"Shakespeare: Editions and Textual Studies in 2023" Not the Year's Work in English Studies
No major scholarly editions of Shakespeare appeared this year. Four monographs relevant to our topic were published, and three of them are from Cambridge University Press's new Cambridge Elements series in its subseries called Elements in Shakespeare and Text, edited by Claire M. L. Bourne and Rory Loughnane.
The first monograph we consider is Holger Schott Syme's Theatre History, Attribution Studies, and the Question of Evidence. Syme asks what the recent authorship attribution discoveries about Shakespeare mean for our understanding of how theatre companies worked. For instance, if Shakespeare contributed Additions to The Spanish Tragedy, what theatre company was he doing that for?
The standard narrative is that The Spanish Tragedy appears in Henslowe's Diary as the Lord Strange's Men's play Jeronimo (variously spelt) being performed at the Rose in 1592-1593 and then again at the Rose by the Admiral's Men, and marked as "ne", in 1597. Then Ben Jonson is paid for additions for a revival of the play in 1601 or 1602, by which time the Admiral's Men are at the Fortune theatre.
But is Jeronimo the same play as The Spanish Tragedy? Syme sets out to show that The Spanish Tragedy had no connexion with the Admiral's Men. Richard Burbage was repeatedly identified with the role of Hieronimo in: a funeral elegy; 2 The Return from Parnassus; and an allusion in the play Wit's Triumvirate (1635) by William Cavendish. In the Induction by John Webster to John Marston's play The Malcontent, Henry Condell (metatheatrically portraying himself) says that the King's Men took this play from the Children of the Queen's Revels just as the Children took an Hieronimo play from the King's Men.
Perhaps Joronymo -- which is Syme's name for what the Admiral's Men performed at the Rose in 1597 -- was not The Spanish Tragedy but "a new play based on the Hieronimo character (p. 9). With The Spanish Tragedy disconnected from the Admiral's Men, there is no mystery in Shakespeare's adding to it. Many of the founders of the Chamberlain's Men in 1594 were old Strange's Men, and they brought The Spanish Tragedy with them to Shakespeare's new company.
For most of the plays that Henslowe marks as "ne", 57 out of 65, there is no reason to suppose this abbreviation means anything other than 'brand new play', as we have no prior record of these plays being performed. But in eight other cases, "ne" has been understood as meaning either 'new to this company' or 'newly refurbished'. Syme works through each of the eight to show that there is no reason to reject these also being entirely new plays and Henslowe's "ne" recording that fact.
The trickiest case is Titus Andronicus (Case 1) being recorded as "ne" when performed by Sussex's Men on 23 January 1594, which scholars have taken as too late to be the premiere of this play. Syme suggests that "ne" might mean not quite 'newly written' but 'first performed', noting that the theatres were closed for almost all of the period from mid-1592 to December 1593. Titus Andronicus might have been written in early 1592 and not performed until early 1594.
Cases 2 and 3 are "the frenshe Comodey" being "ne" on 11 February 1595 and "a frenshe comodey" being "ne" on 18 April 1597. These might simply be explained by their being two different plays. On the other hand, Cases 4 and 5 of "elexsander & lodwicke" being "ne" on 14 January 1597 and "elexsander & lodwecke" being "ne" on 11 February 1597 are clearly entries for one play. Syme notes the fairly long gap between the first and second performances and suggests that the play had to be withdrawn for some reason and relicensed, hence the second "ne" for 'newly licensed'.
What about Cases 6 and 7 of "tambercame" ("ne" on 6 May 1596) and "2 parte of tambercame" ("ne" on 11 June 1596)? We know of the two-part "Tamar Cham" being performed by Strange's Men at the Rose in 1592, so what does "ne" mean for these two in 1596? Syme suggests that the plays performed in 1592 and 1596 are in fact different two-parters -- the former owned by Strange's Men and the latter by the Admiral's Men -- both on the topic of Tamar Cham.
That leaves just Case 8: "Joronymo" being "ne" on 7 January 1597 despite "Jeronymo" being performed by Strange's Men at the Rose in 1592-1593. Syme believes that he has established that "ne" really does mean 'newly licensed play'. The simple explanation for the 1597 "Joronymo" owned by the Admiral's Men is that it is not the same play as the "Jeronymo" performed by Strange's Men in 1592-1593, and only the latter is The Spanish Tragedy. This helps explain why "Joronymo" did not make much money and why the Admiral's Men paid for two rounds of revisions to it, in 1601 and 1602; it was not as good as Thomas Kyd's version.
But would Henslowe really record two different plays with nearly identical titles? Syme shows that Henslowe did this in other cases. Syme thinks that the "Friar Bacon" plays in Henslowe's Diary are not all one play, and might actually be a cluster of three distinct plays about this figure. Likewise the multiple "Fortunatus" plays.
We know that Henslowe recorded under the single name of "Knack" two plays owned by different companies: A Knack to Know a Knave (Strange's Men) and A Knack to Know an Honest Man (Admiral's Men). Henslowe recorded a "Henry V" play that is not Shakespeare's and probably not the Queen's Men's Famous Victories of Henry 5 either. Syme has further examples of ambiguity in Henslowe's naming of plays.
Syme points out that his argument about Henslowe's practices forms a contrast to practices in the publishing industry. A stationer who owned the rights to the King Lear story owned the rights to all versions of that story. In the theatre, there existed multiple clusters of plays with each cluster sharing topics and protagonists and near-identical titles, each owned by a different company. Or even, in the case of the King's Men's non-Shakespearian Richard 2 plays, owned by the same company. We know from the described action that the Richard 2 seen by Simon Forman at the Globe in 1611 was not Shakespeare's play of that name.
Syme summarizes his key points: what we usually call the play of "Jeronymo" in Henslowe's Diary refers to different plays owned by at least four companies. The Spanish Tragedy belonged to the Chamberlain's/King's Men from 1594. The revisions to The Spanish Tragedy must have been made before Antonio and Mellida, which alludes to them, was entered in the Stationers' Register on 24 October 1601 and therefore these additions are not the ones Jonson was paid for as recorded by Henslowe on 25 September 1601 and 22 June 1602.
Syme thinks it unlikely that Shakespeare wrote for the rival company, therefore The Spanish Tragedy was a Chamberlain's Men's play, not an Admiral's Men's play, when he contributed to its Additions. Brian Vickers suggested that both companies played it, but we know that this was not allowed by the playing companies and that the Master of the Revels enforced exclusivity. Thus the Admiral's Men's play that Henslowe recorded as "ne" in January 1597 was some other play called "Joronymo".
Syme's important conclusion is that there were multiple plays about Hieronimo inspired by the success of The Spanish Tragedy, which was a Strange's Men's play and was never owned or performed by the Admiral's Men. When members of Strange's company went into the Chamberlain's Men they took The Spanish Tragedy with them. Thus Shakespeare's writing of Additions to The Spanish Tragedy was not a case of his moonlighting for the rival company.
The second Cambridge Elements volume to consider this year is Paul Salzman's Facsimiles and the History of Shakespeare Editing. Salzman promises to consider the effect upon the editing of Shakespeare of the changing technologies of facsimile reproduction that made early and rare research materials available to ordinary readers from the eighteenth to the twenty-first centuries. In fact, though, this is largely a history of facsimile technologies with only occasional links to editorial practice.
Salzman claims that "almost every copy" in an early printed edition was "unique" because of stop-press correction and "minute variations in the wear on type, patterns of binding, and pagination" (p. 14). This rather over-states the effect of stop-press correction, which as Joseph A. Dane has shown (PBSA 90 (1996): 272-320) tended to create exemplars that were more or less corrected as a whole, rather than uncorrected and corrected states of formes being mixed randomly, which if true would tend to make unique exemplars.
Moreover, if we take wear into account then even the exemplars of modern editions are unique, since the machines that make them are also prone to wear, and indeed no two sheets of paper are identical in their structure. Just what differences Salzman is referring to as those of "binding, and pagination" is mysterious. This is not a good start for a book on the precise meanings of sameness and difference in relation to text.
Salzman gives a summary of the seventeenth- and eighteenth-century editions of Shakespeare and claims that the growing desire to return to the original documents for particular readings -- even as the modern editions looked decreasingly like the first editions in layout -- created the demand for facsimiles. Type-facsimiles of Shakespeare started with one of a Shakespeare Folio in 1807. Salzman describes the brief period of hand-traced facsimiles and then the rise of photographic facsimiles in the nineteenth century.
W. W. Greg made the point that a type-facsimile might actually be more useful than a photographic one because its creator could pick amongst an edition's variant readings and present the correct ones, whereas a photographic facsimile necessarily reproduces either all of one exemplar or at least whole pages from different exemplars. Salzman does not mention it, but in principle a photographic facsimile could combine pictures of different parts of one page from multiple exemplars, although it would be difficult in practice and I think no one has attempted it.
Salzman points out that for the reason Greg gives about selecting readings in variants, a type-facsimile is more like an edition than a facsimile (p. 65). The early Malone Society Reprints were, Salzman reports, type-facsimiles offering a "diplomatic transcription" (p. 66). This is not quite right, since by definition a diplomatic transcription follows the words and punctuation of its source but does not respect its prose line endings and page breaks, whereas a type-facsimile does respect them.
There follows a series of further errors in Salzman's account that are worth noting. Salzman claims that Charlton Hinman's selection of pages from different exemplars of the Folio for his facsimile of it produced "a book that did not actually exist in real life" (p. 70). In fact, we do not know this since only about a third of the print run survives and the lost two-thirds might have contained an exemplar that had the same combination of page-states that Hinman selected. Hinman thought this possible but unlikely.
Certainly, in bringing together from different exemplars pages that belong to the same forme, Hinman did something that the book's printers could not have done. But this does not mean that the results of this process could not have existed: we just do not know what combinations of page-states happen to have existed in the lost two-thirds of the print run.
Salzman reports that microfilm readers have "screens with resolutions and light levels that strained the user's eyes and patience" (p. 86). Actually, microfilm reader screens do not have a 'resolution' of any kind as they do not digitize the image: they merely project it from the film by a purely optical process, and the brightness of the image depends on the rating of the lamp supplied by the manufacturer.
Salzman writes that "Chadwyck-Healey launched the first manifestation of the EEBO interface in 2003 . . ." (p. 86). This is wrong: the present reviewer and John Jowett co-authored a review of it that appeared in the journal Early Modern Literary Studies in January 2001. Salzman repeatedly refers to the high cost of ProQuest's EEBO database, which is certainly how it seems to readers outside the United Kingdom.
The situation in the United Kingdom is somewhat different. Because the government's Joint Information Systems Committee (JISC) wisely bought all the EEBO images with a one-off payment to ProQuest in the early 2000s, universities here get such a large discount that EEBO is affordable by almost all of them, with take-up near 100%. This bears on Salzman's closing remarks about the benefit of greater access to rare materials that facsimiles produce.
The last of the Cambridge Elements to be considered this year is Tiffany Stern's Shakespeare, Malone and the Problems of Chronology. Stern believes that we need to acknowledge that we do not know -- we cannot know -- the order in which Shakespeare wrote his plays, because of the inherent instability in writing and rewriting, which makes the matter undecideable. Her book is, she writes, aimed at "anyone who finds liberating the notion of chronological instability" (p. 1).
Three versions of Edmond Malone's foundational chronology of Shakespeare were published, in 1778, 1790, and 1821. Stern details Malone's process of research leading to the dates of particular plays and points to his inconsistent application of his own principles. An example is what he made of Francis Meres not mentioning a play in his Palladis Tamia in 1598. Only in some instances did Malone read absence from Meres's list as evidence that a certain play did not yet exist.
Stern believes that external evidence is "always about plays already in existence -- being entered in the S[tationers'] R[egister], or published, or put on at court, or seen or read by contemporaries" (pp. 23-24). This happens to be true in Shakespeare's case but is not a condition of the evidence being external. After all, the external evidence of Henslowe's Diary and his correspondence with writers gives us accounts of the progress of the writing of some non-Shakespearian plays.
Stern rightly objects to much of Malone's use of topical allusions, on the grounds that something can also be alluded to at times when it is not topical. Her general theme is that Malone got into his head an idea about a play's date and then marshalled evidence to support that date and ignored evidence that questioned it. She does not identify it as such, but this is the human cognitive tendency called Confirmation Bias.
Stern writes that Malone "knew that Shakespeare had become a shareholder of the Globe playhouse in 1603" (p. 35), but of course Shakespeare was a shareholder in that theatre from its inception, signing the document of syndication on 21 February 1599. If Stern is trying to communicate Malone's misconception then that should be made explicit; otherwise it sounds like her own error.
Stern finds Malone playing fast and loose with the evidence from the appearance of actors' names in early printed editions, particularly those whose time with the King's Men started after Shakespeare's death. The obvious inference, which Malone avoided, was that the plays were continually revised up to the point in 1623 when they got printed. Stern details the many examples of apparently late musical adaptation in the Folio scripts, which, she points out, would not necessarily have needed a dramatist's involvement at all.
I would have thought that if these really were just musical alterations then we are entitled to largely ignore them and date the plays from their first composition as dialogue. That is, only if dramatists such as Thomas Middleton were involved in the adaptation must we worry about Stern's wider point that for adapted plays there is no single date of creation to be found.
Stern finds Malone ignoring signs that the texts in the Folio contain 'late' versions of the plays, including signs of revision. On the question of how often and by how much plays usually got revised, Stern cites Richard Dutton and N. W. Bawcutt reaching different conclusions. Just how much departure from the licensed script would the Master of the Revels tolerate?
Whereas Dutton thought that "even minor changes" (Dutton's words quoted on Stern's p. 63) to a play needed relicensing, Stern uses Bawcutt's point that the Master of the Revels sometimes called for revisions but without demanding to see the script a second time. Rather, he simply warned that his licence was valid only if the changes he demanded were made. Thus, according to Stern, the Master of the Revels did tolerate revisions he had not seen, so clearly not all changes required relicensing.
The obvious retort is that tolerating changes that he had demanded is quite different from tolerating just any changes the actors wanted to make. Stern seems to be clutching at a straw here to defend her view -- repeated across her published works -- that plays were endlessly revised by playing companies because they were free to make changes without needing to get the script relicensed.
Stern reports that the Complete Signet Classic Shakespeare edition of 1972 edited by Sylvan Barnet presents the plays in chronological order (p. 66), but it does not. Barnet's chronology on his page 5 is clearly at odds with his 'Chronological Table of Contents' on page vii. The differences are that in the Table of Contents (and hence the order of presentation in the book) Barnet reorders the Henry 6 plays into their regnal order, reverses the order of Romeo and Juliet and Richard 2, and moves Timon of Athens from its position in the chronology between Measure for Measure and King Lear to a slot between Coriolanus and Pericles. In fact, the first complete-works edition to present the plays to the reader in the order the editors think they were written was the 1986-87 Oxford Complete Works.
On her pages 69-70, Stern challenges Gary Taylor's dating of the adaptation of All's Well that Ends Well from its line "the brooch and the toothpick, which wear not now", meaning that they are unfashionable. Taylor understood this in the light of evidence that brooches and toothpicks became unfashionable only after Shakespeare's death. Using EEBO-TCP, Stern finds positive references to brooches and toothpicks after Shakespeare's death. She acknowledges that it is uncertain, however, whether her examples should be read as ironic.
Stern is shaky on the technical aspects of just what the Text Creation Partnership project did to make EEBO into EEBO-TCP. She writes that many works in EEBO "have not been rendered into modern typeface and cannot therefore be searched" (p. 71). Typeface rendering has nothing to do with it.
The achievement of TCP was its keyboarding of the words that human readers could make out when looking at digital pictures of books. When a human operator presses a single key on a keyboard for each perceived letter or punctuation mark, each letter of the alphabet or piece of punctuation is henceforth represented by exactly one agreed string of binary digits. It is this reduction of a picture of a printed shape to a single agreed string for that letter that turns digital images into searchable transcriptions.
On the topic of whether Swiss visitor Thomas Platter saw Shakespeare's Julius Caesar at the Globe in 1599 or some other Caesar play, Stern reports that it must have been Julius Caesar at the Globe because ". . . the Rose was apparently shut at the time" (p. 81). For this she cites page 528 of Taylor and Loughnane's chronology essay in the New Oxford Shakespeare Authorship Companion. On that page, Taylor and Loughnane actually come to the opposite conclusion -- based on a note by the present reviewer -- that we cannot tell if the Rose was closed and hence we do not know which theatre Platter went to, nor (for that reason) whose play he saw.
Stern ends by proposing that instead of asking about when a play was first written we should ask about when it was first performed, since this is a question "for which there actually is an answer" because "there was a first performance" (p. 85). I can see how this simplifies matters, but it does not seem to avoid all the problems that Stern has shown to emerge from the fact of authorial and non-authorial revision of plays.
For instance, for King Lear, which date of first performance do we seek: that of the quarto-version history or the Folio-version tragedy? For Hamlet, do we look for the first performance of the version underlying the 1603 first edition, the 1604-5 second edition, or the 1623 Folio? For Macbeth, do we look for the first performance of the version that we think Shakespeare wrote but we no longer have, or the version as adapted by Middleton, for which we have the Folio text?
Stern acknowledges these problems but nonetheless offers her Table 2 showing possible entries in such a chronology for the plays Othello and Twelfth Night. It has columns for evidence of 'Earliness' and 'Lateness' and cells for these that cite scholarly sources ("MacDonald P. Jackson", "Helmut Ilsemann"). The problem is that these sources would, for some plays, run to dozens of entries. If completed, such a chronology would tend towards exactly the kind of long essay, with sections in date order, that we already have in such works as the New Oxford Shakespeare Authorship Companion and Martin Wiggins's British Drama 1533-1642: A Catalogue.
One more monograph this year is tangentially connected to our topic: Amy Lidster's Publishing the History Play in the Time of Shakespeare: Stationers Shaping a Genre. This should have been noticed last year but was overlooked. Lidster is concerned with construction of the genre of 'history play', in which a key event is the categorization in the 1623 Shakespeare Folio, although Lidster also traces its precursors. This is a history of publication that is interesting in itself but does not throw light on the texts so will be briefly noticed here.
Chapter One called "'True' Histories: Thomas Creede's Looking Glasses and the Print Identity of the Queen Elizabeth's Men" (pp. 43-85) gives the non-Shakespearian pre-history of the history-play genre. Chapter Two "Authorizing Histories: Andrew Wise and Shakespeare's English History Play" (pp. 86-131) expands on Lidster's essay "At the Sign of the Angel" reviewed in NYWES for 2018. Chapter Three "United Histories: Nathaniel Butter and his Newsworthy Playbooks" (pp. 132-179) likewise expands on Lidster's essay "Publishing King Lear (1608) at the Sign of the Pied Bull", which is noticed below in this review.
Chapter Four "Collecting Histories: The Jaggard-Pavier Collection (1619) and Shakespeare's First Folio (1623)" (pp. 180-229) is about how these collections helped shape our inherited sense of the new genre of history play. Lidster's conclusion recapitulates her main point that the history-play genre was itself constructed by print practices of selection and presentation rather than being something pre-existing in the theatre that the printed books neutrally transmitted. Print publication was, Lidster insists, its own medium with affordances and conventions distinct from those of the theatre.
The journal Shakespeare Quarterly produced two special issues to mark the 400th anniversary of the publishing of the Shakespeare First Folio in 1623. Despite this focus on the 1623 Folio, most of the essays in Shakespeare Quarterly this year are irrelevant to our concerns. We here attend to the five that concern our topic.
Aaron Pratt shows that the 1623 Folio was probably first sold without the preliminary leaves containing the poems by Leonard Digges and I. M. and the list of actors ("Through the First Folio, Darkly: When is a 'Perfect' Copy Perfect?", SQ 74[2023] 395-404). In his 1902 Census of First Folios, Sidney Lee came up with a classification of how good a condition an exemplar could be in:
Class 1-A Perfect, unrestored
Class 1-B Perfect, restored from other F1 exemplars
Class 1-C Perfect, restored from F1-F4 exemplars
Class 2 Lacking some leaves
Class 3 Defective, fragmentary
Class 4 Unknown condition
Pratt alludes to subdivisions of Class 2 but does not identify them.
