The Shakespeare Canon and the NOS

The Shakespeare Conference: SHK 30.008  Saturday, 5 January 2019

 

[1] From:        Jeffrey Robert Wilson <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 4, 2019 at 12:47:12 PM EST

     Subj:         Shakestats 

 

[2] From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 4, 2019 at 1:17:23 PM EST

     Subj:         Re: SHAKSPER: NOS 

 

[3] From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 5, 2019 at 1:44:30 AM EST

     Subj:         Re: NOS 

 

[4] From:        Pervez Rizvi <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 5, 2019 at 8:23:29 AM EST

     Subj:         Re: The Shakespeare Canon and the NOS 

 

[5] From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 5, 2019 at 8:36:35 AM EST

     Subj:         Re: SHAKSPER: NOS 

 

 

[1]-----------------------------------------------------------------

From:        Jeffrey Robert Wilson <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 4, 2019 at 12:47:12 PM EST

Subject:    Shakestats

 

SHAKSPEReans,

 

If you find our on-going discussion on computer-aided Shakespeare studies a little flat, I invite you to take a look at an essay of mine published over the holiday break:

 

Shakestats: Writing About Shakespeare Between the Humanities and the Social Sciences Early Modern Literary Studies 20.2 (2018)

https://extra.shu.ac.uk/emls/journal/index.php/emls/article/view/395

 

I’m quite enjoying the authorship conversation. The point of the article, however, is that statistics and computers allow us to ask the more important - literary - questions about interpretation, meaning, and value that brought us to literary studies in the first place. 

 

Regards,

 

Jeffrey R. Wilson, Ph.D.

Harvard University

This email address is being protected from spambots. You need JavaScript enabled to view it. 

 

[2]-----------------------------------------------------------------

From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 4, 2019 at 1:17:23 PM EST

Subject:    Re: SHAKSPER: NOS

 

Dear SHAKSPERians

 

Pervez Rizvi continues to object to the idea that naming a person (printer or publisher) in the siglum that stands for an early edition of Shakespeare (as in, say, 1DANTER) is more useful than naming the format (as in Q1). The logic of doing so is that it privileges human agency over book format, but Rizvi thinks that this logic is “wishful thinking” since “We have no way of knowing” whether the scribe, compositors, proof-reader, printer, publisher, or reporter of a memorial reconstruction was “the most significant agent”.

 

The defence of using human names doesn’t rest on the claim that we can figure out in each case who was “the most significant agent”. It rests on preferring to name human agents over formats. The agency of scribes, compositors, proof-readers, printers and publishers was also at work in the making of the editions of the 18th-century and later, but Rizvi does not object to anyone referring to those by sigla such as POPE, MALONE, STEEVENS, CHAMBERS, and so on. Would he really prefer that these were given sigla such as Q17, F19, and so on? If not, what is the objection to an early edition being named for its printer or publisher? One possible objection is that we have no reasonable way to decide whether to use the printer or the publisher’s name. I think there is a reasonable way.

 

Rizvi asserts that “It was arbitrary of the NOS editors to single out the printer for some books and the publisher for others”. But the rationale for switching from printer (for first editions) and publisher (for reprints) is not arbitrary: it rests on the uses the modern editor will make of these editions. Any readings that we decide to adopt from a reprint are departures of that reprint from its copy text and more likely to come from the publisher than the printer. That’s not arbitrary. It’s a statement about the balance of probabilities. It is not an assertion that we have worked out the contributions of all the agents and chosen the most significant.

 

Rizvi remarks that in the 1986-87 Oxford Complete Works, “Jowett renamed all the quartos of ‘1 Henry IV’”. Rizvi is referring to the anomalous situation in which the late discovery of part of the first edition of this play caused scholars to label this fragment Q0 so as not to disturb the existing designations of Q1, Q2, Q3 and so on. That is, in the scheme that Rizvi is defending (against Jowett’s revision of it), Q1 was actually the second edition, Q2 was the third, Q3 was the fourth, and so on. This illogical scheme was rightly rejected by Jowett. Far from being an example of the Oxford editors having “form” (slang for previous criminal convictions) “for this kind of thing” (that is, the arbitrary renaming of editions), I would offer this as an example of the absolute necessity of changing our nomenclature when the facts as we know them require it. Sticking with nomenclature that has merely the weight of tradition to recommend it leads scholars to defend absurdities such as using Q1 to denote the second edition of a work.

 

Rizvi says he will read Chiaki Hanabusa’s account of the shared printing of the first edition of ‘Romeo and Juliet’, but even before doing so he thinks it “a bit of a stretch to say that Danter was ‘the leading agent’” in this collaboration, since Allde printed more than half of the book. It really isn’t a matter of who printed the most sheets; that is not what is meant by taking the lead.

 

I’m grateful to Rizvi for pointing SHAKSPERians to Darren Freebury-Jones’s response to my review of his (Freebury-Jones’s) article responding to Gary Taylor and John V. Nance’s article on distinguishing imitation from collaboration. Freebury-Jones makes two valuable corrections to my review. The correct name of his article (co-written with Marcus Dahl) is “The limitations of microattribution” not “The limits of microattribution” and the correct name of the journal in which it appeared was” Texas Studies in Literature and Language” not “Texas Studies in Language and Literature”.

 

It was generous of Freebury-Jones’s to attribute these errors to my work being “hastily-written”, but the truth is that I’m quite capable of such howlers even in my most carefully considered writing. I’m grateful to have them pointed out to me so I can correct them before the review appears. The rest of Freebury-Jones’s response engages with nothing of substance from my review so I’m happy for readers to be able to compare the review as it stands with Freebury-Jones’s response to it and to make their own judgements.

 

Regards

Gabriel Egan

 

[3]-----------------------------------------------------------------

From:        Gerald E. Downs <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 5, 2019 at 1:44:30 AM EST

Subject:    Re: NOS

 

I happened onto this podcast (first time for everything) from the Folger about NOS attribution, from which transcript I’ve removed some function and lexical words for brevity.

 

Michael Witmore, the Folger’s director.

