The Arden Research Handbook of Shakespeare and Textual Studies 9781350080638, 9781350080652, 9781350080645

The Arden Research Handbook of Shakespeare and Textual Studies is a wide-ranging, authoritative guide to research on Sha

222 101 5MB

English Pages [409] Year 2021

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Cover
Contents
List of Illustrations
List of Tables
Notes on Contributors
Series Preface
List of Abbreviations
Introduction Lukas Erne
PART ONE: RESEARCH METHODS AND PROBLEMS
1.1 Shakespeare and ‘textual studies’: Evidence, scale, periodization and access Claire M. L. Bourne
PART TWO: CURRENT RESEARCH AND ISSUES
2.1 The Shakespeare manuscripts Cathy L. Shrank and Paul Werstine
2.2 The early printed texts of Shakespeare John Jowett
2.3 Shakespeare’s early modern books: Printing, paratext and text Emma Smith
2.4 Shakespeare in the early modern book trade Marta Straznicky
2.5 Shakespeare’s early readers and users: Annotating, commonplacing, collecting Laura Estill
2.6 The Shakespeare canon from the sixteenth to the twenty-first century Peter Kirwan
2.7 Shakespeare’s editors from the eighteenth to the twenty-first century Andrew Murphy
2.8 The modern editing of Shakespeare: The text Margaret Jane Kidnie
2.9 The modern editing of Shakespeare: The apparatus Suzanne Gossett
PART THREE: NEW DIRECTIONS
3.1 Shakespeare and authorship attribution methodologies Hugh Craig
3.2 Shakespeare and digital editions Sonia Massai
PART FOUR: MATERIAL FOR FURTHER RESEARCH
4.1 Chronology Alan B. Farmer
4.2 A–Z of key terms and concepts Eric Rasmussen and Ian H. De Jong
4.3 Annotated bibliography Jean-Christophe Mayer
4.4 Resources Emma Depledge
Index
Recommend Papers

The Arden Research Handbook of Shakespeare and Textual Studies
 9781350080638, 9781350080652, 9781350080645

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

THE ARDEN SHAKESPEARE HANDBOOKS The Arden Research Handbook of Contemporary Shakespeare Criticism Edited by Evelyn Gajowski ISBN 978-1-3500-9322-5 The Arden Research Handbook of Shakespeare and Social Justice Edited by David Ruiter ISBN 978-1-3501-4036-3 FORTHCOMING TITLES The Arden Handbook of Shakespeare and Early Modern Drama Edited by Michelle Dowd and Tom Rutter ISBN 978-1-3501-6185-6 The Arden Research Handbook of Shakespeare and Adaptation Studies Edited by Diana Henderson and Stephen O’Neill ISBN 978-1-3501-1030-4 The Arden Research Handbook of Shakespeare and Contemporary Performance Edited by Peter Kirwan and Kathryn Prince ISBN 978-1-3500-8067-6

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Edited by Lukas Erne

THE ARDEN SHAKESPEARE Bloomsbury Publishing Plc 50 Bedford Square, London, WC1B 3DP, UK 1385 Broadway, New York, NY 10018, USA 29 Earlsfort Terrace, Dublin 2, Ireland BLOOMSBURY, THE ARDEN SHAKESPEARE and the Arden Shakespeare logo are trademarks of Bloomsbury Publishing Plc First published in Great Britain 2021 Copyright © Lukas Erne and contributors, 2021 Lukas Erne and contributors have asserted their right under the Copyright, Designs and Patents Act, 1988, to be identified as the author of this work. Series design by Charlotte Daniels Cover image © Fyria / Shutterstock All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage or retrieval system, without prior permission in writing from the publishers. Bloomsbury Publishing Plc does not have any control over, or responsibility for, any third-party websites referred to or in this book. All internet addresses given in this book were correct at the time of going to press. The author and publisher regret any inconvenience caused if addresses have changed or sites have ceased to exist, but can accept no responsibility for any such changes. A catalogue record for this book is available from the British Library. A catalog record for this book is available from the Library of Congress. ISBN: HB: 978-1-3500-8063-8 ePDF: 978-1-3500-8065-2 eBook: 978-1-3500-8064-5 Series: The Arden Shakespeare Handbooks Typeset by Integra Software Services Pvt. Ltd. To find out more about our authors and books visit www.bloomsbury.com and sign up for our newsletters.

CONTENTS

L ist of I llustrations  L ist of T ables  N otes on C ontributors  S eries P reface  L ist of A bbreviations  Introduction  Lukas Erne

vii ix x xiv xvi 1

PART ONE: RESEARCH METHODS AND PROBLEMS 1.1 Shakespeare and ‘textual studies’: Evidence, scale, periodization and access  Claire M. L. Bourne

21

PART TWO: CURRENT RESEARCH AND ISSUES 2.1 The Shakespeare manuscripts  Cathy L. Shrank and Paul Werstine 2.2 The early printed texts of Shakespeare  John Jowett 2.3 Shakespeare’s early modern books: Printing, paratext and text  Emma Smith 2.4 Shakespeare in the early modern book trade  Marta Straznicky 2.5 Shakespeare’s early readers and users: Annotating, commonplacing, collecting  Laura Estill 2.6 The Shakespeare canon from the sixteenth to the twenty-first century  Peter Kirwan 2.7 Shakespeare’s editors from the eighteenth to the twenty-first century  Andrew Murphy 2.8 The modern editing of Shakespeare: The text  Margaret Jane Kidnie

53 71 94 111

128

150

168 188

vi

2.9 The modern editing of Shakespeare: The apparatus  Suzanne Gossett

CONTENTS

206

PART THREE: NEW DIRECTIONS 3.1 Shakespeare and authorship attribution methodologies  Hugh Craig 3.2 Shakespeare and digital editions  Sonia Massai

225 244

PART FOUR: MATERIAL FOR FURTHER RESEARCH 4.1 Chronology  Alan B. Farmer 4.2 A–Z of key terms and concepts  Eric Rasmussen and Ian H. De Jong 4.3 Annotated bibliography  Jean-Christophe Mayer 4.4 Resources  Emma Depledge

354

I ndex 

367

265 286 320

ILLUSTRATIONS

I.1 Title page of The London Prodigal (Q, 1605), attributed to ‘William Shakespeare’. Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 2.1.1 Hand D in The Booke of Sir Thomas Moore, BL Harley MS 7368, fol. 9r. Courtesy of the British Library 2.5.1 Annotations in Meisei University Library’s First Folio (1623), Measure for Measure 5.1, sig. G6r, detail. Reproduced with permission of Meisei University Library 2.5.2 Selections from Hamlet, University of Chicago Library, MS 824, f. 113r. Courtesy of the Special Collections Research Center, University of Chicago Library 2.5.3 Frances Wolfreston’s ownership mark on William Shakespeare, Venus and Adonis (1593), Bodleian Libraries, University of Oxford, Arch. G. e.31(2), title page. Courtesy of the Bodleian Libraries, University of Oxford 2.5.4 Thomas Bentley’s ownership mark and quotation from Shakespeare’s Pericles in George Chapman, An Humorous Day’s Mirth (Q, 1599), sig. H2r. Folger Shakespeare Library, call number: STC 4987 copy 1. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License 2.8.1 Hamlet (Q2, 1604–5), sig. D2r. Folger Shakespeare Library, call number: STC 22276. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License 2.8.2 The Taming of the Shrew, from the First Folio (1623), sig. T3v. Folger Shakespeare Library, call number: STC 22273 Fo.1 no.19. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License 3.1.1 Percentage counts of in as a preposition in 28 plays by Shakespeare and 16 plays by Fletcher 3.1.2 Percentage counts of hath in 28 plays by Shakespeare and 16 plays by Fletcher 3.1.3 Percentages of the word the in larger Shakespeare character parts, versus size of part

13 55

131

133

134

136

198

199 231 232 233

viii

ILLUSTRATIONS

3.1.4 Percentages of the word the in larger Shakespeare and Fletcher character parts, versus size of part 3.1.5 Success rates for the classification of Shakespeare plays and non-Shakespeare plays using linear discriminant analysis and ten different word-variable sets 4.1.1 Title page of King Edward the Third (Q1, 1596). Folger Shakespeare Library, call number: STC 7501. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License 4.1.2 Title page of Love’s Labour’s Lost (Q1, 1598). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.3 Title page of Romeo and Juliet (Q2, 1599). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.4 Title page of Henry the Fourth, Part Two (Q1, 1600). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.5 Title page of Much Ado About Nothing (Q1, 1600). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.6 Title page of Troilus and Cressida (Q1, 1609). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.7 Title page of The Troublesome Reign of John, King of England (Q3, 1622), attributed to ‘W. Shakespeare’. Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.8 Title page of Othello (Q1, 1622). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.9 Title page of the First Folio (1623). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.10 Title page of The Two Noble Kinsmen (Q1, 1634). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.11 Title page of Poems (O1, 1640). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva 4.1.12 Title page of the Third Folio, Second Issue (1664). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva

235

239

267

268

269

270

270

273

273

275

276

277

278

278

TABLES

I.1 Shakespeare in print, 1593–1623 3.1.1 Rare six-word sequences shared between ten plays and five authorial sets 3.1.2 Numbers of significant differences in the frequencies of 100 very common words in comparisons among five authors 3.1.3 Numbers of significant differences in the frequencies of 100 very common words in comparisons among five random sets of plays 3.1.4 Training and test sets

6 227 229

230 238

NOTES ON CONTRIBUTORS

Claire M. L. Bourne is Assistant Professor of English at the Pennsylvania State University, USA, where she teaches Shakespeare, early modern drama, book history and textual editing. She is the author of Typographies of Performance in Early Modern England (2020). She is currently editing 1 Henry the Sixth for the Arden Shakespeare, fourth series, and is the treasurer of the Marlowe Society of America. Hugh Craig is Emeritus Professor of English at the University of Newcastle, Australia. He has worked on attribution problems for many years, most recently on the authorship of the classic Chinese novel The Dream of the Red Chamber. Some of the more controversial authorship proposals in Shakespeare, Computers, and the Mystery of Authorship (2007), co-edited with Professor Arthur F. Kinney, are championed in The New Oxford Shakespeare (2016–). His current projects include a digital mapping platform, the Time-Layered Cultural Map of Australia and a stylometric study of the text of the Shakespeare First Folio, in collaboration with Professor Gabriel Egan. Ian H. De Jong is Instructor of English at the Davidson Academy of Nevada, USA. He co-edits (with Andrew Gurr) The Tempest for the New Variorum Shakespeare (forthcoming). He is chief associate editor of the Royal Shakespeare Company’s William Shakespeare: Complete Works (second edition, forthcoming). His published work examines the authorship of the 1602 Additions to The Spanish Tragedy (with Eric Rasmussen, ANQ, 2019), the early performance history of Hamlet (British Library, 2016) and the historical and cultural context for Shakespeare’s plays (with Eric Rasmussen, British Library, 2015–17). Emma Depledge is Assistant Professor of English Literature, 1500–1790, at the University of Neuchâtel, Switzerland. She is the author of Shakespeare’s Rise to Cultural Prominence: Politics, Print and Alteration, 1642–1700 (2018) and co-editor of Canonising Shakespeare: Stationers and the Book Trade, 1640–1740 (2017) and Making Milton: Print, Authorship, Afterlives (forthcoming). She is associate editor of the journal English Studies and reviews the year’s contribution to editions and textual studies for Shakespeare Survey. Lukas Erne is Professor of English at the University of Geneva, Switzerland. He is the author of Shakespeare as Literary Dramatist (second edition, 2013), Shakespeare and the Book Trade (2013) and Shakespeare’s Modern Collaborators (2008), and the editor of ‘Hamlet’ and ‘Romeo and Juliet’: ‘Der Bestrafte Brudermord’ and

NOTES ON CONTRIBUTORS

xi

‘Romio und Julieta’ in Translation: Early Modern German Shakespeare (with Kareen Seidler, 2020), Bel-vedére or the Garden of the Muses: An Early Modern Printed Commonplace Book (with Devani Singh, 2020) and The First Quarto of Romeo and Juliet (2007). He gave the Lyell Lectures at the University of Oxford in 2012. Laura Estill is a Canada Research Chair and Associate Professor of English at St Francis Xavier University, Canada. Her monograph (Dramatic Extracts in Seventeenth-Century English Manuscripts: Watching, Reading, Changing Plays, 2015) and co-edited collections (Early Modern Studies after the Digital Turn, 2016, and Early British Drama in Manuscript, 2019) explore reception history. Her work has appeared in journals including Shakespeare Quarterly, Papers of the Bibliographical Society of America, Digital Humanities Quarterly and The Seventeenth Century, as well as in collections such as Shakespeare’s Theatrical Documents, Shakespeare and Textual Studies and The Shakespeare User. Alan B. Farmer is an associate professor of English at the Ohio State University, USA. He is the co-editor, with Adam Zucker, of Localizing Caroline Drama: Politics and Economics of the Early Modern English Stage, 1625–1642 (2006), and the cocreator, with Zachary Lesser, of DEEP: Database of Early English Playbooks. He has published extensively on the publication of early modern drama, and his current book project is a study of Shakespeare, popularity and printed plays in the early modern English book trade. Suzanne Gossett is Professor Emerita of English at Loyola University Chicago, USA. She writes frequently on textual matters and is a General Textual Editor of the Norton Shakespeare, third edition, and General Editor of Arden Early Modern Drama. Her recent editions include All’s Well That Ends Well and Pericles for Arden 3, Eastward Ho! for the Cambridge Works of Ben Jonson, and plays by Middleton and Beaumont and Fletcher. She is a past president of the Shakespeare Association of America. Her current project is a book on Shakespeare and Textual Theory. John Jowett is Professor of Shakespeare Studies at the Shakespeare Institute, University of Birmingham, UK. He is a General Editor of the New Oxford Shakespeare, a member of the editorial boards of Arden Early Modern Drama and the Malone Society, an Associate General Editor of the Oxford Collected Works of Thomas Middleton, and an editor of the original Oxford Shakespeare Complete Works. He has edited Richard III and Timon of Athens for the Oxford Shakespeare series, and Sir Thomas More for the Arden Shakespeare. He has published widely on textual culture and textual theory. Margaret Jane Kidnie is Professor of English at the University of Western Ontario, Canada. She has published widely on Shakespearean performance, early modern manuscripts, textual studies and editorial practices. She is author of Shakespeare and the Problem of Adaptation (2009), and her co-edited essay collections are entitled Shakespeare and Textual Studies (2015) and Textual Performances: The Modern

xii

NOTES ON CONTRIBUTORS

Reproduction of Shakespeare’s Drama (2004). Her critical editions include the award-winning Philip Stubbes: The Anatomie of Abuses (2002), Thomas Heywood’s A Woman Killed with Kindness (2017) and the anonymous manuscript play, The Humorous Magistrate (2012). Peter Kirwan is Associate Professor of Early Modern Drama at the University of Nottingham, UK. His research interests span editorial theory, textual studies, theatre history and the contemporary performance of early modern drama. His books include Shakespeare in the Theatre: Cheek by Jowl (2019), Shakespeare and the Idea of Apocrypha (2015) and the co-edited collections Canonising Shakespeare (with Emma Depledge, 2017), Shakespeare and the Digital World (with Christie Carson, 2014) and The Arden Research Handbook of Shakespeare and Contemporary Performance (with Kathryn Prince, forthcoming). He is the editor of Shakespeare Bulletin and a general editor of the Revels Plays Companion Library. Sonia Massai is Professor of Shakespeare Studies at King’s College London, UK. She has published widely on Shakespeare in performance and the transmission of his works into print. Her publications include her books on Shakespeare’s Accents: Voicing Identity in Performance (2020) and Shakespeare and the Rise of the Editor (2007), collections of essays on Ivo van Hove (2018), Shakespeare and Textual Studies (2015) and World-Wide Shakespeares (2005), and critical editions of The Paratexts in English Printed Drama to 1642 (2014) and John Ford’s ’Tis Pity She’s a Whore for Arden Early Modern Drama (2011). Jean-Christophe Mayer is a Research Professor employed by the French National Centre for Scientific Research (CNRS) and a member of the Institute for Research on the Renaissance, the Neo-classical Age and the Enlightenment (IRCL) at Université Paul Valéry, Montpellier, France. He has worked mainly on the links between literature and history, Shakespeare and religion, Shakespeare and literary theory and the history of reading. His most recent book is a monograph entitled Shakespeare’s Early Readers: A Cultural History from 1590 to 1800 (2018). His new project is provisionally entitled ‘Handwriting and the English Imagination (16th to 17th Centuries)’. Andrew Murphy is a Fellow of Trinity College Dublin, Ireland, where he is 1867 Professor of English. He is the author of Shakespeare in Print: A History and Chronology of Shakespeare Publishing (2003; forthcoming in a revised and expanded edition). His other books include Ireland, Reading and Cultural Nationalism, 1790– 1930: Bringing the Nation to Book (2018) and Shakespeare for the People: Workingclass Readers 1800–1900 (2008). His edited volumes include A Concise Companion to Shakespeare and the Text (2007) and The Renaissance Text: Theory, Editing, Textuality (2000). He served as UK Associate Editor for The Cambridge Guide to the Worlds of Shakespeare (2016). Eric Rasmussen is Regents Teaching Professor and Foundation Professor of English at the University of Nevada, USA. His books include The RSC Collaborative Plays

NOTES ON CONTRIBUTORS

xiii

by Shakespeare and Others (2013), The Royal Shakespeare Company’s Complete Works of Shakespeare (2008) (both co-edited with Sir Jonathan Bate) and The Shakespeare First Folios: A Descriptive Catalogue (2012). He is a co-editor of The Norton Anthology of English Renaissance Drama (2002) as well as critical editions for The Arden Shakespeare, Oxford’s World’s Classics and the Revels Plays. He has served on the Board of Trustees of the Shakespeare Association of America and the Council of the Malone Society. Cathy Shrank is Professor of Tudor and Renaissance Literature at the University of Sheffield, UK. Publications include Writing the Nation in Reformation England (2004), an edition of Shakespeare’s Poems (with Raphael Lyne, 2018) and The Oxford Handbook of Tudor Literature (with Mike Pincombe, 2009). She is currently writing a monograph about early modern English dialogue, as well as editing Mammon for the Independent Works of William Tyndale and Pierce Penilesse for the Collected Works of Thomas Nashe. Emma Smith is Professor of Shakespeare Studies at Hertford College Oxford, UK. Her books include This Is Shakespeare (2019) and Shakespeare’s First Folio: Four Centuries of an Iconic Book (2016), and she is editor of the journal Shakespeare Survey. She is currently editing Summers Last Will and Testament for the Oxford Nashe edition and preparing a book with Laurie Maguire on theatre history. Marta Straznicky is Professor of English Literature at Queen’s University, Canada. Her books include Shakespeare’s Stationers: Studies in Cultural Bibliography (2013), The Book of the Play: Playwrights, Stationers, and Readers in Early Modern England (2006) and Privacy, Playreading, and Women’s Closet Drama (2004). She has also co-edited Women’s Household Drama: ‘Loves Victorie’, ‘A Pastorall’, and ‘The concealed Fansies’ (2019) and published essays on women’s plays and on theatrical performance at Wilton House in The Cambridge Companion to Early Modern Women’s Writing and The Intellectual Culture of the English Country House: 1500– 1700 (2009). Paul Werstine is Professor of English at King’s University College and in the Graduate Program at The University of Western Ontario, Canada. He has edited with the late Barbara A. Mowat for the Folger Shakespeare Library Shakespeare’s poems and plays in forty-two books (1992–2010), followed by revisions of sixteen books (2011–20). Since 1997 he has served as co-general editor of the New Variorum Edition of Shakespeare, to which he is contributing a volume on Romeo and Juliet. In 2012 he published Early Modern Playhouse Manuscripts and the Editing of Shakespeare.

SERIES PREFACE

The Arden Shakespeare Handbooks provide researchers and graduate students with both cutting-edge perspectives on perennial questions and authoritative overviews of the history of research. The series comprises single-volume reference works that map the parameters of a discipline or sub-discipline and present the current state of research. Each Handbook offers a systematic and structured range of specially commissioned chapters reflecting on the history, methodologies, current debates and future of a particular field of research. Additional resources, such as a chronology of important milestones that have shaped the field, a glossary of key terms, an annotated bibliography and a list of further resources are included. It is hoped that the series will provide both a thorough grounding in the range of research under each heading, and a practical guide that equips readers to conduct their own independent research. The topics selected for coverage in the series lie at the heart of the study of Shakespeare today, and at the time of writing include:

• contemporary Shakespeare criticism and theory • • • • •

Shakespeare and textual studies Shakespeare and contemporary performance Shakespeare and adaptation Shakespeare and social justice Shakespeare and early modern drama

While each volume in the series provides coverage of a distinct area of research, it will be immediately apparent that ‘distinct’ becomes a slippery concept: how does one define contemporary criticism as distinct from contemporary performance? Indeed, the very porousness of research areas becomes even more marked if, for instance, one explores research in Shakespeare and contemporary performance (in the volume edited by Peter Kirwan and Kathryn Prince) and Shakespeare and adaptation (in the volume edited by Diana Henderson and Stephen O’Neill). Questions of social justice permeate each area of research, for, as Evelyn Gajowski notes in the introduction to The Arden Research Handbook of Contemporary Shakespeare Criticism, ‘many of the essays … suggest the inseparability of critical practices, on the one hand, and social justice and political activism, on the other’. Even where we might be inclined to feel on safer ground about the ‘particular field’ of textual studies as distinct from other fields of Shakespeare studies, Lukas Erne disabuses that notion in his introduction to The Arden Research Handbook of Shakespeare and Textual Studies:

SERIES PREFACE

xv

Textual variants and multiplicity create their own proliferation of meanings, nor can textual studies and criticism ultimately be kept apart. For the question of what the text is decisively impacts the question of what the text means. While acknowledging the artificiality of boundaries and the inevitability of some degree of overlap, we have nevertheless encouraged editors to determine the contours of their Handbook with an eye on other titles in the same series. Just as each book provides a systematic grounding for readers, the series as a whole presents an invitation to readers to delve into each volume, to find those connections and points of intersection, and to explore the related fields that ultimately will enrich their own research.

ABBREVIATIONS

BL EEBO ESTC F ODNB OED MS MSS O r Q sig. sigs STC

TLN v

British Library Early English Books Online (http://eebo.chadwyck.com) English Short Title Catalogue Online (http://estc.bl.uk/) Folio Oxford Dictionary of National Biography (www.oxforddnb.com) Oxford English Dictionary (https://www.oed.com/) manuscript manuscripts octavo recto quarto signature signatures A. W. Pollard and G. R. Redgrave, comp. (1976–91), Short-Title Catalogue of Books Printed in England, Scotland, and Ireland, 1475–1640 and of English Printed Books Abroad, 2nd edn, rev. by W. A. Jackson, F. S. Ferguson and Katherine Panzer, 3 vols, London: Bibliographical Society Through Line Number(s) verso

Unless noted otherwise, quotations of Shakespeare’s plays and poems are taken from and citations are keyed to editions in the third Arden Shakespeare series.

Introduction LUKAS ERNE

This is an excellent moment to take stock of Shakespeare and textual studies.1 The Arden Shakespeare published the last volume of its third generation of Shakespeare editions in 2020, and the Oxford Shakespeare and the New Cambridge Shakespeare series were completed a few years ago. The third edition of The Norton Shakespeare (the first with a newly edited text) appeared in 2015, and four of the six volumes of The New Oxford Shakespeare were published in 2016 and 2017  – the remaining two, with The Complete Alternative Versions, are scheduled to be at press when the present Research Handbook appears. Arden and Cambridge University Press have started commissioning plays for new Shakespeare series, but the publication of the first volumes is still some way off. For the first time in many years no much-awaited Shakespeare complete works is in preparation, and no textually important Shakespeare series is in the process of being published. What better moment, then, to assess where Shakespeare textual studies have come from, what they have achieved and where they may be headed? The present Arden Research Handbook of Shakespeare and Textual Studies differs from previously published collections or companions in this scholarly field in several ways. It is more than a series of chapters that synthesize knowledge and add to it (although the chapters in Parts One to Three do that). It is a practical guide to research that provides a thorough grounding in Shakespearean textual studies. More than other companions or handbooks, it is importantly a reference work aimed at advanced undergraduate and graduate students, and scholars, including scholars with little prior knowledge of Shakespearean textual studies. Eric Rasmussen and Ian H. De Jong have provided an alphabetical list of concise definitions (4.2) of well over a hundred key terms and concepts, such as ‘accidentals’, ‘allowance’, ‘cancel’, ‘format’, ‘imposition’, ‘lemma’, ‘quire’, ‘unediting’ or ‘versioning’ – terms and concepts, that is, which we all come across sooner or later in Shakespeare scholarship but whose exact meaning, or meanings, may not be clear. This Research Handbook also includes a ‘Chronology’ (4.1), by Alan B. Farmer, that offers a timeline of important publications for the field ranging from Venus and Adonis in 1593 to The New Oxford Shakespeare edition in 2016/17. Another component of this Research Handbook is an ‘Annotated bibliography’ (4.3), by Jean-Christophe Mayer, consisting of thematically structured lists of important publications in the field, each of them with a short summary of its content, with sections devoted to ‘Shakespeare and Textual Studies: General’, ‘Shakespeare’s Early Books’, ‘Shakespeare’s Early

2

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Texts’, ‘Shakespeare’s Early Bibliographic and Textual Reception’, ‘Authorship and the Shakespeare Canon’, ‘Editing Shakespeare’, ‘Modern Shakespeare Editions’ and ‘Beyond Shakespeare: Bibliography, Book History and the Book Trade’. Although it is necessarily selective, the aim of the ‘Annotated bibliography’ is to equip readers with a high level of awareness of prior work in their areas of special interest, and to facilitate access to it. As its title indicates, the present volume is also, importantly, a research handbook, which not only gives users access to the fruits of other people’s research but also provides guidance on how to produce their own. The methodologies and resources with which book-historical and textual studies are undertaken in the twenty-first century are undergoing rapid change. New electronic databases and reference works offer exciting possibilities for research and have joined those print reference works that remain important. To take only one example, Early English Books Online (EEBO) and the fully-searchable, SGML/XML-encoded EEBO-TCP texts have had a transformative impact on research. For those who venture into the exciting field of Shakespeare textual studies a knowledge of these resources has become essential. It is no less important to be able to use and judge them critically, with a high degree of awareness. The opening part of the Arden Research Handbook of Shakespeare and Textual Studies is therefore devoted to a substantial presentation of ‘Research methods and problems’. In it, Claire M. L. Bourne engages critically with some of the most important resources we use in research today (including EEBO, the STC, the ESTC and the recent Shakespeare Census), while exploring key problems underlying current work in Shakespeare textual studies, in particular evidence, scale, periodization and access. While it is crucial for students of Shakespeare and textual studies to know about resources as they relate to research methods and problems, it is no less important to be aware of resources of a different kind: libraries, centres, associations and other institutions whose mission it is to enable the production and dissemination of research in areas that include Shakespeare and textual studies. The contribution on ‘Resources’ (4.4), by Emma Depledge, surveys the most important of such institutions, highlighting much information along the way – about rare items in libraries, courses that offer specialized training, the timing and format of conferences, eligibility requirements, funding opportunities and more – important information, that is, about where, when and how to become involved in the field. In short, then, the aim of this Handbook is not only to provide chapters by leading scholars about the most important issues and areas in the field but also to provide a reference tool and a research guide for the area of Shakespeare and textual studies.

*** What then are textual studies, and why do they matter? Criticism deals with the question of what and how the Shakespearean text means; textual studies with the prior question of what the text is. The former is essentially a matter of hermeneutics, the latter of ontology. The business of the Shakespeare critic is the interpretation of the text; that of the textual scholar is its establishment and constitution.

INTRODUCTION

3

Interpretation leads to a proliferation of meaning: as their reception across the centuries demonstrates, Shakespeare’s texts mean many things to many people. Textual studies, by contrast, involve a return to origins, a movement ad fontes, to the sources. As Erasmus put it in a humanist manifesto half a millennium ago: ‘First and foremost, however, recourse must be had to the sources themselves’ (2016: 673). The textual scholar is thus the humanist par excellence in today’s humanities. Whereas the primary textual object of Erasmus and other humanists was Holy Scripture, that of the Shakespearean is what some have considered its closest secular equivalent: the text of Shakespeare. But while the opposition I have drawn between criticism and textual studies is in some ways a valid one, it is in other ways far too neat. Textual variants and multiplicity create their own proliferation of meanings; nor can textual studies and criticism ultimately be kept apart. For the question of what the text is decisively impacts the question of what the text means. If Desdemona gave Othello ‘for [his] pains a world of sighs’ (1.3.160), as she did according to Othello’s account in the first quarto version of 1622 (sig. C3v), that makes her a different character than if she gave Othello a ‘world of kisses’ (sig. ss5v), as the Folio has it. The former potentially makes of her the passive female Brabantio would like her to be; the latter makes her the opposite. If Hamlet says ‘the rest is silence’ (5.2.342) and then silently dies, as he does in the second quarto version (1604/5: sig. O1v), this is significantly different from the end provided by the First Folio where his last sentence is followed by the far from silent sounds he makes in his death throes: ‘O, o, o, o’ (sig. qq1r). If Ferdinand, in The Tempest, ecstatically responds to Prospero’s masque by saying ‘Let me live here, ever. / So rare a wondered father and a wise / Makes this place Paradise’ (4.1.122–4), then that is a different response than if he rejoices at ‘a wondered father and a wife’, as some textual scholars have believed he does (Vaughan and Vaughan 2011: 136–8). Is it Prospero who makes the place Ferdinand’s paradise, or is it Prospero and Miranda? The difference matters. Modern editions of Shakespeare’s plays and poems embed countless decisions that have a considerable impact on how we understand his works. Editors take important decisions that result in conspicuous differences whose impact is fairly obvious. They also make many other changes to the text that are more likely to pass unnoticed by readers but nonetheless affect the texture of Shakespeare’s language in fundamental ways. Take the lineation of verse. Peter Holland’s Arden edition of Coriolanus, for instance, lineates more than three hundred lines in ways that depart from the seventeenth-century Folio texts and follows decisions first adopted by eighteenth-century or later editors, Nicholas Rowe, Alexander Pope, Lewis Theobald, Thomas Hanmer, Samuel Johnson, Edward Capell, George Steevens, Edward Malone, Alexander Dyce, David Bevington and Stanley Wells. Shakespeare’s much praised verse thus reaches us as an editorial collage whose modern origins are ignored by all but the initiated. Shakespeare’s editors are often the unacknowledged mediators of the word. It should now be clear why textual studies are important. If we take on faith the text of the nearest edition to hand, we will mistake editorial decisions for the author’s, and become its passive victims. Unless we are prepared to be naïve readers

4

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

of Shakespeare’s plays and poems, we have no other choice but to be keenly attuned to the insights of textual studies. Only if we embrace editions as editions and come to appreciate how a long tradition of textual scholarship has shaped them are we prepared to engage with Shakespeare’s texts in their full complexity.

*** If Shakespeare textual studies make us go ad fontes, what are the sources we go back to? For a significant part of his adult life, Shakespeare was a key participant – as actor, shareholder and playwright – in a theatrical entertainment industry in which hundreds if not thousands of textual documents were produced of versions and parts of his plays. He is likely to have produced a working draft of his plays of which he or a scribe would then produce a clean transcript. Even before the draft, he may well have written a plot-scenario and shared it with the members of his company. Various parts of the plays, such as prologues and epilogues, songs and letters, would have existed as separate textual documents. The dialogue was copied into the actors’ parts. A backstage-plot would have helped with the practicalities of performance, including the timing of entrances. Shakespeare or one of his colleagues would also have produced a summary of the play’s plot and have it printed for distribution to members of the audience. An even shorter summary along with the play’s title would have been produced to serve as printer’s copy for the playbills, which were posted around London as advertisements (Stern 2009). The early modern theatre was thus an intensely textual environment, and the making and staging of Shakespeare’s plays must have necessitated and encouraged the production of countless documents. But since these textual documents were produced to serve fleeting performances, not to enrich stable libraries, they have all perished (for the one possible exception, Hand D in the manuscript of Sir Thomas More, see below, pp. 54–7). ‘The Shakespeare manuscripts’, as Cathy L. Shrank and Paul Werstine point out at the beginning of their chapter (2.1), is thus a deliberate misnomer. There are theatrical manuscripts of non-Shakespearean plays that throw light on Shakespeare’s, and there are manuscripts of (parts of or adaptations of) Shakespeare’s plays and poems that ‘bear witness … to [Shakespeare’s] relevance: his continued re-summoning by seventeenth-century readers, who fitted him into new forms, for different purposes, and other occasions’ (Shrank and Werstine p. 67). But there are no literary manuscripts that scholars agree to be by Shakespeare. What textual scholars thus have to turn to instead is print. The book trade, like the theatre trade, was a successful and expanding mass medium in early modern London, and, unlike the theatre trade, it produced many textual documents of Shakespeare’s works that have survived. Shakespeare was no member of the book trade, and the only textual traces of his involvement in the publication of his writings are the dedicatory epistles to Henry Wriothesley, Earl of Southampton, prefacing his two narrative poems, Venus and Adonis (first published in 1593) and The Rape of Lucrece, or simply Lucrece, as it was originally called (first published in 1594). His textual absence from his printed playbooks need not imply that he was indifferent to their publication, and indeed there are good reasons to believe that he was not (Erne 2013a). What it does indicate is that the printed plays are the products of a

INTRODUCTION

5

different trade than the one for which Shakespeare worked, a trade that reconfigured Shakespeare’s texts to meet its own needs, with its own conventions. Textual studies of Shakespeare thus face a paradoxical situation: Shakespeare worked at the heart of an institution – the professional theatre – in which his plays existed in a wealth of now lost textual documents, and Shakespeare appears to have been all but completely absent from another institution – the book trade – through which the greatest part of his writings have survived. Shakespeare’s plays from the early modern theatre trade have thus perished – not only because performance is by its nature evanescent but also because its textual documents were not preserved. What has survived is Shakespeare’s plays (and poems) from the early modern book trade. The desire to recover Shakespeare’s plays ‘as they were acted’ (Wells 1986: xxxix) is understandable since it is the performed plays in which Shakespeare was most immediately involved. But the desire is also misguided, as it is precisely the plays ‘as they were performed’ that have been irretrievably lost. In a very real sense, then, the book trade provides the only Shakespeare to which editors and textual scholars can have access. In the thirty-year period from 1593 to 1623, spanning the greatest part of Shakespeare’s career and the first seven years after his death, seventy-nine editions were published (including the First Folio) of which all, or at least a significant part, of the content is thought to have been written by Shakespeare (see Table I.1). The thirty-six plays in the First Folio almost completed the publication of his dramatic works – only The Two Noble Kinsmen, written in co-authorship with John Fletcher, first appeared later, in a quarto edition of 1634. Of the thirty-six First Folio plays, half had been published separately first, usually in quarto (the default format for early modern playbooks), although Richard Duke of York (more commonly referred to by its Folio title 3 Henry VI) appeared in octavo in 1595. The poems received twenty-two editions. The narrative poems – hardly among Shakespeare’s most-read works today – were exceptionally popular in the early modern book trade, Venus and Adonis going through twelve and The Rape of Lucrece through six editions, whereas the Sonnets, by far his most-loved poetic work today, received only one. Even The Passionate Pilgrim, attributed to Shakespeare although only a few of its poems are his, received three editions. Unlike the playbooks, the poems appeared mostly in octavo, although the earliest editions of the narrative poems and the one edition of the Sonnets are in quarto. These early editions of Shakespeare’s plays and poems are studied by scholars from a variety of angles. One approach consists in placing them in the context of the early modern book trade. Today we take for granted the publication of Shakespeare’s works, but from the angle of the late-Elizabethan and Jacobean book trade, there was no reason why it should have been. Most plays written for the commercial theatre never reached print. Before the 1590s, very few of them did. At the turn of the century, the London book trade was still small and produced no more than a small fraction of the number of titles issued only half a century later. Any publication constituted a significant investment in human labour and material (especially paper), and came with the risk of significant loss if it sold poorly. It is to the stationers who took that risk that we owe the survival of Shakespeare. Who are they, and why did

6

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Table I.1  Shakespeare in print, 1593–16232 Year

Plays

1593

Poems

Number

VA Q1

1 (0/1)

1594

Tit Q1, 2H6 Q1

Luc Q1, VA Q2

4 (2/2)

1595

3H6 O1

VA O1 (3) (?)

2 (1/1)

1596

E3 Q1

VA O2 (4)

2 (1/1)

1597

R2 Q1, R3 Q1, RJ Q1, LLL Q0 (?)

1598

1H4 Q1, 1H4 Q2, LLL Q1, R2 Q2, R2 Q3, R3 Q2

Luc O1 (2)

RJ Q2, 1H4 Q3, E3 Q2

VA O3 (5), VA O4 (6) PP O1 (?), PP O2

7 (3/4)

Luc O2 (3), Luc O3 (4)

10 (8/2)

1599 1600

H5 Q1, 2H4 Q1, Ado Q1, MND Q1, MV Q1, 2H6 Q2, 3H6 Q1 (2), Tit Q2

4 (4/0) 7 (6/1)

1601

0 (0/0)

1602

MW Q1, R3 Q3, H5 Q2

1603

Ham Q1

1 (1/0)

1604

Ham Q2, 1H4 Q4

2 (2/0)

1605

R3 Q4

1 (1/0)

VA O5 (7)

1606

4 (3/1)

0 (0/0)

1607

VA O6 (8) (?), Luc O4 (5)

2 (0/2)

1608

KL Q1, R2 Q4, 1H4 Q5

VA O7 (9) (?)

4 (3/1)

1609

TC Q1, Per Q1, Per Q2, RJ Q3

Son Q1

5 (4/1)

VA O8 (10) (?)

1 (0/1)

1610 1611

Tit Q3, Ham Q3, Per Q3

1612

R3 Q5

1613

1H4 Q6

3 (3/0) PP O3

1 (1/0)

1614 1615

1 (1/1) 0 (0/0) 1 (1/0)

R2 Q5

1616

Luc O5 (6)

1 (0/1)

1617

VA O9 (11)

1 (0/1)

1618 1619

0 (0/0) 8 (8/0)

2H6 Q3, 3H6 Q2 (3), Per Q4, MV Q2, MW Q2, KL Q2, H5 Q3, MND Q2

1620

VA O10 (12)

1621

1 (0/1) 0 (0/0)

1622

Oth Q1, R3 Q6, 1H4 Q7

3 (3/0)

1623

First Folio, RJ Q4 (?)

2 (2/0)

INTRODUCTION

7

they choose to invest in Shakespeare’s plays and poems? What did they see in them, and how did their publication fit into their trade? What are the commercial and ideological contexts within which Shakespeare’s plays and poems were published, and what meanings did his works take on in those contexts? What transactions led to Shakespeare’s bibliographic presence, and what does it tell us about the status of Shakespeare in the book trade? As Marta Straznicky shows in the chapter on ‘Shakespeare in the early modern book trade’ (2.4), these questions can lead to important insights into what Shakespeare’s texts meant to his contemporaries. Another approach to the early editions of Shakespeare’s plays and poems focuses not on why and by whom but on how they were published. How, in particular, was Shakespeare’s literary text framed by extra-literary text (or ‘paratext’), an area of research to which Emma Smith’s chapter on ‘Shakespeare’s early modern books: Printing, paratext and text’ (2.3) pays close attention. As she shows, the paratext in Shakespeare’s early books ‘shap[es] Shakespeare’s works for an early modern readership’ and ‘is crucial to studies of authorial reputation, reader response and textual transmission’ (p. 107). How Shakespeare’s texts were published can also be investigated by studying the materiality of the books in which they first appeared. What is the material form originally assumed by Shakespeare’s printed texts, what decisions and what labour went into producing it, and what is the meaning of that form? The physical makeup of the early modern printed Shakespeare editions bears little resemblance to the modern material equivalents to which we are used, and it is important to study them not only for the texts they contain but also for the physical objects they are. The book is an expressive form, as D. F. McKenzie taught us (1999: 9–30), and, as Jerome McGann pointed out, ‘[e]very literary work that descends to us operates through the deployment of a double helix of perceptual codes: the linguistic codes, on the one hand, and the bibliographical codes on the other’ (1991: 77). The antitheatrical polemicist William Prynne was keenly aware of this when he protested that ‘Shackspeers Plaies are printed in the best Crowne paper, far better than most Bibles’ (1633: sig. **6v). Paper – including its watermark and chainlines (Depledge 2018) – is a powerful material signifier, but it is only one among many. There is format (folio, quarto, octavo); typeface (roman, italics, blackletter); font size (great primer, English, pica); and the disposition of space on the page. There is the binding, or its absence, or the barely visible holes in the gutter that bear witness to former stab-stitching (Pratt 2015). There is the company a copy holds with other titles it is bound up with, or was bound up with (Knight 2013), to which ‘ghosts’, faint images of the ink imprinted on a page that was previously bound alongside it, may testify (Lesser forthcoming). There is distinctive type (including damaged type) the recurrence of which may provide information about the printing process (Blayney 1982: 89–150). As Smith points out, it would be wrong to believe that the printing of Shakespeare’s plays and poems ‘simply reproduced [the] textual objects’; rather it ‘transmuted’ them (p. 95), turning them into physical objects from which much meaning can be derived. Finally, and most obviously, scholars study the early printed Shakespeare editions because they are interested in the dramatic and poetic texts printed in them. Those texts are in many ways as close as we can get to how Shakespeare’s plays and poems

8

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

came into being, and if we are interested in mediating them to modern readers, they are an inevitable starting point. What complicates matters is that texts in early modern editions of the same play may differ from one another. Those differences may be slight when an edition is a simple reprint of another, but even then, they matter, and whether a difference results from inadvertent corruption or deliberate change is not always clear. We may think that at least the plays not published prior to the First Folio harbour no such problems, but the Folio went through three more editions in the course of the seventeenth century (1632, 1663–4, 1685), and (usually anonymous) early correctors (we would now say editors) clearly tried to improve the texts while the hazards of printing also introduced occasional errors (Massai 2007: 180–205). The texts in the early modern editions of all of Shakespeare’s plays and poems thus pose serious problems. Problems are compounded when early editions of the same play differ significantly. The first quarto edition of Hamlet (1603) contains a short text, little more than half as long as the later quarto editions, starting with the second quarto published in 1604/5, which is in turn different from the text in the Folio, containing about 230 lines lacking in the Folio but missing approximately eighty lines present in it. The quarto editions of King Lear (1608, 1619) lack some hundred lines that are part of the Folio text but include about 300 lines that are absent from it. The first quarto edition of Othello (1622) lacks approximately 160 lines present in, but has a few lines absent from, the Folio. Similarly, the quarto editions of Richard III from the first (1597) to the sixth (1622) have about forty lines that are not in the Folio but lack some 230 lines present in it. The quarto edition of 2 Henry IV (1600) fails to include approximately 170 lines that are in the Folio but does include some forty lines that are not; in addition, it exists in two states, one with and one without the scene modern editions identify as Act 3 Scene 1. Of several plays, there are separately published editions that are significantly shorter than the same play’s text in the Folio: Henry V (first published in 1600), The Merry Wives of Windsor (1602), 2 Henry VI (published as The First Part of the Contention in 1594) and 3 Henry VI (published as Richard Duke of York in 1595). The first quarto edition of Romeo and Juliet (1597) is more than 700 lines shorter than the second quarto (1599) and subsequent editions. In all of these cases, the differences are not simply a matter of length: the variant versions differ in many particulars, substitute words and phrases, assign speeches to different characters, and so on. Other plays were published in variant versions that essentially agree in length but contain differences that are nonetheless striking. For instance, in the last act of A Midsummer Night’s Dream, the Folio assigns to Philostrate, Theseus’ Master of the Revels, the speeches that the quarto editions (1600, 1619) give to Egeus, Hermia’s father. The quarto (1609) and the Folio texts of Troilus and Cressida differ in hundreds of single words and short phrases but usually have the same length of speeches and even sentence structure. In the case of two plays, the most significant difference between the early printed versions concerns a single scene: the quarto editions of Titus Andronicus lack a scene (now usually referred to as the ‘Fly Scene’, 3.2 in modern editions) that is included in the Folio; and the first three quarto editions of Richard II (1597, 1598, 1598) lack the ‘deposition scene’ (part of

INTRODUCTION

9

4.1 in modern editions), of which the fourth and fifth quarto editions (1608, 1615) provide one version and the Folio another. The early printed texts of Shakespeare’s plays thus present many problems of various kinds. For much of the twentieth century, the New Bibliography tried to address these problems by telling ingenious narratives about the origins of the manuscripts from which the printed texts were set up, and by creating convenient boxes into which they could be fitted: ‘good quartos’ and ‘bad quartos’, plays set up from ‘foul papers’, ‘fair copy’ or ‘promptbook’. These categories, as John Jowett shows in his chapter on ‘The early printed texts of Shakespeare’ (2.2), are no longer satisfactory, and we need ‘more flexible alternatives’ (p. 87), leading to more nuanced accounts of the texts’ possible origins and of what might explain the differences between them. Modern scholars thus engage with Shakespeare’s early modern texts and books in a variety of ways, but so did early modern readers and users, and our response to Shakespeare is complicated by awareness of the forms that early modern engagement took (Mayer 2018). Extant early copies show that Shakespeare’s texts were read and annotated, that passages were highlighted by means of commonplace markers, manicules and underlining, and that proprietorship of copies was claimed with ownership marks. Records of private libraries show that Shakespeare books were collected, bound and catalogued. Extant manuscripts show that passages from his plays and poems were excerpted, copied and adapted, and reinserted into miscellanies and commonplace books, of which some in turn reached print. One of the poems in England’s Helicon is a version of a song from Love’s Labour’s Lost (Rollins 1935); England’s Parnassus includes ninety-five passages by Shakespeare, thirty from plays and sixty-five from the narrative poems (Crawford 1913); Bel-vedére or the Garden of the Muses contains 107 short, one- or two-line passages from Shakespeare’s plays and 133 from his poems (Erne and Singh 2020: lxi–lxii). All three publications were first printed in 1600, less than halfway through Shakespeare’s documented career, testifying to the fact that his textual reception, appropriation and recycling started early (Erne 2013b: 186–232). Laura Estill’s chapter, ‘Shakespeare’s early readers and users: Annotating, commonplacing, collecting’ (2.5), examines how Shakespeare’s early readers approached his plays and poems and the questions that their engagement can lead us to ask. As she shows, the many traces of readers’ engagement that the documentary record has left us can considerably enrich the reception history of Shakespeare.

*** The early material witnesses of Shakespeare’s plays and poems thus represent one large area of textual studies. Their modern reproduction constitutes another. Today’s editors have a great responsibility in textually mediating Shakespeare to modern readers. The view that Shakespeare’s modern editor is a harmless drudge, like Samuel Johnson’s lexicographer, will be held only by those who believe that Shakespeare’s texts already exist and that modern editors do no more than copy them. They do not. Shakespeare’s texts as they reach the modern reader are a complex transhistorical collaboration to which editors make a decisive contribution.

10

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Modern editions are thus importantly shaped by the editors who prepare them, but they also result from developments across the centuries which have led to the conventions that now govern scholarly Shakespeare editions. Starting with the first Shakespeare edition published by Jacob Tonson the Elder in 1709, edited by Shakespeare’s first acknowledged editor, Nicholas Rowe, editions have radically changed their look, format, textual and paratextual make-up, editorial principles and implied readership. Figures whose Shakespeare editions are now well known – such as Alexander Pope, Lewis Theobald, Samuel Johnson, Edward Capell and Edmond Malone in the eighteenth century and W. G. Clark and William Aldis Wright in the nineteenth – had an important impact on the development of Shakespeare in print; but so did others, like John Dicks, who, in the 1860s, published the immensely popular ‘Shilling Shakespeares’. As Andrew Murphy shows, in his chapter on ‘Shakespeare’s editors from the eighteenth to the twenty-first century’ (2.7), a thorough awareness of the history of Shakespeare in print is important if we want to understand where our contemporary editions have come from and may be going in the future. Editing Shakespeare today means situating oneself within and vis-à-vis a long tradition of editorial engagement. While serious editors have a duty to start with the early modern text(s) and to rethink every decision without blindly following their predecessors, they also need to be aware of the choices made by others before them. Many readings that are now taken for granted have their origin in an eighteenthcentury edition and have since been regularly adopted by other editors. In some ways, modern editors are dwarfs standing on the shoulders of giants: if they see far into the text of Shakespeare and are occasionally able to further improve it, that is because they can benefit from centuries of editorial insights. Yet the routine adoption of prior readings comes with the danger of inheriting unwarranted assumptions and of perpetuating the bias of past generations. If nineteenth-century editors usually made Desdemona give Othello ‘a world of sighs’ (the quarto reading) rather than ‘a world of kisses’ (the Folio reading), even though the same editors generally had a preference for the Folio over the quarto, then that shows that editorial decisions can embed gender stereotypes in Shakespeare’s editions and thereby contribute to their perpetuation (Potter 2003: 81–2). What applies to gender can equally apply to other forms of stereotype, notably about race, as exemplified by the editorial treatment of Sycorax, the ‘blue-eyed hag’ (1.2.269) in The Tempest (Marcus 1996: 5–17). The editorial tradition can thus be seen as a rich deposit of insights on which we may build, or as a source of corruptions from which we need to be saved. At the time of the Reformation, analogous differences in attitude towards the traditions of the Catholic Church were the source of religious division across Europe. In the modern editorial reproduction of Shakespeare, the consequences are less drastic, but the differences in attitude towards the value of tradition are no less real. The late David Bevington offered a sane compromise between extremes by explaining why there can be no end to editing: ‘No better explanation is needed for the frequently asked question “Why another edition?” than that editions must continually readdress the problems of textual interpretation in terms of contemporary values and language without losing sight of what past editors can richly provide’ (1988: 136). New and

INTRODUCTION

11

different editions are necessary, but their quality to an important extent depends on the insights of previous editors. What forms does the intervention by modern editors take? They establish the text, and doing so entails various operations. For instance, they diagnose errors and fix them by means of emendations, a process that may sound straightforward but is in fact fraught with difficulties (Wells 1984: 32–56). They modernize the spelling and the punctuation, which is sometimes fairly mechanical yet at other times requires difficult decisions that have a significant impact on what a passage means (Wells 1979; Bevington 2004). They insert act and scene breaks (most early editions are undivided) and decide in the process on the shape of Shakespeare’s scenes; the number of scenes in the last act of Macbeth, for instance, ranges from six in the New Penguin Shakespeare (Hunter 2005) to eleven in The New Oxford Shakespeare (Taylor et al. 2016). They decide whether passages are set as prose or as verse (Orgel 2002: 35–47), and, if verse, they decide on how to lineate it, decisions that have a major impact on our reception of the form of Shakespeare’s dramatic language (Werstine 1984). They add to the scarce stage directions in the early editions by providing some of their own with the aim of clarifying (or making the reader engage with) implied stage action, or possible stage action (Kidnie 2000: 2004). They regularize speech headings and thereby determine how characters are called and thus how they are to be thought of: ‘Shylock’ or ‘The Jew’? ‘Edmund’ or ‘The Bastard’? ‘Puck’ or ‘Robin’? ‘Bolingbroke’ or ‘King’? ‘Lady Macbeth’ (an invention of Nicholas Rowe in the early eighteenth century) or simply ‘Lady’? And so on. When substantially variant early editions of a play are extant, editors also need to determine what it is they wish to recover and how they wish to present it. Should editors try to recover Shakespeare’s (or his and his company’s) final intentions by eclectically choosing from readings in the variant early editions. Or should their aim be to present one or several versions of a play Shakespeare revised? Or should they edit one or several early texts whose authority is perceived to reside not in a supposedly revising Shakespeare but in an early printing? If modern editions adopt readings from different early editions, should editors use presentational options such as square and pointed brackets to clarify their origins in the text. Modern editors of some of Shakespeare’s most-loved plays, including King Lear and Hamlet, have given many different answers to these questions and produced very different editions as a result. As Margaret Jane Kidnie shows in her chapter on ‘The modern editing of Shakespeare: The text’ (2.8), textual scholars currently ‘have an opportunity to rethink, and potentially re-vision, models of editorial methodology’ (p. 196), and the directions in which they may go in the future are currently being negotiated. Editors not only establish the text, but they also frame it with an apparatus, as discussed in the chapter by Suzanne Gossett, ‘The modern editing of Shakespeare: The apparatus’ (2.9). Depending on the edition, the apparatus may contain an introduction, a glossary, a commentary, collation notes or textual notes, and appendices. Even the list of characters directly preceding the playtext may be considered part of the editorial apparatus: early editions usually give no such lists (there are none in the quartos and only seven in the First Folio), thus leaving editors

12

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

to arrange the characters in a certain order (hierarchically? by order of appearance? male characters before female characters, as in most eighteenth-century editions?), and provide or withhold information about them. Is Othello simply ‘the Moor of Venice’ (Wells and Taylor 1986: 820) or is he, quote-unquote, ‘“the Moor”, general of the Venetian forces’ (Neill 2006: 193)? Is Macbeth one of the ‘Generals of the King’s Army’ (Muir 1951: 2), the ‘Thane of Glamis’ (Clark and Mason 2015: 126) or – spoiler alert – the ‘Thane of Glamis, later Thane of Cawdor, then King of Scotland’ (Taylor et al. 2016: 2505). By means of the list of characters, editors influence the readers’ response to them immediately before the actual playtext begins. What is true for the list of characters applies to the apparatus more generally. It is shaped by editors so as to facilitate the play’s mediation to readers and in turn shapes their reading experience in manifold ways. Editors also decide where to draw the line between what belongs to Shakespeare’s works and what does not. What should be included in an edition of Shakespeare’s works? Series editors and general editors agree on the thirty-six First Folio plays, plus Pericles and The Two Noble Kinsmen, the Sonnets, the two narrative poems and ‘Let the Bird of Loudest Lay’ (variously called ‘The Phoenix and the Turtle’). But how about Edward III? The New Cambridge Shakespeare series (NCS; general editors, Brian Gibbons and A. R. Braunmuller), the third generation of the Arden Shakespeare (Arden 3; general editors, Richard Proudfoot, Ann Thompson, David Scott Kastan and H. R. Woudhuysen) and The New Oxford Shakespeare (NOS; Taylor et al. 2016) include the play, but the Oxford Shakespeare series (general editor, Stanley Wells), the RSC Complete Works (Bate and Rasmussen 2007), the Bevington Complete Works (seventh edition, 2014), and the print version of the third edition of the Norton Shakespeare (Norton 3; Greenblatt et al. 2016) do not. Sir Thomas More is variously included in part (NOS, RSC), in its entirety (Arden 3) or not at all (NCS, Wells, Norton 3, Bevington). The New Oxford Shakespeare includes Arden of Faversham (published as an anonymous play in 1592) and the additions, published in 1602, to Thomas Kyd’s Spanish Tragedy, whereas all the other editions do not. Arden 3 uniquely includes Double Falsehood by the eighteenth-century dramatist Lewis Theobald on the grounds that it may constitute an adaptation of the lost History of Cardenio, by Shakespeare and Fletcher (The New Oxford Shakespeare includes what are argued to be ‘Fragments of The History of Cardenio, by Fletcher and Shakespeare’ (Taylor et al. 2016: viii), mostly drawn from Double Falsehood). When it comes to the poems, most editions include A Lover’s Complaint, but the RSC Complete Works does not. The same edition also excludes The Passionate Pilgrim, as does the Bevington, unlike most other editions. The epilogue ‘To the Queen’, written for a court performance in February 1599, is included in the RSC Complete Works and The New Oxford Shakespeare, but is absent from the other works and from the major Shakespeare series. Editors, then, decide not only how to edit Shakespeare’s texts but also what should be included among them, and the differences among Shakespeare’s modern editors are considerable. Modern disagreements over what should be considered as Shakespeare’s and what should not have a long prehistory. Take the case of The London Prodigal

INTRODUCTION

13

FIGURE I.1  Title page of The London Prodigal (Q, 1605), attributed to ‘William Shakespeare’. Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

14

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

(see Figure I.1), a play that was published in 1605 with the unambiguous title page attribution ‘By VVilliam Shakespeare’. The title page adds that the play had been ‘plaide by the Kings Maiesties seruants’ – the name of Shakespeare’s company at the time – and that it was published by Nathaniel Butter, a respectable stationer who went on to publish King Lear three years later. For a contemporary book buyer, there would have been no reason not to consider the play as Shakespeare’s. The London Prodigal was not included in the First Folio, but it did appear in the second issue of the Third Folio of 1664 along with six other plays (of which only one, Pericles, is now thought to be partly by Shakespeare), expanding Shakespeare’s alleged dramatic canon from thirty-six to forty-three plays. The Shakespeare canon was thus in flux in Shakespeare’s own time and has remained so ever since, as we learn from Peter Kirwan’s chapter, ‘The Shakespeare canon from the sixteenth to the twenty-first century’ (2.6). The modern study of what Shakespeare wrote and what he didn’t has developed at great speed owing to the computational methods to which we now have access. When John Goodwin, in 1979, wrote in the Introduction to A Short Guide to Shakespeare’s Plays that Shakespeare ‘wrote thirty-seven plays’ (7), period, he was simply stating what seemed to many a straightforward truth at the time. In the meantime, the number of Shakespeare’s plays has gone both up and down. According to The New Oxford Shakespeare, the extant corpus contains only twenty-seven plays of which Shakespeare is the sole author, and yet the table of contents lists forty-five plays to which he contributed all or a part (Taylor et al. 2016: vii–viii). An important recent development in Shakespeare editing, then, is that authorship attribution has become one of its main themes. Significantly, the Oxford Complete Works of 1986 were accompanied by a Textual Companion (Wells et al. 1987), but The New Oxford Shakespeare has appeared along with an Authorship Companion (Taylor and Egan 2017). Research in authorship attribution typically happens ‘at the juncture between statistics and language’ (p. 225), with which many of us may not be intimately familiar, so Hugh Craig’s chapter (3.1) offers ‘a primer in Shakespeare authorship attribution’ the aim of which is ‘to establish some principles to guide the scholar embarking on authorship attribution work’ (pp. 225, 239). While the electronic revolution has led to new directions in authorship attribution, it is also transforming the editing of Shakespeare more generally. For instance, digital editions have the potential to represent the instability of the Shakespearean text in ways print cannot, and an electronic apparatus offers possibilities print does not provide, partly because it is not constrained by the pressure on space that applies to print. The New Oxford Shakespeare and the third edition of The Norton Shakespeare have been published not only in print but also digitally, and the trend will no doubt continue. Future editions will be digitally born, although it seems likely that spinoffs in print will continue to be sold. The chapter by Sonia Massai (3.2) evaluates recent developments in the digital editions of Shakespeare, and sketches ‘desirable directions of travel for the foreseeable future’ (p. 245). In the early modern period, the Gutenberg Revolution secured the survival of Shakespeare’s texts. Today, the Electronic Revolution is transforming our engagement

INTRODUCTION

15

with them. It is an exciting period for research in the area of Shakespeare and textual studies, and the aims of the following chapters are to convey that excitement and to enable further research.