Pratt describes the order of leaves in the Folio preliminaries (gathering πA), and the peculiar fact that three leaves were added to it to accommodate, in the Pforzheimer exemplar at least, the title page and blank on πA1+1, the Digges and I. M. poems on πA5+1, and the actors list and blank on πA5+2. Just where Digges's and I. M.'s poetry gets inserted differs in different exemplars.
These preliminaries could all have been fitted into a regular three-sheet gathering if blanks had not been used. Pratt concludes that the Digges material and the actors lists "was unavailable and/or unplanned until work on the core preliminary quire was already well underway in the printing house" (p. 400). From the evidence of other exemplars, it is plausible that some were originally sold without the Digges & I. M. material and actors list.
Pratt traces Carl H. Pforzheimer and Henrietta C. Bartlett figuring all this out, and suggests the former had a financial incentive to suppress the facts. Pforzheimer was motivated by his exemplar lacking the original Digges and I. M. material and actors list and having two facsimile leaves there instead. Thus by Lee's criteria Pforzheimer's Folio was a Class 2 exemplar that Pforzheimer could promote to Class 1 by getting some original Digges and I. M. and actors list leaves, which he then did.
Thus "In the end, by helping his First Folio measure up to those of top collectors, Pforzheimer rather ironically obscured the very thing that most distinguishes it" (p. 404). That is, Pforzheimer obscured the evidence it gives us that the Folio was originally sold without the Digges and I. M. poetry and actors list leaves.
Tara Lyons usefully debunks the commonly made claim that until it acquired a Shakespearian Folio in 1624 the Bodleian Library did not collect plays, treating them as mere "baggage books" ("'Baggage Bookes' and the Shakespeare First Folio: Towards a Critical Historiography of the Book", SQ 74[2023] 387-95). Yes, Sir Thomas Bodley did in 1612 tell his librarian Doctor Thomas James not to admit "baggage books" such as plays. But Bodley died in 1613 and James did then admit "English playbooks, almanacs, proclamations, poetry books, news pamphlets, joke books, advice literature, husbandry manuals, and guides to angling and playing chess" (p. 388).
Lyons discovered this simply by reading the Bodleian catalogues, accession records, and binding orders of the time. Lyons details the scholarly fallacy that not until the Shakespeare Folio made them into respectable Literature did the Bodleain admit books of plays. In fact the library had already accessioned the 1616 Jonson folio, which included his plays. Another fallacy is that the Bodleian's 1623 Shakespeare Folio was expensively bound. An interesting minor detail is that Bodley did not have to buy the books for his library: he had "a gift agreement with the London Stationers' Company" (p. 388).
Lucy Munro shows that the Folio preliminaries list of "The Names of the Principal Actors" is not in fact a reliable record of who was in the company in the early 1620s: it is more of a constructed, imaginary King's Men's company ("The Names of the Actors: The First Folio and Theater History", SQ 74[2023] 314-38). Munro traces the company members' individual biographies using other documentary evidence we have, including the occasional mention in stage directions and speech prefixes.
Necessarily, Munro has to string the known facts together using a series of "probably did" and "likely to be" phrases to say anything new. She gives details of the 1625 John Heminges's lawsuit against Elizabeth Gough, the widow of King's Men Robert Gough that was found by C. W. Wallace but not published before now.
The extraordinarily large research team of Christopher N. Warren, Samuel V. Lemley, D. J. Schuldt, Elizabeth Dieterich, Laura S. DeLuca, Max G'sell, Taylor Berg-Kirkpatrick, Kari Thomas, Kartik Goyal, and Nikolai Vogler has figured out the identities of the three printers of the 1685 Fourth Folio ("Who Rpinted Shakespeare's Fourth Folio?", SQ 74[2023] 139-46). Their essay's title alludes to the head-title of Hamlet in the Fourth Folio, which calls it the "THE TRAGEDY OF HAMLET RPINCE OF DENMARK".
We know that the job of printing the Fourth Folio was divided between three printers and Fredson Bowers identified the printer of the first part. This essay confirms Bowers's finding and identifies the printers of the other two parts. Using computational analysis of digital images of books, the authors of this essay have established that Robert Roberts printed part one, Robert Everingham printed part two and John Macock printed part three, the last including the "RPINCE" gaffe.
Lastly from Shakespeare Quarterly, Andrew Mattison makes essentially a literary critical study of the art of writing glossarial notes to Shakespeare, with examples from eigthteenth-century and modern editions ("Shakespeare, Steevens, and the Fleeting Moon: Glossing and Reading in Antony and Cleopatra", SQ 74[2023] 90-113). I cannot detect a larger coherent argument emerging from Mattison's local details.
Duncan Salkeld believes that Shakespeare was the sole author of the play Arden of Faversham ("Shakespeare in Arden: Pragmatic Markers and Parallels", ShSurv 76: Digital and Virtual Shakespeare[2023] 163-79). He begins by setting out his three theses: i) that Kyd had an influence on Arden of Faversham but did not write it; ii) that consistent use of pragmatic/hesitation markers points to the whole play having a single author; and iii) that these markers and some verbal parallels suggest that Shakespeare wrote the play (p. 164).
Salkeld surveys the recent claims for Shakespeare's authorship of the middle of the play (especially by Jackson) and for Kyd's authorship of all of it (especially by Vickers). In his account of these debates Salkeld overlooks a substantial number of the contributions.
Salkeld objects to the practice, common in authorship-attribution studies, of accepting as a match a phrase that is found in the work to be attributed and in some candidates' work even though in one of the two occurrences it is split between two speakers. Salkeld does not elaborate his objection but presumably it is that the change of speaker breaks the authorial mental continuity that characterizes a single phrase.
A possible defence of the practice Salkeld rejects is that at least some writers were in the habit of first composing the speeches for a scene and only later going back to insert the speech prefixes. In such a practice, the first words of a new speech would be the first thing a dramatist wrote after the last words of the preceding speech, maintaining the continuity that Salkeld thinks is lost in a change of speaker.
Salkeld thinks that "For obvious reasons, most 1-gram (i.e. one-word) matches are likely to have no value for attribution, unless the single word is particularly rare (like 'wistly', 'quondam' or 'brabble' -- all unusual words in Arden)" (p. 167). Salkeld seems oblivious to the large volume of scholarship that has shown the utility for authorship attribution of counting the frequencies of single unexceptional words.
The pragmatic/hesitation markers that Salkeld tracks are such things as 'Ah', 'Ay', 'Marry', 'Ha', and 'I'faith'. In particular he notes that Arden of Faversham "scenes 1, 2, 3, 5, 8, 9, 10, 11, 12, 13 and 14 all tend to favour the expressions 'Ay', 'Why' or 'Ah' at the start of lines" (p. 169). Salkeld describes his method for finding markers in early modern plays and it is not as systematic as one would hope..
Salkeld's corpus of plays by Christopher Marlowe is just 1 Tamburlaine, 2 Tamburlaine, Edward 2, and The Jew of Malta, leaving out Dido Queen of Carthage and Doctor Faustus as likely collaborative. His Kyd corpus is The Spanish Tragedy, Soliman and Perseda, and Cornelia. His Shakespeare corpus is The Two Gentlemen of Verona, The Taming of the Shrew, The Contention of York and Lancaster, Richard Duke of York, 1 Henry 6, Titus Andronicus, Richard 3, The Comedy of Errors, Love's Labour's Lost, Richard 2, Romeo and Juliet, and A Midsummer Night's Dream, chosen because they were written in 1590-1595.
Salkeld explains that he keeps his Shakespeare corpus to 12 plays so as not to dwarf the other canons. But of course this Shakespeare corpus does still dwarf the others and moreover it contains four plays (out of 12) that are widely agreed to be only partly by Shakespeare: The Contention of York and Lancaster, Richard Duke of York, 1 Henry 6, and Titus Andronicus. 1 Henry 6 in particular is largely not by Shakespeare.
Salkeld records that the word 'Come' followed by 'let us' or 'go' or 'away' occurs in Arden of Faversham in scenes 1, 2, 3 (twice), 4 (twice), 5, 6 (twice), 8, 9 (thrice), 10, 11, 12, 13 (twice), 14, 16, and 18, and about half of those are to clear the stage at the end of a scene. Particularly strong is the 'Come' followed by 'go' pattern: 11 occurrences in Arden of Faversham and a great many in Shakespeare but hardly any in Marlowe or Kyd. Likewise 'I pray you' occurs 13 times in Arden of Faversham, many times in Shakespeare, but hardly at all in Marlowe or Kyd.
Likewise, 'what' at the start of a question occurs 19 times in Arden of Faversham. Salkeld is unclear just what he is claiming about this example, since he does not give the count for Shakespeare but notices that the construction is fairly common in Marlowe and Kyd. I think his point is that 19 occurrences being spread fairly evenly across Arden of Faversham points to one author. In this section, about the word 'what', Salkeld confusingly mentions a few other links of Arden of Faversham to Shakespeare but without giving all the numbers for all the authors in his study.
The word 'zounds' occurs 10 times across Arden of Faversham and is absent from the works of Marlowe and Kyd and it is fairly common in Shakespeare: 6 occurrences in Salkeld's 12-play dataset. Salkeld is inconsistent about which Shakespeare plays he takes evidence from. For instance, 1 Henry 4 is not in his corpus of 12 Shakespeare plays but he refers to it here. This casual approach to which plays represent Shakespeare's style makes it impossible to think entirely quantitatively about the evidence he is presenting. And again, in the 'zounds' section Salkeld refers in passing to other links of Arden of Faversham to Shakespeare links without treating them comprehensively.
The phrases 'how now . . . news' and 'how now . . . whither' occur in total 16 times across Arden of Faversham. This construction is rare in Kyd and Samuel Rowley and absent in Marlowe. Salkeld gives no total for Shakespeare but quotes six occurrences, although curiously having named Richard Duke of York as one of his 12 Shakespeare plays he here quotes from 3 Henry 6. Thus it is not clear if he is using the 1595 octavo Richard Duke of York or the 1623 Folio 3 Henry 6 in his dataset; both contain the phrase Salkeld is quoting.
The phrase 'I pray thee' occurs 16 times in Arden of Faversham, in scenes 1-4 and 8-14. It is absent from Marlowe and occurs only once in Kyd and 24 times in the 12-play Shakespeare corpus. Salkeld reports that it occurs twice in 2 Henry 6 and so it does in the Folio text of that play. But it does not occur in The Contention of York and Lancaster, which is the name he gives this play in his dataset.
Next Salkeld turns to 'Lexical Parallels in Arden and Shakespeare' and he lists some of those. But for many of them Salkeld does not tell the reader which, if any, are found in the Marlowe and Kyd canons. The effect of this inconsistent approach to the evidence is to make Salkeld's study seem subjective and indeed amateurish.
The unfortunate effect is heightened by Salkeld repeatedly supporting his claims by reference to Shakespeare works not in his 12-play Shakespeare dataset, such as Venus and Adonis, The Merchant of Venice, As You Like It, Hamlet and The Winter's Tale. The last parallel Salkeld finds is the notion of 'earing' the land, present in Arden of Faversham and across Shakespeare's works, including his dedicatory epistle to Venus and Adonis. From all this, and by a route that he does not entirely logically spell out, Salkeld concludes that one man wrote Arden of Faversham and it was probably Shakespeare.
Jonathan P. A. Sell argues that in Henry 5 Mistress Quickly really does say of the dying Falstaff that "his nose was a sharp as a pen on a table of green fields" ("Quickly's Rawhide Notebook: Desentimentalising the Crux at Henry V, 2.3.15-16", Shakespeare 19[2023] 355-76). Sell starts with a scathing attack on a dozen literary critics and authors, from John Keats to Harold Bloom and Laurie Maguire whom he thinks have sentimentalized Falstaff and the plays' tavern world.
Sell wants us to condemn the "boozy, aggressive philanderer, capable of organising armed assault on innocent victims but too cowardly to take active part" (pp. 358-359). Sell thinks we should not consider Prince Harry "as a disloyal blackguard" but rather "a young prince who tires of what he has sufficient maturity, intelligence and moral pride to recognise as an arid and senseless life of Riley" (p. 359). We have all been led to this "momentous, collective loss of moral bearings" (p. 359) in part by Lewis Theobald's solution to the "table of green fields" crux.
Sell gives an account of the actual Boar's Head tavern in London and then reminds the reader of all the bad things that happen in the fictional one in the Henry 4 plays, summed up as "Bad debts, coarse humour and shabby sex" (p. 361). Sell concludes that "between merry swaggering and alcoholised sexual exploitation there is a very fine line indeed" (p. 373).
The Folio has Mistress Quickly say of the dying Falstaff that "his Nose was as sharpe as a Pen, and a Table of greene fields", which Theobald emended to "his nose was as sharp as a pen and a babbled of green fields". Sell tells the story of Pope's suggestion that a properties man named Greenfield was being mentioned in a stage direction that got mistaken for dialogue: "His nose was as sharp as a pen. A Table of Greenfield’s". Theobald mentioned an edition owned by "a Gentleman sometime deceas'd" having the marginal correction "and a' talked of green Fields" that led Theobald to his solution.
In his book Critical, Historical and Explanatory Notes on Shakespeare of 1754, Zachary Grey rejected Theobald's emendation and offered "For his nose was as sharp as a pen upon a table of green fells", in which a 'table' is a pocket book. Peter Stallybrass showed that these tables could be made of erasable paper. Sell points out that such a table would be useful to Mistress Quickly for keeping her record of bar tabs and other debts, which she refers to.
Sell glances at Salkeld's claim (reviewed in YWES for 2004) that "a table of green fields" alludes to the points on a backgammon board. Sell objects that the points on a backgammon board were not green and that several words would have to have dropped out for Salkeld to be right, since Quickly says Falstaff's nose is like the table not like the points on a table.
Sell also glances at Cyrus Mulready's defence of "a table of green fields" (reviewed in NYWES for 2018), which linked Quickly's line to the habit of commonplacing: writing someone's quotable expression in a pocket book. Such collections are sometimes called 'florilegia', and according to Mulready this prompts Quickly's remark, a couple of lines before the crux, about the dying Falstaff playing with flowers. As Sell rightly remarks, Quickly would have to be mistaking her words to an extraordinary degree, even for her, in order to be referring to commonplacing. And why would Quickly "be conversant with florilegia" (p. 370)?
Returning to Grey's solution of "For his nose was as sharp as a pen upon a table of green fells", Sell asks just what 'fells' are and how this word became the early edition's 'fields'. Grey claimed that 'fells' means skin and refers to the animal skin covering of Quickly's table book, and that the word got mistranscribed (or misspoken by Quickly) as 'fields'. Sell adds that 'green' could just mean untanned animal skin, rather than its colour. Finally, the Folio's 'and' is an easy mistake for 'on', and ". . . a pen might be expected to lie on a notebook, ready for use" (p. 372).
Sell endorses Grey's solution to the crux, although he would accept Quickly saying 'fields' for 'fells' as a malapropism. He thinks her phrase "need not have presented insuperable difficulties to contemporary audiences" (p. 372), but since the line struck Theobald, a mere 100 years later, as manifestly wrong this is not obviously true.
Bradley J. Irish claims that a supposed crux in Coriolanus is not really one ("Close Reading: The Inveighing, Envying Martius in Coriolanus 3.3.94", Shakespeare 19[2023] 377-86). In Coriolanus, Sicinius says that Martius has "Enui'd against the people", and the first word of this is usually emended to 'Inveighed'.
Irish finds that EEBO-TCP gives examples of 'envy against'. Most are not exactly like the Coriolanus case, but these are: "accompanied with envying against the raisements of other Saints", "you would ever have envied against a good work", and "It doth not become thee to Envy at other mens good labours". Thus Irish is right that "to envy against was a perfectly possible construction in early modern English" (p. 380).
Moreover 'to envy' had a wider meaning than it does now. It could mean a general malice and it was said to accompany pride, which of course Martius suffers from. And envy could also be directed towards inferiors who seem to be rising, which exactly covers Martius's attitude towards the plebians and their Tribunes including Sicinius. Complicating all this is the fact that 'envy' and 'inveigh' were not quite two words as they are now: the former was a valid spelling of the latter (but not vice versa).
Besides Salkeld's essay reviewed above, a second on the topic of the authorship of Arden of Faversham is Darren Freebury-Jones's claim that, contrary to Jackson's argument, this play shares more trigrams with Kyd's plays than with Shakespeare's ("Kyd, Shakespeare, and Arden of Faversham: Rerunning a Two-horse Race", ANQ: A Quarterly Journal of Short Articles, Notes, and Reviews, 36[2023] 347-50). Freebury-Jones's target is Jackson's essay "New Research on the Dramatic Canon of Thomas Kyd" (reviewed in YWES for 2009), which aimed to refute Vickers's claim (reviewed in YWES for 2008) that Kyd wrote Arden of Faversham. Freebury-Jones's method is to rerun Vickers's and Jackson's claims using Pervez Rizvi's Collocations and N-Grams online dataset and results.
Freebury-Jones sets the dates for comparable plays to 1580-1596, using Wiggins's chronology, and looks for what Rizvi calls 'formal' trigrams. This means that a three-word phrase counts as a hit when all three words in the same order find a match, but also gives an additional hit when one of the component bigrams within the trigram finds a match and yet another when one of the component single words within the trigram finds a match.
Freebury-Jones takes the trigrams in Arden of Faversham and looks for them in plays from 1580-96 and Kyd's The Spanish Tragedy, Soliman and Perseda, and Cornelia and 2 Henry 6 and The Taming of the Shrew. A hit would be a match found in Arden of Faversham and in one of the three Kyd plays or one of the two Shakespeare plays but not in any of the other plays from 1580-1596.
An obvious objection to this experiment is that in the almost 20 years since Jackson was writing there have been published multiple studies showing that 2 Henry 6, one of the plays counted here, was written by Shakespeare, Marlowe, and possibly one other dramatist. No one would set out today to test an authorship attribution claim using as one's 'ground truth' for Shakespeare's writing a play that is now widely agreed to be co-authored by him and others. (But see below Vickers's article rejecting the claim that 2 Henry 6 is co-authored.)
Freebury-Jones finds 38 hits for Arden of Faversham with 2 Henry 6 and 43 for Arden of Faversham with The Taming of the Shrew (this is the case for Shakespeare), and finds 34 for Arden of Faversham with The Spanish Tragedy, 66 for Arden of Faversham with Soliman and Perseda, and 13 for Arden of Faversham with Cornelia (this is the case for Kyd). That is, 81 for Shakespeare and 113 for Kyd. When Freebury-Jones makes an adjustment for The Spanish Tragedy at 21,450 tokens being shorter than 2 Henry 6 at 25,348 tokens the 34 hits of Arden of Faversham with The Spanish Tragedy is a higher ratio than the 38 hits of Arden of Faversham with 2 Henry 6: one hit per 631 tokens for The Spanish Tragedy versus one hit per 667 tokens for 2 Henry 6.
Freebury-Jones then tests Arden of Faversham against, on the one hand Kyd's The Spanish Tragedy, Soliman and Perseda, and Cornelia (totalling 54,624 tokens) and on the other hand nine Shakespeare plays: The Two Gentlemen of Verona, The Taming of the Shrew, Richard 3, The Comedy of Errors, Love's Labour's Lost, Richard 2, Romeo and Juliet, A Midsummer Night's Dream, and King John (totalling 186,790 tokens). The Kyd plays have 113 matches or one for every 483 tokens, and the Shakespeare plays have 352 matches or one for every 531 tokens. Freebury-Jones's conclusion is that Jackson was wrong: Arden of Faversham has more links to Kyd's style than to Shakespeare's.
Brian Vickers argues that Jowett and Taylor had no strong evidence for their claim that the young Shakespeare used the spellings 'Oh' and 'O' indifferently but then around 1593 switched to a strong preference for 'O' ("Compositors' Spelling Preferences and the Integrity of 2 Henry VI", The Library 24[2023] 141-53). Vickers finds that this undermines Jowett and Taylor's claim that Folio 2 Henry 6 is co-authored.