 

In the New Oxford Shakespeare, the plays Henry VI, Parts 1, 2, and 3 would no longer be listed as having been written by Shakespeare alone. Instead the title pages will say: “By William Shakespeare and Christopher Marlowe.”

 

As you'll hear, advances in computer science have enabled scholars to find, with much greater certainty, the fingerprints that we think tell us definitively who wrote which plays, and even who wrote which acts . . . throwing open the idea that Shakespeare was a solo genius in charge of it all.

 

Throwing open the idea that fingerprints come from fingers, or DNA from cigars.

 

RASMUSSEN: Well . . . it used to be that people were assigning shares of Henry VI to playwrights other than Shakespeare because they thought they were bad.

 

Overstated, but some people think they’re bad, and some are.

 

RASMUSSEN: Well . . . . you know, Marlowe writes Tamburlaine and Shakespeare says, “Wow. . . . And then he writes Henry VI then, and Marlowe says, “Wow . . .”

 

And we know . . . Shakespeare collaborated on . . . Arden of Faversham . . . . So this is the nice... this is a nice way to think . . .

 

WITMORE: I think . . . that multiple methods have now converged, so that we can say things about authorship right down to the level of act within plays. . . . There have been editors who’ve . . . said, “You know . . . .” But at this point . . . using scannable texts, where we compare patterns within the words of each writer, have yielded the same conclusions. And . . . it's a moment when computers and human beings are working basically in parallel . . .

 

BOGAEV: And that is a technological difference. Before we get deeper into that, if we could look at the marketing aspect of this. . . . [LAUGHS]

 

RASMUSSEN: [LAUGHS] Let's talk about marketing.

 

BOGAEV: Is it also important because this was a major publication taking a stance on this subject?

 

BOGAEV: And Oxford taking this different direction on the authorship issue—there's a pretty long history there, isn’t there?

 

WITMORE: Well, I think the Oxford editors were bold in saying even that different editions—a quarto edition or a Folio edition of Shakespeare’s plays—were distinct works. . . . And in the case of Lear, you know, one published edition is entitled The Tragedy of King Lear and the other uses the word History.

 

Wow . . .

 

BOGAEV: And . . . this idea of the business of Shakespeare. This whole topic made me wonder who has hats in the ring, in this fight over whether Shakespeare wrote alone or collaborated. And this gets us into this issue of something or someone called disintegrators, correct, Eric?

 

RASMUSSEN: Sure. . . . And that has then evolved, largely due to influences like the Oxford Shakespeare, who said in their first edition that “Hey, Shakespeare probably collaborated . . . and we know that he did. I've always like that idea that beginning in his career, especially, he's collaborating as sort of apprentice playwright. And then at the end of his career he's collaborating . . . .

 

In between he wrote and revised alone.

 

BOGAEV: And Mike, do you agree with this... with Eric, on the players and their investment in answering this authorship question one way or another?[?]

 

WITMORE: I do. It's also probably the case that there's more prestige to lose than to gain. . . . So, in a way, I think those of us who study Shakespeare and talk about him enter an environment where everyone has something to gain, and there are institutions and publishing houses that make the bottom line by selling their editions. But it does matter when someone can show why they think what they think. And I think Eric’s right to say that one of the interesting things about the Oxford attribution is that we've got a different way of saying that it's true now and not only they have made this case, but now others have done it as well.

 

BOGAEV: . . . And for those of us who aren’t Shakespeare scholars, it's really hard for us to conceive of a time when the idea of a solo genius just didn’t wash. It didn’t compute.

 

BOGAEV: Well, it's interesting . . . scholars have approached this authorship question by looking at the First Folio.

 

J A G G A R D, for those who are Shakespeare scholars.

 

BOGAEV: Okay. . . . do you agree that those actors, they were probably good witnesses and authorities when it comes to “Oh, is this a solo job versus this wasn’t?” Eric, I know you have... I can hear you having thoughts there.

 

RASMUSSEN: Well I think it's... it is fascinating that, you know, when Ben Jonson brought out his folio . . . . And, you know, we know, for instance . . . .

 

Having said that, many people will now say that . . . . And I just... it's fascinating that after the publication of the First Folio [I A G G A R D}, when the Second Folio [I F O R G E T] came out in 1632, the publishers, the Cotes brothers, also published, at the same time, the First Quarto edition of The Two Noble Kinsmen [1 C O T E O R 2?] and a reprint of Pericles. [3 C O T E S] And I'll bet anything that they were bundling these collaborative plays along with the Folio. So you get the... you sort of get the . . . more or less, in the Folio volume.

 

BOGAEV: Mike, does it also come down to this… idea of a collaboration that was different then? And Heminges and Condell might have had a different conception of what a collaboration is?

 

WITMORE: . . . And it's very hard for us to roll back the tape on history and say, “Well, would it have been noteworthy for them to even comment on it?” We're certainly interested in it, but that doesn’t mean that they were.

 

H & C did say Qs were “stolen & surreptitious” and I A G G A R D the cure; we’re certainly not interested in what they did say, but what did they know?

 

WITMORE: What about the writers' room? I mean you could imagine them going in, you know . . . you know . . . it's very interesting to think about how they might have gotten together and said, “You know . . . .” But the whole idea that they might be egging each other on, learning from each other, I think is exciting . . . .

 

We can imagine what we know, as we are reminded daily.

 

BOGAEV: . . . Mike, I know this is up your alley . . . can you please explain what an algorithm is, and how these algorithms help scholars tell whose voice we're reading. How do they work?

 

WITMORE: Well, an algorithm is a very boring kind of robot that goes in and counts things and comes back and says, “I saw . . . and here’s where I saw them.” And it allows you to take an entire play or a scene or all the works by a given author and say, “Here, in quantitative terms, are the kinds of features that I see in this writer’s writing.” And those features could be anything from a unique word that no one else uses or a phrase that is more likely to occur in someone’s writing than someone else’s. It could also be a mundane common word like “of,” “the,” “for,” or “to”—words that happen over and over again. We just can't do without them. . . . And they are only as good as the sample they are given.

