Doubts about Thomas

Between 1570 and 1642, Elizabethan dramatists sold their manuscripts direct to a theatre company, a simple arrangement but one that has left many problems for posterity. If a company then published the play, either to raise money, or to publicize a revival, they would use the title page to advertise themselves, rather than the dramatist. With rare exceptions, such as Shakespeare, this resulted in many plays being published without the author’s name. Deprived of reliable external evidence, scholars wishing to identify an author have had to search for internal evidence, the play’s language. The survival of unidentified manuscripts raises the same problem.

There are two principal approaches to authorship attribution, which I shall call “open” and “closed”. Open approaches have been used since the 1880s, as scholars have developed methods to identify and discriminate between dramatists, studying their vocabulary, minute features of prosody, and (more recently) their use of favourite repeated phrases. These approaches are empirical. Closed approaches, by contrast, are those where enquirers already know who the author is.

This distinction can be tested by recent studies of a celebrated item in the British Library collection known as “Egerton 1994”, a play lacking a title page and a conclusion, subsequently called either Richard II, Part One or Thomas of Woodstock. Derived mainly from Holinshed’s Chronicles, it depicts an earlier stage of that king’s reign, immediately prior to the opening of Shakespeare’s play. Its hero is Thomas of Woodstock, Duke of Gloucester and Lord Protector, uncle to the king, who opposes the king and his corrupting favourites. The embodiment of responsible government, “Plain Thomas” is undermined by the flatterers and murdered, but his death provokes a purging of the favourites. Thomas of Woodstock is a well-structured play, with several striking scenes, such as the boar-hunting masque in which Thomas is the real prey.

On Woodstock’s first publication in 1870, J. O. Halliwell described it as “a composition anterior to Shakespeare’s tragedy”, having noted passages that resemble some in Richard II. In his outstanding 1961 Arden edition of Richard II, however, Peter Ure observed that “There is a very large number of verbal echoes; but we do not really know which play was written first”. In Richard II Gaunt delivers a celebrated denunciation of an England corrupted by the king and his flatterers:

This land of such dear souls, this dear, dear land
Dear for her reputation through the world,
Is now leased out – I die pronouncing it –
Like to a tenement or pelting farm.

In Woodstock Richard himself juxtaposes this image with his father Edward’s wars in France:

And we his son, to ease our wanton youth,
Become a landlord to this warlike realm,
Rent out our kingdom like a pelting farm.

Who was the innovator, who the imitator?

Scholars used to date Woodstock to the 1590s, preceding the 1597 quarto of Richard II, but in 1983 David Lake showed that Woodstock is the later play. Earlier, in 1970, Lake had published The Canon of Thomas Middleton’s Plays, a pioneering revaluation using a radically new method. As he argued, reliable internal evidence must be “objective, unambiguously defined features, recognition of which is not a matter of opinion”, and “quantifiable, since differences between authors are mostly matters of relative frequency, not of invariable use or non-use”. He also realized that to define the individuality of an author it is essential to compare him with other writers using the same genre in the same period.

In pursuing both approaches Lake developed methods not suited to the light-hearted. He put together a corpus of 132 plays from the period of Middleton’s literary career (1597–1627), including works by Middleton, and subjected it to micro-analysis, counting the occurrence of 201 linguistic items. He showed that Middleton’s linguistic style was unusually colloquial, using a wide range of contracted forms: I’m, I’ve, sh’as (for “she has”), ’tas and ’tad (for “it has”, “it had”), y’are, they’ve and all the common contractions. Middleton preferred ’em to them, has to hath, does to doth, and had favourite exclamations (pish, push, tut, why) and favourite oaths – faith, troth, ’sfoot (“God’s foot”) and cuds me. Over 300 pages Lake compiled tables for every play, meticulously itemizing all significant linguistic minutiae whose presence or absence gave reliable evidence of Middleton’s authorship. This was by all criteria a work of scientific criticism, its results indisputable.