In the Appendix II in their book Shakespeare Reshaped, on Shakespeare's mature preference for 'O' over 'Oh', Taylor and Jowett made so many exceptions about anomalies (cases that contradict their main hypothesis) that the argument -- in which counter-evidence is attributed to scribal or compositorial interference -- falls apart. Vickers claims that Taylor and Jowett's counts for 'O'/'Oh' in Venus and Adonis and The Rape of Lucrece are wrong, giving Henry Woudhuysen's correct figures.
In fact, Woudhuysen agrees with Taylor and Jowett regarding Venus and Adonis and it is only for The Rape of Lucrece where Taylor and Jowett are slightly wrong in their counts. For the hypothesis that the young Shakespeare used 'Oh' but then stopped, Taylor and Jowett rely on Richard Field's editions of Venus and Adonis (about evenly split between 'O' and 'Oh') and The Rape of Lucrece (exclusively 'O'), which editions they assumed followed authorial copy. Vickers shows that Field's compositors imposed their own spelling and hence his editions cannot be used to infer authorial preferences. In Folio 2 Henry 6 we find 'O'/'Oh' roughly evenly used.
Vickers's account of the physical structure of the Folio is imperfect. Where he writes about "the three pages folded inside each other" (p. 147) to make a quire, he means three sheets, each of which holds four pages. Vickers also gets wrong the order of typesetting (composition) of each quire, writing: "Composition would begin on the outer sheet, with pages 1 and 12, followed by the reverse side, pages 2 and 11, then 3 and 10, 4 and 9, 5 and 8, finally 6 and 7" (p. 147).
Vickers cites Charlton Hinman's The Printing and Proof-Reading of the First Folio of Shakespeare for this account but in fact in that book Hinman showed that the usual order was precisely the opposite, working from inside to outside the quire: first pages 6 and 7, then pages 5 and 8, then 4 and 9, 3 and 10, 2 and 11, and finally 1 and 12. Vickers gets this right in his Table 1 but wrong in his description.
Vickers shows that if we look at the pages in the order the compositors set them, and who the compositors were -- using Peter W. M. Blayney's summary in the second edition of the Norton facsimile -- then we find that ". . . compositors sometimes changed their preferred spelling from one page to the next, or even within the same page" (p. 147). Vickers gets his counts from Rizvi's database of the Folio.
Overall, Compositor A and Compositor B who set Folio 2 Henry 6 both used 'O' and 'Oh' spellings. According to Vickers we are unable to "see the indifference showed by both compositors to spelling choices" (p. 149) if we look at the play by scenes. But of course if the compositors alternate in their spelling of this word in a pattern that matches the alternation of scenes, that constitutes evidence that in fact they were consistently following their copy in its alternation between 'O' and 'Oh' spellings. Whenever we can overlay upon a pattern that seems random -- here, the 'O'/'Oh' alternation in the printed book -- another pattern that is directly connected to the same activity -- here, the division of the play into scenes -- we have to take seriously the hypothesis that this second pattern explains the first.
Vickers points to a few scattered examples of Jowett and Taylor in other works accepting that Folio compositors were inconsistent about their spelling practices. Then he describes Rizvi's work on the entire body of Folio compositor analysis being flawed (reviewed in NYWES for 2016). Vickers asserts that a "series of studies in peer-reviewed journals has thoroughly discredited the methods on which the New Oxford Shakespeare’s attributions have been based" (p. 155) and in a footnote he lists them. In fact, most have been explicitly answered and shown to be wrong.
Also in The Library this year, Lukas Erne reports on a previously undocumented early exemplar of Pericles in Zurich ("The Two Gentlemen of Zurich: Marcus Stapfer and Johann Rudolph Hess, Swiss Travellers to England (1611-13,) and Their Shakespeare Quartos", The Library 24[2023] 51-67). The Central Library in Zurich has an exemplar of the 1611 third quarto of Pericles that does not appear in the censuses or the English Short Title Catalogue. It seems to have been bought from 4 pence since "4 den" is written on the title page.
Also written on the title page is the owner's name, Marco Stapfero, and that name also appears on the Yale University Elizabethan Club's third quarto edition of Hamlet. Erne gives an account of what we know about Stapfero, which is quite a lot as Stapfero wrote about his travels across Europe. Erne details his travels (including to the Bodleian) with his friend Johann Rudolf Hess. They both bought English play quartos.
Lastly from The Library this year, Anthony James West details "all sales and discoveries of Folios since 12 July 2000" in order to update his book The Shakespeare First Folio, The History of the Book, Volume 2: A New Worldwide Census of First Folios, published in 2003 ("Update of the Shakespeare First Folio Census", The Library 24[2023] 454-64). In some cases, Folio exemplars that were known to exist but were lost have been found, but there are no additions to the census itself. That is, no new exemplars not previously known have turned up.
This year, six articles address revision in Shakespeare's plays. Gary Taylor argues that the 1600 quarto of Henry 5 is not based on a corrupted version of the Folio version but is simply Shakespeare's first attempt at the play, which he later revised to make the Folio version ("Transeditions: Shakespeare's Two French English Lessons", Letras 33 (Number 67)[2023] 3-36). The play's English-language lesson is one place this is discernible.
The English-language lesson in Henry 5 is the only scene in Shakespeare written almost entirely in another language, and the versions in the 1600 quarto and 1623 Folio are markedly different. The scene is hard to edit because we want to distinguish mistakes about language that Shakespeare intended his characters to make -- the "dramatizations of erroneous translation" (p. 5) -- from unintended mistakes in the play's creation and transmission: "errors by scribes, printers, or Shakespeare himself" (p. 5).
Eighteenth-century editors were inclined to deny that Shakespeare wrote the English-language lesson, but Samuel Johnson thought it Shakespearian and explained the infelicities of the quarto version by its being poorly printed; Malone and Edward Capell agreed. Thus began what became the orthodox view, that the Folio's version was based on a good authorial manuscript, and the quarto's version was based on a performance-related and defective later manuscript. That is, until recently no one seriously suggested that the Folio version was in fact an authorial expansion of the quarto version.
Then, in a book reviewed in NYWES for 2016, Dutton did just that, arguing that Shakespeare routinely expanded his commercial-theatre plays for court performance and that this explains various quarto/Folio relationships. Others had dismantled the memorial-reconstruction theory and the claim that short quartos reflect their plays being abridged. Thus we have to treat the Folio version of the English-language lesson in Henry 5 as an authorial revision of the quarto version and edit accordingly.
The quarto version of the language lesson does not provide a reason for Katherine learning English, while in the Folio she says she that she has to learn it. Only the Folio has the sexualized language of the Harfleur victory, and Taylor sees the Folio's explanation in the same light: ". . . Katherine is a victim of English sexual violence" (p. 12).
Taylor claims that an editor's spelling the princess's name as the Anglophone Katherine rather than the French Catherine "will affect how some readers interpret and how some actors pronounce her name" (p. 13). I am not seeing how the 'C'/'K' choice could affect pronunciation, nor why 'C___' is French and 'K___' is English.
All the French words that one needs to know to understand the language lesson are nouns that the actors can point to (hand, arm, neck, and so on), and it is easy to infer that Katherine is asking for the English word for some French word ('main', 'bras', 'col'). Thus the scene is also in a way a French lesson for the English audience.
The most commonly used French textbook for English speakers was Claudius Hollyband's The French Littleton and its influence on the scene is particularly noticeable in the Folio version. This suggests that Shakespeare turned to this textbook to expand the scene, since a recollector who was turning the version underlying the Folio into the version underlying the quarto would hardly omit just the bits indebted to Hollyband.
How, then, is an editor to edit the scene? The French in it is imperfect by modern standards but was acceptable early modern French, and the scene is about two French people mispronouncing English. Because modern readers are used to modernized editions of Henry 5 based on the Folio and have never seen a modernized edition based on the quarto, the quarto's variants look particularly alien and wrong.
Taylor argues that what Andrew Gurr called the "rough phonetic" spelling (p. 21) of the quarto's version of the scene is in fact a phonetic spelling that Shakespeare adopted to help ensure that his two boy actors said what he wanted them to say in order to sound French. English printers were notoriously prone to error when typesetting foreign-language texts.
Thomas Creede printed the 1600 quarto for Thomas Millington and John Busby and when he reprinted it in 1602 for Thomas Pavier he not only failed to correct the printing errors in the language-lesson scene but actually made them worse. The third quarto of 1619 added yet more errors.
Where the 1600 quarto uses "vou" for modern 'vous' it is reflecting the actual pronunciation the boy actors had to make. Where the 's' should be sounded, in "vous aues ettue en Englatara", the quarto gives the terminal 's'. In the quarto the lesson begins with Katherine's "Allice venecia, vous aues cates en" and Taylor suggests that the meaningless 'cates' should be 'este' or 'estes', an early modern form of 'été'.
Taylor makes a few other suggested emendations of the 1600 quarto on the assumption of fairly routine printing errors and possible slips by Shakespeare. He finds that the French of the language lesson is, in the quarto and Folio, "more oral than bookish, more demotic than academic, and more insular than continental" (p. 28). Taylor identifies the French people in England whom Shakespeare knew, who were his main source of knowledge of the language.
Fernando Cioni argues that authorial revision separates The Taming of a Shrew, published anonymously in 1594, from The Taming of the Shrew in the 1623 Folio ("The Background of The Shrew: Texts and Intertexts", Cahiers Élisébethains 112[2023] 77-90). He starts with a claim that the Second, Third, and Fourth Folios of Shakespeare have some pages with the running title "The Taming of a Shrew", but this is not quite right. One page of the Second Folio (1632) has a similar error in that its running title is "The Taming of a the Shrew" (sig. S4r), but none lack the "the". The error that Cioni identifies appears only in the Third (1663-4) and Fourth (1685) Folios.
Cioni briefly traces the history of scholarship about the A Shrew play's relationship to The Shrew, and how eighteenth-century editors conflated the two. He finds the 1594 quarto of A Shrew to be unlike other bad quartos in that it is "more derivative" (p. 81) of the good version than is usually the case.
Cioni also claims (and quotes Ann Thompson claiming) that the material of the good version "has been handled more freely and rewritten more extensively (especially in the subplot) than is usually the case with memorially reconstructed text" (Thompson quoted on Cioni's p. 81). This appears to be a contradiction in Cioni's account, since if the 1594 quarto were "more derivative" of the good version than bad quartos usually are then it would be less (not more) extensively rewritten.
Cioni surveys the various theories about the origins of the bad quartos, in particular stenography and memorial reconstructions, and finds them generally inadequate to the problem. Cioni sketches Shakespeare's sources for The Shrew, focussing in particular on Plautus' Mostellaria. Cioni counts that Plautus' play has three similarities with A Shrew, four with The Shrew, and seven with both.
Cioni comments:
It seems that, if we accept the revision theory, Shakespeare wrote a version following the folk tradition (the three sisters' theme), Plautus, and Ariosto; then, revising the play, he went back to Ariosto, deciding to keep only two sisters as in I suppositi, changing the setting from Athens to Padua, and changing the names of all the characters, except Kate, Grumio, and Tranio. (p. 86)
I cannot find Cioni giving reasons why we should accept the revision theory, only that if we do then his is a plausible way for it to have happened.
Also in the journal Cahiers Élisabéthains this year, Warren Chernaik argues that authorial revision turned the play underlying the 1594 quarto The Contention of York and Lancaster into Folio 2 Henry 6 and the play underlying the 1595 octavo Richard Duke of York into Folio 3 Henry 6 ("Rival Monarchs: The Two Versions of 2 Henry VI and 3 Henry VI", Cahiers Élisabéthains 110[2023] 23-47). Chernaik finds The Contention of York and Lancaster/2 Henry 6 and Richard Duke of York/3 Henry 6 cases to be somewhat like the quarto/Folio case for King Lear, and announces that he wants to show that, as many believe happened with King Lear, the short first edition got revised and extended to make the Folio version.
Chernaik summarizes the history of claiming the opposite, that the first editions were based on cut-down versions of the full-text versions represented by the Folio, whether shortened by failure of recall (as in the memorial reconstruction theory) or intentional cutting (as held by Gurr and Erne). Chernaik considers the two original full titles of the plays: "The First Part of the Contention betwixt the two famous Houses of York and Lancaster" and "The true Tragedie of Richard Duke of Yorke, and the death of good King Henry the sixt, with the whole contention between the two Houses Lancaster and Yorke, as it was sundrie times acted by the Right Honourable Earle of Pembrooke his servants".
Chernaik claims that in these titles the contents "are announced explicitly as a two-part play" (p. 30), whereas the Folio makes them parts of a trilogy. I cannot see the two-ness that Chernaik discerns in these titles and he does not point it out. The Contention of York and Lancaster and Richard Duke of York have essentially the same scenes as those found in the Folio versions but the scenes are shorter. Also, particular roles are much longer in the Folio, especially Queen Margaret, King Henry, and (in The Contention of York and Lancaster but not Richard Duke of York) the role of York.
Chernaik thinks that memorial reconstruction "is scarcely credible" (p. 32) as an explanation for this but without saying why. In fact, the differences in certain roles' lengths as found in the quartos and Folio was a key piece of evidence used by supporters of memorial reconstruction: the roles that differ least are the ones taken by the actors doing the memorial reconstruction, since actors remember their own lines better than they remember the lines of other actors.
Chernaik quotes passages from 2 Henry 6 scene 3.1 and asserts that ". . . there can be no doubt that the differences in the two texts are the result of conscious choice" (p. 33) but without stating why this must be the case; he simply asserts that this is self-evident. The argument, as I infer it, is that the Folio version adds entirely new and coherent idea-clusters that are not present in the 1594 edition, and since random forgetting would be unlikely to alight on such coherent pieces and take them out entirely the more logical proposition is that they were added to the short version.
The obvious retort is that memorial reconstructors would be trying to create meaningful speeches, so if they could not remember enough of a whole new idea-cluster to render it meaningfully, they would leave that whole new idea out entirely. Similarly, those looking to abridge a play -- as in Erne's theory -- would remove whole idea-clusters in toto, not just parts of them.
Chernaik builds his argument with more examples of the same phenomenon, and they are all susceptible to the same critique. What is missing from Chernaik's examples -- and is found in the King Lear case he uses as a parallel -- are examples of something in the quarto being replaced (rather than added to) by something in the Folio. That is, the King Lear case for revision is built on examples of someone rethinking the drama -- for instance via a series of connected changes that alter the nature of the English/French war in the play -- rather than just expanding it.
In 3 Henry 6 there are lines that Richard Duke of York lacks and that anticipate Richard Crookback's rise in Richard 3, and Chernaik offers these as evidence that Shakespeare revised the play underlying Richard Duke of York once he knew he was going to write (or had written) the play Richard 3. Chernaik finds that Henry 6's long melancholic mid-battle disquisition about wanting a private life, found only in the Folio, is reminiscent of Richard 2's prison soliloquy, and beyond that the pre-Azincourt soliloquy of Henry 5, and hence was composed around 1595, about five years after Richard Duke of York was first composed.
Chernaik concludes that Folio 2 Henry 6 is a revision of quarto The Contention of York and Lancaster and Folio 3 Henry 6 is a revision of octavo Richard Duke of York. He contrasts these cases with that of quarto/Folio Henry 5, for which he cites Taylor in 1984 concluding that quarto Henry 5 represents a cut-down version of the play underlying the Folio. As we saw in NYWES for 2022 and in the essay reviewed above, Taylor has since then reversed his position and now claims that Folio Henry 5 represents an authorial expansion of the quarto version.
Peter C. Herman argues that the plays Richard 2, 1 Henry 4, 2 Henry 4, and Henry 5 were revised by Shakespeare to make the serial history we find in the 1623 Folio ("Shakespeare and the Book of Henry", Humanities 12.ii[2023] 1-17). Disconcertingly, Herman starts with a chronological confusion of his own. He dates Erne's claim that Shakespeare wrote for readers as well as for the theatre to Erne's 2013 book Shakespeare as Literary Dramatist (reviewed in YWES for 2013).
In fact, the 2013 edition was a second edition of Erne's book of the same title first published in 2003, and indeed the claim first appeared in Erne's 2002 essay "Shakespeare and the Publication of his Plays" (reviewed in YWES for 2002). By Herman's chronology (in which Erne makes the claim in 2013), Douglas A. Brooks's 2006 comment on Erne's claim seems to precede the claim it is responding to.
It is not simply that Herman cites the wrong edition of Erne's book, since he really does commit to his mistaken sequence in the history of ideas: ". . . in 2013, Lukas Erne dropped a bomb by proposing that 'Shakespeare wrote his plays not only with performance but also with a readerly reception in mind'" (p. 2). Furthermore, for his argument that Richard 2, 1 Henry 4, 2 Henry 4, and Henry 5 in the 1623 Folio represents Shakespeare's own revision of these plays to make a coherent and unified tetralogy Herman should engage with Nicholas Grene's 2002 book Shakespeare's Serial History Plays on the same topic, but Herman makes no mention of Grene.
Herman claims that the deposition scene in Richard 2, which is absent from the first three editions, "gets restored for the 1608 quarto" (p. 2). Of course, the deposition scene's absence from the first three quartos might instead be explained by its being written as a late addition to the play. Herman's wider point is that the Folio's deposition scene is substantially different again from the 1608 fourth-quarto's version, reflecting authorial revision, and he sketches other evidence that the Folio Henriad reflects authorial revision.
Herman points out that the end of Richard 2 leads seamlessly into the start of 1 Henry 4, the end of 1 Henry 4 leads seamlessly into the start of 2 Henry 4, and the end of 2 Henry 4 leads seamlessly into the start of Henry 5. In this I think he over-simplifies somewhat, since as critics have often noted Prince Harry seems reformed at the end of 1 Henry 4, having answered the call to arms and overcome Hotspur, but seems to have slipped back into his old ways at the start of 2 Henry 4, in which play he has to reform all over again.
Richard 2 mentions Sir Thomas Erpingham as one of those who support Bolingbroke in his return from banishment, and he is not heard of again until he turns up in Henry 5's army. The purpose of this inter-connexion of the plays is "reminding the reader that the events of the earlier play lead inexorably to this moment in the later one" (p. 4), and this happens only in Folio Henry 5.
Likewise, in 2 Henry 4 Shallow recalls the young Falstaff/Oldcastle being Thomas Mowbray's page, for no other purpose but to remind us of Mowbray's part in Richard 2. And in 2 Henry 4, Prince Harry and Poins briefly recall the "Anon, anon, sir" joke they played on Francis in 1 Henry 4. Herman finds that all these echoes are more noticeable when you read the plays than when you see them performed.
Herman considers these to be "theatrically insignificant echoes" (p. 5) and goes on to detail other echoes that are thematically significant. In 2 Henry 4, King Henry quotes from Richard 2 ("Richard . . .| Did speak these words"). Herman also notes echoes of the words 'heart' and concepts of time across the plays.
Like the chronicles on which they are based, the Henriad plays present the same events retold multiple times with different nuances each time; we are invited to compare how different people think differently about these events and how individuals -- for instance, Northumberland regarding Richard 2's deposition -- change their minds about them.
When the dying Henry 4 tells Prince Harry to "busy giddy minds| With foreign quarrels", Herman understands him to be admitting that he lied when he promised at the end of Richard 2 to expiate the sin of deposing Richard by ridding the Holy Land of Muslims. The real reason for this promise was to distract those who might oppose his new reign. The same irony arises at the start of Henry 5: a perceptive reader will understand that when Henry asks the clerics about Salic Law he is not seeking genuine legal and moral justification for attacking France. Rather, he wants a cover story.
Across the sequence of plays, Shakespeare increasingly represents the nation's most oppressed people, those in "the tavern world" (p. 11) and especially the pressed soldiers. The latter are not mentioned in Richard 2, are described in 1 Henry 4, are depicted in the muster scene in 2 Henry 4 but with jokey names, and then given real names and voices in Henry 5, but only in Folio Henry 5. Overall, I would say that Herman has convincingly shown the revision of the play underlying the 1600 quarto of Henry 5 to make Folio Henry 5, but for the other plays the quarto/Folio differences are not sufficiently clearly explored other than for the Deposition Scene in Richard 2.