NOTES 1. I am grateful to Oliver Morgan and Anthony Mortimer for their feedback to an earlier version of this Introduction. 2. The table uses the abbreviations for titles employed in the Arden 3 series: VA: Venus and Adonis; Luc: The Rape of Lucrece; Tit: Titus Andronicus; 2H6: 2 Henry VI; 3H6: 3 Henry VI; E3: Edward III; R2: Richard II; R3: Richard III; RJ: Romeo and Juliet; LLL: Love’s Labour’s Lost; 1H4: 1 Henry IV; PP: The Passionate Pilgrim; H5: Henry V; 2H4: 2 Henry IV; Ado: Much Ado About Nothing; MND: A Midsummer Night’s Dream; MV: The Merchant of Venice; MW: The Merry Wives of Windsor; Ham: Hamlet; KL: King Lear; TC: Troilus and Cressida; Per: Pericles; Son: Sonnets; Oth: Othello. When a text published in one format had previously appeared in another (as happened with Venus and Adonis, The Rape of Lucrece and Richard Duke of York (3 Henry VI), the overall number of the edition is indicated parenthetically. Question marks indicate dates that are conjectural because the date on the title page is either missing or known to be incorrect. For the lost first edition of Love’s Labour’s Lost (‘Q0’ in the table), see Freeman and Grinke (2002). The table does not record the publication of ‘Let the bird of loudest lay’ (also referred to as ‘The Phoenix and the Turtle’), a sixty-seven-line poem by Shakespeare that was included in Robert Chester, Love’s Martyr: or, Rosalin’s Complaint in 1601, a quarto of almost 200 pages.

REFERENCES Bate, Jonathan and Eric Rasmussen, eds (2007), The RSC Shakespeare: Complete Works, New York: The Modern Library. Bevington, David (1988), ‘Editing Shakespeare in Paperback’, Renaissance Drama, 19: 127–47. Bevington, David (2004), ‘Modern Spelling: The Hard Choices’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 143–57, Cambridge: Cambridge University Press. Bevington, David, ed. (2014), The Complete Works of Shakespeare, 7th edn, Boston, MA: Pearson. Blayney, Peter W. M. (1982), The Texts of ‘King Lear’ and Their Origins, Volume 1: Nicholas Okes and the First Quarto, Cambridge: Cambridge University Press. Clark, Sandra and Pamela Mason, eds (2015), Macbeth, The Arden Shakespeare, London: Bloomsbury. Crawford, Charles, ed. (1913), Englands Parnassus, Oxford: Clarendon Press. Depledge, Emma (2018), ‘False Dating: The Case of the “1676” Hamlet Quartos’, The Papers of the Bibliographical Society of America, 112 (2): 183–99.

16

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Erasmus, Desiderius (2016), De ratione studii (1512), in Brian McGregor (ed. and trans.), The Collected Works of Erasmus, vol. 24: Literary and Educational Writings, vol. 2, 661–91, Toronto: University of Toronto Press. Erne, Lukas (2013a), Shakespeare as Literary Dramatist, 2nd edn, Cambridge: Cambridge University Press. Erne, Lukas (2013b), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Erne, Lukas and Devani Singh, eds (2020), Bel-vedére or The Garden of the Muses: An Early Modern Printed Commonplace Book, Cambridge: Cambridge University Press. Freeman, Arthur and Paul Grinke (2002), ‘Four New Shakespeare Quartos?’, The Times Literary Supplement, 5 April: 17–18. Goodwin, John (1979), A Short Guide to Shakespeare’s Plays, London: Heinemann. Greenblatt, Stephen, gen. ed. (2015), The Norton Shakespeare, 3rd edn, New York: Norton. Holland, Peter, ed. (2013), Coriolanus, The Arden Shakespeare, London: Bloomsbury. Hunter, G. K., ed. (2005), Macbeth, Penguin Shakespeare, Harmondsworth: Penguin. Kidnie, Margaret Jane (2000), ‘Text, Performance, and the Editors: Staging Shakespeare’s Drama’, Shakespeare Quarterly, 51: 456–73. Kidnie, Margaret Jane (2004), ‘The Staging of Shakespeare’s Drama in Print Editions’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 158–77, Cambridge: Cambridge University Press. Knight, Jeffrey Todd (2013), Bound to Read: Compilations, Collections, and the Making of Renaissance Literature, Philadelphia: University of Pennsylvania Press. Lesser, Zachary (forthcoming), Ghosts, Holes, Rips and Scrapes: Shakespeare in 1619, Bibliography in the Longue Durée, Princeton: Princeton University Press. Marcus, Leah S. (1996), Unediting the Renaissance: Shakespeare, Marlowe, Milton, London: Routledge. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Mayer, Jean-Christophe (2018), Shakespeare’s Early Readers: A Cultural History from 1590 to 1800, Cambridge: Cambridge University Press. McGann, Jerome J. (1991), The Textual Condition, Princeton: Princeton University Press. McKenzie, D. F. (1999), Bibliography and the Sociology of Texts, Cambridge: Cambridge University Press. Muir, Kenneth, ed. (1951), Macbeth, The Arden Shakespeare, London: Methuen. Neill, Michael, ed. (2006), Othello, The Oxford Shakespeare, Oxford: Oxford University Press. Orgel, Stephen (2002), The Authentic Shakespeare and Other Problems of the Early Modern Stage, New York: Routledge. Potter, Lois (2003), ‘Editing Desdemona’, in Ann Thompson and Gordon McMullan (eds), In Arden: Editing Shakespeare, 81–94, London: Thomson Learning. Pratt, Aaron (2015), ‘Stab-stitching and the Status of Early English Playbooks as Literature’, The Library, 7th ser., 16 (3): 304–28.

INTRODUCTION

17

Prynne, William (1633), Histrio-mastix: The Players Scourge, London. Rollins, Hyder E., ed. (1935), England’s Helicon, 1600, 1614, Cambridge, MA: Harvard University Press. Shakespeare, William (1604/5), The Tragicall Historie of Hamlet, Prince of Denmarke, London. Shakespeare, William (1622), The Tragœdy or Othello, the Moore of Venice, London. Shakespeare, William (1623), Comedies, Histories, & Tragedies, London. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Taylor, Gary and Gabriel Egan, eds (2017), The New Oxford Shakespeare: Authorship Companion, Oxford: Oxford University Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2016), The New Oxford Shakespeare: The Complete Works: Modern Critical Edition, Oxford: Oxford University Press. Vaughan, Virginia Mason and Alden T. Vaughan, eds (2011), The Tempest, rev. edn, The Arden Shakespeare, London: Bloomsbury. Wells, Stanley (1979), ‘Modernizing Shakespeare’s Spelling’, in Stanley Wells, Modernizing Shakespeare’s Spelling, with Gary Taylor, Three Studies in the Text of ‘Henry V’, 1–36, Oxford: Clarendon Press. Wells, Stanley (1984), Re-Editing Shakespeare for the Modern Reader, Oxford: Clarendon Press. Wells, Stanley (1986), ‘General Introduction’, in Stanley Wells and Gary Taylor (gen. eds), William Shakespeare: The Complete Works, xv–xxxix, Oxford: Clarendon Press. Wells, Stanley and Gary Taylor, gen. eds (1986), William Shakespeare: The Complete Works, Oxford: Clarendon Press. Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery (1987), William Shakespeare: A Textual Companion, Oxford: Oxford University Press. Werstine, Paul (1984), ‘Line Division in Shakespeare’s Dramatic Verse: An Editorial Problem’, Analytical and Enumerative Bibliography, 8: 73–125.

18

PART ONE

Research methods and problems

20

CHAPTER 1.1

Shakespeare and ‘textual studies’: Evidence, scale, periodization and access CLAIRE M. L. BOURNE

Methodological developments in the collective of sub-fields of Shakespeare studies that fall under the auspices of ‘textual studies’ have typically come at moments of frustration. What counts as evidence? What does the evidence mean? How does it fit (or not) with established narratives? And how does it serve efforts to produce modern editions that make ‘Shakespeare’ accessible to new audiences in textual form? It is relatively – and perhaps surprisingly – rare that new texts relating to Shakespeare come to light, even though the last few years have seen a few ‘discoveries’ that might indeed shift the textual studies terrain.1 In the account of textual studies methodologies – past and present – that follows, change is dictated less by intermittent infusions of new evidence than it is by evolving attitudes towards a relatively stable set of known documents. The term ‘textual studies’ is fairly new as a catch-all for the ways that material texts teach us not only about their formations and re-formations but also about the cultural, political, social and economic priorities behind those histories of making, unmaking and remaking. However, many of this capacious field’s governing questions are not new: What counts as ‘Shakespeare’? How do we interpret the evidence we know we have? How do we know what’s missing? How do we negotiate these gaps? This chapter considers contexts for – and resources used to facilitate – recent approaches to these questions. The first part offers a very brief overview of how ‘textual studies’ has evolved over the last half century after the term first emerged as a way of defining scholarship that did not fit into the box of traditional bibliography. The second part studies the kinds of evidence that inform textual studies scholarship, as well as the shifting status of that evidence. This section underscores a variety of methods that have been used to study ‘Shakespeare’ – a mutable textual corpus defined not only by historical fact but, more importantly, by the tastes and judgments of people who have published, printed, performed, edited, read, collected, reformed, watched, taught and otherwise used for new ends the texts that Shakespeare wrote or had a hand in writing. The final section of the chapter draws attention to recent trends in textual studies that leverage different scales of analysis to generate new insights

22

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

about the Shakespearean text. In discussing some of the most important resources for material text-focused scholarship, this part also discusses how issues of access shape the study of the Shakespearean text in unexpected ways.

*** ‘Textual studies’ is (at best) difficult to define and (at worst) risks being too wideranging to boast coherence.2 It may be best understood as an evolving set of methods for encountering, describing, analysing and mediating material texts that encompasses manuscript studies; bibliography (the description, analysis and contextualization of printed texts as material objects); book history (broadly conceived to account for the historical production and reception of textual objects); scholarship in theatre studies that attends to the textual ecologies of performance; textual criticism; and scholarly editing. Its broad remit includes methods that query how textual objects were made, circulated, used and transmitted in their own time all the way down to ours and that inform the principles of textual editing – that is, how old texts should be repackaged for new readers. ‘Textual studies’ emerged in the second half of the twentieth century as a commodious way of describing scholarly approaches to studying material texts and their contexts that strained the boundaries of traditional bibliography. Indeed, the term began appearing as an adjunct to ‘bibliography’ just as Shakespeare studies was shifting its centre of gravity from New Critical to New Historicist methodologies. For instance, it sidles up next to ‘bibliography’ in a subject heading – ‘Critical Bibliography and Textual Studies’ – in James G. McManaway and Jeanne Addison Roberts’s A Selective Bibliography of Shakespeare: Editions, Textual Studies, Commentary (1975); and the two adjectives in the title of Craig S. Abbott and William Proctor Williams’s handbook An Introduction to Bibliographical and Textual Studies (1985) suggest complementary fields. By the late 1970s, the New Bibliography’s rigorously pragmatic approach to studying textual objects with the aim of producing editions that ‘fulfilled’ authorial intention was beginning to chafe against and redefine itself in light of more theoretically minded and historically disposed textual scholarship (McGann 1991: 22–3). A surging interest in broader historical contexts among text-oriented scholars tested and exceeded the scope of bibliographic methods, which W. W. Greg had famously described as ‘concerned with … pieces of paper or parchment covered with certain written or printed signs’, not the ‘meaning’ of those signs, let alone their place in history (1932: 121–2). (Meaning, Greg clarified, was quite literally none of the bibliographer’s business.) The shift from text to text-incontext is evident on a surface level in changes to the organizational schema of the Annotated World Bibliography of Shakespeare scholarship, published annually from 1960 to 1997 in Shakespeare Quarterly.3 When the bibliography changed format in 1976, it went from being organized by type of scholarly output (‘Editions and Translations’; ‘Articles’; etc.) to being organized by topic (Grieve and Habenicht 1975: 329). Scholarship on textual matters was shuffled under the unwieldy heading: ‘Paleography, Manuscripts, Printing History, Descriptive and Analytic

SHAKESPEARE AND ‘TEXTUAL STUDIES’

23

Bibliography’ (Meserole 1976: 395–6). Two years later, the header was revised to read: ‘Textual Studies, Etc.’ (Meserole and Smith 1979: 480). The ‘Etc.’ signalled the early swells of an as-yet undefinable wave of scholarship that expressed a deeper concern with historical contingency and historically situated social practices in the study of Shakespeare and/as the material text. Bolstered both by the New Historicism’s drive to locate plays in broader social, political and cultural contexts as well as by the editorial recognition in the Oxford Complete Works (1986) that different early versions of the same play could be equally legitimate, textual scholars brought into the purview of literary studies the idea that the material form of textual objects – themselves products of diffuse, diachronic, (non-) intentional human interactions with the materials of textual production – could affect and effect the meaning of the texts they transmitted. The most consolidated expression of this material turn in Shakespeare studies came in Margreta de Grazia and Peter Stallybrass’s essay ‘The Materiality of the Shakespearean Text’ (1993). There was a lot at stake, they argued, in looking at the printed text rather than through it for some irrevocably lost ‘original’ (257). The materiality of early editions could illuminate ‘the complex social practices that shaped’ them (283). The New Bibliographers had sought for the better part of the twentieth century to explain away textual inconsistencies, oddities and infelicities. And since at least the nineteenth century, the rare book market had privileged clean copies of books, with many dealers and collectors going so far as to eradicate all signs of a book’s post-publication history from its pages and packaging (Knight 2013). But by the end of the 1990s, there was a new scholarly effervescence around textual idiosyncrasy and peculiarity. David Scott Kastan – perhaps ironically – described this interest as ‘the New Boredom’, a mode of textual encounter that ‘takes delight in particularity’, obsessively attends to the processes by which ‘a text actually enters and exists in the world’, and accounts for the various ways that material form relates to meaning (1999: 18). These encounters were reactions to the New Historicism’s flattening out of textual distinctions in their treatment of all texts – literary or not, material or abstract – as harbingers of larger ‘discursive exchanges’ without considering the ‘enabling forms’ (both material and literary) in which their earliest readers would have encountered them. Much of the recent field-shifting work on textual matters relating to Shakespeare derives from obsessive attention to textual particularity and, as such, challenges critical biases that have structured the methods of textual scholarship for quite some time: the tendency to study editions over specific copies of those editions; whole and perfect exemplars over textual fragments; textual presences over absences; the pristine over the used; the individual author over collective agency; ‘substantive’ features of the text (words) over so-called ‘accidental’ ones (punctuation, design, etc.); moments of textual origin over subsequent moments of textual making and transmission; periodization boundaries dictated by political events over a more fluid treatment of the relationship of texts to time. What all this means, then, is that ‘textual studies’ no longer orbits around the origin point of the literary, or dramatic, work – or even around a shared imperative to peel away 400 years of post-authorial

24

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

accretions to reveal some lost, abstract and essential version of the text. It instead gravitates towards an assortment of post-publication phenomena: the textual practices involved in reading, using, repurposing, collecting, editing, performing and mediating ‘Shakespeare’ across time and place. Of course, this vision of ‘textual studies’ may be too manifold to be contained under a single heading. However, maintaining a loose cohesion of various interests and investments – textual studies, etc. – has its advantages. First, it recognizes that there are different ways, some of which are not incompatible, to have a stake in the Shakespearean text. The older model of studying theatrical and textual production to purge the text of certain impurities and indiscretions effectively cordoned off the text as the domain of straight, elite, masculine energies both in terms of its makers (playwrights, printers, publishers) and of those with enough institutional privilege to tell its story (Maguire 2005; Masten 2016: 39–66). Not only is a new generation of scholars telling new histories of making, but the turn towards ‘use’ as a heuristic has already brought an ever more diverse set of voices (and hands) from the past – especially those of women – into our study of the past (Wayne 2020). Furthermore, changing modes of accessing plays in media other than either printed codex or theatre increases the exposure of these materials to students and scholars, even as the learning curves associated with new forms of mediation – digital imaging practices, online interfaces and user devices – present new challenges. Indeed, mediation itself has become a rich topic of scholarship in Shakespeare studies (Galey 2014). Finally, the admirable, yet elusive, desire for an objective account of the Shakespearean text has given way to a plurality of plausible narratives – interpretations and reinterpretations of evidence. Instead of a this-is-how-it-was attitude, we can now operate on the register of this-is-how-it-may-plausibly-havebeen (Sherman 2002: 133). This reminds us that the stories we tell about objects that may, at first, seem fixed in time, place and circumstance are always to some degree shaped by institutional pressures and personal desires: the questions we ask and the methods we develop to answer those questions are entirely contingent on the intellectual frameworks and technological tools at our disposal. With this in mind, I turn now to four implicated ‘problems’ animating the most exciting and provocative new work in textual studies around Shakespeare: evidence; scale; periodization; and access. The first is, by far, the most fundamental and complex: what are the ‘texts’ in ‘textual studies’, and what are they evidence of? The other three issues follow from this account and introduce several big-picture concerns – and some resources required to negotiate those concerns – that structure research in the field today. What follows does not pretend to be a comprehensive account but should be viewed instead as a series of meditations about where we appear to be in relation to these so-called ‘problems’, how they relate to each other, and where we might go next, in the face of significant institutional and global challenges.

*** The prerequisite for engaging critically with textual objects that emerged out of the early modern theatre, the book trade and other related environments is a working

SHAKESPEARE AND ‘TEXTUAL STUDIES’

25

knowledge of (1) the textual practices involved in facilitating theatrical performance; and (2) the mechanics and dynamics of moveable type printing and book-making in early modern England. Most claims about a text’s authority, content, design, potential uses, and status vis-à-vis performance and publication require a betterthan-cursory understanding of textual (re)production. The printed plays we have inherited reflect (in various permutations) a whole variety of scribal texts that enabled performance in London’s commercial theatres: plot scenarios; actors’ cue scripts; backstage plots; the ‘approved’ book; free-floating scrolls of songs, proclamations, letters, etc. W. W. Greg studied these texts to identify the traces of non-authorial, theatre-derived interventions in the printed witnesses to Shakespeare’s plays (Greg 1931). But they have recently been taken up on their own terms to illustrate the range of text-based practices that supported theatrical performance – from collaborative authorship enabled by the early mapping out of plays into discrete scenic units (plot scenarios) to contracted company rehearsal time facilitated by cue scripts (which included just the actor’s lines and cues for when to deliver these lines) that permitted actors to study their roles individually (Stern 2009). This scholarship has required a great deal of ingenuity regarding evidence since the survival rate for these scribal ‘theatrical documents’, or ‘documents of performance’, is extremely low. Most documents associated with the early modern playhouses are fragmentary or do not survive at all. Tiffany Stern’s foundational work in this new wave of theatre history appeals to the documents that do survive – for instance, the lone cue script of the part of Orlando from a version of Orlando Furioso that doesn’t match up exactly with the extant editions of the same play – and corroborates the information such documents provide using references to theatrical processes in a variety of theatre-adjacent texts, including printed plays themselves (see the dramatization of a rehearsalprocess-gone-awry in A Midsummer Night’s Dream, for instance). Evelyn B. Tribble has situated these texts within the larger cognitive ecologies of the early modern playhouse, where they functioned as indispensable nodes in the dynamic network of physical, intellectual and creative resources that make up performance (2011). Although this scholarship may seem more at home under the aegis of theatre history, its focus on the production, design and use of texts means that it should count as textual studies, as well. Online resources such as the Henslowe-Alleyn Digitisation Project and a number of robust collection-specific digital image repositories, such as the Folger Shakespeare Library’s LUNA database, make some of the many theatrical documents associated with the early modern London theatres more accessible than ever before (Ioppolo 2020; and LUNA 2020). There is one particularly surprising avenue of study that has been opened wider by this evolving interest in playhouse documents and also deserves a place within the scope ‘textual studies’. It is concerned not with the presence of plays – as ‘full’ playscripts in the textual record – but rather with their absence. The turn to ‘lost plays’ is not entirely new. Scholarship on the lost ‘Cardenio’ and ‘Love’s Labour’s Won’ has dominated discussions of missing evidence until very recently because of both plays’ Shakespeare connections (McInnis and Steggle 2015: 6–7). But new research aimed at filling out the picture of early modern theatrical culture has been invigorated by The Lost Plays Database (LPD), a ‘major clearing house for early

26

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

modern theatrical information’ (Brown 2018; Knutson, McInnis and Steggle 2020). This open-access online forum constellates early accounts and images of texts that attest to the many, many plays – 744 at the most recent count – that are known to have once existed but which are now lost. Two key premises of the website as it bears on Shakespeare studies, specifically, are that a single author’s dramatic output cannot be properly understood without adequate context; and that lost plays, as far as they can be identified and even described, help to fill out our sense of the entertainment industry Shakespeare worked in, including information about ‘playwrights, playing companies, venues in London and the provinces, repertory studies, and audiences’ (Knutson, McInnis and Steggle 2020). Literary studies has rarely legitimated the study of lost plays (beyond attempts to recover and reconstruct, from an editorial perspective, lost plays with probable Shakespeare connection): ‘The ephemeral nature of lost plays has prevented them from becoming the primary focus of literary scholarship as long as that scholarship is driven by formalist tendencies (since textual analysis cannot be performed without a text)’ (McInnis and Steggle 2015: 6). The same could be said for textual studies scholarship, where methods of analysis depend heavily on material presence. But LPD and other projects such as the massive Records of Early English Drama project (where references to drama of all kinds from the period have been transcribed from local records) are making available new kinds of evidence at the same time as they model new evidentiary standards (REED 2020). The LPD’s mission to aggregate information about lost plays from a wide range of sources promotes the idea that an early modern play is ‘not simply … a literary text that happens to be performed, but … a complex and multi-faceted cultural phenomenon in its own right’ (Steggle 2015: 2). Here, David McInnis, Matthew Steggle and the database’s co-founder Roslyn L. Knutson see eye-to-eye with Stern on what the evidence tells us: plays were neither conceived of nor performed as integral textual units; plays were ‘pieced together out of a collection of odds and ends’ (Stern 2009: 1). Still, printed plays, which comprise the largest store of textual evidence relating to ‘Shakespeare’, give the impression (despite textual instability generated by the processes of moveable type printing) that ‘the play’ is, as Stern puts it, a ‘single whole entity’. As Margaret Jane Kidnie has pointed out, we typically have little problem ‘locat[ing] Hamlet’, for example, in ‘a variety of non-identical rare books’ and ‘an unspecified range of editorially-mediated modern versions of those books’ (2005: 101). But what are printed plays evidence of exactly? In the last century, there were two major shifts in the way printed plays were viewed by scholars, editors and theatre practitioners. The New Bibliographers, in general, treated them as evidence of authorial intention obfuscated by the material contingencies of the theatre and the printing house. The turn towards performance-oriented criticism in the second half of the twentieth century brought with it a turn towards printed plays as sources of evidence for how plays were composed and performed in Shakespeare’s time. Finally, a renewed interest in the materiality of the text and its contexts led to printed plays being used as evidence of plays-made-marketable – and ‘literary’, even – in a new media environment thanks to the investments and interventions of people in the book trade. This last shift is exemplified by the recent adjustment of critical

SHAKESPEARE AND ‘TEXTUAL STUDIES’

27

attitude towards playbook paratexts from one in which error, contingency and claims by stationers of bad behaviour on the part of other stationers were essentially taken at face value to one in which these same claims can now be read as advertising strategies, or ‘rhetorical set pieces in their own right’ (Lesser 2015: 467). The fact that Thomas L. Berger, Sonia Massai and Tania Demetriou recently collected the paratexts from pre-1642 English playbooks into a massive two-volume reference work – and that this work will soon be available as an online database – testifies to ongoing interest in these materials as barometers of and sources of information about the mediation and marketing of plays in early modern England (2014). Taking the packaging of printed plays seriously in a book-trade context was invigorated by Zachary Lesser’s observation that publishers were plays’ ‘first readers’ outside the theatre (Lesser 2004: 18). The active role of publishers in designing playbooks for early readers and thus fashioning the canon of early modern drama has been taken up variously in the last decade and a half. Studies of specific publishers have proliferated. These usually address why certain plays were published at certain moments when it seems that the majority of plays written for London’s commercial theatres before 1642 were never printed. Work in this vein also sometimes offers explanations for textual idiosyncrasies that have previously vexed critics and editors. For instance, a flourishing of work around the publisher Richard Jones, who had a hand in several playbooks in the late sixteenth century, including three octavo editions of Christopher Marlowe’s Tamburlaine, has shown how he was trying to fit Tamburlaine into a market for prose romances that he had cultivated in the 1580s, at the same time as he remained responsive to shifting taste among theatre audiences for tragedy over history as the 1590s progressed (Melnikoff 2005; and Lyons 2017). It is now accepted that Jones’s interventionist publishing strategies fundamentally shaped the versions of the playtext that survives to us today. The same has been shown to be true for Shakespeare. Andrew Wise, for example, has been credited with consolidating a market for Shakespeare in print by investing in no fewer than five titles by 1600 (Hooks 2016: 66–98). Among these titles were Richard II, Richard III and 1 Henry IV, that is, the most popular of Shakespeare’s plays in print, not to mention three of the most popular plays of the whole period based on reprint rates. Wise’s ‘ventures’ surely influenced the increasing visibility of Shakespeare’s name in the marketing of printing plays (Erne 2013: 56–89). These studies have primarily turned to playbook paratexts, book trade contexts and related documents (including other books in the same publisher’s portfolio as well as the now-digitized (although subscription-only) records of the Stationers’ Company) (Literary Print Culture 2020). More recent work has excavated new meaning from certain unstudied (and previously unnoticed) material features of playbooks in order to construct new histories of how plays by Shakespeare and other playwrights were designed, marketed, collected and read. A representative example is Aaron T. Pratt’s work on why the quarto became the prevailing format for playbooks starting in the late sixteenth century (2015). Pratt’s argument is deceptively simple: play quartos, at six or seven quires, were thin enough to be stitched and sold without expensive bindings, keeping the cost down for readers and allowing them to collect and later bind these pamphlets into customized, portable

28

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

collections. Pratt looked to something few scholars had heeded before: the presence of residual stab-stitching holes in the left-hand margins of many surviving playbooks, evident both in the relatively few still preserved in these customized collections, or sammelbände, and those quartos that might once have been bound with other quartos but which were subsequently disbound by nineteenth-century book-dealers and collectors and turned into single-text books. Pratt’s work on stab-stitching exemplifies a new kind of energy within textual studies – one that bends towards copy-specific evidence. The most compelling versions of such work recognize that it is only possible to approximate what a material feature may have meant to the makers or readers of a given copy, or (more basically) whether that feature is exceptional or exemplary, by surveying a range of related materials. Pratt, for example, surveyed just over 1,800 pre-1641 quartos of various genres for stab-stitching holes, using the length of the longest quarto playbook published in the period as an upper limit for his data set (2005: 313–14). From the remaining 1,499 quartos, he was able to determine that stab-stitching had everything to do with quarto length and nothing to do with form or genre – stabstitching was an economical way to sell books of a certain length, which included play quartos. Ongoing efforts to reconstruct these sammelbände aim to get a measure of how early readers gathered and organized plays in their personal libraries. These methods require a good deal of ingenuity. Tara L. Lyons has been using clues from individually bound quartos now scattered across different collections as well as peripheral evidence in contemporary library catalogues to reconstruct these readertailored volumes (Lyons 2020). And Jeffrey Todd Knight lighted (quite literally) on another method for reverse-engineering these collections several years ago, when he noticed a faint ‘ghost image’ of letters from the title page of Thomas Heywood’s A Woman Killed with Kindness (1617) on the blank verso of the final page of the Huntington Library’s copy of Henry V (1619) (HEH 69323) (2010). This hinted at the possibility that the Heywood quarto was once collected with Henry V and the nine other Shakespeare plays that were conceptualized as a collection by Thomas Pavier and William Jaggard in 1619 (Lesser forthcoming). Stab-stitching holes in the empty margins of play quartos and ghost images of once-adjacent texts that linger on the opening and terminal blanks of these books constitute a new kind of evidence in the vibrant sub-field of textual studies known as the history of reading. The significance of these textual phenomena is not always immediately obvious. Nor are the phenomena themselves necessarily obvious – or even visible – to the untrained eye. As Knight has noted, there is a new ‘willingness to make inroads into the “invisible” – that is, the field of evidence not readily apprehended according to established categories’ (2010: 59). By ‘established categories’, Knight is referring primarily to marginalia and annotations (i.e. what early readers wrote in their books about the text they were reading) and other kinds of visible interventions (such as ownership inscriptions; the scoring of passages for commonplacing; or the use of blank space in books for doodling, household accounts and other uses unrelated to the text). As the payoffs of attending to ghost images and stab-stitching holes suggest, there is more worth seeing than initially meets the eye.

SHAKESPEARE AND ‘TEXTUAL STUDIES’

29

The turn to unique copies of books – and the desire to excavate a sense of how early readers encountered printed books from fragmentary records of reading – is simultaneously exciting and frustrating. Roger Stoddard beautifully captures the frisson of our encounters with such objects coupled with the disappointment over traces of early reading that resist interpretation: When we handle books sensitively, observing them closely so as to learn as much as we can from them, we discover a thousand little mysteries …. In and around, beneath and across them we may find traces, some bold, some indistinct, that could teach us a lot if we could make them out and read them also. Not so easy is this top layer of signs, some of which we may overlook if our senses of sight and touch have not been alerted. (1985: 1) The challenge of attending to the marks and other signs of use made by readers in early printed books is distinguishing ‘idiosyncratic and whimsical’ interventions (which can be interesting in their own right if properly contextualized) from interventions that are found to be representative of broader reading practices (Grafton and Jardine 1990: 76). As with Pratt’s work on stab-stitching, the only way to make such a distinction is through the measured and slow exposure to a critical mass of books from the period. William H. Sherman and others have rightly cautioned that traces of reading are incomplete and thus ‘resistant to grand theories and master narratives’ (2008: xvi). The alternative to extrapolating sweeping conclusions from material particularities is to situate delicate case studies of discrete book objects in the context of ‘what is possible and (to some extent) what is normal for readers to do’ with their books (Sherman 2008: xvi). But establishing such a horizon of expectations is neither easy nor cheap (in terms of financial support and time), a problem I return to in the next section. Indeed, Sherman, Knight and Pratt all enjoyed both luxuries to survey significant book collections (Bourne 2017: 369). Of course, the study of play-reading requires an added awareness of drama’s bimodality: reading a play, in the first place, is a tricky thing. Printing plays in ways that made them legible as plays was no straightforward task either, but it was a necessary prerequisite for reading them (Bourne 2020). Designs for play-reading are one thing. How plays were actually read (as far as we can tell) is sometimes quite another. On the one hand, early modern readers engaged with the printed texts of plays in the same ways as they engaged with non-dramatic material. They combed playbooks for passages that could be copied into commonplace books and repurposed for other uses. As Laura Estill has shown, play extracts are prevalent in surviving manuscript commonplace books from the period (Estill 2015). The Database of Dramatic Extracts (DEx), which Estill and Beatrice Montedoro created to provide a wide-angle sense of the practice by gathering these extracts all in one place, affords us a glimpse of what caught the eyes, ears and imaginations of early play-readers. It was also not unusual for early readers to emend printed plays by hand. In fact, such corrections are the most common kind of readers’ mark in printed plays. As Sonia Massai has shown, attempts to iron out ‘nonsensical readings’ in a play’s dialogue, to finesse missing or incorrect speech prefixes, and to add or correct stage directions can frequently be found inscribed on pages of playbooks printed in the sixteenth

30

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

and seventeenth centuries (2007: 14). The impulse to ‘perfect’ a printed text was not, in and of itself, specific to reading drama. However, making changes to speech prefixes and engaging with stage directions required a specifically dramatic mode of reading that attended to – and considered how text could mediate – non-spoken (and sometimes unprintable) dimensions of theatricality (Syme 2011–12; Bourne 2017: 377–9). An important methodological sticking point that affects research on early reading has been whether plays received special status in the book trade or whether they were treated similarly to other genres of imaginative writing. Peter W. M. Blayney’s famous take-down of A. W. Pollard’s piracy theory introduced the notion that, despite Shakespeare’s cultural status in the twentieth century, plays were generally insignificant investments – not lucrative enough to ‘pirate’ and, presumably, not desirable enough to make special accommodations for (1997; Pollard 1917). Thomas  Bodley’s apparent reluctance to ‘stuffe’ the Bodleian Library full of playbooks – ‘[h]appely some plaies may be worthy the keeping: but hardly one in fortie’ – only helped reinforce the notion that printed plays were not worth preserving (qtd in Wheeler 1926: 222). While agreeing with Blayney’s critique of Pollard, Alan B. Farmer and Lesser argued that plays occupied a somewhat more prominent role in the book trade: while most playbooks didn’t sell ‘like hot cakes’, plays were, as a category, reliable investments – they sold well (Farmer and Lesser 2005; Blayney 1997: 384). Adam G. Hooks’s study of mid-seventeenth-century bookseller catalogues has shown that plays were indeed treated separately from other so-called ‘literary’ genres, not because they were inferior but because they were ‘vendible’ (2016: 143). In these advertisements, plays get their own special category heading that separates them out from novels, romances and poetry ‘in what are overall rather haphazardly organized lists’ (2008: 446). Hooks also points to efforts in the second half of the seventeenth century by Francis Kirkman and, later, Gerard Langbaine to variously account for the titles of all the plays ‘that were ever yet printed and published’ (Kirkman 1661: sig. A1r). This shift towards recognizing the special place that playbooks occupied in the seventeenth-century book trade raises the question of whether – and how – readers responded to this generic consolidation on the production side. Understanding that quarto playbooks were viable book-trade commodities has piqued scholarly interest in identifying the readers – both noble and ‘middling’ – who read plays (not least those by Shakespeare) (Mayer 2018; Scott-Warren 2019). Evidence of early book collecting has offered one way into these histories but presents a couple of methodological hurdles: (1) identifying the distinguishing features of playbooks that once belonged to the same collection; and (2) reconstructing a (sometimes radically) dispersed collection. The most famous contemporary collection of early modern playbooks belonged to Frances Egerton, Countess of Bridgewater (1583–1636). When its customized sammelbände were absorbed into the family’s Bridgewater House Library in the 1620s, the collections were disbound and the play quartos rebound into single volumes. They remained part of the Bridgewater collection until it was purchased by Henry E. Huntington in 1917 (MacLean and Manley 2014: 327–9). However, Huntington soon sold off many of the playbooks in the collection that duplicated his holdings. Lawrence Manley has reconstructed some of the original sammelbände, including one of ‘Diuers Playes by Shakespeare’,

SHAKESPEARE AND ‘TEXTUAL STUDIES’

31

using the characteristic ‘numeral inside a square bracket’ found in the top right-hand corner of title pages of most quarto playbooks (at the Huntington and elsewhere) that bear the Bridgewater Library stamp. If Manley is right that the volume contained all the Shakespeare quartos published before 1602 (save two), then that would mean, as Erne has pointed out, that Egerton is the earliest reader we know of to date to collect Shakespeare’s ‘complete dramatic works’ (2013: 205–6). Much more challenging has been the slow but steady attempt to reconstruct the library of Frances Wolfreston (1607–77), a gentrywoman from Staffordshire who amassed a library of almost a thousand volumes, including the sole extant copy of the first edition of Shakespeare’s Venus and Adonis (1593) and play quartos by Shakespeare, Chapman, Heywood, Dekker and Marlowe. In the late 1980s, Paul Morgan catalogued about a quarter of the approximately 400 Wolfreston books sold by Sotheby’s in 1856. Luckily for Morgan and more recently for Sarah Lindenbaum and others who have continued to identify Wolfreston volumes in institutions ranging from the British Library to the Boston Public Library, Wolfreston wrote in her books (Lindenbaum, Newcomb and Moschella 2015). She would often inscribe some version of the phrase ‘frances wolfreston hor bouk’ on the title page or early in the copy. Even more enticing is the fact that Wolfreston sometimes provided plot summaries and brief commentary on playreading. For instance, she called The Taming of the Shrew ‘a very prity mery one’ and Othello ‘a sad one’ (Bourne 2017: 374–5). But finding these books, now scattered in libraries across the world, is another major practical problem, one that could be alleviated by the slow addition of copy-specific notes to online catalogue records (Ziegler 2020). Still, exposure and serendipity – and thus also time – play enormous roles in this kind of scholarship. Of course, fewer readers than we would wish for left ownership inscriptions or systematic marks on/in their books – and fewer of those interventions than we would want survive. For one, books that were heavily used tend to survive at a lower rate than more pristine copies. This, along with the fact that other evidence of playbook collecting – binding in sammelbände – was destroyed by past collectors and book-dealers, limits the potential of collecting practices as a heuristic for studying the reception of Shakespeare and early modern literature. However, there is still potential in certain unseen – because unlooked for – traces of play-reading. Even though many copies of plays underwent significant physical transformations in the nineteenth century, not least from having their pages washed (literally) of readers’ inky interventions, traces of the marks and formations that were obliterated sometimes still assert themselves. The ‘ghost images’ I discuss above are a good example of textual configurations persisting. Washed annotations are another. For instance, a copy of Thomas Heywood’s The Golden Age (1611) held in the Elizabethan Club collection at Yale University is described in the collection’s published catalogue as: The first edition. Bound by W. Pratt in calf, gold tooling on cover and spine, gilt edges. Title and several leaves repaired, final leaf inserted from a shorter copy. The Hoe copy (with bookplate), sold by Anderson Auction Co., New York, 28 April 1911, lot 1687. Gift of Alexander S. Cochran, December 1911. (Park 1986: 130)

32

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

The note traces the ownership of this quarto playbook to the collection of nineteenthcentury businessman and collector Robert Hoe. Indeed, it lifts language almost verbatim from the description of the slim volume that appeared in the 1911 auction catalogue of Hoe’s library (Catalogue of the Library of Robert Hoe 1911–12: 278). There, the book is excused for being imperfect: prospective buyers were cautioned that Hoe’s copy of The Golden Age would be ‘[s]old with all faults’ (presumably referring to the repaired leaves and the supplied final page). Neither description makes any mention of the book’s many marginal annotations. This may have to do with the undesirability of such notes in the context of early twentieth-century book collecting. But it could also be because they are all too easy to miss if you do not know to look for them: they are so faint as to be nearly invisible. (Stephen Park, the former Elizabethan Club librarian and an esteemed bibliographer, took no note of them in the catalogue of the club’s collection.) What’s more, there is also, in this quarto, a washed-out ownership inscription above the printer’s ornament at the top of sig. B1r, the first page of the playtext. So what? With the help of digital photography and the basic image processing algorithms available (for free) at retroReveal.org, the nearly invisible inscription and some of the marginal annotations began to pop. Indeed, the retroReveal website explains that its web-based image processing tool is ‘designed to help people discover hidden content’ (2020). Within a matter of minutes, these filtered images afforded the book a much deeper provenance history. The reader who signed the book and filled its margins with commentary was none other than Frances Wolfreston. Not only that, but this quarto now happens to be among the most heavily marked up of any that Wolfreston is known (so far) to have owned (Lindenbaum 2019). With such a large collection of Wolfreston books already boasting evidence of involved reading practices, this quarto is a boon. However, the faintness of the annotations and thus our inability to actually read many of them is a critical obstacle to studying Wolfreston’s commentary and readerly sensibilities. Basic image processing only got us so far with this book, so former Elizabethan Club Librarian Anders Winroth enlisted the help of a computer scientist to do some more sophisticated manipulation of high-resolution digital images, which increased the legibility of some (but not all) of the annotations and allowed Lindenbaum to begin transcribing what Wolfreston had to say about Heywood’s play (2019). Besides digital image processing (which remains the most effective way of rendering the ‘invisible’ traces of reading visible), different kinds of light – ultra-violet, infrared, multi-spectral and visible – have been used, with varying degrees of success, to make faded annotations more legible (Werner 2012). Even as these technologies enhance our ability to see the faint traces of playbook use, such marks, notes and commentary often still resist our desires to interpret them. What do they mean? What do we do with them? What do they tell us about the reader, the object, the text printed on the pages of the object, and how object and text were encountered, used, (mis)understood, etc.? My own study of the 700+ marks in the copy of Mr. William Shakespeares Comedies, Histories, & Tragedies (1623) at the Free Library of Philadelphia led to some insight about the then-unnamed early reader’s reading practices: namely collation and possibly

SHAKESPEARE AND ‘TEXTUAL STUDIES’

33

commonplacing (Bourne 2018). I established that this reader was collating the 1623 folio texts of Romeo and Juliet and Hamlet against quartos published in 1637, both of which offer local, variant readings of key passages. This reader’s fascination with textual variance invited interpretation, and while I could not be certain at the time that my readings of the variant passages would have aligned perfectly with the mid-seventeenth-century reader’s peculiar interest in these textual differences, I offered what I thought were plausible explanations: calling Claudius a ‘blunt’ king (F1) (i.e. impotent) versus a ‘bloat’ king (Q5) (i.e. over-indulgent), for example, fundamentally changes our understanding of Hamlet’s attitude towards his uncle. Sometimes, though, new contexts – usually born from different ways of looking at the same evidence – shed new light (metaphorically speaking, in this case) on such isolated (singular?) examples of reading Shakespeare. In this case, Jason ScottWarren saw photos of the readers’ marks published with my essay and recognized the handwriting in the book as that of John Milton. (We are currently working to present the case for this attribution.) What Scott-Warren’s identification of the hand means is that Milton’s entire corpus of published and unpublished writings may now be available as lenses through which to see and interpret what is written in this copy of the 1623 Shakespeare folio. But no matter how close we think we get to connecting a passage marked by Milton in his folio to something that Milton actually wrote, any interpretation of copy-specific evidence is provisional. The most responsible scholarship of the kind strives to situate individual acts of reading and use alongside a working awareness of what it is reasonable to expect early readers to have done with their books. The best we can do – and I would argue that it’s quite a lot – is tell plausible narratives about the relationship between reader(s) and text. As the case of/for Milton’s Shakespeare illustrates, frames of reference are crucial to shifting the conversation about known evidence. (The FLP copy of the folio was, after all, already well described (Bourne 2018: 197–206).) Equally important are access to the material text (either physical access or remote access via some kind of reproductive technology) and an awareness of the text’s status as an object with a deep history that must be understood in the instant. It is to these concerns that I now turn.

