Cambridge Catalogue  
  • Help
Home > Catalogue > The Struggle for Shakespeare's Text
The Struggle for Shakespeare's Text


  • 2 tables
  • Page extent: 332 pages
  • Size: 228 x 152 mm
  • Weight: 0.7 kg

Library of Congress

  • Dewey number: 822.3/3
  • Dewey version: 22
  • LC Classification: PR3071 .E38 2010
  • LC Subject headings:
    • Shakespeare, William,--1564-1616--Criticism, Textual
    • Shakespeare, William,--1564-1616--Bibliography
    • Shakespeare, William,--1564-1616--Criticism and interpretation--History--20th century
    • Drama--Editing--History
    • Transmission of texts

Library of Congress Record


 (ISBN-13: 9780521889179)

The Struggle for Shakespeare's Text
Cambridge University Press
9780521889179 - The Struggle for Shakespeare's Text - Twentieth-Century Editorial Theory and Practice - By Gabriel Egan


At the height of the trial in the cinematic court room drama The Verdict, a nurse acting as witness for the plaintiff offers as evidence a photocopy of a hospital admission form showing that the victim of the alleged medical malpractice was known to have eaten just one hour earlier and so should not have been anaesthetized (Lumet 1982). Yet she was anaesthetized, which made her vomit into her face mask, causing brain damage from lack of oxygen. The original admission form shown to the court recorded that the victim ate nine hours earlier (and so could be anaesthetized), but the nurse claimed that she photocopied the form before the anaesthetist (realizing his error) forced her to change the numeral 1 to a 9. On an established legal preference for original documents over photocopies, the jury is instructed to forget it ever heard about the nurse and her photocopy. Happily, the jury ignores this instruction and awards damages against the hospital.

The principle that one should ordinarily prefer an original of something over its copy is central to much of our thinking about textual authenticity, although of course there are circumstances under which it should be set aside, as when one suspects that the original was altered after the copy was taken. If the original was altered, one has to ask why and make a judgement based on one's best attempt at an answer. Originals should normally be preferred to copies because copying introduces errors, some random and some predictable. We may leave aside for the moment the new technologies that allow digital copying with perfect bit-for-bit fidelity, since these at the very least blur our convenient distinction between original and copy and perhaps even undermine our notions of what constitutes property.

The means by which early modern books were reprinted made errors of transmission inevitable. For many early modern books, the second edition was a reprint of the first and the third a reprint of the second, so that errors accumulated rather as they do in the children's game that Britons call Chinese Whispers and Americans call Telephone. When the first collected edition of the plays of Shakespeare, the First Folio (F or F1) of 1623, was reprinted in 1632 (F2), 1663–4 (F3) and 1685 (F4), each edition was based on its immediate predecessor, and error was piled upon error. These reprints’ publishers and printers attempted to restore sense where they could, as indeed players in a circle of Chinese Whispers do: almost unconsciously players turn the whispered sounds into words that cohere to make at least grammatical sense. But just as in the children's game, without access to the original words these attempts at improvement are overwhelmed by the corruption. The fun arises when the resulting words are grammatically plausible but wildly and comically inaccurate.

That such degeneration-by-repetition is also true of the later Folios of Shakespeare was observed by Samuel Johnson in the middle of the eighteenth century and he decried his fellow editors’ complacency in basing their editions on later Folios rather than returning to the First, the ultimate source (Shakespeare 1765, l). And yet, describing this seemingly sensible complaint from Johnson, one of today's leading theorists of textual transmission sees a darker motive at work:

Because the twentieth century's dominant textual theory raises up the ideal of recovering in an edition the full authorial presence that is now believed to lie just behind some of the earliest printed texts, the eighteenth-century preference for an edition that has benefited from cumulative editorial attentions (each removing us further from the earliest printings) has been slighted by our century's textual theorists (e.g., Wells, Taylor, et al. 55).

(Werstine 1995, 257).
Werstine implies that it is not awareness of the Chinese Whispers problem that motivates modern editorial preference for the First Folio over its reprints, but rather the delusion that the best early editions bring one fully into the presence of the author. Werstine rightly points out that Johnson did not scrupulously abide by his own counsel of perfection (he used reprints like everyone else) but in revealing this failing Werstine seems to abandon the central principle that Johnson was sketching. Werstine accuses Stanley Wells and Gary Taylor of following Johnson's precept for the wrong reason: not because it minimizes error but because it helps us commune with the dead.