Also on the topic of revision, but denying it, is MacDonald P. Jackson's claim that the prologue in the 1599 second quarto of Romeo and Juliet cannot represent a revision of the prologue in the 1597 first quarto ("The Prologue to Romeo and Juliet in Q1 (1597) and Q2 (1599): Purpose, Precedence, and Value", ANQ: A Quarterly Journal of Short Articles, Notes, and Reviews, 36[2023] 339-43). The familiar prologue to Romeo and Juliet is the one from the second quarto and it forms a 14-line Shakespearian sonnet. The prologue to the first quarto has just 12 lines comprising two quatrains rhyming abab cdcd and then two couplets rhyming ee ff.
The first quarto's prologue calls the Capulets and Montagues "Two household friends", which makes no sense in relation to the play's plot. The second quarto's "Where civil blood makes civil hands unclean" is rendered in the first quarto as "Whose civil war makes civil hands unclean", in which 'Whose' has no antecedent.
The second quarto's prologue is in the present tense throughout but the first quarto's slips into the past tense, as if the action of the play were over. Jackson considers the diction of the first quarto's version and finds it significantly inferior to that of the second quarto's version. The first quarto's version fails to mention that the lovers' deaths will heal the feud between their families.
Jackson finds that the second quarto's prologue is closer than the first quarto's to its source, Arthur Brooke's poem Romeus and Juliet, for instance in using many of Brooke's own words. It is inherently unlikely that Shakespeare wrote the incoherent first-quarto version and then revised it to make the coherent second-quarto version that also brought the play more in line with its source in Brooke. Thus the first quarto's prologue is simply a garbling of the version represented in the second quarto.
Heejin Kim published a comprehensive demonstration that one of our theories about the effects of revision is wrong ("Revision and Duplication in Early Modern Plays: A Reevaluation of the 'Minus' Hypothesis", PBSA 117[2023] 271-309). Greg proposed that when we find a duplicated line in a printed play it might have been caused by an addition or a deletion having been made to the underlying manuscript. Kim shows that this phenonemon is caused only by addition.
Scholars have long observed that in printed plays, including Shakespeare's, we sometimes find a certain line duplicated with another set of lines in between. For instance in the 1603 Hamlet, Hamlet says to Horatio and Marcellus:
I am very glad to see you, good euen sirs:
But what is your affaire in Elsenoure?
Weele teach you to drinke deepe ere you depart.
To this Horatio replies "A trowant disposition, my good Lord" and Hamlet responds:
Nor shall you make mee truster
Of your owne report against your selfe:
Sir, I know you are no trowant:
But what is your affaire in Elsenoure?
Scholars have hypothesized that the cause of such duplication (here, of "But . . . Elsenoure?") could be the insertion of additional lines to the manuscript, or else that some lines were marked for deletion.
In the case of insertion, the inserted matter (perhaps on a different piece of paper) could have ended by quoting the existing line that the insertion was meant to proceed, as in:
ORIGINAL MANUSCRIPT TEXT
aa aaa aaaaa aaaaaaaa aaa
bbbbb bbb bbbbbb bbbbbbb
ccc ccc cc ccccc cccccccccc cMANUSCRIPT INSERTION
xxxxxx xxxx xxxxxx xx
yyy yyyyyyy yyyy yyyyyy
zzzz zzzzzz z zz zzzzz zz
bbbbb bbb bbbbbb bbbbbbb
Notice that the insertion is meant to put lines "xxxxxx . . . zz" before the original manuscript's line "bbbbb . . .". A compositor might accidentaly begin the insertion after rather than before the original manuscript's line "bbbbb . . ." and fail to omit the insertion's final line of "bbbbb . . ." (which is merely the marker of where the insertion is wanted). If so, this would produce:
PRINTED RESULT
aa aaa aaaaa aaaaaaaa aaa
bbbbb bbb bbbbbb bbbbbbb
xxxxxx xxxx xxxxxx xx
yyy yyyyyyy yyyy yyyyyy
zzzz zzzzzz z zz zzzzz zz
bbbbb bbb bbbbbb bbbbbbb
ccc ccc cc ccccc cccccccccc c
Here we have line "bbbbb ..." repeated.
In the deletion scenario, the manuscript was perhaps marked thus:
MANUSCRIPT SHOWING CUT
aaaaa aa aaaaa aaa aaaaaa aaaa
bb b bbbbb b bbb b bbb bbbbb
| ccc ccccc c cc cccc cccc cccc ccc ° fff ffff ffff ffff fffffff
| ddddddd dd ddd ddd d ddd dd d
| eeeee eeee ee eeeee ee eeee ee ee
fff ffff ffff ffff fffffff
gggggg ggg ggggg ggg ggggg gggg
Notice that the cut is marked by a vertical line to the left of the material to be omitted and by the repetition, on the right of the head of the cut, of the line ("fff . . . fffffff" in our example) that is where the text should continue after the cut. If a compositor ignored or overlooked the cut, he might easily set:
PRINTED RESULT
aaaa aa aaaaa aaa aaaaaa aaaa
bb b bbbbb b bbb b bbb bbbbb
cc ccccc c cc cccc cccc cccc ccc
fff ffff ffff ffff fffffff
ddddddd dd ddd ddd d ddd dd d
eeeee eeee ee eeeee ee eeee ee ee
fff ffff ffff ffff fffffff
gggggg ggg ggggg ggg ggggg gggg
Now we have line "fff . . . fffffff" repeated.
Kim traces how these two scenarios were hypothesized and discussed by John Dover Wilson and Greg in the early twentieth century, on the basis of printed and manuscript plays. Kim sets out to show that the second scenario, the deletion hypothesis, is not supported by evidence.
Kim takes the one case that Greg offered as evidence for a duplication-from-deletion, in The Honest Man's Fortune, and shows that although it has a cut marked by a vertical line and it has the kind of interlined addition at the start of the cut shown above (our "° fff . . . fffffff"), this interlinear addition is not a marker of where to start again after the cut and so it would not produce duplication. That is, Greg's account is misleading and in fact he had no evidence for his duplication-from-deletion theory.
Kim has looked at 51 playbook manuscripts (essentially, all of the extant ones) and found 385 passages marked for omission, 70 additional passages in margins or additional leaves, and 5 false starts. In none did he find markings that support the duplication-by-deletion scenario above.
Also, Kim has found that false starts can explain duplication without needing the addition or deletion scenarios. Kim takes the reader through the details of the 15 most revealing playbook manuscript cases before concluding that ". . . duplicate lines in manuscripts are found only when additions are inserted in the course of composition or transcription" (p. 303).
Several essays appeared this year on the topic of how we use computers to analyze the language of Shakespeare's plays and those of his contemporaries. Pervez Rizvi published a two-part critique of the Word Adjacency Network (WAN) method that was co-created by the present reviewer and a team of electrical engineers from the University of Pennsylvania ("An Analysis of the Word Adjacency Network Method -- Part 1 -- The Evidence of its Unsoundness", DSH 38[2023] 347-60; "An Analysis of the Word Adjacency Network Method -- Part 2 -- A True Understanding of the Method", DSH 38[2023] 361-78).
This reviewer and the Pennsylvanian team published a response to this critique (Mark Eisen, Santiago Segarra, Alejandro Ribeiro and Gabriel Egan "'I Would I Had That Corporal Soundness': Pervez Rizvi's Analysis of the Word Adjacency Network Method of Authorship Attribution", DSH 38[2023] 1494-507). Rizvi attempts to replicate the WAN method for authorship attribution and show that it does not produce the new knowledge claimed for it. In our response, we show that Rizvi misrepresents fundamental aspects of the WAN method, that his attempted replication fails not because the method is flawed but because he erred in replicating it, and that Rizvi misunderstands key aspects of the mathematics of Information Theory that the method uses.
Another response to a critique by Rizvi was published by David L. Hoover, who died in 2023 ("Zeta Revisited: A Rejoinder", DSH 38[2023] 559-70). This responds to Rivzi's essay "The Interpretation of Zeta Results: A Supplement" reviewed in NYWES for 2022, and argues that Rizvi's critique of the Zeta method of authorship attribution is based on exaggerations.
Hoover continues the debate begun by Rizvi about whether the Zeta method's separation of the texts it tests into two distinct sets is an inevitable consequence of the method itself, as Rizvi maintains. Hoover points out that Rizvi admits that this separation does not always happen and Rizvi treats as rare exceptions the cases where it does not.
In fact, Hoover shows, what Rizvi calls the exceptional situations -- having few or short text segments, having texts that are highly alike such as multiple editions of the same work, or having more segments in one of the two starting sets (the 'base' and 'counter') -- are usual occurrences in real-world applications of Zeta, and hence not exceptions.
But what of the fact that Zeta can be made to produce two distinct clouds of dots from two sets of texts even when there is no meaningful difference between them, for instance "texts by blue-eyed and green-eyed authors" as Hoover puts it (p. 560)? Hoover thinks that "with thousands of words in play, some of them show enough consistency of use, by chance, to allow Zeta to distinguish the base and counter segments" (p. 560).
This sounds rather like conceding Rizvi's point that the method is biased towards finding differences even when there is no real difference to find. But then, the same phenomenon is also true of many machine-learning clustering algorithms: no matter what is the input data, if you ask the k-means algorithm to give you two clusters, it always will do so even if no meaningful distinctions are present.
Rizvi's performed an experiment in which he randomly selected words from Jane Austen's Pride and Prejudice and then let Zeta put them into two clusters even though they are by the same author. As Hoover demonstrates with an example, this randomization makes words that in the novel were concentrated in one part of the book seem more distinctive of authorship, since it distributes them across many segments, and makes words that were consistently spread across the book seem less distinctive because it bunches them. Thus, almost paradoxically, it is Rizvi's randomization that caused Zeta to cluster the segments.
Rizvi next randomly picked words from Shakespeare plays and John Fletcher plays to make segments that are thorough mixtures of their two styles, and yet again Zeta separated them into two clusters. Hoover shows that although Zeta will do this, a much stronger separation is created if the two sets really are differentiated by authorship. So, Hoover does to an extent concede Rizvi's central point that Zeta itself tends to separate texts even when there is no meaningful difference between them.
As Hoover puts it, "Zeta often produces a separation of base and counter segments even when there is no reason to believe that the two groups are truly distinct, but it produces a 'meaningful' separation . . .when the two groups 'are' truly distinct . . ." (p. 562). Hoover offers a few more randomized Zeta experiments that he thinks more correctly show Zeta's strengths and weaknesses than Rizvi's tests do.
The second part of Rizvi's critique of Zeta concerned the use of a single bisector line as a way to define the 'zones' of a scatterplot belonging to each author. Part of the problem, according to Hoover, is that Rizvi was critiquing Burrows's original form of Zeta not the one as revised by Hugh Craig that everyone uses now.
Moreover, Rizvi's proposed solutions to the bisector-line problem use subjective language: scores being 'like' or 'unlike' other scores. Rizvi's critique did not address the fact that in multiple blind validation experiments, the Zeta method using a bisector line has shown itself to be a powerful classifier of authorship.
A third essay responding to Rizvi on points of technical detail about how we do authorship attribution by computational means is by Nathan Dooner ("Principal Component Analysis and Authorship", DSH 38[2023] 1482-93). In his essay "Shakespeare and Principal Components Analysis" (reviewed in NYWES for 2021), Rizvi critiqued several investigators' use of Principal Component Analysis (PCA). One essay that drew Rizvi's attention was Jack Elliott and Brett Greatley-Hirsch's contribution to the New Oxford Shakespeare Authorship Companion, which attributed the play Arden of Faversham to Shakespeare.
Dooner argues that Elliott and Greatley-Hirsch's essay does indeed have limitations, but they are not the ones identified by Rizvi. For comparison with Arden of Faversham, Elliott and Greatley-Hirsch chose 34 sole-authored, well-attributed plays from 1580 to 1594. They modernized the spelling using the VARD2 software, removed proper nouns and stage directions, and divided the dialogue into non-overlapping 2000-word-token blocks, discarding the last one because it is necessarily smaller than 2000 tokens.
Within these blocks Elliott and Greatley-Hirsch counted the occurrences of the 500 most-frequent words from all 34 plays and also counted the occurrences of 221 pre-chosen function words. The encoding of the plays distinguished between different words spelt the same way, "such as 'like' the adverb from 'like' the preposition" (p. 1482).
Elliott and Greatley-Hirsch asked themselves how they could be sure that the true author(s) of Arden of Faversham was/were among those who wrote their 34 plays. On this point, Dooner shows, they misled themselves. They produced PCA scatterplots for their 500 most-frequent word and 221 function-word counts for their 34 plays and for the blocks of Arden of Faversham, with the first Principal Component along the x axis and second Principle Component along the y axis.
Because the points for the blocks of Arden of Faversham fell within the clouds of points for the 34 plays, Elliott and Greatley-Hirsch concluded that the true author of Arden of Faversham was among the authors of those 34 plays. This conclusion was faulty, as Dooner shows by repeating the experiment with i) some plays from the 1590s that Elliott and Greatley-Hirsch did not use, ii) 13 plays by Jonson, and iii) 9 plays from the nineteenth and twentieth centuries.
Where Elliott and Greatley-Hirsch omitted to identify the settings for essential parameters, Dooner tries multiple settings to get his scatterplots looking as much like theirs as possible. No matter which dataset Dooner uses, Arden of Faversham's points fall within the overall cloud of points from that dataset, even when in case (iii) the author of Arden of Faversham could not be present.
Rizvi objected that Elliott and Greatley-Hirsch did not disclose the eigenvalues from their PCAs, which are numbers giving us a measure of how much of the variation in the data is present in each Principal Component presented in a scatterplot. Rizvi suggested that in fact Elliott and Greatley-Hirsche's study's low eigenvalues make their results unreliable. Dooner shows that in fact we would expect low eigenvalues from the kinds of experiment that Elliott and Greatley-Hirsch were doing.
Most importantly, Elliott and Greatley-Hirsch were counting many more variables -- 500 in their most-frequent word experiment -- than they had data samples (around 260 blocks of 2000-word-tokens each). In general with PCA it is best to have more data samples than variables. At the ratio used by Elliott and Greatley-Hirsch, no single Principal Component is likely to capture much of the variation in the data.
Dooner points out that with as many as 500 words to count, there will be a large number of zero counts, where blocks have no occurrences of some of these 500 words. The frequency distribution across the blocks for the counts of a common word such as 'the' will be roughly Gaussian, but the frequency distribution for an uncommon word such as 'prince' will not be Gaussian because of all the zero counts.
Necessarily, with all these zero values, Elliott and Greatley-Hirsch's Principal Components were not capturing much of the variation in the data and Dooner estimates that their eigenvalues were in the range 21-30%. Dooner explains the notion of 'culling' -- which is excluding words from an experiment unless they appear in at least some chosen proportion of the tested texts -- and shows that of their 500 most-frequent words, only 25 actually feature in all of their 2000-word-token chunks.
Despite this, insists Dooner, Rizvi's objection to low eigenvalues on principle is mistaken. Dooner shows that roughly the same scatterplots can be generated using different levels of culling that produce eigenvalues as low as the tens and as high as the eighties.
Dooner's point is that Elliott and Greatley-Hirsch's PCA scatterplots simply do not show the zero-inflated data. And high culling thresholds, yielding high eigenvalues, can easily make the number of word-types being counted so small that real authorial-style information is lost and the distinctions between authors become less clear.
Thomas Merriam published three essays this year. In the first, he argues that the characters of Suresby in Sir Thomas More and Silence in 2 Henry 4 are strikingly similar and that Shakespeare wrote them both ("Suresby and Shallow", ANQ: A Quarterly Journal of Short Articles, Notes and Reviews 36[2023] 327-33). Both men are self-important elderly magistrates suffering from palilalia, the compulsive repetition of words or phrases.
Current dating puts 2 Henry 4 as the earlier play, in which case Suresby is modeled on Shallow. But since Suresby is also modelled on a source for Sir Thomas More, Thomas Stapleton's Tres Thomae of 1588, Merriam instead proposes that in fact Sir Thomas More preceded 2 Henry 4.
But if Sir Thomas More was written before 2 Henry 4 this presents a problem for Anthony Munday's authorship of Sir Thomas More. Munday et al.'s play 1 Sir John Oldcastle was meant to restore the reputation of Oldcastle after Shakespeare's traducing of him in 1 Henry 4 and 2 Henry 4, so it is hard to see Munday basing his Suresby on Shallow from 2 Henry 4, since to do so "would seem a respectful bow toward the object of Shakespeare's ridicule" (p. 330).
Merriam tries to show that Suresby and Shallow share empirically demonstrable habits of speech. He divides 2 Henry 6, Richard 3, 2 Henry 4, Julius Caesar, Cymbeline, and The Winter's Tale into 1000-token contiguous blocks, and likewise divides Munday's John a Kent and John a Cumber, and takes a single 1000-token block of Munday's original text of Sir Thomas More from scenes 2 and 3 in which Suresby appears.
Merriam counts the 8000 most-frequent character 4-grams in these texts and applies PCA to the results. He finds that the parts of Sir Thomas More containing Suresby's role test like the parts of 2 Henry 4 containing Silence's role and that neither test much like Munday's John a Kent and John a Cumber. For Merriam, this is further evidence for his long-argued conviction that Shakespeare not Munday wrote the original text of Sir Thomas More.
Merriam's second essay is on the same topic of Shakespeare writing all of Sir Thomas More without Munday's input ("Word Collocations and Character N-grams Agree That Munday Did Not Compose Sir Thomas More", ANQ: A Quarterly Journal of Short Articles, Notes and Reviews 36[2023] 15-23). Merriam starts by sketching prior work -- his own and that by others -- that casts doubt on Munday's (and perhaps Henry Chettle's) writing of the original text of Sir Thomas More.
Merriam then divides Sir Thomas More's original text into 1000-word contiguous blocks, and does the same for Munday's John a Kent and John a Cumber and for the parts of 2 Henry 6 that prior work, based on long n-grams shared with Sir Thomas More, has led him to suspect that Shakespeare not Munday wrote the original text of Sir Thomas More.
In these blocks Merriam counts the top 5000 most-frequent character 4-grams and lets the software package called Stylo for the R language generate a cluster analysis called a dendrogram. In this visualization, all of John a Kent and John a Cumber forms one of the two main branches, and the original text of Sir Thomas More and 2 Henry 6 form the other branch.
Thus Sir Thomas More's original text tests more like Shakespeare than Munday. To show that the method he uses is reliable, Merriam reprints a couple of dendrograms from an earlier publication of his, showing that his method successfully distinguishes George Peele's style from Shakespeare's.
Then Merriam turns to shared phrasing in the Sir Thomas More original text and Shakespeare's Cymbeline, and also likeness of themes, including Belarius privileging the simple and natural life over the court life, much as More does. Merriam adds Cymbeline to his 5000 most-frequent-character 4-grams analysis and finds that it clusters with Sir Thomas More not John a Kent and John a Cumber.
In his third essay, Merriam argues that Shakespeare created his Henry 5 by adapting an existing play by Marlowe ("Is it Time to Reconsider Henry V?", DSH 38[2023] 1158-65). Merriam surveys his previous publications arguing that Shakespeare and Marlowe co-wrote Henry 5. For the present essay, he counts the lengths of words in the verse and in the prose parts (except the French-language scene) of Henry 5 and finds a consistent difference.
To a statistically significant degree, the words in verse are longer at a mean of 4.27 letters than the words in prose at a mean of 3.97 letters. Moreover, this prose distribution in Henry 5 (mean, median, and inter-quartile range) is close to the distribution for prose in 32 other Shakespeare plays. But the verse distribution in Henry 5 is unlike the distribution for verse in 32 other Shakespeare plays, and is like the distribution found in Marlowe's plays. Merriam's preferred explanation is that Shakespeare revised a Marlowe play after the latter's death.