 

But we know the samples are good because “now we know.”

 

And I think that’s important for people to understand. Unless there had been scholars who had gone though the works of Marlowe and Shakespeare and others and said, “This is a play by Shakespeare,” you wouldn’t be able to create the sample, which is what the Oxford team did, that you use to train this little algorithm or robot and say, “Look at this, look at this, are you noticing any patterns?” But the algorithm will always honor the sample and so you have to have a human being in there somewhere saying, “Okay, here’s your sample of Shakespeare, here’s your sample of Marlowe, learn everything you can. Now I'm going to show you something that we don’t have a decision on. You tell me what it's closer to.”

 

The Wizard of O Z ?

 

BOGAEV: And let me see if I understand this correctly, because you're saying the algorithm keeps track in a quantitative way in which humans can't really... like our ear is not attuned to the “tos,” and the “ofs,” and the “fors,” and the “thes,” these very boring words that amount to—when you look at it mathematically—they amount to Shakespeare.

 

WITMORE: I think that’s a great explanation. There's only so much you can consciously attend to, so having an algorithm...

 

You can be unconscious, and have time to teach.

 

BOGAEV: Well, it reminds me of an incredibly scientific handwriting analysis, [& which is that?] but I'm thinking . . . a playwright . . . a great one, is going to be writing in all of these voices . . . . It's not like writing an essay or speech. . . . So how much of voice survives . . . ?

 

BOGAEV: Well, it's really fascinating to think that possibly you could be missing the person that everyone modeled his work after, right? . . . . And the algorithm only knows the playwrights that it sees. . . . Shakespeare likely collaborated with actors, and actors were so steeped in Marlowe . . . that the actors Shakespeare could have been collaborating with read like Marlowe.

 

WITMORE: That’s the area where we really have to think hard. . . . I think that’s where the trail gets lost.

 

BOGAEV: Okay, so this is a vulnerability of this form of analysis you're saying?

 

WITMORE: . . . But if you’ve got a great sample for two writers, and you’ve got an unknown one for a sample . . . that’s the ideal situation which you can say, “Given what you know already, tell me who you think wrote this?”

 

If you’ve got bad samples, then what? Siri? Alexa? One at a time, please—and don’t forget to replicate.

 

. . . . And it’s a fine point, but it's important, that there are some things we can know with these techniques, but they always depend first on a human being saying, “Here’s what I think the original samples mean.” And they also depend on our ability to take only the surviving texts. . . . These are powerful techniques, but there is a point in which they won’t work.

 

The techniques depend on human beings saying the techniques are good; what a human thinks is good may not be, first and last. Yet the techniques may ‘work’ for the “bottom line,” depending on marketing algorithms.

 

BOGAEV: And stepping out of the technical for a moment . . . .

 

RASMUSSEN: . . . But one of the limitations of the current attribution programs is that it can't distinguish between whether Marlowe and Shakespeare . . . would sit down in the writers’ room . . . or maybe Marlowe wrote the play on his own and then maybe Shakespeare came in some years later and revised it. . . . And I think this is . . . something that we've got to think about. . . . Well, is that a collaboration? . . . Is that an adaptation? And so the Oxford editors have hedged their bets in saying that the original of the Henry VI plays was probably written by Shakespeare and Marlowe and maybe Thomas Nashe and maybe others. And that Shakespeare then revised these base texts . . . .

 

The reader now “knows” what? My questions and observations are piling up. I’ll start with those “maimed” playtexts. There were agencies before 1 W I S E, U N W I S E, and M O S T U N W I S E got ahold of the Shakespeare Canon (1 T A Y L OR). At least I’m getting pretty good at the “Name Game.”

 

Gerald E. Downs

 

[4]-----------------------------------------------------------------

From:        Pervez Rizvi <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 5, 2019 at 8:23:29 AM EST

Subject:    Re: The Shakespeare Canon and the NOS

 

Larry Weiss wrote:

 

...I predict that high school and undergraduate English teachers are unlikely to assign it as the class text. 

 

It’s happened already to some extent. On YouTube, viewers will find a video of Anna Pruitt, one of the NOS associate editors, using the modern edition to teach Titus Andronicus to a class at her university. 

 

I liked Al Magary’s humour, to lighten the mood on what is unavoidably a tough thread. He wonders if “scholars have to keep a NOS quick-reference table in their pockets”. Many a true word is spoken in jest. More than once in the past I’ve had to make my own lookup table to help me convert traditional Globe-based act-scene references to the ones in the Oxford editions, old and new. As with the sigla for quartos, they changed the established scene boundaries, without a care for the impact on the people who actually use them. Trying to find a line in the Oxford editions by using a reference you’ve found in another book or essay is a bit like taking a Ryanair flight: sometimes you land a long way from where you want to be. 

 

Music scholars have managed to agree the K-numbers for Mozart’s works, and everyone uses them. They have been changed a few times in the past, but always to correct problems, not just because someone wanted to be iconoclastic. Meanwhile, Shakespeare editors are still messing about with book abbreviations, play titles, play title abbreviations, act-scene boundaries, and line numbering systems; and they have made things less, not more, settled in the past thirty-odd years. They are moving backwards and calling it cutting-edge scholarship. 

 

[5]-----------------------------------------------------------------

From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 5, 2019 at 8:36:35 AM EST

Subject:    Re: SHAKSPER: NOS

 

Dear SHAKSPERians

 

Larry Weiss is unconvinced that naming the printer or publisher of an early edition in the siglum is more useful than naming the format, although he hasn’t yet said why he finds the format information more useful. He sees a missed opportunity, though, since “it might be useful to identify the particular F1 compositor who set the portion of a play for which JAGGARD is cited, as their tendencies are fairly well known. That would look something like JAGGARD/B, or JAGGARD/D(?) if the assignment to Compositor D is uncertain. But NOS does nothing like that”.

