Is This Site Anti-Oxfordian?
The charge: The site refuses to privilege Oxford even though the works point to him more clearly than to anyone; it rules out the plays, and above all the Sonnets, as evidence of his hand; and it keeps a field of rival candidates alive when the evidence, honestly read, should have narrowed to one. A recent correspondent put it plainly; he found the site “significantly biased against Oxford” and declined to list it on that ground.
The answer begins with a fact that sits awkwardly beside the charge. On the site’s central measure, Edward de Vere leads the field; and on the retired profile that first singled him out as the author (J.T. Looney’s), he scores a perfect mark.
* * *
Where de Vere actually stands on this site
The convergence analysis (the site’s central measure, drawing on methodologically diverse strands of external evidence) places de Vere at the top of its leading band, ahead of Florio and Mary Sidney Herbert. The retired Looney author-profile, scored at its most generous, now returns de Vere a perfect 15 out of 15, ahead of Marlowe at 14½, and a full 18 out of 18 once every criterion is counted. And on every framework the site runs, the man from Stratford scores last: zero across all ten of Diana Price’s documentary categories, and a single convergence point for having been alive during the composition years, the least discriminating mark in the table.
A site built to sink Oxford would not seat him at the head and hand him a clean sweep of the very profile the Oxfordian case began with. The scores are not buried; they are stated plainly on the findings page, in the tables, and on de Vere’s own assessment page. If the aim were to damage his candidacy, this is a strange way to pursue it.
The one place he doesn’t lead, and why the site says so
There is one framework de Vere does not lead. On Price’s literary paper trail, which counts personal primary-source evidence of a writing life, Bacon scores highest at 10 out of 10, with Florio and Mary Sidney Herbert close behind, and de Vere lower. The site records that too, in the same place and the same plain terms as his convergence lead. No single candidate leads both live frameworks; this is the site’s central null result, and de Vere is no exception to it.
This can look like the site withholding a verdict at the last moment. What it is doing is narrower and more consistent: reporting each framework’s result without adjusting it to a preferred conclusion. De Vere wins where the evidence puts him first and places lower where it does not, and the site declines to average the two into a verdict the evidence has not delivered. That refusal is not aimed at Oxford. It is applied to Bacon, who leads Price and trails on convergence, in the same way.
The corrections have gone both ways
The clearest test of bias is not what a site says about itself but which way it moves when a critic pushes. This site keeps an errata page, and the record there is the answer to the charge.
De Vere’s scores have been corrected upward at the request of Oxfordian critics. The most recent instance is criterion 8 of the Looney profile: the site had scored him ◑ on “a lyric poet of recognised talent,” on the reasoning that his verse showed no talent “of the Shakespearean order.” That reasoning was circular; it measured his own-name poetry against Shakespeare’s, which assumes the disputed works are already his, and when the correspondent pressed the point, the mark went to ✓. That single change lifted de Vere to a perfect profile score. The correspondent was right, and the page now says so over his name.
But the same standard has moved de Vere’s marks and his favourable facts the other way whenever a source failed to hold. The convergence line crediting him with pre-1603 access to Florio’s specific Montaigne phrasings was corrected from ◑ to ~ when the cited study turned out to describe post-publication use, not manuscript access, a score lowered against the Oxfordian reading. And the long-repeated claim that Southampton paid de Vere a £5,000 penalty for refusing to marry his daughter was struck as an unsupported rumour with no fiscal document behind it (the penalty, had it been paid, owed to Burghley as guardian rather than to Oxford at all), a favourable “fact” retired even though no score had hung on it. Neither correction helps the Oxfordian case, and both were made anyway.
Oxfordian and alternative-candidate correspondents have prompted a good share of the corrections on that page. That is not the behaviour of a site closed against Oxford; it is the behaviour of a site that treats Oxfordian scrutiny as exactly what it is, scrutiny to act on, and that cuts in both directions when it does.
What the site declines to do, to everyone
If the scores favour de Vere and the corrections run both ways, what is left of the charge? What remains is a set of things the site declines to do, and reading them one by one shows that each refusal applies equally to every candidate, the man from Stratford included.
The site will not privilege Oxford by assertion. It narrows the field by method or not at all: no candidate is ruled in or out by assertion, which is the only arrangement under which Shakspere can be held to the same standard as the rest. The site will not reduce the field to sharpen his case, because reducing the candidate list is a conclusion, and the site reaches its conclusions through the frameworks rather than ahead of them. And the site will not convert the strongest circumstantial case into a proof, for Oxford or for anyone, because circumstantial strength and documentary proof are different things, and the distance between them is the distance the evidence has not yet crossed, for every candidate.