Our last authorship-attribution essay this year is the least convincing (Hartmut Ilsemann "R Stylo and the Authorship Determination of Henry V", DSH 38[2023] 1577-81). Ilsemann thinks that Merriam is right (in the essay reviewed above) that Marlowe perhaps wrote the verse parts of Henry 5 and Shakespeare the prose parts. Ilsemann makes from Henry 5 a new text file in which all the prose parts come first and then come all the verse parts.
Ilsemann uses R Stylo to run a Rolling Delta experiment across this text and "100 reference texts" (p. 1577). Ilsemann does not mention what the reference texts are, nor where he got his digital transcriptions of them.
In the experiment, Ilsemann looks for most-frequent-character trigrams using a 5000-word-token window advancing at 250 words per step. Although the body text says that he uses 100 reference texts, the legend to Table 1 giving the results for this experiment refers to "123 reference texts" (p. 1578).
Ilsemann finds that the prose first-half of his re-ordered Henry 5 tests most like Hamlet and The Merchant of Venice until about half-way through when the closest matches are to Marlowe's 2 Tamburlaine. Thus Merriam was right.
Ilsemann gets essentially the same results using the Rolling Classify software within R Stylo, with the classifiers being Nearest Shrunken Centroid, Support Vector Machines, and Delta. For unstated reasons, in this second round Ilsemann seems to have tried first 5000-word-token windows and then 8000-word-token windows. The latter takes the window up to nearly a third of the whole play. Ilsemann does not mention whether he also changed the step factor but it is implied that he kept it at 250-word-tokens.
Ilsemann seems to think that his second round of experiments gives him a separate attribution for "each 250-word section" (p. 1577p)` of his test text. It does not: the attributions are for each window, not for the steps by which the window advances.
Next Ilsemann tries the General Imposters method included in the R Stylo software package. He quotes at length from Maciej Eder about the implementation of the General Imposters methods in R Stylo and then discusses some refinements to it that he used.
But Ilsemann never mentions who, in his application of the General Imposters method, are "the authors that could not have written the text to be assessed" (p. 1580). Without knowing which candidates Ilsemann uses, his results are impossible to evaluate.
Jacques Savoy shows that Shakespeare uses a style of speaking for his female characters that differs from the style of his male characters, so we can make a classifier that identifies a character's sex just from what that person says ("Stylometric Analysis of Characters in Shakespeare's Plays", DSH 38[2023] 1238-46). We know that real men and women speak differently in measurable ways, so the interesting question is whether Shakespeare's men and women do.
Real women use personal pronouns more frequently than men and talk more often about family and social relationships. Savoy notes that "They are also employing more cognitive words (e.g. consider, think, believe, and it seems), certainty expressions (e.g. always and must), negations (e.g. never and not), and verbs, particularly auxiliary forms (e.g. is, have, and did)" (p. 1239). Men use more long words, more nouns, more definite determiners, more prepositions, more numbers, and more swear words than women.
For this study, Savoy takes 29 Shakespeare plays. He does not mention how he selects the ones to omit, but noticeably he leaves out Macbeth, 1 Henry 4, 2 Henry 4, Henry 5, 1 Henry 6, 2 Henry 6, 3 Henry 6, Henry 8, and The Two Noble Kinsmen, so the Histories in particular are under-represented. There are several co-authored plays among the 29 he uses: Titus Andronicus, Measure for Measure, Pericles, and Timon of Athens.
For his transcriptions Savoy uses the Folger Shakespeare Library XML files because these are marked up for the sex of the speaker. Strangely, for the dates of composition Savoy relies on Wikipedia rather than, say, the Database of Early English Playbooks, which gives both Alfred Harbage's Annals's dates and Wiggins's Catalogue's dates. Savoy ignores characters with small roles.
Savoy lists the top-13 most-frequent words for men and for women across the plays and they are the same words: 'a', 'and', 'I', 'in', 'is', 'me', 'my', 'not', 'of', 'that', 'the', 'to', and 'you'. What differs is the rank order: for men the top item is 'the' and the second is "I' and for women the top item is 'I' an the second is 'the'. These 13 types account for almost a quarter of all tokens for both sexes.
Savoy puts together all the male characters' speeches to make one file and then all the female characters' speeches to make another and then splits each of the two files into non-overlapping 1000-word-token blocks and then non-overlapping 500-word-token blocks. There are about three times as many segments for the male characters as female, since there are more male characters overall.
Savoy uses t-tests to compare the results for 19 stylistic markers -- including number of long words and mean number of letters per word -- for the male and female segments. His null hypothesis is that there is no difference between the male group's and the female group's values, and he starts with setting his threshold (α) at 5%. Since he has 38 markers (19 for 1000-word-token segments and 19 for 500-word segments) he expects to incorrectly reject the null hypothesis about twice. To reduce this error he uses Bonferroni correction -- appropriate where the same data is being tested multiple times -- in which you divide α by the number of tests (here 38) to get 0.13% as his new α.
The language features that are significantly differently used by male and female characters are: rate of use of long words (men use more); rate of use of personal pronouns (women use more); rate of use of first-person-singular pronouns (women use more); rate of use of second-person-plural pronouns (men use more); rate of use of definite determiners 'a', 'an', and 'the' (men use more); rate of use of prepositions (men use more); rate of use of auxiliary verbs (women use more); and rate of use of negatives (women use more).
Savoy groups the various pronoun forms to show that female characters talk more about themselves than male characters do, and talk less about other people than male characters do. Shakespeare definitely makes his female characters speak differently from his male characters in ways that studies have shown us that real men and women do.
Savoy builds two classifiers based on two-thirds of his Shakespeare data and uses them to 'predict' the sex of the speakers of the remaining one-third. One classifier uses Logistic Regression and the other Random Forests. Overall the accuracy of these classifiers was not terribly good.
Savoy then takes just the speeches of three of Shakespeare's female characters: Juliet, Rosalind, and Cleopatra. These three characters' speeches are removed from the training set and the classifiers recreated before being used to 'predict' the sex of these three characters from their words. Depending on which linguistic features Savoy uses, the classification was 'female' with at best the confidence of 92% for Juliet, 89% for Rosalind, and 83% for Cleopatra.
Scott Maisano believes that Rosaline in Love's Labour's Lost is a black character and that editors need to take this into account ("Against Editing 'Whitely': Restoring Shakespeare's Interracial Comedic Couple to Love's Labour's Lost", Renaissance Drama 51[2023] 55-83). Maisano thinks that "no one in Love's Labour's Lost directly comments on Rosaline’s skin color until act 4, scene 3" (p. 63), and he can see that this might cause scepticism. Why does no one mention it before? Maisano's answer is that "anti-Black sentiment rarely gets directed at Black characters" (p. 63).
To make his case, Maisano selectively quotes from the exchange in 4.3 between Biron and his friends (principally the King) about Rosaline's blackness. His reading omits the rhetorical and dramatic force of the exchange, which is that the more that Biron praises Rosaline's fairness the more the others say that she is black. Both sides use increasingly absurd exaggerations.
Specifically, Rosaline is described as "an attending star", without whom "day would turn to night", with "a fair cheek", and she is likened to "the sun that maketh all things shine". In response to this hyperbole from Biron, the King says that Rosaline is "as black as ebony". Biron replies that "in black my lady's brows be decked" because redheads dye their hair black "to imitate her brow", in both cases using 'brow' to mean eyebrows.
Maisano considers Dominic Dromgoole's production of Love's Labour's Lost at the London ,replica Globe theatre which cut quite a lot of this exchange about Rosaline's appearance but retained "DUMAINE I never knew man hold vile stuff so dear.| LONGUEVILLE (showing his foot) Look, here's thy love -- my foot and her face see".
Maisano thinks that these lines were kept "because they do not explicitly mention Blackness per se and because the lines lose their racist charge as a result of Dromgoole's casting of Thomasin Rand, a white woman, a brunette, to play the part of Rosaline" (p. 65). Thus Maisano finds that these lines are part of the reason for thinking that Rosaline is black -- even though he admits that they do not say that -- and that Dromgoole de-raced the lines by casting a white actor as Rosaline.
Maisano writes at length about just why almost everybody has hitherto understood Rosaline as a white character, and the core of the problem is that editors have been editing "whitely". This adjective comes from Biron's description of Rosaline as white: "A whitely wanton, with a velvet brow,| With two pitch balls stuck in her face for eyes" (3.1.191-192).
Maisano rightly points out that in editions before the Third Folio the word 'whitely' was spelt 'whitly'. Maisano thinks that this "an unusual word in early modern English" (p. 68) but in fact it is not: EEBO-TCP has 90 occurrences. Maisano believes 'whitly' to be an error and that Shakespeare wrote 'whitty wanton'.
One problem of course is that the next line is also about colour, referring to Rosaline's eyes as black balls. Another problem, perhaps fatal to Maisano's argument, is that there is exactly one occurrence of 'whitty' in EEBO-TCP other than as the family name of a person, and that one is part of the rare Scottish phrase 'whitty whatty' meaning nonsense. This is not the meaning required bys the play.
Next Maisano suggests that 'whitly' might actually be 'wittolly, which suffers from the same problem of not fitting the context. Since a wittol is a man who knows and does not mind that his wife is unfaithful, a man can be wittolly (as Falstaff says Ford is in The Merry Wives of Windsor) but a woman cannot. Maisano toys with the idea that Shakespeare meant that Rosaline is "'a wightly wanton,' implying a certain alacrity or briskness with regard to impropriety or promiscuity" (p. 69).
Maisano approvingly quotes Timothy Love's argument that Rosaline being described as 'whitely' "could serve as confirmation, rather than contradiction, of Rosaline's identity as a Black woman" (p. 70) because it "testifies to her mimicry of whiteness" (Love quoted on Maisano's page 70). Or perhaps, suggests Maisano, 'whitely wanton' means that Rosaline suffers a pallor arising from unsatisfied sexual desire.
Maisano expends considerably effort trying to show that Rosaline's "velvet brow" does not refer to the smoothness of her skin or the black colour of her eyebrows, but to her black forehead. He acknowledges that Biron four times refers to Rosaline's white hand, and he considers a couple of ways around this obstacle. One is that Biron "throughout the play lavishes white supremacist praise upon her glove" (p. 75), presumably knowing that the hand inside it is black.
Even more bizarrely, Maisano suggests that perhaps Biron "thinks he is being funny by repeatedly referring to Rosaline's 'white hand' despite knowing her hand is black" (p. 75). This last suggestion contradicts Maisano's earlier claim that "anti-Black sentiment rarely gets directed at Black characters" (p. 63) since if Biron is throughout the play making a joke at Rosaline's expense he would be repeatedly doing it to her face.
Asked about Biron's letter to Rosaline, the latter says that Biron has drawn her picture in his letter. Asked whether the likeness is good, Rosaline replies "Much in the letters" (meaning the penmanship) but "nothing in the praise". The Princess responds "Beauteous as ink. A good conclusion" and Catherine chips in "Fair as a text B in a copy book", meaning that Rosaline is therefore as beautiful as a formally written letter 'B' in what we would call an exercise book.
Maisano reads this as Catherine saying that Rosaline is like the blackest-letter "the inkiest of them all" (p. 77) and hence Rosaline is black. Of course, the joke would work just as well if Rosaline were merely dark-haired.
Maisano finds seriousness and venom in the witty exchange between Catherine and Rosaline in 5.2 ("What’s your dark meaning . . . is still past care") and considers ". . . Katherine’s attitude toward Rosaline as motivated by and taking shape in reaction to Rosaline as a Black woman" (p. 79). For the stage direction "Enter blackamoors with music" Maisano suggests that perhaps black musicians were used. Since we know of no black musicians working for early modern theatre companies, this seems particularly implausible.
Maisano's theory requires that Shakespeare wrote a major character in one of his plays as black woman but -- unlike his practice elsewhere -- he made no one in the play refer to it openly. Instead, Shakespeare concealed this fact in a series of verbal ciphers that no one until the twenty-first century cracked the real meaning of.
Sarah Neville discerns an undisclosed intention in Greg's celebrated essay "The Rationale of Copy-Text", which made the influential distinction between substantive readings in a document (essentially, the choice of words) and the accidental readings (such things as punctuation) ("The Accidentals Tourist: Greg's 'Rationale of Copy-text' and the Dawn of Transatlantic Air Travel", Textual Cultures 14.ii[2021] 18-29). What Greg really wanted to do, Neville argues, was get editors to write textual notes that enabled the reader to reconstruct the copy text so that scholars needed to travel less to see original documents.
Neville claims that the Unediting movement of the 1980s and 1980s was a reaction to Greg's "Rationale" distinguishing between the authority for accidentals and the authority for substantives. Thus she writes that "Since the 1980s, proponents of 'unediting' have . . . chipped away at Greg’s 'Rationale', speculating that its accidental/substantive division is deceptively reductive, as even minor variants can have major implications" (p. 23).
In fact this argument against Greg's "Rationale" appeared well before the 1980s and is unconnected to the Unediting movement. Paul Baender in his essay "The Meaning of Copy-Text" in Studies in Bibliography in 1969 began the critique of Greg's accidentals/substantive distinction. Others joined in, including Tom Davis in his essay "The CEAA and Modern Textual Editing" in The Library in 1977 and D. F. McKenzie in his conference paper "Typography and Meaning: The Case of William Congreve" delivered in 1977. Neville's chronology of this debate is wrong.
Neville's central claim is that Greg wanted to ensure that modern scholarly editions had textual notes that were detailed enough to enable the reader to reconstruct the copy text (p. 23), thereby saving the reader a trip to a research library. Neville spends considerable space on the changing technologies of air travel in the twentieth century, as piston-engined aeroplanes were surpassed by jet-liners. But there is nothing in Greg's "Rationale" about how to write textual notes so that the copy text can be reconstructed; rather, Greg was concerned with the readings (accidental and substantive) that an editor admits, and not their documentation.
In a second article this year, Tara Lyons finds that Shakespeare's 1623 Folio has similarities to an edition of Seneca's collected plays from 1581 ("Shakespeare and the English Seneca in Print: Collections, Authorship, Collaboration, and Pedagogies of Play-reading", Memoria di Shakespeare: A Journal of Shakespearean Studies 10[2023] 1-42). The edition of Seneca's Ten Tragedies in 1581 was the first printed collection of plays in English. The second was the Shakespeare Folio of 1623. (This perspective of course excludes mixed-genre collections such as the Jonson folio of 1616.)
In citing the Shakespeare Folio preliminaries, Lyons uses a 'p' where I would expect a superscripted π symbol to indicate that the preliminaries use the same letter ('A') as the main sequence of gatherings. Thus she cites "pA2r-A3r" (p. 5) where I think she means πA2r-A3r and presumably an 'A' has dropped out so that her "p4r-A6r+1" (p. 5) should be πA4r-A6r+1. Lyons reports that the 1623 Folio preliminaries contain "a 'Catalogue' or table of contents that divided the thirty-six play titles into three dramatic genres" (p. 5). Actually, the Folio Catalogue gives only 35 play titles: it makes no mention of Troilus and Cressida.
Lyons finds commonalities between the Seneca collection and the 1623 Shakespeare Folio that previously have gone unnoticed. Both books try to capture the author's intentions, both were preceded by editions of suspect accuracy, and both somewhat acknowledge the collaborative nature of their contents: Seneca's by acknowledging translators, Shakespeare's by acknowledging theatrical origins. Lyons gives detailed accounts of the two books' preliminaries and their organizing principles.
Comparing the quarto and Folio versions, Gerald E. Downs tries to make sense of the unclear handicapping system used in the betting on the fencing match in the final scene of Hamlet ("The Hamlet Wager: Ods A'th Weeker Side", ANQ: A Quarterly Journal of Short Articles, Notes, and Reviews, 36[2023] 194-202). In the 1604-5 quarto, Osric says of the bet against Hamlet that "hee hath layd on twelue for nine", and Downs seems to believe that here 'hee' is Laertes; in context I would have thought that 'hee' is the king.
In the Folio, Osric says "He hath one twelue for mine" which does not seem to mean anything. In the 1603 quarto the equivalent line is "Leartes . . . do not get three oddes of you", the meaning of which is unclear. Downs finds that in EEBO-TCP there are no occurrences of the phrase "one odds" before 1650 but he is using ProQuest's inferior front end to this dataset, and EarlyPrint shows several occurrences and in exactly the sense in Hamlet. For instance Gerard Langbaine's A Review of the Covenant of 1643 has "three to one odds against it".
Downs writes of such phrases as "five to one" that they "will not quite have expressed probability in Shakespeare's day" (p. 194). I cannot make out what he means by this. Perhaps he means that this is almost correct as an expression of probability. If so, one wants to know in what ways it is right and what ways wrong.
The problem of Downs's incomprehensibility gets worse as the paper progresses. For instance, Downs writes:
The handicap is set apart from attacking hits as defensive exceptions, or odds. . . . Hamlet and Claudius understand that no more than two defensive hits will be scored for Laertes. (p. 196)
Downs does not explain what he means by "defensive exceptions" or "defensive hits". Having read his article several times I am defeated and have no idea what his solution to this crux consists of.
This year one collection of essays is largely on our topic and will be treated as a unit: Claire M. L. Bourne's Shakespeare / Text: Contemporary Readings in Textual Studies, Editing and Performance. This is part of the Arden Shakespeare's series called Arden Shakespeare Intersections and in their Series Preface (pp. xvi-xvii) the series editors Sonia Massai, Farah Karim-Cooper, Gordon McMullan, and Lucy Munro explain that they want to look at "alternative configurations for the field of Shakespeare criticism" (p. xvi).
The series editors encourage consideration of "intersections", by which they mean the name 'Shakespeare' yoked to a topic, such as 'Shakespeare/Sex' or as here 'Shakespeare/Text', in which various binary oppositions are considered. This perhaps explains why the title of every essay in this collection is a binary opposition. A few essays in the collection that are not on our topic will be passed over in silence.
In "Fair / Foul" (pp. 29-49), B. K. Adams proposes the four-term homology "Shakespeare is to text as fair is to foul" (p. 31) and explores what follows from this. The answer seems to be nothing of interest, which presumably is why no one has proposed this homology before.
Adams follows Kim F. Hall in finding that the term 'fair' was, in Shakespeare's time, "an endogenous somatic marker of race and difference" (p. 32). The evidence she adduces for this is that people used 'fair' in its ordinary senses of 'just' and 'desirable' and 'clean' but also used it for the whiteness of skin.
Out of this trivial fact of language, Adams spins a substantial claim of white supremacism in early modern English. One might just as easily spin a claim of black supremacism out of the fact that African languages use the same term for blackness of skin, such as dúdú in Yoruba, and for generally desirable characteristics such as health and beauty.
Because paper and skin were often likened -- whiteness being valued in both -- Adams links what she sees as a racial hierarchy to considerations of writing, but the link is tenuous. Adams barely acknowledges that paper was merely the substrate for writing, and that writers recurrently observed that writing was made possible only by the addition of black ink in pens and printing presses.
For Adams, the social climbing of the kind that Shakespeare was accused of in Greene's Groatsworth of Wit in 1592 "is similar to racial passing" (p. 37). Adams insists that wherever we find something being praised for fairness it is therefore also being praised "by extension [for] whiteness" (p. 37). With such extensions, almost anything can be whiteness: "Like fairness, sweetness is an achievable type of whiteness . . ." (p. 38).
Adams turns to studies of the 1623 Folio, finding the extraordinary explosion of new knowledge about Shakespeare's works arising from the New Bibliography as something suspect. "This relentless mission", she writes, "was frequently disguised as evidence-based theories and ongoing friendly conversations among male colleagues" (p. 39).
It is hard to see what Adams means here by "disguised". The New Bibliography really was a set of evidence-based theories that where discussed by experts when academia in general was largely a male preserve. Adams offers no evidence to substantiate her denigration of the New Bibliography.
Adams surveys a book by Evelyn May Albright and implies that the New Bibliographers paid little attention to it because she was a woman and they were sexist men. Rather, the New Bibliographers paid little attention to Albright's work because it was of low quality. By contrast, Alice Walker's work was widely discussed because of its excellence. In championing Albright, Adams ignores the gross errors of fact that mar her work.