 

The reason the New Oxford Shakespeare does nothing like that is that during its creation the general editors became aware of a remarkable essay that convinced them that almost all existing Folio compositor attributions are unreliable. That essay was:

 

Pervez Rizvi “The Use of Spellings for Compositor Attribution in the First Folio” Papers of the Bibliographical Society of America 110 (2016): 1-53

 

In the midst of all the dislike of the New Oxford Shakespeare being aired on SHAKSPER, it is a pleasant duty to acknowledge the high regard in which its general editors hold this essay by Rizvi.

 

Al Magary also doesn’t like the New Oxford Shakespeare’s sigla, remarking in relation to ‘Richard II’ that “Q1 is named for Simmes and Qs 2 & 3 for Wise” even though all three were printed by Simmes for Wise. I don’t have a new way of explaining that this is because editors rightly treat reprints differently from first editions. If someone would like to positively assert that reprints should be treated the same way as first editions, I’d be happy to engage with that argument. But just being “alarmed” at a naming practice, as Magary says he is, is not in itself a critique, so I’m given nothing to engage with here.

 

Brian Vickers simultaneously objects to the printed Datasets section of the New Oxford Shakespeare Authorship Companion being too small (“the provision of datasets is inadequate”) and in the case of one essay, Gary Taylor’s study of the manuscript “Rawlinson Poetry 160”, being too big (“no less than seven datasets”). This is wasted space, according to Vickers, because the dataset supports an essay about that manuscript’s attribution of “Shall I Die?” to Shakespeare. According to Vickers, there is no need to discuss this 17th-century attribution to Shakespeare since all one has to do is quote a bit of “Shall I Die?” to convince everyone that “such banal lines” aren’t Shakespeare’s.

 

Vickers’s method here is much like Jane Lee’s in her essay “On the Authorship of the Second and Third Parts of ‘Henry VI’” (Transactions of the New Shakspere Society 4 (1876): 219-79, 293-306. Lee did not count or tabulate the metrical features on which her argument rested. Instead—much like the famous proof of Pythagoras’s Theorem reproduced by Bhaskara of India in the twelfth century with no working out and just the caption “Behold!”—Lee merely quoted passages that she thought would “serve to illustrate what these metrical differences are” and left them unanalysed (Lee, p. 222).

 

The New Oxford Shakespeare rejects on principle the “Behold!” school of authorship attribution. There isn’t any point in trying to debate evidence and analysis with someone who adheres to this school since they privilege their sensibility over their reasoning.

 

Regards

Gabriel Egan

 

 

 

The Shakespeare Canon and the NOS

The Shakespeare Conference: SHK 30.007  Friday, 4 January 2019

 

[1] From:        Brian Vickers <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 3, 2019 at 12:05:37 PM EST

     Subj:         Re: SHAKSPER: NOS 

 

[2] From:        Larry Weiss <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 3, 2019 at 3:35:23 PM EST

     Subj:         Re: SHAKSPER: NOS 

 

[3] From:        Al Magary <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 3, 2019 at 4:37:15 PM EST

     Subj:         Re: SHAKSPER: NOS 

 

 

[1]-----------------------------------------------------------------

From:        Brian Vickers <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 3, 2019 at 12:05:37 PM EST

Subject:    Re: SHAKSPER: NOS 

 

Letter to Shaksper 3.1.19

 

Gentle SHAKSPERians,

 

A Happy New Year to you all!

 

I begin 2019 with some more complaints about Gabriel Egan’s defence of the NOS, first the variable provision of ‘Datasets’ in the Authorship Companion. Pervez Rizvi pointed out that only 5 of the 25 chapters were graced with additional documentation and that 

 

A reader of any chapter of the book is left to fend for herself if she wants to know whether the data is given somewhere in the chapter itself, is given in the Datasets section at the back, or not given at all.

 

Egan feels we should be grateful to OUP for the amount of space made available. My problems have to do both with those omitted and those included. For the former, consider the essay by Anna Pruitt: “Refining the LION Collocation Test: A Comparative Study of Authorship Test Results for Titus Andronicus 4.1” (AC, 92-106). Pruitt states that

 

The full dataset of my results is provided on the Oxford University Press website that supports the New Oxford Shakespeare edition (99)

 

Like Larry Weiss, I had assumed that, as an owner of the NOS (a review copy) I would be able to access this database, but this turned out not to be the case. I twice wrote to Dr Pruitt asking her to share her data with me, but she failed to reply. Fortunately, a kind editor at OUP allowed me access, and I was able to check her evidence for rejecting the ascription of Titus Andronicus 4.1. to Peele. 

 

Like so many NOS scholars, Pruitt had found enough matching phrases in the favoured author—here Shakespeare, elsewhere Middleton—but a distinct lack of evidence for the author attribution they were rejecting—here Peele, elsewhere Kyd or Dekker. As it happens, all three of whom were the authors I had suggested. In connection with which I should cite a passage from the Preface to the AC by Gary Taylor, John Jowett, Terri Bourus, and Gabriel Egan drawing attention to the fact that

 

In the course of this work, the contributors necessarily respond, time and again, to the most widely read and influential voice in the field of Shakespearean authorship attribution, Brian Vickers. Their repeated disagreements with his conclusions are a sign of a healthy, growing subject-discipline forming a consensus as new knowledge is created by different methodologies. (vi)

 

(Nothing personal, then.) Using anti-plagiarism software, recently supplemented by Pervez Rizvi’s database, I found that in her dataset Pruitt, a devoted follower of MacDonald Jackson’s use of LION to collect verbal matches, had somehow missed 33 matches with Peele in this scene of 121 lines. This oversight could be due to the deficiencies of the search engine, or it could be an instance of careless scholarship by a group convinced of their superior knowledge. I have found similar deficiencies in the work of Taylor, Nance, and Jackson.

 

While agreeing with Rizvi that the provision of datasets is inadequate, and with Weiss that they should be freely available, not hidden behind a paywall, Taylor and Egan privilege one contribution over all others in the provision of data, Taylor’s own essay 0n the Bodleian MS “Rawlinson Poetry 160: The Manuscript Source of Two Attributions to Shakespeare” (218-30). In this miscellaneous anthology an unknown scribe attributed  two poems to Shakespeare, one of them (beginning “When God was pleas’d”), I have never seen in print; the other is the notorious lyric, “Shall I Die?”, which Taylor came across in the Bodleian (its presence was not a secret), announcing in the Sunday Times in November 1985 that he had 

 

found the literary equivalent of Sleeping Beauty, a nameless poem awakening from the ancient sheets in which it had lain undisturbed for centuries…. [As soon as he had] finished copying the poem down … I felt in my guts that it was Shakespeare.

