in theory help us assess how likely it is that a given candidate wrote works published as by “Shakespeare.” For a general overview, see Ramon Jiménez’s 2011 essay, “Shakespeare by the Numbers: What Stylometrics Can and Cannot Tell Us” (reviewing Craig & Kinney, Shakespeare, Computers, and the Mystery of Authorship, 2009). See also this 2011 article by W. Ron Hess and this 2017 book review of Oxford University Press’s New Oxford Shakespeare (an ironic title, Oxfordians note with amusement).
As suggested by that key qualification (“known and undisputed bodies of writing”), extremely difficult problems arise right off the bat in applying stylometry to the SAQ. The traditionally anointed candidate — William Shakspere of Stratford-upon-Avon — is effectively exempt from stylometric scrutiny. That is due to one of the central and most puzzling historical facts giving rise to the SAQ in the first place: We have no other known writings by Shakspere of Stratford (not so much as a letter survives) to compare with the “Shakespeare” canon, whose authorship is the very point in question.
No one needs to take the word of the SOF or any authorship skeptic for that basic point. Sir Stanley Wells, staunch Stratfordian and one of the leading Shakespeare experts in the world, has conceded that during Shakspere of Stratford’s lifetime, “despite the mass of evidence” we have relating to mundane business and legal transactions, “there is none that explicitly and incontrovertibly identifies” him as the author (Shakespeare Beyond Doubt, Edmondson & Wells eds. 2013, p. 81).
Shakspere’s will, which he presumably dictated, appears to be the only significant writing directly and personally linkable to him. But Stratfordians don’t seem to think his will bears any resemblance to “Shakespearean” writing or merits any stylometric study — or apparently any study at all. Laughably, the Edmondson-Wells book ignores it altogether. And no wonder. On multiple grounds, the will appears extremely difficult to reconcile with the Stratfordian authorship claim, as Bonner Miller Cutting has shown (among other scholars).
With Shakspere of Stratford thus conveniently set aside, Stratfordian stylometrists have been free to fire potshots at Edward de Vere (Earl of Oxford) and other alternative authorship candidates. But despite decades of trying, and many pages of dense and jargon-laden arguments, they have failed to eliminate Oxford.
Most notably, Ward E.Y. Elliott (a political science professor) and Robert J. Valenza (a mathematics professor), both at Claremont McKenna College, co-authored articles in 2000, 2007, and 2010, arguing that stylometric comparisons of the Shakespeare canon with certain writings by Oxford ruled him out as a Shakespearean candidate. Their articles were based largely on stylometric studies conducted during the 1980s and 90s by Claremont McKenna’s “Shakespeare Clinic.” Elliott & Valenza also co-authored an article on the subject in the Tennessee Law Review, v. 72, p. 323 (2004) (part of an important symposium on the SAQ, not freely available online but accessible in law libraries and fee-access legal databases).
Some mainstream Stratfordian scholars, such as Professor Donald Foster of Vassar College, have harshly criticized such stylometric studies and claims. In a 1998 article in the scholarly journal Computers and the Humanities, Foster noted “doubts about accuracy, validity, and replicability that have dogged the [Claremont McKenna] Clinic’s work from the outset.” As summarized in an abstract of the article, “Foster details extensive and persistent flaws in the Clinic’s work: data were collected haphazardly; canonical and comparative text-samples were chronologically mismatched; procedural controls for genre, stanzaic structure, and date were lacking.” (Foster later changed his view, but his original concerns remain valid.)
On the limited nature of the Oxfordian comparison writings cited by Elliott & Valenza, in terms of scope, genre, and dating (merely a fraction of Oxford’s juvenilia, a small number of poems and song lyrics), see also, e.g., “Oxford’s Known Poetry as Juvenilia.”
The Elliott-Valenza arguments were systematically refuted in two major articles co-authored by independent scholars John M. Shahan and Richard F. Whalen in 2006 and 2009 (and a short 2010 reply). Shahan & Whalen noted that the available and conceded Oxfordian writings are an inadequate comparison sample in terms of scope, genre, and dating.
For still another refutation of Elliott & Valenza, focusing on Shakespearean parallels in one early Oxford poem, see Robert R. Prechter’s 2012 article in The Oxfordian. An early critical commentary on the Claremont McKenna Clinic studies is by Peter R. Moore, originally published in 1990 and available in the 2009 collection of his essays, The Lame Storyteller, Poor and Despised (see pp. 282–87, 345).
Shahan earned degrees in psychology and public health from UCLA and worked as a public health expert for the California state government. He is perhaps best known as the principal author of the Declaration of Reasonable Doubt and as co-editor of Shakespeare Beyond Doubt? (the book that responded to the 2013 Edmondson-Wells anthology). Whalen, a prolific author, earned degrees from Fordham, the Sorbonne, and Yale Graduate School and worked professionally as a reporter and editor for the Associated Press and as a communications executive for IBM. His 1994 book remains one of the most powerful and concise summaries ever published of the evidence against the Stratfordian theory and in favor of the Oxfordian theory.
Academic journals and institutions devoted to Shakespeare studies have almost uniformly censored anti-Stratfordian arguments from their publications and conferences, and have generally shied away from the authorship debate. By contrast, the SOF scholarly peer-reviewed journal which published the Elliott-Valenza and Shahan-Whalen articles linked above — The Oxfordian — has often published Stratfordian defenses of Shakspere and attacks on the Oxfordian and other anti-Stratfordian theories. As Elliott & Valenza graciously noted (2007, 142), it is “to the credit of the Oxfordians … that The Oxfordian has been open to both sides of the debate.”