These are not anti-Oxford rules. They are the rules of the method, and they bear on Bacon, Marlowe, Florio, Mary Sidney Herbert and Shakspere in precisely the same way they bear on de Vere. None of them is aimed at Oxford; the rules would read as hostile to whichever candidate a reader had come to defend, because what they withhold is the verdict, and a defender wants the verdict.
Circumstantial evidence is not the poor relation
A particular version of the charge is that the site leans so hard on documentary evidence that it treats circumstantial evidence as second-class, and that this cuts against Oxford, whose case is largely circumstantial. The premise about the law is correct, and the site grants it at once. As Tom Regnier has emphasised in his lectures on the rules of evidence, the law draws no rank between circumstantial evidence and the testimony of an eyewitness, and juries convict beyond reasonable doubt on circumstantial evidence every day. Nothing on this site disputes that, and nothing on it ranks circumstantial evidence below documentary evidence as a class.
The line the site actually draws runs somewhere else. It is not about circumstantial versus documentary evidence; it is about whether a piece of evidence singles one candidate out or merely fits him. A jury convicts on the circumstantial fact that points to the defendant and away from everyone else, the fingerprint on the weapon, the tyre-track that matches one car in the county. It does not convict on the circumstantial fact that the defendant owned a coat the colour of the one a witness saw, because that fact, however true, fits half the town. Both are circumstantial. Only one narrows the field, and the narrowing is the whole of the evidential work.
Most of what the authorship debate offers as circumstantial evidence is the coat, not the fingerprint. That Oxford travelled in Italy, knew the law, kept falcons, read widely and moved among courtiers is all true and all circumstantial, and all of it fits many well-read, well-travelled noblemen of the period just as snugly. It is consistent with his authorship; it does not discriminate his authorship from the alternatives, because it does not discriminate him from a dozen other men. Consistency is a real result, since it rules out the inconsistent, but it is a weaker one than identification. Consistent facts of this kind do not accumulate into identification the way a stack of independent clues can, because each fits the rivals as well as it fits Oxford, and so leaves the field exactly as wide as it found it. Fifty facts that every candidate satisfies do not multiply into one that only Oxford does.
Where circumstantial evidence does narrow, the site weighs it fully, and this is the part the charge overlooks. The whole purpose of the convergence analysis is to credit the discriminating circumstantial strands: Stritmatter’s annotations in de Vere’s own Geneva Bible, Roe’s on-the-ground Italian topography where it outruns the printed sources, North’s source-text fingerprint. Each of these, if it holds, speaks to the narrow population who could have supplied a specific input rather than to biographical resemblance, which is what makes it the discriminating kind of evidence, whatever one finally concludes about any single case. The site does not discount circumstantial evidence; it grades it by whether it narrows the field, which is the question a court asks too, and by that measure Oxford’s best circumstantial evidence is credited, not dismissed. The distinction, and the handful of arguments on the site that meet it, are set out on the findings page.
None of this smuggles documentary evidence back above circumstantial. A circumstantial case can close a question when it excludes the alternatives, exactly as a court’s can; the reason this one has not is that the circumstantial evidence on offer overwhelmingly fits the rivals too, not that circumstantial evidence is the weaker kind. Where this page later names a document as the thing that would settle the matter, it names the clearest discriminator, not the only admissible one; discriminating circumstantial evidence, if it ever excluded the rivals, would do the same work.
The works, and the Sonnets
The hardest form of the objection is that ruling out the works as evidence (and the Sonnets above all, the author’s only openly personal writing) is itself an act of bias against the candidate the works most resemble. Answering it takes more care than the rest, and it is where the site’s reasoning is easiest to mistake for hostility.
The site does not rule the works out as evidence. It declines to treat “the works fit my candidate” as an identification, because the same method, applied by a Baconian or a Marlovian with equal care, delivers each of them their man with equal conviction. A work this large and varied mirrors any sufficiently rich life; the reflection arrives for whoever stands in front of it, which is exactly why the reflection cannot settle who was standing there first. That the Sonnets fit Oxford is not in dispute. That they fit Oxford and fail the alternatives, the thing that would make them evidence rather than resemblance, is what no reading has yet shown.