 

(cit. Brian Vickers, ‘Counterfeiting’ Shakespeare (2002), “Prologue: Gary Taylor finds a poem”, pp. 1-53). As most members of this forum will know, the response to Taylor’s gut feeling has been one of ridicule and universal disbelief that Shakespeare could ever have written such banal lines as these: 

 

Next her haire forehead faire

Smooth and high next doth lye without wrinckle 

Her faire browes under those…

 

Now, thirty-two years later, Taylor seeks to justify his youthful attribution in an essay that cites such trivial “evidence” as the number of other poems lacking the definite article the but containing the pleonastic do or doth. He then feels justified in adding no less than seven datasets (pp. 608-27), comprising a table of contents of the Bodley MS, an index of authors, an index of titles, an index of first lines, one “Chronological Index (by apparent date of composition)” and another “(by date of print publication)”, concluding with an index of “Rhyme-Parallels” with the Pyramus and Thisbe playlet in Midsummer Night’s Dream, divided into “comic” rhymes, “not comic”, “not obviously comic”, “an intentionally? ridiculous rhyme”, and one “tragic instance”. These twenty pages are irrelevant to an attribution that nobody takes seriously, and together with his essay amount to a third of the space allocated to supporting data. Taylor’s attempt to claim scholarly respectability is privileged at the expense of other contributors’ datasets and reduces OUP to the level of vanity publishing. How will Egan defend it?

 

[2]-----------------------------------------------------------------

From:        Larry Weiss <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 3, 2019 at 3:35:23 PM EST

Subject:    Re: SHAKSPER: NOS

 

I am grateful to Gabriel Egan for his detailed response to my last post. I have only a few clarifications to add for the sake of the record:

 

 

1. My dismay at the unavailability to individual scholars of the online edition is only partly a result of its cost. I might be willing to pay $500-600 for a lifetime subscription, but to ask an individual scholar to pay that much per year when s/he already has the complete print edition, and the digital version is identical except for correction of a few typos, pays no heed to the concept of economic utility. It was probably for that reason that OUP seems to have decided to restrict the digital version to libraries and other institutions, or at least it is pitching its sales efforts in that direction.

 

My main point is that it looks as if the OUP spiffed us when it encouraged us to buy the new edition by offering what I was given to understand was a lifetime online NOS subscription (or at least recurrent annual subscriptions at a fair price) if we ordered the Mod Crit Ed prior to the release date. Allow me to present my recollection of the history: At the 2016 SAA, a prototype or advance copy of the Mod Crit Ed was exhibited (but not sold, as it was still a work in progress), and the senior editor described the project in some detail and urged us to order a copy before the official release date as that would give us a free subscription to the online version of the entire four volumes. My recollection was that he said that this would be a lifetime subscription, but I might be misremembering that. He pointed out that the digital version would be corrected and updated more frequently than the print version could be, so it was useful for a serious scholar to have it at hand. I am not accusing the senior editor of deliberately misrepresenting the facts; he was probably just as misinformed as the rest of us. I believe there were also brochures and ads that offered or implied the same thing. What OUP has done strikes me as bad commercial practice, even apart from its ethics. The structure of the NOS is not likely to attract casual readers—its sheer weight is off-putting—and I predict that high school and undergraduate English teachers are unlikely to assign it as the class text. Its target audience must be the scholarly community, which increasingly relies on digitized texts.  So to make the online edition available only to institutions who can afford the steep annual cost will probably reduce sales of both the online and print editions.

 

2. I am not “confused” by the revised citation form for the substantive editions of the works; that term suggests a degree of cognitive impairment. The NOS revised style is cumbersome and unhelpful, especially to those of us who have spent a lifetime watching our Qs and Fs. Sure, the ordinal numerals preceding the names of the printers or publishers correspond to the Q numbers, but why should it be necessary or even helpful to change the convention. As for the folios, no numbers precede the names of the printers, so the reader will have to memorize that A L L O T is F2, etc. (Gabriel points out that we can identify the folios by consulting the list of Works Cited, which tells us, for example, that C H E T W I N D E refers to F3. But that is only useful if we start with the name of the editor; if we want to know how F3 is cited, we have to look for C H E T W I N D E, which is not intuitive. That is analogous to searching the definition of a word in the dictionary to find out what the word is.)

 

“Foregrounding” the contribution of the printer provides no information most textual or critical scholars use; if one printing house has a greater or lesser reputation for accuracy and care, that is a point to make in a monograph on the subject or a detailed analysis of the text of a particular work, it is of no use in a volume of collected works. Perhaps the idea is to give credit (or blame) where it is due, to the workmen who actually made the physical book. But, even a died-in-the-wool Marxist would have to acknowledge that the NOS style doesn’t accomplish that; the name of the printer doesn’t identify the compositor or typesetter, only the capitalist who owned the print shop. Even if the compositor were named, who after 400 years would feel better?

 

Arguably, it might be useful to identify the particular F1 compositor who set the portion of a play for which J A G G A R D is cited, as their tendencies are fairly well known. That would look something like J A G G A R D/ B, or J A G G A R D / D(?) if the assignment to Compositor D is uncertain. But NOS does nothing like that.

 

3. I observed in my prior post that NOS departs from the convention of citing a collected edition by the name of the general editor(s), and instead cites only the name of a editor of the particular work being considered. That is actually useful, and I don't object. Citing the general editor makes sense when he or she is actually responsible for the final product.  When the individual editors have authority to make the final decision, they should be cited. Where, as in the case of NOS, there are multiple general editors and only one reviews the contribution of an individual editor, it could be helpful to specify this in some fashion. The previous editions of the Oxford Shakespeare did this in its Textual Companion by signing the essays on the individual works with the initials of the main editor followed by those of the reviewing editor; for example, 1 Henry VI is signed "G.T./(J.J.)" which tells us that Gary Taylor edited the play and his work was reviewed by John Jowett.