Subsequently Lake reworked his data in terms of chronology, showing that in 1599–1600 there was a marked increase in the use of colloquialisms and contractions in the works of Middleton, Ben Jonson, John Marston and Thomas Dekker. These changes were set out in a table of seventy-seven extant plays from the London commercial theatre, 1585–98. In a separate table Lake gave the comparative figures for linguistic markers that identified the common authorship of Woodstock and Samuel Rowley’s When You See Me You Know Me (1605). Both plays shared an unusually high liking for the pronoun ye, which has 231 occurrences in Woodstock and 220 in Rowley’s play, far more frequent than elsewhere in Elizabethan drama. Both works preferred hath to has (respectively 25:20 and 47:41), both used the contractions i’th’, ’eth and ’oth, and both liked the oath ’sfoot (“God’s foot”), Woodstock fourteen times, Rowley’s play twice.

Lake emphasized that the linguistic evidence for Rowley’s authorship of Woodstock was decisive: “there are 14 instances of the oath ’Sfoot, well scattered through the text … and used by several different speakers”. Further, the high incidence of I’m (on fifteen occasions, used by many different speakers) compared to “22 instances of the uncontracted I am, cannot be paralleled for any 16th-century play”. Both plays give their leading characters characteristic oaths. Woodstock has “afore my God” very frequently, while Rowley’s play has both “Gods holy mother” and “mother of God” (the variant preferred by a recent television detective). When You See Me has fourteen instances of the phrase I warrant followed by a personal pronoun, Woodstock has twelve; it has three instances of I assure followed by a pronoun while Woodstock has nine. These are unusually high frequencies, indicating common authorship.

In 2001 MacDonald Jackson confirmed Lake’s attribution of Woodstock to Rowley in an article for Medieval and Renaissance Drama in England, drawing on both linguistic and prosodic evidence, all of which placed Woodstock in the early seventeenth century. He showed that the oath ’sfoot first appeared in a stage play in 1598. Other contractions that date Woodstock to the period 1604–10 are shall’s (“shall we”), byth or bith (meaning “by the”) and th’are (“they are”). Jackson also selected thirty-nine words used in Woodstock and traced their first usage in OED, which overwhelmingly pointed to the early seventeenth century. Metrical evidence (run-on lines, feminine endings, pause patterns) confirmed that dating. Finally, Jackson confirmed Lake’s suggestion that Samuel Rowley wrote Woodstock, drawing additional evidence from When You See Me You Know Me. In addition to their unique fondness for the pronoun ye, both plays share expletives appealing to the Virgin Mary, and both use the colloquialisms I warrant and I assure with unusual frequency – fifty-one instances, more than any other plays of the period. Jackson summed up the “conclusive” evidence:

Woodstock exerted no influence whatever upon Richard II, because Woodstock was written after Richard II, probably at least ten years after Shakespeare’s play had first been performed and at least eight years after it had been first published in a Quarto of 1597.

Lake and Jackson were admirable exponents of what I have called the “open”, or empirical approach to authorship attribution studies, based on close observation of the texts. “Closed” approaches, by contrast, are those where enquirers are convinced that they know who the author is. Some are locked into a belief that Shakespeare’s plays were written by Francis Bacon, using a special cipher; others favour Edward de Vere, 17th Earl of Oxford (1550–1604). Readers may wonder how de Vere was able to publish and perhaps act in the six or seven plays that Shakespeare had yet to produce, but as we know from contemporary evidence, for conspiracy theorists all things are possible.

The only critic who disputed Jackson’s attribution of Woodstock to Rowley was Michael Egan, who had a much more prestigious candidate: Shakespeare. Egan was so convinced by his ascription that in 2006 he published The Tragedy of Richard II, Part One, described as “A Newly Authenticated Play by William Shakespeare” (authenticated by himself). The edition filled three volumes in four parts, over 2,000 pages, with Volume One (685 pages) devoted to “aesthetic” criticism, comparing the play to Shakespeare in terms of its “complexities, dramatic subtleties, and aesthetic achievement”. Given the play’s simple dramatic structure, the length of that discussion is alarming. Most authorship attribution studies run to thirty or forty pages: to exceed that norm thirty-fold arouses suspicion of special pleading, a scholar heaping up quotations to bolster a weak case.