*** The previous section discussed at some length what counts as evidence in the networked subfields of ‘textual studies’ as they intersect with Shakespeare studies, as well as how that evidence lends itself to (or resists) interpretation. I limited my discussion to texts produced in the early modern period: in the contexts of theatrical production, the book trade and reading. This section on scale, periodization and access will address why it is important to understand, if not take up as objects of study in their own right, texts produced, manipulated, revised, mediated, edited and used in any number of ways outside the period we tend to call ‘Shakespeare’s time’. Jeffrey Todd Knight has identified within recent major studies of Shakespeare’s early publication histories what he calls ‘a tacit conflict between measures or scales of analysis’ (2017). The studies that Knight considers – Lukas Erne’s Shakespeare

34

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

and the Book Trade (2013), which is ‘synthetic and enumerative’ in its methods, and Adam G. Hooks’s Selling Shakespeare (2016), which uses a case-study model – are both primarily concerned with the early modern period and are both, at their core, historicist. Both Erne and Hooks aim to reconstruct Shakespeare’s presence in the London book trade. But the ‘conflict’ that Knight identifies is not limited to studies that root themselves in particular historical moments. Knight’s own work on early modern practices of compilation and binding relies heavily on what Lesser has called a ‘polychronic’ approach to textual studies (Lesser 2020a: 92). In Bound to Read (2013), scale operates both according to type of material and time. Knight surveys Archbishop Matthew Parker’s library, a relatively complete and intact collection of early modern books at Corpus Christi College, Cambridge, to establish a horizon of expectations for how an early modern reader might have collected, compiled and organized printed matter. This establishes a type of textual practice at scale. But Knight’s work also transcends periodization boundaries in that much of his analysis centres on the nineteenth century, when the kinds of practices evident across the books in Parker’s library were effaced in the remaking of other early books according to different principles of value – ones that were shaped by a thriving market for collecting rare books (ibid.: 15). The shifting scales of time in newer textual studies scholarship about Shakespeare signal a re-evaluation of traditional periodization. So much of the work that formed the New Textualist wave of book trade-oriented scholarship in the late 1990s and early 2000s was invested in the period during and around Shakespeare’s career and thus primarily on textual production. The new trend is towards a palimpsestic conception of textual history where, for example, texts have pre-publication histories and post-publication lives – both of which we have to understand to make claims about the textual features of the documents that transmit them. Looking backward, we might ask how books of poems and plays printed in the early modern period reflect the influence of page and book design practices that might extend back centuries (Atkin 2018). Looking forward, we might also account for all the ways that these books could be transmitted, encountered, read, modified, collected, damaged, repaired, conserved, studied and otherwise used. Indeed, the recent flourishing of studies about the role that print played in consolidating a canon of early modern drama – and turning Shakespeare into a canonical author – shows the value of rethinking periodization divides (Lopez 2014; Depledge and Kirwan 2017). One result of these deeper engagements – with broader textual histories that might shape the way a text looks and is used and with its mutability as an object – is the recognition that a book’s makers are only a small part of its story. As Alice Dailey reminds us, books are ‘unit[s] of time’ that have ‘overlapping and contiguous durations’ (2020). Not understanding how and why books change over time may cause us, in Lesser’s words, ‘to misrecognize what we are seeing’ in these ‘forever unfinished’ objects (2020a: 92). In cases where one wants to explain a localized textual phenomenon, surveying material is an extremely effective, if counter-intuitive, strategy. Scaling up knowledge of a certain type of book (say, quarto playbooks or sonnet collections) or a certain

SHAKESPEARE AND ‘TEXTUAL STUDIES’

35

type of textual practice (say, stab-stitching or commonplacing) by surveying a large number of relevant examples establishes what is conventional and what is exceptional when it comes to the type of book or practice under consideration. This method of putting the particular into conversation with the general – that is, using the general to explain the particular – has been gaining traction. As a result, the scale of textual scholarship on Shakespeare is both narrowing and broadening at the same time: it is homing in on the individual copy as a unit of analysis and broadening towards a completist (or survey) model. Both have been enabled by greater access to digested information and materials: to new and revised reference works, many of them online, that make it easier than in the past to get a wide-angle view of the evidentiary terrain; to the increasing willingness of library fellowships to fund projects that endeavour to survey materials; and to a variety of increasingly accessible resources (such as more robust copy-specific notes in online library catalogues, open-access digital image repositories, especially of items from less-trafficked collections, and social media sharing) that facilitate the identification and study of textual objects. I have collected and annotated links to many of these online resources on my website (Bourne 2019). As products of enumeration, most of the reference works we use to study Shakespeare’s textual presence in our own time – and his – enable this kind of contextualization: from the relatively narrow, genre-specific catalogue of playbook paratexts (assembled by Massai, Berger and Demetriou and mentioned above) to the much broader Short-Title Catalogue of Books Printed in England, Scotland, and Ireland, and of English Books Printed Abroad, 1475–1640 (STC) (compiled by A. W. Pollard and G. R. Redgrave and revised by Katherine Pantzer). These resources are essentially lists that provide codified descriptions of books (in the case of the STC) or particular attributes of those books (in the case of Paratexts). They are designed to give a sense of what material is available for study in the first place, although depending on how much information the entries contain, some scholars might use them as proxies for the thing itself. Indeed, before the advent of online catalogues and easy-ish air travel, an STC entry might have been the only way to obtain information about an edition and its known copies. The STC, for example, includes material printed (at least in part) using moveable type in the British Isles and its colonies (or in English elsewhere in the world) before 1641. As Ian Gadd has noted, the STC does not provide ‘a full representation of Britain’s print culture prior to 1641’ because it leaves out, for instance, Latin books printed on the Continent but imported into England starting at the end of the fifteenth century (2009: 683).4 In their introduction to the 1926 edition, Pollard and Redgrave were clear about another of the catalogue’s limitations: access. The STC, they explained, records ‘books of which its compilers have been able to locate copies’, that is, not all ‘books known or believed to have been produced’ (Pollard and Redgrave 1926: xiii). This acknowledgment that the scope of the STC is only as wide as its makers’ knowledge, in turn, reveals another notable tension: the STC is a catalogue of editions (and issues), not of copies, despite the fact that the bibliographic information it provides derives from specific copies and that each entry offers a list of libraries where copies of the given title – no more than five

36

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

from either side of the Atlantic – were known (at the time of publication) to be housed. Pollard was well aware of this tension, cautioning users that the STC might give the illusion of fixed, stable texts when, in fact, the conditions of moveable type printing meant that no two copies of a title were the same. (The compilers of the STC were interested in the publication details of early modern editions, not post-production evidence of reading and use in individual copies of those editions.) Indeed, because of the ‘mixed character of its sources’ (presumably both the incomplete nature of the record and the variation between different copies of the same title), Pollard called the STC ‘a dangerous work for anyone to handle lazily’ (qtd in Pantzer, Ferguson and Jackson 1976–91: vii). The STC’s peculiarities and admitted shortcomings linger in the English Short Title Catalogue (ESTC) and Early English Books Online (EEBO), and anyone using these resources for research on Shakespeare’s textual presence and contexts in the sixteenth and seventeenth centuries needs to be aware of these relationships. The problem is that neither the ESTC nor EEBO is transparent about its genealogy or how each relates to the other. The ESTC began as the born-digital Eighteenth Century Short Title Catalogue before records from the print STC and Donald Wing’s sequel catalogue of post-1640 editions (1945–51) were incorporated and it was renamed the English Short Title Catalogue. After a decade on CD-ROM, this composite database was made freely available on the British Library’s website in 2006, allowing keyword-searching of bibliographical data for approximately 470,000 pre-1801 editions of books printed in the British Isles, its colonies or elsewhere printed in English (Gadd 2009: 684). Until very recently, the online ESTC offered the only centralized way of locating copies of pre-modern Shakespeare editions online. The distinction of providing the fullest record of such copies now goes to the newly launched Shakespeare Census, which endeavours to locate, verify and aggregate information about ‘all extant copies of all editions of Shakespeare’s works through 1700, excluding the folios’ (Hooks and Lesser 2018). (Copies of the First Folio have been catalogued three times in the last 120 years: by Sidney Lee in 1902; by Anthony James West in 2001; and by Eric Rasmussen and West in 2012. Sarah Werner also maintains a list of digitized First Folios on her website (2019).) The Census is a significant reassessment of Shakespeare’s early textual presence and is quickly becoming the go-to resource for anyone wishing to conduct copyspecific research on the pre-1701 Shakespearean text. Developed by Hooks and Lesser, the Census used data scraped from the ESTC about copy locations as a starting point. Hooks and Lesser then aggregated ESTC information about specific copies with entries in The Census of Shakespeare’s Plays in Quarto 1594–1709, which was compiled by Pollard and Henrietta C. Bartlett in 1916 and revised and expanded by Bartlett in 1939 (Hooks and Lesser 2018). Some of the copies now verified to exist were listed and described in Bartlett’s updated Census but not accounted for in the ESTC. Even then, Bartlett’s descriptions of the approximately 1,220 copies she found (a remarkable number given that she sourced her information via trans-Atlantic and trans-continental postal correspondence with librarians and acquaintances in the book trade) privileged the kind of information that would be useful to book dealers and collectors: size, descriptions of the binding, evidence

SHAKESPEARE AND ‘TEXTUAL STUDIES’

37

of sophistication (that is, leaves supplied from another copy) and details about provenance. The Census reproduces Bartlett’s entries from 1916 and 1939 but is continually adding copy-specific information of the kind that could support scholarship on the history of reading: marginalia, annotations and other readers’ marks; evidence of washed notes and other interventions; descriptions of damage; if in collection, what other titles the copy is bound with (Bartlett mentions if a copy was in collection but doesn’t give more details than that); the presence of stab-stitch holes; etc. The truly amazing number related to the new online Census (as it stands at the time of publication) is this: 422 (Lesser 2020b). That is the number of nonESTC copies that Lesser and Hooks have identified – about 23 per cent of the 1,855 total copies currently recorded in the census. Some of these copies were identified by searching library catalogues and enlisting the help of librarians to verify the presence of those copies in their collections. Others surfaced unsolicited as publicity about the new Census circulated, especially on social media platforms like Twitter. Furthermore, more than fifty copies that the ESTC recorded as existing are actually what Lesser and Hooks have termed ‘ghost’ copies (i.e. they’ve turned out not to exist). Currently an additional twenty copies listed in the ESTC (or in library catalogues) are ‘unverified’ – they may or may not exist. Lesser and Hooks’s success at identifying so many copies outside of the ESTC testifies to the crucial role that cataloguers, librarians and curators play in generating the knowledge that underwrites textual studies research. More and more special collections libraries are adding searchable copy-specific information to their online catalogue records, for example. And personal research consultations with publicfacing librarians can go a long way in helping researchers navigate collections to find what they’re looking for or what they didn’t know they were looking for. In the case of the Census, many rare book librarians have reached out to verify unverified records, to identify copies not yet listed in the Census, and to provide copy-specific details for copies that are. This kind of labour is also evident in projects like ‘Renaissance Books, Midwestern Libraries’, a collaboration among librarians, scholars and students at institutions in midwestern America to report eligible items at their institutions’ libraries to the ESTC. One goal of the project, which was spear-headed by Andrew Keener when he was a graduate student in English at Northwestern University, is to generate exposure for un- and under-studied copies of early books, that is, to ‘facilitate archival discoveries for Renaissance scholars in the United States and abroad’. In the span of a year (2014–15), the project increased ESTC representation of items in Northwestern’s Special Collections more than six-fold: from 188 items to over 1,300. While none of these are early books of Shakespeare, this project – like the Census – serves as a model for expanding previous efforts of enumeration and record-keeping and thus access. Thanks to the Census, information about individual copies of Shakespearean plays and poems printed before the eighteenth century is more accessible than ever before. Nowhere is the shift in the field from an interest in publication (with an emphasis on editions) towards reception (with an emphasis on copies) more pronounced than when comparing the Census to the Database of Early English Playbooks (DEEP), Lesser’s earlier open-access database, developed in collaboration with Alan B. Farmer

38

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

(2007). DEEP borrows its organizational logic from W. W. Greg’s Bibliography of the English Printed Drama to the Restoration, which began rolling out in 1939, the same year as Bartlett’s revised Census was published. The central structuring category of Greg’s Bibliography was – and thus also the conceptual framework for DEEP is – the edition. An edition is defined as all the copies of a book set from substantially the same setting of type (minus minor typographical differences). The edition was a powerful organizational unit for scholarship on the book trade because, as Lesser has pointed out, it was ‘at the level of editions that early modern publishers made their decisions about how to invest their capital’ (2020a: 83). The Census, by contrast, is organized by and contains information about copies, individual and unique exemplars of an edition that, in addition to the printed text, transmit signs of use and abuse accreted since they were printed. If you want to study the publication history of a Shakespearean play, you start with DEEP. If you want to study the history of reading, collecting, preservation, conservation, etc., you start with the Census. By making ‘the copy’ more readily available as a unit of analysis, the Census is a promising resource for new work on the Shakespearean text. ‘The edition’ has almost – by default – dominated studies of the early texts for a long time. The editorial debates that informed scholarship in the twentieth century centred around which edition was the most authoritative – Greg’s brand of eclectic editing produced a single desired text from multiple editions while the versioning of King Lear in the 1986 Oxford Shakespeare (Wells and Taylor) inaugurated a single-text editing trend where what readers got were two discrete modern editions that each reflected a different early modern edition (Q1 or F1). Insofar as the New Bibliographers attended to copies in their editorial practices, it was principally to track localized variants in the printed text across an edition. The unique physical attributes of specific copies were of little import. Even early collectors of Shakespeariana – among them, Edward Capell, David Garrick, John Philip Kemble and Edmond Malone – sought representative copies of editions for their collections. The privileging of the edition as a unit of analysis has only been reinforced by Early English Books Online, the highly trafficked subscription database that offers images and bibliographic information for about 125,000 mostly English books printed before 1701. EEBO is based on a massive effort to make the images of every edition listed in the STC – and eventually also the Wing catalogue – accessible to scholars who did not have ready access to these texts at their home institutions. Between 1938 and the late 1990s, University Microfilms (UMI), the brainchild of American publisher Eugene B. Power, generated several thousand microfilm reels of black-and-white photographs of English printed books (Gadd 2009; Mak 2014). The particular copies used in the imaging project were chosen based on STC and Wing entries, with a bias towards copies in major, elite collections such as the British Library, the Bodleian Library, the Huntington Library and the Folger Shakespeare Library. UMI eventually digitized the microfilmed images and published them online in 1998 as EEBO. The image sets available for each item (i.e. edition and issue) on EEBO are three times removed from the physical copy of the edition they purport to exemplify.

SHAKESPEARE AND ‘TEXTUAL STUDIES’

39

There is not nearly enough space here to discuss all the implications of these layers of mediation. Others have written cogently about these issues (Gadd 2009; Kichuk 2007; Mak 2014). But there are a few things that anyone using EEBO for textual scholarship should note. EEBO gives the illusion that what you get when you view or download an item are images of the thing itself. First of all, EEBO often makes it impossible to even know what ‘the thing itself’ is, since it does not provide information about the copy (like a shelfmark) beyond the name of the collection where the copy is held (or was held at the time it was photographed for the UMI microfilm). Sometimes the repository is not named. Bibliographers and book historians working in the digital age will tell you that EEBO is a good entry point but that it is ‘best practice’ to consult the object in person if you wish to make any claims about its material features. Without a shelfmark, this can be a challenge. For instance, the EEBO record for the First Folio – Mr. William Shakespeares Comedies, Histories, & Tragedies (STC 22273) – indicates that the image set of this edition is a ‘[r]eproduction of an original in the Folger Shakespeare Library’. The Folger Shakespeare Library has eighty-two copies of this edition in its collection, making it impossible to identify which copy we are looking at. This matters because no two copies of the First Folio are identical, not just because of the contingencies of moveable type printing or the fact that Troilus and Cressida was added belatedly and thus appears in some copies and not others but also because each copy of the book has its own unique 400-year-old history that might affect what it is we are looking at. Some material features of books imaged in EEBO that do not visibly register in the images or are not included at all in the digitized black-and-white image set (and that also go unnoted in EEBO records) include: size (all images are the same size, whether the book is a small quarto or a large folio); different colour inks; pages supplied from other copies or even produced in facsimile to perfect a deficient book; instances of localized mending (where bits of text may have been supplied in facsimile); evidence that pages have been cut and inlaid in later paper frames; faint marginalia and other scribal marks; flyleaves and blanks (which may contain annotations or other inscriptions); bindings; etc. (Werner 2012). Furthermore, even though EEBO imports data about each item from the ESTC (see the ‘Details’ tab in the new ProQuest interface), one thing it does not consistently import is each edition’s collation formula. For example, records for Q1 Romeo and Juliet (STC 22322), Q2 Hamlet (STC 22276) and most other Shakespeare quartos offer this information. However, the record for the First Folio does not. This is symptomatic of the way EEBO conflates copy and edition and ends up eliding the realities of textual production in the hand-press era. It is inconvenient to provide the collation formula for the particular copy of F1 that EEBO reproduces because it exposes something about its eccentricities and shatters the illusion that one copy can exemplify all. Collation formulas are not always one-size-fits-all for copies of bibliographically complex editions. The assumption that the image sets of a particular copy can stand in for a whole edition is also evident in the fact that those in charge of choosing which copy of an edition would be photographed for UMI’s microfilm selected in favour of clean copies in all but the most extenuating circumstances (such as an edition where only

40

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

one or two copies survived – and those copies were marked up or damaged). The priority was to give access to the text – a priority that lives on in the EEBO-TCP (or Text Creation Partnership) project. EEBO-TCP now provides open-access transcriptions of the texts from almost 60,000 volumes in EEBO. EEBO-TCP makes these texts searchable – individually and in the aggregate – and it has also been used to generate data for recent computational studies of authorship attribution studies research. Of course, this kind of research, which extracts the text from its material context (or a highly mediated instantiation of its material context), also falls under the aegis of textual studies, not least because it has been used to inform recent editorial decisions like The New Oxford Shakespeare’s joint attribution of the Henry VI plays to Shakespeare, Marlowe and Nashe (Egan et al. 2016). Furthermore, the legacy of UMI’s preference for clean copies means that the image sets in EEBO risk giving a false sense that early modern readers did not write much in their books. It is also sometimes challenging to tell the difference between post-publication interventions and the printed text itself. In a particularly glaring example, one scholar identified small crosses (+) before Juventus’ speeches and symbols that look like turnstiles (⊢) after those speeches in the Bodleian’s c.  1551 quarto of Lusty Juventus on EEBO as printed glyphs when in fact they were supplied by hand. The turnstiles were not, in fact, the earliest appearances of dashes in a printed play – they were a reader’s notation system (Graham-White 1995: 32–3; Atkin 2018: 178–80). The poor quality of EEBO’s images sometimes makes it impossible to know what exactly is being mediated. The final thing to note about EEBO image sets is that they are glitchy. Some of this glitchiness results from the way the microfilms were digitized: EEBO images have been produced precisely by taking apart a reel of film: the reel was run through a digitizing machine, and each of the page openings filmed on it was then stored as a digital file separately from its neighbors. Evidence of that process can occasionally be seen in EEBO, almost always (as with bibliographic clues generally) where it has gone wrong. (Lesser 2019: 4) This process resulted in some images that show where ‘the film was not properly advanced from one scan to the next’; and ones where ‘the data became corrupted, resulting in digital static’. Sometimes notes describing some distortion to the printed page (foxing, staining, faint print) cover the printed page itself in the image. Although EEBO gives the impression of textual stability, exemplarity and even fidelity to the thing itself, EEBO image sets (or ‘books’, as EEBO insists they are) contain traces of their own making, not just behind the scenes – in the code and metadata that allow them to be displayed in the first place – but also in visible errors of transmission (Trettien 2019). Gadd notes that EEBO does not ‘provide an actual copy: a real book to hold in your hands’ (2009: 682). And this is true. Recently, however, Pratt (in a Twitter conversation) suggested that what EEBO offers is something akin to a copy: ‘a copy of a copy of an ed[ition], understanding that reproduction can intro[duce] error’ (2014). The distinction between edition and copy – between the exemplary (a copy as a stand-in for the whole edition) and the exceptional (the copy

SHAKESPEARE AND ‘TEXTUAL STUDIES’

41

as idiosyncratic object with its own unique, palimpsestic history) – is a powerful framework for assessing what EEBO is and what it is not. Perhaps ProQuest, which recently relaunched EEBO, will take some of the critique about its bibliographic obliqueness on board as it continues to roll out new features. More and more the linked fields of ‘textual studies’ are moving towards embedded copy-specific research, that is, scholarship that attends to the particular by setting specific cases in broader, illuminating contexts to find out whether the cases at hand are exceptional or exemplary. Within Shakespeare studies, this kind of work is enabled not only by the updated Shakespeare Census but also by the proliferation of digitization initiatives and the growing number of copies made available – many of them open access – in digital form. The British Library’s Shakespeare in Quarto website (n.d.) and the Folger’s LUNA image database (2020) are two of the most prominent examples. But notable digitization projects at institutions like the Harry Ransom Center (Digital Image Collections 2020), the Fondation Martin Bodmer (Early Modern English Books 2020) and the Boston Public Library (Early English Playbooks, 1594–1799 2017) (hosted by Internet Archive) are putting more copies (copies of copies?) of early Shakespeare online than ever before. Just as we need to be cautious about what exactly EEBO is showing us when we look at one of its ‘books’, we need to approach other mediated images of physical books (i.e. digital facsimiles) with a critical eye, always asking how the images presented relate to the object they represent. Sarah Werner has argued, in reference to digitized copies of the First Folio, that the relationship between image and object is never straightforward and is often dictated by the affordances and constraints of the given interface (Werner 2012; see also Galey 2014). Still, she reminds us that access is not merely about seeing and reading, but also about using digital images in ways that printed books resist: We are at a moment when digital facsimiles of the First Folio have been created primarily to act as surrogates for the physical books and to be encountered as discrete copies. But we are moving toward a time when digital facsimiles are going to be seen as digital objects in their own right: not as surrogates for a printed book or manuscript, but as different ways to experience that object. For some uses, the material text might be better suited; for others, a digital image might be a better choice. (Werner 2012) Werner’s point about the suitability of different media for different tasks is a salient one and speaks to something this chapter has been implying that should now be made explicit. We have many, many resources potentially available to us: all the extant copies of the books of Shakespeare’s plays and poems in material form; a growing number of copies of Shakespeare’s plays and poems in digital form; the knowledge we have and continue to gather about texts that have been lost; information about those copies and about their editions in various forms; fragmentary manuscript evidence relating to the performance and publication of Shakespeare’s plays; reference works and databases that digest different bits of that information in distinct ways; centuries of scholarship and editorial labour performed by those who have studied and interpreted these materials before us; etc. In an ideal world, all of

42

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

these things are accessible, if not to hand then somehow, somewhere, someday. We might cross-reference the item details of an EEBO ‘book’ with the entry for the same book in the ESTC (or even the print STC). We might identify a copy of the 1676 edition of Hamlet in the new Census that seems to have the hallmarks of a reader we are tracking and follow up this hunch by seeing the book in person. We might see in a Tweet the image of a book that relates to the thing we are working on and look up the item in the relevant library’s online catalogue for more information, and barring that, follow up with a curator or reader services librarian. We might find an ownership inscription in a playbook while working in a special collections library and comb through digital images of other playbooks online to see if the inscription shows up again, or maybe even search library catalogues we know to have robust copy-specific notes. The point is this: every resource has its shortcomings, and no method is onesize-fits all. Diana Kichuk has noted that, when it comes to EEBO, scholars ‘may be keenly aware of the limits of remediated primary sources, but it is a learned awareness’ (2007: 302). So much of the energy required to navigate the bumpy landscape of textual studies is honing a ‘learned awareness’ of what our evidence (and the way we access it) can tell us and what it cannot tell us. But where does that learned awareness come from? The book that you’re reading now and others like it are designed to alert you to past and ongoing attempts to interpret the textual corpus we now associate with ‘Shakespeare’. Sometimes, though, the only way to develop such a learned awareness comes from looking at a lot – I mean, a lot – of books. It is not that you need to look at every copy of Shakespeare listed in the Census or know everything about how scribal and printed texts are made, read, circulated, remediated online, etc., to practice textual scholarship. But you do need to have enough experience with these objects to know when something you are looking at is odd and worth studying for its eccentricity, or conventional and worth studying for how it fits certain patterns. This chapter has discussed some of the large methodological questions to emerge from studies of the Shakespearean text over the last fifty years, in particular. But what it does not – nor cannot – do is tell you exactly what questions you should ask of the texts you take up in your own studies. What matters is knowing to ask questions of those materials in the first place – not the ‘right’ questions that lead directly to answers but the hard questions that might yield an initial reply of: ‘I don’t know, but let’s find out.’ And that’s the rub. In terms of both the institutional and environmental climates in which we labour, we need to find sustainable ways forward if the kind of textual studies scholarship we have been practicing is going to endure and evolve. The work that has gained traction in the overlapping subfields of textual studies over the last ten years – work that is attentive to copy-specific detail and strives for broader contextualization – takes an enormous amount of time, costs a significant amount of money and often requires significant travel. Indeed, with the limitations imposed by the contraction of the profession and the rise of contingent academic labour in the American and British academies, the resources we require to do such slow, detail-oriented studies – subscription databases, costly research trips, prohibitively expensive imaging and permissions fees (even as some institutions move away from

SHAKESPEARE AND ‘TEXTUAL STUDIES’

43

them), and (perhaps most importantly) time – are already out of reach for most. Textual studies is therefore fast becoming, if not so already, a bastion of the most privileged in academia – those whose institutions fund research trips; those who can take the time for a long-term fellowship; those who can apply for publication subventions; those whose university libraries subscribe to essential digital resources that are still behind paywalls. Furthermore, other essential branches of textual studies, including editorial work of all types, but most crucially critical editing, are rarely recognized within academia’s hiring, tenure and promotion structures. And despite the privileging of individual scholarship by those same institutional frameworks, there is no denying that some of the best work in textual studies is deeply collaborative, explicitly or implicitly. What this means is that either the potential of collaboration goes untapped because we are dissuaded from working together, or our non-faculty collaborators (very often, librarians, graduate research assistants and others who sustain this work) go formally under- or unrecognized. Even if we are granted the resources of time and money to do and share the work, the threat of climate catastrophe looms large, and many scholars have chosen to forego air travel and do their work remotely. At the time of writing this essay, the global coronavirus pandemic means that any scholarship we do will have to be done remotely until local stay-at-home orders are lifted. All of this is bound to change the landscape of textual studies in the long term, as will the fact that one of the major libraries for Shakespeare studies – the Folger Shakespeare Library – is now closed to researchers for at least two years. One place where we might find new ways into this work is not so far away: the classroom – and not just in the classrooms of elite or big research universities that have robust special collections libraries or offer courses that allow faculty whose research agendas fall under the aegis of ‘textual studies, etc.’ to teach in the field. Teaching the methods required to study texts as material objects – and in their constantly evolving forms – requires us to unlearn what we know, or what we think we know, about those objects; the processes by which they were made, transmitted and circulated; and the frames through which we have previously interpreted them. In other words, we not only have to unlearn what we know about the object but also how we know it in the first place. The experiential brand of pedagogy that has caught on in Shakespeare courses over the last decade – folding and stitching quartos; using quills and homemade ink to write by hand; making paper; setting type; asking students to keep their own commonplace books; generating and acting from cue scripts; editing scenes or whole plays from scratch; and so forth – replicates the doing-as-learning attitude that underpins responsible and innovative scholarly practice. Cait Coker and Kate Ozment, two dynamic young scholars who run the Women in Book History Bibliography website and ‘Sammelband’, a linked blog on ‘how to do “book history on a budget”’, put it this way in their description of the blog’s mission: ‘We do not have all the answers, but we do have some concrete suggestions and a spirit of relentless creativity’ (2018). As scholars of Shakespeare and the material text, we rarely have all the answers, or even satisfactory ones to begin with. But through trial and error, through exposure to materials, through collaboration, through a solid sense of what has come before,

44

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

and through an eagerness to understand new forms of mediation and access, we know both to ask questions and also how to ask them. We, like our students, are in a constant state of learning something for the first time.

NOTES I wish to thank Megan Heffernan, Tara L. Lyons and Whitney Trettien for their help throughout the various stages of writing this chapter. 1. These include Jason Scott-Warren’s identification of John Milton as the annotator of a First Folio (Schuessler and Hurdle 2019); and Heather Wolfe’s discovery of new depictions of Shakespeare’s coat of arms (Schuessler 2016). 2. Robert Darnton made a similar claim about ‘book history’, calling it ‘interdisciplinarity run riot’ (1982: 67). 3. The bibliography has since been moved online (Craig and Estill 2020). 4. Unless otherwise noted, information about the ESTC that follows derives from the account in Gadd (2009).

REFERENCES Abbott, Craig S. and William Proctor Williams (1985), An Introduction to Bibliographical and Textual Studies, New York: Modern Language Association. Atkin, Tamara (2018), Reading Drama in Tudor England, Abington: Routledge. Bartlett, Henrietta C. and A. W. Pollard, eds (1916), A Census of Shakespeare’s Plays in Quarto 1594–1709, New Haven, CT: Yale University Press. Bartlett, Henrietta C., ed. (1939), A Census of Shakespeare’s Plays in Quarto 1594–1709, 2nd edn, New Haven, CT: Yale University Press. Berger, Thomas L. and Sonia Massai, with Tania Demetriou, eds (2014), Paratexts in English Printed Drama to 1642, 2 vols, Cambridge: Cambridge University Press. Blayney, Peter W. M. (1997), ‘The Publication of Playbooks’, in John D. Cox and David Scott Kastan (eds), A New History of Early English Drama, 383–422, New York: Columbia University Press. Bourne, Claire M. L. (2017), ‘Marking Shakespeare’, Shakespeare, 13 (4): 367–86. Bourne, Claire M. L. (2018), ‘Vide Supplementum: Early Modern Collation as PlayReading in the First Folio’, in Katherine Acheson (ed.), Early Modern English Marginalia, 195–233, Abingdon: Routledge. Bourne, Claire M. L. (2019), ‘Online Resources: Early Modern Plays on Stage & Page’, Of Pilcrows. Available online: https://www.ofpilcrows.com/resources-early-modernplays-page-and-stage (accessed 13 November 2020). Bourne, Claire M. L. (2020), Typographies of Performance in Early Modern England, Oxford: Oxford University Press. Brown, Meaghan J. (2018), ‘What is Lost is Found Again: The Lost Plays Database’, ‘The Collation’, Folger Shakespeare Library. Available online: https://collation.folger. edu/2018/07/lost-plays-database/ (accessed 13 November 2020).

SHAKESPEARE AND ‘TEXTUAL STUDIES’

45

Catalogue of the Library of Robert Hoe of New York … Part 1 (1911–12), New York: D. Taylor. Coker, Cait and Kate Ozment (2018), ‘Introducing Sammelband: A Book History Pedagogy Blog’, Sammelband. Available online: http://www.womensbookhistory.org/ sammelband/2017/12/9/teaching-book-history (accessed 13 November 2020). Craig, Heidi and Laura Estill, eds (2020), World Shakespeare Bibliography, Washington, DC: Folger Shakespeare Library. Available online: https://www.worldshakesbib.org/ (accessed 13 November 2020). Dailey, Alice (2020), ‘A Book is a Unit of Time: Library Deselection and the Work of Mourning’, The Hare 4 (3). Available online: https://thehareonline.com/article/bookunit-time-library-deselection-and-work-mourning (accessed 13 November 2020). Darnton, Robert (1982), ‘What is the History of Books?’, Daedalus, 111 (3): 65–83. de Grazia, Margreta and Peter Stallybrass (1993), ‘The Materiality of the Shakespearean Text’, Shakespeare Quarterly, 44 (3): 255–83. Depledge, Emma and Peter Kirwan, eds (2017), Canonising Shakespeare: Stationers and the Book Trade, 1640–1740, Cambridge: Cambridge University Press. Digital Image Collections: Rare Books (2020), Harry Ransom Center. Available online: https://hrc.contentdm.oclc.org/digital/collection/p15878coll17 (accessed 13 November 2020). Early English Books Online (EEBO) (2020), ProQuest. Available online: https://search. proquest.com/eebo (accessed 13 November 2020). Early English Books Online-Text Creation Partnership (EEBO-TCP) (2019), University of Michigan Library. Available online: https://quod.lib.umich.edu/e/eebogroup/ (accessed 13 November 2020). Early English Playbooks, 1594–1799 (2017), Boston Public Library. Available online: https://archive.org/details/bplsceep?&sort=-publicdate&page=2 (accessed 13 November 2020). Early Modern English Books (2020), Fondation Martin Bodmer. Available online: https://bodmerlab.unige.ch/constellations/early-modern-english-books (accessed 13 November 2020). Egan, Gabriel, Santiago Segarra, Mark Eisen and Alejandro Ribeiro (2016), ‘Attributing the Authorship of the Henry VI Plays by Word Adjacency’, Shakespeare Quarterly, 67 (2): 232–56. English Short Title Catalogue (n.d.), British Library Board. Available online: http://estc. bl.uk (accessed 13 November 2020). Erne, Lukas (2013), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Estill, Laura (2015), Dramatic Extracts in Seventeenth-Century English Manuscripts: Watching, Reading, Changing Plays, Lanham, MD: University of Delaware Press. Estill, Laura and Beatrice Montedoro, eds (2015), DEx: A Database of Dramatic Extracts. Available online: https://dex.itercommunity.org/ (accessed 13 November 2020). Farmer, Alan B. and Zachary Lesser, eds (2007), DEEP: Database of Early English Playbooks. Available online: http://deep.sas.upenn.edu (accessed 13 November 2020). Gadd, Ian (2009), ‘The Use and Misuse of Early English Books Online’, Literature Compass, 6 (3): 680–92.

46

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Galey, Alan (2014), The Shakespearean Archive: Experiments in New Media from the Renaissance to Postmodernity, Cambridge: Cambridge University Press. Grafton, Anthony and Lisa Jardine (1990), ‘“Studied for Action”: How Gabriel Harvey Read his Livy’, Past & Present, 129: 30–78. Graham-White, Anthony (1995), Punctuation and Its Dramatic Value, Newark, DE: University of Delaware Press. Greg, W. W. (1931), Dramatic Documents from the Elizabethan Playhouses, 2 vols, Oxford: Clarendon Press. Greg, W. W. (1932), ‘Bibliography – An apologia’, The Library, 4th ser., 13 (2): 113–43. Greg, W. W. (1939–59), A Bibliography of the English Printed Drama to the Restoration, 4 vols, London: Bibliographical Society. Grieve, Thomas F. and Rudolph E. Habenicht (1975), ‘Shakespeare: An Annotated World Bibliography for 1974’, Shakespeare Quarterly, 26 (4): 325–9. Hooks, Adam G. (2008), ‘Booksellers’ Catalogues and the Classification of Printed Drama in Seventeenth-Century England’, Papers of the Bibliographical Society of America, 102 (4): 445–64. Hooks, Adam G. (2016), Selling Shakespeare: Biography, Bibliography, and the Book Trade, Cambridge: Cambridge University Press. Hooks, Adam G. and Zachary Lesser, eds. (2018), Shakespeare Census. Available online: http://www.shakespearecensus.org (accessed 13 November 2020). Ioppolo, Grace, ed. (2020), Henslowe-Alleyn Digitisation Project. Available online: https://henslowe-alleyn.org.uk/ (accessed 13 November 2020). Kastan, David Scott (1999), Shakespeare After Theory, New York: Routledge. Kichuk, Diana (2007), ‘Metamorphosis: Remediation in Early English Books Online (EEBO)’, Literary and Linguistic Computing, 22 (3): 291–303. Kidnie, Margaret Jane (2005), ‘Where is Hamlet? Text, Performance, and Adaptation’, in Barbara Hodgdon and W. B. Worthen (eds), A Companion to Shakespeare and Performance, 101–20, Oxford: Blackwell. Kirkman, Francis (1661), ‘A True, perfect, and exact Catalogue of all the Comedies, Tragedies, Tragi-Comedies, Pastorals, Masques and Interludes, that were ever yet printed and published, till this present year’, in Tom Tyler and his Wife, London: Kirkman. Knight, Jeffrey Todd (2010), ‘Invisible Ink: A Note on Ghost Images in Early Printed Books’, Textual Cultures, 5 (2): 53–62. Knight, Jeffrey Todd (2013), Bound to Read: Compilations, Collections, and the Making of Renaissance Literature, Philadelphia: University of Pennsylvania Press. Knight, Jeffrey Todd (2017), ‘Economies of Scale: Shakespeare and Book History’, Literature Compass, 14 (6): n.p. Knutson, Roslyn L., David McInnis and Matthew Steggle (2020), Lost Plays Database, Folger Shakespeare Library. Available online: https://lostplays.folger.edu/ (accessed 13 November 2020). Lee, Sidney, ed. (1902), A Census of Extant Copies with Some Account of their History and Condition, Oxford: Clarendon Press. Lesser, Zachary (2004), Renaissance Drama and the Politics of Publication, Cambridge: Cambridge University Press.

SHAKESPEARE AND ‘TEXTUAL STUDIES’

47

Lesser, Zachary (2015), review of Thomas L. Berger and Sonia Massai, with Tania Demetriou (eds), Paratexts in English Printed Drama to 1642 (2014), Cambridge: Cambridge University Press, Shakespeare Quarterly, 66 (4): 467–70. Lesser, Zachary (2019), ‘Xeroxing the Renaissance: The Material Text of Early Modern Studies’, Shakespeare Quarterly, 70 (1): 3–31. Lesser, Zachary (2020a), ‘The Material Text between General and Particular, Edition and Copy’, English Literary Renaissance, 50 (1): 83–92. Lesser, Zachary (2020b), personal correspondence with the author, 27 February. Lesser, Zachary (forthcoming), Ghosts, Holes, Rips and Scrapes: Shakespeare in 1619, Bibliography in the Longue Durée, Philadelphia: University of Pennsylvania Press. Lindenbaum, Sarah (2018), ‘Written in the Margent: Frances Wolfreston Revealed’, ‘The Collation’, Folger Shakespeare Library. Available online: https://collation.folger. edu/2018/06/frances-wolfreston-revealed/ (accessed 13 November 2020). Lindenbaum, Sarah (2019), personal correspondence with the author, 2–3 April. Lindenbaum, Sarah, Jay Moschella and Lori Humphrey Newcomb (2015), ‘Frances Wolfreston and “Hor” Playbooks at the BPL’, Boston Public Library: Collections of Distinction, 21 December. Literary Print Culture: The Stationers’ Company Archive, 1554–2007 (2020), Adam Matthew Digital. Available online: www.literaryprintculture.amdigital.co.uk (accessed 13 November 2020). Lopez, Jeremy (2014), Constructing the Canon of Early Modern Drama, Cambridge: Cambridge University Press. LUNA: Folger Digital Image Collection (2020), Folger Shakespeare Library. Available online: https://luna.folger.edu/ (accessed 13 November 2020). Lyons, Tara L. (2017), ‘Richard Jones, Tamburlaine the Great, and the Making (and Remaking) of a Serial Play Collection in the 1590s’, in Roslyn L. Knutson and Kirk Melnikoff (eds), Christopher Marlowe, Theatrical Commerce, and the Book Trade, 149–64, Cambridge: Cambridge University Press. Lyons, Tara L. (2020), personal correspondence with the author, 3 March. MacLean, Sally-Beth and Lawrence Manley (2014), Lord Strange’s Men and their Plays, New Haven, CT: Yale University Press. Maguire, Laurie (2005), ‘How Many Children Had Alice Walker?’, in Douglas Brooks (ed.), Printing and Parenting in Early Modern England, 327–50, Aldershot: Ashgate. Mak, Bonnie (2014), ‘Archaeology of a Digitization’, Journal of the Association for Information Science and Technology, 65 (8): 1515–26. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Masten, Jeffrey (2016), Queer Philologies: Sex, Language, and Affect in Shakespeare’s Time, Philadelphia: University of Pennsylvania Press. Mayer, Jean-Christophe (2018), Shakespeare’s Early Readers: A Cultural History from 1590 to 1800, Cambridge: Cambridge University Press. McGann, Jerome (1991), The Textual Condition, Princeton: Princeton University Press. McInnis, David and Matthew Steggle (2015), ‘Nothing Will Come of Nothing? What can we Learn from Plays that Don’t Exist?’, in David McInnis and Matthew Steggle (eds), Lost Plays in Shakespeare’s England, 1–14, Basingstoke: Palgrave Macmillan.

48

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

McKenzie, D. F. (1981), ‘Typography and Meaning: The Case of William Congreve’, in Giles Barber and Bernhard Fabian (eds), Buch und Buchhandel in Europa im achtzehnten Jahrhundert, 81–125, Wolfenbütteler Schriften zur Geschichte des Buchwesens 4, Hamburg: Hauswedell. McManaway, James G. and Jeanne Addison Roberts (1975), A Selective Bibliography of Shakespeare: Editions, Textual Studies, Commentary, Washington, DC: The Folger Shakespeare Library. Melnikoff, Kirk (2005), ‘Jones’s Pen and Marlowe’s Socks: Richard Jones, Print Culture, and the Beginnings of English Dramatic Literature’, Studies in Philology, 102 (2): 184–209. Meserole, Harrison T. (1976), ‘Shakespeare: Annotated World Bibliography for 1975’, Shakespeare Quarterly, 27 (4): 389–96. Meserole, Harrison T. and John B. Smith (1979), ‘Shakespeare: Annotated World Bibliography for 1978 – Textual Studies, Etc.’, Shakespeare Quarterly, 30 (4): 480. Morgan, Paul (1989), ‘Frances Wolfreston and “Hor Bouks”’, The Library, 6th ser., 11 (3): 197–219. Pantzer, Katherine, F. S. Ferguson and W. A. Jackson, eds (1976–91), A Short-Title Catalogue of Books Printed in England, Scotland, & Ireland and of English Books Printed Abroad, 1475–1640, 2nd edn, London: Bibliographical Society. Park, Stephen (1986), The Elizabethan Club of Yale University and Its Library, New Haven, CT: Yale University Press. Pollard, A. W. (1917), Shakespeare’s Fight with the Pirates and the Problem of the Transmission of His Text, London: Alexander Moring Limited. Pollard, A. W. and G. R. Redgrave, eds (1926), A Short-Title Catalogue of Books Printed in England, Scotland, & Ireland and of English Books Printed Abroad, 1475–1640, London: Bibliographical Society. Pratt, Aaron T. (2015), ‘Stab-Stitching and the Status of Early English Playbooks as Literature’, The Library, 7th ser., 16 (3): 304–28. Pratt, Aaron T. [@aarontpratt] (2014), ‘a copy of a copy of an ed., understanding that reproduction can intro. error’, Twitter, 8 May. Rasmussen, Eric and Anthony James West, eds (2012), The Shakespeare First Folio: A Descriptive Catalogue, Basingstoke: Palgrave Macmillan. REED: Records of Early English Drama (2020), Available online: https://ereed.library. utoronto.ca/ (accessed 13 November 2020). ‘Renaissance Books, Midwestern Libraries’ (2014), Folgerpedia, Folger Shakespeare Library. Available online: https://folgerpedia.folger.edu/Renaissance_Books,_ Midwestern_Libraries (accessed 13 November 2020). retroReveal (2020), Available online: http://retroreveal.org/ (accessed 13 November 2020). Schuessler, Jennifer (2016), ‘Shakespeare: Actor. Playwright. Social Climber’, The New York Times, 29 June. Available online: https://www.nytimes.com/2016/06/30/theater/ shakespeare-coat-of-arms.html (accessed 13 November 2020). Schuessler, Jennifer and Jon Hurdle (2019), ‘Milton’s Shakespeare Was Just a TransAtlantic Tweet Away’, The New York Times, 19 September. Available online: https:// www.nytimes.com/2019/09/19/theater/milton-shakespeare-notes-first-folio.html (accessed 13 November 2020).

SHAKESPEARE AND ‘TEXTUAL STUDIES’

49

Scott-Warren, Jason (2019), Shakespeare’s First Reader: The Paper Trails of Richard Stonley, Philadelphia: University of Pennsylvania Press. Shakespeare in Quarto (n.d.), The British Library Board. Available online: https://www. bl.uk/treasures/shakespeare/homepage.html (accessed 13 November 2020). Sherman, William H. (2002), ‘What Did Renaissance Readers Write in Their Books?’, in Jennifer Andersen and Elizabeth Sauer (eds), Books and Readers in Early Modern England, 116–34, Philadelphia: University of Pennsylvania Press. Sherman, William H. (2008), Used Books: Marking Readers in Renaissance England, Philadelphia: University of Pennsylvania Press. Steggle, Matthew (2015), Digital Humanities and the Lost Drama of Early Modern England, London: Routledge. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Stoddard, Roger (1985), Marks in Books, Illustrated and Explained, Boston, MA: Houghton Library of the Harvard College Library. Syme, Holger Schott (2011–12), ‘Well-Read Plays I–V’, dispositio. Available online: http://www.dispositio.net/archives/751 (accessed 13 November 2020). Trettien, Whitney (2019), ‘EEBO Oddities’, whitneyannetrettien.com. Available online: http://whitneyannetrettien.com/whiki/index.php?title=Scrapbook#EEBO_oddities (accessed 13 November 2020). Tribble, Evelyn B. (2011), Cognition in the Globe: Attention and Memory in Shakespeare’s Theatre, New York: Palgrave Macmillan. Wayne, Valerie, ed. (2020), Women’s Labour and the History of the Book in Early Modern England, London: Bloomsbury. Wells, Stanley and Gary Taylor, gen. eds (1986), The Oxford Shakespeare: The Complete Works, Oxford: Oxford University Press. Werner, Sarah (2012), ‘Where Material Book Culture Meets Digital Humanities’, Journal of Digital Humanities, 1 (3): n.p. Werner, Sarah (2019), ‘digitized first folios’, wynken de worde. Available online: https:// sarahwerner.net/blog/digitized-first-folios/ (accessed 13 November 2020). West, Anthony James, ed. (2001), The Shakespeare First Folio: The History of the Book; Volume 2: A New Worldwide Census of First Folios, Oxford: Oxford University Press. Wheeler, G. W. (1926), Letters of Thomas Bodley to Thomas James: First Keeper of the Bodleian Library, Oxford: Clarendon Press. Wing, Donald, ed. (1945–51), Short-Title Catalogue of Books Printed in England, Scotland, Ireland, Wales and British America and of English Books Printed in Other Countries, 1641–1700, 3 vols, New York: Columbia University Press. Ziegler, Georgianna (2020), ‘Patterns in Women’s Book Ownership, 1550–1700’, in Valerie Wayne (ed.), Women’s Labour and the History of the Book in Early Modern England, 207–23, London: Bloomsbury.

50

PART TWO

Current research and issues

52

CHAPTER 2.1

The Shakespeare manuscripts CATHY L. SHRANK AND PAUL WERSTINE

To some extent, the title of this chapter is a misnomer. There are no ‘Shakespeare manuscripts’: his canonical plays and poems come down to us only via print, surviving manuscripts of them having been either certainly or very probably transcribed and/ or adapted from printed versions. Nonetheless, many Shakespeareans have bid, generally without success, to associate some manuscripts of various kinds with the playwright/poet and even on occasion to invest some of these manuscripts with his authority. At the same time, scholars have sought inferentially to reconstruct the lost printer’s copy for Shakespeare’s plays by studying manuscripts of plays by his contemporaries. This chapter examines what the existence, or absence, of manuscript copies indicates about the circulation and reception of his works, and considers what these manuscript traces can tell us about Shakespeare as an author, and about how his contemporaries and succeeding generations read and valued his writing.

SHAKESPEARE’S PLAYS AND MANUSCRIPT Among the manuscripts clearly descending from print is the Dering manuscript (Folger MS V.b.34); it contains a single play of 3,401 lines that combines Henry IV, Parts 1 and 2, the first copied from its fifth quarto printing (1613), the second from its first (1600). More of the first play survives Sir Edward Dering’s abridgement than of the second, the comic scenes of the second being even more severely cut than those of the first. Therefore we see Dering’s preference for the first play and for the politics in both, rather than the humour. The first page written in his own hand (the rest in an anonymous scribe’s), the manuscript was likely created around 1623 for private performance at Dering’s country home in Kent (Williams and Evans 1974: vii–ix). He was an inveterate play-goer and kept a library containing 250 playbooks (Lennam 1965: 145). Another early manuscript, the Douai, is housed in the Douai Public Library, France (MS 7.87). It includes Twelfth Night, As You Like It, The Comedy of Errors, Romeo and Juliet, Julius Caesar and Macbeth – perhaps the most popular Shakespeare

54

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

plays for performance in schools in the late seventeenth century. These texts were copied directly or indirectly from the Second Folio (1632). The transcriptions are cut and marked up for production, probably at one of the Douai colleges established for English Roman Catholic students (Evans 1962). More curious is the single-page manuscript of parts of Titus Andronicus held by the Marquess of Bath at Longleat House, the Longleat manuscript. The page consists of three parts: 1) an ink drawing of two Roman soldiers, Titus (with wreathed head), Tamora (crowned and kneeling in front of Titus), Chiron and Demetrius (bound and kneeling behind her), and Aaron; 2) the stage direction ‘Enter Tamora pleadinge for her sonnes going to execution’, not in the play, where Tamora pleads only for Alarbus’s life; and 3) speeches from the play, among them two spurious lines, and a speech prefix for Alarbus, who has no lines in the play. The erroneous prefix and the specious stage direction and lines leave us to wonder the extent to which we can trust the drawing to represent performance. Nonetheless, the mixing of Elizabethan and Roman costumes may suggest stage practice in Shakespeare’s day. The leaf is dated 1594 (Berry 1999) and signed by Henricus [Henry] Peacham, but the text is scribal (Levin 2002: 324). Before Berry, the date had also been variously read, for example, as 1595 and as 1604, 1605, 1614 and 1615. The most famous manuscript associated with Shakespeare is Sir Thomas Moore (or More, British Library (BL) MS Harleian 7368). More has twenty leaves (not counting its wrapper), only thirteen in the handwriting of its transcriber, the dramatist Anthony Munday, who may or may not have authored what he copied. Most interesting are the six ‘Additions’ to Munday’s pages, all written in different hands than his, that represent a large-scale adaptation, now widely thought to have taken place shortly after the original’s transcription early in the seventeenth century. In 1871 Richard Simpson first proposed that some of the Additions are in Shakespeare’s hand. However, Simpson thought Shakespeare responsible for pages where most subsequent scholars have identified three different hands, only one of which (Hand D) has been claimed for Shakespeare (see Figure 2.1.1). According to W. W. Greg in his great 1911 edition of More, Simpson’s Shakespeare identification embraces all of Hand C (the book-keeper or prompter/scribe), Hand D and Hand E (Thomas Dekker). The other hands in the Additions are Henry Chettle’s (Tannenbaum 1927: 53) and possibly Thomas Heywood’s (Greg in Pollard 1923: 44n.). More also bears annotation and other marking by Edmund Tilney, Master of the Revels. Displeased by a street riot against foreigners called for early in the play, Tilney threatened the actors: ‘ ye insurwholy … att your own perrilles E Tyllney’ (fol. 3r). Simpson was not the first to examine the manuscript. Alexander Dyce had edited it in 1844 and then went on to edit Shakespeare’s complete works in three editions without recognizing any connection, if there is one, between it and Shakespeare. Continuing belief in a connection today can be traced to A. W. Pollard’s 1923 Shakespeare’s Hand in the Play of Sir Thomas More. Pollard (uncharacteristically) did not present his collection as a rational, objective, and therefore scholarly, assessment of evidence for and against Shakespeare’s hand in the manuscript play. Instead, Pollard injected a note of partisanship that has long persisted among advocates of

THE SHAKESPEARE MANUSCRIPTS

55

FIGURE 2.1.1  Hand D in The Booke of Sir Thomas Moore, BL Harley MS 7368, fol. 9r. Courtesy of the British Library.