 

[3]-----------------------------------------------------------------

From:        Al Magary <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 3, 2019 at 4:37:15 PM EST

Subject:    Re: SHAKSPER: NOS

 

Once again, I come as the nonspecialist on the list with my three cents. I used to be alarmed at anything Trump says, now I’m alarmed at anything the NOS editors say regarding their editorial methods. In particular, after reading co-editor Gabriel Egan’s last post (Jan. 3) about the renaming of, say, the Richard II quartos, including—

 

1SIMMES (1597), printed by Simmes for Wise = 'Q1'

2WISE (1598), printed by Simmes for Wise = 'Q2'

3WISE (1598), printed by Simmes for Wise = 'Q3'...

 

All are “printed by Simmes for Wise,” yet Q1 is named for Simmes and Qs 2 & 3 for Wise. Is this arcane renaming for Elizabethan unknowns really supposed to help scholars and all who may write about, and need to refer in a coherent way to, different editions—even students in AP English Lit? Will scholars have to keep a NOS quick-reference table in their pockets? But Gabriel says (twice), “I’m just not seeing the confusion that Weiss reports...” and “I’m not seeing how anyone would find this unusual or confusing.” The latter bafflement is after explaining about an edition of Richard II called JOHN. 

 

Luckily, I see Babbel.com, the online language learning folks, have added a new course, “How to Understand NOSese.” Here’s their blurb:

Unlock the new Newoxfdordshakespearespeak!

 

At Babbel, we worked for years with the editors of the New Oxford Shakespeare because we obsess over crafting the tools you need to completely relearn Shakespeare and start having practical, everyday conversations with your Shakespearean colleagues. We believe the sooner you begin to speak a new language, the sooner you’ll open yourself up to a world that’s bigger, richer and more inspiring.

 

Cheers,

Al Magary

 

 

 

The Shakespeare Canon and the NOS

The Shakespeare Conference: SHK 30.006  Thursday, 3 January 2019

 

[1] From:        Pervez Rizvi <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 3, 2019 at 3:48:20 AM EST

     Subj:         Re: The Shakespeare Canon and the NOS 

 

[2] From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 3, 2019 at 9:01:47 AM EST

     Subj:         Re: NOS 

 

 

[1]-----------------------------------------------------------------

From:        Pervez Rizvi <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 3, 2019 at 3:48:20 AM EST

Subject:    Re: The Shakespeare Canon and the NOS

 

Gabriel Egan says that “the siglum in each case acknowledges the human agency at work.” This is wishful thinking, reminiscent of New Bibliography. All the quartos and Folios come to us through the agency of several people, not just one: the scribe, if any; the compositors; and the proof-reader. We have no way of knowing which of these roles, if any, the master printer or the publisher played. If memorial reconstruction took place, for example for a play like Merry Wives, then the most significant agent was the reporter. It was arbitrary of the NOS editors to single out the printer for some books and the publisher for others and use his name alone for the siglum. 

 

By the way, the Oxford editors have form for this kind of thing. In the 1986 edition, Jowett renamed all the quartos of 1 Henry IV. They have now been renamed again. 

 

To answer Larry Weiss’ question, the Second Folio has been renamed ALLOT, the Third Folio has been renamed CHETWINDE, and the Fourth Folio has been renamed HERRINGMAN (see p.xiii). Observe that the numbers 2, 3 and 4 were not used, but then the arbitrary application of an arbitrary policy is not much to complain about.

 

Rizvi’s account of a textual mishap in the printing of the New Oxford Shakespeare is pure fantasy.

 

OK. I may continue to rely on the immortal words of Mandy Rice-Davies.

 

Rizvi has been misled by Brian Gibbons’s out-of-date account...

 

Thanks for the information about Hanabusa’s articles. I’ll be interested to read them.

 

From what we know, Danter was the leading agent in this venture and hence it was quite correct for the New Oxford Shakespeare editor, Francis X. Connor, to use the siglum 1DANTER in this case.

 

Danter printed four sheets and Allde printed six, so it’s a bit of a stretch to say that Danter was “the leading agent”. 

 

Readers may be interested to know that Freebury-Jones has responded to Egan’s misrepresentations, on his own website: https://darrenfj.files.wordpress.com/2017/11/A-RESPONSE-TO-GABRIEL-EGAN.pdf

 

[2]-----------------------------------------------------------------

From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 3, 2019 at 9:01:47 AM EST

Subject:    Re: NOS

 

Dear SHAKSPERians

 

I sympathize with Larry Weiss’s unhappiness at the cost of the online version of the New Oxford Shakespeare. Equally, I sympathize with anyone who finds books expensive to buy. I’m not aware of the terms of the subscription changing, but I can’t swear that they haven’t because I don’t know what he was told when he ordered.

 

I do hope that readers are aware that these matters are not the fault of the editors who write the books, since they are decided by the marketing and sales divisions of the publishing house. I have passed along Weiss’s remarks, though, since of course the publisher wants readers like him to be happy.

 

The case of ‘Richard II’ for which Weiss finds confusing the New Oxford Shakespeare’s sigla used in textual discussions is an inherently confusing one because the deposition episode first appeared in print in the fourth edition and the version of this episode in the 1623 Folio comes from a different source again.

 

But I’m just not seeing the confusion that Weiss reports when he claims that in the New Oxford Shakespeare “Q2 is 4WHITE”. Instead, I’m seeing on page 360 of Anna Pruitt’s textual introduction this table:

 

 1SIMMES (1597), printed by Simmes for Wise = 'Q1'

 2WISE (1598), printed by Simmes for Wise = 'Q2'

 3WISE (1598), printed by Simmes for Wise = 'Q3'

 4WHITE (1608), printed by White for Law = 'Q4'

 5LAW (1615), printed by White for Law = 'Q5'

 

That seems to follow the principle I previously mentioned: the ordinal numbers from the old sigla Q1, Q2, and so on, are retained in the New Oxford Shakespeare sigla and the name used is the printer for the first edition and the publisher for reprints.