The purpose of this book

The story of how modern textual theorists have come to hold such divergent views about the same raw materials and processes is one of the central narratives of this book. It aims to trace the debates about Shakespeare's texts as they have developed in the past century or so since the emergence of what is known as the New Bibliography. It presents a history of a set of ideas, but not impartially, for it argues that authors are the most dominant agents in the constellation of forces (personal, cultural, political and institutional) that come together in the publication of books. This does not mean that the author is sovereign, autonomous, or splendidly isolated, those being notions of authorship that modern literary theory tells us were invented by the Romantics. (Of course the theory might be wrong and the Romantics rather more subtle about authorship than they are usually given credit for; see Pechter 2001.) When it worked properly, publication in Shakespeare's time, just as publication now, invoked a hierarchy of agencies with the author at the top, supported by the labour of copyists and printshop workers. Readers unfamiliar with recent debates about the nature of authorship may be surprised to learn that this assertion is contentious and that making it opens one to a charge of conservatism, even elitism.

This book aims to help push the pendulum back from a currently fashionable dispersal of agency and insist upon authors as the main determinants of what we read. The ‘struggle’ of the book's title has two senses. The first alludes to the Herculean tasks of scholarship undertaken by bibliographers to extract knowledge from the surviving early editions of Shakespeare, as when Charlton Hinman compared each of the 900 pages in one exemplar of the 1623 Folio with the same page in each of fifty-four other exemplars, looking for the small differences that arose during the printing. By revealing the details of the printing process, Hinman hoped to offer editors better means for determining how it misrepresented what Shakespeare wrote, so they could undo the harm.

A second sense of ‘struggle’ in this book's title alludes to the arguments between scholars about how far we can hope to undo the harms of reproduction in order to recover what Shakespeare wrote. One branch of modern textual theory, identified in this book as New Textualism, accuses another, older branch, the New Bibliography, of over-optimism about seeing beyond the early editions to the manuscripts from which they were made. The hope that we might get an editorially recovered glimpse of those manuscripts (what Werstine means by ‘the full authorial presence’) is, according to many recent studies, delusional. According to Randall McLeod, the argument between scholars of editorial theory about how to recover what Shakespeare wrote is itself a constructive act, for it makes the very object that it would pursue. McLeod expressed this as ‘The struggle for tne text is the text’ (McLeod 1991, 279). (The n in tne was an intentional error made as part of a larger, witty argument about the ineluctability of error.) This book argues that McLeod is mistaken and that editors may reasonably pursue the objectively existing (now lost) readings of Shakespeare's manuscripts: they are not simply inventing readings from their imaginations and their struggle is worthwhile not for itself but for the recovered texts. McLeod generously agreed to the appropriation of his phrase in this book’s title, knowing that its argument would oppose his.

A currently popular view is that the early editions were so collectively (rather than individually) constructed and so imperfectly printed that the connection with Shakespeare's authorial intentions is all but lost. If this is the case, we must treat the early editions as social phenomena rather than the products of a single consciousness. In terms of the children's game of Chinese Whispers, this is akin to observing that the sentence emerging at the end of the circle is the collective product of all the whisperers and that, once the game has broken up, asking each participant what she heard will produce as many answers as there were players. To continue with the analogy, a textual optimist would be someone who, undiscouraged by the collective and corruptive process of transmission, attempted to work out the order in which the whisperers sat and so differentiate the more corrupted sentences from the less. Such an optimist would give most credence to the evidence of the first whisperer without necessarily falling into an idealist delusion of perfection; she would be expressing a relative preference for better over worse reproductions.

The subtitle of this book refers to the theory and practice of editing Shakespeare, but the book contains considerably more of the former than of the latter. This is because there simply is more theory than theoretically derived practice to describe, and because a comprehensive history of the facts of Shakespeare publication in the twentieth century already exists (Murphy 2003, 208–60). What remains to be described are the theoretical ideas embodied in the most progressive editions. Although certain editions are discussed in passing as the theories develop, a full account of the relationship of theory to practice is relegated to Appendix 3, with cross-references indicating where in the main text the associated theoretical ideas appear. The reader will find that the editions impinge more noticeably upon the main theoretical narrative towards the end of the story. This happens because for most of the century the theory was so far ahead of the practice as to be virtually out of sight. There was no edition of all of Shakespeare overtly executed according to New Bibliographical principles until John Dover Wilson's New Shakespeare series for Cambridge University Press was completed in 1966, and this edition was far from the New Bibliographical mainstream. Earlier complete works editions were in part shaped by New Bibliography, but none explained its editorial principles to the reader. Particular volumes in the mid-century Arden Shakespeare series showed the influence of New Bibliography, but only Wilson made a sustained effort to re-examine the entire textual situation for the whole canon from the new perspectives. However, theory and practice started to become contemporaneous in the 1980s, when there appeared several new editions formed along highly divergent lines.