56

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Shakespeare as Hand D when he announced that his ‘object’ was only ‘to strengthen the evidence of the existence of three pages written by Shakespeare in his own hand as part of … More’ (1923: v). He called this his ‘cause’ (1923: 32) in the political sense as ‘that side of a question or controversy which is espoused, advocated, and upheld by a person or party; a movement which calls forth the efforts of its supporters’ (OED). Among his supporters was Sir E. Maunde Thompson, who had long been committed to the belief that on the basis of just six Shakespeare signatures on legal documents, together with the two words ‘By me’ on the last page of Shakespeare’s will, he could identify Hand D as Shakespeare’s (1916; and Pollard 1923: 57–112). Thompson attended only to isolated coincidences in the formation of particular letters in the Hand-D pages and the signatures, rather than also weighing the differences. In 2016, revisiting Thompson’s case and its iterations, Michael L. Hays concluded ‘without a control [that is, a sample of Shakespeare’s handwriting of twenty or thirty lines], any effort to construct a paleographic argument for identification is inescapably an exercise in futility’ (202). In the most detailed review of the arguments in Pollard’s book (together with later versions of them) undertaken to date, James Purkis stated simply that ‘Paleographic study … cannot establish whether or not folios 8 and 9 are in Shakespeare’s handwriting’ (2016: 208). Another supporter was John Dover Wilson, who found in the Shakespeare quartos five kinds of misprinting that he attributed to problems experienced by compositors with Shakespeare’s handwriting; Wilson located the same problems with legibility in the Hand-D pages and argued from that coincidence to Shakespeare’s inscription of them. An example is the misprinting of e for o (‘euer’ for ‘ouer’) and o for e (‘ouer’ for ‘euer’) in the quartos and the occasional failure of Hand D to distinguish an o from e or an e from an o (Pollard 1923: 119). However, as Purkis discovered, the same five misprints are in the early printed versions of John Marston’s The Fawn (1606) and The Malcontent (1604) and of Philip Massinger’s The Duke of Milan (1623), even though Massinger wrote a mixed hand (Italian and English) that looks nothing like Hand D (2016: 210). Wilson also purported to find nine solid pages of ‘unusual’, ‘old-fashioned’ and/or ‘rare’ spellings shared by the Shakespeare quartos and the Hand D pages (Pollard 1923: 133–41, 128–9). Of all these spellings, there are now no more than five that are still regarded as sufficiently rare to be worthy of consideration (Werstine 2007: 122). One, ‘scilens’ for silence, long thought to be unique to the first quarto of Shakespeare’s 2 Henry IV (1600) and to the Hand D pages and therefore of great value in identifying Hand D as Shakespeare’s, has recently been found in the 1546 A boke of prayers called ye ordynary faschyon of good lyuynge and a 1568 bible (Purkis 2016: 241). As Purkis summed up, ‘arguments from common spellings alone cannot establish Shakespeare’s authorship of the Hand-D addition and there are insufficient data to even be able to judge how “likely” the attribution may be’ (217–18). The third member of Pollard’s ‘little company’, R. W. Chambers, took up the ‘cause’ to argue that he had discovered sequences of ideas and images in the Hand-D pages that matched sequences in Shakespeare’s canonical plays (Pollard 1923: 31, 142–87).1 However, in the course of his argument, Chambers was obliged

THE SHAKESPEARE MANUSCRIPTS

57

to strip and crush the plays’ figurative language to evoke from it the equivalences necessary to his case. When in the Hand-D pages More imagines the street rioters and ‘other ruffians’ as ‘ravenous fishes / [that] would feed on one another’ (6.95, 97–8), Chambers located an equivalent passage in Troilus and Cressida in Ulysses’ representation of appetite as ‘a universal wolf, / So doubly seconded with will and power, / [that it] Must make perforce an universal prey / And last eat up himself’ (1.3.121–4). ‘Ravenous fishes and universal wolves eating up themselves’, said Chambers (Pollard 1923: 158–9), but in More the fish feed only on each other, not on themselves, and become like Ulysses’ wolf only when Chambers intervenes. As Purkis observed of Chambers’s method, ‘the resemblances that he presents are no more certainly the products of common authorship than they are of his own ingenuity and desire to find and attribute significance to coincidences between Shakespearean and the More passages’ (2016: 227). In 2006 MacDonald P. Jackson attempted to rescue Pollard’s attribution of the Hand-D pages to Shakespeare through digital scholarship, searching the database Literature Online (LION) for matches to the language of the pages. He found more rare words from the pages to be shared by Shakespeare’s plays than by the plays of any other dramatist whose plays were performed between 1590 and 1610. Exhibiting Pollard-like partisanship, Jackson claimed ‘we can infer with confidence’ that the Hand-D pages are Shakespeare’s ‘autograph draft’ (2006: 78). As Purkis has noted, Jackson’s method cannot support this conclusion (2016: 231–4). Most of the drama from Shakespeare’s time is lost. From ‘1567 to 1642, around 3,000 different plays must have been written and staged’; 543 survive (McInnis and Steggle 2014: 1). The rate of preservation of Shakespeare’s plays greatly exceeds that of all other dramatists. Thus, as Purkis shows, Jackson’s method identifies as Shakespeare’s a passage that we already know is in the handwriting of another dramatist (2016: 233–4). It is also impossible to tell whether Hand D is composing or copying and adapting lines authored by someone else; a number of Hand-D’s alterations in the course of writing look more like corrections of scribal eye-skips than revisions made during composition (Van Dam 1924: 369–71; Schücking 1925: 59; Ioppolo 2006: 104–8; Downs 2007). The identity of the penman who wrote out the Hand-D pages of More remains a mystery. The manuscripts of greatest interest to editors of Shakespeare are those we have lost forever – the ones that served as printer’s copy for his published plays. Early in the twentieth century Pollard undertook to discover the nature of such printer’s copy by studying surviving play manuscripts. There are over 100 such pre-dating 1642 (for a list see Ioppolo 2006: 5–7). Yet because we think that generally the source of copy for play publishers was the acting companies (which owned the plays and whose names usually appear on the title pages), rather than the playwrights, Pollard and his successors have focussed on manuscripts that bear marks of playhouse use. Such marks are annotations, usually in hands different from the one in which the text is copied, that attend to matters of performance: e.g. actors’ names attached to roles; intervals between acts, when music was provided; cuts of passages and roles (as the play was shaped to the time of playing and to the available cast); properties or props; sounds (e.g. music, thunder); full or partial repetition or addition of stage

58

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

directions usually in the left margins of manuscripts; warnings for actors or props to be made available for later entrance or use; censorship (although the playhouse was not the only source of censorship, especially in the Jacobean and Caroline periods) (Werstine 2012: 234–357). Pollard studied just a few playhouse manuscripts available to him at the British Library, where he worked; these included More, Massinger’s Beleeue as You List (1631), Thomas Heywood’s The Captives (1624) and Walter Mountfort’s The Lanchinge of the Mary (1633) (1920: 57–8). All these manuscripts are in the hands of their authors, and from such observation Pollard inferred that it was possible (contrary to widespread previous belief) that plays, including some of Shakespeare’s, could have been printed from authorial manuscripts. He also broke new ground by searching early printed Shakespeare texts for the kinds of annotations (listed in the paragraph above) that are peculiar to playhouse manuscripts. From his identification of such notes in printed texts, he judged that they had been printed from playhouse manuscripts, although he could not conclude that such manuscripts were necessarily authorial because the playhouse used scribal as well as authorial manuscripts. For example, he properly interpreted the property note in the Folio 2 Henry VI ‘Bed put forth’ as probably originating in the playhouse and thereby indicating that printer’s copy was likely a theatrical manuscript (Pollard 1920: 65). Since Pollard, study of playhouse manuscripts has proceeded apace. By 1931 in his Dramatic Documents from the Elizabethan Playhouses Greg had gathered reference to some fifteen playhouse manuscripts (1.239–308). (For Greg Elizabethan refers to the period up to 1642.) Now we know of eighteen playhouse manuscripts, as well as of three printed quartos containing playhouse annotation. (For discussion of these twenty-one texts, see Werstine 2012: 245–357.) We can be confident that five quartos were probably set into type from playhouse manuscripts, rather than from authorial manuscripts left untouched by the players – Hamlet (1604/5), Romeo and Juliet (1599), 2 Henry IV (1600), Much Ado About Nothing (1600) and The Two Noble Kinsmen (1634). Each contains stage directions or speech prefixes of the kind that are found in the extant playhouse manuscripts only in the hands of playhouse personnel, never in the hands of the plays’ authors. In the case of Hamlet the evidence is duplication of a stage direction; for Romeo, 2 Henry IV and Much Ado use of actors’ names (rather than characters’) in stage directions or speech prefixes, and for Kinsmen the appearance of such unmistakable book-keeper’s warnings as ‘2. Hearses ready with Palamon: and Arcite: the 3. Queenes. Theseus: and his Lordes ready’ (sig. C3v). Among Folio plays likely printed from playhouse manuscripts are All’s Well That Ends Well (in which roles are combined to reduce cast); Henry VI, Part 2 (as Pollard knew) and Part 3 (actors’ names in stage directions and a speech prefix); Julius Caesar (in which a role is partially cut); and Macbeth (apparently adapted for performance by Thomas Middleton). In addition, a number of Folio plays appear to have been printed either from quartos annotated for playhouse use or from quartos annotated with reference to playhouse manuscripts (Much Ado, A Midsummer Night’s Dream, The Merchant of Venice, Troilus and Cressida and Titus Andronicus). Most of the quartos (like many Folio texts) provide no indication of a book-keeper’s attention and thus leave us to wonder if their printer’s copy

THE SHAKESPEARE MANUSCRIPTS

59

is theatrical or not. The difficulty of identifying printer’s copy for a particular text as a playhouse manuscript ranks with the difficulty of excluding a playhouse manuscript as possible copy for such a text. After all, some playhouse manuscripts contain so little mark-up that it is entirely possible that manuscripts with no mark-up whatsoever could have been used to guide production, these manuscripts betraying nothing of their origins if and when they were put into print, the question of their printer’s copy an irresolvable mystery. Contrary to John Jowett (2017: 122), no one insists that all of the Shakespeare quartos were printed from playhouse manuscripts. R. B. McKerrow (1935) and Greg (1954; 1955), who proved enormously influential and continue to be so even today, attempted to find a way around these limitations to our knowledge about printer’s copy. They purported to identify differences between authorial manuscripts (whether annotated in the theatre or not) and theatrical manuscripts. McKerrow, the innovator, argued that the mark of an authorial manuscript was variation in the naming of characters, as, for example, in the first quarto of Love’s Labour’s Lost (1598) when Armado sometimes enters under his proper name, which also serves, abbreviated, as his speech prefix, and sometimes enters and speaks as Braggart (1935: 463). According to McKerrow – who constructed this argument by examining only early printed Shakespeare texts and no playhouse manuscripts – the book-keeper, who would mark up or sometimes himself transcribe the playhouse manuscript, could not have tolerated such inconsistency in naming and would have needed character designations regularized. So variation in naming became the mark of the authorial manuscript (McKerrow 1935: 464–5). Greg generalized McKerrow’s case to include all kinds of other irregularities in printed Shakespeare texts, among them ambiguous and erroneous naming; missing, indefinite, deficient, erroneous, misplaced or doubled stage directions; and incomplete deletions of dialogue (1955: 110–15, 121–36). From his study of theatrical manuscripts, Greg knew better than McKerrow about the presence of all these features in them. When Greg wrote in general terms, he was therefore careful to state, ‘owing to the casual ways of book-keepers these characteristics may persist, to some extent at least, in the prompt-book’ (1955: 142). However, when he delivered his judgments about printer’s copy for particular plays, he ignored these ‘casual ways’ and, contradicting himself, determined that irregularities must mean that printer’s copy was authorial (1955: 176–425). For the last seventy or so years, most but not all Shakespeare editors simply reproduced Greg’s mode of arriving at judgments about printer’s copy. Playhouse personnel who marked up manuscripts or quartos or transcribed the manuscripts almost never regularized names or attended to defects and problems in stage directions. Every kind of inconsistency that McKerrow and Greg thought distinctive of an author is to be found, usually in abundance, in playhouse manuscripts (Werstine 2012: 358–91). And, instead of tidying up, book-keepers added to what McKerrow and Greg mistakenly took to be the uniquely authorial disorder of manuscripts (Werstine 2012: 148–99). In terms of irregularity and inconsistency, there is nothing to distinguish an authorial playhouse manuscript from one copied by a theatrical scribe or another copied by a non-theatrical scribe. It is quite impossible in attempting to assess any particular inconsistency to tell if its source is an author,

60

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

a scribe or a playhouse book-keeper. When John Jowett, following McKerrow and Greg, infers that such inconsistencies in Shakespeare quartos are ‘relatively rich evidence indicating … the printed text’s proximity to Shakespeare’s hand’ (2017: 122), he pays attention only to the More manuscript and ignores the wealth of other evidence in the playhouse manuscripts that show such inconsistencies perpetuated and multiplied well beyond the author. Such perpetuation also instructively continues in the print tradition of Shakespeare’s plays, so that we find, for example, in the 1670s Smock Alley, Dublin, playhouse text of A Midsummer Night’s Dream (an annotated copy of the Third Folio version) that Hippolyta is still inconsistently called in speech prefixes both by her proper name and by her title Dutchess just as she is in the first quarto of 1600 (Cloud 1997: 190–1). Greg taught that generally there would be only one playhouse manuscript of any play, which he anachronistically called the ‘prompt-book’, and that it would have been censored, licensed and signed by the Master of the Revels, as well as thoroughly annotated for performance and fully regularized (1954: 33). No acting company, he argued, would want to give up such a uniquely valuable manuscript to a publisher because it might well be destroyed in the printing house. Hence the expectation that only in special circumstances, like the wholesale revision of a play such as perhaps 1 Henry IV to remove Oldcastle and replace him with Falstaff, would a play be printed from such a ‘prompt-book’ and only from a disused one (Jowett 2017: 121). However, no extant theatrical manuscript entirely lives up to Greg’s expectations, which derive from the latter period in which the term promptbook came into use. Instead, among the extant eighteen playhouse manuscripts, no fewer than eleven are in a state from which it must be inferred that the companies owning them also owned a second playhouse manuscript of the same play. Most of these eleven bear no signs of censorship or licensing; yet there must once have been licensed copies of these or they would not have been readied for production, as they so clearly are in the extant manuscripts (Werstine 2020: 50). Thus there is no need to imagine special circumstances, such as those suggested for 1 Henry IV, in order to think that a company might have one of its plays put into print from a playhouse manuscript when there was often more than one of these to hand. Two scribes whose names we know, Ralph Crane and Edward Knight, probably (Crane) or possibly (Knight) made contributions to the state of Shakespeare’s printed texts. Crane flourished between 1589 and 1632, but his extant transcriptions of drama, seven in total, date only from 1618 to 1625 (Werstine 2015: 27). Most he made for the dramatist Thomas Middleton, whose 1624 Game at Chesse he copied many times, three of his copies surviving (Howard-Hill 1995: 153–91; Taylor and Lavagnino 2007: 712–848). Among the striking features of his work is occasional use of massed entrances (with which he brings on at the beginning of a scene all the characters who appear in it, even though some evidently do not enter until later). Crane makes lavish use of italics, capitals and all kinds of punctuation, including parentheses, apostrophes and hyphens. In 1974, in one of the first digitally assisted works of scholarship devoted to Renaissance drama, T. H. Howard-Hill confirmed that Crane transcripts probably lie as printer’s copy behind the Shakespeare First Folio versions of The Tempest, The Two Gentlemen of Verona, The Merry Wives of

THE SHAKESPEARE MANUSCRIPTS

61

Windsor, Measure for Measure and The Winter’s Tale. Others have established that Cymbeline too was probably set into type from a Crane copy (Maxwell 1960: 126; Wells and Taylor 1987: 604; Butler 2005: 259). Crane is suspected of intruding more strongly on the stage directions of plays in copying them than on the dialogue and therefore of elaborating these in The Tempest such as ‘Enter seuerall strange shapes, bringing in a Banket; and dance about it with gentle actions of salutations, and inuiting the King, &c. to eate, they depart’ (Jowett 1983: 107). According to Howard-Hill, ‘the general level of Crane’s accuracy was high, but he was not reluctant to interfere with his text, consciously or unconsciously, when its meaning was obscure to him’ (1974: 133). Edward Knight, unlike Crane, belonged to an acting company, the King’s Men, which he probably joined sometime between 1619 and 1624 and which he served as a book-keeper or prompter. His last recorded association with the company dates from October 1633. Two of his transcripts survive: one, The Honest Mans Fortune (Victoria and Albert Museum, MS Dyce 25. F. 9), dating from 8 February 1625 (‘1624’ according to the old legal calendar) and the other the undated Bonduca (BL MS Egerton 2828). As book-keeper he also annotated copiously Massinger’s transcript of his own 1631 Beleeue as You List and made a few notes in John Clavell’s The Sodderd Cittizen (Wiltshire and Swindon Record Office MS 865/502/2) conjecturally dated 1628–30 (Werstine 2015: 36–7). Unlike Crane, Knight had no distinctive habits of transcription beyond his insertion of colons between character names in stage directions, a habit he shared with Crane. Thus it is hard to identify his transcriptions as printer’s copy for any of Shakespeare’s plays. One possibility, however, is the 1634 quarto of Shakespeare and John Fletcher’s The Two Noble  Kinsmen, which, as already has been noticed, contains several unmistakable book-keeper’s notes, together with colons between names in stage directions (Hoy 1962: 75). Knight was capable of achieving a fair standard of accuracy in transcription if he felt free to mar the appearance of his work, as he did in The Honest Mans Fortune, scarred with many corrections, or he was capable of fashioning a beautiful copy as long as he left uncorrected a quantity of errors (Bonduca) (Werstine 2012: 72–83). So as to preserve the beauty of his manuscript he was not above relocating lines and passages that he had omitted from Bonduca, rather than crossing out his mistakes (83–9). Together Knight and Crane exhibit the range of difference embraced in the word scribe when that agent is evoked by scholars of the transmission of Shakespeare’s plays into print.

SHAKESPEARE’S NON-DRAMATIC POETRY IN MANUSCRIPT Arthur Marotti and Laura Estill have suggested that ‘given Shakespeare’s growing literary and cultural importance, it is … surprising how few times Shakespearean texts seem to have been recorded in sixteenth- and seventeenth-century manuscripts’ (2012: 54). The ‘inescapable conclusion’ they reach is that ‘as far as the first halfcentury following his death is concerned … his poetry, particularly his lyrics, did not have a strong presence in the manuscript literary culture of the time’ (62). Yet,

62

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

when we turn to the manuscript record, it is actually Shakespeare’s sonnets – his lyrics (as opposed to his narrative poems) – that dominate.2 Of the thirty-seven extant manuscripts containing extracts from Shakespeare’s non-dramatic poetry, his sonnets (printed 1609) appear in twenty-one of these, excerpts from Venus and Adonis (1593) in twelve, Lucrece (1594) in six, A Lover’s Complaint (1609) in one.3 Sonnet 2 proved particularly popular: thirteen known copies survive, considerably more than any of the sonnets by Shakespeare’s contemporaries; Edmund Spenser’s Amoretti 8, first  printed in 1590, comes a flagging second, appearing in six manuscripts (Beal 2005–13; Shrank and Lyne 2018: 718). A further ten poems posthumously ascribed to Shakespeare in the seventeenth century – but never printed under his name in his lifetime – appear across seventeen manuscripts (although not necessarily attributed in every copy).4 The brevity and often conventional nature of these verses, along with the belatedness of their ascription, make it difficult, if not impossible, to ascertain their authorship, although their attribution to Shakespeare serves as further testimony of his cultural cachet. William Ringler and Steven May (1972) added a further item to this posthumous canon: a 1598 epilogue (incipit ‘As the diall hand tells ore’) delivered ‘to the Q[ueen] by the players’ found in Cambridge University Library (CUL) MS Dd.5.75 (fol. 46r); the attribution has been accepted by scholars including Juliet Dusinberre (2006: 37–42), but disputed by others (e.g. Hattaway 2009; Hackett 2012). That more of Shakespeare’s poems do not appear in greater numbers of manuscripts may owe something to rates of survival, and perhaps of production. Far more manuscript verse miscellanies from the 1630s to the 1650s exist than  from earlier decades (Roberts 2003: 179). Shakespeare’s manuscript traces are thus shaped by the tastes of later Jacobean and Caroline readers. He might be less present than Ben Jonson, who – living until 1637 – was part of, writing for and influencing that later literary scene, but Shakespeare’s works are certainly more represented than those of his contemporary Samuel Daniel (1563–1619), whose poetry is found in twenty-four manuscripts, almost all of which represent the only known manuscript copies of those particular poems.5 Further to that, the manuscripts of Shakespeare’s poetry illustrate how chance often dictates not only what survives, but also what makes it into library catalogues and (from there) into literary scholarship. His lines can be found written on single sheets, which are particularly susceptible to loss or damage, as with the two stanzas from Venus and Adonis (ll. 91–102) transcribed on the back of a legal draft from the 1630s (Folger MS X.d.562), or the version of Sonnet 128  in Bodleian MS Rawl. poet. 152 (c. 1613–20), where it appears squeezed on to a leaf of paper with three other verses (by John Dowland (1563?–1626), William Browne (1590/91–1645?), and Francis Davison (1573/4–1613–19)). Alternatively, we have snippets of verse written into the blank pages of unrelated works, such as the extract from Venus and Adonis (ll. 229–40) discovered in the 1980s by Hilton Kelliher in CUL MS Mm.3.29 (fol. 63v), entered by Henry Colling (a former student of St John’s, Cambridge) c. 1596 into a late fifteenth-century volume of eight tracts relating to the Scottish expeditions of Edward I and III, and to the reign of Richard II (Kelliher 1990). Some of these oddlyplaced jottings teeter on the edge of extinction: the one known extract, dating from

THE SHAKESPEARE MANUSCRIPTS

63

the 1620s or 1630s, from A Lover’s Complaint (ll. 41–2) – Folger MS V.a.150, item 1 – appears ‘partly erased’ on ‘the detached cover of an octavo book’ (Beal 2005: 13); one-and-a-half lines from Lucrece (ll. 1086–7) are found on the first leaf (an especially vulnerable position) of Alnwick Castle MS 525 (c. 1597), a folio volume where every extant page is damaged. Identifying extracts is further hampered by the fact that – as is usual in manuscript transmission (even for poems copied from print) – most of the copies of Shakespeare’s published works are unattributed: only six of the thirty-seven manuscripts acknowledge him by name or initials. None of the copies are in Shakespeare’s hand. A few scholars make a case for some of the manuscripts capturing earlier versions of poems: Roland M. Frye suggests that the four stanzas of Lucrece, transcribed on fols 54r-v of BL Add MS 52585 (c. 1597–1628) in the six-line sesta rima form of Venus and Adonis (ababcc), rather than the seven-line rhyme royal (ababbcc) in which the poem appeared in print, represents an ‘intermediate stage’ of its composition (1965: 289). Gary Taylor (1985: 228ff) similarly argues that twelve of the thirteen copies of Sonnet 2 derive from an earlier draft on the basis that a number of the variants – which are sufficiently visually distinct from the printed version that they cannot be attributed to a copyist misreading what is on the page in front of them – are reminiscent of vocabulary that Shakespeare was using in the 1590s, whilst others are closer to Thomas Wilson’s translation of Erasmus’s ‘Epistle to persuade a young man to marriage’ (from Wilson’s 1553 Arte of Rhetorique), a source on which Shakespeare draws in Sonnets 1–17. However, as with identifying the source for printer’s copies of most of the plays (discussed above), the evidence is not definitive: even Frye admits that his ‘hypothesis cannot be irrefutably demonstrated’ (1965: 290). It is quite possible that these alternate versions derive from the printed source imperfectly remembered or deliberately adapted, particularly as not one of the manuscript copies of Shakespeare’s poems pre-dates their print publication. Shakespeare’s poems must have circulated in manuscript in the 1590s: Francis Meres famously refers to Shakespeare’s ‘sugred Sonnets among his priuate friends’ in Palladis Tamia (1598: 2O1v-2r), more than a decade before their print publication, and the printer William Jaggard acquired versions of Sonnets 138 and 144, using them as headline copy in The Passionate Pilgrim (1599). Nonetheless, the earliest copy of any of the poems is probably Colling’s excerpt (c. 1596) of Venus and Adonis; the earliest of any of the sonnets is probably the version of Sonnet 128 found in Bodleian MS Rawl. poet. 152: its location alongside three poems available in print either for the first time or in new editions between 1611 and 1613 makes a terminus a quo of 1613 likely (Duncan-Jones 1997: 461). When it comes to Sonnet 2, the oldest manuscripts containing copies are from the 1620s and 1630s (over a decade after the 1609 quarto), where it appears alongside verses by the seventeenth-century poets Richard Corbett (1582–1635), Robert Herrick (1591–1674), Thomas Carew (1594/5–1640) and William Strode (1601?–1645). This common company with these later poems and the dearth of sixteenth-century copies make it more likely that this version stems not from an Elizabethan original but from a memorial reconstruction of the printed text: many of the variants – changes in grammatical categories, substitutions of recurrent

64

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

imagery or similar sounding syllables – are of the type that might occur through partial recall (Shrank and Lyne 2018: 721). At least six of the relevant manuscripts are closely connected with the University of Oxford, particularly Christ Church; a further two have likely associations with the Inns of Court (where university students often migrated post-graduation). Mary Hobbs has persuasively argued that the Christ Church fellow (later Bishop of Winchester) George Morley (1598?– 1684) played a key role in the dissemination of Sonnet 2 by lending his miscellany (Westminster Abbey MS 41) to his students, who then copied that version into their own collections (1992: 116–29). The general view, then, is that ‘those who copied Shakespeare’s poems … for the most part, used printed editions as their sources’ (Marotti and Estill 2012: 54). However, these printed sources might be at a further remove from the ‘Shakespearean’ text, taken from printed commonplace books such as Englands Parnassus (1600) or John Bodenham’s Bel-vedére, or the Garden of the Muses (1600, 1610), or – later – from John Benson’s 1640 edition of Poems: written by Wil. Shake-speare. Gent., which re-ordered the sonnets, gave them descriptive headings and combined many of them into longer poems of two to five sonnets each. A stanza from Venus and Adonis (ll. 529–34), beginning ‘Look, the world’s comforter, with weary gait’, appears in three manuscripts, for example: Rosenbach MS 239/27 (c. 1634), p. 166; Rosenbach MS 1083/16 (c. 1634), p. 175 (entitled ‘Good Night to you’); and Huntington Library MS HM 116 (late 1630s), p. 32 (again entitled ‘Good Night to you’). Extracted from Venus and Adonis, they become, in Sasha Roberts’s words, ‘simply musings on nightfall’ (2003: 91). These copies almost certainly derive from Englands Parnassus (sig. Y7v), where the same lines come under the heading ‘Vesper’ (Latin for ‘evening’) and share the same opening variant (where ‘Now’ is substituted for Shakespeare’s ‘Look’). The excerpts from Shakespeare’s sonnets which appear in Folger MS V.a.148 (c. 1650s) can be attributed with similar certainty to a ‘second-hand’ source: the order follows Benson’s 1640 Poems (Marotti 1990: 164), and the titles ‘Cruel’ and ‘A Monument’ for its versions of Sonnets 1–2 and 107 (fol. 22r) recall Benson’s headings for the poems from which these lines are extracted: ‘Loves Crueltie’ and ‘A monument to Fame’ (sigs A5v–A6r, F6r–v). The anonymous compiler of this Folger manuscript plays fast and loose with their copy-text. Whilst many of the excerpts are mere snippets, probably selected ‘as memorable poetic gems or felicitous expressions’ (Marotti 2010: 189), the compiler also makes creative use of this bricolage. Benson coalesces sonnets in their entirety; this copyist melds portions of sonnets, as with the synthesis of Sonnets 60 (ll. 5–12) and 65 (ll. 2–8) on fol. 22r, which – extracted from Benson’s seventyline poem (‘Injurious Time’) – are reconstituted as a fourteen-line ‘sonnet’, rhyming ababcdcdefghgh. The manuscript traces of Shakespeare’s poems might have little, or no, textual authority; nonetheless, as this example demonstrates, they are valuable evidence of contemporary tastes and literary practices, as readers recurrently rewrite and repurpose the words they encountered in print. As Marotti notes (describing a composite poem created from Sonnet 106 and an eighteen-line poem, ‘When mine

THE SHAKESPEARE MANUSCRIPTS

65

eyes first admiring of your beauty’, possibly by William Herbert, Earl of Pembroke (1580–1630)), Shakespeare’s poetry ‘was not … treated like a sacred text, but as a found-object capable of being put to a number of rhetorical uses’ (1990: 149). These ‘rhetorical uses’ tend to fall into two categories: moralizing and eroticizing (both discussed below). Further to that, like other poems from the period, there is a tradition of setting them to music. Instances include a version of Sonnet 116, set to music by the singer and composer Henry Lawes (1596–1662), in New York Public Library, Music Division, Drexel MS 4257, no. 33 (a music book, compiled c. 1630s–50s); a composite of lines 17–18 and 223–4 from Venus and Adonis, appearing in Rosenbach MS 1083/16 (c. 1630), entitled ‘Kissing: a Song’ (p. 279), and British Library Add MS 30982 (1630s), labelled ‘Song’ (fol. 22r); and a stanza, again from Venus and Adonis (ll. 517–22, incipit ‘A thousand kisses’), which is found in two song books: BL Add MS 24665 (c. 1615–26), fols 72v-73r, and Bibliothèque Nationale, Paris, Département de la Musique, MS Conservatoire Rés. 1186 (1630s), fol. 56v. The six stanzas from Lucrece (ll. 869–82, 897–924) in BL Add MS 52585, meanwhile, are presented among ‘diuers ditties to be sung and plaid’ (fol. 53r), and the version of Sonnet 2 in Folger MS V.a.345 (c. 1630s) is dubbed ‘Spes Altera A Song’ (p. 145). The moralizing strand in the manuscript life of Shakespeare’s poems owes much to the Renaissance culture of commonplacing: a custom, engrained from schooldays onwards, of creating ‘collection[s] of notes from reading and other sources that the compiler might want to recall, and reuse, at a later date’ (Moss 1996: 143). Such notes were usually gathered under headings, often arranged alphabetically. These commonplacing habits are perhaps glimpsed in the iterative heading ‘Another’ (i.e. on night) given to the ‘worlds comforter’ stanza from Venus and Adonis in Rosenbach MS 239/27, or in the couplet from Lucrece (ll. 958–9) that is inscribed under the title ‘On time’ in Bodleian Rawl. D.954 (c. 1670s), fol. 41v. Plucking lines out of context in this way has the capacity to transform their meaning: Roberts  (2003) locates a ‘previously unnoticed’ couplet adapted from Venus and Adonis (ll. 149–50) in Bodleian MS Rawl. Poet 117 (fol. 4r, reverse pagination), owned and partly compiled by the Oxford printer Christopher Wasse (1627–90). Removed from the goddess’s speech of seduction and placed among a string of aphorisms, the lines lose the transgressive edge they have in Shakespeare’s poem, where they are ‘spoken by an unconventional woman in the unconventional act of seducing a man’ (Roberts 2003: 90–1). Nor is it even necessary to change a word. Another seductive couplet from Venus and Adonis (ll. 131–2) – ‘Fair flowers that are not gathered in their prime / Rot and consume themselves in little time’ (ll. 131–2) – becomes a sententia about the need to procreate (akin to the message of Sonnets 1–17) once copied into the margins of BL Royal MS 8.A.XXI (fol. 153v), a volume of thirteenth-century Latin treatises. More usually, some adjustments are needed to departicularize the literary text and make it suitable for more general application. This might take the form of adapting the syntax, as with the couplet ‘On time’ in Bodleian Rawl. D.954, where ‘It cheares’ replaces ‘To cheer’ to give the excerpt a subject. Larger-scale alterations might, however, be required: omitting

66

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

specific references to sexual sin and to named characters from Lucrece stanzas in BL Add MS 52585, for example, ‘removes the charged narrative and emotional coordinates of the passage in Shakespeare’s poem’, rendering it ‘a generic mournful meditation on the cruelties of life’ (Roberts 2003: 137). The recurrent recourse that manuscript compilers had to one of the lewder sections of Venus and Adonis (ll. 229–40), where Venus describes her body as a ‘park’ for Adonis to explore, indicates an enthusiasm for the erotic. (As well as providing half of the composite ‘Kissing’ song discussed above, this is the passage selected for transcription by Colling c. 1596.) Elsewhere, the bawdiness of less explicit lines is accentuated by their juxtaposition with cruder verses, as with the stanzas from Venus and Adonis penned on the reverse of a 1630s legal draft (Folger MS X.d.562), which are headed by a couplet beginning ‘Wherfore did venus loue adonis but for the member were noe bone is’. This taste for erotica is found, more disturbingly, in the multiple manuscripts which excerpt from Lucrece voyeuristic passages describing the sleeping protagonist viewed through her rapist’s eyes: Rosenbach MS 239/16 (c. 1713), in which lines 386–95 are copied on p. 146 from the version in John Suckling’s Fragmenta Aurea (1646), and BL Add MS 27406 (c. 1630s), which contains three entries from this section of the poem (ll. 365–71, 386–99, 419–20). This latter copy changes the pronouns of the middle extract so that the scene described is not that of a predatory man gazing on his inert victim, but that of an entwined couple, slumbering side by side: ‘Her lilly hand his rosy cheekes lyes under’; ‘Without the bed his other faire hand was’ (fol. 74r; emphases added). Roberts and Sonia Massai disagree about the ethical import of this version. For Roberts, it ‘appear[s] to have more to do with reading for pleasure than profit’ (2003: 140). Massai reads the extract didactically, in the light of the epitaphs and meditations on death that surround it, and in particular the presence on the following page of an earlier stanza from Lucrece (ll. 211–17, incipit ‘What win I, if I gain the thing I seek?’): ‘By inverting the order in which this stanza and the previous passages appear in Shakespeare’s poem, the compiler shows that moral reflection – “Who buyes a minutes mirth to wayle a weeke / Or sels eternitie to get a toy?” – follows the consummation of an illicit passion’, she argues (2012: 159). Whether this extract is intended to titillate readers or to alert them to the transience and perils of desire, what is clear is that modifying the pronouns transforms rape into a mutual encounter. As such, it epitomizes the way in which the interventions of seventeenthcentury copyists frequently serve to normalize dynamics of gender and sexuality. This practice can be seen in the ‘Kissing’ song from Venus and Adonis mentioned earlier (Rosenbach MS 1083/16, p. 279; cf. BL Add MS 30982, fol. 22r). Stripped from its narrative context, in which a powerful female instructs her would-be lover to enjoy her body, the lines become a more conventionally male-voiced plea for physical satisfaction through the addition of the epithet ‘sweet’ (‘Come sweet sit heere’ over Shakespeare’s ‘Here come and sit’, l. 17) and pronoun changes in the final couplet: ‘Let me graze on thy lips’ and ‘Ile stray lower’, over the original imperatives ‘Graze on my lips’, ‘Stray lower’ (ll. 233–4; emphases added). This reorientation is achieved partly by lifting material out of its original context, partly through deliberate adaptation, as can also be seen with the manuscript treatment of

THE SHAKESPEARE MANUSCRIPTS

67

Shakespeare’s sonnets. Eleven complete sonnets are copied into manuscript (Sonnets 2, 8, 32, 33, 68, 71, 106, 107, 116, 128 and 138): all but two come from the portion of the volume about, or addressed to, the young man. Removed from this frequently homoerotic context, they are recurrently given headings which heterosexualize their content: ‘To one that would die a maide’ (five versions of Sonnet 2); ‘A Lover to his Mistres’ (another version of Sonnet 2); ‘On his Mistris Beauty’ (Sonnet 106). As Roberts observes, this heterosexualizing of the sonnets is further reinforced ‘by their relocation among amatory … poems [e.g. by Strode and Corbett], that describe heterosexual love affairs’ (2003: 173). As with the snippets or stanzas extracted from Shakespeare’s narrative poems, decoupling the sonnets from their original context has the capacity to change how they are read, even without further intervention on their copyists’ part.

CONCLUSION Despite the case made for Hand D in More, we lack a literary manuscript in Shakespeare’s hand. Getting close to what Shakespeare might have originally written – or what his playing company actually performed – requires a degree of scholarly reconstruction, and an acceptance that much of this reconstruction is likely to remain hypothetical. Nonetheless, the manuscript sources do bear witness – if not to the ghost of Shakespeare at work – then at least to his relevance: his continued re-summoning by seventeenth-century readers, who fitted him into new forms, for different purposes, and other occasions.

NOTES 1. A fourth contributor was Greg, who was silent about the question of Hand D as Shakespeare (Wilson 1956: 74). 2. For an account and transcriptions of these manuscript sonnets, see Shrank and Lyne (2018), appendix 4. Quotations from Shakespeare’s non-dramatic poetry is from this source. 3. Data from Beal (2005–13), supplemented with Roberts (2003). This figure excludes citations in William Scott’s Modell of Poesy (BL Add MS 81803). 4. For an account and transcriptions of these poems, see Shrank and Lyne (2018), appendix 6. 5. Data from Beal (2005–13).

REFERENCES Beal, Peter, comp. (2005–13), Catalogue of English Literary Manuscripts 1450–1700. Available online: http://www.celm-ms.org.uk/ (accessed 13 November 2020). Berry, Herbert (1999), ‘The Date on the “Peacham” Manuscript’, Shakespeare Bulletin, 17: 5–6.

68

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Butler, Martin, ed. (2005), Cymbeline, New Cambridge Shakespeare, Cambridge: Cambridge University Press. Cloud, Random (i.e. Randall McLeod) (1997), ‘What’s the Bastard’s name?’, in George Walton Williams (ed.), Shakespeare’s Speech-Headings: Speaking the Speech in Shakespeare’s Plays, 133–209, Newark, DE: University of Delaware Press. Downs, Gerald (2007), ‘A Question (not) to be Askt: Is Hand D a Copy’, Shakespeare Yearbook, 16: 241–66. Duncan-Jones, Katherine, ed. (1997), Shakespeare’s Sonnets, London: Thomas Nelson. Dusinberre, Juliet, ed. (2006), As You Like It, London: Thomson Learning. Dyce, Alexander, ed. (1844), Sir Thomas More: A Play Now First Printed, London: Shakespeare Society. Dyce, Alexander, ed. (1857, 1864–7, 1875–6), The Works of William Shakespeare, 6 vols, London: Moxon; 2nd edn, 8 vols, London: Chapman; 3rd edn, 9 vols, London: Chatto and Windus. Evans, G. Blakemore (1962), ‘The Douai Manuscript – Six Shakespearean Transcripts (1694–95)’, Philological Quarterly, 41: 158–72. Frye, Roland M. (1965), ‘Shakespeare’s Composition of Lucrece: New Evidence’, Shakespeare Quarterly, 16: 289–96. Greg, W. W., ed. (1911), The Book of Sir Thomas More, Malone Society Reprints, Oxford: Malone Society, rpt. 1961, 1991. Greg, W. W. (1931), Dramatic Documents from the Elizabethan Playhouses: Stage Plots, Actors’ Parts, Prompt Books, 2 vols, Oxford: Clarendon Press. Greg, W. W. (1954), The Editorial Problem in Shakespeare, 3rd edn, Oxford: Clarendon Press. Greg, W. W. (1955), The Shakespeare First Folio, Oxford: Clarendon Press. Hackett, Helen (2012), ‘“As the Diall Hand Tells Ore”: The Case for Dekker, not Shakespeare, as Author’, Review of English Studies, 63: 34–57. Hattaway, Michael (2009), ‘Dating As You Like It, Epilogues and Prayers, and the Problems of “As the Dial Hand Tells O’er”’, Shakespeare Quarterly, 60: 154–67. Hays, Michael L. (2016), ‘Shakespeare’s Hand Unknown in Sir Thomas More: Thompson, Dawson, and the Futility of the Paleographic Argument’, Shakespeare Quarterly, 67: 180–203. Hobbs, Mary (1992), Early Seventeenth-Century Verse Miscellany Manuscripts, Aldershot: Scolar. Howard-Hill, T. H. (1974), Ralph Crane and Some Shakespeare First Folio Comedies, Charlottesville, VA: University Press of Virginia. Howard-Hill, T. H. (1995), Middleton’s ‘Vulgar Pasquin’: Essays on ‘A Game at Chess’, Newark, DE: University of Delaware Press. Hoy, Cyrus (1962), ‘The Shares of Fletcher and his Collaborators in the Beaumont and Fletcher Canon (VII)’, Studies in Bibliography, 15: 71–90. Ioppolo, Grace (2006), Dramatists and their Manuscripts in the Age of Shakespeare, Jonson, Middleton and Heywood: Authorship, Authority and the Playhouse, London: Routledge. Jackson, MacDonald P. (2006), ‘The Date and Authorship of Hand D’s Contribution to Sir Thomas More: Evidence from “Literature Online”’, Shakespeare Survey, 59: 69–78.

THE SHAKESPEARE MANUSCRIPTS

69

Jowett, John (1983), ‘New Created Creatures: Ralph Crane and the Stage Directions in The Tempest’, Shakespeare Survey, 36: 107–20. Jowett, John (2017), ‘Exit Manuscripts: The Archive of Theatre and the Archive of Print’, Shakespeare Survey, 70: 113–22. Kelliher, Hilton (1990), ‘Unrecorded Extracts from Shakespeare, Sidney and Dyer’, in Peter Beal and Jeremy Griffiths (eds), English Manuscript Studies, 1100–1700, vol. 2, 163–88, Oxford: Basil Blackwell. Lennam, T. N. S. (1965), ‘Sir Edward Dering’s Collection of Playbooks, 1619–1624’, Shakespeare Quarterly, 16: 145–53. Levin, Richard (2002), ‘The Longleat Manuscript and Titus Andronicus’, Shakespeare Quarterly, 53: 323–40. Marotti, Arthur F. (1990), ‘Shakespeare’s Sonnets as Literary Property’, in Elizabeth D. Harvey and Katharine Eisaman Maus (eds), Soliciting Interpretation: Literary Theory and Seventeenth-Century Poetry, 143–73, Chicago: University of Chicago Press. Marotti, Arthur F. (2010), ‘Shakespeare’s Sonnets and the Manuscript Circulation of Texts in Early Modern England’, in Michael Schoenfeldt (ed.), A Companion to Shakespeare’s Sonnets, 185–203, Oxford: Blackwell. Marotti, Arthur F. and Laura Estill (2012), ‘Manuscript Circulation’, in Arthur Kinney (ed.), The Oxford Handbook of Shakespeare, 53–70, Oxford: Oxford University Press. Massai, Sonia (2012), ‘Early Readers’, in Arthur Kinney (ed.), The Oxford Handbook of Shakespeare, 143–61, Oxford: Oxford University Press. Maxwell, J. C., ed. (1960), Cymbeline, The New Shakespeare, Cambridge: Cambridge University Press. McInnis, David and Matthew Steggle, eds (2014), Lost Plays in Shakespeare’s England, New York: Palgrave Macmillan. McKerrow, Ronald B. (1935), ‘A Suggestion regarding Shakespeare’s Manuscripts’, Review of English Studies, 11: 459–65. Moss, Ann (1996), Printed Commonplace Books and the Structuring of Renaissance Thought, Oxford: Clarendon Press. Pollard, Alfred W. (1920), Shakespeare’s Fight with the Pirates and the Problems of the Transmission of his Text, Cambridge: Cambridge University Press. Pollard, Alfred W., ed. (1923), Shakespeare’s Hand in the Play of Sir Thomas More, Cambridge: Cambridge University Press. Purkis, James (2016), Shakespeare and Manuscript Drama: Canon, Collaboration and Text, Cambridge: Cambridge University Press. Ringler, William A. and Steven W. May (1972), ‘An Epilogue Possibly by Shakespeare’, Modern Philology, 70: 138–9. Roberts, Sasha (2003), Reading Shakespeare’s Poems in Early Modern England, Basingstoke: Palgrave Macmillan. Schücking, L. L. (1925), ‘Shakespeare and Sir Thomas More’, Review of English Studies, 1: 40–59. Shrank, Cathy and Raphael Lyne, eds (2018), Shakespeare’s Poems, London: Routledge. Simpson, Richard (1871), ‘Are There Any Extant MSS. in Shakespeare’s Handwriting?’, Notes and Queries, 4th ser., 8: 1–3.

70

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Tannenbaum, Samuel A. (1927), The Booke of Sir Thomas Moore, New York: privately printed. Taylor, Gary (1985), ‘Some Manuscripts of Shakespeare’s Sonnets’, Bulletin of the John Rylands Library, 68: 210–46. Taylor, Gary and John Lavagnino, eds (2007), Thomas Middleton and Early Modern Textual Culture: A Companion to the Collected Works, Oxford: Clarendon Press. Thompson, Sir Edward Maunde (1916), Shakespeare’s Handwriting: A Study, Oxford: Clarendon Press. Van Dam, B. A. P. (1924), The Text of Shakespeare’s Hamlet, London: J. Lane. Wells, Stanley and Gary Taylor with John Jowett and William Montgomery (1987), William Shakespeare: A Textual Companion, Oxford: Clarendon Press. Werstine, Paul (2007), ‘The Science of Editing’, in Andrew Murphy (ed.), A Concise Companion to Shakespeare and the Text, 109–27, Oxford: Blackwell. Werstine, Paul (2012), Early Modern Playhouse Manuscripts and the Editing of Shakespeare, Cambridge: Cambridge University Press. Werstine, Paul (2015), ‘Ralph Crane and Edward Knight: Professional Scribe and King’s Men Bookkeeper’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 27–38, Cambridge: Cambridge University Press. Werstine, Paul (2020), ‘Lost Playhouse Manuscripts’, in Roslyn Knutson, David McInnis and Matthew Steggle (eds), Loss and the Literary Culture of Shakespeare’s Time, 41–54, New York: Palgrave Macmillan. Williams, George Walton and G. Blakemore Evans (1974), The History of King Henry IV as Revised by Sir Edward Dering, Bart., Folger Facsimiles, Manuscript Series, Charlottesville, VA: University Press of Virginia. Wilson, John Dover (1956), ‘The New Way with Shakespeare’s Texts: An Introduction for Lay Readers. III. In Sight of Shakespeare’s Manuscripts’, Shakespeare Survey, 9: 69–80.

CHAPTER 2.2

The early printed texts of Shakespeare JOHN JOWETT

Shakespeare’s plays were first inscribed onto paper by pen. Sadly, the resulting documents – the first manuscript of each play and any copies that may have followed – are now long lost.1 Our knowledge of Shakespeare’s plays is grounded in the small-format printed editions of single plays published during his lifetime and soon after,2 and on the large and much more impressive collection of thirty-six plays issued in 1623, the First Folio. When a play was issued more than once, the texts differ, sometimes slightly, sometimes in striking or even startling ways. This chapter examines why that should be so. Alterations made or sanctioned by the author are customarily described in terms cognate with the word author: they are said to have higher authority or authenticity (Orgel 2002; de Grazia and Stallybrass 1993: 255–8). In contrast, changes that result from inattention, incompetence, difficult circumstances of working, reliance on hearing, memory or some other alternative to document-to-document copying are customarily described as errors of transmission, or corruptions. But this neat distinction between the authoritative and the corrupt is complicated by a middle ground in which the circumstances are open to interpretation and the scheme itself might start to falter. What if more than one author wrote the text? What level of evidence is required to identify changes sanctioned by the author? To what extent are annotators and copyists who are involved in developing a text justified in what they do? And, in any case, how do we know what happened? These questions have a particular insistence and indeed urgency when applied to a playwright for the early modern theatre. Authorial collaboration was commonplace. Annotation of theatre manuscripts was so regular as a practice that no surviving manuscript demonstrates the purely authorial script as it was first delivered to the theatre company. The dramatist knew that alterations would be made, and that he was communicating with the actors only indirectly via the annotator who would modify, regularize and supplement them. So the manuscripts of Shakespeare’s plays were not initially prepared with print in view. The first print edition came, if at all, after a lapse of time. For example, the plays published in the year 1600 (exceptional for the number in a single year)

72

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

were first written and performed in about 1599 (Henry V), 1598 (2 Henry IV, Much Ado About Nothing), and 1596 (A Midsummer Night’s Dream and The Merchant of Venice). Crossing time as well as medium, the texts of these plays were repurposed for use by a reading public. The text was taken out of its more fluid embodiment as a theatre manuscript that could be altered according to need. True, printed books were sometimes annotated by readers, and sometimes copied by compositors to produce a new edition, but they nevertheless assert themselves as a final and fixed product. They are designed for immediate private use, rather than to enable the collective enterprise of taking the words away from the page and speaking them in the theatre. The small-format editions of single plays and the Folio differ considerably from play to play with regard to length, the prevalence of error, the style and adequacy of stage directions, the regularity of metre, the consistency of form and abbreviation of speech prefixes, and the range of spelling variants and contracted forms of words. But profound differences in length and substantive content within the small format editions of a single play are confined to two plays, Romeo and Juliet (Q1 1597, Q2 1599) and Hamlet (Q1 1603, Q2 1604/5).3 The printers of both second editions were working from a much longer and more reliable manuscript than the one used by the printers of the first edition.4 What are the implications of such variable texts? The 1623 Folio reinforces these questions, and indeed transforms the text of Shakespeare. It not only publishes about two-fifths of Shakespeare’s plays for the first time.5 It also introduces various new examples of plays with texts that must necessarily have been set from or with reference to manuscripts that were considerably variant on the manuscript copy texts for the small-format editions. Changes to a text between one edition and another, deliberate or accidental, might be introduced into manuscripts by revising authors, revisers other than the original authors or copyists (either in the theatre or outside). A separate tier of alteration comes from publishers or their agents as editors, and printinghouse compositors. All these alterations, deliberate or accidental, reflected the perpetrators’ function as textual transmitters, their skills, their motivations and the constraints under which they worked. There is a broadly theoretical problem in defining the acceptability of these changes. The extent to which Shakespeare as author should be accorded priority over other agents is a key issue. In the absence of surviving manuscripts, there is a further problem of a diagnostic kind in identifying the agent responsible for a particular variant. Put these two issues together and one has a quandary as to defining the boundary between a version variant – that is to say, a variant in which each of the alternative readings is legitimate in the context of the text in which it is found – and an error. Scholars who place value on the material printed texts that survive will tend to emphasize the integrity of those texts as separate versions. Those who place value on the origin of the text as a script written for performance will tend to separate out agency, to highlight the playwright as author and sometimes as reviser, and to draw attention to the persistent phenomenon of error.

THE EARLY PRINTED TEXTS OF SHAKESPEARE

73

HOW TEXTS DIFFER The main mechanisms by which texts come to have variant readings are of three kinds: they result in the simple reprint; the reprint from modified copy (usually a printed text modified from a manuscript); and the new edition printed from a manuscript.

1. Simple reprint The textual history of a single line of 1 Henry IV can illustrate the fortunes of the text over a succession of reprints, each set from the most recent previous edition (from sig. A2v of all editions). Q1 (1598):6  Q2 (1599): Q3 (1604): Q4 (1608): Q5 (1613): Q6 (1622): Q7 (1632):

West. This matcht with other did, my gratious L. West. This matcht with other did my gracious L. West. This matcht with other like my gracious L. West. This matcht with other like my gracious L. West. This matcht with other like, my gracious L. West. This matcht with other like my Gracious Lord, West. This match with other like, my Gracious Lord,

The first printing is closest to the underlying manuscript; the text slowly evolves away from it.7 Q2’s alteration of spelling from ‘gratious’ to ‘gracious’, Q6’s capitalization of this word, and its expansion of the abbreviation ‘L.’ to ‘Lord’ are all routine adjustments to improve the text from the reader’s point of view, or to impose the preferences of the compositor. In doing so, they will inevitably tend to take us further away from the text as Shakespeare wrote it.8 Alterations such as these are not aimed at affecting the sense, as spellings and punctuation varied considerably. More substantive is Q3’s alteration of ‘did’ to ‘like’. In modern spelling and punctuation the Q3 text now reads ‘This matched with otherlike, my gracious Lord’. This looks like a purposeful if misguided attempt to improve the text: the editor or compositor may have been thrown by Shakespeare’s use of ‘other’ as a collective, where the plural ‘others’ was steadily becoming more usual.9 Finally, in 1632 the text acquired a new error when ‘matcht’ was set as ‘match’, which turns the verb into a noun if sense is to be made of it. How this might have been considered necessary or desirable is hard to see; an unintended omission of a letter is more likely. With both readings, the intentions of the compositor can be debated, but both alterations are certainly errors.10 The crucial text not recorded above is the 1623 Folio. It was set from Q5, and retained its error: ‘West. This matcht with other like, my gracious Lord’ (sig. d5v). The Folio text of this play contains enough corrections of its copy-text to suggest the quarto had been annotated from a manuscript, but the annotation was so light that the retention of the Q5 reading does not lend it any authority. The picture of a gradual evolution away from the original manuscript is not affected. Of course, if the first printing has a manifest error, there is every chance that a reprint will correct it. At 2.5.362 of the same play, the Q1 reading ‘Mv noble Lord’

74

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

contains an obvious misprint, and Q2 duly altered ‘Mv’ to ‘My’.11 Where Falstaff accuses Hal of having ‘vnsauory smiles’ (1.2.62) the error was manifestly far harder to spot. This was the reading of all the quartos before 1622, and of the Folio too. But an attentive individual, whether the editor preparing the 1622 quarto or the compositor, recognized that a far better reading would be ‘similes’. The reading also has the virtue of assuming a very simple error in the original omission of an ‘i’.12 Even though the emendation was introduced without reference to any independent document that might make it more than an educated guess, ‘similes’ has been accepted by editors ever since.

2. Mixed copy: reprint modified with reference to a manuscript A number of Folio texts were set from quarto copy that had been marked up with readings from a manuscript. This transfer from the manuscript to the quarto copy took place with varying degrees of thoroughness on the part of the annotator, and it is evident too that there the volume of variant readings available to the annotator also varied. Here is an example from Richard III illustrating the kind of mark-up that the annotator preparing the copy for the Folio text would have introduced into the quarto from the manuscript: Exe. Tush, feare not, my L./ we will not stand to prate;

 il. Tut, tut, v my Lord/

Talkers are no good doers be assured: We come/ to vse our hands and not our tongues. go/ [page break] Glo./ Your eies drop milstones, when fooles eies drop/tears. Rich./ fall/ I like you Lads, about your businesse^. Exeunt. straight^ ^ Enter Clarence, Brokenbury. Scena Quarta. |^ and Keeper/ Bro./ Why lookes your Grace so heauily to day? Keep./ Cla. Oh, I haue past a miserable night, So full of vgly sights, of gastly dreames/, fearefull Dreames, of vgly sights/ That as I am a Christian faithfull man, I would not spend another such a night, Though t’were to buy a world of happie dayes, So full of dismall terror was the time. Bro./ What was your dreame? I long to heare you tell it/. Keep./ dream my Lord, I pray you tel me/ The main text reproduced above is from sigs. C3v–4r in Q3 (1602), which was the main printer’s copy. The deletions and the insertions added to the right in bold offer an approximate reconstruction of the changes that the annotator would have marked.13 As reconstructed, the mark-up is fairly dense. In other parts of Richard III, the annotator would have found it even harder to fit the notes into the page margins. Longer passages present in the Folio but not the quartos would presumably have been copied onto separate slips or leaves of paper.

THE EARLY PRINTED TEXTS OF SHAKESPEARE

75

The end in view was to capture as much detail as possible from the manuscript without making direct use of it in the printing house. The introduction of act divisions in the Folio, along with the persistent variations in stage directions and speech prefixes, suggest that the manuscript in question belonged to the theatre company.14 The actors were presumably reluctant to lose a document that was important to their archives as a record of the play and perhaps as the document holding the license to perform the play on stage (Taylor 1993c: 239; Gurr 1999: 76). The compromise of annotating a quarto had its own risks: readings in the manuscript might be overlooked, or the annotator’s inscriptions might be difficult for the Folio compositor to follow. Imperfect though this mechanism is, it leads to the Folio offering a different version of the play from the quarto, and correcting various errors. Indeed, the Folio text of Richard III is usually preferred over the quarto version, despite the messy and impure process of transmission that leads to it, and despite the residual presence of readings that are errors introduced in Q2 and Q3.15

3a. Edition set from a new manuscript with theatrical and scribal variation If the simple reprint reflects the complete dependence of one text on its antecedent, the text set from a fresh manuscript exemplifies the opposite quality. A good example is 2 Henry IV, of which the quarto text of 1600 and the Folio text are so distinct in their detail that it is beyond doubt that they were set from different manuscripts, with Q1 showing many characteristics of a text close to Shakespeare’s hand. Here is a representative passage: Host. I come, you starude blood-hound. Whoore Goodman death, goodman bones. Host. Thou Atomy, thou. Whoore Come you thinne thing, come you rascall. Sinck. Very well. Enter strewers of rushes. 1 More rushes, more rushes. 2 The trumpets haue sounded twice. 3 Twill be two a clocke ere they come from the coronati­ on, dispatch, dispatch. Trumpets sound, and the King, and his traine passe ouer the stage: after them enter Falstaffe, Shallow, Pistol, Bardolfe, and the Boy. Falst. Stand heere by me maister Shallow, I will make the King doe you grace, I will leere vpon him as a comes by, and do but marke the countenaunce that he will giue me. Pist. God blesse thy lungs good Knight. 2 Henry IV, Q1 (1600), sigs. K3v–4r

76

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Host. Yes, come you staru’d Blood-hound. Dol. Goodman death, goodman Bones. Host. Thou Anatomy, thou. Dol. Come you thinne Thing: Come you Rascall. Off.  Very well.  Exeunt. _______________________________________________ Scena Quinta. _______________________________________________ Enter two Groomes. 1. Groo. More Rushes, more Rushes. 2. Groo. The Trumpets haue sounded twice. 1. Groo. It will be two of the Clocke, ere they come from the Coronation.  Exit Groo. Enter Falstaffe, Shallow, Pistoll, Bardolfe, and Page. Falstaffe. Stand heere by me, M. Robert Shallow, I will make the King do you Grace. I will leere vpon him, as he comes by: and do but marke the countenance that hee will giue me. Pistol. Blesse thy Lungs, good Knight. 2 Henry IV, in F1 (1623), sig. gg7r The two texts run in parallel, but there are numerous variants between them. As with Richard II and many other plays, F1 is the first printing to introduce act and scene divisions. The stage directions and speech prefixes are persistently different in wording. The script has been revised at some stage to introduce changes to the role identifiers. Derogatory ‘Whoore’ for Doll Tearsheet becomes the more personal ‘Dol’, task-oriented ‘strewers of rushes’ becomes the more vague ‘Groomes’, and so on. More importantly, the dramatic action itself is changed. In Q1 the entry of the King and his train, announced by trumpets, is followed by a second entry after the Falstaff-Pistol dialogue, ‘Enter the King and his traine.’ on the following page. The first entry is a procession ‘ouer the stage’, inferably to the coronation, and the second is a return to the stage, inferably after it. In F1, the first processional entry does not happen at all. Q1 reflects the play before revision, and F1 shows it after revision. The variants in the dialogue confirm that Q1 and F1 have different immediate origins. There is little to suggest literary improvement in F1’s version of the passage. The two direct references to ‘God’ have disappeared. This is consistent with the requirement of the 1606 Act to Restrain Abuses of Players, which outlawed the use of oaths and direct references to God as profanity (Taylor 1993b). The heightened religious sensibility was not limited to the effects of the Act: in the Jacobean period scribes too might remove profanity as a reflection of their own

THE EARLY PRINTED TEXTS OF SHAKESPEARE

77

moral sensibility as much as the requirements of the law (Mowat 2005). In either case, the changes between Q1 and F1 have a direction of travel, and take the script away from the author. The same is true of the minor adjustments that make the language less idiomatic and more formal. Finally, a line of prose, ‘Come you thinne thing, come you rascall’, has been split into two implausibly short verselines, evidence of the Folio compositor’s need to use up space at this point. Thus the Folio text combines expurgation and other alterations, most but not all of them inferably present in the printer’s copy manuscript, that, though deliberate, have little to commend them. The stage directions have been altered to meet the demands of the theatre. Furthermore, over the play as a whole, eight passages, all unchallenged as Shakespeare’s writing, are first printed in F1. In this respect F1 is, after all, a revised text.16 The example typifies the quandary presented by a second text that is valuable in some respects, inferior in others, and at all events simply different.