According to Adams, widespread bigotry meant that others had it even worse than Albright. "For decades", she writes, "the libraries and universities in which researchers have interacted with these various texts or held symposia or had uneasy discussions about this work were not particularly welcoming to white women and certainly not hospitable or open to Black, Indigenous or other People of Colour interested in the field" (p. 43).
Adams offers no evidence for her claim that libraries were closed to black, indigenous, or other people of colour. It is true that race quotas were until recently used to keep some people out of American universities, which practice ended with a Supreme Court ruling (Students for Fair Admissions versus Harvard, 600 U.S. 181 (2023)) upholding a complaint brought by Asian Americans. Adams ends where she began, with the assertion that the scholarly effort to find out exactly what Shakespeare wrote -- the pursuit of writings based on fair rather than foul scripts -- has long been a racist endeavour, since 'fair' means white.
The topic of "Text / Paratext" (pp. 50-65) by Hannah August is the significance of the presence of an author's name on an early modern play edition's title page, and occasionally elsewhere in the book. August looks for genre markers on the title pages of early Shakespeare editions and finds that whereas tragedies and histories are generally identified as such (with some awkward overlaps), comedies mostly are not.
August proposes that the name 'Shakespeare' itself was enough to tell readers, unless otherwise informed, that the contents were comic. Mentions of Shakespeare in commendatory verse after his death show the same thing: he was most often named as a writer of comedy.
In "Canon / Apocrypha" (pp. 102-120), Aleida Auld argues that Shakespeare's poems were in a strangely liminal position in the canon in the eighteenth century. Readers could buy the complete plays in major editions, and then buy an additional volume containing the poems. But in Dublin at least, from 1726, they could also buy an edition in which the poems were integrated.
Theobald was the first to use parallel passages from Shakespeare's poems to emend or make sense of words in the plays (p. 109). Auld glances at the Shakespeare Apocrypha once or twice in this essay, but really her essay is about how the poems got published, so her two-word title is rather misleading.
In "Miscellany / Sequences" (pp. 143-162), Megan Heffernan shows that the terms we use for collections of poems, such as 'miscellany' and 'sonnet sequence', were not in use in Shakespeare's time. Heffernan's considers mise-en-page and how sonnet sequences were numbered. The upshot is that we should think beyond the miscellany/sequence binary. I am not clear how Heffernan wants us to do this since her examples are almost entirely non-Shakespearian.
Margaret Jane Kidnie starts "Life / Afterlife" (pp. 203-220) by sketching the argument of Greg that a phenomenon that Wilson diagnosed as a sign of additional passages being added to a play (what Wilson called 'repetition brackets') can also be a sign of a passage being marked for omission from a play. (Kim's comprehensive exploration of this phenomenon is reviewed above.)
The standard narrative is that with Hamlet the passages found only in the 1604-5 second quarto (hereafter Q2) are ones that got cut to make the version underlying the Folio rather than being passages that got added to make the version underlying Q2. But what about the three Folio-only passages? Kidnie considers why these are not in Q2.
The first Folio-only passage is where Hamlet says that he regrets insulting Laertes at Ophelia's grave because he now realizes that their cases of grief are alike and plans to apologize. By contrast, in a Q2-only passage Hamlet is told by a messenger that Gertrude wants him to apologize to Laertes.
Kidnie reports that the 1603 first quarto (hereafter Q1) "includes a fragment of each of these passages separately printed in Q2 and F" (p. 207). But this is not so. Q1 has indeed got a couple of lines that convey the Folio-only motivation for Hamlet apologizing to Laertes: "it greeues mee much Horatio,| That to Leartes I forgot my selfe" but it has nothing that is equivalent to the Q2-only motivation that Gertrude wants Hamlet reconciled to Laertes.
Certainly, Q1 has a gentlemen (equivalent to Osric) say that Claudius and Gertrude and the court are arriving, but he says nothing of Gertrude wanting a reconciliation. Thus I cannot understand why Kidnie concludes that Q1 "complicates any suggestion that these F-only and Q2-only passages in 5.2 always were, and must necessarily be understood as, textual alternatives" (p. 207).
A second Folio-only passage is where Hamlet tells Rosencrantz and Guildenstern that Denmark is like a prison. That is, where Q2 has "Then is Doomes day neere, but your newes is not true;| But in the beaten way of friendship, what make you at Elsonoure?", the Folio has "Then is Doomesday neere: But your newes is not true. Let me question . . . " followed by 30 Folio-only lines ending with "but in the beaten way of friendship,| What make you at Elsonower?"
Another Folio-only passage is in the same scene, where Hamlet alludes to boy playing companies. Where Q2 has "No indeede are they not. Ham. It is not very strange, for my Vncle is King of Denmarke", the Folio has "No indeed, they are not. Ham. How comes it? doe they grow rusty?" followed by 26 Folio-only lines ending with "Hercules & his load too. Ham. It is not strange: for mine Vnckle is King of Denmarke".
After discussion of the Q2 and Folio versions of these moments, and what Q1 has at the same places, Kidnie offers no new claims or argument: she just concludes that we cannot reliably tell an addition from a cut. (Kim's essay reviewed above, however, concludes that repetition braces occur only where there are additions or false starts.) Near the end of her essay Kidnie alludes to "Q2's doubled stage direction in the final scene" (p. 214) but it is not clear what she means as she has not referred to this stage direction earlier in her essay.
Perhaps Kidnie means the stage direction that follows what Osric calls Hamlet's palpable hit in the fencing: "Drum, trumpets and shot.| Florish, a peece goes off". Kidnie discussed this stage direction in an essay (reviewed in NYWES for 2020) on Q2 being based on a theatrical document and perhaps she once intended to discuss it in this essay too. Kidnie ends by remarking that the New Bibliographers were unduly certain about what they inferred from printed books and manuscripts and that we now need to be more circumspect.
In "Book / Theatre" (pp. 223-244), Holger Schott Syme claims that once the 1623 Shakespeare Folio appeared ". . . all his plays but The Two Noble Kinsmen had a textual identity and were available, at least in principle, to actors and audiences independently of performance" (p. 224). This seems obviously untrue.
There is irrefutable evidence that Shakespeare and Fletcher's Cardenio had been performed but we have no evidence that it was printed, and Sir Thomas More likewise was not then in print. It could be argued that Shakespeare made only a small contribution to Sir Thomas More, but the same can be said for 1 Henry 6 that the Folio attributes to Shakespeare alone.
Syme traces how various versions of plays bifurcated across the eighteenth century, such as the individual Hamlets that were advertised as being based on what was being acted in London and those appearing in the major complete-works editions. Syme details a war of words between Robert Walker, who claimed theatrical provenance for his editions, and Jacob Tonson, who claimed to have the best texts. Syme then turns to what David Garrick did to prepare the book for his Hamlet of 1772, and the vagaries of subsequent eighteenth-century editions.
"Fact / Fictions" (pp. 281-298) by Adam G. Hooks tells the story of the fragment of an otherwise lost edition of The Passionate Pilgrim that turned up in the 1930s and the discussion about whether this lost edition preceded or followed the 1599 edition. From this, Hooks launches a familiar (and essentially correct) narrative about how the New Bibliographers were over-confident and built what they thought were facts upon what were really suppositions. Hooks then turns to the forgeries of John Payne Collier and the recent novel The Tragedy of Arthur by Arthur Phillips.
In "Black / White" (pp. 319-342 ), Miles P. Grier claims that as Marcus in Titus Andronicus says the words "'behold the child" (meaning Tamora and Aaron's) he is "holding aloft" (p. 320) the baby. He repeats the claim in an endnote (". . . what matters most is that this progeny of black and white is raised on high as a banner and a warning", p. 333n4), and again at the end of his essay: ". . . Marcus holds the progeny of black and white aloft . . ." (p. 332). Actually, nothing in the script indicates that the baby is held up by anyone.
Marcus's general attitude towards the child is not indicated in the script. But Grier thinks that Marcus's words "Of this was Tamora deliverèd,| The issue of an irreligious Moor,| Chief architect and plotter of these woes" can be read as suggesting that "the issue of his [Aaron's] union with Tamora [i.e. the baby] is 'chief architect and plotter' of the tragedy" (p. 320). It is hard to credit that Marcus really accuses the baby of planning the terrible events of the play; he obviously means that Aaron did.
Grier assumes that Cleopatra in Antony and Cleopatra was played by a boy in black makeup and that this is why she says "seek no colour for your going", which is her way of denying him a kiss that would stain his face. But the script does not indicate an attempt at a kiss here and the word 'colour' means simply 'pretext' or 'excuse'. Moreover, Cleopatra's being black is hard to reconcile with Antony calling her fair and her hand white, and Caesar chlding her for blushing.
When Caesar calls Cleopatra's children "unlawful issue", Grier detects a pun on 'issue' in the sense of print publication, "incalculable circulating pages in which Antony's tainted character can be read" (p. 324). Certainly, Shakespeare sometimes likens sexual reproduction to printing, but this seems an obscure example. Grier remarks that he could find other examples of ink and paper standing in for sex: "I could argue that The Tempest analogizes Prospero's books and his daughter Miranda as objects that he must protect from falling into the hands of the (painted) Algerian Caliban" (p. 325).
Grier explains that Tamora was not pregnant at the start of Titus Andronicus, since Saturninus would not have chosen her if she seemed to be. Grier thinks that there is not enough time for Tamora to get pregnant and have Aaron's child by Act 4. Grier remarks that at the start of the play Tamora "need not, of course, be a virgin" (p. 327). Since she is accompanied by her sons, it is unclear where Grier detects any uncertainty on this point.
Grier thinks we need an explanation for "the critical impulse to say that she and Aaron have had a son" (p. 327). But it is not just the critics who say this but Aaron, telling Demetrius that "this boy" is his "brother" (4.2.87-89). Grier believes that when Lavinia and Bassianus find Aaron and Tamora together they see the letter that Aaron has given Tamora "in her hand" (p. 329). But again there is nothing in the script to substantiate this claim. They might see it, or else Tamora might have pocketed it.
Grier's argument seems to need Lavinia and Bassianus to see the letter in order to find that in this play white paper with black ink on it stands for miscegenation. Thus Grier reinterprets the stage direction that reads "Enter Nurse with a blackamoor child". Although the Nurse says that it is a baby and Aaron says that it is a baby, Grier proposes that what the Nurse really brings on is "not merely like the books but was made of (and made into) reading material" (p. 331).
That is, the Nurse brings in merely a bundle of rags, the material used to make paper. Grier locates a similar moment in The Winter's Tale when the Shepherd finds Perdita and remarks "A boy or a child, I wonder?", where of course 'child' is a dialectal word for 'girl'. Grier comments that "The inhumanity of the textual infant is evident in the fact that it does not even have a sex" (p. 332). But this child does have a sex -- the audience knows that it is the girl Perdita -- and the Shepherd is merely wondering what it is because he cannot see it yet.
Grier's surprising conclusion about Titus Andronicus is worth quoting at length:
Her [Tamora's] relationship with Aaron has never been consummated. There is no indication that nine months have passed from her capture to this delivery. What has passed, however, is a 'fatal plotted scroll' [the letter incriminating Titus's sons] from Aaron to Tamora, in their only direct interaction in the script. (p. 332)
According to Grier, it is this, the "purloined scroll" (p. 333), that Marcus holds up at the end of the play when he says "Behold the child".
Scott A. Trudell's "Extant / Ephemeral" (pp. 343-359) starts with the existence of a manuscript from 1540 with the lyrics and music to a song, including the words "as I am all ye shall be| Therefore come after and dance with me| For all a green willow is my garland". Trudell thinks this might be the Willow Song from Othello although "The words are not quite the same" (p. 343).
Indeed, all they have in common is "green willow [is | must be] my garland". This is, however, an especially rare collocation in printed books, the nearest match being "green willow garland" in Stephen Batman's The Golden Book of Leaden Gods of 1577, STC 1583. None of this is Trudell's discovery but has been known for decades.
As Trudell goes on to acknowledge, we have the words and music of the Willow Song from before Othello was written, so the 1540 manuscript is not the origin of the song. Trudell argues that the complex relationship between songs in the plays and songs circulating in the wider culture (especially popular ballads) makes them "neither extant nor ephemeral but extempore, literally 'out of the time'" (p. 344).
Trudell goes on to consider Jaques's creation of additional lines for Amiens's song in As You Like It and Arviragus and Guiderius's speaking of a funeral song for Fidele in Cymbeline, both being reworkings of existing songs. According to Trudell, songs were always open to reworking. Whatever we have in the printed script -- such as the speaking of a song in Cymbeline -- is just a fossilization of a particular way of doing it in one production, not something fixed by Shakespeare.
Shakespeare knew that "Variability in performance conditions was the rule . . ." (p. 350) and hence he had no "single, uniform intention" (p. 351) about songs. According to Trudell, this indeterminacy is what he means by songs being neither quite extant nor ephemeral but rather "on a continuum" between the two (p. 352).
In a similar vein is "Lost / Found" (pp. 360-382) by Misha Teramura. Some plays lie between the conditions of existing and not existing, such as Cardenio which we somewhat have via Theobald's Double Falsehood. And of course there is slipperiness in our 'having' Hamlet since each of the three early editions contains things the other two lack.
Teramura sketches the cases of Love's Labour's Won, which might be an actually lost Shakespeare play or just an alternative title for one we have, and the lost 1597 edition of Love's Labour's Lost. According to Teramura, the fact that early modern printings are never perfect means that all plays are to some extent lost, and of course it is possible that the authorial manuscript itself contained errors and left certain details incomplete.
In "Paper / Ink" (pp. 383-401), Emma Depledge sketches the ways that paper and ink are used metaphorically in the drama -- skin being like paper, sin being like ink, for instance -- and gives an account of some of Shakespeare's depictions of anxieties about the trustworthiness of writing. She also sketches how handwritten and printed documents are used in some of his plays and the various metaphors and similes he constructs from the mechanics of writing. Depledge ends with an account of some past and recent discoveries arising from close study of the paper in printed books, especially the Jaggard-Pavier Quartos, and she is refreshingly willing to credit the New Bibliography with generating new knowledge this way.
The final essy in the book is "Material / Digital" (pp. 401-423) by Zachary Lesser and Whitney Trettien. A signature that looks like Shakespeare's appears on an exemplar of Laurence Nowell's Archaionomia (1568). Lesser and Trettien detail how they were misled by a digital image of this book that was taken when the book was disbound.
Then they turn to how microfilms, which are the origins of the images in EEBO, used to be viewed on analogue equipment and how the aspect ratio of the film readers differs from all the varying standards for computer-screen aspect ratios. Computer screens have got wider over time and purveyors of digital images of books are more likely now than they used to be to present book images as openings -- two pages side by side -- rather than single pages.
Lesser and Trettieu go into considerable detail about particular pieces of image-viewing software used by Folger Shakespeare Library. Software that automatically puts two page images side by side on the screen can, if you are not careful, give the appearance of an opening that never existed, bringing together images from different books (p. 411) or from one book but with "recto to the left of verso" (p. 418).
Historically, libraries have been divided on whether to catalogue book facsimiles such as microfilm as if they were instances of the original book or as separate objects in their own right. Lesser and Trettieu think that the latter approach is better. That is, ". . . we need more metadata that treats digital objects as textual objects in themselves . . ." (p. 416)
Four relevant chapters in books that otherwise do not concern us appeared this year. In the most significant, Jason Hogue argues that the play underlying the 1595 octavo of Richard Duke of York was revised to make the play underlying Folio 3 Henry 6, with particular attention to developing the play's imagery of plants ("The Botanical Revisions of 3 Henry VI", in Staub, ed. Shakespeare's Botanical Imagination, pp. 243-66).
There are many more plant images in Folio 3 Henry 6 than in octavo Richard Duke of York, and these appear not only in the whole sections that the Folio has that the octavo lacks but also in variant choices of words in sections that both editions have. This suggests that the Folio version represents a revision of the play underlying the octavo rather than the octavo being based on a memorial reconstruction (or some other derivative corruption) of the script underlying the Folio, or an abridgement of it. As Hogue asks, "Why go to the trouble of intentionally scrubbing botanical terms [in the Folio] for non-botanical ones [in the octavo]?" (p. 245).
Hogue risks overstating his case when he claims that before the Folio the history plays "were not necessarily . . . even thought of as being related to one another in genre, series, trilogy, or tetralogy" (p. 246). In fact, The Contention of York and Lancaster and Richard Duke of York were thought of as belonging together in that they were printed as The Whole Contention in the 1619 Jaggard-Pavier quartos collection. And, as Amy Lidster shows in the book briefly reviewed above, thinking of history plays as serial works preceded the 1623 Folio.
Among Hogue's many examples of expanded botanical language is "lets pull him down" (octavo) becoming "lets pluck him down" (Folio), the word 'pluck' being especially suited to plants. Hogue has two more such 'pluck' (Folio) for 'pull' (octavo) examples, used in particular for deposing a monarch. Hogue's argument is that the image is of plucking fruit from a tree, although this does not seem an especially good metaphor for uprooting a usurper.
The Folio adds 41 lines to the soliloquy for Richard Gloucester in the octavo, and they are full of botanical imagery, with Richard calling himself a "shrub" and a "trunk". (The latter could of course just mean torso.) Richard also says that he is surrounded by a "thorny wood". Hogue notes the Folio using the more botanical "wither" in place of the octavo's "perish".
Hogue finds multiple examples of the cutting down of great warriors being likened to the chopping down of great trees, and this appears more frequently in the Folio than the octavo. In exhorting his brothers not to show respect to the body of the slain Clifford, Richard of Gloucester in the Folio uses the imagery of "lopped", "hewing", "leaves put forth", "the root", and "tender spray", all referring to Clifford's victims among the house of York, and not present in the octavo.
Queen Margaret speaks of the remembrance of one's own children inducing mercy when killing enemies in a war, using in the Folio but not the octavo the phrase "How sweet a plant have you valiantly cropped". And then there is the famous change in the final scene, from Richard of Gloucester's "And that I loue the fruit from whence thou| Sprangst, witnesse the louing kisse I giue the child" in the octavo to "And that I loue the tree fro<m> whence yu sprang'st:| Witnesse the louing kisse I giue the Fruite" in the Folio, which makes the thing even more botanical.
Sent to France to win the hand of Lady Bona for King Edward, Warwick refers to Edward's love as "an eternall plant" in the octavo and "an externall Plant" in the Folio, and modern editions tend to use the octavo's reading. Hogue wonders if in fact "externall" might be right in the sense that King Edward's love is the above-ground (and hence external) manifestation of a feeling of which the "root was fixed in virtue's ground", as Warwick says in the next line.
Hogue spoils the end of his essay by offering incoherent remarks about the merits of not picking one or other reading where there are variants. Instead, ". . . a revision of an earlier text can be thought of as a kind of proliferation of ideas, which flourish variously depending on the given textual 'environment'" (p. 262). Hogue is explicit that he promotes a deconstructionist approach to editing.
Thus according to Hogue "There is no 'true' or 'correct' text, no meaning 'behind' texts, no intention that might be sought or discovered" (p. 263). These final remarks undermine his preceding scholarship, since there is, by this final claim, no intention in the revision of the play underlying the octavo to make the play underlying the Folio.
Two of this year's chapters in books that otherwise do not concern us are by Amy Lidster. The first should have been noticed in NYWES for 2021. In it, Lidster argues that the publishers Nathaniel Butter and Matthew Law set up the area near Saint Austin's Gate in Saint Paul's Churchyard as a place where buyers knew that certain kinds of politically topical books would be for sale ("Publishing King Lear (1608) at the Sign of the Pied Bull", in Altman & Buckner, eds. Old Saint Paul's and Culture, pp. 293-318).
The printer of the 1608 first quarto of King Lear was Nicholas Okes, but the less-studied figure is Nathaniel Butter at the sign of the Pied Bull near Saint Austin's Gate in Saint Paul's Churchyard. Butter and John Busby entered King Lear in the Stationers' Register on 26 November 1607. Butter went on to become one of the first serial news publishers in England and was mocked by Jonson (amongst others) for it. Butter specialized in anti-Catholic and topical publications.