The term bibliography derives from the self-reflexive practice of writing about books, although it is most commonly used to mean simply a list of books. The larger, but effectively the more specialist, sense of bibliography discussed in this book has two main varieties. Enumerative bibliography is concerned with establishing lists of books, such as all the works of one writer (perhaps published under various names), or of one centre of writing (say, the mediaeval abbey at Barking), or written about one subject. Analytical bibliography, on the other hand, is concerned with studying and describing books and their linguistic content, and divides into descriptive or physical bibliography (concerned with the book as a made object, including such things as its binding, its paper and the way sheets are folded), historical bibliography (concerned with the contexts for book publishing, such as the operations of various institutions that support it), and textual bibliography, also known as textual criticism, concerned with establishing the correct words of a writer by removing the errors of transmission. The bibliography with which this book is concerned is analytical bibliography in all its forms: descriptive/physical, historical and textual. Clearly, textual bibliography – establishing the words of Shakespeare – is the main concern, but as will become apparent the boundaries between the fields are permeable. Much of the early twentieth-century excitement about recovering Shakespeare's writings arose because the New Bibliographers championed multidisciplinary and interdisciplinary skills that crossed or erased these boundaries in the effort to remove errors of transmission.

Editing Shakespeare up to the end of the nineteenth century

To help make sense of the developments in editorial theory and practice in the twentieth century, the following sketch of the preceding history is offered. In the seventeenth century Shakespeare was not edited in the sense that we mean today. As we shall see, certain editions of Shakespeare (most especially the 1623 Folio) were prepared with considerable care to combine manuscripts and existing print editions, but the textual principles that characterize modern editing had not been developed. (For an argument dissenting from this view, based on the objection that many editor-like interventions were made in seventeenth-century reprintings, see Massai 2007.) As well as the four Folio collections (F1 to F4, each of the last three based on its predecessor), individual plays were printed in the smaller quarto format, typically one-per-volume, and for a given play the successive quartos (whether or not reprinting a predecessor) are abbreviated to Q1, Q2, Q3 and so on. Appendix 2 lists the editions of Shakespeare up to 1623 and who made them. The first collected works of Shakespeare that was edited in anything like the modern fashion was Nicholas Rowe's six-volume edition of 1709. Margreta de Grazia's account of the developments in editorial theory and practice in the eighteenth century, and especially of Edmond Malone's groundbreaking edition of 1790, is highly polemical and brilliantly argued (de Grazia 1991), while the developments in the nineteenth century are handled by Murphy rather more drily and without contentious philosophical assumptions (Murphy 2003, 188–207).

The intellectual development of eighteenth- and nineteenth-century editions can be characterized as an increasing regard for historical context and a willingness to undertake systematic comparison of the early editions to ascertain their relative authority. For the starkest contrast in these matters we may take an early and a late example: Alexander Pope's edition of 1723–5 and the Cambridge–Macmillan edition of 1863–6. In preparation for his editorial work, Pope published a newspaper note asking anyone who had editions of The Tempest, Macbeth, Julius Caesar, Timon of Athens, King John and All is True printed before 1620 to bring them to his publisher's office. As Murphy observed, ‘Tonson and Pope might have waited until doomsday for the requested texts to be delivered to them, since all of these plays had ... appeared in print for the first time in the 1623 folio’ (Murphy 2003, 64). Convinced that large portions of the early editions were not written by Shakespeare, Pope either cut them entirely or relegated them to the bottom of the page. According to Murphy, the greatest contribution made by Pope's edition was that in reaction to it other editors were determined to tackle the problems more systematically and to seek objective knowledge about print transmission before relying on subjective judgements about dramatic quality (Murphy 2003, 8).

One hundred and forty years later, the Cambridge–Macmillan edition was the first produced by university-employed scholars using an openly expressed bibliographical methodology arrived at after examining afresh the entire textual situation of Shakespeare (Murphy 2003, 202–6). Its editors, W. G. Clark, John Glover and W. Aldis Wright, compared each early printing with the others (a process called collating) in order to establish textual priority (which editions were reprints of which) and used this knowledge to help decide what to put in their edition where the early editions differed. Thus although their edition of Hamlet was mainly based on Q2 of 1604–5, the one they thought had the highest authority in general, they used the Folio text for the line ‘O, that this too too solid flesh would melt’ (1.2.129). In their collation notes at the foot of the page the Cambridge–Macmillan editors wrote ‘129. solid] Ff. sallied (Q1) Qq. sullied Anon. conj’ (Shakespeare 1866, 16), meaning that in line 129 their reading solid came from the Folios, that the quartos all read sallied (although Q1 differs significantly elsewhere on the same line), and that the reading sullied has been conjectured by persons unknown. This kind of attention to detail was new in the editing of Shakespeare, and the Cambridge–Macmillan editors were explicit about their application of processes that were established and refined for the editing of classical texts in Latin and Greek (Murphy 2003, 203).