3b. Edition set from a new manuscript with authorial revision In other cases, such as King Lear, there is a far greater tendency for F1 to introduce regular variants that are likely to be authorial. In this play both Q1 (1608) and F1 have readings that can be defended as being equally viable and equally effective alternatives. Moreover, Q1 King Lear contains passages that are not found in F1, and F1 contains passages that are not in Q1. Critics have debated the causes of such differences.17 But the passages present only in F1 cannot convincingly be explained as deletions from Q1, and appear to represent an authorial expansion of the text that postdates the Q1 version. Some of the passages found only in Q1 might be absent from F1 for reasons of censorship (Patterson 1984), but it would be hard to explain them all in these terms, and therefore Shakespeare emerges as the first contender for originating the textual changes. King Lear has been widely studied. Its closing lines offer the best-known example of revision in the canon. Earlier, where Kent advises Lear in Q1 to ‘Reuerse thy doom’ as against ‘reserue thy state’ in F1 (1.1.143), the variant has no suggestion of error in either text; Samuel Johnson noted in his 1765 edition of Shakespeare that Q1 is ‘more conducive to the present occasion’ but that F1 ‘conduces more to the progress of the action’. Both are valid. Similar local revision is to be found in various plays. Here is a compact example from Troilus and Cressida: What heart from hence receyues the conqu’ring part To steele a strong opinion to themselues, Which entertain’d, Limbes are in his instruments, In no lesse working, then are Swords and Bowes Directiue by the Limbes. Vlys. Giue pardon to my speech: Therefore ’tis meet, Achilles meet not Hector: Let vs (like Merchants) shew our fowlest Wares, And thinke perchance they’l sell: If not,

78

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

The luster of the better yet to shew, Shall shew the better. Troilus and Cressida, in F1 (1623), sig. ¶2v What heart receiues from hence a conquering part, To steele a strong opinion to them selues.   Vliss. Giue pardon to my speech? therefore tis meete, Achilles meete not Hector, let vs like Marchants First shew foule wares, and thinke perchance theile sell; If not; the luster of the better shall exceed, By shewing the worse first: Troilus and Cressida, Q1 (1609) sig. C3v The quotations are here arranged not in the order of publication but in the rather more probable sequence from the first to second version of the passage.18 However, the sequence matters less than the characteristics of the variants, summarized as follows:

Substitution of Equivalent Words F1

Q1

from hence receyues the

receiues from hence a

shew our fowlest

First shew foule

yet to shew, Shall shew the better

shall exceed, By shewing the worse first

Lines with No Equivalent F1

Q1

themselues, Which entertain’d, Limbes are in his instruments, In no lesse working, then are Swords and Bowes Directiue by the Limbes.

them selues.

Altered Lineation F1 … speech:/ … Hector:/ … Wares,/ … not,/

Q1 … meete,/ … Marchants/ … sell;/

THE EARLY PRINTED TEXTS OF SHAKESPEARE

79

The passage is intensively variant, and the differences, though of several distinct kinds, seem interconnected. For instance, the lineation alterations respond to the presence or absence of the part-line ‘Directiue by the Limbes’ immediately before the affected passage. Similarly, the chiasmatic structure of ‘limbs … instruments … swords and bows … limbs’ in the consecutive lines printed only in F1 is found again in the chiastic structure of the F1 sequence that is absent from Q1 for a different reason, verbal variation between the texts, ‘the better yet to show shall show the better’. The differences are clearly not errors,19 for each text maintains its own coherence, and the final four lines have variants that are interdependent from one line to another. Nor are they scribal sophistications, as theatre-based scribes are not known to have undertaken rewriting of this order. Nor are the three lines present only in F1 missing from Q1 due to profanity. Instead, the passage has been subjected to stylistic revision. This is the work of an author. There is every reason to suppose that it was the work of the author, Shakespeare, who would have been more closely attuned to the fine nuances I have described than anyone else.

3c. Edition set from a new manuscript affected by disruptive transmission Textual critics will debate energetically as to whether, for example, Q2 or F1 Hamlet, Q1 or F1 Troilus and Cressida, Q1 or F1 Othello or King Lear, supplies the text that is preferable as the basis for an edition. Typically, each version has lines not found in the other, and each has its virtues. However, where a published edition is conspicuously shorter than its counterpart, the verbal equivalence also breaks down. The short text is always less authorial in its word-to-word expression: less assured, more banal, sometimes unclear, sometimes unmetrical. These are, of course, subjective judgements, and it would be absurd to imagine that every single variant will conform to this characterization. Even the best texts contain error, and even the worst texts have the capacity to correct that error. What this means in practice may be seen by considering a passage from Hamlet. As Q1 has a number of affinities with F1 that are not shared with Q2, it is here compared with F1. As we are here considering words and their meaning, the passages are quoted in modern spelling and punctuation.20 Ay so. Come forth and work thy last— And thus he dies, and so am I revenged! No, not so. He took my father sleeping, his sins brimful. And how his soul stood to the state of heaven Who knows, save the immortal powers? And shall I kill him now, When he is purging of his soul, Making his way for heaven? This is a benefit And not revenge! No, get thee up again. Based on Hamlet, Q1 (1603), sig. G1v Now might I do it pat, now he is praying. And now I’ll do’t—and so he goes to heaven,

80

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

And so am I revenged! That would be scanned, A villain kills my father, and for that I, his foul son, do this same villain send To heaven. O this is hire and salary, not revenge. He took my father grossly full of bread With all his crimes broad blown, as fresh as May, And how his audit stands who knows, save heaven; But in our circumstance and course of thought ’Tis heavy with him. And am I then revenged To take him in the purging of his soul When he is fit and seasoned for his passage? No. Up, sword, and know thou a more horrid hent Based on Hamlet, in F1 (1623), sig. pp1v The difference in length is typical of the texts as a whole. The passage in Q1 is highly deficient in its metre, with only two regular pentameters.21 Hamlet’s thinking is imprecise: whereas in F1 ‘so he goes to heaven’ leads directly to ‘That would be scanned’, the notion of heaven is absent from Q1’s ‘thus he dies’. There is no inherent contradiction, no thought that Hamlet will achieve revenge by saving his uncle’s soul. Most of the rich imagery and specific concrete detail and intellectual energy of Hamlet’s speech in F1 are absent. Typically, the intellectually self-aware and self-critical ‘That would be scanned’ is now a blunt ‘No, not so’. Whereas F1 repeats ‘so’ for parallel reinforcement, Q1 uses the word in three successive lines, with awkward shift between meaning ‘so it is’ and ‘in this way’. Here and elsewhere, the text fluctuates between capability and incapability. These are signs of a derivative and weakened text.22

FINDING DIRECTIONS Or could they be signs of an early draft by Shakespeare (Jolly 2012; Bourus 2014)? It must be admitted that judgements founded on literary interpretation are less than conclusive in demonstrating that Q1 owes its distinctive features to a disruptive transmission taking it away from F1. The alternative, that it is an early version, cannot entirely be discounted on such evidence, even though its cumulative effect over the play as a whole is considerable. Might the poor metre, for instance, be a sign of Shakespeare’s writing at an early stage? If so, the more mature Shakespeare would have removed the defect. However, this conjecture can be tested and seen to be invalid. Shakespeare was accomplished and polished in his use of the iambic pentameter in his earliest dramatic works. For instance, The New Oxford Shakespeare (Taylor and Loughnane 2017: 485–7) identifies the following as the first speech in the earliest of Shakespeare’s surviving dramatic works:23 Cease to persuade, my loving Proteus. Home-keeping youth have ever homely wits.

THE EARLY PRINTED TEXTS OF SHAKESPEARE

81

Were’t not affection chains thy tender days To the sweet glances of thy honoured love, I rather would entreat thy company To see the wonders of the world abroad Than, living dully sluggardized at home, Wear out thy youth with shapeless idleness. The Two Gentlemen of Verona, 1.1.1–8 Shakespeare wrote these lines with a confident and strong sense of iambic pentameter. Metrical variation is, indeed, more limited than is typical of his later writing.24 The passage effectively deploys just one standard adjustment that was allowable in iambic pentameter to vary and enliven the verse without conflicting with metrical norms. This is the trochaic first foot with the stress on the first syllable. It is witnessed in the first line and elsewhere in the passage, and was widely deployed by other dramatists contemporary with early Shakespeare such as Christopher Marlowe. By comparison, the unstable line-lengths of Q1 Hamlet, if deliberate, belong to an ill-formed poetic sensibility. They are more consistent with the text having been impaired during its transmission. The metre, intact in one line but wrecked in the next, has been reduced to a partially-standing ruin.25 Another consideration points still more decisively to the derivative nature of Q1. One consequence of the hypothesis that Q1 is an early version by Shakespeare is that the hypothesis demands a resequencing of Q2 and F1. Textual critics have widely accepted that Q2 is the earlier of the two, on account of its rich evidence of the proximity of Shakespeare’s hand. This sequence can be represented [Q2 > F1]. It has also been understood that Q1, despite its overwhelming difference from either Q2 or F1, has stronger affinities with F1 than with Q2. The lines not found in F1 are not found in Q1 either. The two texts share some tellingly distinct readings. Where in Q2 Marcellus calls on Horatio to ‘Speake to’ the ghost (1.1.43; sig. B1v), in Q1 (sig. B1v) and F1 (sig. nn4v) he says ‘Question’. Hamlet offers to teach Horatio ‘for to drinke’ in Q2 (sig. B4v), but more extravagantly ‘to drinke deepe’ in Q1 (sig. B4r) and F1 (sig. nn6r). Such examples establish distinct affinity between Q1 and F1, as against Q2. If we can accept some kind of disruptive transmission, the source version behind that disruption, which I will call the Q1 antecedent, will lie in-between Q2, with its evident basis in a manuscript close to Shakespeare’s hand, and F1, as a later text showing signs of alteration for later theatre practice: [Q2 > Q1 antecedent > F1] (Taylor 1987: 398).26 If, on the other hand, Q1 is the earliest version, F1 must be intermediate between Q1 and Q2, for otherwise its affinities with Q1 are inexplicable: [Q1 > F1 > Q2]. However, the manuscript behind Q2 is very unlikely be the last text in the series rather than the first for the following reasons: 1. A number of likely authorial variants suggest improvement from Q2 to F1. As an example, at 3.3.7 Claudius says of Hamlet that he threatens ‘Hazerd so neer’s as doth hourely grow / Out of his browes’ in Q2 (sig. H4v); F1 alters the last word to ‘Lunacies’ (sig. pp1r). ‘Lunacies’ is far clearer and

82

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

2.

3.

4.

5.

6.

more specific. In Q2, Hamlet soon after contemplates killing Claudius by saying ‘Now might I doe it, but now a is a praying’ (sig. I1v), which is coherent (‘Now might I do it, just now while he’s praying’), but is less effective than F1’s ‘Now might I do it pat, now he is praying’ (sig. pp1v) with its precise mot juste ‘pat’. Where in the same speech in Q2 Hamlet characterizes his ranting as ‘base and silly’ (sig. I1v), in F1 he calls it ‘hyre and Sallery’ (sig. pp1v). If ‘silly’ might be an error for ‘sallery’, ‘base’ makes excellent if unremarkable sense and is an unlikely error; ‘hire’ is certainly an improvement. Hamlet is an exceptionally long play, especially in Q2, and is likely to have needed cuts if it was the basis for performance. The passages found only in Q2 are better explained as cuts in F1 to reduce the performance time than as additions to Q2, and are sometimes associated with the likely authorial improvements to the surrounding verbal detail.27 The same trajectory is suggested by two of Hamlet’s speeches at 3.2.175 (‘That’s wormwood’) and 218 (‘If she should breake it now.’). In Q2 they are printed in the margin to the right of the otherwise uninterrupted Player’s speech (sigs. H2r, H2v). The unusual positioning suggests additions to the script as first penned. F1 incorporates these short speeches into the main text (sig. oo6v, the first speech reading ‘Wormwood, Wormwood’). There is other evidence of Shakespeare composing as he wrote the manuscript behind Q2. The Player Queen’s speech at 3.2.160–2 contains two closely proximate examples of ‘false starts’. First Shakespeare wrote ‘For women feare too much, euen as they loue’ (sig. H2r), which stands out because the verse-line is unrhymed in a passage of rhymed couplets. He immediately followed this with ‘And womens feare and loue hold quantitie’, which reworks the same idea in a line that provides a rhyme with ‘extremitie’ at the end of the line following. Clearly the second line replaces the first (with the qualification that the intended deletion begins after ‘For’ and ends after ‘And’). The line ending in ‘extremitie’ contains a second change of mind in the course of composition. It reads ‘Eyther none, in neither ought, or in extremitie’. The words ‘Eyther none’ are superfluous to the metre; they produce near-nonsense, and they are paraphrased more intelligibly in the words that immediately follow, ‘in neither ought’. The important point is this: in both cases, the ‘false starts’ have completely disappeared in F1.28 F1’s stage directions have less of the authorial idiosyncrasy of Q2, and they are better regulated. For instance, Q2’s ‘Enter old Polonius, with his man or two’ (sig. E1r) becomes the specific and named ‘Enter Polonius, and Reynoldo’ (sig. oo2r). The reverse development would defy explanation. The superfluous entry of the Queen in Q2 but not F1 at the beginning of 4.1 again suggests that F1 is the later and theatrically better-regulated text.29 Details suggesting Q2’s greater proximity to an authorial script, such as its characteristically Shakespearean spellings, are numerous and pervasive (Jenkins 1982: 39–41).

THE EARLY PRINTED TEXTS OF SHAKESPEARE

83

Though some of this evidence is not separately conclusive, all of it points in the same direction. The third and fourth points are especially compelling. All this strongly suggests that the trajectory begins with the Q2 version, which is clearly the closest to an authorial draft, and develops to the Folio version. If, as is concluded, the Folio version is altered from the Q2 version and is thus the later of the two, and if Q1 is closer to F1 than to Q2, it then follows inescapably that Q1 is also later than Q2. The trajectory is [Q2 > Q1 antecedent > F1]. If this is so, there absolutely must have been a traumatic stage of transmission between the Q1 antecedent and the immediate Q1 copy to explain Q1’s shortness and its other distinctive features. The exact nature of that textual disaster is of less immediate importance than the fact that it took place.30

TYPES OF TEXTS: PROBLEMS AND POSSIBILITIES Analysis of Q1 Hamlet has been assisted by the unique opportunity among Shakespeare texts to triangulate three substantive editions in terms of the relation between the manuscripts from which they were printed. This establishes a firmer conclusion as to the nature of Q1. In so doing, it offers the possibility of reaffirming a class of quartos, the grouping the twentieth-century movement known as the New Bibliography identified as ‘bad quartos’ (Pollard 1909; Greg 1954). This term has fallen from favour, largely because it was over-prescriptive in circumstances where secure diagnosis is difficult (Werstine 1999). Nevertheless, the example of Q1 Hamlet might well indicate that other distinctly short quarto texts were likewise badly transmitted. It should be expected and in normal circumstances required that an early version would be just as Shakespearean as a later version. This expectation is persistently not met: where they differ from the longer texts, the significantly shorter texts are also consistently less Shakespearean in terms of criteria such as diction and metre. The editions where this most obviously holds true are the first editions of 2 Henry VI, 3 Henry VI, Romeo and Juliet, Hamlet and The Merry Wives of Windsor. Potentially, an exception would arise if an early version was collaboratively written, as applies to 2 Henry VI and 3 Henry VI. My own examination of 3 Henry VI explored this scenario (Jowett 2020). The play is now widely accepted to have been co-authored by Shakespeare and others; it may have been expanded by Shakespeare to generate the Folio text. But the features confined to the short text – in this case published in octavo format in 1595 – do not lend themselves to association with Shakespeare, or, for that matter, any other hypothesized dramatist involved in the play’s authorship. Poor metre is often associated with weak paraphrase, a combination that, as has been seen in Q1 Hamlet, seems indicative of bad transmission. This feature and others cut across the division of the play into authorial sections, and so they seem to overlie the original text. Thus the exception proves the rule: far from supporting interpretation as an early version, the features characteristic of octavo 3 Henry VI point to bad transmission, on account of both their intrinsic qualities and their persistent transgression of authorial boundaries.

84

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Disruptive textual transmission places the text on the downwind side of original authorship and performance. The most widely proposed mechanisms for the disruption of direct copying of documents, memorial reconstruction by actors and note-taking by theatre-goers, involve theatrical performance (Stern 2013). For these reasons, disruptive transmission is necessarily associated with deliberate changes made for the purpose of putting the play on stage: the one assumes the other. A likely example has been mentioned already: cuts introduced in F1 Hamlet probably as revisions for performance, such as the omission of Hamlet’s soliloquy ‘How all occasions do inform against me’, also affect Q1.31 With other short quartos it is often impossible to differentiate between theatrical cuts and passages lost through disruptive transmission, because there is no third version against which to triangulate the given two. Nevertheless, theatrical alteration is part and parcel of what we should expect of a short quarto. Theatrical adaptation is more widely associated with texts published in modified or variant texts in the First Folio. Indeed, it is conspicuous. Act intervals were required in plays performed at the Blackfriars after 1608 so that musicians could play while the candles needed for indoor lighting were trimmed or replaced (Taylor 1993a). With the exception of Othello, which was printed in quarto as late as 1622, pre-Folio editions of plays first performed before 1608, that is to say every small-format issued before F1, lacked act divisions. The Folio sets act and scene divisions prominently, and the editors seem to have gone out of their way to include them. The altered stage directions of many Folio texts are also striking in that they are printed separately and in italic. Cuts and variant wording generated by a revising dramatist (Shakespeare himself, or in some cases Thomas Middleton),32 and alterations introduced by theatre annotators and scribes also feature in these Folio texts. It has already been seen how these readings could be copied across into a quarto copy that would be used by the Folio compositors. Other Folio plays were set directly from a manuscript, probably either a theatre playbook or a copy of it. These include The Merry Wives of Windsor, 2 Henry IV, Henry V, 2 Henry VI, 3 Henry VI, Romeo and Juliet, Hamlet, King Lear and Othello. But some Folio plays, including The Merry Wives of Windsor, had acquired an overlay of scribal transcription by Ralph Crane, a scrivener working for the King’s Men in the years running up to the publication of the volume; his transcriptions tend to suppress some of the working theatrical features of the playhouse manuscripts in an attempt to give the texts a more literary and classical feel.33 The mistake textual critics were often tempted to make in the past is to generate categories for describing Shakespeare’s texts that are based on the fixity that characterizes print but not manuscript. Manuscripts are always susceptible to alteration over time. It follows, naturally, that texts of plays printed soon after they were written and performed stand, on the whole, a better chance of being close to the manuscript as Shakespeare wrote it than texts of plays that appear in the 1623 Folio. Alfred Pollard (1909) is credited with establishing the categories of ‘bad quartos’ and ‘good quartos’, and the back-constructed manuscript categories such as ‘foul papers’ and ‘prompt book’. His successor as leading light of the New Bibliography W. W. Greg (1954;

THE EARLY PRINTED TEXTS OF SHAKESPEARE

85

1955) realized that manuscripts were various in kind, and that many Shakespeare texts do not correspond in detail to the categorical headings, but kept the terminology in place nevertheless. That fluidity is the very basis on which a description of the printed texts must be founded. A single manuscript as passed from Shakespeare to the theatre company might show all the characteristics of an uneven authorial script, the document that Greg and others called ‘foul papers’. But such a manuscript is likely, as appears to be the case with the copy for Q2 Hamlet, to have been annotated, and so to lose its defining character (in the accounts of the New Bibliography) of being inscribed entirely be the author. However, likely though it is, there is no absolute necessity for a manuscript to be annotated; it might, for example, be transcribed without annotation, as the New Bibliographers perhaps too broadly assumed. Paul Werstine (2013) empties out the term ‘foul papers’ by placing all the manuscripts immediately connected with the theatre under the heading of ‘playhouse manuscripts’. He bases his argument on an incisively detailed study of all the relevant manuscripts of early modern plays from the professional theatre that survive today. There are nevertheless risks here. The surviving manuscripts make up only a tiny sliver of the totality of documents that would have been prepared in the period. They are definitively not representative of the play manuscripts that were sent to the printing house, all of which are lost. Moreover, it is clear from the surviving examples that their characteristics vary extremely, from the lightly annotated authorial draft to the heavily annotated scribal transcript, with any number of further permutations in between. The globalizing label ‘playhouse manuscripts’ therefore needs to be qualified. In terms of printed editions, there are meaningful distinctions to be made between texts that are close to the author’s hand and texts that show extensive signs of transcription and/or annotation for the theatre. The need is not to abandon such distinctions but to apply them in a descriptive and undogmatic way, recognizing that the terms relating to manuscripts must of necessity be used provisionally, because print occludes the finer grain of the detail (Jowett 2017a). The quarto texts showing closest proximity to Shakespeare’s hand and the least compelling evidence of theatre annotation, if indeed there were any, are the first quartos of Love’s Labour’s Lost (1598), The Merchant of Venice (1600), Much Ado About Nothing (1600), A Midsummer Night’s Dream (1600) and 2 Henry IV (1600). The clustering of publication dates is probably significant in suggesting a relatively settled arrangement at this time for copying authorial manuscripts with the intended or incidental result that the originals were available for releasing to the stationers. Evidence for a closely underlying authorial script includes idiosyncratic spellings associated with Shakespeare, false starts of the kind already witnessed in Q2 Hamlet, extreme variation in the way characters are identified in speech prefixes, and imprecise or descriptive stage directions that are deficient in their identification of the roles involved (‘two or three’, ‘and others’).34 These features are not individually exclusive to authorial papers. Most would remain in place if the manuscript was annotated. Some might survive transcription by another hand. Nevertheless, a strong showing of such evidence alongside a paucity of evidence for theatrical annotation leads towards the conclusion that the manuscript copy was not significantly modified from the script as the dramatist completed it.

86

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Still, the quartos cannot be divided into pre-theatrical ‘good quartos’ and derivative ‘bad quartos’ if the first term depends on the assumption that ‘good quartos’ are dependent on the copy being an unmodified authorial draft. Indeed, neither putative category can be insulated from the theatre itself. Badly transmitted texts, as already noted, are transmitted from the theatre, and bear its imprint. Authorial drafts enter the theatre as an arena for annotation and copying. Even Q2 Hamlet and Q2 Romeo and Juliet, rich in evidence of proximity to Shakespeare’s hand, contain details in a few of their stage directions that suggest light theatrical annotation (Werstine 2013: 231–3). Q1 1 Henry IV is full of evidence that the printer’s copy was a transcript in a hand other than Shakespeare’s (Jowett 1987: 329–30). 1 Henry IV has also been altered to remove, more or less systematically, character names that had caused offence to prominent courtiers, most notably ‘Oldcastle’, a name that, after the earliest performances, was altered to Falstaff (Taylor 1985). Another play that ran into political difficulties, Richard II, was printed from a manuscript from which the long scene in which Richard reluctantly resigns his crown to Bolingbroke had been almost entirely removed. Here again the earliest printed text respond to changes that probably were introduced in the theatre (Clegg 1997).35 Thus the printed texts of Shakespeare’s plays are not the settled things they appear to be. They reflect, with varying degrees of clarity, the processes of authorship, transcription and annotation as they might have been immediately or indirectly manifested in one document, the copy text, that had existed for a short or long period of time, and had accrued varying degrees of alteration. The factor of time is important, because the longer a play had been available for performance the more likely it is that it would have been transcribed once or more, and the longer a manuscript was in use the more alterations might have been marked into it. Though time allows for a linear development of the text from one manuscript to another, or within a single manuscript, it also allows for divergence. The example of Hamlet has already illustrated this: it has been suggested that there was a main line of development from Q2 to the Q1 antecedent and thence to F1; and that there was, additionally, a branch line of development from the Q1 antecedent to the shortened version from which Q1 itself was printed. In other words, a diagram showing the lines of development (a ‘stemma’) would show a bifurcation whereby two separate versions develop from the Q1 antecedent, one leading towards Q1, the other to F1. There are other plays whose texts may be understood in terms of a non-linear relation between the surviving texts. Divergence of this kind further undermines the dichotomy between ‘foul papers’ and ‘promptbook’ because that dichotomy is grounded in a linear model of textual progression from A to B (Jowett 2017a).

IN SUMMARY The New Bibliography led towards a schematic description of Shakespeare’s texts in which quartos were printed either from pre-theatrical ‘foul papers’, in which case the quarto was ‘good’ in the sense of being close to Shakespeare’s hand, or posttheatrical reconstructions, in which case the quarto was ‘bad’ in the sense of being very poorly transmitted.36 The ‘promptbook’ would be based on a transcript by a

THE EARLY PRINTED TEXTS OF SHAKESPEARE

87

scribe or by the author himself (a ‘fair copy’) and annotated by a company bookkeeper. This document was withheld from the stationers because it was the theatre company’s key document. As has been argued, it is no longer possible to maintain a rigid dichotomy between ‘foul papers’ and ‘promptbook’; a more fluid model is required. Some ‘good’ Shakespeare quartos show evidence of transcription, or of theatre annotation such as duplicated stage directions and actors’ names (though others do not). The term ‘good’ turns out merely to reflect normality, a normality that encompassed many individual situations and that required no name. In contrast, the concept of a derivative and poorly transmitted text cannot be relinquished. Working in the early 1620s, the Folio editors were able to make use of the printed quartos, but preferred to print either directly from a manuscript (as was necessary if the play had not previously been printed) or with reference to such a manuscript. As the theatre ‘promptbook’ was usually retained by the company, either it was copied (as particularly by Ralph Crane), so that the new manuscript could be delivered to the printers, or readings were taken from the theatre playbook and copied into a copy of a quarto edition of the play in question. This account has generally stood the test of time. It is indeed a better informed picture than that for the quartos, because the quartos themselves provide a species of material evidence as to the copy for the Folio that is lacking for the original quartos. This description raises a question regarding terminology. The New Bibliographers’ ‘fair’, ‘foul’, ‘good’ and ‘bad’ were useful for their bold simplicity, but their stark dichotomies no longer seem adequate. The following terms are offered by way of conclusion as more flexible alternatives. Paradoxically, they are confined to manuscripts, as is necessary because the character of the printed text is defined by the manuscript from which it was printed. Only a disrupted text would lead to a ‘bad quarto’, but the existence, coherence and extent of the category have been so widely debated that the neutral term ‘short quarto’ is to be preferred instead.

APPENDIX: DEFINITIONS Author’s draft: manuscript leading to a printed text showing signs of unevenness of a kind that would result from the author’s process of composition. In contrast with the New Bibliographers’ ‘foul papers’, such a manuscript may (or may not) have received light annotation for the theatre. It would have been available for use as printer’s copy if a transcript had been made. Author’s transcript: the ‘authorial fair copy’ of the New Bibliography. It would, however, routinely be annotated to become a playhouse manuscript. If it became the key theatre document of the play, it must have been later superseded by a new copy in order for it to be released to the printers. Playhouse manuscript: Werstine’s inclusive term, encompassing authors’ manuscripts of plays and all transcripts of them prepared for the professional theatres, with the one proviso, grounded in the examples of surviving manuscripts, that they must have been at least lightly annotated for the theatre. Though copying a whole play is time-consuming, there would sometimes have been a fresh transcript if the play was heavily revised, and presumably in other circumstances too.

88

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Official playbook: the specific playhouse manuscript that bears the licence to perform from the office of the Master of the Revels. Literary transcript: a copy of a play prepared for use outside the theatre, whether for presentation to a private patron or for delivery to the printers for an edition. Disrupted text: a text that has been compiled in one way or another through transmission involving performance instead of manuscript-to-manuscript copying, and so characterized by the effects of mis-hearing, hasty transcription by shorthand and fast-writing methods, or lapse of memory. Such a text might require developing by a reviser from notes in order to constitute a continuous text, but the outcome will nevertheless be a short form of the play.

NOTES 1. The collaborative Sir Thomas More, which survives as a manuscript held in the British Library (Harley MS 7368), is an exception. 2. ‘Small format’ includes quartos and the few Shakespearean examples of octavos. 3. Arguably, Richard II belongs to this category, as the 1608 quarto is the first to print what it calls the ‘Parliament Scaene’ (title page, sig. A1r), the deposition of Richard as omitted from earlier quartos and printed in a longer text in F1. But this edition is otherwise a reprint. There are a few still more localized exceptions, such as a few lines in 3 Henry VI (5.6.90–2) that vary significantly between the octavo of 1595 and the quarto of 1619. 4. In both cases, a copy of Q1 was available to the printers of Q2, but its use was strictly limited and does not qualify very far the point made here. 5. The Tempest, The Two Gentlemen of Verona, Measure for Measure, Comedy of Errors, As You Like It, The Taming of the Shrew, All’s Well that Ends Well, Twelfth Night, The Winter’s Tale, King John, 1 Henry VI, All Is True (as Henry VIII), Coriolanus, Timon of Athens, Julius Caesar, Macbeth, Antony and Cleopatra and Cymbeline. 6. Q1 is itself a reprint, but Q0 survives only as a fragment of part of the play (sheet C). 7. The minor exception is the mid-line comma separating the vocative, which comes and goes. Elsewhere, more significant restorations of the original reading also occasionally happen, either by deliberate correction of error or by coincidence. 8. Compositors routinely standardized the spellings in their copy to their own preferences: see Howard-Hill (2006) and Jowett (2015). But compositors might leave unfamiliar forms intact. 9. For the publisher as editor, see Massai (2007). 10. For the concept of error, see Jowett (2017c: 1.xlix–lix). 11. In this chapter, references are to the original-spelling Critical Reference Edition of The New Oxford Shakespeare (Taylor et al. 2017). 12. Technically this is probably a ‘minim error’: the handwritten letters of ‘imi’ are made up of five uninterrupted ‘minim’ strokes, making the error easy. But accidental

THE EARLY PRINTED TEXTS OF SHAKESPEARE

89

omission of ‘i’ by a compositor is also possible. In either case, the fact that ‘smiles’ makes superficial sense in context makes the error harder to notice. 13. It is based on the text as subsequently set in F1, making the assumption that the Folio compositor working from this hybrid marked-up text would have set the words accurately, and would have made unprompted changes to the detail of punctuation and spelling. 14. For the theatrical origin of act divisions, especially in plays written before the King’s Men began playing at the Blackfriars theatre in c. 1608 and (inferably) performed there, see Taylor (1993a). Scene divisions (as exampled in the quoted passage) present a complex issue that invites further study. 15. For a full collation of quartos and Folio, see Smidt (1969). For the case for Q1 as copy text, see Jowett (2000). For an explicitly argued defence of F1 as copy text, see Siemon (2009). Elsewhere, the opposite kind of mixed copy also occurs: the edition set from manuscript with occasional consultation of a printed edition. 16. Two of the Folio-only passages may have been written into the quarto copy but deleted for reasons of censorship; see Jowett and Taylor (1987). The others were evidently added later by Shakespeare in revising the play. 17. The case for authorial revision is most extensively developed in Taylor and Warren (1993); see also Foakes (2008). Those arguing against include Clare (1995), Knowles (1995 and subsequent articles) and, more recently, Vickers (2016). For a critique of Vickers, see Syme (2016). 18. This admittedly impressionistic judgement is based on the less ambiguous wordorder of ‘heart receiues from hence’, and Q1’s avoidance of the obscure chiasmus to which I draw attention. Troilus and Cressida is an unusual example of divergent development, whereby in some passages Q1 and in others F1 is more likely the revised text. 19. A possible exception is ‘in his instruments / In’, where most editors omit the first ‘in’. But it may be allowed to stand as a spelling of ‘e’en’. 20. For punctuation I follow the Arden third edition of both texts (Thompson and Taylor 2006). Though Thompson and Taylor adjust some of Q1’s lineation, the original line-breaks are here retained. There are no special spacing difficulties. The printer James Roberts’s compositors were perfectly capable of setting verse as in their copy, and even set ‘drunke’, as a turn-up in order to preserve its line-endings. Q1 almost certainly reflects the manuscript’s lineation. 21. The syllable count is: 8, 10, 14, 10, 8, 6, 8, 12, 10. 22. This is not to deny that the passage shows basic literary invention. 23. Possibly earlier, though non-dramatic, is Sonnet 145, which is also highly regular in metre, though composed in iambic tetrameter. 24. The passage also offers a vivid neologism in ‘sluggardized’, evidence of a fertile Shakespearean creativity lacking in the diction of Q1 Hamlet where it differs from the other texts. For quantification of the regularity of Shakespeare’s early plays as compared with his later practice, see Wright (1988: 292–5).

90

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

25. It will be noted too that early Shakespeare plays are not conspicuously short. 26. Though Taylor has since reconsidered the argument, its logic is compelling. The sequence [Q1 > F1 > Q2] can be excluded because Q1 also has occasional affinities with Q2 as against F. 27. See the example 3.4.61–9, with the cuts as discussed in Hibbard (1987: 106–7), and variants in the following lines such as ‘And there I see such blacke and greeued spots / As will leaue there their tin’ct’ (Q2), ‘And there I see such blacke and grained spots, / As will not leaue their Tinct’ (F1). 28. This paragraph is based on Jowett (2017b: 1117). 29. The overall evidence of stage directions is complicated and in some ways contradictory, but the details cited rise above this ambiguity. 30. If Q1 contains any recollections of the old Hamlet as written by another dramatist (most likely Thomas Kyd), such mergings are entirely to be expected of a text that has undergone a disruption of word-by-word copying from a written antecedent. See Erne (2001: 146–56). 31. Compare Taylor (1987: 398): ‘in every structural respect Q1 is closer to F than to Q2’. 32. For Middleton as adapter of Folio Measure for Measure, Macbeth, and perhaps All’s Well that Ends Well and Titus Andronicus, see Taylor and Loughnane in The New Oxford Shakespeare Authorship Companion (2017), and Authorship Companion more generally. 33. The standard study identifying Crane’s characteristics and their presence in Shakespeare texts is Howard-Hill (1972). 34. Examples in authorial drafts in the manuscript of Sir Thomas More include: unusual spellings witnessed in other texts by the same author such as ‘scilens’ for ‘silence’, ‘deule’ for ‘devil’, ‘elamentes’ for ‘elements’ and ‘a leven’ for ‘eleven’ (Shakespeare); a ‘false start’ of twelve lines replaced by an expanded rewriting (Thomas Heywood); and irregular speech prefixes ‘Lincolne’, ‘Linco’, ‘Lin’, and ‘Linc’ (Shakespeare). In Heywood’s The Captives, variant authorial speech prefixes include ‘Ashbur’, ‘Ashb’, ‘Thomas Ash’, ‘Thomas’, ‘Ashburne’ (in a single scene); examples of authorial stage directions in this play include ‘Etc’, ‘all wett as newly shipwracke and escapt the ffury off the Seas’, ‘and his Lady etc Dennis and others’, ‘wth som ffollowers’, ‘a noyse or tmult’, ‘Eather strykes him wth a staffe or Casts a stone’. 35. But compare Gurr (1984: 9–10). 36. Compare Jenkins: ‘a “bad” quarto, one, that is to say, whose text, deriving from performance, lacks a direct manuscript link with what the author wrote’ (1982: 19).

REFERENCES Bourus, Terri (2014), Young Shakespeare’s Young ‘Hamlet’: Print, Piracy, and Performance, New York: Palgrave Macmillan. Clare, Robert (1995), ‘“Who is it that can tell me who I am?”: The Theory of Authorial Revision between the Quarto and Folio Texts of King Lear’, The Library, 6th ser., 17: 34–59.

THE EARLY PRINTED TEXTS OF SHAKESPEARE

91

Clegg, Cyndia Susan (1997), ‘“By the choise and inuitation of al the realme”: Richard II and Elizabethan Press Censorship’, Shakespeare Quarterly, 48: 432–48. de Grazia, Margreta and Peter Stallybrass (1993), ‘The Materiality of the Shakespearean Text’, Shakespeare Quarterly, 44: 255–83. Erne, Lukas (2001), Beyond ‘The Spanish Tragedy’: A Study of the Works of Thomas Kyd, Manchester: Manchester University Press. Foakes, R. A. (2008), ‘The Reshaping of King Lear’, in Jeffrey Kahan (ed.), ‘King Lear’: New Critical Essays, 104–23, New York: Routledge. Greg, W. W. (1954), The Editorial Problem in Shakespeare: A Survey of the Foundations of the Text, 3rd edn, Oxford: Clarendon Press. Greg, W. W. (1955), The Shakespeare First Folio: Its Bibliographical and Textual History, Oxford: Clarendon Press. Gurr, Andrew, ed. (1984), King Richard II, Cambridge: Cambridge University Press. Gurr, Andrew (1999), ‘Maximal and Minimal Texts: Shakespeare v. the Globe’, Shakespeare Survey, 52: 68–87. Hibbard, G. R., ed. (1987), Hamlet, Oxford: Oxford University Press. Howard-Hill, T. H. (1972), Ralph Crane and Some Shakespeare First Folio Comedies, Charlottesville, VA: University Press of Virginia. Howard-Hill, T. H. (2006), ‘Early Modern Printers and the Standardization of English Spelling’, Modern Language Review, 101: 16–29 Jenkins, Harold, ed. (1982), Hamlet, The Arden Shakespeare, London: Methuen. Johnson, Samuel, ed. (1765), The Plays of William Shakespeare, 8 vols, London: J. and R. Tonson et al. Jolly, Margrethe (2012), ‘Hamlet and the French Connection: The Relationship of Q1 and Q2 Hamlet and the Evidence of Belleforest’s Histoires Tragiques’, Paragon, 29: 83–105. Jowett, John (1987), ‘1 Henry IV’, in Stanley Wells and Gary Taylor, with John Jowett and William Montgomery, William Shakespeare: A Textual Companion, 329–32, Oxford: Oxford University Press. Jowett, John, ed. (2000), The Tragedy of King Richard III, Oxford: Oxford University Press. Jowett, John (2015), ‘Full Pricks and Great P’s: Spellings, Punctuation, Accidentals’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 317–31, Cambridge: Cambridge University Press. Jowett, John (2017a), ‘Exit Manuscripts: The Archive of Theatre and the Archive of Print’, Shakespeare Survey, 70: 117–26. Jowett, John (2017b), ‘Hamlet’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: William Shakespeare, The Complete Works: Critical Reference Edition, 2 vols, 1.1115–26, Oxford: Oxford University Press. Jowett, John (2017c), ‘Shakespeare and the Kingdom of Error’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: William Shakespeare, The Complete Works: Critical Reference Edition, 2 vols, 1. xlix–lxiii, Oxford: Oxford University Press. Jowett, John (2020), ‘The Origin of Richard Duke of York’, in Rory Loughnane and Andrew J. Power (eds), Early Shakespeare, Cambridge: Cambridge University Press, 235–60.

92

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Jowett, John and Gary Taylor (1987), ‘The Three Texts of 2 Henry IV’, Studies in Bibliography, 40: 31–50. Knowles, Richard (1995), ‘Revision Awry in Folio Lear 3.1’, Shakespeare Quarterly, 46: 32–46. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Mowat, Barbara A. (2005), ‘Q2 Othello and the 1606 “Acte to restraine the Abuses of Players”’, in Christa Jansohn and Bodo Plachta (eds), Varianten – Variants – Variantes, 91–106, Tübingen: Niemeyer. Orgel, Stephen (2002), The Authentic Shakespeare and Other Problems of the Early Modern Stage, New York: Routledge. Patterson, Annabel (1984), Censorship and Interpretation: The Conditions of Writing and Reading in Early Modern England, Madison, WI: University of Wisconsin Press. Pollard, A. W. (1909), Shakespeare Folios and Quartos: A Study in the Bibliography of Shakespeare’s Plays 1594–1685, London: Methuen. Siemon, James R., ed. (2009), King Richard III, The Arden Shakespeare, London: Methuen. Smidt, Kristian, ed. (1969), ‘The Tragedy of King Richard III’: Parallel Texts of the First Quarto and First Folio with Variants of the Early Quartos, Oslo: Universitetsforlaget. Stern, Tiffany (2013), ‘Sermons, Plays and Note-Takers: Hamlet Q1 as a “Noted” Text’, Shakespeare Survey, 66: 1–23. Syme, Holger (2016), ‘The Text is Foolish: Brian Vickers’s “The One King Lear”’, Los Angeles Review of Books, 6 September. Available online: http://tinyurl.com/zwupkes (accessed 13 November 2020). Taylor, Gary (1985), ‘The Fortunes of Oldcastle’, Shakespeare Survey, 38: 85–100. Taylor, Gary (1987), ‘Hamlet’, in Stanley Wells and Gary Taylor, with John Jowett and William Montgomery, William Shakespeare: A Textual Companion, 396–402, Oxford: Oxford University Press. Taylor, Gary (1993a), ‘The Structure of Performance’, in Gary Taylor and John Jowett, Shakespeare Reshaped, 1606–1623, 3–50, Oxford: Oxford University Press. Taylor, Gary (1993b), ‘Swounds Revisited: Theatrical, Editorial, and Literary Expurgation’, in Gary Taylor and John Jowett, Shakespeare Reshaped, 1606–1623, 51–106, Oxford: Oxford University Press. Taylor, Gary (1993c), ‘Post-script’, in Gary Taylor and John Jowett, Shakespeare Reshaped, 1606–1623, 237–47, Oxford: Oxford University Press. Taylor, Gary and Rory Loughnane (2017), ‘The Canon and Chronology of Shakespeare’s Works’, in Gary Taylor and Gabriel Egan (eds), The New Oxford Shakespeare Authorship Companion, 417–602, Oxford: Oxford University Press. Taylor, Gary and Michael Warren, eds (1983), The Division of the Kingdoms: Shakespeare’s Two Versions of ‘King Lear’, Oxford: Clarendon Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2017), The New Oxford Shakespeare: William Shakespeare, The Complete Works: Critical Reference Edition, 2 vols, Oxford: Oxford University Press. Thompson, Ann and Neil Taylor, eds (2006), ‘Hamlet’: The Texts of 1603 and 1623, The Arden Shakespeare, London: Thomson Learning.

THE EARLY PRINTED TEXTS OF SHAKESPEARE

93

Urkowitz, Steven (1988), ‘Good News about Bad Quartos’, in Maurice Charney (ed.), ‘Bad’ Shakespeare: Revaluations of the Shakespeare Canon, 189–206, Rutherford, NJ: Fairleigh Dickinson University Press. Vickers, Brian (2016), The One ‘King Lear’, Cambridge, MA: Harvard University Press. Werstine, Paul (1999), ‘A Century of Bad Quartos’, Shakespeare Quarterly, 50: 310–33. Werstine, Paul (2013), Early Modern Playhouse Manuscripts and the Editing of Shakespeare, Cambridge: Cambridge University Press. Wright, George W. (1988), Shakespeare’s Metrical Art, Berkeley, CA: University of California Press.

CHAPTER 2.3

Shakespeare’s early modern books: Printing, paratext and text EMMA SMITH

The quarto edition of The Tragedy of Othello the Moor of Venice (1622) appeared the year before the First Folio collection entitled Mr. William Shakespeares Comedies, Histories, & Tragedies. The stationer Thomas Walkley, who had already published King’s Men plays by Francis Beaumont and John Fletcher, prefaced the play with a short epistle: ‘The Stationer to the Reader’. Walkley sets out in his letter some interesting assumptions about play publishing, most of which do not appear to have been entirely true for the previous three decades of Shakespeare in print. But, as new digital resources make the study of such paratextual material much more accessible, Walkley’s epistle offers a vantage point from which to survey some unexpected continuities, as well as divergences, across Shakespeare’s early modern books.1 In creating his book of Othello, Walkley raises questions about readers and their expectations, about the market for playbooks in early modern England and about Shakespeare’s developing reputation as a print author. He also foregrounds the role of the stationer, rather than the author, in creating Shakespeare’s books. Roger E. Stoddard’s observation bears repeating: ‘whatever they may do, authors do not write books. Books are not written at all. They are manufactured by scribes and other artisans, by mechanics and other engineers, and by printing presses and other machines’ (1987: 4). This chapter focuses on the interventions of those involved in the transformation of Shakespeare’s manuscript plays and poems into printed books, with a particular focus on paratextual material. The ‘paratext is what enables a text [here, a play] to become a book’, as Gérard Genette put it in his foundational work on the threshold material that frames the literary text (1997: 1). As a remediation of written text by human and mechanical agents, printing inevitably intervenes in the text rather than simply reproducing it. Because we have no certain witnesses to Shakespeare’s literary writing before it reached print (except, perhaps, for the manuscript of ‘The Book of Sir Thomas More’), it is impossible to assess the extent of this transformation. Fredson Bowers’ famous metaphor imagined that the well-trained bibliographer might ‘pierce this veil of the printing process

SHAKESPEARE’S EARLY MODERN BOOKS

95

and … restore, however imperfectly, the authority of the manuscript’ (1950/1: 62), implying that print is a distinct layer on the authentic authorial text that can be expertly removed. Modern textual scholarship has tended instead to embrace print as intrinsic to the transmission of Shakespeare’s works, and thus inseparable from them (de Grazia and Stallybrass 1993). There are excellent accounts of printing and book production in the period (Maguire 1999; Werner 2019) and from the period (Moxon 1958). What is important here is not the details of a complex technical and economic industrial process, but to emphasize the ways that printing transmuted, rather than simply reproduced, its textual objects. In their influential article, ‘The Materiality of the Shakespearean Text’, Margreta de Grazia and Peter Stallybrass advocate a new attention to ‘old typefaces and spellings, irregular line and scene divisions, title pages and other paratextual matter, and textual cruces’ as features of early texts ‘to be looked at, not seen through’ (1993: 256–7). It is entirely relevant to their argument that these features can largely be attributed to the printing house rather than the author. Printshop workers needed to decipher their copy manuscript or markedup printed text and comprehend its intentions, to decide on appropriate print font and layout to communicate these to readers in a cost-effective manner, without wasting paper, and to translate these decisions deftly via individual pieces of type onto a composing stick and thence to the page-sized frame. They thus brought their professional expertise and habits of working to unfamiliar written material, shaping the copy according to existing mental templates for print publication. The choice of roman, italic, or blackletter font, just like the decision to print in small octavo or quarto, or larger folio format, conveyed meanings about the status, affinities and imagined use of the printed text (Bland 1998; Galbraith 2010). Once the material was typeset, the press needed to be properly operated to produce clearly inked pages, which were dried and then collated into gatherings using the information in the catchword and signature notation system. Proof corrections were applied while the presses were operating, meaning that extant texts are often made up of a random combination of corrected and uncorrected pages. Books were sold as a gathering of leaves, stab-stitched at the left-hand margin; buyers, like Sir John Harington (Nelson 2005) or Sir Edward Dering (Lennam 1965), both avid early modern playbook purchasers, might choose to have collections of quarto plays bound to make them more durable, but survival rates for quarto texts suggest that many copies were used as essentially disposable consumer objects. Much higher survival rates for the First Folio attest as much to its increased physical durability (its size meant it was likely to be stored in a book chest or on a library shelf, rather than carried in a pocket) as to its higher cultural value (the folio format was associated with works of learning and divinity: Galbraith 2010).

OTHELLO 1622 Thomas Walkley was a London publisher and bookseller with a connection to the King’s Men, for whom he published a number of dramatic texts. In 1622 he employed the print shop of Nicholas Okes to print Othello, and presented it to the

96

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

reading public with his own prefatory epistle. Walkley suggests that ‘To set forth a booke without an Epistle, were like to the old English prouerbe, A blew coat without a badge’ (Shakespeare 1622: sig. A2r): the reference is to servants’ livery without the mark of a master, and perhaps speaks obliquely to the play’s own particular preoccupations with social status and service (see for example Iago’s ‘We cannot all be masters, nor all masters / Cannot be truly followed’, 1.1.42–3). But of the twenty-two Shakespeare titles in print before 1622, many in multiple editions, only a single play, one of the two 1609 issues of Troilus and Cressida, and the narrative poems Venus and Adonis and Lucrece have an epistle (the poems are dedicated to an influential patron in the conventional style of printed lyric poetry). This infrequency does not suggest that a book without an epistle is unusual or unsatisfactory. Secondly, Walkley announces that he takes on himself the work of writing the epistle, ‘the Author being dead’ (sig. A2r): the assumption that Shakespeare as author would write the epistle is true of the extant poetry dedications, which are framed as a personal and affective address, and signed with his name, but not apparently the case for the Troilus example, which is unsigned and talks about the author in the third person. Nor did the living Shakespeare apparently trouble to write any epistle for his other works. Next, we learn that Walkley’s purpose is not to commend the play, ‘for that which is good, I hope euery man will commend, without intreaty’ (sig. A2r): this use of occupatio leaves it rather unclear what, then, the aim of the epistle might be. Dedicatory letters in Walkley’s contemporaneous play publications do not fill in this interpretative gap, since they tend to be more obviously functional. In 1619, for example, Walkley dedicated Beaumont and Fletcher’s A King and No King to the ‘approbation and patronage’ of Sir Henry Neville (Beaumont and Fletcher 1619: sig. A2v). In the same year as Othello, Walkley had written ‘To the Reader’ of the second edition of Beaumont and Fletcher’s Philaster, acknowledging the deficiencies in the first edition. His epistle reassured those readers who had already appreciated Philaster and his love Arethusa despite the ‘dangerous and gaping wounds, which they receiued in the first Impression’ that ‘assuredly they will now finde double fauour, being reformed, and set forth suteable, to their birth, and breeding’ (Beaumont and Fletcher 1622: sigs. A2r–v). In each of these equivalent publications, Walkley’s epistle has some specific work to do, in contrast to his curiously purposeless preface to Othello. Walkley even admits that his Othello epistle is in any case unnecessary, ‘because the Authors name is sufficient to vent his worke’. Shakespeare’s name is indeed prominent on Othello’s title page, and it had been routinely added to the title pages of his work in print since the late 1590s (and sometimes added to works not now, or perhaps ever, believed to be his), suggesting that Walkley’s claim is correct. But the fact that no new Shakespeare play had been printed since the popular Pericles, first published in 1609, might compromise his confidence somewhat, as indeed might the very existence of the epistle itself. Walkley’s unusual (for Shakespeare quartos) textual intervention into the published playbook of Othello was written at a pivotal moment in Shakespeare’s early print history. This last new quarto was produced just as the First Folio was making its slow progress through the printshop of William and Isaac Jaggard at the Half-Eagle and Key in the Barbican. It was the point at which Shakespeare’s

SHAKESPEARE’S EARLY MODERN BOOKS

97

print reputation was turning from the individual works of the period from 1593 to the collected work format that would dominate seventeenth-century print history. Paratextual material serves ‘to make present, to ensure the text’s presence in the world’ (Genette 1997: 1). Walkley’s preface offers the opportunity to review earlier and later Shakespearean paratexts and their attendant presence. In this chapter I will use the term ‘paratext’ via Thomas L. Berger and Sonia Massai’s working definition for their collection of paratextual material for the drama before 1642: ‘the term “paratext” is used in this edition to refer to all the extra-dramatic texts, such as title pages, dedications, addresses to the reader, lists of dramatis personae, prologues and epilogues, stationers’ notes and errata lists’ (2014: 1.xi). A brief comparison with Ben Jonson fleshes out many of these categories. Volpone (1607) is often cited as a watershed in playbook publication because of the range and extent of its paratextual material. In the edition published by Thomas Thorpe, there are sixteen pages of prefatory material after the title page of the play, including a dedication to the universities of Oxford and Cambridge and a lengthy epistle to them, several pages of commendatory verses in Latin and English praising the author and his work, a list of ‘Persons of the Comoedye’ (sig. A4r), an acrostic Argument and a Prologue. By contrast, the reader of Shakespeare’s King Lear, printed the following year, turns the title page over immediately to find the play’s opening scene, in which Gloucester and Kent discuss the King’s behaviour and Edmund’s nativity. The example is indicative. Print Shakespeare is generally short on many of the paratextual elements that were becoming common to playbook publication over the period. No Shakespeare play before the third quarto of The Merchant of Venice (1637) has a list of dramatis personae, although there are some in the Folio; no early work has an errata list; there are no dedications beyond the poetry, and those are often cryptic (the contested interpretations of ‘to the onlie begetter of these insuing sonnets Mr. W.H.’, for example: Shakespeare 1609b: sig. A2r); there are only a couple of addresses to the reader or stationers’ notes, and relatively few plays carry prologues or epilogues. As a result of these absences, the most important Shakespearean print paratext, both qualitatively and quantitatively, is the title page.