Lidster speculates about what in King Lear might have appealed to Butter as potentially worth publishing, and alights on its topicality regarding the uniting of dissevered kingdoms. The particularly long and detailed imprint of the 1608 King Lear -- "Printed for Nathaniel Butter, and are to be sold at his shop in Pauls Church-yard at the signe of the Pide Bull neere St Austins Gate. 1608" -- suggests to Lidster that Butter wanted his name and his shop's identity to be strongly associated with this publication.
Only Butter and Matthew Law were in the Saint Austin's Gate area and they established it as a place for strongly Protestant political publications. In 1603-8, the Saint Austin's Gate area was the main place to buy new Shakespeare editions. Lidster proposes parallels between the fourth quarto of Richard 2 published by Law in 1608, with its addition of the deposition episode, and Butter's publication of King Lear in 1608.
In the Richard 2 deposition episode, the notion of the disposed king as nothing is emphasized. Also, the deposition episode switches to the speech prefix "Rich." instead of "King", emphasizing his loss of power. Lidster rather hedges her bets, though: "While it is not my intention to propose that the parliament scene was added because of its parallels with King Lear and Jacobean politics, the simultaneous publication of these two plays in 1608 at neighbouring bookshops establishes interpretative connections between them" (p. 310).
Lidster's second chapter-in-book concerns some variants between the first quarto (Q1, 1596) and the second quarto (Q2, 1599) of the play Edward 3, to which Shakespeare contributed ("Making Sense of Error in Commercial Drama: The Case of Edward III", in Candal, Grafton & Sachet, eds. Printing and Misprinting: A Companion to Mistakes and In-house Corrections in Renaissance Europe (1450-1650), pp. 415-28). Lidster begins with the fact that the first errata list in a printed playbook from the professional stage appears at the beginning of the 1602 edition of Thomas Dekker's Satiromastix.
After Satiromastix, only five more plays were printed with errata lists before 1650, and a few more have discussions of errors without lists of them, but these include also errors of judgement by the author. To make the Q2 Edward 3, the printers did not consult a new manuscript or the manuscript used to print Q1 but rather they relied on an exemplar of Q1.
From the Q1/Q2 differences we can spot what publishers thought was worth correcting. The corrector seems not to have known the proper noun 'Hellespont', since Q2 repeats Q1's error of "hellie spout" as "helly spout". Lidster itemizes some corrections that clarify such things as ambiguous speech prefixes and odd arrangements of stage directions.
Some small changes of wording are read by Lidster as having perhaps more significance than they deserve. For instance, where in Q1 Prince Edward calls French King John 2 "the king" when talking to a French herald, in Q2 he calls him "thy king", which Lidster reads as a change highlighting the play's interrogation of just who is the rightful king.
Likewise, the people of Calais will not open "their gates" in Q1 but "the gates" in Q2, which Lidster understands as "a variant which tempers the suggestion that the town and its defences belong to the people and complements the sections of the play where even the French characters admit the correctness of Edward's claim" (p. 424). Speaking to Edward, Queen Phillipa calls the people of France "her people" in Q1 but "thy people" in Q2. Lidster admits that it is hard to distinguish meaningful differences that separate Q1 and Q2 from merely unintentional variation.
Our last chapter-in-book this year comes from the same collection as Lidster's and also concerns the topic of error in printed plays ("Misprinting and Misreading in The Comedy of Errors", in Candal, Grafton & Sachet, eds. Printing and Misprinting: A Companion to Mistakes and In-house Corrections in Renaissance Europe (1450-1650), pp. 399-414). Here, Alice Leonard recapitulates the argument in her book Shakespeare in Error: Error in Shakespeare (reviewed in NYWES for 2020) that the many misprints in Folio The Comedy of Errors should be relished because the play is about confusion. One of her examples is the Folio's misprinting of "chalkie" (meaning chalky) as "chalkle", and she remarks that "the alternative 'chalkle' is not particularly meaningful in itself" (p. 410). In fact it is not meaningful at all.
Leonard contrasts some such errors that are, in her view, unproductive with ones we should retain, such as the Folio's speech-prefix confusion about the play's twins, by which "the audience is placed in a position of sustained mistaken identity" (p. 410). This version of Leonard's argument is no less frivolous than the version in her monograph, nor has she corrected her misspelling of Edmond Malone's name. But then, the principles she promotes are inimical to correction.
Finally, to the round-up from Notes and Queries. We start with authorship attribution matters and two notes by Thomas Merriam about King John being co-authored by Shakespeare and Peele. In the first, Merriam describes an unpublished experiment by the present reviewer that divided King John into multiple 2000-word 500-step rolling-window blocks and used the Word Adjacency Network method to attribute each block to one of six authors (Thomas Merriam "Co-authorship in King John: Further Evidence", N&Q 268[2023] 66-72).
Blocks 5-8 (from 1.1.239 to 2.1.392) and Block 16 (from 3.1.39 to 3.1.283) were most like Peele, Blocks 9-10 (from 2.1.205 to 2.1.513) and Blocks 34-37 (from 4.3.97 to 5.6.4) were most like Marlowe and the rest most like Shakespeare. Merriam replicates my division of King John into 2000-word, 500-step blocks, but wisely omits the stage directions, so the extents of his blocks do not quite match mine.
Because of a unique verbal parallel of ten words (with Peele's The Troublesome Reign of King John), Merriam gives Blocks 9-10 to Peele instead of Marlowe. But otherwise my work largely agrees with Merriam's prior attributions.
Merriam then runs a new experiment. He uses Peele's The Arraignment of Paris, The Battle of Alcazar, David and Bethsabe, Edward 1, and The Troublesome Reign of King John. For the Shakespeare set he uses 31 Folio plays, being all of them except the three Henry 6 plays, Titus Andronicus, and Henry 5. This means he includes as Shakespeare's work the undoubtedly co-authored Timon of Athens and Henry 8, and also the disputed Measure for Measure, Macbeth, and All's Well that Ends Well.
Merriam counts the frequencies of the 206 most-frequent words found in the combined Peele and Shakespeare training sets and applies Principal Component Analysis to the 206 results. Principal Component 1 accounts for 19.4% of the variation and puts all the Peele plays on the left and quite distinct from all the Shakespeare plays on the right. Unfortunately, Merriam does not present this result as a visualization or table.
Then Merriam tries the 2328 most-frequent character 4-grams found in the Peele and Shakespeare sets and in two further texts: Blocks 5-10 and Blocks 19-25 of the rolling-windows-divided King John. In a Principal Component Analysis, Blocks 5-10 fall nearest the Peele texts and Blocks 19-25 nearest the Shakespeare. Merriam considers this as further evidence that King John was a Shakespeare and Peele collaboration and that this helps explain its inconsistent position on the legitimacy of King John's reign.
In his second note on this topic. Merriam takes as his training data the Peele plays The Battle of Alcazar, David and Bethsabe, Edward 1, The Arraignment of Paris, The Troublesome Reign of King John, and Titus Andronicus Act 1 and Act 2 Scene 1, and the Shakespeare plays The Tempest, The Merry Wives of Windsor, As You Like It, A Midsummer Night's Dream, The Comedy of Errors, and The Taming of the Shrew ("King John is Written By Shakespeare and Peele", N&Q 268[2023] 250-3). His test data is King John divided into two parts: the bits he thinks are by Peele and the bits he thinks are by Shakespeare.
The six Shakespeare plays were chosen "for their relative shortness in order to compare more fairly with the two divisions of King John" (p. 251). Merriam counts the 198 most-frequent words in all of them (100% culling) and plots the results on a Principal Component Analysis scatterplot. Principal Component 1, accounting for 26.3% of the variation, entirely separates the known Peele and Shakespeare works across the horizontal axis.
This Principal Component puts the bits of King John that Merriam thinks are Peele with the known Peele works and the bits of King John that Merriam thinks are Shakespeare with the known Shakespeare works. Merriam details how the play displays conflicting attitudes towards the moral rectitude of King John and the Pope's representative Pandulph.
Then Merriam divides the play into 26 sections along the same lines as he did in his essays "Feminine endings in King John" (reviewed in YWES for 2009) and "Feminine Endings and Function Words in King John" (reviewed in NYWES for 2016). It would have been helpful for him to have actually stated the divisions here.
In these 26 sections Merriam counts the frequencies of the words 'and' and 'I' and plots their percentages along the x and y axes of a scatterplot respectively, labelling each data point as P[eele] or [S]hakespeare according to his prior view. There is a distinct clustering of Peele in the bottom-right corner (high 'and', low 'I') and of Shakespeare in the top-left corner (low 'and', high 'I'), but with some overlap.
Moreover, the two centroids of these two clouds of points are "close to the global averages for Peele and Shakespeare" (p. 253). For Merriam this is further corroboration of his claim that King John is co-authored by Shakespeare and Peele.
Merriam published a third note this year on a different topic: the end of Titus Andronicus being by Peele not Shakespeare ("'The Limitations of Stylometry': Unconscious Versus Conscious Use of Frequent Words", N&Q 268[2023] 65-6). In "Titus Andronicus: Peele, Shakespeare, Peele" (reviewed in NYWES for 2022), Merriam showed that the frequencies of 'and' at the end of Titus Andronicus are at the same anomously high rate as they are at the beginning, which is widely accepted to be by Peele.
On page 214 of his book Shakespeare, Co-Author (reviewed in YWES for 2002), Vickers accepted Jackson using 'and' frequency as a marker of authorship. Later, in his essay "The Limitations of Stylometry" (reviewed in NYWES for 2022), Vicker claimed that we cannot use this marker because Shakespeare sometimes consciously overused 'and' for rhetorical effect.
In response, Merriam reruns his tests for most-frequent-words and most-frequent-character-tri-grams on texts in which he has deleted all the occurrences of 'and', and he gets the same result: the beginning and end of Titus Andronicus test like Peele, the middle like Shakespeare.
In 2017, Michael Gilleland identified the Latin source for the distinctive five-line speech "This is no age . . . do melt away" in Sir Thomas More, which are lines 8.210 to 8.214 in Jowett's edition. John Klause attempts to build on this discovery to show that Shakespeare was responsible for this part of the play, and not Dekker as Jowett thougt ("Shakespeare, Mantuan, and The Book of Sir Thomas More", N&Q 268[2023] 164-5). The source is the Eclogues of the Italian poet Baptista Mantuanus (born 1447, died 1516) at that work's lines 5.153-158.
Klause finds that the sentiment about the proper object of the Muse's attention in Mantuanus's eclogue and in Sir Thomas More is the same as in Shakespeare Sonnet 100. I would say that Sonnet 100 is about beauty inspiring the Muse whereas the point in Mantuanus and Sir Thomas More is about heroism inspiring the muse, so the connexion is only loose.
Klause lists some phrases in Sir Thomas More 8.120-214 that he thinks distinctly Shakespearian. Here are Klause's phrases with his Shakespearian claims for them and the other plays I have found them in, with the word 'perhaps' used where authorship is uncertain:
"melt away" Klause finds this in Shakespeare's Titus Andronicus, Richard 2, and Coriolanus. I find that it also occurs in Jonson's Every Man in his Humour, Epicoene, and Bartholomew Fair.
"make you master of my" Klause finds this in Shakespeare's The Comedy of Errors but there it is "make him master of thy". I find that it also occurs in Francis Beaumont and Fletcher's The Scornful Lady.
"we'll banquet" Klause finds this in Shakespeare's 1 Henry 6 but there it is "let us banquet". I find that it also occurs as "we banquet" in Marlowe's 1 Tamburlaine and Thomas Heywood's The Brazen Age and as "let vs banquet" in Marlowe's 1 Tamburlaine.
"fresh delight" Klause finds this in Shakespeare's Romeo and Juliet but there it is "such delight| Among fresh fennel buds". I find that it also occurs as "fresh appetite| To my delights" in Beaumont, Fletcher and Philip Massinger's Love's Cure, as "to breed fresh delight" in Beaumont and Fletcher's The Woman Hater, and as "Fresh and delight" in John Mason's The Turk.
"staid (as an adjective)" Klause finds this in Shakespeare's Cymbeline but there it is "my staider senses". I find that it also occurs as "some staid Chamber-maid" in Fletcher's The Woman's Prize, as "more staid respect" in Fletcher and (perhaps) James Shirley's Wit Without Money, as "quiet speritts, staid, and temperate" in Fletcher and Massinger's Sir John van Olden Barnavelt, as "A staid, and mature judgement" in Nathan Field, Fletcher, and Massinger's The Knight of Malta, as "my staid affections" in Fletcher and (perhaps) Beaumont's The Noble Gentleman, as "Staid judgement" in Beaumont, Fletcher, and Massinger's Thierry and Theodoret, as "staid, discreet, and understanding people", as "the most staid, and knowing" in Fletcher's Valentinian, and as "staid discretion" in Fletcher's The Humorous Lieutenant.
"cheer our sprites" Klause finds this in Shakespeare and Middleton's Macbeth but there it is "cheer we up his sprites". I find that it also occurs as "cheere thy hoping sprites" in Dekker's The Shoemaker's Holiday.
"the cates must be but mean" Klause finds this in Shakespeare's The Comedy of Errors but there it is "though my cates be mean". I find that it also occurs as "Meane cates are welcome" in Jonson's Every Man Out of His Humour.
"How now, master" Klaus finds this in Shakespeare and Peele's Titus Andronicus but there it is "how now, my masters", and in Shakespeare's The Merry Wives of Windsor, and his Antony and Cleopatra, but there it is "How now, masters?" I find that it also occurs as "how now, Master Thomas" in W. S.'s Thomas Lord Cromwell, as "How now master Strumbo" in W. S.'s Locrine, as "how now master Gumwater?" in Middleton's A Mad World, My Masters, as "How now master Browne?" in Anonymous's A Warning for Fair Women, as "How how Master Churmes" in Anonymous's Wily Beguiled, as "How now, master Kno'wel" in Jonson's Every Man In His Humour, and as "how now Master Little-wit" in his Bartholomew Fair.
"a suitor to" Klause finds this in Shakespeare's The Taming of the Shrew. I find that it also occurs in Webster The Devil's Law Case, Jonson's Bartholomew Fair and A Tale of a Tub, and Cyril Tourneur's The Atheist's Tragedy, and (perhaps) John Cumber's The Two Merry Milkmaids.
"the fellow with" Klause finds this in Shakespeare's 2 Henry 4. I find that it also occurs in Beaumont and Fletcher's The Coxcomb, and Jonson's Every Man Out of His Humour.
"long hair" Klause finds this in Shakespeare's Venus and Adonis but there it is "long, dishevelled hair". I find that it also occurs in Munday and Chettle's The Death of Robert, Earl of Huntingdon, and (perhaps) Heywood's How a Man May Choose a Good Wife from a Bad, and Jonson's The Staple of News and Bartholomew Fair.
"I'll hear you then" Klause finds this in Shakespeare's The Winter's Tale but there it is "I'll hear you by and by". I find that it also occurs as "I wil heare you anone" in George Chapman, Jonson, and Marston's Eastward Ho!, and as "I will heare you out" in Beaumont and Fletcher's A King and No King.
"the foolish knave" Klause finds this in Shakespeare's The Taming of the Shrew. I find that it also occurs in Scene 3 of Anonymous's Arden of Faversham (in a part of the play not suspected of being by Shakespeare), and in Lording Barry's Ram Alley.
"submitted himself" Klause finds this in Shakespeare's King John but there it is "submit himself". I find that it also occurs as "submit himself" in Robert Greene's Alphonsus, Emperor of Germany, Kyd's Cornelia, Peele's The Battle of Alcazar, and Chapman's All Fools, and as "submitting himself" in John Lyly's Midas.
"to the mercy of" Klaus finds this in Shakespeare's A Midsummer Night's Dream, and 2 Henry 6 but omitted it occurring in Shakespeare and Fletcher's The Two Noble Kinsmen and Henry 8. I find that it also occurs in Marston's Antonio and Mellida, (perhaps) John Ford's The Laws of Candy, (perhaps) William Rowley and (perhaps) Heywood's The Thracian Wonder, Anonymous's Lust's Dominion and Look About You, Marlowe's The Jew of Malta, Fletcher and (perhaps) James Shirley's Wit Without Money, Heywood's The Silver Age, Fletcher and Massinger's The Little French Lawyer, Beaumont and Fletcher's Philaster, Field, Fletcher, and Massinger's The Knight of Malta, Chapman's May Day, and Chettle and John Day's The Blind Beggar of Bethnal Green.
"make a . . . vow before" Klause finds this in Shakespeare's Hamlet but there it is "Makes vow before". I find that it also occurs as "made by vow| Which should be fought before" in Anonymous's Clyomon and Clamydes.
"live civil" Klause finds this in Shakespeare's The Merry Wives of Windsor but there it is "I live again, but in honest, civil". I find that it also occurs as "liue with him bee ciuill" in Beaumont and Fletcher's The Scornful Lady, and as "Live then I say famous for civill" in Fletcher's The Mad Lover.
It seems to me that Klause has not shown any distinctive Shakespearian phrasing in Sir Thomas More in the lines around the quotation from Mantuanus.
It might be argued that writing from Shakespeare can be paralleled with every one of the above phrases while no other playwright has parallels with all of them. However, this is not sound logic since the phrases in the above list were chosen precisely because they can all be parallelled from Shakespeare. For even-handedness we would have to hunt for every phrase in a series of lines from another part of Sir Thomas More and count whose canons they are also found in.
Klause ends by showing Shakespeare explicitly quoting Mantuanus in Love's Labour's Lost 4.2 and loosely paraphrasing the beginning of his sixth eclogue in his Winter's song at the close of that play. This shows that Shakespeare knew Mantuanus but not that he wrote more of Sir Thomas More than we thought.
Our last note on the subject of authorship is Darren Freebury-Jones's argument that Kyd was the sole author of the play Arden of Faversham ("The Stylistic Unity of Arden of Faversham: Shakespeare, Thomas Kyd, and Pragmatic Markers", N&Q 268[2023] 274-7). Freebury-Jones agrees with Salkeld's article on Arden of Faversham (reviewed above) in its claim that the play had a single author, but he disagrees that the single author was Shakespeare.
Freebury-Jones claims a remarkable 98% success rate for Rizvi's experiments in 2018 at "correctly identifying authors" (p. 274) and gives an URL for Rizvi's website. But in fact Rizvi on his website does not claim 98% accuracy overall, and indeed it is clear from Rizvi's results that the real author of any particular work is found far less frequently than this by his method. Perhaps here Freebury-Jones is doing what he does elsewhere in treating as a correct attribution any occurrence of Rizvi's method putting the correct author of a play into the top 12 rank order of the most-likely candidates.
Freebury-Jones makes some of the same criticisms of Salkeld's article that this review makes. One is that of Salkeld using a larger Shakespeare canon than Kyd canon, for the purposes of counting matching words and phrases with Arden of Faversham. Another is that Salkeld should not have included co-authored plays in the 'ground truth' Shakespeare canon.
Freebury-Jones works his way through Salkeld's evidence, finding that Salkeld overlooked phrasing that Arden of Faversham shares with Kyd plays. In part, Freebury-Jones relies on phrases shared by Arden of Faversham and King Leir because the latter has been "long attributed" to Kyd (p. 275) and appears in his own edition of Kyd's works.
Freebury-Jones finds it significant that "The phrase 'To let thee know'" found in Arden of Faversham "cannot be found in Shakespeare’s entire dramatic corpus" (p. 276). But in fact, with variation in the direct object, it can: "to let Beatrice know" (Much Ado About Nothing), "To let you know" and "To let them know" (Hamlet), and "To let you know" (Othello).
Freebury-Jones disputes Salkeld's count of 16 occurrences of "How now" in Arden of Faversham, finding that there are only 15. But Salkeld is right, and, with the New Oxford Shakespeare edition's line numbering, these are: #1 "How now, Adam" (1.106); #2 "How now, Michael" (1.141); #3 "How now, Alice" (1.535); #4 "How now, Clarke" (1.588); #5 "How now, fellow Bradshaw" (2.14); #6 "How now, Michael" (3.18); #7 "How now, Master Shakebag" (3.122-123); #8 "How now, Will" (5.18); #9 "How now, Alice" (8.45); #10 "How now, Bradshaw" (8.151); #11 "How now, Michael" (10.49); #12 "how now? Who is this" (12.22-23); #13 "How now, what's the matter" (14.255); #14 "How now, Mistress Arden" (14.267); #15 "How now, Alice" (14.328); and #16 "How now, Master Mayor" (14.357).