The techniques used by the Cambridge–Macmillan editors were first formalized by the German philologist Karl Lachmann (1793–1851) for his edition of the Greek New Testament. Lachmann refined the genealogical process known as recension, in which the comparison of the surviving documents (all textual witnesses to the lost original, the author's manuscript) leads to a pictorial stemma that shows the family-tree relationships between them. The making of stemmata remains common in Shakespearian textual criticism even though it was developed not for printings that followed shortly after composition (as with Shakespeare) but for manuscripts made long after composition. Shakespearian stemmata are complicated in certain cases by the printers’ copy being an existing book that was annotated by comparison with an authoritative manuscript before being reprinted, which annotation injected new authority into the genetic line of an otherwise derivative reprint; several of the debates with which we are concerned here arise from this complication. The process of recension allows the editor to determine which of the surviving witnesses is the most authoritative and should be the basis for a modern edition, for which R. B. McKerrow coined the convenient term copy-text (Nashe 1904, xi). Thereafter comes emendation, the correcting of errors in this witness.

The Lachmannian approach stressed recension over emendation and encouraged editors to try to make sense of their copy-text rather than depart from it, and if departing from it was unavoidable then the next-best witness in the family tree should be consulted for its reading. This was essentially the process followed by the Cambridge–Macmillan edition, as they explained:

The basis of all texts of Shakespeare must be that of the earliest Edition of the collected plays, the Folio of 1623 ... This we have mainly adopted, unless there exists an earlier edition in quarto, as is the case in more than one half of the thirty-six plays. When the first Folio is corrupt, we have allowed some authority to the emendations of F2 above subsequent conjecture, and secondarily to F3 and F4; but a reference to our notes will show that the authority even of F2 in correcting is very small. Where we have Quartos of authority, their variations from F1 have been generally accepted, except where they are manifest errors, and where the text of the entire passage seems to be of an inferior recension to that of the Folio.

(Shakespeare 1863, xi)
The Cambridge–Macmillan edition was widely received as the culmination of all possible efforts to recover Shakespeare's true words, and it spawned a single-volume edition, the Globe Shakespeare, that sold nearly a quarter of a million copies and became the standard edition for the purposes of referencing for almost 100 years (Murphy 2003, 175–7). A sense of just how successfully the Cambridge–Macmillan editors conveyed the impression that there was nothing left to be done can be had from Horace Howard Furness's comment in his edition of Love's Labour's Lost:

Ever since the appearance, forty years ago, of The Cambridge Edition of SHAKESPEARE, followed by its offspring, The Globe Edition, this whole question of Texts, with their varying degrees of excellence, which had endlessly vexed the Shakespearian world, has gradually subsided, until now it is fairly lulled to a sleep as grateful as it is deep.

(Shakespeare 1904b, vi–vii)

Scope and plan of this book

This book is concerned only with Shakespeare's plays and leaves aside his poetry, this being principally his early narrative poems Venus and Adonis and The Rape of Lucrece and his Sonnets. The founders of New Bibliography also worked on other early modern dramatists – R. B. McKerrow edited Thomas Nashe, W. W. Greg edited Christopher Marlowe, Fredson Bowers edited Francis Beaumont and John Fletcher – and its principles were later applied to eighteenth- and nineteenth-century novels and poetry. However, in order to tell a coherent story of how the editing of Shakespeare has been theorized and practised, these additional contexts can only be mentioned in passing; New Bibliography has a larger history than can be told in this book's account of its origins and development. The entire subject of editorial theory and practice is now commonly placed within the even broader context of l’histoire du livre (the history of the book), which emerged as a distinct academic discipline in the middle of the twentieth century and was given its initial shape by the French historians Lucien Febvre and his protégé Henri-Jean Martin. Except where it impinges directly on Shakespeare – as it does tangentially when bibliographers discuss whether they should privilege certain authors or treat all books alike (pp. 84–6 below) – these broader contexts could not be incorporated into this book's narrative without doubling its length.