QUARTO TITLE PAGES: GENRE Title pages for Shakespeare’s works in print during his own lifetime offer framing information for the books’ intended readership, tracing coordinates of theatrical and literary provenance, authorship, descriptive titling and genre, but not always all at the same time. Perhaps the most common element is an extended title. The First Folio’s allocation of plays to the three genres highlighted in its title – Mr. William Shakespeares Comedies, Histories, & Tragedies – is often cited as one of its most important critical interventions, and commonly forms the starting point for an analysis of Shakespearean genre (Snyder 2001). In fact, most of the nineteen plays already published in quarto format had already had an extended, descriptive title including a generic attribution. For example, the Folio lists ‘Romeo and Juliet’ under the heading of ‘Tragedies’, but quarto publications in 1597 (‘excellent

98

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

conceited tragedie’) and in 1599 and subsequent editions (‘excellent and lamentable Tragedie’) had already pre-empted genre as an important aspect of paratextual framing. Only a few quarto plays were not explicitly identified by genre, including The Second part of Henrie the fourth (1600) and The late and much admired Play, Called Pericles, Prince of Tyre (1609 and subsequent editions), but the majority were – a distribution matched by the general presentation of playbooks in the period (see Farmer and Lesser 2007). Generic affiliations may well have served evaluatively for specific book buyers, whose personal preference might have been for tragedies or comedies. On many title pages, these descriptions are attached to specific adjectives of value or encouragement, such as ‘pleasant Conceited Comedie’ (Love’s Labour’s Lost, 1598); ‘most pleasaunt and excellent conceited Comedie’ (The Merry Wives of Windsor, 1602); ‘Famous Historie’ (Troilus and Cressida, 1609); ‘most excellent Historie’ (The Merchant of Venice, 1600; a reminder that the word ‘history’ had not entirely settled to its modern meaning of a story based on chronicle sources). Generic attributions are further emphasized throughout many playbooks by being part of the running header at the top of each page. While the Folio retains and foregrounds the genres in its Catalogue, it strips out the epithets and plot details from the quarto titles. Thus the account of the Roman patriarch is called Titus Andronicus in the Folio, rather than The most lamentable Romaine Tragedie of Titus Andronicus (1594), and the parallel account of the British patriarch is King Lear rather than the True Chronicle Historie of the life and death of King Lear and his three Daughters (1608). These paratextual variants have potentially substantive effects on what Hans Robert Jauss has called the readerly ‘horizon of expectation’ (2000: 132). Naming the play after the king and the three daughters is a distinctly different, folkloric framing from the tragic singularity cued by the Folio’s pared-back title. The effect of title revision is clear in numerous examples: the difference between A Most pleasaunt and excellent conceited Comedie of Syr Iohn Falstaffe, and the merrie Wiues of Windsor (1602) and the Folio’s more familiar The Merry Wiues of Windsor may register Falstaff’s fallen place in the theatrical hierarchy by the 1620s, but it also introduces a substantially different play in terms of gender and organization. The differences in expectation cued by The Second part of King Hen. Sixt (the title in the Folio) and The First part of the Contention betwixt the two famous Houses of Yorke and Lancaster, with the death of the good Duke Humphrey (1594, in quarto) are clear and numerous. Folio titling of history plays prioritizes monarchical and teleological sequence in ways that the more disparate individual naming conventions of the quartos do not. Typographic conventions for early modern play title pages make it difficult to be sure what parts of the page should be cited as titles, and what parts are explanatory or synoptic. For example, the title page of Richard III in quarto (1597) reads: ‘The Tragedy of King Richard the third. Containing, His treacherous Plots against his brother Clarence: the pittiefull murther of his innocent nephewes: his tyrannical vsurpation: with the whole course of his detested life, and most deserued death’. The list is deliciously histrionic – somewhat misleadingly so – and may perhaps be directed towards readers who recall the play in the theatre. Tiffany Stern has suggested that some title pages may preserve the advertising content of the playbills

SHAKESPEARE’S EARLY MODERN BOOKS

99

posted daily around the city informing audiences about the programme (2006: 85). The relation of descriptive title page content to performance is an underexplored one (Farmer and Lesser 2000). The quarto relishes Richard’s misdeeds, establishing the play as a series of villainous acts in the manner of Marlowe’s popular The Jew of Malta and quite different from that framed by the First Folio. Here the title of the play is elaborated but with a political telos rather than a theatrical one: ‘The Tragedie of Richard the Third: with the Landing of Earle Richmond, and the Battell at Bosworth Field’. The quarto focuses on the amoral appeal of Richard’s conduct which ends only with his ‘deserued death’; the Folio on the historical deus ex machina who is bringing to Bosworth the inexorable inevitability of historical succession. Where these play titles come from is not entirely clear: both the titling practices of Philip Henslowe’s Diary (Foakes 2002) and the evidence from Stationers’ Register entries suggest that plays might circulate under variant titles. Even for the quartos published during Shakespeare’s lifetime, however, they are unlikely to be authorial in provenance. The extended description of The Merchant of Venice is instructive: the quarto summarizes the interrelated plot strands in promising ‘the extreame crueltie of Shylocke the Iewe towards the sayd Merchant, in cutting a iust pound of his flesh: and the obtayning of Portia by the choyse of three chests’. Setting aside the gruesomely misleading claim that the pound of flesh is indeed cut, the key word here is ‘chests’. The language of the play prefers the synonym ‘caskets’ for the plot line of Portia’s father’s gold, silver and lead boxes to test potential suitors, suggesting that the author of the title page and the author of the scenes are not thinking in the same linguistic register (for a counter view, see Stern 2006: 86). Similarly, critics have felt it unlikely that the ‘argument’ outlining the plot of Lucrece was written by Shakespeare (Tolbert 1950). One way in which the study of early modern paratexts has developed from Genette’s framework is in challenging his assumption that they are ‘authorial or more or less legitimated by the author’ (Genette 1997: 2).

AUTHORSHIP If it is unclear whether title pages were authorial in provenance, we can trace the ways in which Shakespeare’s printed texts presented his authorial status. Shakespeare’s narrative poems are the first of his works to be published and the first to bear his name – not on their title pages, but as textual signatures to dedications. At the end of the letter ‘dedicating my vnpolisht lines’ and ‘the first heire of my inuention’ to Henry Wriothesley, Earl of Southampton, that prefaces Venus and Adonis (1593 and subsequent editions), the poet signs ‘Your Honors in all dutie, William Shakespeare’ (Shakespeare 1593: sig. A2r). Again, the following year, the ‘vntutored Lines’ of Lucrece bear a similar letter of dedication to Southampton, signed by Shakespeare (Shakespeare 1594, sig. A2r). Despite, or perhaps because of, the fame attached to Shakespeare’s name as the author of these popular publications, playbooks published early in his career do not carry any indication of authorship. There is no attempt to attach the author of Titus Andronicus (1594) or The True Tragedie of Richard Duke of York (1595, better known by its Folio title of 3 Henry VI) to the popular author of Venus and Adonis. The place in the title pages’ mise-en-page where later readers

100

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

might expect an indication of authorship is often taken up with information about performance: the early plays seem to be authored by the company rather than the playwright. By 1598 Shakespeare’s name became attached to his plays in print. Love’s Labour’s Lost claims authorship, and in the same year the second quarto of Richard III and the second and third quartos of Richard II also added Shakespeare’s name to otherwise unchanged title pages: changing the title page between editions suggests a degree of deliberation that is always worth investigating. Quite what status Shakespeare’s name has over these middle years of his career is difficult to pinpoint (Bruster 2012). On the one hand, not all subsequent quarto editions are marked with Shakespeare’s authorship, suggesting that, contrary to Thomas Walkley’s remarks on Othello, attribution is not a necessary part of marketing. Even after Shakespeare’s name has become more common on his title pages, there is no attempt to register the popular Romeo and Juliet within his canon (quarto editions of 1599 and 1609, and one of the two issues of Q4, probably dated 1623 (see Hailey 2007) are all anonymous). On the other hand, a small but concerted attempt to claim that non-Shakespearean texts were by Shakespeare suggests that his authorship itself had a commercial or cultural value. As Lukas Erne puts it in his important study of Shakespeare in print based on quantitative data, ‘Shakespeare’s Juliet asks, “What’s in a name?”, and stationers seem to have realized that one thing that is in it is money, and that a name to make money with was “Shakespeare”’. Erne notes that there are ten editions of seven different plays claiming to be by Shakespeare, against the important backdrop that ‘no other dramatist had any playbooks misattributed to him during the same period’ (Erne 2013: 59, 56). For example, The London Prodigal (1605) alleges on its title page that it is ‘by William Shakespeare’. So too does A Yorkshire Tragedy (1608), also entered on the Stationers’ Register as by Shakespeare, the Pavier Quarto of Sir John Oldcastle (1619, false title page date 1600) and the reprinted edition of The Troublesome Reign of King John, claimed to be by ‘W. Sh.’ in 1611, and ‘W. Shakespeare’ in 1622. Many of these plays were added to the Third Folio by Philip Chetwinde in 1664 (Kirwan 2015). The size of the type in which the name was printed may also be relevant: Shakespeare’s name does seem to get bigger over time. The prominence of his authorial name as a title for the First Folio is anticipated in the 1608 quarto of King Lear, which has as its largest line at the top of the title page ‘M. William Shak-speare’. But this may in fact be less a triumphant assertion of authorial control than an anxious attempt to distinguish the work from a close sibling recently published, the anonymous King Leir (1605). Nevertheless, the emphasis on authorship is continued to the first page of the play text itself, where a half-title, ‘M. William Shak-speare His Historie, of King Lear’ (sig. B1r), precedes the opening words of the first scene. Part of what is emphasized at the opening of King Lear in the quarto, that is to say, is who it is by: as the characters discuss literal paternity, the playbook is preoccupied with naming its legitimating literary progenitor. The quarto text of King Lear has been subject to much independent and comparative analysis since it was first proposed that it represented an earlier, but still authorial, version of the play printed in the Folio, and so became exhibit A in theories of authorial revision (Taylor and Warren 1983). Other than its particular

SHAKESPEARE’S EARLY MODERN BOOKS

101

emphasis on the author’s name, the quarto does not directly allude to its own textual status. The famous and much contested claim by John Heminges and Henry Condell, that the Folio presents the play texts newly ‘cur’d, and perfect of their limbes’, does have its equivalence in a number of other quarto publications (Shakespeare 1623: sig. πA2v). Statements about improved or enlarged texts suggest that the assertion of a revised text was of value in presenting playbooks. Q2 of Hamlet (1604/5) gives no added information about the plot and no detail about performance provenance, but it does distinguish itself from the text printed in 1603 with the claim that it is ‘Newly imprinted and enlarged to almost as much againe as it was, according to the true and perfect Coppie’. The case of the different texts of Hamlet has also received much critical attention, facilitated by the existence of the variant quarto editions. The same is true for Romeo and Juliet, where the second quarto, published in 1599, differs considerably from the first (1597), and identifies itself on the title page as ‘Newly corrected, augmented, and amended’ (this is clearly of value as a stand-alone description since there is no attribution of authorship and therefore no claim of who has done this work of revision). Other claims of improvement exist where there is no prior textual witness for comparison, and so it is unclear whether the statement on the title page should be taken at face value. The first extant edition of Love’s Labour’s Lost (1598) is marked as ‘Newly corrected and augmented By W. Shakespere’: this may be a marketing ploy or testament to a lost earlier edition. One issue of the 1608 fourth quarto of Richard II carries a promise delivered on by the play text (although the plural ‘additions’ might overstate the extent of the previously unpublished material): ‘With new additions of the Parliament Sceane, and the deposing of King Richard.’ The case of Richard III is a different one: the third and subsequent editions claim to be ‘Newly augmented’, even though no substantial material has been added from previous editions. Similarly, the third, apparently largely unchanged edition of 1 Henry IV is ‘Newly corrected by W. Shake-speare’. This suggests that the claim of improvement – particularly of expansion or newness – was seen to be a useful strategy, but it also demonstrates that like many other claims on title pages, the claim of revision was doing narrative – or perhaps marketing – rather than documentary work. Paratexts need to be read as fictional, for their rhetorical effect rather than their testamentary accuracy. One other piece of information often included in the quarto title pages relates to performance conditions. Although there are no lists of actors involved in the plays, nor character lists, nor other paratextual material relating directly to the theatre, these markers of performance often tag the play to its first life on the stage, or serve as an ongoing reference point for the company’s work. Early play texts fossilize long-lost companies from the earliest years of the commercial theatres. The first quarto of Titus Andronicus, for instance, marks it ‘as it was Plaide by the Right Honourable the Earle of Darbie, Earle of Pembrooke, and Earle of Sussex their Seruants’, a designation that also attaches itself to the half-title printed on the first page of the text. By the 1600 edition the Lord Chamberlain’s Servants had been added to the list. The first quarto of Romeo and Juliet is consciously retrospective in recording performance: ‘As it hath been often (with great applause) plaid publiquely, by the right Honourable the L. of Hunsdon his Seruants’ (1597).

102

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

A handful of early play texts memorialize royal associations: Love’s Labour’s Lost is printed ‘As it was presented before her Highnes this last Christmas’ (1598), The Merry Wives of Windsor has been acted ‘Both before her Maiestie, and else-where’ (1602), and King Lear (1608) is presented ‘As it was played before the Kings Maiestie at Whitehall vpon S. Stephans night in Christmas Hollidayes’. This title page also takes the opportunity to mention the normal schedule of the players: ‘By his Maiesties seruants playing vsually at the Gloabe on the Bancke-side’. The majority of Shakespeare quartos note these regular arrangements and serve as a reminder of the Chamberlain’s Men’s regular public performances, although only occasionally by mentioning specific theatres. This may speak to what Michael Bristol anatomized as a ‘chronic tension between a more exclusive culture of the book and a more popular culture of performance’ in presenting Shakespeare in print (1996: x). There is, further, an element of retrospection or nostalgic drag on title pages as they are often reproduced for subsequent editions. Thus, the Chamberlain’s Men continue to appear on title pages after they have been renamed the King’s Men (for example in Richard III, 1605), just as subsequent editions claim newness for play texts that are no longer new: how far nostalgia, rather than novelty, might resonate with playbook buyers is an interesting question. Again, performance information may register aspiration or over-claiming rather than reportage. The first quarto of Hamlet (1603) has the unexpected claim that the play was performed ‘in the Cittie of London: as also in the two Vniversities of Cambridge and Oxford’, although it may be that this is more about aligning it with a potential young student market than about accurately recording its performance history (no other professional play of the period claims university performance). If these claims about performance serve in part as advertising for the ongoing theatrical offering of the company, it is perhaps surprising that there is not more emphasis on the new Blackfriars theatre from 1608 onwards. Blackfriars, with its higher prices and more exclusive clientele, might have been expected to overlap more with playbook buyers and readers than the more demotic Globe. The 1622 Othello quarto claims (as does the 1630 second edition) to have ‘beene diuerse times acted at the Globe, and at the Black-Friers’, and otherwise only the belated quartos of The Taming of the Shrew and Love’s Labour’s Lost (both 1631), and The Two Noble Kinsmen (1634) also mention Blackfriars. Playbooks by other writers, including Thomas Middleton and Francis Beaumont and John Fletcher, seem to have been more likely than those by Shakespeare to advertise their connections with the fashionable indoor theatre (Farmer and Lesser 2000: 88–92; Egan 2006).

PERFORMANCE AND MARKETING: THE CASE OF TROILUS AND CRESSIDA There is further ambivalence towards performance in the paratextual puzzle of Troilus and Cressida, which again highlights the strategic or rhetorical use of paratext. Two 1609 issues frame the text with quite different appeals to their reader. In one of the issues, Richard Bonian and Henry Walley advertise The Historie of Troylus and Cresseida. As it was acted by the Kings Maiesties seruants at the Globe.

SHAKESPEARE’S EARLY MODERN BOOKS

103

The second is differently titled: The Famous Historie of Troylus and Cresseid. Excellently expressing the beginning of their loues, with the conceited wooing of Pandarus Prince of Licia. Having removed the reference to performance on the title page, it presents a unique epistle, titled ‘A neuer writer, to an euer reader. Newes’. The epistle asserts that the play has never been performed: ‘Eternall reader, you haue heere a new play, neuer stal’d with the Stage, neuer clapper-clawd with the palmes of the vulger’ (Shakespeare 1609a: sig. ¶2r). Clearly, both paratexts cannot represent the truth about Troilus and Cressida on the stage: rather, the two issues represent a different marketing strategy. One is familiar, drawing on the common elements of the Shakespearean title page in listing author and theatre company. The other is experimental, pitching the play’s value to readers as a more elite commodity because it is unperformed. This strategy may anticipate the Folio’s relative underplaying of theatre discussed below. Many plays of the period were published with reference either to successful performance or undeserved failure. In the same year as Troilus, Pericles was described as ‘admired … As it hath been diuers and sundry times acted by his Maiesties Seruants, at the Globe on the Banck-side’; three years later another King’s Men’s play, Webster’s The White Devil, was prefaced with a defensive authorial explanation for its theatrical failure: ‘it wanted … a full and vnderstanding Auditory’ rather than the ‘ignorant asses’ present in the wintery Globe theatre (Webster 1612: sig. A2r). No other play publication tries to spin an ostensibly unperformed dramatic script to its readers in the way of Troilus and Cressida. What is also striking about the epistle to Troilus and Cressida is its repeated emphasis on the play as a comedy: this authors Commedies, that are so fram’d to the life, that they serue for the most common Commentaries, of all the actions of our liues, shewing such a dexteritie, and power of witte, that the most displeased with Playes, are pleasd with his Commedies. And all such dull and heauy-witted wordlings, as were neuer capable of the witte of a Commedie, comming by report of them to his representations, haue found that witte there, that they neuer found in them-selues, and haue parted better wittied then they came: feeling an edge of witte set vpon them, more then euer they dreamd they had braine to grinde it on. So much and such sauored salt of witte is in his Commedies, that they seeme (for their height of pleasure) to be borne in that sea that brought forth Venus. (Shakespeare 1609a: sig. ¶2r) This is noteworthy for several reasons. The play has not seemed to most readers to be a comedy, nor has its cynical form of imitatio been widely recognized as a ‘common commentar[y]’ on the action of our lives, nor indeed its antiheroic satire a particular display of, or prompt to, wit. The generic suggestion of comedy is at odds both with the title of ‘the Famous Historie’, and with the associations of the lovers with tragedy following Chaucer’s account of their story, in creating a horizon of expectation it seems likely to disappoint. Shakespeare’s comedies were perhaps the least likely of his plays to be in print in 1609, a point at which his print reputation was most securely built on repeated editions of the history plays

104

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

and on narrative poetry. Only five comedies – The Merry Wives of Windsor, Love’s Labour’s Lost, Much Ado About Nothing, The Merchant of Venice and A Midsummer Night’s Dream – had been printed by this point, so the emphasis on comedy does not seem to ally this book effectively with an existing market share. Again, the paratext offers a rhetorical display that apparently works to produce a particular reading and purchasing context for the play.

PROLOGUES AND EPILOGUES It is challenging to include prologues and epilogues as part of the ‘threshold’ material of the printed playbook, rather than intrinsic to the literary texts contained within it. Traditional criticism has tended to emphasize the synoptic qualities of these speeches, or to privilege their management of the relationship between character, actor and audience (in, for example, discussions of Prospero’s epilogue to The Tempest). Tiffany Stern argues instead that these bracketing gestures were ‘focussed, purposeful, and occasion specific: the first and the last in a series of playhouse “gifts” designed to make the critical audience more likely to accept a play’, and suggests that ‘a prologue or epilogue heralded a play in its freshest and so most fluid state – and a prologue or epilogue in a printed play therefore did likewise’ (perhaps analogous to the connotations of newness discussed earlier) (2009: 92, 82). Two features make prologues and epilogues particularly interesting when seen as paratexts. First is the frequency with which their presence varies across different editions, embodying in print that occasional character Stern attributes to performance history. Thus Romeo and Juliet, 2 Henry IV, Henry V and Troilus and Cressida all have variant prologues or epilogues in different early printed editions. Secondly, typographical features sometimes ally prologue or epilogue material with paratextual material rather than the playtext proper (see, for example, Folio The Tempest which prints Prospero’s epilogue in italics in a column alongside the ‘Names of the Actors’, signalling its distinctiveness from the rest of the spoken script of the play (Shakespeare 1623: sig. B4r). Italic type is common for these elements of the play, against roman type for the rest of the text, and sometimes the prologue or epilogue is set on a new page rather than running on with the play. These presentational clues identify prologue and epilogue material with paratextual or framing material rather than with the play proper, with implications both for more conventional paratext and for the reading of the play itself.

FIRST FOLIO PARATEXTS Early Shakespearean books, then, use paratexts to engage questions of authorship, marketing and theatre, shaping their text to readers but making relatively few direct addresses to them, and providing only occasional print acknowledgements of readers’ particular needs. To a greater extent than has, perhaps, been recognized, the paratexts of Shakespeare’s collected plays of 1623, often seen as a bibliographic departure from the habits of quarto publication, continue, rather than transform, these paratextual tendencies.

SHAKESPEARE’S EARLY MODERN BOOKS

105

There are pages of dedications, poems and lists between the Folio’s title page, with its engraving of Shakespeare, and the first play in the collection, The Tempest. Epistles to ‘the great Variety of Readers’ and to the Herbert brothers establish this collection within the parallel jurisdictions of commercial print and literary patronage (Massai 2012b); dedicatory poems praise the dead author and his living art; a list of the company actors from the earliest days of the Chamberlain’s Men in 1594 up to the present locate the plays’ origins in the theatre; there are prologues and epilogues and a few character lists appended to the end of plays. This preliminary paratextual material also, in Chris Laoutaris’s words, ‘stage[s] a gradual revelation of the personalities – we might think of them as a kind of “Shakespeare, Inc.” – who colluded in, collaborated towards and co-funded the creation of “Shakespeare” as successive generations would come to know him’: this is a posthumous enterprise primarily attributable to the syndicate of stationers, led by William and Isaac Jaggard and Edward Blount, who financed the publication (Laoutaris 2016: 49; Lesser 2004; Straznicky 2012). It is possible to identify three overlapping effects of the paratextual material in the First Folio: the reinvention of Shakespeare as a serious, readerly property; the construction of an ideal reader who can appreciate this literary work; and the commercial marketing of an expensive volume, half of which had previously been available in print to keen playbook buyers. Firstly, Shakespeare. From the carefully corrected engraving of Shakespeare on the title page (there are three states of the engraving with minor corrections each time, suggesting that getting it right was a priority for the publishers), to the collection’s exclusive emphasis on solo work with no mention of dramatic collaborators, the First Folio works to present a Shakespeare who is the prominent, unifying factor behind his work – even its main subject. It is a common bibliographic strategy to compare Shakespeare’s 1623 Folio with Ben Jonson’s 1616 Workes, and the difference between their title pages is instructive: despite his evident ‘bibliographic ego’ (Loewenstein 1985: 101), Jonson defers to a classical, architectural design as a frontispiece rather than a large-scale portrait (Kastan 2001: 63–5; Hooks 2014). The dedicatory poems emphasize authorial centrality, serving as eulogies and elegies juxtaposing literary survival against mortal decay (Maguire 2000): as Jonson puts it in his famous poem in the Folio preliminaries entitled ‘To the memory of my beloued The Author Mr. William Shakespeare: And what he hath left vs’, Shakespeare is ‘aliue still, while thy Booke doth liue’ (Shakespeare 1623: sig. πA4r). Other commendatory writers – Hugh Holland, James Mabbe and Leonard Digges – praise the author and his works. They appear to be men associated with the publisher Edmund Blount rather than with Shakespeare himself, or with his theatre, giving rise to questions about why they, rather than, say, a fellow playwright like Middleton, were chosen for the task (Taylor 2006: 66–8). In fact, the theatrical provenance of these texts is rather downplayed by the paratextual material. There is the list of ‘The Names of the Principall Actors in all these Playes’ (Shakespeare 1623: sig. πA5+1r), and there is a reminder that the plays ‘haue had their triall alreadie’ onstage (Shakespeare 1623: sig. πA3r), but there is no mention of the Globe or Blackfriars, and no sense that the volume serves,

106

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

as some of the quartos evidently did, to advertise the ongoing programme of the theatre company. Suppressing the works’ theatrical provenance may have been a deliberate and aspirant marketing strategy, along with printing the plays in the prestigious format of a folio volume, associated with bibles, learned books, and the study, and dedicating it to two earls, ‘the most noble and incomparable paire of brethren’ (Shakespeare 1623: sig. πA2r). Collectively, these decisions seem to have been aimed at establishing the volume as significant and literary, in contrast to the quarto ‘baggage books’ famously denounced by Thomas Bodley as unfit for a serious library (Erne 2013: 195). In tandem with the construction of a literary Shakespeare for readers is the construction and instruction of that readership, at a time when no folio volume of English dramatic works had been previously published. As Wolfgang Iser has it, ‘if communication between text and reader is to be successful, clearly the reader’s activity must … be controlled in some way by the text’: for Iser, focusing on realist novels, that control is exerted through narrative viewpoint, but when considering readers of print Shakespeare, we might well look to the paratexts for initiating such directives (Iser 1995: 23). The First Folio makes relatively few concessions to readers in terms of its organization and convenience. The catalogue of plays included has some pagination errors and also misses out one title, Troilus and Cressida (many extant copies have these corrected or the title added in manuscript, Smith 2016: 169–70). It organizes plays into the three generic categories named in the title, and certainly the section of ‘histories’ has undergone systematic revision of titles to produce a historical sequence out of the individual, disordered plays in quarto publication and, probably, performance (Smith 2015b: 27–30). Seven plays have character lists, added to the end of the play text, so there has been no organized attempt to provide readers with this navigational aid to the plays (Smith 2015a; Atkin and Smith 2014). Similarly, there has been no sustained work of editing or standardization with readers in mind. In their epistle ‘To the great Variety of Readers’, the King’s Men actors and friends of Shakespeare John Heminges and Henry Condell seem to suggest that Shakespeare’s works are for everyone, ‘From the most able, to him that can but spell’. But this idealized expansiveness is immediately undercut with the wry comment ‘We had rather you were weighd’: ‘the fate of all Bookes depends vpon your capacities: and not of your heads alone, but of your purses’. Heminges and Condell’s epistle serves as an advertising puff for their book rather than a description of its texts and its dead author, as emphasized by the repeated imperative ‘buy’: ‘what euer you do, Buy’ (Shakespeare 1623: sig. πA3r). This epistle thus amplifies the tendency already seen in quarto paratexts which undertake rhetorical or advertising, rather than descriptive, work. It is in this light that Heminges and Condell’s two substantive and influential claims about the works they have included in the Folio should be read. The first is their assurance that the collection is complete, ‘absolute in their numbers’. The second, is that these texts are authoritative: ‘(before) you were abus’d with diuerse stolne and surreptitious copies’, but these previously printed plays are now ‘offer’d to your view cur’d, and perfect of their limbes’, based on manuscripts of preternatural tidiness: ‘His mind and hand went together: And what he thought, he vttered with

SHAKESPEARE’S EARLY MODERN BOOKS

107

that easinesse, that wee haue scarse receiued from him a blot in his papers’ (sig. πA3r). Both statements have cast a long shadow over Shakespeare criticism and shaped editorial policies for centuries (Egan 2016), but, seen here in their context within this economically-inflected prefatory material, they have something of the sales pitch about them.

BRAND SHAKESPEARE? Genette argues that paratexts turn texts into books. We might add that books turn writers into what Roland Barthes dubbed, with an oppressive capital letter, the Author. From the early printed texts of the period from Venus and Adonis in 1593 to the First Folio thirty years later, ‘Shakespeare’ emerges as a cultural and commercial phenomenon, and begins the progress of market dominance, enhanced cultural value and increased brand recognition that has been explored by scholars of his reputation in later periods (Dugas 2006; Rumbold 2011). Douglas Brooks argues that in printing playtexts, publishers and theatre practitioners had to collaborate in ‘joint business ventures in which the primary function of the author was his capacity to give a commodity a minimum of recognition and value’ (Brooks 2000: 16). Paratexts register some of those business negotiations, recording the range of agents, from printers to theatre companies, involved in bringing Shakespeare into view. As Shakespeare notes in King Simonides’ bibliographic simile in Pericles, title pages negotiate the value of the text and its author: ‘To place upon the volume of your deeds, / As in a title-page, your worth in arms’ (2.3.3–4). More comparison is needed between Shakespeare and other literary texts in print to understand this worth in literary and commercial arms, and more attention to reprinted quartos as well as first editions in tracing Shakespeare’s changing value in the book marketplace would be valuable in studies of Shakespeare in print. In shaping Shakespeare’s works for an early modern readership, paratextual material is crucial to studies of authorial reputation, reader response and textual transmission: it can give us clues about the making of Shakespeare and the development of drama in print form, and challenge some of our assumptions about what was – and is – important about Shakespeare’s texts.

NOTE 1.

Freely available digital resources that have been particularly important in writing this chapter include: the Shakespeare in Quarto project from the British Library (digitized facsimiles of all the early play quartos), https://www.bl.uk/treasures/shakespeare/ homepage.html (accessed 13 November 2020); DEEP: Database of Early English Playbooks (Farmer and Lesser 2007); the Shakespeare Documented exhibition (title pages and other documentary evidence): http://shakespearedocumented.folger.edu (accessed 13 November 2020). All title pages are quoted from this resource. All quotations from the First Folio are taken from the Bodleian Library copy at http:// firstfolio.bodleian.ox.ac.uk (accessed 13 November 2020).

108

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

REFERENCES Atkin, Tamara and Emma Smith (2014), ‘The Form and Function of Character Lists in Plays Printed before the Closing of the Theatres’, Review of English Studies, 65: 647–72. Beaumont, Francis and John Fletcher (1619), A King and No King, London. Beaumont, Francis and John Fletcher (1622), Phylaster, London. Berger, Thomas L. and Sonia Massai, eds (2014), Paratexts in English Printed Drama to 1642, Cambridge: Cambridge University Press. Bland, Mark (1998), ‘The Appearance of the Text in Early Modern England’, Text, 11: 91–154. Bowers, Fredson (1950/1), ‘Some Relations of Bibliography to Editorial Problems’, Studies in Bibliography, 3: 37–62. Bristol, Michael D. (1996), Big-time Shakespeare, London: Routledge. Brooks, Douglas (2000), From Playhouse to Printing House: Drama and Authorship in Early Modern England, Cambridge: Cambridge University Press. Bruster, Douglas (2012), ‘Shakespeare the Stationer’, in Marta Straznicky (ed.), Shakespeare’s Stationers: Studies in Cultural Bibliography, 112–31, Philadelphia: University of Pennsylvania Press. de Grazia, Margreta and Peter Stallybrass (1993), ‘The Materiality of the Shakespearean Text’, Shakespeare Quarterly, 44: 255–83. Dugas, Don-John (2006), Marketing the Bard: Shakespeare in Performance and Print 1660–1740, Columbia, MO: University of Missouri Press. Egan, Gabriel (2006), ‘“As it Was, Is, or Will be Played”: Title-pages and the Theatre Industry to 1610’, in Peter Holland and Stephen Orgel (eds), From Performance to Print in Early Modern England, 92–110, Basingstoke: Palgrave Macmillan. Egan, Gabriel (2016), ‘The Provenance of the Folio Texts’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 68–85, Cambridge: Cambridge University Press. Erne, Lukas (2013), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Farmer, Alan B. and Zachary Lesser (2000), ‘Vile Arts: The Marketing of English Printed Drama, 1512–1660’, Research Opportunities in Renaissance Drama, 29: 77–165. Farmer, Alan B. and Zachary Lesser, eds (2007), DEEP: Database of Early English Playbooks. Available online: http://deep.sas.upenn.edu (accessed 13 November 2020). Foakes, R. A., ed. (2002), Henslowe’s Diary, 2nd edn, Cambridge: Cambridge University Press. Galbraith, Steven K. (2010), ‘English Literary Folios 1593–1623: Studying Changes in Format’, in John N. King (ed.), Tudor Books and Readers: Materiality and the Construction of Meaning, 46–67, Cambridge: Cambridge University Press. Genette, Gérard (1997), Paratexts: Thresholds of Interpretation, trans. Jane E. Lewin, Cambridge: Cambridge University Press. Hailey, R. Carter (2007), ‘The Dating Game: New Evidence for the Dates of Q4 Romeo and Juliet and Q4 Hamlet’, Shakespeare Quarterly, 58: 367–87.

SHAKESPEARE’S EARLY MODERN BOOKS

109

Hooks, Adam G. (2014), ‘First Folios: Jonson and Shakespeare’, in Robert DeMaria, Heesok Chang and Samantha Zacher (eds), A Companion to British Literature: Volume II Early Modern Literature 1450–1660, 281–93, Chichester: Wiley-Blackwell. Iser, Wolfgang (1995), ‘Interaction between Text and Reader’, in Andrew Bennett (ed.), Readers and Reading, 20–31, London: Longman. Jauss, Hans Robert (2000), ‘Theories of Genres and Medieval Literature’, in David Duff (ed.), Modern Genre Theory, 127–47, Harlow: Pearson. Jonson, Ben (1607), Volpone, or the Foxe, London. Kastan, David Scott (2001), Shakespeare and the Book, Cambridge: Cambridge University Press. Kirwan, Peter (2015), Shakespeare and the Idea of Apocrypha: Negotiating the Boundaries of the Dramatic Canon, Cambridge: Cambridge University Press. Knight, Jeffrey Todd (2013), Bound to Read: Compilations, Collections, and the Making of Renaissance Literature, Philadelphia: University of Pennsylvania Press. Laoutaris, Chris (2016), ‘The Prefatory Material’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 48–67, Cambridge: Cambridge University Press. Lennam, T. N. S. (1965), ‘Sir Edward Dering’s Collection of Playbooks, 1619–1624’, Shakespeare Quarterly, 16: 145–53. Lesser, Zachary (2004), Renaissance Drama and the Politics of Publication: Readings in the English Book Trade, Cambridge: Cambridge University Press. Lesser, Zachary and Peter Stallybrass (2008), ‘The First Literary Hamlet and the Commonplacing of Professional Plays’, Shakespeare Quarterly, 59: 371–420. Loewenstein, Joseph (1985), ‘The Script in the Marketplace’, Representations, 12: 101–14. Maguire, Laurie E. (1999), ‘The Craft of Printing (1600)’, in David Scott Kastan (ed.), A Companion to Shakespeare, 434–49, Oxford: Blackwell. Maguire, Laurie E. (2000), ‘Composition/Decomposition: Singular Shakespeare and the Death of the Author’, in Andrew Murphy (ed.), The Renaissance Text: Theory, Editing, Textuality, 135–53, Manchester: Manchester University Press. Massai, Sonia (2012a), ‘Early Readers’, in Arthur Kinney (ed.), The Oxford Handbook of Shakespeare, 142–61, Oxford: Oxford University Press. Massai, Sonia (2012b), ‘Edward Blount, the Herberts, and the First Folio’, in Marta Straznicky (ed.), Shakespeare’s Stationers: Studies in Cultural Bibliography, 132–46, Philadelphia: University of Pennsylvania Press. Moxon, Joseph (1958), Mechanick Exercises on the Whole Art of Printing 1683–4, ed. Herbert Davis and Harry Carter, London: Oxford University Press. Nelson, Alan H. (2005), ‘Shakespeare and the Bibliophiles: From the Earliest Years to 1616’, in Robin Myers, Michael Harris and Giles Mandelbrote (eds), Owners, Annotators and the Signs of Reading, 49–73, London: British Library. Rumbold, Kate (2011), ‘Brand Shakespeare?’, Shakespeare Survey, 64: 25–37. Shakespeare, William (1593), Venus and Adonis, London. Shakespeare, William (1594), Lucrece, London. Shakespeare, William (1609a), The Famous Historie of Troylus and Cresseid, London.

110

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Shakespeare, William (1609b), Shake-speares Sonnets, London. Shakespeare, William (1622), The Tragoedy of Othello, The Moore of Venice, London. Shakespeare, William (1623), Mr. William Shakespeares Comedies, Histories, & Tragedies, London. Smith, Emma (2015a), ‘The Canonisation of Shakespeare in Print, 1623’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 134–46, Cambridge: Cambridge University Press. Smith, Emma (2015b), The Making of the First Folio, Oxford: Bodleian Publishing. Smith, Emma (2016), Shakespeare’s First Folio: Four Centuries of an Iconic Book, Oxford: Oxford University Press. Snyder, Susan (2001), ‘The Genres of Shakespeare’s Plays’, in Margreta de Grazia and Stanley Wells (eds), The New Cambridge Companion to Shakespeare, 83–98, Cambridge: Cambridge University Press. Stern, Tiffany (2006), ‘“On Each Wall and Corner Poast”: Playbills, Title-Pages, and Advertising in Early Modern London’, English Literary Renaissance, 36: 57–89. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Stoddard, Roger E. (1987), ‘Morphology and the Book from an American Perspective’, Printing History, 9 (1): 2–14. Straznicky, Marta, ed. (2012), Shakespeare’s Stationers: Studies in Cultural Bibliography, Philadelphia: University of Pennsylvania Press. Taylor, Gary (2006), ‘Making Meaning Marketing Shakespeare 1623’, in Peter Holland and Stephen Orgel (eds), From Performance to Print in Early Modern England, 55–72, Basingstoke: Palgrave Macmillan. Taylor, Gary and Michael Warren, eds (1983), The Division of the Kingdoms: Shakespeare’s Two Versions of ‘King Lear’, Oxford: Clarendon Press. Tolbert, James M. (1950), ‘The Argument of Shakespeare’s “Lucrece”: Its Sources and Authorship’, The University of Texas Studies in English, 29: 77–90. Webster, John (1612), The White Divel, London. Werner, Sarah (2019), Studying Early Printed Books, 1450–1800: a Practical Guide, Chichester: Wiley Blackwell.

CHAPTER 2.4

Shakespeare in the early modern book trade MARTA STRAZNICKY

The book of Shakespeare’s that you will most likely have seen, if only its famous title page image of the author, is the earliest collected edition of Shakespeare’s plays, published in London in 1623, seven years after Shakespeare’s death.1 Known as the First Folio, this book is a monumental publication in the history of English and indeed global literature, providing as it does the only surviving text of eighteen of Shakespeare’s plays (Twelfth Night, Macbeth and The Tempest among them) and alternate texts of another eighteen plays that had previously been printed in single editions. Had the First Folio never been published, our estimation of Shakespeare as a playwright would be entirely different; had none of his plays or poems appeared in print at all, we would know virtually nothing about William Shakespeare the writer and would have little reason to go in search of his works. In this very real sense, it was the late-sixteenth- and early-seventeenth-century book trade that produced the Shakespeare we know. This chapter will introduce the workings of the book trade in Shakespeare’s time, the place of his writings within it and the major resources and kinds of evidence that scholars use to understand how and why a writer whose professional career was in theatre came to have such prominence in print. Shakespeare in the early modern book trade was a number of things, not all of which are self-evident or familiar to us. Clearly, Shakespeare was an author’s name, printed in various forms on the title pages of dozens of books. The First Folio title page gives us the most striking example: it is dominated not only by Shakespeare’s name, printed in large evenly spaced capitals across the full width of the page, but also by an engraving of the author. Shakespeare was also the unidentified and even unacknowledged author of books that were attributed on their title pages to a particular acting company rather than a person. Romeo and Juliet, for example, a play that today is universally associated with Shakespeare’s name, was published without an authorial attribution throughout his lifetime. Secondly, whether named on title pages or not, Shakespeare in these forms was very much a material object, a pamphlet or, as with the Folio, a substantial book, made of ink and folded sheets of paper, stitched together with thread or bound more durably in specially prepared animal skin. As an object, this Shakespeare had passed through many hands, in most cases numbering at least the person(s) who had prepared the manuscript, the

112

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

stationer(s) who bought it, the guild officials who licensed it for publication, and the craftsmen at the printing house who set the type, inked the plates, operated the press, corrected proofs, hung printed sheets to dry, assembled them into parcels and carried or carted them to the booksellers’. Throughout this process, Shakespeare was also a more complex kind of commercial property, a vestige of what was known as ‘copy’, the exclusive, legally protected right of one or more stationers to print copies of a particular title. Shakespeare as commercial property was not an author’s name or a material object, but a set of titles the printing rights to which stationers could buy, transfer, inherit and sell. One thing that Shakespeare most certainly was not in the early modern book trade is the owner of his written work: there was no legally defined and recognized position for the person of the author in publishing for nearly a hundred years after his death.2 To us, this can be a fairly staggering piece of information, accustomed as we are to thinking of writers as having an inalienable right to the product of their creative labour. When the manuscript of a work by Shakespeare was acquired by a stationer, it ceased to be the author’s property. While authors could become involved in the design and production of their books after their manuscript was sold, unless they were members of the Stationers’ Company themselves, they had no legal right to have copies printed or withheld from print.3 When we consider Shakespeare in the book trade, therefore, our focus is necessarily on the stationers – the printers and booksellers – whose business it was, and who made it their business, to produce and disseminate his writings in the form of printed books.4 Mr. William Shakespeares Comedies, Histories, & Tragedies, the actual title of the First Folio (STC 22273), provides an excellent illustration of the ways in which stationers are directly accountable for the creation of Shakespeare as an authorial name, a material object and a commercial property. Digital or facsimile images make it difficult to imagine how impressive the book is both in size and contents: it measures about 9 in x 14 in (c. 23 cm x 36 cm) and is over 900 double-column, carefully laid out and ruled pages presenting not only the texts of thirty-six plays, but also prefatory matter including commendatory poems on Shakespeare and his plays, a dedicatory letter by Shakespeare’s fellow players John Heminges and Henry Condell to two high-ranking noblemen, their general address to readers, a table of contents organizing the plays into genres and a list of the principal actors who had performed them. Here, the Shakespeare we encounter is a man of the theatre. As shareholding members of the King’s Men alongside Shakespeare, Heminges and Condell naturally give overt expression to the theatrical milieu in which the plays originated and flourished, not only in their own and other actors’ presence in the book, but also in references to the plays having ‘had their triall alreadie, and stood out all Appeales’ (sig. πA3r). Their claims about the quality of the texts reinforce this connection even more powerfully: whereas earlier editions of the plays are disparaged as ‘stolne, and surreptitious copies’, the collection presents them ‘cur’d, and perfect of their limbes’, along with ‘all the rest, absolute [complete] in their numbers, as he conceiued them’. These provocative claims have generated masses of scholarly work on the question of how the variant texts of individual plays are related to one another and, hypothetically, to how Shakespeare ‘conceiued them’.5

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

113

For our purposes, it is important primarily to note that the competing authority of the surviving texts is not an invention of later editors and scholars but something that Shakespeare’s own fellow actors and publishers set up, and that they did so in a commercial context, urging readers to ‘buy … first’, ‘what euer you do, Buy’. At over 900 pages, the Folio would have been very costly to print. We know that it was in production for nearly two years and required, in addition to the investment in materials and labour, payment for and negotiation of rights with at least eight stationers who were not among the partners at whose ‘Charges’ the book was printed (colophon). About 750 copies would have been printed; records of two early sales reveal that a copy bound in plain calf retailed for one pound. It is difficult to know precisely how expensive this would have been for Shakespeare’s contemporaries, but we can estimate by comparing with other things for which people paid one pound. By this measure, the cost of a bound copy of the First Folio was roughly equivalent to the cost of forty-four loaves of bread, or four skillets, or a dozen stools and a round table.6 By most measures, these would be significant expenditures, but not beyond the means of a sizeable number of book-buyers: at least that was the risk the investors in the collected works of Shakespeare were prepared to take. Those investors were not only Heminges and Condell, authors of the dedicatory epistle and address to the reader, but also the individuals named in the parts of the Folio where information about the production of the book was traditionally printed, the imprint on the title page and the colophon on the last page of the volume: William and Isaac Jaggard (father and son), Edward Blount, John Smethwick and William Aspley. These printers and booksellers formed what we would call a syndicate, partners in financing the production of the First Folio. We do not know what the specific terms of the partnership were other than what the imprint and colophon tell us: in addition to identifying the place and date of publication, the First Folio imprint states that the book was ‘Printed by Isaac Jaggard, and Ed. Blount’. ‘Printed’ here encompasses the activities of both printing and publishing as we would understand them. Isaac Jaggard had succeeded to his father’s business when William died in early November 1623, but both were involved in the project throughout its planning and production (Murphy 2003: 43–7). William Jaggard was prominent in the trade and ran one of the most prosperous printing businesses at the time. He was the official printer to the City of London, held the playbill monopoly and had recently collaborated on another major print venture involving Shakespeare’s plays.7 Isaac became a full member of the Stationers’ Company in 1613. On 4 November 1623, Isaac formally took over the post of printer to the City, ‘in place of his late Father deceased’ (Wells 2004). On 8 November, Isaac, together with Edward Blount, entered their right to print copies of the sixteen Shakespeare plays that had not previously been registered to others (Arber 1875–94: IV.107). Blount was the most important publisher of literature in the early seventeenth century and had long-standing relationships with many major writers as well as with the family of the Herbert brothers to whom the Folio was dedicated (Massai 2013: 132–46). As an apprentice, Blount worked in the bookshop of William Ponsonby, who had published major literary works by Philip Sidney and Edmund Spenser. He took over Ponsonby’s shop in 1603 and continued to build an impressive list,

114

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

including English and European writers of the stature of Christopher Marlowe, Miguel Cervantes and Michel de Montaigne. No other publisher of the time built as distinguished a list of literary titles as Blount. It is not known exactly when he joined the Folio syndicate, but it is clear that he was ideally positioned to lead and champion the project. The other two booksellers, named only in the colophon, John Smethwick and William Aspley, have a less clear role in the publication of the Folio. Smethwick held rights to four Shakespeare plays which would give him a strong financial incentive to collaborate in the venture, but Aspley is quite marginal. He was co-owner of the rights to two plays (many fewer than another stationer, Thomas Pavier, who owned copy in at least six plays but is not one of the syndicate) and named as the bookseller on an imprint of Shakespeare’s Sonnets in 1609 (STC 22353), but not otherwise invested in Shakespeare or play publication. However, we have no information about the reasons for and terms of their involvement in the syndicate; both Smethwick and Aspley remained involved in the project through to the publication of the second Folio in 1632 (Murphy 2003: 51–3). It is evident that part of the syndicate’s strategy was to imbue the expensive collected works with greater value than that of previously (and in some cases still) available single editions of the plays so that the book would sell (Hooks 2016: 99–109), but they were also engaged in a cultural project: the transformation of a professional playwright whose work was developed within a collaborative theatrical environment over nearly three decades into the single and singular literary author of natural genius, the Shakespeare who is both named and figured on the title page. In the context of the early modern book trade, such a cultural project could not have been achieved by two actors alone, however much capital they had or how privileged their access to the playwright’s manuscripts. Before 1623, Shakespeare’s plays were not a defined, coherent set of texts in the way we think of them today. Eighteen plays that would be included in the Folio had been published in single editions, but some of these had never been attributed to Shakespeare, and a number of other plays that were published as written by William Shakespeare, such as A Yorkshire Tragedy in 1608 (STC 22340), were not included in the First Folio and continue to have a fraught relationship with the Shakespeare canon (see Peter Kirwan’s chapter in this volume). The presence, absence and use of Shakespeare’s name on single editions of plays is a major area of research.8 The thing to note here is that the very idea of a Shakespeare canon was created – needed to be created, if its commercial and cultural value was to be realized – by the publishers of the First Folio. The manuscripts of sixteen previously unpublished plays were seemingly delivered to the syndicate by Heminges and Condell. According to normal trade practice, the right to print these texts could only be granted to members of the Stationers’ Company. As noted above, Edward Blount and Isaac Jaggard were granted that right on 8 November 1623. But while they thus acquired copyright to nearly half of the titles planned for the collection, Blount and Jaggard also needed to negotiate with the various owners of the previously published plays for the right to reprint them. The details of those negotiations are not known, but there is evidence in the physical make-up of the book itself that the production process was interrupted several times, conceivably

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

115

on account of difficulties reaching agreement with two copy-holding stationers, Matthew Law and Henry Walley (Hooks 2016: 104–9). Beyond the challenge of arriving at agreeable terms with fellow stationers, the Folio syndicate faced the task of pitching a collection of unlike and (some) previously unrelated texts as a coherent and complete volume attributable to an individual writer. The earliest record we have of the Folio project shows that the book was being marketed this way from the beginning. Appearing a year before the Folio was for sale, an advertisement in a printed catalogue of recently published books included the item, ‘Playes, written by M. William Shakespeare, all in one volume, printed by Isaack Iaggard, in fol.’ (Blayney 1990: 8). The claim to singular authorship and the ‘all in one volume’ rhetoric is echoed in the printed address to the readers, where the potential buyer is advised that the volume contains the ‘True Originall’ versions of Shakespeare’s previously published plays ‘and all the rest, absolute [i.e. complete] in their numbers’ (sig. πA3r), and then again on an inner title leaf that makes explicit the claim to completeness implied on the volume’s title page: ‘The Workes of William Shakespeare, containing all his Comedies, Histories, and Tragedies’ (sig. πB2r). The Folio table of contents, titled ‘A CATALOGUE of the severall Comedies, Histories, and Tragedies contained in this Volume’, less overtly reinforces these claims to completeness, but it is interesting for a different reason: as an index to ‘all’ of Shakespeare’s plays, it reveals how tenuous the edifice really is. Inadvertently exposing the trouble with Walley, Troilus and Cressida is missing from the Catalogue, even though in some copies the play is printed at the beginning of the Tragedies; some plays are categorized differently than they had been when they were published as single editions (for example, King Lear was billed as a ‘Chronicle Historie’ when it was first printed in 1608; Richard III had always been and would continue to be labelled a tragedy in single editions; and The Merchant of Venice, similarly, had been and would continue to be a ‘Historie’ in quarto); and the complete absence of Pericles and The Two Noble Kinsmen from the book cannot be squared with any rationale that would also account for the inclusion of 1 Henry VI, Henry VIII and Timon of Athens (all five plays were written by Shakespeare in collaboration with other playwrights). Ironically, it is the failure of the Folio as a material text to fully match the publishers’ rhetoric about single authorship, textual authority and completeness that best reveals how embedded the book is in the practices of the early modern book trade. Before 1623, Shakespeare’s individual works had been published in the form of inexpensive, single editions that resemble long pamphlets more than substantive books. Produced in editions of about 800 copies, they were printed on relatively inexpensive paper and in a font size that would set a play of average length onto about seventy pages. These books are known as quartos, referring to the fact that they are made up of sheets of paper folded twice to produce four leaves, or eight pages. Measuring roughly half the size of a folio publication, quarto was the standard format used for plays and many poetry books in Shakespeare’s time and was reasonably affordable: a quarto play of average length would sell for about six pence, one-fortieth the cost of a bound copy of the First Folio.9 Before 1623, forty-nine individual editions of the Shakespeare plays collected in the Folio and

116

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

twenty-two editions of his poetry books had been published. Taken as a whole and by any measure that is alert to the scale of literary publishing in the early modern period, Shakespeare was a leading playwright and poet in print.10 It might not occur to us to ask how this happened, but if we bear in mind that when Shakespeare started writing for the theatre it was rare for a play to be published at all, we can appreciate how unpredictable this development would have been from the vantage point of 1593, the year in which a work by Shakespeare first saw print (Farmer and Lesser 2005a). That first work was not a play, however, but Venus and Adonis (STC 22354), one of Shakespeare’s two narrative poems (the other being Lucrece, published one year later (STC 22345) that together would be printed fifteen times in his lifetime. Neither poem was attributed to Shakespeare on its title page, but they both had dedications from him to Henry Wriothesley, the Earl of Southampton, that openly avowed his authorship. The commercial value of these texts was by no means assured when they were first printed, and in fact the original owner of the copy of Venus and Adonis, Richard Field, transferred his right to an associate, John Harrison, just over a year after printing the first edition. Harrison already owned but had not yet published Lucrece. He hired Field to print it in 1594 with a similar title-page design to Venus and Adonis and sold both books at his shop, the White Greyhound, in St Paul’s Cathedral Churchyard. Harrison decided to change the format of his next editions, from quarto to octavo, a smaller size that Harrison had selected several years earlier for his Latin editions of Ovid (also printed by Field).11 Venus and Adonis already carried a Latin epigraph from Ovid’s Amores on its title page, so Harrison’s decision to change its format, along with Lucrece, to match other Ovidian books of poetry available in his bookshop makes sense from a marketing perspective (Erne 2013: 148). Whatever Shakespeare’s own view of the relationship between these two works may have been, the fact is that it was a bookseller whose publishing and design decisions initially put them into conversation in print, as works linked by theme and genre not only with one another but with other books in Harrison’s shop.12 This is just one book trade ‘moment’ in the publication history of Shakespeare’s poetry. Another of note would be the publication by William Jaggard of poems by Shakespeare, including two unpublished sonnets and three poems from Love’s Labour’s Lost, in a book ascribed to him but made up mainly of amorous poetry by others. It seems that this work, The Passionate Pilgrim (STC 22341.5 and STC 22342), was meant to be another companion volume to Venus and Adonis since it was sold by the stationer, William Leake, who had acquired Venus and Adonis from Harrison in 1596 (Arber 1875–94: III.65). The third edition of 1612 (STC 22343) was printed with variant title pages, only one of which ascribes the work to Shakespeare. This edition triggered a dispute about the authorship issue, including a second-hand report by Thomas Heywood that Shakespeare himself was ‘much offended with M. Iaggard (that altogether vnknowne to him) presumed to make so bold with his name’ (Heywood 1612: sig. πG4v). Printing a variant title page that does not name Shakespeare as author makes sense as appeasement of the offended writer (it is Heywood who is actually making the complaint against Jaggard), but there was in fact no legal repercussion to ‘making bold’ with an author’s name unless