Freebury-Jones's wider point about single authorship is that the features he tracks are evenly spread across the play -- it is "stylistically homogenous" (p. 277) -- whereas we would expect co-authorship to produce unevenness. But the evidence from "How now" rather contradicts this claim, since this feature is not evenly distributed across the play: half of the occurrences (eight of the 16) come from just two scenes: the first and the 14th.
The play Locrine was attributed to "W. S." on the title page of its first edition, a quarto of 1595, and was included in the second impression of the Shakespeare Third Folio in 1664. Nobody takes this attribution to Shakespeare seriously now, but who did write it? Freebury-Jones thinks it was Greene, but Jackson argues against this ("Locrine and Robert Greene's Dramatic Canon", N&Q 268[2023] 154-9).
Freebury-Jones used Rizvi's Collocations and N-grams (CAN) dataset to make his attribution, and Jackson helpfully explains its contents and terminology. Jackson finds CAN useful if you employ all 527 early plays it contains, but Freebury-Jones confined himself to plays from certain year-spans.
Jackson shows that using Freebury-Jones's own data, "Each canonical play by Greene has at least one of the other three ranked within the twelve closest matches" (p. 156). But Locrine and the play Selimus do not have one of the canonical Greene plays in their top 12 matches. Locrine has three Marlowes in the top 12, so perhaps he wrote it. The play that shares most matches with Locrine is Selimus and the play having the most matches with Selimus is Locrine.
Jackson explores some other verbal evidence -- including some counts of repetitive phrasing made by F. G. Hubbard -- and it too points away from Greene because Locrine has much more of it than was usual for him. Jackson explores some of the other quantitative evidence for the play's style, and for its possible early date (before 1586).
Brian Vickers presents his solution to an old problem ("The Spanish Tragedy: Now in Five Acts!", N&Q 268[2023] 249-50). The 1592 quarto of The Spanish Tragedy divides the play into four acts. Editors since Greg have understood that the third act, at 1580 lines (almost half the play), should have been divided, and that the label "Actus Quartus" should have been "Actus Quintus". But where to put the break in Act 3?
Vickers argues that an act break should mark a sequence of actions ended and a new sequence begun. Thus in his edition of The Spanish Tragedy he puts the break between the existing scenes 3.6 and 3.7 so that Act 3 ends with the hanging of Pedringano and Act 4 starts with Hieronimo's soliloquy "Where shall I run to breathe abroad my woes".
Regarding the same play, Vickers argues that what has been taken as an error in transmission is in fact intentional and expressive ("Epistrophe and the 'Lost Line' in The Spanish Tragedy 20.7.23", N&Q 268[2023] 160-4). The Spanish Tragedy contains this pair of lines in Bel-Imperia's complaint to Hieronimo:
With what dishonour, and the hate of men,
From this dishonour and the hate of men?
Most editions treat this as an accidental near-duplication and suggest an alternative reading for one of the lines. Vickers defends the repetition as a meaningful epistrophe: repetition at the ends of adjacent or nearby lines or sentences. Vickers shows some examples of Kyd's use of epistrophe.
Emelye Keyser thinks that the prologue to Henry 5 was inspired by some writing by King James ("Shakespeare's Muse of Fire", N&Q 268[2023] 72-4). The prologue to Dekker's Old Fortunatus is much like the opening chorus to Henry 5 in drawing attention to the playhouse being too small for the space depicted in the play and asking for the audience's imagination to make up the difference.
If the chorus was present in Henry 5 as originally written then it was the inspiration for Dekker's prologue. But Keyser thinks that King James's Essays of a Prentice, written in the mid-1580s, might be the source for both. In the first few sonnets in this collection, James asks Jove to help him create in words the grand visual images he wants his poetry to conjure, an effect that George Puttenham in The Art of English Poesie called hypotyposis. Keyser finds James's reference to Jove's "might and thundring fyre" close to Shakespeare's "muse of fire" in the Henry 5 prologue.
Keyser also finds James's "when combats manyfold| I do descriue, [let readers] see two champions braue| With armies huge approaching" to be like the "imagined armies of France and England that Shakespeare's Chorus conjures" (p. 74). Dekker's prologue refers to "the laws of poesy" and James's book includes his treatise on Scottish poetic laws. We know that William Cecil, Henry Carey, and Gabriel Harvey each had a copy of James's book, so Shakespeare and Dekker had access to it.
Rhema Mei Lan Hokama revisits Shakespeare's use of the term 'Cathayan' ("Shakespeare's Cathayans: Twelfth Night, Maritime Exchange, and Early Modern China", N&Q 268[2023] 254-9). Shakespeare uses it once in The Merry Wives of Windsor and once in Twelfth Night. In the latter, Maria requests Sir Toby to keep quiet because otherwise "my lady" (meaning Olivia) will call on Malvolio to eject him. Sir Toby replies "My lady's a Cathayan, we are politicians".
Timothy Billings, in an essay reviewed in YWES for 2003, argued that in explaining 'Cathayan' as a strongly derogatory term eighteenth-century editors projected their own racial prejudices onto the Elizabethans. But Billings, Hokama argues, managed only to show that the word was somewhat less pejorative than eighteenth-century editors made out: it still connoted "trickery and deceit" (p. 256).
In an note reviewed in YWES for 2013, Zhiyan Zhang argued that in fact Sir Toby is being approving of Olivia in calling her a Cathayan. Hokama thinks this overstated: Sir Toby does mean it as a criticism. David Bevington argued that Shakespeare was the writer who first made 'Cathayan' into a pejorative term.
Hokama thinks that the point of the remark in Twelfth Night is that it compares ". . . Olivia's refusal to admit marriage suitors to her court to China's infamous practice of refusing to admit European merchants and sailors . . " (p. 257). Hokama supports this claim of Chinese rejection of merchants from accounts reproduced in Richard Hakluyt's Principal Navigations.
The Chinese permitted trade with docked foreign ships, but would not let the Europeans off their ships. Likewise, Olivia lets suitors stay at her house but will not give them access to herself. Even before she is seen, the play makes repeated reference to the difficulty of envoys getting an audience with Olivia.
Roger Stritmatter presents two notes using the Bible to illuminate Shakespeare. In the first he argues that the Parable of the Talents is alluded to in the plot of Measure for Measure ("'In the Ambush of My Name': The 'Parable of the Talents' in Measure for Measure", N&Q 268[2023] 74-7). The Duke is like the estate owner who leaves his wealth to his servants to manage, but also he is like the servant who buried the one talent he was given. The word 'talent' meant a coin, as does the word 'Angelo'.
Stritmatter also hears the Parable of the Talents in the Duke's lines "Heaven doth with us . . . glory of a creditor" in the opening scene, but does not spell out the likeness. He finds some other minor biblical allusions in the speech too.
The play is full of the imagery of coinage, such as "test . . . of my mettle| Before . . . a figure | Be stamped upon it", and that too is a link to the coin called a talent. Stritmatter finds that "Duke Vincentio and his reluctant deputy Angelo suffer from the same absence of faith as the third servant who buried his talent in the ground" (p. 76), but Stritmatter does not explain what that absence of faith consists of or is shown by.
In his second note, Stritmatter argues that Shakespeare repeatedly alludes to 2 Corinthians 11.14 ("'Satan is an Angel of Light': II Corinthians 11.14 and 'Ocular Proof' in Shakespeare", N&Q 268[2023] 77-9). This note starts as a survey of references to vision -- especially mistaken vision -- in Shakespeare, but becomes just a list of moments when Shakespeare's characters reflect on the nature of deceit.
Stritmatter writers that "Angelo soliloquizes, 'O cunning enemy . . ." and that "Isabella responds by describing her tormentor as . . ." (p. 78). But that cannot be called her response since she cannot have heard Angelo's speech, as it was a soliloquy, and she is speaking 342 lines (about 17 minutes) later.
Stritmatter repeatedly quotes lines from Shakespeare that he says are indebted to the biblical remark that "Satan himself is transformed into an Angel of light" (2 Corinthians 11.14). For instance, he finds Shakespeare "evidently recalling" (p. 178) this part of the bible in Antonio's remark in The Merchant of Venice that "The devil can cite Scripture for his purpose . . . what a good outside falsehood hath!". I cannot see a link beyond the fact that both mention the devil.
Stritmatter then quotes 25 lines from The Merchant of Venice that he attributes to Measure for Measure and supports this with a reference to a 29-line span of the play; he offers no remark on the lines or explanation of what they are doing in his note. Overall, Stritmatter claims 17 Shakespearean allusions to 2 Corithians 11.14, which together show that Shakespeare's great interest was "the discrepancy between truth and appearance" (p. 79).
Zixian Xu finds a source for part of Sonnet 19 in Edmund Spenser's work ("The Brood-devouring Earth: A Possible Source for the First Quatrain of Shakespeare's Sonnet 19", N&Q 268[2023] 16-7). The second of the following lines of Sonnet 19 is tricky:
Devouring time, blunt thou the lion's paws,
And make the earth devour her own sweet brood;
Pluck the keen teeth from the fierce tiger's jaws,
And burn the long-lived phoenix in her blood.
What is that "sweet brood"? Also, the second line does not fit well with the other three, which are about time weakening or taming wild things.
Xu finds the explanation for this in Spenser's The Visions of Petrarch, in which the fourth sonnet mentions a spring that is "sweetly" echoed (by Muses and nymphs) and which "the gaping earth" did "devour". In the previous sonnet Spenser mentions lions and in the next he mentions the phoenix.
Robert Easting thinks that Shakespeare got the idea of vanishing into thin air, in The Tempest, from Virgil's Aeneid ("'Into Thin Air': Shakespeare's The Tempest and Virgil", N&Q 268[2023] 79-80). Easting reports that the OED credits The Tempest with "the earliest recorded use of the now commonplace expression, to vanish 'into thin air'" (p. 79). This is a misunderstanding of the OED's use of quotations, which are meant to be illustrative but not record the earliest use of a word or phrase.
Easting finds a source for Shakespeare's phrase in Virgil's Aeneid: "in tenuem ex oculis evanuit auram" (meaning "it vanished from sight into thin air"). Easting finds the same image in Virgil's Georgics and wonders whether Shakespeare got it from there. Since The Tempest is elsewhere indebted to Virgil's Aeneid, Easting reckons that Shakespeare got the phrasing about vanishing into air from there.
But in fact, as Easting acknowledges, Shakespeare's phrasing is "melted into air, into thin air" not "vanished . . .". Moreover, the phrase "vanished away into the thin air" appeared in Konrad Lykosthenes's The Doom Warning All Men in 1581, so it was not unknown.
In a second note, Easting finds the source for Prince Escalus's "The sun for sorrow will not show his head" in Romeo and Juliet in Virgil's Georgics: "The sun, when Caesar fell, had sympathy for Rome--| That day he hid the brightness of his head in a rusty fog" ("'The Sun for Sorrow Will Not Show His Head': Shakespeare's Romeo and Juliet and Virgil", N&Q 268[2023] 174-5). Virgil uses the word 'caput' (Latin for 'head'), just as Shakespeare uses 'head'. (A claimed source in Ovid's Metamorphoses has instead 'vultus', meaning 'face'.) With Romeo banished to Mantua, it would have been natural for Shakespeare to think of Virgil, who was from there.
Brett Jones has a solution to a longstanding puzzle ("Twelfth Night's MOAI: Le Solution Pour Maria's 'Fustian Riddle'", N&Q 268[2023] 170-4). In Twelfth Night, the letters 'M', 'O', 'A', 'I' that Malvolio reads aloud are, Jones claims, meant to sound like 'M, O, et I' (using the French 'et' for English 'and') to spell out the French word 'moi' meaning 'me'. Thus Malvolio can read the note as referring to himself. Jones thinks that their ongoing commentary as Malvolio tries to figure out the riddle shows that Sir Toby and Fabian have cracked the puzzle, but the linguistically challenged Andrew Aguecheek cannot.
Jones finds this essentially a repeat of the earlier pun in "her very c's, her u's, and her t's, and thus makes she her great P's", in which the word 'and' stands for the letter 'n' so that Malvolio spells out 'cunt', from which Olivia pees. In both riddles, the third letter of four is given by 'and'.
Jones acknowledges that he is not the first to see 'moi' as the solution, but no one has spotted the 'A' = 'et' part of the solution before. In the Folio text, Maria earlier says of Malvolio that she will "gull him into an ayword" and editors since Nicholas Rowe have emended that last word to 'nayword', meaning a byword. That is, Maria says that Malvolio will make his own name into a byword for stupidity. Jones thinks that we should not emend, since Maria in fact means that she will gull him into a 'I-word', a word ('moi') that refers to himself.
B. J. Sokol shows that Shakespeare knew the works of Euripides ("Euripides' Bacchae and Shakespeare's Theseus in A Midsummer Night's Dream", N&Q 268[2023] 166-70). In A Midsummer Night's Dream, Theseus rejects the proferred entertainment of "The riot of the tipsy bacchanals| Tearing the Thracian singer [Orpheus] in their rage", saying that it was played when he came from Thebes as a conqueror.
Sokol points out that neither Plutarch (in his Life of Theseus) nor Euripides (in The Suppliant Women) says that Theseus set out to actually conquer Thebes. Rather, Theseus merely wanted to recover the bodies of the slain warriors that Creon refused to release for burial.
In Euripides' Bacchae the Theban king Pentheus is torn apart by Maenads, as Orpheus in the entertainment rejected by Shakespeare's Theseus is torn apart by Bacchantes. Like Pentheus, Shakespeare's Theseus denies supernatural forces. Later, Shakespeare's Theseus redeems himself by accepting at least the power of imagination to do strange things.
Earlier in A Midsummer Night's Dream, Hippolyta anachronistically mentions hunting with Cadmus, who was grandfather to Bacchus and Pentheus and who plays an important role in Euripides' Bacchae. Sokol suggests from all this that Shakespeare knew Euripides' works rather better than we have thought.
Lisa Hopkins has a source for something obscure said in Hamlet ("'Now Let the Strucken Deer Go Weep': What Does it Mean or Think?", N&Q 268[2023] 254). Hamlet says "Why, let the strucken deer go weep,| The hart ungallèd play;| For some must watch while some must sleep,| Thus runs the world away". It has been conjectured that here Hamlet is quoting a ballad but none suitable has been found.
Hopkins thinks that Shakespeare is remembering the ballad "Blow Thy Horn, Hunter" by William Cornysh in the court of Henry 8. The links are that Cornysh's song is about an oversexed woman (as Hamlet thinks his mother is), it comes from the court of Henry 8 (who raised the question of whether a man may marry his brother's widow), and it explicitly asks the reader to make an interpretation (as Hamlet asks Horatio). Also, its line "Now the construction of the same -- what do you mean or think" is a galloping fourteener, as is Hamlet's speech, and both have 'struc' as the stressed fourth syllable.
Ceri Sullivan explicates a remark in As You Like It ("Orlando's Beard and Love Cures in As You Like It", N&Q 268[2023] 15-6). Rosalind-as-Ganymede says that Orlando lacks the marks of a man in love, including "a beard neglected, which you have not", and then she adds "but I pardon you for that, for simply your having in beard is a younger brother's revenue".
In her assumed disguise as Ganymede, Rosalind should not know that Orlando is a younger brother. Sullivan thinks this a double insult, since Orlando lacks a neglected beard (so, is not in love) and almost lacks a beard at all (so, is not fully a man). The insult motivates Orlando's acceptance of Ganymede's proferred cure for love.
Lastly, Akihiro Yamada lights an ignus fatuus that leads Vickers astray ("'Maimed, and Deformed': Shakespeare's First Folio and Beaumont and Fletcher's Philaster", N&Q 268[2023] 18). Yamada declares his purpose to be to "suggest that the famous phrase 'maimed, and deformed' in Heminge and Condell’s address 'To the great Variety of Readers' of the First Folio of Shakespeare published in 1623 originates in Thomas Walkley’s address 'To the Reader' of the second edition of Beaumont and Fletcher's Philaster published in 1622".
Walkley used the phrase "maimed and deformed" to describe the poor state of the text of the previous edition (the first) of Philaster, and of course in the Folio preliminaries the phrase is used to describe the poor state of the text of the preceding editions of Shakespeare's plays. Yamada draws attention to this because, he thinks, ". . . none of so many editors of the play and the Folio, to the best of my knowledge, has ever publicly discussed the relation between the Folio editors' address and Walkley's".
In fact, Suzanne Gossett in her 2009 Arden Early Modern Drama edition of Philaster discussed at length this phrase "maimed and deformed" providing a connexion between Philaster and the 1623 Shakespeare Folio (her pages 76-102). An additional Shakespearian link is that Walkley also published Othello in 1622, and Gossett pursued this too.
Shortly after Yamada's note was published, Vickers published a letter in the Times Literary Supplement based on his (Vickers's) misunderstanding of it. Vickers thought that Yamada was claiming that ". . . the phrase 'maimed, and deformed' was first used by Thomas Walkley in the preface to his 1622 edition of Beaumont and Fletcher's Philaster, or Love lies a-bleeding'" (Times Literary Supplement, Letters 24 March 2023).
For Vickers, the phrase "maimed and deformed" being used by Walkley about the manifestly inferior first edition of Philaster and about pre-Folio editions of Shakespeare was evidence that there really was a category of "unauthorized [bad] quartos" even though this is "doubted by some modern scholars".
In fact, Yamada nowhere claims that Walkley's was the first use of the phrase "maimed and deformed", which is just as well because it was not. The phrase was common, especially in religious texts. A Martin Marprelate pamphlet of 1589 used it three times (STC 17456, sig. D4r, E1r, E1v). Richard Bancroft used it in his 1593 Survey of the Pretended Holy Discipline (STC 1352, sig. Ee3v). Richard Hooker used it in his 1604 Laws of Ecclesiastical Polity (STC 13713, sig. H6r). Nicholas Byfield used it in his 1618 Cure of the Fear of Death (STC 4213, sig. D12r). And John Heigham used it in his 1622 Treatise of Auricular Confession (STC 13036, sig. N12r)
Books reviewed
Altman, Shanyn and Jonathan Buckner, eds. Old Saint Paul's and Culture, Early Modern Literature in History. Palgrave Macmillan [2021]. 366 pp., £107.40, ISBN 978-3030772697
Bourne, Claire M. L., ed. Shakespeare / Text: Contemporary Readings in Textual Studies, Editing and Performance, Arden Shakespeare Intersections. Bloomsbury [2023]. 464 pp., £33.81, ISBN 978-1350344556
Candal, Geri Della Rocca de, Anthony Grafton and Paolo Sachet, eds. Printing and Misprinting: A Companion to Mistakes and In-house Corrections in Renaissance Europe (1450-1650). Oxford University Press [2023]. 600 pp., £105, ISBN 978-0198863045
Lidster, Amy Publishing the History Play in the Time of Shakespeare: Stationers Shaping a Genre. Cambridge University Press [2022]. 300 pp., £24, ISBN 978-1316517253
Salzman, Paul Facsimiles and the History of Shakespeare Editing, Cambridge Elements: Elements in Shakespeare and Text. Cambridge University Press [2023]. 114 pp., £17.10, ISBN 978-1009228244
Staub, Susan C., ed. Shakespeare's Botanical Imagination. Routledge [2023]. 302 pp., £42.99, ISBN 978-9463721332
Stern, Tiffany Shakespeare, Malone and the Problems of Chronology, Cambridge Elements: Elements in Shakespeare and Text. Cambridge University Press [2023]. 110 pp., £17.10, ISBN 978-1009224727
Syme, Holger Schott Theatre History, Attribution Studies, and the Question of Evidence, Cambridge Elements: Elements in Shakespeare and Text. Cambridge University Press [2023]. 110 pp., £17.10, ISBN 978-1009227414