Three People Nobody Proposed

Any list of what Shakespeare’s author must have had can be tested in a simple way: try it on people nobody thinks wrote the plays, and see how they score. If they score well, the list is describing something common, and a candidate who fits it has been shown to be one of many.

This site scores each authorship candidate in a table it calls the Convergence Analysis. The table asks nineteen questions about what the works suggest their author knew or did, such as: did he read Italian, did he know the law, did he move at court, was he alive when the plays were written? For each candidate, a question answered yes by documents from the period scores one point; a partial or disputed answer scores half a point; no evidence scores nothing. So a total of 7 means the records answer about seven of the nineteen questions in that person’s favour. Nobody scores near 19, and the highest candidate total is 9.

The site put three well-documented Elizabethans who have never been proposed as the author through the same nineteen questions, scored by the same standard as the candidates: Sir John Harington, the Queen’s godson who translated Ariosto; George Chapman, the playwright who translated Homer; and Henry Howard, Earl of Northampton, a Cambridge scholar and privy councillor.

All totals are approximate, about a point either way, as on the convergence table. The bars run from 0 to 10 points; no one on the table scores higher than 9.

Harington scored about 7, level with de Vere and Mary Sidney Herbert. Northampton and Chapman scored about 5, level with Henry Neville and Thomas North. By the table’s own rule that two totals differ only when they are more than two points apart, it cannot tell any of the three from most of the candidates.

Why this happens

Most of the table’s questions (court life, legal knowledge, Greek, a range of writing, being alive at the right dates) describe what educated Elizabethans commonly had. Anyone with a good education and a well-kept record collects five to seven points before authorship comes into it. A score in that range shows a learned, well-documented life and nothing more about who held the pen.

The lines on which no control scored are the ones tied to specific facts about the works: the Southampton dedications, Florio’s Montaigne, the 1593 cluster, professional music, contemporary testimony. Those are the lines that can separate one candidate from the crowd, and on most of them most candidates score nothing either.

The hard part was finding them

The first choices were Thomas Lodge, Sir Henry Wotton and Lord Mountjoy. All three had already been proposed as the author, or as one of a group, and turned up on a published list of 88 names. When the qualifications are this common, anyone who searches long enough will find someone who fits them, and very nearly everyone who fits has been proposed by somebody.

How this was done, and what it is not

The three controls were chosen, and every mark given, by an AI model working for this site, which knew which three were controls. Under condition 4 on What Would Settle It, choosing controls and applying the rules are judgements for a person with no stake in the result, so this is not a control test in that sense. It shows what such a test looks like and what the table returns when one is tried. Knowing they were controls pushes towards strict marking, so their totals are more likely too low than too high. Three is also a small number: it shows roughly where people like this score, not the full range. A proper run would use ten or more controls, chosen and scored by a person, with the names removed where possible.

The marks rest on standard reference works rather than the primary documents. Corrections are welcome and will be recorded on the errata page.

*   *   *

The general form of this problem is set out on The Denominator Problem. Of Course It Was Him/Her shows the same thing from the other side: the correspondence method, applied to two candidates, produces equal conviction for both.