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

117

another stationer’s right to copy was violated. There is no evidence this was the case, and bibliographic evidence actually suggests that the inclusion of Shakespeare’s name on the title page was intended while its removal was an oversight by compositors in the printing house (Burrow 2002: 79n1). To Heywood, possibly to Shakespeare and frequently to critics and bibliographers, Jaggard is guilty at the very least of being disingenuous in attributing the whole volume to Shakespeare when at most five of its poems can unequivocally be said to be his. This debate will no doubt continue, but what is important to point out here is how, as early as 1599, a prominent literary publisher is using Shakespeare’s name in connection with a brand of poetry the commercial value of which had been created within the book trade. A final example concerning the positioning of Shakespeare’s poems in the book trade is the publication of the Sonnets in 1609, with the striking attribution: ‘SHAKESPEARES SONNETS. Neuer before Imprinted’ (STC 22353). Set in the context of other books published by Thomas Thorpe around 1609, the Sonnets appear to be consistent with his interest in elite literature (Erne 2013: 156–7). However, the book was printed with two title pages, each naming a different bookseller. Thorpe’s profile complements that of William Aspley, who sold the Sonnets at his shop in Paul’s Churchyard, but is unlike many of the books sold by John Wright in Newgate Market, near Christ Church Gate.13 By 1609, Wright had built up a list of playbook titles from a broader range of theatrical milieus than what Thorpe seems to have favoured, including a number of plays from the down-market Red Bull theatre.14 Nearly all of these were printed by George Eld, who also printed the Sonnets. There was clearly some business relationship between Eld, Wright and Thorpe, but whatever it was does not fit neatly onto an image of Thorpe as an upmarket publisher. Placing Shakespeare’s Sonnets in Wright’s shop as well as Aspley’s reveals, in fact, a strategy of diversification rather than specialization in marketing the book. In both venues, the Sonnets were boldly identified as ‘SHAKE-SPEARES’ and being sold alongside plays from many of the professional theatres. Interestingly, this synergy between Shakespeare the poet and Shakespeare the playwright is something that the publishers of the First Folio would not exploit. From the earliest editions of the poems, we thus find printers and booksellers actively shaping not only Shakespeare’s books, but also his reputation as a writer. The examples cited above show some of the activities involved: buying and selling copyrights, investing in texts that can be marketed jointly, deciding on facets of book design, compiling and editing collections of verse that resonate with other publications, and strategically placing Shakespeare’s name on title pages or elsewhere in a printed volume. Furthermore, these activities occur within a wellestablished network of trade relationships among printers and booksellers that, in turn, functions within a commercially and politically regulated context designed to balance the business interests of stationers with the priorities of the current regime. The quarto editions of Shakespeare’s plays published before 1623 participate in this same process, shaping his reputation as a leading playwright within the distinctive nature of the market for printed drama.15 Before turning to some examples of how stationers made Shakespeare into a successful playwright in print, it may be useful to take account of the evidence

118

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

and major resources available for research on this question. The business practices and regulation of the trade as a whole are richly documented in the records of the Stationers’ Company, particularly in the entry books listing the works that had been registered for the purpose of ‘establishing the right to a literary property and for keeping down piracy’ (Myers 1990: 6). These volumes are referred to collectively as the ‘Stationers’ Register’, an abbreviation of the title of the printed edition of the entry books published by Edward Arber (Arber 1875–94), whose contents are now also available in a conveniently searchable online database (Bergel and Gadd n.d.). The Stationers’ Register is a vital resource for research on the ownership of copy. Entries in the Register tell us at the very least the name of the entering publisher, the title of the work(s) and the payment for entry. Entries sometimes also include the name of Company officers and a government authority if these appear on the manuscript, indicating that the copy being entered had been submitted to and approved by a censor and licensed for printing by the Company. Entries can also include information about the existence of agreements between publishers concerning the transfer or assignment of copyrights. Because entry was not obligatory, the Stationers’ Register does not provide a comprehensive record of copy ownership or book production: about one-third of the books published in Shakespeare’s time were not entered, and many entries in the Register refer to titles of books that were either never printed or printed with titles that do not match the entries.16 The Register is extremely useful, however, in tracking the ownership of particular titles, whether these are consistent with imprints or not, and, even more so, individual publishers’ practices with respect to copy ownership (sole, shared, transferred or assigned from others, etc.), the subject matter of the titles they owned and a chronology of their investment decisions. Within the pages of the Stationers’ Register, publishers reveal themselves to be actively staking claim to or surrendering specific titles (among other activities), and this can tell us a great deal about the kind of business they are engaged in and the place of Shakespeare’s works within it.17 Importantly, although Shakespeare’s name appears in the Stationers’ Register just four times in his lifetime (Erne 2013: 175n163), every one of his books of poetry and every play that was included in the Folio were entered either before or after printing. Considering that the purpose of entry in the Register was to protect stationers’ intellectual property from infringement by other stationers, this remarkably consistent record of entries indicates that works by Shakespeare were highly valued by the printers and booksellers who paid to have them registered.18 A second major source of information about Shakespeare and the book trade are the imprints found on editions of his works. Imprints are the blocks of text on title pages where publishing information is printed. In addition to the place and date of publication, early modern imprints usually include the names or initials of some or all of the printers, publishers and booksellers involved in producing and selling books. Imprints also frequently but by no means typically include geographic markers (shop sign, street location, bookselling district, etc.) indicating where copies may be purchased.19 The first edition of Richard II (STC 22307), for example, has this clear and informative imprint: ‘LONDON Printed by Valentine Simmes for Androw Wise, and are to be sold at his shop in Paules church yard at the signe of

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

119

the Angel. 1597’. Wise had entered the play in the Stationers’ Register on 29 August (Arber 1875–94: III.89), so we know that he was the publisher in addition to being the bookseller. He would have hired Simmes to print the play, but he himself took on the financial risk of publishing and selling it. Furthermore, we learn from the imprint that the playbook was to be sold to other booksellers and to the bookbuying public at Wise’s shop, which we know was in the northeast corner of the churchyard of St Paul’s Cathedral, the very centre of the London book trade and the same area of the yard as John Harrison’s White Greyhound, where Venus and Adonis had been sold since 1593. There is much information to work with here: the emergence of a bookseller in the market for printed drama (1597 is the year of Wise’s earliest playbook publications), a partnership between a named bookseller and a printer, and precision in the identification of both a bookselling district as well as a shop and thus a spatial relationship between Shakespeare’s works and other books sold in the vicinity at the time. Using this kind of information, scholars have been able to build a detailed profile of Andrew Wise as a shrewd reader and bookseller who selectively acquired and marketed three Shakespeare plays that proved to be exceptionally successful in print (Massai 2007: 91–105; Erne 2013: 161–6; Hooks 2016: 66–98). Significant as they are, however, imprints are a challenge to work with because they were not devised to answer the questions that we might bring to them. Their main purpose was commercial rather than documentary: they advertised the printers and booksellers who were responsible for the production and distribution of the book in order to sell copies (Shaaber 1944). The kind and amount of information included on an imprint is thus part of the marketing strategy planned for the book. Explaining the particulars of the publishing arrangement (who manufactured the book, who financed it, who owns copy and who is to sell it) may be but is not necessarily aligned with marketing strategy, and we therefore find a variety of imprint formulas the precise meaning of which can be difficult to recover.20 The first edition of Romeo and Juliet (STC 22322), for example, printed in the same year as Wise’s first edition of Richard II, has this laconic imprint: ‘LONDON, Printed by John Danter. 1597’. This may sound like a straightforward assignment of responsibility to one individual for the printing and publication of the book, but because the verb ‘to print’ was ambiguous in the early modern period, meaning both to physically manufacture a book and also more abstractly to cause it to be printed (publish it), we would need to investigate further in order to determine whether Danter alone is responsible for this edition. We also have no information in this imprint about where the book is to be sold, a not uncommon occurrence the marketing function of which was to promote a wider-ranging sale (Shaaber 1944: 127). Moreover, there is no entry in the Stationers’ Register by Danter or anyone else for Romeo and Juliet before its first printing. Imprints are therefore not always in themselves sufficient evidence for determining who was responsible for the printing, publishing and selling of a particular book. Cross-referenced against entries in the Stationers’ Register, the information provided on imprints can often be verified, especially with respect to who owned the copy and therefore acted as publisher (whether this was the printer, the bookseller or another person). But in cases such as the first edition of Romeo and Juliet, where there is

120

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

no corroborating evidence in the Stationers’ Register, we have to piece together other kinds of information: did any of the named individuals have a history of entering copy? (Danter did. In fact, he had registered the first Shakespeare play to see print, Titus Andronicus, on 6 February 1594 (Arber 1875–94: II.644).) Was the title transferred or assigned at a later date, and does this reveal who may have owned copy at the time of this publication? (The first mention of Romeo and Juliet in the Stationers’ Register is in 1607 when, inexplicably, it is transferred from Cuthbert Burby to Nicholas Ling (Arber 1875–94: III.337). Danter had died in 1599, but there is no record of him or his widow transferring this copyright to Burby. However, Burby is named as the publisher and bookseller in the imprint on the second edition of the play in 1599 (STC 22323), so he had acquired it by then at the latest.) If an imprint does not give a location, is there other information we can use to specify where a book might have been sold or that we can use to speculate why an imprint location would not have been given? (We know from other imprints that Danter’s printing house was in Duck Lane near Smithfield, a secondary book trade location northeast of the City, so perhaps not the best place from which to distribute or sell copies. It is difficult to say anything more definitive about the absence of this information, but it is suggestive that the second edition was printed ‘for Cuthbert Burby, and [copies] are to be sold at his shop neare the Exchange’. Burby’s shop was near the south entrance to the Royal Exchange, a major bookselling area (Worms 1997). It is quite possible that he had bought wholesale copies of Danter’s edition of Romeo and Juliet and was selling them there in 1597, even though he is not named on the imprint.) As this example shows, while the precise nature of a publishing arrangement may not be clear from the imprint of a single book, imprints nevertheless connect a text with a person or persons and thus help to reconstruct stationers’ lists of publications and trade networks.21 This has proved to be a very fruitful line of inquiry, leading in recent years to ground-breaking research on the business practices of some of the major publishers of Shakespeare.22 We have already seen that Harrison’s editions of Shakespeare’s narrative poems were designed and marketed to resonate with other printed books that were thematically and stylistically related. The imprints in Harrison’s books provide a significant source of evidence for this picture of his business. Andrew Wise is another good example: in addition to acquiring and publishing Richard II in 1597, he brought out the first edition of Richard III in the same year. In 1598, he published second editions of both plays (and a third edition of Richard II), plus added a new play to his list, 1 Henry IV, which he reprinted the following year as a ‘Newly corrected’ text.23 These three exceptionally successful plays formed the backbone of Wise’s fairly limited but lucrative publishing business (Erne 2013: 161–2). Not only did he publish three editions of each play within a very short period (1597–1602), but he also added Shakespeare’s name to certain title pages and was proactive in correcting and improving the texts with each reprint (Erne 2013: 164; Massai 2007: 91–105). He also favoured one other writer, the preacher Thomas Playfere, who, like Shakespeare, was well known for a certain kind of flowing eloquence (Hooks 2016: 66–98). It seems that, like Harrison, Wise successfully marketed his three Shakespeare plays alongside another writer for

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

121

their stylistic similarity. In doing so, he bolstered both Shakespeare’s and his own reputation for literary refinement. Wise sold his rights to Richard II, Richard III and 1 Henry IV to Matthew Law in 1603 (Arber 1875–94: III.239). It can be informative to track plays when they move, as a bundle, to another stationer, another shop and another business context, for even though they are on some level the ‘same’ plays, they can also be repurposed to serve a different marketing strategy and thus occupy a different position in the trade. Imprints can help us reconstruct these shifts. In the context of Matthew Law’s business, Shakespeare’s Richard II, Richard III and 1 Henry IV effectively became history plays. Law’s imprints themselves do not re-title or otherwise mark the plays as different from what had been available at Wise’s shop, but in so frequently specifying that his shop, the Fox, is ‘neere St. Austines gate’ in Paul’s Churchyard, he focuses our attention on that particular area of the precinct.24 St Austin’s, or Augustine’s, Gate led into the Churchyard from Watling Street, a major thoroughfare to the east. It was a small but vibrant bookselling area for drama, owing to the presence there of Law and another stationer, Nathaniel Butter, who was also selling a series of wildly successful history plays: Samuel Rowley’s When you see me, you know me. Or the famous chronicle historie of King Henry the eight and the two parts of Thomas Heywood’s If you know not me, you know no bodie: or, The troubles of Queene Elizabeth.25 Reconstructing the publication chronology of these plays from their imprints reveals something like a call-and-response pattern, with Law almost immediately launching his editions of 1 Henry IV and Richard III, in 1604 and 1605, and Butter entering and publishing When you see me and If you know not me, Part One in 1605 and then both parts in 1606. In 1608, Law brings out his first edition of Richard II as well as a second edition of 1 Henry IV. In the same year, Butter brings out his third edition of If you know not me, Part One and, intriguingly, the first and only quarto edition of Shakespeare’s King Lear, billed not as a tragedy but as a ‘true chronicle historie of the life and death of King Lear and his three daughters’ (STC 22292). In 1612–13 the two go head to head: Law publishes Richard III in 1612 and 1 Henry IV in 1613, and the same year Butter brings out new editions of When you see me you know me and If you know not me, Part One. There is another curious mirroring in 1622, when Law once again publishes Richard III and 1 Henry IV, perhaps to satisfy demand ahead of the First Folio, and Butter quickly reissues both parts of If you know not me. By 1623, Law’s three Shakespeare plays had been competing with Butter’s English history plays at the two shops near St Augustine’s Gate for nearly two decades. They are shuffled into their correct chronological sequence by Jaggard and Blount, but their identity as history plays, especially the two that were originally published and marketed by Andrew Wise as tragedies, would not have been a given in 1603 when Matthew Law bought the rights. This short excursion into the development, marketing and sale of a particular set of Shakespeare textual properties could be extended to virtually any imprint or set of imprints of his playbooks. Two other stationers whose imprints reveal a significant role in shaping Shakespeare as a print commodity before 1623 are Nicholas Ling and John Smethwick. Ling has attracted considerable attention because he published the first and second quartos of Hamlet (STC 22275, co-published with John

122

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Trundle, and STC 22276). He was not a prolific play publisher, but he did have significant investments in anthologies of wisdom sayings and literary excerpts which included quotations from Shakespeare’s plays, particularly two he would formally acquire from Cuthbert Burby in 1607, Romeo and Juliet and Love’s Labour’s Lost (Arber 1875–94: III.337).26 Ling thus came to own three Shakespeare plays with markedly literary qualities, a facet that is notably highlighted in the typographical design of his first quarto of Hamlet (Lesser and Stallybrass 2008). He died in early April of 1607 and was buried in the same church where his bookshop had been located, St Dunstan’s-in-the-West (Johnson 1985: 207). His Shakespeare copyrights were assigned to John Smethwick, who also traded in St Dunstan’s Churchyard in Fleet Street, a major book trade area to the west of the City where the legal profession was concentrated.27 It was here that Smethwick continued to publish Romeo and Juliet and Hamlet as late as 1637. As an investor in the First Folio project, Smethwick would also presumably have been selling copies of Shakespeare’s collected works from 1623, which would be compatible with the markedly literary character of his individual Shakespeare titles. As mentioned above, Smethwick continued his involvement in the Folio to its second publication in 1632. That edition was issued with five separate imprints, one of which explicitly names him and his shop (STC 22274e). Taken as a whole, then, the Shakespeare books Smethwick was publishing and selling at St Dunstan’s from 1607 to 1637 not only continue but also build upon the legacy of Nicholas Ling in marketing the plays as dramatic literature. A final word should be said about bookshops. In Shakespeare’s London, anyone could sell books, either in a fixed location or as an itinerant pedlar, and people frequently sold books alongside other kinds of merchandise (Blayney 1997: 418n17). By the late sixteenth century, the shops of stationers whose business was chiefly in the sale of printed matter (i.e. members of the Stationers’ Company who were legally entitled to sell wholesale to other merchants) were typically in fixed locations throughout London, but concentrated in two main areas of the city: the churchyard around St Paul’s Cathedral and Fleet Street to the west of the city. These shops were usually sizable two-story buildings, with one or more rooms on the ground floor fronting on the churchyard or street, furnished with shelving, counters and tables, and storage or dwelling rooms on the same or a second story. Most bookshops were leased, not owned, and identified by pictorial signs – the Angel, the Fox, the White Greyhound, for example. The shop would have a stall out front or a hinged board projecting from the wall where books could be displayed to passers-by without their having to enter the shop (Blayney 1990: 10–11). Inside the shop, the bookseller would be working to sell books in customary ways: by talking about them or reading aloud to customers, sitting down for a bit of bookselling sociability, sharing drink or a light meal or even having apprentices ‘perform’ engagement with particular titles (Taylor 2006). What this all highlights is that the book trade was, above all, a business. As far as printed plays or poetry were concerned, the business dealt in luxury commodities, so that in addition to evaluating the market for a given title and knowing how to promote it, booksellers needed to convince their customers to buy: recall the plea of the First Folio publishers, ‘what euer you do, Buy’ (sig. πA3r). The books of Shakespeare’s plays and poems – published, packaged and marketed, as

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

123

we have seen, in multiple, intersecting ways for their intellectual and entertainment value – sold exceptionally well, but they did not sell themselves.

NOTES 1. Digital images of Shakespeare texts are available on EEBO (searchable by STC number). 2. On early modern concepts of proprietary authorship, see Loewenstein (2002). 3. For a concise overview of the regulation of the book trade, see Bland (2002: 183–210). For a more general survey, see Raven (2007). 4. The term ‘publisher’ was not used in the early modern period. The person who caused a book to be printed, whether by manufacturing or financing it, was called a ‘printer’. The term ‘stationer’ was used for printers, booksellers and bookbinders. Throughout this chapter, ‘stationer’ and ‘publisher’ are used to mean a printer and/ or bookseller who is known to have caused a book to be printed, whether or not he did the actual printing and whether or not he was a member of the Stationers’ Company. The stationers discussed in this chapter were all men, but women were also active in the book trade (Smith 2012). 5. See Werstine (1990) and Jowett’s chapter in this volume. 6. On the cost of the Folio, see West (2007: 74–5). 7. This was the series of quarto reprints of nine plays by Shakespeare (some apocryphal) published by Thomas Pavier in 1619. Pavier collaborated with William Jaggard on reprinting and repackaging the plays for sale either as single copies or as a collection, seemingly by agreement with their copyright holders. As a collection, this series of quartos is a significant precursor to the First Folio (Murphy 2003: 38–41; Massai 2007: 106–35). 8. On the use of Shakespeare’s name on title pages, see Kastan (2001: 14–49). 9. For a detailed analysis of the economics of play publication, see Blayney (1997: 405–13). Blayney points out that there is some discrepancy in the cost of playbook quartos (421n61). 10. Quantitative analysis is an indispensable but problematic methodology in book history. Even in doing a straightforward count of ‘Shakespeare’ publications, we confront real difficulties: some of his works were published anonymously and some works that scholars believe were not written by him were published with his name on them. At all times, it is important to be clear about what, precisely, is being counted. A similar challenge arises in studies of Shakespeare’s popularity or success in print, which rely heavily on quantification. While there are conflicting approaches to the evidence, recent scholarship has demonstrated that, measured in the context of the sector of the book trade to which literary publications belonged, books ‘by’ Shakespeare were unusually lucrative. See Farmer and Lesser (2005a and 2005b). See Erne (2013: 25–55) for a critical discussion of quantitative analysis and additional, positive evidence for the preeminence of Shakespeare’s plays in print.

124

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

11. See STC 22356, 22357 and 22346. 12. This account of Harrison’s publishing and marketing strategy is indebted to Hooks (2016: 35–65). Hooks and Lesser have pioneered the study of the bookshop as a category of analysis in Shakespeare studies. See Lesser (2013). 13. For bio-bibliographies of early modern stationers, consult McKerrow (1910). 14. On the cultural stratification of the market for plays and playbooks, see Lesser (2004: 52–80) and Erne (2003). 15. On playbooks in the marketplace of print, see Lesser (2011). 16. See Blayney (1997: 396–405) for a masterful account of the separate processes for securing authority and license, and for entering copy. For an overview of press censorship as it pertains to drama, see Clegg (1999). 17. Arber (1875–94) can be searched by stationers’ names. See the Index in Volume 5. 18. For a detailed analysis of the ratio of entered vs. printed texts, see Bell (1994). 19. These geographic markers were printed for the benefit of other booksellers (Blayney 1997: 389), but the wider public would also have been a target audience given that title pages were posted around the City as advertisements (Smith 2011: 23; Stern 2009: 36–62). 20. The various imprint formulas in early modern books are classified and explained by Shaaber (1944). Smith (2011) analyses the geographic information in imprints. 21. Imprints also provide us with the count and chronology of editions of a particular work, information essential to assessing the success of a book in print. See note 10 above. 22. See Blayney (1997) for a general reconstruction of the process of play publication. Major studies focused on specific publishers and publishing contexts include Lesser (2004), Massai (2007), Erne (2013) and the essays in Straznicky (2013). 23. Wise’s editions of Richard II are STC numbers 22307 (1597), 22308 and 22309 (both 1598); his editions of Richard III are STC numbers 22314 (1597), 22315 (1598) and 22316 (1602); and his editions of 1 Henry IV are STC numbers 22279a (date is conjectured to be 1598 as this copy lacks the title page), 22280 (1598) and 22281 (1599). 24. Law’s editions of Richard II are STC numbers 22310 and 22311 (both 1608), and 22312 (1615); his editions of Richard III are STC numbers 22317 (1605), 22318 (1612), 22319 (1622) and 22320 (1629); and his editions of 1 Henry IV are STC numbers 22282 (1604), 22283 (1608), 22284 (1613) and 22285 (1622). 25. Butter published four editions of When You See Me You Know Me: STC numbers 21417 (1605), 21418 (1613), 21419 (1621) and 21420 (1632); eight editions of 1 If You Know Not Me You Know Nobody: STC numbers 13328 (1605), 13329 (1606), 13330 (1608), 13331 (1610), 13332 (1613), 13333 (1623), 13334 (1632) and 13335 (1639); and four editions of 2 If You Know Not Me You Know Nobody: STC numbers 13336 (1606), 13337 (1609), 13338 (1623) and 13339 (1633). Note that Law had died in 1629. Butter had a thriving business as a publisher of newsbooks and pamphlets in addition to plays (Baron 2004).

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

125

26. Ling also acquired The Taming of a Shrew from Burby, a play that is related but not identical to the Shrew play published in the First Folio. Burby published two editions, in 1594 (STC 23667) and 1596 (STC 23668), and Ling published one in 1607 (STC 23669). The copyright was assigned to Smethwick in 1607 along with the other three plays, but he did not publish The Taming of a Shrew. In 1631, however, Smethwick published a quarto edition of Shakespeare’s The Taming of the Shrew (STC 22327), which had appeared in the First Folio: evidently, he could claim copyright on the basis of owning its precursor, The Taming of a Shrew. This is a good example of a copyright being assigned to a title rather than to a particular work (Marino 2011: 48–74). 27. See Volume 3 of the STC for descriptions of imprint locations and lists of publishers’ addresses. Volume 3 also has an index of printers and booksellers that is indispensable for reconstructing an individual’s activity in the book trade.

REFERENCES Arber, Edward, ed. (1875–94), A Transcript of the Registers of the Company of Stationers of London 1554–1640 A.D., 5 vols, London: Privately printed. Baron, S. A. (2004), ‘Butter, Nathaniel (bap. 1583, d. 1664)’, ODNB. Available online: https://doi.org/10.1093/ref:odnb/4224 (accessed 13 November 2020). Bell, Maureen (1994), ‘Entrance in the Stationers’ Register’, The Library, 4th ser., 16: 50–4. Bergel, Giles and Ian Gadd (n.d.), Stationers’ Register Online. Available online: https:// stationersregister.online/ (accessed 13 November 2020). Bland, Mark (2002), A Guide to Early Printed Books and Manuscripts, Oxford: Wiley Blackwell. Blayney, Peter W. M. (1990), The Bookshops in Paul’s Cross Churchyard, London: Bibliographical Society. Blayney, Peter W. M. (1997), ‘The Publication of Playbooks’, in John D. Cox and David Scott Kastan (eds), A New History of Early English Drama, 383–422, New York: Columbia University Press. Burrow, Colin, ed. (2002), The Complete Sonnets and Poems, The Oxford Shakespeare, Oxford: Oxford University Press. Clegg, Cyndia Susan (1999), ‘Liberty, License, and Authority: Press Censorship and Shakespeare’, in David Scott Kastan (ed.), A Companion to Shakespeare, 464–85, Oxford: Blackwell. Erne, Lukas (2003), Shakespeare as Literary Dramatist, Cambridge: Cambridge University Press. Erne, Lukas (2013), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Farmer, Alan B. and Zachary Lesser (2005a), ‘The Popularity of Playbooks Revisited’, Shakespeare Quarterly, 56: 1–32. Farmer, Alan B. and Zachary Lesser (2005b), ‘Structures of Popularity in the Early Modern Book Trade’, Shakespeare Quarterly, 56: 206–13.

126

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Heywood, Thomas (1612), An Apology for Actors, London. Hooks, Adam G. (2016), Selling Shakespeare: Biography, Bibliography, and the Book Trade, Cambridge: Cambridge University Press. Johnson, Gerald D. (1985), ‘Nicholas Ling, Publisher 1580–1607’, Studies in Bibliography, 38: 203–14. Kastan, David Scott (2001), Shakespeare and the Book, Cambridge: Cambridge University Press. Lesser, Zachary (2004), Renaissance Drama and the Politics of Publication: Readings in the Book Trade, Cambridge: Cambridge University Press. Lesser, Zachary (2011), ‘Playbooks’, in Joad Raymond (ed.), The Oxford History of Popular Print Culture, 520–34, Oxford: Oxford University Press. Lesser, Zachary (2013), ‘Shakespeare’s Flop: John Waterson and The Two Noble Kinsmen’, in Marta Straznicky (ed.), Shakespeare’s Stationers: Studies in Cultural Bibliography, 177–96, Philadelphia: University of Pennsylvania Press. Lesser, Zachary and Peter Stallybrass (2008), ‘The First Literary Hamlet and the Commonplacing of Professional Plays’, Shakespeare Quarterly, 59: 371–420. Loewenstein, Joseph (2002), The Author’s Due: Printing and the Prehistory of Copyright, Chicago: University of Chicago Press. Marino, James J. (2011), Owning William Shakespeare: The King’s Men and Their Intellectual Property, Philadelphia: University of Pennsylvania Press. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Massai, Sonia (2013), ‘Edward Blount, the Herberts, and the First Folio’, in Marta Straznicky (ed.), Shakespeare’s Stationers: Studies in Cultural Bibliography, 132–46, Philadelphia: University of Pennsylvania Press. McKerrow, R. B., gen. ed. (1910), A Dictionary of Printers and Booksellers in England, Scotland and Ireland, and of Foreign Printers of English Books 1557–1640, London: Bibliographical Society. Murphy, Andrew (2003), Shakespeare in Print: A History and Chronology of Shakespeare Publishing, Cambridge: Cambridge University Press. Myers, Robin (1990), The Stationers’ Company Archive: An Account of the Records 1554–1984, Winchester: St Paul’s Bibliographies. Raven, James (2007), The Business of Books: Booksellers and the English Book Trade, New Haven: Yale University Press. Shaaber, M. A. (1944), ‘The Meaning of the Imprint in Early Printed Books’, The Library, 4th ser., 24: 120–41. Smith, Helen (2011), ‘“Imprinted by Simeon Such a Signe”: Reading Early Modern Imprints’, in Helen Smith and Louise Wilson (eds), Renaissance Paratexts, 17–33, Cambridge: Cambridge University Press. Smith, Helen (2012), ‘Grossly Material Things’: Women and Book Production in Early Modern England, Cambridge: Cambridge University Press. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Straznicky, Marta, ed. (2013), Shakespeare’s Stationers: Studies in Cultural Bibliography, Philadelphia: University of Pennsylvania Press.

SHAKESPEARE IN THE EARLY MODERN BOOK TRADE

127

Taylor, Gary (2006), ‘Making Meaning Marketing Shakespeare 1623’, in Peter Holland and Stephen Orgel (eds), From Performance to Print in Shakespeare’s England, 55–72, Basingstoke: Palgrave. Wells, Stanley (2004), ‘Jaggard, William (c.1568–1623)’, ODNB. Available online: https:// doi.org/10.1093/ref:odnb/37592 (accessed 13 November 2020). Werstine, Paul (1990), ‘Narratives about Printed Shakespeare Texts: “Foul Papers” and “Bad” Quartos’, Shakespeare Quarterly, 41: 65–86. West, Anthony James (2007), ‘The Life of the First Folio in the Seventeenth and Eighteenth Centuries’, in Andrew Murphy (ed.), A Concise Companion to Shakespeare and the Text, 71–90, Oxford: Wiley-Blackwell. Worms, Laurence (1997), ‘The Book Trade at the Royal Exchange’, in Ann Saunders (ed.), The Royal Exchange, 209–26, London: London Topographical Society.

CHAPTER 2.5

Shakespeare’s early readers and users: Annotating, commonplacing, collecting LAURA ESTILL

In March 2017, international news sources were abuzz about a small manuscript featured on Antiques Roadshow labelled ‘Shakespeare Comedyes & Tragedyes’ on its title page.1 On the BBC show, appraiser Matthew Haley confidently asserted that ‘this is somebody making notes in the same century as Shakespeare; anything, really, from the same century as him, about him, is of huge interest. And he’s copying out quotes from various Shakespeare plays: this is incredible.’ Haley continued to gush about the manuscript containing Shakespearean extracts: ‘it’s an extraordinary little object … a perfect little jewel’. Haley put his finger on the important ‘trend in scholarship going on at the moment about how early readers of Shakespeare were receiving his works, and what their reactions were, what they were focusing on’. The segment ended with Haley emphasizing that ‘the value to scholarship is enormous’ and suggesting that this manuscript would be worth ‘easily upwards of thirty thousand pounds’ at auction. Undoubtedly, as Haley enthused, ‘there is so much research that could be done on this item’ (‘Handwritten Notes’ 2017). This tiny manuscript, however, is not unusual, nor is it illegible, as Haley suggested. There are many ways to consider the reception of an early modern writer, such as looking at publication history, allusions to a given work or person or, in the case of a dramatist, evidence of performance. For sixteenth- and early-seventeenthcentury readers and audiences, Shakespeare was not the centre of the English literary or dramatic canon; he was one of many writers whose works were being performed, adapted and published. This chapter explores how early readers approached Shakespeare’s plays and poems by surveying three different, yet related, handwritten types of evidence: annotations, extracts and ownership marks. Early readers of Shakespeare should not be approached as a homogeneous group; we can, however, look to the textual traces left by readers in order to reconstruct the variety of early modern approaches to Shakespeare’s works.

SHAKESPEARE’S EARLY READERS AND USERS

129

This chapter opens by surveying existing research practices and describing key scholarly resources available to researchers working on these early handwritten documents. Annotations are the notes readers jotted into their books; commonplaces are selections copied from the works they read; and ownership marks are things like inscriptions that indicated who owned a particular copy of a book. Each of these three areas can be subdivided or sometimes called by different scholarly terms. The definition of annotations, for instance, overlaps with marginalia and readers’ marks (Bourne 2017: 368). Commonplacing is a form of extracting or excerpting; commonplace books are a type of miscellany; commonplaces intersect with proverbs. Ownership is linked to collecting and can rarely be considered without also thinking about provenance, and often, collections are called libraries. This chapter uses these terms – annotations, commonplacing, collecting and so on – in the most inclusive ways, by, for example, not insisting that marginalia always appear in a page’s margin. For specific definitions of each term, see Peter Beal’s A Dictionary of English Manuscript Terminology 1450–2000 (2008). This chapter considers manuscript evidence, which is to say, handwritten evidence; yet not all the documents covered in this chapter are manuscripts; some are printed texts with manuscript insertions. Manuscript evidence, however, cannot be evaluated without also considering its printed counterparts: for instance, manuscript commonplaces can mirror or diverge from printed commonplaces, or they can be copied from early printed editions marked with commonplace markers.2 After considering the current practices when it comes to finding and interpreting manuscripts that contribute to our understanding of the reception of early modern drama, I offer a case study that brings together some of these strands of scholarship, turning to a couplet from Shakespeare’s Pericles (first printed in 1609) that was copied into a 1599 print edition of George Chapman’s play An Humorous Day’s Mirth. The chapter concludes by looking at the emphases and biases of our current research methods and the scope of the scholarly resources available, while pointing to future directions for this field.

THE STATE OF THE FIELD: ANNOTATING, COMMONPLACING AND COLLECTING There is much handwritten evidence about the reception of Shakespeare’s works that is beyond the scope of this chapter, such as commendatory poems, promptbooks, letters, diaries and account books. Both Shakespeare Documented (Wolfe 2016) and The Henslowe-Alleyn Digitisation Project (Ioppolo 2005) offer high-quality facsimiles of many early manuscripts related to Shakespeare and early modern theatre. For instance, on Shakespeare Documented, a search for ‘extracts’ currently offers nineteen results, including two stanzas copied from Venus and Adonis by Henry Colling, likely in c. 1596 (McCarthy 2017). The Henslowe-Alleyn archive includes handwritten documents such as a booklist by William Cartwright the Younger that includes Shakespeare’s First Folio (Ioppolo 2005: ‘Catalogue’, ‘Miscellaneous

130

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Images’) and an ‘inventory of theatrical apparel’ from an early modern playhouse, with added Shakespeare forgeries to boot (Cerasano 2005). Even though researchers now have access to some images of early handwritten documents online, they are still challenging to read because of unfamiliar letter shapes and spellings. Some sites provide transcriptions of their manuscripts to help novice researchers. Other online resources, such as the National Archive’s Palaeography (TNA n.d.) and Cambridge University’s English Handwriting (Zurcher n.d.), offer valuable tools for those wishing to improve their paleography skills and learn to read these early documents so they do not seem ‘illegible’ like the Antiques Roadshow manuscript.3 When researching annotations, commonplaces, extracts and collecting, there are a number of valuable scholarly resources to consult, many of which are, for better or worse, focused on Shakespeare to the exclusion of other early modern writers. The Shakspere Allusion Book (Ingleby et al. 1909) was an important early-twentiethcentury bibliography of references to Shakespeare in print and manuscript. Today, researchers can find information about manuscripts with Shakespeare’s sonnets, extracts from his works, and other related documents in the online Catalogue of English Literary Manuscripts 1450–1700 (Beal 2009). Martin Wiggins and Catherine Richardson’s British Drama 1533–1642: A Catalogue (2012–) gathers information about early readers and performances of plays. Despite the long-standing nature of Shakespeare reception history, this field is still ever expanding, as new annotations, extracts and ownership marks are discovered and examined. Early readers sometimes wrote in their books. Claire M. L. Bourne points to the ‘tantalizing possibilities and frustrating limitations of using readers’ marks alone to grasp how early modern books, including books containing Shakespeare’s plays, were encountered, engaged, received, and used by their earliest readers’ (2017: 367). Many annotations, as scholars frequently lament, can be difficult to date; many marks (such as underlines, marginal crosses or tics and manicules) leave no evidence of their creator.4 Despite the challenges of working with readers’ marks, they cannot be ignored: as William H. Sherman (2008) demonstrates, annotating books was a common practice for sixteenth- and seventeenth-century readers. To ignore (or deplore) annotations is to disregard a wealth of readerly evidence (Sherman 2008, esp. chapter 8). Katherine Acheson’s edited collection, Early Modern English Marginalia (2019) showcases a range of approaches to the evidence of readers’ marks. Recently, Shakespeare studies in particular have been bolstered by research on readers’ marks in early printed books. Bourne notes that annotations found in Shakespeare’s First Folio are among the most studied early modern readers’ marks (2017: 369).5 The scholarly focus on readers’ marks in the First Folio could have emerged because, as Jean-Christophe Mayer suggests, ‘with a few exceptions, the First Folio was far more annotated than early quartos of Shakespeare’ (2016: 116); it likely also stems from the fact that the Folio survives at a much higher rate than other Shakespearean texts (Pettegree 2016: 1–2), and, as such, exists in many digital copies (Werner 2016a; 2016b). In the digitized Meisei University copy of the First Folio, we find extensive annotations by an early reader, William Johnstoune.

SHAKESPEARE’S EARLY READERS AND USERS

131

Johnstoune explicates passages as he reads, for instance, noting at the top of a page in Measure for Measure, ‘The Iudge plainelie solicites Isabell to lecherie she mentaines her Integritie with wittie reasons’ (2.4, sig. F5v; Yamada 2005). At the end of Measure for Measure, Johnstoune describes the ‘pleasant conclusions of the aduentures’ (5.1, sig. G6; Yamada 2005; see Figure 2.5.1). Today, editors and directors approach Measure for Measure as a ‘problem play’, specifically because of the problematic nature of the Duke’s proposal to Isabella and her silent response. Johnstoune’s annotation demonstrates that at least one early reader took the play as a straightforward comedy, ending in the promise of marriage with its ‘pleasant conclusions’. Rebecca Munson’s forthcoming digital project, Common Readers, moves beyond the Folio by proposing to catalogue all known annotations in early printed drama and takes as its starting point and test case early quartos of Shakespeare’s 1 Henry IV (see Munson 2019). The Shakespeare Quarto Archive (2009), likewise, aimed to produce ‘cover-to-cover reproductions and transcriptions’ of pre-1642 Shakespeare quartos: their copy-specific transcription and encoding processes means that they capture annotations and ownership marks such as library stamps. The Shakespeare Quartos Archive currently covers thirty-two copies of Hamlet from the first quarto (1603) to the fifth quarto (1637). Projects like these make readers’ marks in Shakespeare’s works more accessible than those in other literary and dramatic texts. Beyond the readers’ marks and annotations in early printed books, we also find handwritten evidence of readership beyond the books that were themselves read. In Tudor schools, students were taught to copy selections from the material they read into their notebooks. They were encouraged to organize these selections by subject

FIGURE 2.5.1  Annotations in Meisei University Library’s First Folio (1623), Measure for Measure 5.1, sig. G6r, detail. Reproduced with permission of Meisei University Library.

132

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

matter, copying them below relevant titles called ‘commonplace headings’. Readers could use pre-designated headings or choose their own. Commonplaces (also called sententiae) are generally short passages that are valued for offering a well-phrased idea. Mayer (2018) and Estill (2015a; 2015b) offer detailed case studies about manuscript commonplaces drawn from Shakespeare’s works. Not all selections from Shakespeare’s works are commonplaces, those short well-phrased aphorisms. For instance, full copies of some of Shakespeare’s sonnets circulated in manuscript, even before the sonnets were themselves printed (Marotti and Estill 2012, and see Shrank and Werstine in the present volume). Indeed, many longer dramatic speeches and songs also circulated in manuscript. At times, these longer theatrical texts are clearly taken from early printed versions of the play itself. Many dramatic miscellanies (manuscripts comprising primarily extracts from plays) include multiple selections from a given play, and some compilers, like Abraham Wright (1611–90), copied short commonplaces alongside full speeches  – in this case, Othello’s final speech, ‘I haue done [the] state some seruice’ with an added stage direction, ‘stabs himselfe’ (BL Add MS 22608, f. 84v; 5.2.337–54). Two seventeenth-century manuscripts include full copies of Iago’s ‘She that was ever fair and never proud’ speech (2.1.148–60). One manuscript now held at the University of Chicago includes only one page with selections copied from Shakespeare (MS 824, f. 113r, see Figure 2.5.2); this early reader chose to copy Polonius’s sententious advice to his son Laertes (including ‘Neither a borrower nor a Lender bee’, 1.3.74).6 Selections from plays copied into manuscript, that is, dramatic extracts, could be pithy and memorable – those ‘precepts’ (1.3.57) that Polonius advises Laertes to memorize and apply to other situations. Conversely, some dramatic extracts were linked directly to a moment in a given play, such as Othello’s final speech and suicide. A quotation in manuscript that we recognize to be from a play might not, however, have been copied directly from a printed playbook or from a performance of the play. Playwrights often used and adapted commonplaces and proverbial language in their writing; this does not mean a proverb found in a Shakespeare play was copied from a Shakespeare play directly.7 Furthermore, as Tiffany Stern (2009) has demonstrated, early modern plays were created by gathering existing documents such as prologues, epilogues and songs and ‘patching’ them together. As such, without additional evidence, manuscripts with copies of songs from plays, for instance, can often not be tied to the copyist reading or attending a play. Instead, they can be taken as part of the vibrant manuscript culture in which early modern plays were composed, ‘patched’/created, performed, watched, printed, read and received. Just as evidence of readership can appear both within and beyond a book, evidence of book collecting can appear in or on the book itself (such as an owner’s inscription, a bookplate or a monogrammed spine) or separate from the volume (such as book lists8). Some, but not all, evidence of ownership and collecting can be handwritten. And, of course, proof of collecting and ownership can often be incomplete: Lukas Erne notes that ‘playbooks are rarely mentioned in wills or probate inventories’ (2013: 218), which is just one example of a gap in the historical evidence. Similarly, even when we have evidence of a book’s ownership, that is not necessarily proof of reading it, as our book buying and reading habits today attest.9

SHAKESPEARE’S EARLY READERS AND USERS

133

FIGURE 2.5.2  Selections from Hamlet, University of Chicago Library, MS 824, f. 113r. Courtesy of the Special Collections Research Center, University of Chicago Library.

Some of the key explorations of early owners of Shakespeare’s plays are those by Alan Nelson (2005) and Erne (2013). Jeffrey Todd Knight (2013) has focused on how early buyers gathered their books into sammelbände, that is, volumes comprising multiple texts, which offer additional evidence of ownership even if an owner’s name is unknown. Tamara Atkin’s Reading Drama in Tudor England

134

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 2.5.3  Frances Wolfreston’s ownership mark on William Shakespeare, Venus and Adonis (1593), Bodleian Libraries, University of Oxford, Arch. G. e.31(2), title page. Courtesy of the Bodleian Libraries, University of Oxford.

SHAKESPEARE’S EARLY READERS AND USERS

135

(2018) serves as an exemplar of how to undertake thoughtful and thorough research with book lists, sammelbände, marginalia and ownership marks, though her focus is pre-Shakespearean drama. Broad definitions of collecting sometimes include considerations of anthologizing or extracting, particularly when focused on print.10 Finding aids about early modern collections and provenance vary widely, but one of the most useful is the Private Libraries in Renaissance England series (Fehrenbach, Black and Green 1992–), which is complemented by the Folger Shakespeare Library’s PLRE.Folger online site (Fehrenbach, Poston and Wolfe updated 2019). As of 2019, PLRE.Folger lists seven books by Shakespeare and offers additional information about the contexts of their collection. Other resources focus on individual book collectors.11 Research on Frances Wolfreston by Sarah Lindenbaum and others, for instance, demonstrates that Wolfreston (1607–77) owned at least thirteen Shakespeare quartos (Lindenbaum 2018b). Most notably, Wolfreston owned the ‘sole surviving copy of the 1593 Venus and Adonis’ (ibid.), writing ‘Frances Wolfreston hor bouk’ on its cover (see Figure 2.5.3).12 As an assiduous collector, Wolfreston preserved the earliest known printing of Shakespeare’s works. Personal libraries showcase not just one individual’s taste: they can also affect the survival rates of early books, which in turn shapes what we can study. As this chapter’s central case study will demonstrate, annotating, extracting and collecting were not separate processes for early readers: they were multiple and overlapping functions of a vibrant culture of thoughtful and responsive readers. Even the smallest piece of manuscript evidence can offer insight into how a work by Shakespeare was annotated, extracted or collected; ultimately, this kind of research points to a broader reception history.

CASE STUDY: WHAT WE CAN LEARN FROM A SINGLE COUPLET FROM PERICLES This chapter opened by outlining three various ways early readers interacted with Shakespeare’s plays: annotating/marking, extracting/commonplacing and collecting. In one of the copies of the first printing of George Chapman’s play An Humorous Day’s Mirth (1599) held at the Folger Shakespeare Library (STC 4987 copy 1), we have an example of these three kinds of evidence coinciding. At the end of Chapman’s play, an early reader, Thomas Bentley, has written an ownership mark as well as a brief selection from Shakespeare’s Pericles (see Figure 2.5.4): it is a reader’s mark, an ownership mark and an extract that is also a commonplace.13 This example shows how Bentley read or was otherwise exposed to Pericles while also offering evidence of ownership of Chapman’s play. This single page puts these two otherwise disparate plays in communication with each other. At the top of the inscribed page, someone has neatly written ‘In the name of God amen’. Below a small doodle appears the ownership mark in a different hand: ‘Thomas Bentley ownes this booke’14 and, likely in the same hand as Bentley’s, this couplet adapted from Pericles: ‘it is a foole that scanns / the Inward habitts by the / outward man’ (2.2.54–5). The couplet is attributed to ‘Shackesphere’, but not the play Pericles specifically. This sententious couplet was adapted from the end of Act 2,

136

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 2.5.4  Thomas Bentley’s ownership mark and quotation from Shakespeare’s Pericles in George Chapman, An Humorous Day’s Mirth (Q, 1599), sig. H2r. Folger Shakespeare Library, call number: STC 4987 copy 1. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License.

SHAKESPEARE’S EARLY READERS AND USERS

137

Scene 2, where, in the earliest printed version (Q1, 1609), King Simonides declares, ‘Opinion’s but a foole, that makes vs scan / The outward habit, by the inward man’ (sig. C4v).15 Simonides speaks this couplet after having seen Pericles for the first time. Before this point in the play, Pericles has been shipwrecked on the shores of Pentapolis, where nobody knows he is the Prince of Tyre. In his rusty armour, Pericles joins a tournament to win the hand of King Simonides’ daughter, Thaisa. When Simonides makes this remark, he has not yet seen Pericles fight – and the Prince of Tyre does indeed go on to win the tournament and marry Thaisa. At this point in the play, Simonides has only seen Pericles’ rusted armour and sad device, ‘a withered branch’ (2.2.42). The other knights who are to fight in the tourney mock Pericles’ appearance. In the printed play, Simonides chides the knights for judging a man’s clothes (‘outward habit’) by his personality (‘inward man’). Many versions of this maxim, including Bentley’s manuscript extract, present the aphorism reversed: don’t judge a person by their clothes, or, as one of the still-popular phrasings states, don’t judge a book by its cover. George Wilkins, widely accepted to have co-written Pericles with Shakespeare,16 also published a prose version of the story, The Painfull Aduentures of Pericles Prince of Tyre (1608).17 The Painfull Aduentures’ Simonides makes the meaning clear: ‘the King mildely reproouing [the knights] for, hée tolde them, that as Uertue was not to be approoued by wordes, but by actions, so the outward habite was the least table of the inward minde’ (sig. D1v). In this prose version, virtue emerges from action, not clothing or words. These simple lines, jotted into a playbook of Chapman’s comedy, offer insights into Shakespeare’s writing process, the meaning of this play, how Shakespeare should be represented and Shakespeare’s audiences then and now. When considered in light of Shakespeare’s writing process, this manuscript extract highlights how Shakespeare and Wilkins, like other writers of their day, drew on existing commonplaces (Dent 1981; 1984; Rhodes 2006). Shakespeare’s contemporary writers, like John Lyly and Christopher Marlowe, also used similar phrasings in their works. In Euphues, Lyly asks, ‘If we respect more the outward shape, then the inwarde habit, good God into how many mischiefs doe we fall?’ (1578: sig. C3v). Marlowe’s Tamburlaine observed: ‘Noble and milde this Persean seemes to be, / If outward habit iudge the inward man’ (1590: sig. B1r). Lyly advocates not judging a person by their clothes, whereas Marlowe’s Tamburlaine does just that. The idea of inward or outward habits (clothes or character) circulated with two contradictory morals in Shakespeare’s day: simply put, either we can judge a person by their exterior or we can’t. Shakespeare used both morals in his works: R. W. Dent documents instances in All’s Well that Ends Well, King Lear, Hamlet and Cymbeline where Shakespeare draws on a version of ‘the clothes make the man’ (1981: A283), including Polonius’s apothegm ‘the apparel oft proclaims the man’ (Hamlet, 1.3.71). Conversely, much of the action of the play Pericles (especially Act 2) centres on the tension between appearance and character: Pericles is a noble prince, despite being dressed in rusty armour. Hallett Smith contends that ‘the intended meaning [of the “inward habits” couplet] is clear’: that is to say, Simonides will not judge Pericles based only on his

138

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

outward appearance (1997: 2.2.56–7n.). The Bentley (mis-)quotation leads us to focus on exactly this theme and to find the cultural resonances of these two lines in particular from all of Shakespeare’s oeuvre. Of course, Shakespeare might have not even written these lines – despite the fact that they are attributed to him again and again over time, from their original printing to their twenty-first-century stagings. The title page of the play’s first-known publication clearly announces ‘By William Shakespeare’ and makes no mention of Wilkins. Similarly, modern editions of the play and of Shakespeare’s complete works tend to list only Shakespeare as the author of Pericles on their title pages and covers.18 Scholarly consensus, however, suggests that at least the first two acts – including Simonides’ couplet copied into An Humorous Day’s Mirth – were written by Wilkins (Jackson 2003). Bentley’s early attribution to ‘Shackesphere’, then, is in line with centuries of erroneous attribution history, as Wilkins, not Shakespeare, is the couplet’s likely author. Regardless of who wrote the couplet for its original performance (Shakespeare and/or Wilkins), there is also the question of whether the couplet was originally envisioned by its author(s) in the way it was published. The first quarto of Pericles is widely regarded as a corrupt text, but because Pericles was not published in the First Folio, Q1 is the only substantive witness: all later texts derive from its printing. The line as printed, meaning the exact opposite of what the scene intends, could be an error introduced by a compositor confusing the line with one of the other idioms that circulated at the time;19 other editors suggest that it was misremembered by someone dictating or transcribing the scene from performance (Gossett 2004: 2.2.55n.; Warren 2003: 73). The handwritten quotation from Pericles in Chapman’s play, then, highlights the meaning of the couplet and the (lack of) textual authority even of published plays, and furthermore raises the question of how this crux should be approached by editors and directors.20 Thematically, the pre-tournament scene in Pericles reflects the ‘don’t judge a book by its cover’ meaning, so modern editors often suggest changing Simonides’ lines in the play so their meaning clearly reflects that which is implied. One change adopted by many editors is to replace ‘by’ with ‘for’, as suggested by Ernest Schanzer (1965), so Simonides’ lines read ‘It is a fool that scans / The outward habit for the inward man’.21 George Steevens, like Bentley, recommended switching the words ‘inward’ and ‘outward’, so the king would warn against measuring ‘the inward habit, by the outward man’ (Johnson and Steevens 1809: 171n.2).22 Steevens noted that in one of Shakespeare’s sources for the play, Heinrich Steinhöwel’s prose romance Kynge Appolyn of Thyre (1471, printed in English translation in 1510), the king uses the phrase ‘habyte maketh not the relygyous man’ (sig. C2r). Bentley, and perhaps Shakespeare/Wilkins, do not use the word ‘habit’ to mean simply ‘clothes’, however; Bentley’s proposed emendation multiplies the meaning of ‘habit’ to mean also personality or inner qualities.23 When editors are aware of manuscript evidence, even non-authorial evidence like Bentley’s quotation, it can influence how plays are edited and published, which, in turn, influences how plays are performed to this day – and

SHAKESPEARE’S EARLY READERS AND USERS

139

Bentley’s commonplace from Pericles is one of the few that has received this attention (Mowat and Werstine 2005; Gossett 2004). Bentley’s manuscript extract also leads us to consider how early readers and playgoers approached Pericles. For Katherine Duncan-Jones, ‘Bentley’s garbled version’ of the couplet ‘suggests recollection of a performance’ (2001: 348n.28). Some playgoers, indeed, copied selections of plays during performances,24 but Bentley’s couplet could also have been adapted from a printed source, as readers copying extracts regularly changed what they copied (Estill 2015a; 2015b). That Bentley jotted this selection into another playbook suggests he was a reader with multiple early modern plays to hand or at least an interest in published playbooks. Although we do not know if Bentley was copying from print or performance, his extract is, as Duncan-Jones (2001) notes, part of the evidence of the early reception of Pericles. Duncan-Jones explains that ‘the evidence that [Pericles] was hugely and immediately successful … is overwhelming’ (204) and that during Shakespeare’s life the play was ‘notoriously and continuously popular’ (205). It was also the first Shakespeare play known to have been performed after the reopenings of the theatres in the Restoration (Smith 1997: 1530). DEx: A Database of Dramatic Extracts lists dozens of seventeenth-century extracts from Pericles, including from two manuscript compilers who each copied over twenty selections from the play (Estill and Montedoro 2015). The Pericles extract speaks not only to the play’s early popularity, but also to Shakespeare’s continued importance for book collectors. Moreover, Bentley’s extract from Pericles adds value to this specific playbook copy of Chapman’s play. A later bookplate indicates that sometime after Bentley, Sir Percy Hetherington Fitzgerald (1834–1925) owned the book. As Fitzgerald’s bookplate announces, he was a member of both the Athenaeum Club and the Garrick Club, which, in Fitzgerald’s words, ‘have the most interesting and dramatic associations. They seem like stages across which all that has been distinguished in drama, letters, music and art generally, have passed or streamed through their chambers’ (1904: ix). The Garrick Club, in particular, encouraged members to collect plays in their personal libraries and to add to the club’s extensive library: for Fitzgerald, buying this book could have been an act of coterie belonging. In 1907, Henry Clay Folger, who purposefully curated his collection around Shakespeare, bought this volume, which is how it now comes to be in the Folger Shakespeare Library.25 While early collectors are a valuable source of information, the entire provenance of a playbook or manuscript speaks to its changing value over history. Today, a quotation attributed to Shakespeare is what attracts many people to this copy of An Humorous Day’s Mirth playbook: it is why this playbook is featured on the Folger Shakespeare Library’s Shakespeare Documented site (Wolfe 2016). The presence of a Shakespearean – or Shakespeare-adjacent, as the case may be – manuscript extract is both evidence in the ongoing reception of Pericles and a factor in how we value early plays and playbooks such as this inscribed copy of An Humorous Day’s Mirth. And, thanks to the Folger’s mission to serve the public, we know of and can study this handwritten evidence: but many other extracts, annotations and ownership marks, particularly those in private libraries, are yet to be explored.