 

So, why for the 1608 edition does the New Oxford Shakespeare use William White’s name to make the siglum 4WHITE instead of using Matthew Law’s name? The answer is that in respect of the deposition episode this 1608 edition has fresh authority. For the deposition episode this 1608 edition is the first edition, so we go back to identifying the printer (White) rather than the publisher (Law), as we do for all first editions.

 

Weiss writes that he has “not found how F2-F4 are cited”.  If he turns to the master list of Works Cited on pages xiii-xviii, he will find that it begins:

 

 JAGGARD Comedies, Histories, & Tragedies (F1, London, 1623)

 ALLOT Comedies, Histories, and Tragedies (F2, London, 1632)

 CHETWINDE Comedies, Histories, and Tragedies (F3, London, 1663)

 HERRINGMAN Comedies, Histories, & Tragedies (F4, London, 1685)

 

As Weiss observes, the sigla used in the critical editions of the works are not used in the Authorship Companion to the New Oxford Shakespeare. The Authorship Companion contains essays by scholars who did not edit works in the edition:

 

MacDonald P. Jackson, Doug Duhaime, Jack Elliott, Brett Greatley-Hirsch, John Burrows, Hugh Craig, David L. Gants, John V. Nance, Farah Karim-Cooper, Roger Holdsworth, Marina Tarlinskaja, and Giuliano Pascucci. The citational needs of these essays about attribution are unlike those of the textual apparatuses to the original-spelling critical editions.

 

Essays on authorship attribution do not make the kind of repeated comparison of the textual minutiae of one early edition with those of another that the textual apparatuses of the critical editions require. Whereas in the textual apparatuses to the works there are multiple sigla on every page, the essays in the Authorship Companion use about one siglum every 10,000 words (= 37 times in all across 386,000 words). For these reasons, it was decided not to apply the sigla of the main edition to the Authorship Companion.  Because the ordinal numbers are the same for the traditional sigla and those used in the textual apparatuses, no one should have any trouble following which edition is being referred to.

 

I believe it is quite correct to refer to Ivor B. John’s 1912 Arden1 edition of ‘Richard II’ as JOHN while referring to the edition of that play in a multi-editor collection by the name of the particular editor who did the editing rather than by the general editor, so that John Jowett’s ‘Richard II’ in the 1986-87 Complete Works is correctly JOWETT. I’m not seeing how anyone would find this unusual or confusing.

 

Regards

Gabriel Egan

 

 

 

Book Announcement

The Shakespeare Conference: SHK 30.005  Thursday, 3 January 2019

 

From:        Julia Griffin <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 2, 2019 at 2:02:28 PM EST

Subject:    Re: SHAKSPER: Announcements

 

Thanks to Brett Gamboa for alerting us to the publication of his book, which sounds most interesting.

 

Happy New Year!

 

Julia Griffin

 

 

 

The Shakespeare Canon and the NOS

The Shakespeare Conference: SHK 30.004  Monday, 2 January 2019

 

[1] From:        Larry Weiss <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         December 31, 2018 at 4:56:25 PM EST

     Subj:         Re: SHAKSPER: NOS

 

[2] From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

     Date:         January 1, 2019 at 9:54:57 AM EST

     Subj:         Re: SHAKSPER: NOS

 

 

[1]-----------------------------------------------------------------

From:        Larry Weiss <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         December 31, 2018 at 4:56:25 PM EST

Subject:    Re: SHAKSPER: NOS

 

I concur with Pervez Rizvi’s objection to NOS’s renaming of the 16th and 17th Century editions from Q2, F1, O1, and the like, to the names of its printers (or publishers, for reprints following the initial edition), preceded by an Arabic numeral to denote the printing’s “order in the sequence of quarto editions” (NOS Crit Ref Ed. p. xciv). Thus, Richard II Q1 is 1 S I M M E S (printed by Valentine Simmes for Andrew Wise), followed by two reprints the following year, 2 W I S E and 3 W I S E (both printed by Simmes); Q2 is 4 W H I T E (printed b y Wm. White for Matthew Law and dated 1608); Law’s “reprint” of 1615 is 5 L A W, printed by Thos. Purfoot (NOS Crit Ref Ed p. 369); the 1634 quarto is 6 N O R T O N (printer, no publisher identified in NOS); and, of course, F1 is identified by the printer J A G G A R D. (The extra leading in the names is also annoying.) I have not found how F2-F4 are cited, if they are. Eighteenth Century and subsequent editions are referenced by the names of their editors, which is the sensible traditional practice; but, contrary to the usual convention, a work contained in a collection is cited by the name of the editor of that particular play or poem, not the editor(s) of the collection. So, for example, the Arden1 Richard II is J O H N, while that play in the Oxford Shakespeare (a/k/a W E L L S - T A Y L O R) is J O W E T T (ibid.). This mélange of printers, publishers and editors is so confusing that the NOS editors actually revert to the old-fashioned Q & F nomenclature when they discuss the works in the Authorship Companion.

The only justification given by the NOS editors for this bizarre and unhelpful choice is that it “foregrounds the human agency involved in the physical manufacture of the book and its place in book history” (NOS Crit Ref Ed. p. xciv). I can’t see why the book’s place in history is clearer under this system than it is when it is identified as the first, second, etc., quarto or folio. And why is it important to “foreground the physical manufacture of the book,” and how does naming the printer or publisher accomplish this? Whatever the value of such foregrounding, does it justify requiring scholars who have spent their working lives using one form of easily grasped nomenclature to employ another, which has no obvious nexus to the literary issues which concern them?