This bears on biographical resemblance, not on every use of the text. Where the works preserve a specific external input (a book a candidate demonstrably marked, a place rendered in detail beyond the printed sources), that is not the mirror but a fingerprint, and the site credits it, as the circumstantial section showed. Resemblance to a rich life fits any rich life; a recoverable input narrows to whoever could have supplied it. The Sonnets read as self-portrait are resemblance, which is why they do not identify, and why the site weighs them differently from a marked Bible.
The strongest concrete form is the procreation sonnets, 1 to 17, read as urging Southampton to marry, which Oxfordians tie to the proposed match with Oxford’s daughter Elizabeth Vere and press as the one rationale no rival can supply. It is the best of the resemblance arguments, because it reaches for a documented circumstance rather than a mood; but it still turns on reading the young man as Southampton and the urging as a father-in-law’s, both inferred from the poems rather than drawn from a record. Granted in full it shows a motive consistent with Oxford, not one that fails every rival; the reading of the Sonnets is taken up on its own page.
The strongest Oxfordian use of the Sonnets is not the resemblance argument at all but the eliminative one: that the 1609 dedication and certain sonnets fix the author’s death before publication, cutting away every candidate who outlived it. That is a better kind of argument, because it tries to narrow the field rather than merely to fit it. But it fixes the date only on a biographical reading of the poems and of the dedication’s “ever-living”, the very phrase whose meaning is contested, and the site cannot take a contested reading of a poem as a documentary fact. Not for Oxford, and not against him. The same rule that stops the site asserting Shakspere wrote the works stops it asserting the author was dead by 1609. Held in one direction that rule clears the board toward Oxford; held in the other it does not; and the site holds it in both, which is the whole of what “even-handed” means. The 1623 First Folio is a separate matter, and a firmer one, taken up with the endpoint below: it does not read a phrase, it records a death.
The play dates cut both ways
A related complaint is that the site accepts the orthodox dating of the plays, the conventional dates that place several works after Oxford’s death in 1604, and so quietly stacks the chronology against him. It does not. The site refuses composition-dating altogether, in both directions. The orthodox composition dates are inference from topical allusion and source use, not documentation; on the primary record no play’s composition is fixed to a particular year (Kevin Gilvary’s dating audit), and a chronology inferred that way excludes no candidate, Oxford included. What the site does accept, from either side and wherever it falls, is the documentary record of when a play entered the world (a court performance, a Stationers’ Register entry, a first quarto). Several of those records do fall after 1604; the site takes them at face value, and they are exactly what the endpoint constraint below rests on.
The same discipline falls on the Oxfordian early dates, and here a question presses: why should Meres’s 1598 list count as documentary record while the 1589, 1594 and 1596 references to a Hamlet count as inference? Put that way, they don’t. Each is a firmly dated document, and each carries the same kind of fact, that a work of that name was in existence by the date of the mention. The site takes that fact from any of them. Where the two part company is not in their dating, which is sound on both sides, but in what the named work can be shown to be.
Meres names his plays unmistakably as Shakespeare’s, so 1598 is a firm terminus for the plays he names: those existed, under that name, by then. The Hamlet notices name only “Hamlet,” and whether the play meant is the one we have or the lost earlier tragedy often ascribed to Kyd is exactly what is in dispute. As a terminus for a Hamlet the notices are sound; as a terminus for Shakespeare’s Hamlet they hold only if that identification holds, and the identification is an inference, not a record. That is the whole of the difference, not that one source is documentary and the other is not, but that Meres’s work is identified and the early Hamlet’s is not. Ramon Jimenez’s wider early-authorship case is built from links of the same shape: strong if the identification holds, silent until it does. The site refuses the orthodox late dates and the Oxfordian early ones alike; there, a topical allusion read as a date of composition; here, a contested identification read as an established one.
So the site works from what these documents can carry without an added inference: the record that a titled, identified work had entered the world by a given date (Meres in 1598 at the front, the canon’s close around 1613 with the 1634 attribution of The Two Noble Kinsmen to Fletcher and Shakespeare at the back). That endpoint is a genuine, two-sided constraint. It excludes no candidate, but it asks any who died before the canon closed, Oxford among them, for a mechanism by which the last works were finished and released, and the scale of that machinery grows the earlier a candidate died. The site states the constraint rather than assuming it away, and it states it the same for everyone.
