The Denominator Problem

In 1920, an English schoolteacher named J. T. Looney published a book that changed the Shakespeare authorship debate permanently. His method was genuinely rigorous in one important respect: he derived a profile of the author from close reading of the works before naming any candidate. Eighteen characteristics—classical education, Italian enthusiasm, lyric talent, drama enthusiasm, and so on—assembled from what any careful reader would infer about the person who wrote the plays. Profile first. Candidate second. This is the correct order of inquiry.

Then he made an error.

He found that Edward de Vere, 17th Earl of Oxford, matched the eighteen characteristics. He noted the match. And he stopped.

He never asked the question that determines whether a match means anything at all: How many people in Elizabethan England satisfy this profile?

This is the denominator problem.

Suppose a profile produces one match in the relevant population. Then the match is powerful—it effectively identifies your candidate. Suppose it produces twenty matches. Then the match narrows the field to twenty. Useful, but not an identification. Passing a profile establishes that a candidate cannot be ruled out. It does not establish that the candidate is the answer.

Looney computed the numerator. He never calculated the denominator.

When his eighteen characteristics are applied consistently to all the serious candidates—not just de Vere—several of them, including superior classical education, pronounced literary tastes, enthusiasm for drama, and lyric talent of recognised quality, return positive scores for five or more candidates simultaneously. De Vere is not distinguished by these characteristics. He shares them. Mary Sidney Herbert scores 14 out of 18 on Looney’s own profile. John Florio scores comparably. The profile does not produce a single answer. It produces a short list.

It gets more interesting when you look at the characteristics where de Vere seems to score most distinctively: eccentricity, financial improvidence, a mysterious and unconventional manner. These are also the most biographical characteristics—the most dependent on what Looney already believed about the kind of person who wrote the plays, and the most susceptible to the objection that they describe de Vere’s life rather than anything genuinely derivable from the texts. The characteristics that seem to clinch the case are the ones least securely grounded in evidence.

And Looney’s eighteen characteristics don’t exhaust what the works themselves require of their author. Further constraints are derivable by exactly the same inductive method he used. Henry V contains sustained French dialogue—not tourist phrases, but extended dramatic scenes requiring genuine fluency. Applied to the candidates, this favours Florio (born into an Italian-French exile household, French was a working language), North (translated Plutarch via Amyot’s French original), and Mary Sidney (translated Garnier from French)—and places de Vere at best in the partial column. The medical and anatomical knowledge in King Lear, Othello, and All’s Well That Ends Well—precise enough that medical historians have documented it—does not correspond uniquely to any candidate, with one partial exception: Mary Sidney Herbert’s documented medical knowledge, reinforced by her long association with her personal physician and set out in detail by Robin Williams in Sweet Swan of Avon.

The pattern is consistent. Every additional characteristic derivable from the works either expands the candidate pool or points toward a candidate other than de Vere. None of the natural extensions converge on him uniquely.

What this means is that the author profile, applied honestly to any candidate, not just de Vere, does not solve the authorship question. It sharpens it. It narrows the field from everyone who was alive in Elizabethan England to a handful of serious candidates. That is genuinely useful, but it is not what Looney thought he had done.

The works require an author. The profile identifies several people who fit. Determining which of them actually wrote the plays requires a different kind of evidence: not more characteristics, but a document connecting a specific person to the name “Shakespeare” during their lifetime.

That document has not been found. For any candidate.

That is where the authorship question actually stands.

[This post was written with the aid of Anthropic’s AI LLM Claude.