 

I also thank Rizvi for pointing out the imprecise citation of Q1 Romeo & Juliet as 1 D A N T E R, who set only the first four sheets, rather than D A N T E R - A L L D E. However, I pick this particular nit only as prologue to a more serious gripe:

 

This imprecision in citation, like all such minor missteps, can easily be corrected in the next edition of NOS, whenever that will be (if ever). In the meantime, it can be corrected in the online edition. I recall that Gabriel Egan informed us that OUP has already begun to correct misspellings, typos, etc., online. I have proposed that the online edition contain a portal for readers to note mistakes they find, and Gabriel told us that he will take this up with the responsible persons. Of course, this implicates the problem of mutable online texts, which Gabriel and I have mooted here recently. Now for my big issue:

 

My complete four volume set of NOS was ordered prior to the publication date (and received months later) under the impression that it would include a free subscription to the online edition. I recall that I was informed that this was a lifetime subscription as a promotion for ordering the set before the pub date. Be that as it may, what I received was only a one-year subscription. When I went on line to renew it, the website refused to accept a renewal from an individual. It appears that only libraries and other institutions are eligible to subscribe to the site (see http://www.oxfordscholarlyeditions.com/page/20/how-to-subscribe). Not only that, but the fee for an institutional subscription is >$500 per annum, well beyond the means of the typical academic or gentleman scholar. What has changed, and why? This is distressing.

 

[2]-----------------------------------------------------------------

From:        Gabriel Egan <This email address is being protected from spambots. You need JavaScript enabled to view it.>

Date:         January 1, 2019 at 9:54:57 AM EST

Subject:    Re: SHAKSPER: NOS

 

Dear SHAKSPERians

 

Pervez Rizvi deplores the New Oxford Shakespeare’s choice of sigla for identifying the early editions of Shakespeare. He believes that it “destroys continuity with all previously published scholarship”. I would say that it was carefully chosen to maintain that continuity, by retaining the ordinal numbers (“1”, “2”, “3” and so on) that indicate the chronological sequence of publication while providing extra useful information.

The continuity means that in the case of ‘King Lear’ for example:

 

Q1 = 1OKES

Q2 = 2PAVIER

Q3 = 3BELL

 

Notice that the ordinal numbers are the same, but instead of providing the textually insignificant fact of format the siglum in each case acknowledges the human agency at work.

 

In the case of first editions, the name recorded in the siglum is the printer, the agent most directly responsible for the details that are of most significance for the discussions in which the sigla appear. In the case of reprints, the name recorded in the siglum is that of the publisher, since any readings that we actually adopt from a reprint are editorial emendations, and more likely to come from the publisher than the printer.

 

These principles govern all the sigla for all the early editions and have the advantage of creating greater continuity by recording the earliest editions using the same principles that govern the accepted sigla for the editions of the 18th-century and beyond, which get named for their editors as in POPE, DYCE, WILSON and so on.

 

Rizvi believes that the New Oxford Shakespeare editors, lacking the “competence” to apply their own nomenclature, “forgot” that although John Danter printed sheets A to D of the 1597 ‘Romeo and Juliet’ a second printer, Edward Allde, printed sheets E to K. He infers this lapse of memory about Allde’s contribution from the New Oxford Shakespeare’s use of the siglum 1DANTER for this edition, and decides that “They must have remembered when their book had already been typeset” because in the list of Editions Cited the 1597 edition is recorded as “printed by John Danter [and Edward Allde]” and “That parenthetical mention of Allde was apparently added because that was all they had the space to add (the previous page is almost full)”.

 

Rizvi’s account of a textual mishap in the printing of the New Oxford Shakespeare is pure fantasy. There was no last-minute discovery of forgotten shared printing, no panicked insertion of a parenthetical reference because “the previous page was almost full”. (That really isn’t how modern computerized typesetting affects what writers are allowed to do.)

 

Rizvi has been misled by Brian Gibbons’s out-of-date account of how “Danter’s premises were raided while he was printing the play, and his printing press seized and destroyed, making it necessary for the rest of the quarto to be printed by someone else”.  For the latest scholarship on the shared printing of the 1597 ‘Romeo and Juliet’, Rizvi should turn to the work of Chiaki Hanabusa, and in particular his essays:

 

Chiaki Hanabusa “Edward Allde’s Types in Sheets E-K of ‘Romeo and Juliet’ Q1 (1597)” Papers of the Bibliographical Society of America 91 (1997): 423-428

 

Chiaki Hanabusa “A Neglected Misdate and ‘Romeo and Juliet’ Q1 (1597)” Notes and Queries 244 (1999): 229-230

 

As Hanabusa shows, the 1597 edition of ‘Romeo and Juliet’ was completed before the raid on Danter’s shop and the shared printing was performed concurrently (before the raid), not consecutively.  From what we know, Danter was the leading agent in this venture and hence it was quite correct for the New Oxford Shakespeare editor, Francis X. Connor, to use the siglum 1DANTER in this case.

 

Turning to the appendix of “Datasets” in the New Oxford Shakespeare Authorship Companion, Rizvi complains that “It contains the datasets for only five chapters, but this is not mentioned”. I hope readers will understand that some of the huge datasets referred to in certain chapters could not be reproduced in print, and that indeed it was forward-thinking of Oxford University Press to accept the editors’ request to print the 27,000 words of data that we were able to include.

 

I’m unsure what Rizvi thinks was “not mentioned” about the coverage of this appendix, as I took care to label each one in relation to the essay it supported, as in:

 

Dataset _3.1 (in support of Hugh Craig, ‘Shakespeare and Three Sets of Additions’).

 

. . .

 

Dataset _4.1 (in support of Rory Loughnane, ‘Thomas Middleton in ‘All’s Well that Ends Well’? Part One’.  Word Sequences ‘All’s Well that Ends Well’ (TLN __2350-91; 4.3.203-38)

 

and so on.

 

Lastly, Rizvi claims that in my review of Darren Freebury-Jones and Marcus Dahl’s critique of microattribution I misrepresent him “at several points” and do the same to Freebury-Jones and Dahl. If he’d care to identify those points, I’d be happy to defend the fairness of my representation or, if he is able to show that I am mistaken, to correct my review.

 

Regards

Gabriel Egan

 

 

 

Subscribe to Our Feeds

Search

Make a Gift to SHAKSPER

Consider making a gift to support SHAKSPER.