The same record closes the window at the other end. The 1623 First Folio, in Heminge and Condell’s dedication, states that the author was already dead, “by death departed from that right” of setting forth his own works. Unlike the 1609 “ever-living”, this is not a reading to be argued but a plain documentary statement, of the same kind the site accepts at the front of the window; and it constrains the candidates who outlived it exactly as the canon’s close constrains those who died before it. A candidate still living in 1623 (Bacon, Stanley, Florio and Emilia Bassano among them) must hold that the Folio’s attestation names the front-man rather than the true author: a concealment sustained past the candidate’s own death, whose scale grows the longer the candidate outlived 1623. That is a genuine constraint, and the site marks it where it marks the mirror one, on the convergence line for chronological compatibility; those candidates carry a half-mark there, as the candidates who died too early already do. It crowns no one: the 1623 terminus does not reach Oxford, who died in 1604; it is his rivals who outlived it, not he, and it leaves untouched the half-mark he already carries for having died before the canon closed. And it excludes no one; it states the same documentary fact for everyone it touches.
Why a neutral method reads as hostile
There’s a good reason the site can seem to bear down on Oxford harder than on anyone else: there is more of him to bear down on. The Oxfordian case is the oldest of the living alternatives and by a wide margin the most developed, a century of books, biographies and catalogued parallels since Looney named de Vere in 1920. The volume of analysis tracks the size of the case, not any animus toward the candidate, if anything it is a measure of how seriously the case is taken. The thinnest candidacies on the site draw the least comment, and no one mistakes that for favour.
A method that privileges no candidate will read as hostile to any reader who has already settled on one, because it withholds the verdict they have reached, and it withholds it from all alike: anti-Oxford to the Oxfordian for the same structural reason it is anti-Bacon to the Baconian. The refusal falls identically on the candidate a reader opposes and the one they favour, and lifts for no one until the evidence lifts it.
A further sign, though not by itself a proof, is who else objects. The same site that Oxfordians call anti-Oxford is treated by orthodox scholars as a doubters’ site, too willing to entertain the alternatives, too ready to score the Stratford man last. A page that gives Oxford the top convergence score is not the natural ally of the Stratfordian; a page that will not declare Oxford the author is not the natural ally of the Oxfordian.
The test that settles it
There is a way for any reader to check the charge for themselves, and it is the same test the site puts to every position, including the orthodox one. Imagine the non-fraudulent document that would settle the question (a probate inventory, say, professionally authenticated, listing the foul papers of a play in a named hand) and ask what you would do as the name is filled in. If the name is de Vere, you accept it without hesitation. The question is what you do when the name is Marlowe, or Mary Sidney, or Shakspere of Stratford: same paper, same ink, same provenance, only the name changed. If you would accept it when it names your candidate but doubt the identical document when it names another, then it is the name deciding for you, not a principle.
That direction-blind test (set out at greater length for advocates of every candidate on the Proposal for Advocates and in short form on the Self-Check) is the standard this site tries to hold. Applied to Oxford it looks like reluctance; applied to Shakspere it looks like doubt, but applied consistently it is neither. It is the one discipline that lets the site say Shakspere left no personal documentary trace of a writing life, the thing Price’s ten categories measure, and, in the same breath, that no alternative has closed the case, because it refuses to count for Oxford what it will not count against Shakspere.
* * *
One thing easy to lose in a dispute about scores is the thing that matters most. Every exchange of the kind this page has answered (his mark should be higher, this candidate leads that one) is a contest over ranking.
But ranking is not proof.
Proof will be twofold: a primary source documenting who composed the works, and an account of Shakspere’s own part in it, why the man from Stratford left no personal documentary trace of a writing life, and what, if anything, he had to do with the writing. Neither has been found, for de Vere or for anyone. Until both are, the ranking stays a ranking, which is not a hollow thing: it states where the evidence leans, and first place means that the day a document settles this, de Vere is the likeliest name it will carry. What the site will not do is read a lean as a proof, or let the top of a list stand in for the document that is missing. On the ranking, in the meantime, Oxford is a leading contender.
So: is the site anti-Oxfordian? It ranks Oxford the strongest alternative candidate on the evidence it can assess, corrects his scores upward when a critic is right and downward when a source fails, and still declines to call the question closed in his favour, because it declines to call it closed for anyone. If that reads as hostility, it is only the caution of a method that will not reach a verdict without evidence. When that evidence comes, it will move the site for Oxford as readily as for any other name. Until then de Vere, at the top of the list, gets the same verdict as everyone below him: not proven.