Confining our attention just to Shakespeare, his poetry is excluded because it was, for good reason, subject to an almost entirely distinct set of editorial principles in the twentieth century. The major advances in Shakespearian editorial theory emerged from knowledge of the practices of the early modern theatre, and in particular the ways that scripts for performance would be copied, divided, licensed, reworked and printed. These processes simply did not apply to the poems, which were written not for public performance but private consumption and although they probably circulated extensively in manuscript copies nonetheless went into print more or less directly from authorial papers. As we shall see towards the end of this book (pp. 215–22), the assumption that plays were written for public performance rather than private reading has been challenged, but the distinction from poems still holds since the most that can be said is that plays were intended for both kinds of consumption while poems were without doubt essentially a private pleasure. Regarding the plays, the claim (first made by the New Bibliographers at the end of the Great War) that they too were printed directly from Shakespeare's papers was controversial and requires extensive consideration.

The main concern of the narrative offered here is the development of a series of arguments about how best to present the plays of Shakespeare to modern readers. To make sense of the arguments requires knowledge of how early modern books were made, and readers without this (or wanting a refresher) will find that Appendix 1 covers the essential technical details. The story begins with a group of scholars who decided in the 1890s that the Cambridge–Macmillan editors had not achieved the best texts possible, and who invented an entirely new set of methodologies for making better ones. This book will consider the debates from the inside, as it were: how they seemed to the people who were making the arguments at the time. With hindsight it is possible to contextualize such debates by considering what else was happening in society, and a certain kind of historiography would read arguments about Shakespeare editions as symptomatic of other, wider conflicts. In such readings, Shakespearians may not even be aware that they are really arguing about human sexuality, or class, or the effects of technological empowerment (Masten 1997b; Loewenstein 1998; DiPietro 2006). Such studies are valuable, but this book pays its subjects the compliment of taking them at their word and it deals with their overt differences of opinion without trying to discern their unconscious motives. This is how most of us wish our own arguments to be taken: literally, not figuratively, nor interpreted psychologically. The book's self-denying ordinance cuts both ways, and there is no attempt to explain in political terms the reaction against New Bibliography in the 1980s, even though some of its critics were effectively pursuing a well-established left-leaning literary criticism by other means. Rather than seeking to explain the textual debates by reference to the debaters’ politics, the political underpinnings enter the narrative only when they are explicitly part of the arguments being made, as when various kinds of materialism must be distinguished.

Certain people feature rather less prominently in this narrative than they might have, as a result of the economies of selection. There is a case to be made for a feminist revaluation of the work of Alice Walker, and especially her book Textual Problems of the First Folio (1953) that is not much represented here. The materials for a revaluation exist in the archive of Walker's papers at Royal Holloway, University of London, but in truth she did not have much impact on the actual developments of the New Bibliography and after. Certainly, she had no more effect than John Dover Wilson who likewise is essentially tangential to this narrative except in his collaborations with A. W. Pollard and in his New Shakespeare series for Cambridge University Press. This book will for the most part take as read the facts of Shakespeare's co-authorship with other dramatists of certain plays and will not chart the development of the dawning realizations about this in the second half of the twentieth century, after initial progress was retarded by E. K. Chambers's ill-judged attack upon investigation of the subject (Chambers 1924–5). The facts of the matter are well summarized by Brian Vickers (2002), although their impact on editorial practice is as yet limited, as will be discussed in this book's conclusion.

In order to capture the debates as they developed, the structure of the present narrative is essentially a chronological survey of publications about Shakespearian bibliography with minor unchronological departures as necessary. In a few cases, the significance of a particular work was not registered when it first appeared, only to be discovered years later and built upon, and these works appear in the narrative at their delayed moment of impact. To assist the reader there are forward and backward cross-references in the narrative, so she may remind herself of where a previously discussed subject or argument first made its appearance, or skip forward to the point at which it came to fruition or destruction. The aim is not to account for every study of the textual situation of every play of Shakespeare: readers wanting such a comprehensive survey will find one in the Oxford Complete Works edition's Textual Companion (Wells et al. 1987), which may be supplemented (for works appearing since then) by the ‘Shakespeare Editions and Textual Studies’ sections of the annual books Shakespeare Survey and The Year's Work in English Studies. Rather, this book aims to trace closely the developments in the lines of thinking, with special attention to new breakthroughs, consolidations of positions, corroborations, disruptions and reversals. As a resting point in a long story, an Intermezzo is provided that breaks the chronological sequence to tell the full story of one strand of the New Bibliography, the theory that memorial reconstruction of their scripts by actors was the origin of certain early editions. Although it impinges upon it at key moments, and was used to buttress its claims, memorial reconstruction is not integral to the New Bibliography and its eighty-year rise and fall is best traced without disruption to the main narrative.

© Cambridge University Press
printer iconPrinter friendly version AddThis