In 2017 Egan published a brief “Acting Edition” of the play, in which the “hidden protagonist of course is William Shakespeare”. Now he has issued a hybrid volume, The Critical Legacy of Thomas of Woodstock, a collection of secondary literature from Halliwell in 1871 to Egan himself in 2024, with commentary. The aim of the collection is to suggest that a critical tradition has long existed that justifies Egan’s claim for Shakespeare’s authorship, but the claim is unconvincing. It includes an excerpt translated from Wolfgang Keller’s 1899 edition, hailed for having catalogued “more than a dozen striking parallels with Shakespeare”; but since they are not quoted, we cannot judge their significance. Egan claims that the author of Woodstock “deployed literally hundreds of expressions found nowhere else but in Shakespeare”, but he has no means of verifying that claim. If Egan were familiar with attribution studies of Elizabethan plays, he would know that they contain thousands of common collocations (“By the Lord”, “Good morrow, Master”). These have no evidentiary value unless they are unique, occurring in only two plays, a valid argument – to be checked by other tests – that both plays were by the same author. There was no way of identifying unique collocations before 2017, when Pervez Rizvi published Shakespeare’s Text online, a remarkable annotated corpus of 547 plays marked up to highlight every phrasal repetition. It remains the only resource able to identify unique collocations found nowhere else in sixteenth-century drama.

Egan’s contributors are various. They include Rainbow Saari, an independent scholar from New Zealand, who takes seventy pages to itemize commonplace verbal similarities between Woodstock and Shakespeare’s much larger vocabulary. Egan also includes excerpts from earlier scholars, reinterpreting them at will. A passage from A. P. Rossiter’s pioneering edition (1946) is labelled “Rossiter’s Unwitting Case for Shakespeare”, even though Rossiter declared unambiguously: “There is not the smallest chance that [the author] was Shakespeare”. But Egan, gifted with second sight, affirms that “Rossiter did know, somewhere deep in his professional soul”, that Shakespeare wrote it, “an affirmation of Shakespeare’s creative presence, all the more persuasive because Rossiter is supposedly speaking against the motion. In current politispeak [sic] he says the quiet part aloud”. If this practice of interpreting scholars as meaning the opposite of what they say proliferates, it will play havoc with literary history.

Egan’s own 100-page essay makes for painful reading. It attempts to impress by name-dropping: the dynamics of Richard II, Part One “generate a kind of extended Heisenberg effect”; it is “a work of deceptively surreal transitions”; the young King Richard “is precisely and almost contradictorily anti- or un-Nietzschean, driven not by a will to power but to powerlessness”; his portrait is “surprisingly modern”, often “anticipating Freud, Marx and … more modern thinkers”, even Lacan. Egan’s main interpretative concept is the term “dialectic”, which he misuses as a synonym for “doubling” or duplication, in words, scenes or characters. Regrettably, he also confuses it with the rhetorical figure hendiadys, the Greek word meaning “one by means of two”, in which two words (often different parts of speech) are joined by and, forming a semantic unity in which each defines the other in a variety of ways. So Prospero questions Miranda about her memory: “What seest thou else / in the dark backward and abysm of time?”, where “backward” refers to a past period of time, which is doubly “dark” or inaccessible (a straightforward epithet) and related to an abyss (a noun that suggests immeasurable vastness). Although Egan claims that this “characteristic of Shakespeare’s style” closely resembles “Brecht’s verfremdungseffekt, an estranging or making new, as Pound would say”, his examples are simple doublets, involving the same part of speech: “see and shun”, “rude and bitter”, “wished and welcome”.

The scholarly weaknesses of Egan’s volume are evident. Its significance is depressing, testifying to the persistence of anti-Shakespeare beliefs since the mid-nineteenth century. Claims that Bacon wrote Shakespeare were decisively refuted by the Bacon scholar J. M. Robertson in the 600 pages of The Baconian Heresy: A confutation (1913), who confronted “the Baconian fantasy” afresh, while recording that “many of these claims were made years before; but they seem to recur spontaneously”. This is true of all such fantasy attributions. If there are typing facilities in Purgatory, the authorship claimants may work away happily ever after there.

Brian Vickers’s fourth volume of The Collected Works of John Ford and the first first volume of his two-volume Collected Works of Thomas Kyd are reviewed in this issue. His book Thomas Kyd: A dramatist restored will be published later this year

The post Doubts about Thomas appeared first on TLS.