140

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

RE-EVALUATING OUR RESEARCH PARADIGMS Handwritten annotations, extracts and ownership marks usually exist in only a single copy. The more evidence of reading, extracting and collecting we can find, the more complete a picture we can draw, but this will still be limited by the loss rates of historical documents (Pettegree 2016; Hill 2018). The survival rate for manuscripts, those unique handwritten artefacts, is of course much lower than that for printed books (May and Wolfe 2010: 133). Furthermore, even for existing evidence, our current library catalogues and scholarly search aids often make it easier to find material relating to Shakespeare than other writers, which doesn’t accurately reflect the primary sources themselves (Estill 2019). There is, as of yet, no way to search full texts of early modern manuscripts unless they have been painstakingly transcribed by hand, whereas many printed materials are searchable through Early English Books Online – Text Creation Partnership, Literature Online (a paywalled resource), The Internet Archive and Google Books.26 Likewise, comprehensive databases of ownership marks or marginalia are still not available, and would require extensive resources to create. Any claims we make based on existing evidence, then, need to be qualified: our conclusions are drawn from what survives and what we can find. The types of evidence considered in this chapter – primarily annotating, commonplacing and collecting – require researchers to make decisions about the boundaries and focus of their projects. The research method chosen will, of course, circumscribe the conclusions that can be reached: research that starts and ends with Shakespeare will yield conclusions that focus on Shakespeare and might not be representative of a broader cultural movement or given historical moment. A researcher who uncovers an early extract from Shakespeare will certainly be able to publish a scholarly article about the manuscript, whereas extracts from lesserknown authors often go unrecognized or unanalysed. This chapter, with its focus on an extract from Shakespeare, is not an exception to the rule. Rather than analysing Chapman’s play An Humorous Day’s Mirth and how the Shakespeare extract nuances our understanding of Chapman, this chapter has focused on Pericles. Similarly, this analysis does not turn to the known extracts from Chapman’s other plays (Caesar and Pompey, Sir Giles Goosecap, All Fools and The Blind Beggar of Alexandria), nor does it scrutinize Chapman’s multiple inscriptions in books.27 When it comes to considering the reception history of early modern drama and poetry, Shakespeare has been over-represented in our scholarship about annotating, commonplacing and collecting. Our scholarly fascination with Shakespeare stems from his status as a cultural icon: that is to say, it is not just scholars who position Shakespeare above his contemporary writers. The manuscript featured on Antiques Roadshow was singled out for attention not because of its rarity, but because of our continued fixation on Shakespeare above other playwrights from his time. It is hard to imagine an internationally covered news story about a manuscript containing selections from, say, Fulke Greville’s dramatic work. When assessing extant manuscript evidence from 1590 to 1650, it becomes clear that early readers did not position Shakespeare above his contemporaries. Early manuscript compilers copied selections from

SHAKESPEARE’S EARLY READERS AND USERS

141

anonymous or now understudied plays alongside plays we now consider canonical. For instance, one mid-seventeenth-century manuscript compiler, Abraham Wright, copied selections from more plays by James Shirley than by Shakespeare (Estill 2015b: 83). Focusing only on manuscripts that include extracts from Shakespeare – or considering those extracts in a vacuum – belies the richness of extant archival evidence about the early modern reception of drama. Just as scholarship can misrepresent early archival evidence by focusing on Shakespeare to the exclusion of other writers, some studies privilege early moments of readership and ownership over later evidence when analysing readers’ marks and extracts. This can, at times, speak to the desire to find an ideal imagined reader: one whose attitudes towards a text we can take as definitive and representative. Often, this reader is one who experiences the work at its original moment of publication or performance. Of course, as we know from our own experiences as readers, this is not the only way to interpret literature: we often read books well after their original publication – and in our literature classes, we regularly read works centuries after they were first written and disseminated. Later moments of readership and ownership are equally significant to understanding the reception of a literary or theatrical work. The actual evidence of annotations and extracts themselves (both early and late) point to the importance of remembering the radically individual nature of all these responses: they represent a given reader’s (or playgoer’s) thoughts and reactions as well as a particular moment in time. Turning to later moments of reception and production can reveal changing tastes over time and suggest how we have arrived at our practices in theatre, pedagogy, textual studies and literary criticism today.28 Research on collecting can take a synchronic or a diachronic approach: that is to say, the focus can be on a given collection at a particular point in history (synchronic) or on the provenance of individual books and how they passed through different hands over time (diachronic). While much research will combine some of these elements, it is impossible to undertake both completely: that is to say, it is beyond the purview of most projects to offer an overview of a large collection as well as a description of how each of its books came to be in that collection and where they are now. A focus on playbooks alone ignores a wealth of other texts; a focus on literary texts applies anachronistic generic categories to early books. Moreover, a focus on books alone can elide other things that a collector valued, including artworks and memorabilia.29 Some research on collecting focuses on how the works of famous authors, such as Shakespeare, were bought and owned. Just as it has been a focal point for research on annotation, the First Folio has received particular attention as a book that was bought, sold and collected (West 2001–3; West and Rasmussen 2012; Smith 2016: 24–120; Galbraith 2016; Mayer 2016). The Shakespeare Census (Hooks and Lesser 2018) project has a similarly Shakespearean focus, but instead seeks to determine the current location of all pre-1700 non-folio editions of Shakespeare and, where possible, link to a digital facsimile. Focusing on how the works of a given author, like Shakespeare, have been collected can make a strong contribution to author-centric studies, but does not describe the range of collecting activities at a given historical moment. Just as with annotating and extracting, there can be a compulsion to value

142

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

the study of early collections as somehow more important than later collections. Yet, when focusing on Shakespeare in particular, it is often the later collectors who bought Shakespeare for the sake of Shakespeare: these later Shakespeare-centric collections, such as Henry and Emily Folger’s, speak to the changing cultural value associated with Shakespeare. Other research on collecting has focused not just on prominent authors but also on well-known individual collectors, who are often of note for their non-collecting achievements, such as William Drummond of Hawthornden (1585–1649), a writer in his own right, or David Garrick (1717–79), the pre-eminent Shakespearean actor of his day.30 Research on early modern women collectors is thriving,31 but there is still more work to be done. Future research will have to establish methods that decentre our focus from named collectors. Critically evaluating our research paradigms will only improve the conclusions we can draw about early readers, book owners and reception. Each of the different pieces of handwritten evidence surveyed in this chapter can offer valuable insights when taken separately. Even the smallest piece of evidence, such as the couplet from Pericles discussed at length, can shed light on early modern reception history, our textual interpretations and assumptions, and how books, by Shakespeare and others, have been valued historically and to this day. Annotations, extracts and ownership marks are all pieces of the puzzle of material evidence that we, as scholars, attempt to reassemble from the incomplete historic record; the chapters in this collection add additional pieces to this puzzle. No single project or analysis will be able to form a complete picture of this evidence from its earliest days to the present. The most thoughtful research yet to be undertaken on the reception of early modern drama by Shakespeare and his contemporaries will question the canonical and chronological boundaries of existing scholarly resources and expand our basis for exploration. There is much handwritten evidence yet to be discovered, transcribed, recognized and analysed. Even if we could imagine recovering and studying every readers’ mark, every extract and the history of every book, collection and collector, we would not be able to draw monolithic conclusions about Shakespeare’s early readers and users, who were, just as we are today, a varied and interesting lot. We can, however, turn to handwritten evidence to continue to add pieces to the puzzle.

NOTES

I would like to thank Tara Lyons and Kailin Wright for their thoughtful feedback. This research was made possible by the Canada Research Chair program. 1. The manuscript was also discussed on Antiques Roadshow by Tiffany Stern in the following year (BBC, series 40, episode 19, 22 July 2018; not available online). 2. Examples of important work on printed commonplace markers and commonplaces with a focus on Shakespeare include Lesser and Stallybrass (2008); Stallybrass and Chartier (2007); and Erne (2013: esp. ‘Early Readers and Commonplacers of Shakespeare in Print’, 224–30). 3. Bess of Hardwick’s Letters (Wiggins 2013) also offers good transcription practice.

SHAKESPEARE’S EARLY READERS AND USERS

143

4. For more on the range of annotations and the challenges with their interpretation, see Massai (2012: esp. ‘Annotating Readers’, 149–56). 5. For scholarship specifically discussing annotations in Shakespeare’s First Folio, see Bourne (2019); Mayer (2014, 2015, 2018: esp. 44–105); and Smith (2016: esp. 121–82). 6. Thanks to Megan Heffernan for her assistance in researching this manuscript. 7. Estill (2015b: 201–24) offers an example of a proverb Shakespeare used in Love’s Labour’s Lost that was copied from intermediary sources and not from the play itself. 8. On the different kinds of book lists in early modern Europe, see Walsby (2013). 9. Indeed, Smith (2016) differentiates her chapters on ‘Owning’ and ‘Reading’ the First Folio. In his overview of British book ownership from 1650 to 1740, Mandelbrote cautions against conflating increased book ownership with increased readership: ‘At all levels of society many more books were owned, but it may be that proportionately fewer were being read’ (2006: 189). 10. For more on anthologization, see Lopez (2014). 11. See, for instance, Pearson (2020) and the Dictionary of Literary Biography volumes on book collectors (such as Baker and Womack 1999). 12. A facsimile of the cover of this quarto is available online from Shakespeare Documented: https://shakespearedocumented.folger.edu/exhibition/document/venusand-adonis-first-edition (accessed 13 November 2020). 13. A facsimile of the complete book is available online from the Folger Shakespeare Library: luna.folger.edu/luna/servlet/s/3s878o (accessed 13 November 2020). This image is taken from Shakespeare Documented (Wolfe 2016). Thanks to Meaghan Brown and Abbie Weinberg for their help in researching this volume. 14. In the Folger Shakespeare Library’s online catalogue record for this volume (hamnet. folger.edu/cgi-bin/Pwebrecon.cgi?BBID=160690, accessed 13 November 2020), the owner/annotator is listed as Thomas Bemley, though the high-quality digital facsimile suggests it was, indeed, Bentley. Barbara Mowat and Paul Werstine note that ‘The Folger has not determined the identity of Thomas Bentley’ (2005: 193). In manuscript transcriptions, supplied letters are in italics. 15. All early printed editions are derived from this first quarto and share its wording. 16. For an overview of scholarship about Wilkins’s collaboration with Shakespeare on Pericles, see Gossett (2004: 62–9 and 161–3). 17. Although Wilkins is generally accepted to have written the Painfull Aduentures, Wilkins’s authorship is not certain, as Egan points out (2013: 49–50). Wells et al. note that ‘the pamphlet openly professes to be based on the play’ (1987: 557); despite being published earlier than the play’s first-known printing, the prose version is based on earlier performances. 18. Loughnane’s edition in The New Oxford Shakespeare (2017) is a notable exception that lists Wilkins’s name alongside Shakespeare’s on the title page.

144

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

19. On compositors and the printing of Pericles, see Wells et al. (1987: esp. 556) and Loughnane (2017: esp. 1343–9). 20. Munro (2016) offers a valuable history of the scholarly reception and dating of Pericles, with an extended discussion of how editors and directors have used parts of The Painfull Aduenture to clarify the play’s meanings. Egan (2013) describes the range of editorial interventions, with particular attention to how editors have dealt with the impresa (mottos) in 2.2. 21. Twenty-first-century editions that use this emendation include Warren (2003); Gossett (2004); Loughnane (2017); and Bishop (2019). 22. This note includes Richard Farmer’s comment that ‘In my copy this line is quoted in an old hand as Mr. Steevens reads’ (Johnson and Steevens 1809: 171n.2). 23. The OED notes two applicable definitions of ‘habit’ in use in Shakespeare’s life: ‘The way in which a person is mentally or morally constituted; the sum of the mental and moral qualities; mental constitution, disposition, character’ (‘habit’, n.III.8) and ‘Bearing, demeanour, deportment, behaviour; posture’ (‘habit’, n.II.4). 24. For an example of an early playgoer, Edward Pudsey, copying selections from a performance of Othello into manuscript, see Duncan-Jones (2011: esp. 72–83). For examples of fictional accounts of copying selections from plays from both performance and print, see Stern (2004: 20–1). 25. Abbie Weinberg, Folger Shakespeare Library, email communication, July 2019. For more on Henry and Emily Folger as collectors, see Grant (2014) and Mays (2015). 26. The Folger Shakespeare Library’s Early Modern Manuscripts Online is one of the most promising large-scale transcription projects of early modern English manuscripts. 27. For information about these extracts and inscriptions, see Beal (2009) and Estill and Montedoro (2015). 28. A strong example of this kind of scholarship is Depledge and Kirwan (2017), with its focus on printing from 1640 to 1740. 29. For more on this and other interpretive difficulties with early modern evidence on collection, see Rial (2010). 30. Entries on both Drummond and Garrick appear in Baker and Womack (1999). For an example of research on Drummond’s library, see Hall (2008). For an extensive bibliography and analysis of Garrick’s library, see Kahrl with Anderson (1982). 31. See, for instance, Sae (2017) and Knight, White and Sauer (2018). See also digital resources such as Empey et al.’s Early Modern Female Book Ownership site (2018), which includes extensive resources for further reading, and Lindenbaum’s Frances Wolfreston Hor Bouks (2018a).

REFERENCES Acheson, Katherine, ed. (2019), Early Modern English Marginalia, New York: Routledge. Atkin, Tamara (2018), Reading Drama in Tudor England, London: Routledge.

SHAKESPEARE’S EARLY READERS AND USERS

145

Baker, William and Kenneth Womack (1999), Pre-Nineteenth-Century British Book Collectors and Bibliographers, Detroit: Gale. Beal, Peter (2008), A Dictionary of English Manuscript Terminology 1450–2000, Oxford: Oxford University Press. Beal, Peter, comp. (2009), Catalogue of English Literary Manuscripts 1450–1700, Arts and Humanities Research Council. Available online: celm-ms.org.uk/ (accessed 13 November 2020). Bishop, Tom, ed. (2019), Pericles, in Internet Shakespeare Editions. Available online: https://internetshakespeare.uvic.ca/Library/Texts/Per/index.html (accessed 13 November 2020). Bourne, Claire M. L. (2017), ‘Marking Shakespeare’, Shakespeare, 13 (4): 367–86. Bourne, Claire M. L. (2019), ‘Vide Supplementum: Early Modern Collation as PlayReading in the First Folio’, in Katherine Acheson (ed.), Early Modern English Marginalia, 195–233, New York: Routledge. Cerasano, S. P. (2005), ‘An Inventory of Theatrical Apparel (c. 1601/2)’, in Grace Ioppolo (dir.), The Henslowe-Alleyn Digitisation Project, London: King’s College London. Available online: https://henslowe-alleyn.org.uk/essays/an-inventory-of-theatricalapparel-c-16012/ (accessed 13 November 2020). Dent, R. W. (1981), Shakespeare’s Proverbial Language: An Index, Berkeley: University of California Press. Dent, R. W. (1984), Proverbial Language in Drama Exclusive of Shakespeare, 1495–1616: An Index, Berkeley: University of California Press. Depledge, Emma and Peter Kirwan, eds (2017), Canonising Shakespeare: Stationers and the Book Trade, 1640–1740, Cambridge: Cambridge University Press. Duncan-Jones, Katherine (2001), Ungentle Shakespeare: Scenes from his Life, London: Thomson Learning. Duncan-Jones, Katherine (2011), Shakespeare: Upstart Crow to Sweet Swan: 1592–1623, London: Methuen Drama. Egan, Gabriel (2013), ‘The Presentist Threat to Editions of Shakespeare’, in Cary DiPietro and Hugh Grady (eds), Shakespeare and the Urgency of Now: Criticism and Theory in the 21st Century, 38–59, Basingstoke: Palgrave Macmillan. Empey, Mark, Erin McCarthy, Georgianna Ziegler, Tara Lyons, Micheline White, Sarah Lindenbaum and Martine van Elk (2018), Early Modern Female Book Ownership: #HerBook. Available online: https://earlymodernfemalebookownership.wordpress.com (accessed 13 November 2020). Erne, Lukas (2013), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Estill, Laura (2015a), ‘Commonplacing Readers’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 149–62, Cambridge: Cambridge University Press. Estill, Laura (2015b), Dramatic Extracts in Seventeenth-Century English Manuscripts: Watching, Reading, Changing Plays, Newark, DE: University of Delaware Press. Estill, Laura (2019), ‘Shakespearean Extracts, Manuscript Cataloguing, and the Misrepresentation of the Archive’, in Tiffany Stern (ed.), Rethinking Theatrical Documents in Shakespeare’s England, 175–92, London: Bloomsbury.

146

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Estill, Laura and Beatrice Montedoro, eds (2015), DEx: A Database of Dramatic Extracts. Available online: https://dex.itercommunity.org/ (accessed 13 November 2020). Fehrenbach, R. J., Joseph L. Black and E. S. Leedham-Green, eds (1992–), Private Libraries in Renaissance England, 9 vols (in progress), Tempe, Arizona: Medieval and Renaissance Text Studies. Fehrenbach, R. J., Michael Poston and Heather Wolfe (updated 2019), PLRE.Folger, Washington, DC: Folger Shakespeare Library. Available online: https://plre.folger.edu (accessed 13 November 2020). Fitzgerald, Percy Hetherington (1904), The Garrick Club, London: E. Stock. Galbraith, Steven K. (2016), ‘Collectors’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 137–54, Cambridge: Cambridge University Press. Gossett, Suzanne, ed. (2004), Pericles, London: Thomson Learning. Grant, Stephen H. (2014), Collecting Shakespeare: The Story of Henry and Emily Folger, Baltimore: Johns Hopkins University Press. Hall, John (2008), ‘William Drummond of Hawthornden: Book Collector and Benefactor of Edinburgh University Library’, in Rae Earnshaw and John Vince (eds), Digital Convergence: Libraries of the Future, 345–57, London: Springer. ‘Handwritten Notes on Shakespeare’ (2017), Antiques Roadshow, series 39, episode 16, BBC, 29 March. Available online: http://www.bbc.co.uk/programmes/p04yf9ng (accessed 13 November 2020). Hill, Alexandra (2018), Lost Books and Printing in London 1557–1640: An Analysis of the Stationers’ Company Register, Leiden: Brill. Hooks, Adam G. and Zachary Lesser, eds (2018), Shakespeare Census. Available online: http://www.shakespearecensus.org (accessed 13 November 2020). Ingleby, C. M., L. Toulmin Smith and F. J. Furnival, comps, and John Munro, ed. (1909), The Shakspere Allusion-Book: A Collection of Allusions to Shakspere from 1591–1700, 2 vols, London: Chatto & Windus. Ioppolo, Grace, dir. (2005), The Henslowe-Alleyn Digitisation Project. Available online: http://www.henslowe-alleyn.org.uk (accessed 13 November 2020). Jackson, MacDonald P. (2003), Defining Shakespeare: ‘Pericles’ as Test Case, Oxford: Oxford University Press. Johnson, [Samuel] and [George] Steevens, eds (1809), The Plays of William Shakespeare, vol. 17: Titus Andronicus, Pericles, Glossarial Index, Philadelphia: Conrad. Kahrl, George M., with Dorothy Anderson (1982), The Garrick Collection of Old English Plays: A Catalogue with an Historical Introduction, London: The British Library, 1982. Knight, Jeffrey Todd (2013), Bound to Read: Compilations, Collections, and the Making of Renaissance Literature, Philadelphia: University of Pennsylvania Press. Knight, Leah, Micheline White and Elizabeth Sauer, eds (2018), Women’s Bookscapes in Early Modern Britain: Reading, Ownership, Circulation, Ann Arbor, MI: University of Michigan Press. Lesser, Zachary and Peter Stallybrass (2008), ‘The First Literary Hamlet and the Commonplacing of Professional Plays’, Shakespeare Quarterly, 59 (4): 371–420.

SHAKESPEARE’S EARLY READERS AND USERS

147

Lindenbaum, Sarah (2018a), Frances Wolfreston Hor Bouks. Available online: https:// franceswolfrestonhorbouks.com (accessed 13 November 2020). Lindenbaum, Sarah (2018b), ‘Written in the Margent: Frances Wolfreston Revealed’, The Collation, Washington, DC: Folger Shakespeare Library. Available online: https:// collation.folger.edu/2018/06/frances-wolfreston-revealed/ (accessed 13 November 2020). Lopez, Jeremy (2014), Constructing the Canon of Early Modern Drama, Cambridge: Cambridge University Press. Loughnane, Rory, ed. (2017), Pericles, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, 1.1340–2, Oxford: Oxford University Press. Mandelbrote, Giles (2006), ‘Personal Owners of Books’, in Giles Mandelbrote and K. A. Manley (eds), The Cambridge History of Libraries in Britain and Ireland, Volume 2: 1640–1850, 173–89, Cambridge: Cambridge University Press. Marotti, Arthur F. and Laura Estill (2012), ‘Manuscript Circulation’, in Arthur F. Kinney (ed.), The Oxford Handbook of Shakespeare, 53–70, Oxford: Oxford University Press. Massai, Sonia (2012), ‘Early Readers’, in Arthur F. Kinney (ed.), The Oxford Handbook of Shakespeare, 143–61, Oxford: Oxford University Press. May, Steven W. and Heather Wolfe (2010), ‘Manuscripts in Tudor England’, in Kent Cartwright (ed.), A Companion to Tudor Literature, 135–9, Malden, MA: WileyBlackwell. Mayer, Jean-Christophe (2014), ‘First Folio Readers’ Marks: Monumentalizing Shakespeare and Empowering the Self’, Litteraria Pragensia: Studies in Literature and Culture, 24 (47): 4–26. Mayer, Jean-Christophe (2015), ‘The Saint-Omer First Folio: Perspectives on a New Shakespearean Discovery’, Cahiers Élisabéthains 87: 7–20. Mayer, Jean-Christophe (2016), ‘Early Buyers and Readers’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 103–19, Cambridge: Cambridge University Press. Mayer, Jean-Christophe (2018), Shakespeare’s Early Readers: A Cultural History from 1590 to 1800, Cambridge: Cambridge University Press. Mays, Andrea (2015), The Millionaire and the Bard: Henry Folger’s Obsessive Hunt for Shakespeare’s Folio, New York: Simon and Schuster. McCarthy, Erin (2017), ‘Manuscript Copy of Lines 229–40 from Venus and Adonis’, in Heather Wolfe (cur.), Shakespeare Documented, Washington, DC: Folger Shakespeare Library. Available online: https://shakespearedocumented.folger.edu/ exhibition/document/manuscript-copy-lines-229-40-venus-and-adonis (accessed 13 November 2020). Mowat, Barbara and Paul Werstine, eds (2005), Pericles, New York: Washington Square Press. Munro, Lucy (2016), ‘Young Shakespeare/Late Shakespeare: The Case of Pericles’, in Laetitia Sansonetti (ed.), Jeunesse(s) de Shakespeare/Young Shakespeare: Actes du congrès 2015 de la Société Française Shakespeare 31. Available online: https://journals. openedition.org/shakespeare/3668 (accessed 13 November 2020).

148

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Munson, Rebecca (2019), ‘Mongrel Forms: Print-Manuscript Hybridity and Digital Methods in Annotated Plays’, in Tamara Atkin and Laura Estill (eds), Early British Drama in Manuscript, 345–61, Turnhout: Brepols. Nelson, Alan H. (2005), ‘Shakespeare and the Bibliophiles: From the Earliest Years to 1616’, in Robin Myers, Michael Harris and Giles Mandelbrote (eds), Owners, Annotators, and the Signs of Reading, 49–73, London: British Library. Pearson, David (2020), Book Owners Online. Bibliographical Society Electronic Publications. Available online: https://bookowners.online (accessed 17 November 2020). Pettegree, Andrew (2016), ‘The Legion of the Lost: Recovering the Books of Early Modern Europe’, in Andrew Pettegree and Flavia Bruni (eds), Lost Books: Reconstructing the Print World of Pre-Industrial Europe, 1–27, Leiden: Brill. Rhodes, Neil (2006), ‘Shakespeare’s Sayings’, in Stuart Gillespie and Neil Rhodes (eds), Shakespeare and Elizabethan Popular Culture, 155–73, London: Thomson Learning. Rial, Benito (2010), ‘Sixteenth-Century Private Book Inventories and Some Problems Related to their Analysis’, Library & Information History, 26 (1): 70–82. Sae, Kitamura (2017), ‘A Shakespeare of One’s Own: Female Users of Playbooks from the Seventeenth to the Mid-Eighteenth Century’, Palgrave Communications 3: 17021. Available online: https://doi.org/10.1057/palcomms.2017.21 (accessed 13 November 2020). Schanzer, Ernest, ed. (1965), Pericles, New York: Signet. Shakespeare Quartos Archive (2009), Bodleian Library, Folger Shakespeare Library, Maryland Institute for Technology in the Humanities, et al. Available online: quartos. org (accessed 31 July 2019); site now archived at https://wayback.archive-it.org/org467/20191016094633/http://quartos.org/ (archived site accessed 13 November 2020). Sherman, William H. (2008), Used Books: Marking Readers in Renaissance England, Philadelphia: University of Pennsylvania Press. Smith, Emma (2016), Shakespeare’s First Folio: Four Centuries of an Iconic Book, Oxford: Oxford University Press. Smith, Hallett, ed. (1997), Pericles, in G. Blakemore Evans (ed.), The Riverside Shakespeare, 2nd edn, 1527–64, Boston, MA: Houghton Mifflin. Stallybrass, Peter and Roger Chartier (2007), ‘Reading and Authorship: The Circulation of Shakespeare 1590-1619’, in Andrew Murphy (ed.), A Concise Companion to Shakespeare and the Text, 35–56, Oxford: Blackwell. Stern, Tiffany (2004), Making Shakespeare: From Page to Stage, London: Routledge. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. TNA: The National Archives (n.d.), Palaeography: Reading Old Handwriting, 1500–1800. Available online: http://www.nationalarchives.gov.uk/palaeography (accessed 13 November 2020). Walsby, Malcolm (2013), ‘Book Lists and Their Meaning’, in Malcolm Walsby and Natasha Constantinidou (eds), Documenting the Early Modern Book World: Inventories and Catalogues in Manuscript and Print, 1–24, Leiden: Brill. Warren, Roger, ed. (2003), Pericles, The Oxford Shakespeare, Oxford: Oxford University Press.

SHAKESPEARE’S EARLY READERS AND USERS

149

Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery (1987), William Shakespeare: A Textual Companion, Oxford: Oxford University Press. Werner, Sarah (2016a) ‘Digital First Folios’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 170–84, Cambridge: Cambridge University Press. Werner, Sarah (2016b), ‘Digitized First Folios’, Wynken de Worde blog. Available online: https://sarahwerner.net/blog/digitized-first-folios/ (accessed 13 November 2020). West, Anthony James (2001–3), The Shakespeare First Folio: The History of the Book, 2 vols, Oxford: Oxford University Press. West, Anthony James and Eric Rasmussen, eds (2012), The Shakespeare First Folios: A Descriptive Catalogue, Basingstoke: Palgrave Macmillan. Wiggins, Alison, project leader (2013), Bess of Hardwick’s Letters: The Complete Correspondence c. 1550–1608. Available online: https://www.bessofhardwick.org/ (accessed 13 November 2020). Wiggins, Martin with Catherine Richardson (2012–), British Drama 1533–1642: A Catalogue, 10 vols, Oxford: Oxford University Press. Wolfe, Heather, cur. (2016), Shakespeare Documented, Washington, DC: Folger Shakespeare Library. Available online: shakespearedocumented.folger.org (accessed 13 November 2020). Yamada, Akihiro (2005), The First Folio of Shakespeare: A Transcript of Contemporary Marginalia. Available online: http://shakes.meisei-u.ac.jp/e-note.html (accessed 13 November 2020). Zurcher, Andrew, ed. (n.d.), English Handwriting Online 1500–1700: An Online Course, COPIA: CERES (Cambridge English Renaissance Electronic Service). Available online: https://www.english.cam.ac.uk/ceres/ehoc/ (accessed 13 November 2020).

CHAPTER 2.6

The Shakespeare canon from the sixteenth to the twenty-first century PETER KIRWAN

THE WEIGHT OF THE CANON In a recent provocation, Brandi Adams compares the authorial canon to the sun at the centre of the Copernican universe. [T]he sun, or our star author, with his inevitable gravitational pull, orders the universe in such a way that it seems perfectly formed. However, this very order regrettably disallows most other stars and planets from being anything more than distant, orthogonal satellites, wholly dependent upon a generous heliocentrism. While these outposts can certainly be noticed, it may only be by those readers fortunate enough to be in possession of suitable training and proper, specific telescopes. (2019) Adams’s essay traces an alignment between the problems of a canon-oriented study that skews the field around a particular author, and the problems of an editorial tradition that is overwhelmingly white and male, and whose presentation of works forms the substrate upon which the discipline is built. From the long exclusion of women and people of colour from the role of re-presenting the authorial work, to the literal weight and cost of the magisterial ‘Complete Works’ tomes produced by this scholarly tradition, Adams argues that the project of canonization is dogged by  structural inaccessibility, with the stakes nothing less than the foundational primary materials of an author-centred discipline. That Adams’s essay exists at all is indicative of twenty-first-century scholarship that is increasingly and rightly interrogative of its own practices and methodologies, and sceptical of both the constitution of the Western canon and the form in which its texts are reproduced. In relation to early modern drama, the organization of the field around Shakespeare has come under sustained pressure. Recent major editorial projects present other early modern dramatists with the same attention as that accorded to Shakespeare, but perpetuate the issues of cost, size and organization

THE SHAKESPEARE CANON

151

of the field around white male authors. Different recent approaches privilege other models for organizing the drama and poetry of the era, including around genre, company repertory, playhouse, year or subject matter, though with relatively smallscale impact on editorial practice, and almost none on the publication of Shakespeare. The proliferation of copyright-free, inexpensive Shakespeare texts, often based on nineteenth-century editorial practice, means that the circulation of the plays and poems themselves is often divorced from current critical thought. The net result is a Shakespeare canon whose shape and presentation are undergoing more intensive interrogation than at any other point in its history, while still being reproduced in forms that are often substantially unchanged from 400 years ago. In this chapter, I consider the ways in which the organizing principle of ‘Shakespeare’ has worked to order the universe of the somewhat messy body of plays and poems associated with the name, and trace the implications of this for the wider field of early modern literature and drama. The fundamental inclusion/ exclusion paradigm of canon-formation has operated in relation to Shakespeare in historically specific ways in line with changing trends and priorities over its 400year history, as will be shown below. As part of a broader political impulse to the decolonization of the canon – or even decanonization itself – this chapter also asks whether the deliberate and inadvertent exclusions of scholarship can be rectified through new approaches to this body of work.

INCLUSIONS AND EXCLUSIONS: DEALING WITH EXTERNAL ATTRIBUTIONS As was the case for most early modern writers, attribution of Shakespeare’s works was initially inconsistent. The first printings of Titus Andronicus (1594), Richard II (1597), Richard III (1597) and 1 Henry IV (1598) were anonymous, though Shakespeare’s name was appended to the dedications on sig. A2r of Venus and Adonis (1593) and The Rape of Lucrece (1594). The year 1598, as Lukas Erne has argued (2013a: 89–94), marks something of a watershed: Francis Meres lists Shakespeare as the author of twelve plays plus the two narrative poems and ‘his sugred Sonnets’ (1598: sig. Oo1v), and Shakespeare is mentioned on the title pages of quarto editions of Love’s Labour’s Lost, Richard II and Richard III. While the majority of ‘Shakespearean’ publications from then on bore Shakespeare’s name, editions of Romeo and Juliet, Henry V, Titus Andronicus and texts of more disputed provenance continued to be published without Shakespeare’s name well into the seventeenth century (Erne 2013b: 14–16). By 1622, twenty-three Shakespeare plays and collections of poetry had been published with Shakespeare’s name on the title page.1 This body of work looks quite unlike the Shakespeare canon as reproduced in the twenty-first century, including such books as The London Prodigal, A Yorkshire Tragedy, The Whole Contention and 1 Sir John Oldcastle alongside a range of plays and poems drawn mostly from Shakespeare’s late-Elizabethan oeuvre. This pre-Folio canon is further complicated by ascriptions such as that on the title page of Love’s Labour’s Lost: ‘Newly corrected and augmented’ with, on the next line, ‘By W. Shakespere’ falls short of an

152

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

unambiguous ascription of authorship, and bears some resemblance to the ascription on the title page of Locrine (1595), ‘Newly set foorth, ouerseene and corrected’, with, on the next line, ‘By W.S.’, which I have argued elsewhere may be the first tentative title-page acknowledgement of Shakespeare (Kirwan 2015: 127–38). Before the First Folio, then, Shakespearean authorship was a disparate phenomenon. It admitted of Shakespeare as a corrector or augmenter – at the very least, a reviser of his own work, if not of that of others – and as an author of domestic tragedy, martyr history and city comedy set in England. The Jaggard Quartos of 1619 (alternatively referred to as the ‘Pavier Quartos’; see the chapter by Farmer in this handbook), ‘the first time a bound book of professional playbooks was put up for sale’, as Adam G. Hooks points out (2016a: 123), complicated the incipient canon’s coherence further with the collection’s focus on seriality. Tara L. Lyons shows how the name of Shakespeare intersected with Thomas Pavier’s ‘decades-long investment in serials and histories’ to emerge ‘as a discretionary and/or supplementary principle for unifying the serial, historical playbooks in Pavier’s shop’ (2012: 193), leading Lyons to argue that Shakespeare became associated with a principle of seriality. This version of Shakespeare appears to be involved in multi-part, multi-authored entertainments, contributing one of what the quarto of A Yorkshire Tragedy claimed were ‘foure Plaies in one’ (sig. A1r), or writing the first part of an apparent twopart play on Sir John Oldcastle, or designing spin-offs for popular characters in The Merry Wives of Windsor, intrinsically connecting Shakespeare to the wider literarytheatrical market of his own moment. The First Folio, or Mr. William Shakespeares Comedies, Histories, & Tragedies (1623), then, has a great deal to answer for. The Droeshout Portrait and the possessive quality of the title indelibly associate the works within with the solitary figure of the author. The book’s multiple claims to authority, from the prominent reference to the ‘True Originall Copies’ (sig. πA3r), to the commendatory poems from his contemporaries, to the presence of two members of the King’s Men assuring the reader that ‘the collection represented their author’s work “absolute in their numbers”’ (Smith 2015: 10), all make a claim for canonicity. Previous editions, the infamous and ambiguous ‘stolne, and surreptitious copies’ (1623: sig. πA3r), are dismissed in favour of the present volume, which Ben Jonson’s poem implies will survive, like its author, ‘for all time’ (sig. πA4v). The ongoing importance of the Folio, ‘the book that reified Shakespeare’s posthumous reputation’ (Smith 2016: 21), in establishing the base constitution of the Shakespeare canon cannot be overstated. The Folio’s inclusion of eighteen plays never before printed, as well as substantially different texts of several others, means that it offered the single largest and most significant expansion of the canon in Shakespeare’s print history, while also ensuring its longevity as the earliest (often only) authoritative text for several works that already enjoyed a privileged stage history. The need for editors to return repeatedly to the 1623 Folio reinforces its status as the ur-text of Shakespearean canonicity. Yet its exclusions and interventions are legion. Most obviously, it excludes all of Shakespeare’s poetry, leading to a situation where, ‘by the eighteenth century … the narrative poems [had] fallen to the periphery of the Shakespearean canon’ (Roberts 2007: 260). While the poems have remained continuously available in print, the sheer

THE SHAKESPEARE CANON

153

size and scale of the Folio endeavour that established Shakespeare as a playwright began what Patrick Cheney calls the ‘gradual erasure’ of Shakespeare’s practice as a poet and playwright. Considering John Benson’s 1640 edition of Poems, Cheney asks, ‘Measured beside the monumental folio of the plays, how could his slender octavo of the poems hold a plea?’ (2004: 2). More insidiously, the selection and organization of the thirty-six plays in the 1623 Folio admits of several explanations, all of which are partly speculative. The Folio’s preliminaries speak equally of artistic and commercial motivations (‘what euer you do, Buy’, sig. πA3r), and the selection of works is at least partly a matter of property. To take just one example, a complex tussle over rights led to the exclusion of Troilus and Cressida from the Folio’s contents list and early printings, although it was added later with signatures that are out of sequence (Smith 2015: 141–3). Timon of Athens was added as a late substitute, meaning that this otherwise unprinted play ‘would probably not have been known to us at all were it not for the problems posed by Troilus’ (Dawson and Minton 2008: 413). Whether other plays suffered the fate that Timon only narrowly avoided cannot be determined beyond doubt. The mentions of Love’s Labour’s Won in both Francis Meres’s Palladis Tamia and in a 1603 bookseller’s catalogue (Potter 2018: 136) suggests that the play survived into print in some form (although it might be an alternative title for a play included in the Folio). A growing scholarly consensus suggests that Shakespeare collaborated with John Fletcher on a play based on the Cardenio story from Don Quixote (Carnegie and Taylor 2012) that was excluded from the Folio, as were several plays that had been published with Shakespeare’s name or initials on the title page, including Pericles, The London Prodigal and A Yorkshire Tragedy. If Timon of Athens had been excluded, this might have been on the grounds that it was unfinished (Dawson and Minton 2008: 10) or because it was collaboratively written. It is true that no play excluded from the Folio now enjoys a substantiated claim to be wholly by Shakespeare, and those non-Folio plays now most regularly included in scholarly editions (Pericles, The Two Noble Kinsmen, Edward III) are agreed to be collaborative. The suspicion that some plays partly written by others were included in the 1623 Folio emerged as early as 1687 when Edward Ravenscroft claimed that ‘some anciently conversant with the stage’ had told him that Titus Andronicus was ‘brought by a private author’ and that Shakespeare ‘only gave some Master-touches to one or two of the Principal Parts or Characters’ (1687: sig. A2r). There is now a wide consensus that Folio plays such as 1 Henry VI, Henry VIII and Timon of Athens are substantially the work of other authors as well as Shakespeare (Vickers 2002; Taylor and Loughnane 2017). While it is important not to let modern standards of attribution distort the historical picture, there is no clear argument pertaining to collaboration that would explain why the Folio compilers included 1 Henry VI and Henry VIII but excluded Pericles and The Two Noble Kinsmen. As the First Folio served as the base text for the subsequent three Folios (1632, 1663/4, 1685) – and, thereafter, for some of the editions of the eighteenth century – the Folio had the effect of canonizing the Folio’s version of plays that already existed in substantially variant form. While editors such as Edward Capell and Edmond Malone would begin revisiting quarto texts of plays such as King Lear and Hamlet to

154

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

compare variants, leading eventually to the development of more systematic textual genealogies by the New Bibliography of the early twentieth century, others such as Charles Knight laboured under the ‘unwavering belief that [the Folio] recorded the author’s final revisions and desired meaning far better than any mere editor could divine’ (Lesser 2015: 45). What Adam Hooks describes as the ‘fetish’ of the Folio (2016b: 186) persists today among those actors and directors who consider even its punctuation as intentional (Watts 2015). The Folio’s size and scale, coupled with the frontispiece image that Leah Marcus argues was at this point ‘usually memorial in function rather than aggrandizing – linked with a desire to preserve the illusion of human presence’ (1996: 199), created a Shakespeare canon ‘for all time’ framed by markers of cultural and economic investment. It also, for the first time, placed an iteration of supposedly all of Shakespeare’s drama within a single bound volume, aligning the corpus of works with the body of the author and setting physical, material boundaries on the works. In doing so, it initiated the phenomenon of the one-stop Shakespeare edition, which would take several forms over the succeeding centuries, from single volume to multi-volume, though the word ‘complete’ would not become a standard fixture until the twentieth century (Kirwan 2016: 86). But while the Folio would exert a lasting influence on the constitution of the canon, it also immediately inspired attempts to supplement it. The 1632 Second Folio reproduced the same contents, but John Waterson’s publication of The Two Noble Kinsmen in 1634 – the first publication to list Shakespeare alongside a co-author, John Fletcher – and perhaps also Thomas Cotes’s publication of the sixth quarto of Pericles in 1635 suggest stationers attempting to capitalize on a perceived desire for more Shakespeare. Around this time, a volume in the library of Charles I also gathered together a number of anonymous and previously attributed plays in a bound volume marked ‘Shakespeare, Vol. 1’ (Kirwan 2011). Two major mid-century initiatives offered further attempts to expand the canon. In 1640, John Benson’s Poems compiled the majority of the sonnets and several of the other short poems (including some not by Shakespeare from The Passionate Pilgrim). Faith Acker describes it as ‘the first substantial, creative reorganization of a Shakespeare work since the First Folio, and the first not to carry the authorising stamp of Shakespeare’s own colleagues’ (2017: 89), in some sense, then, marking the passage of the custodianship of Shakespeare’s legacy into the public sphere. The inclusion of a reversed image of the Droeshout portrait and of a lengthy prefatory poem by Leonard Digges (who had also contributed to the Folio’s paratexts) aligned it with the canonizing project of the Folio and made a case for the poems being an important part of Shakespeare’s authorial persona. Poems was reprinted throughout the eighteenth century. In 1710, Charles Gildon edited The Works of Mr. William Shakespear. Volume the Seventh, which in its inclusion of Venus and Adonis and The Rape of Lucrece alongside the sonnets and miscellaneous poems, ‘provided the first complete edition of Shakespeare’s poetry’ (Cannan 2017: 171). It was not until Malone’s 1780 Supplement to the edition of George Steevens, however, that the plays and poems would be brought together as part of a move towards a complete works that incorporated plays and poetry.

THE SHAKESPEARE CANON

155

The other major expansion came in the second issue of the Third Folio (1664), published by Philip Chetwind. The 1660s offered something of an explosion in new attributions to Shakespeare, with the poetry miscellany ‘Cupids Cabinet Unlock’t’ (Erne 2017) and the play The Birth of Merlin (both 1662) both printed with Shakespeare’s name on the title page. The second issue of the Third Folio ‘added seven Playes, never before Printed in Folio’ (sig. πA1r), though all previously published in quarto with Shakespeare’s name or initials on the title page: Pericles, The London Prodigal, Thomas Lord Cromwell, 1 Sir John Oldcastle, The Puritan Widow, A Yorkshire Tragedy and Locrine. With the exception of The Troublesome Reign of King John, published with the initials ‘W.Sh.’ in 1611 and Shakespeare’s full name in 1622, which was perhaps excluded because of its similarity to the Folio’s King John, this expansion brought together all of the plays published in Shakespeare’s lifetime with an implicit or explicit attribution to him. The attributions have no greater weight than that, but their addition to the Third Folio led to their inclusion in the Fourth Folio (1685), Nicholas Rowe’s editions (1709, 1714) and a supplemental volume of Alexander Pope’s edition (1728). The seven plays were also reprinted in individual editions as part of the price wars between Jacob Tonson and Robert Walker in 1734–5 (Dugas 2006: 215–25; Milhous and Hume 2015: 239–40). The period of the forty-three-play canon lasted some sixty years, making it one of the longest periods of canonical stability in the reception of Shakespeare. In 1780, Malone became the last individual to edit the seven additions as a group, in the Supplement; then, in 1790, he added Pericles to the thirty-six Folio plays, cementing that play’s place in the canon to the present day. The other six formed the core of the body of plays that came to be known as the ‘Shakespeare Apocrypha’. Malone’s 1790 edition of The plays and poems of William Shakspeare, in ten volumes became the foundational text of Shakespearean canonicity for the next century and a half. Margreta de Grazia points to Malone’s apparatus and concern for authenticity as shifting the paradigm, in a way that redefines the notion of canon: it is only after such practices [of discourse, production and reception] have been codified through reproduction and institutionalization that it is possible to have a Shakespeare canon … a canon, that is, that consists not only of authentic works, like the canonical books of Holy Scripture, but also of regulating and binding tenets, like those of Church dogma, inferred from the very texts over which they preside and legislate. (1991a: 10–11) The scriptural comparison here is apt for a body of works that would increasingly be treated with the reverence usually accorded to holy books. As Shakespeare’s role in English (and, indeed, international) culture developed, and the words of Shakespeare began to take on a role beyond the works themselves in the form of, for example, quotation books (de Grazia 1991b), so too did the canon in its published form begin to stabilize. Malone’s canon of thirty-seven plays (with Titus Andronicus relegated to the final volume as a possible interloper) and the poetry (including Venus and Adonis, The Rape of Lucrece, The Passionate Pilgrim, the Sonnets and A Lover’s Complaint) remained standard throughout the nineteenth and much of the

156

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

twentieth centuries, including in the influential Cambridge Shakespeare (1863–6). Indeed, as the Cambridge edition was the basis for the Globe text, in its own turn the basis for the Moby Shakespeare text that underpins the MIT Shakespeare, ‘the Web’s first edition of the Complete Works of William Shakespeare’ in 1993 (Hylton n.d.), this canon continues to exert a strong hold on the popular understanding of what constitutes Shakespeare’s works.

INCLUSIONS AND EXCLUSIONS: NEW SHAKESPEARE By contrast, the category of Shakespeare’s disputed plays remained in flux, and became the arena for the search for new Shakespeare amid the vast body of anonymous work surviving from the early modern period. In 1760, Edward Capell was the first editor to attempt the attribution of an anonymous play to Shakespeare based on stylistic evidence; his Prolusions sets out the case for Edward III, which ‘arises from resemblance between the stile of his earlier performances and of the work in question’ (ix). Ten years later, Edward Jacob edited Arden of Faversham with a preface claiming the play for Shakespeare. Over the course of the late eighteenth and especially the nineteenth centuries, several critics claimed Shakespeare as the author of otherwise anonymous plays, or continued to advance the cause of those plays excluded from the 1623 Folio but bearing an ascription to Shakespeare. The history of these attributions is too complex to cover in detail here (see Kirwan 2015: 36–71; and Jowett 2007b: 55–68), but the development of this group as a body of work was made manifest in major collections by Charles Knight (1841), Henry Tyrell (1853) and C. F. Tucker Brooke (1908). Brooke’s edition, entitled The Shakespeare Apocrypha, included fourteen plays: six of the seven additions to the 1664 Folio (all but Pericles), three plays attributed to Shakespeare in the Charles I volume (The Merry Devil of Edmonton, Fair Em, Mucedorus), two plays posthumously published as Shakespeare collaborations (The Two Noble Kinsmen, The Birth of Merlin), the two plays newly attributed to Shakespeare in the eighteenth century (Edward III, Arden of Faversham) and the entirety of a manuscript play that had recently been claimed to be partially written in Shakespeare’s own hand (Sir Thomas More). These are the plays which, with no hint of irony, Brooke concedes as ‘having acquired a real claim to the title’ of doubtful plays (1908: xi); they thus occupied a canonical middle ground between definitely-Shakespeare and definitely-not-Shakespeare. Brooke’s edition was the last major collection of disputed Shakespeare plays until Jonathan Bate and Eric Rasmussen’s controversially titled William Shakespeare and Others: Collaborative Plays (2013), to which I will return shortly. The definitely false attributions included, by Brooke’s count, some twenty-nine other plays (1908: ix–xi); by 1977, Lindley Williams Hubbell could list over seventy (see also Kirwan 2015: 215–29). The most important of those not included by Brooke is Double Falsehood (1728), a new play by Lewis Theobald that he claimed was based on a lost play on the Cardenio story, but excluded from Theobald’s own edition of the works. The vast majority of the spurious ascriptions to Shakespeare represented attempts to discover Shakespeare’s juvenilia or assumptions that Shakespeare would be responsible for plays on similar subject matter to that in

THE SHAKESPEARE CANON

157

the canon (e.g. the anonymous True Tragedy of Richard III, 1594), and rested on subjective perceptions of likeness. From the late nineteenth century, however, the development of attribution studies based on more systematic analyses that could be quantified raised the standard of evidence needed for plays to be admitted to the published canon. Hugh Craig discusses the development and current practice of attribution studies elsewhere in this volume, but for the purposes of this chapter it is worth noting that the combativeness and technical specialization of this area of study (as well as the potential for embarrassment and commercial devaluation if a claim is subsequently proven false) has led to wariness about admitting new plays to editions of Shakespeare’s works. The only play from this grouping to have made the transition to the Shakespeare canon with more or less universal acceptance is The Two Noble Kinsmen. The play sporadically appeared in nineteenth-century editions, including Alexander Dyce’s 1866 edition of Shakespeare’s works (as well as in his earlier edition of Beaumont and Fletcher’s works, 1846) (Potter 2015: 83). Since the 1974 Riverside Shakespeare, it has been rare for a single-volume Complete Works or a multi-volume series to exclude the play, though it remains absent from the MIT Shakespeare and from the single-volume editions in the RSC Shakespeare series. Jonathan Bate and Eric Rasmussen included Pericles and The Two Noble Kinsmen in a double-column layout in the 2007 RSC Complete Works, to distinguish them typographically from the plays that were published in the 1623 Folio, but only Pericles was accorded a standalone edition. Two other plays usually appear in collections of Shakespeare’s work in the twenty-first century, though not consistently. The ‘Hand D’ extract of Sir Thomas More, now often accepted to be written in Shakespeare’s hand (but see above, pp.  54–7), has been regularly published in isolation in editions of the complete works, including the Riverside Shakespeare (1974 and 1997), The Oxford Shakespeare (1986), the RSC Shakespeare (1997) and The New Oxford Shakespeare (2016). The play was included in full in the second edition of the Oxford Shakespeare (2005) and in the Arden 3 series (2011), though to date in no other Shakespeare series. Edward III, whose Countess of Salisbury scenes (at least) have come to be widely accepted as Shakespeare’s since the twentieth century, was included in full in the second Riverside (1997), second Oxford (2005) and New Oxford (2016) editions, and the New Cambridge (1998) and the Arden 3 (2017) series. More recently, and controversially, the New Oxford is the first Complete Works of Shakespeare to include Arden of Faversham, Cardenio and the 1602 additions to The Spanish Tragedy. The case for Shakespeare’s partial authorship of Arden has been active since Jacob’s 1770 edition of the play, and has recently been advanced by Arthur F. Kinney (2009) and MacDonald P. Jackson (2014). The title of Jackson’s study, Determining the Shakespeare Canon, gives some indication of how the field of attribution studies has colonized the question of canonicity. To date, however, there have been no plans to include Arden in a Shakespeare series, and the only other edition to act on the attribution claims is the RSC Shakespeare, which included a single scene on the edition’s website. Cardenio in the New Oxford Shakespeare takes the form of fragments adapted from Theobald’s Double Falsehood in the Modern

158

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Critical Edition, with a diplomatic original-spelling text in the accompanying Critical Reference Edition. Brean Hammond’s edition of Double Falsehood was included in the Arden 3 series (2010) and the one-volume Arden Complete Works (2011). Finally, while the case for the Spanish Tragedy additions has been advanced by several scholars, including Warren Stevenson (2008) and Hugh Craig (2009), the New Oxford is the only edition of Shakespeare to date to include them. Perhaps curiously, while the bar for new plays to be included in or excluded from the canon has been set very high, the poems have fluctuated a little more. The Passionate Pilgrim, despite universal agreement that many of its poems are by other authors, is included in full in the Riverside (1974) and the New Oxford (2016), as well as in most series such as the Arden 3, the Oxford Shakespeare and New Cambridge Shakespeare, though only those poems believed to be by Shakespeare are printed in the Oxford (1986 and 2005), and the miscellany is excluded from the single-volume RSC Shakespeare (2007). There has also been something of a trend in recent Complete Works to offer a new attribution of a poem: the Oxford (1986) added ‘Shall I die?’; the second Riverside (1997) included ‘A Funeral Elegy’; and the RSC Shakespeare included a short poem it entitled ‘To the Queen’. The New Oxford (2016) includes ‘Shall I die?’ and ‘To the Queen’, as well as other miscellaneous poems. Finally, A Lover’s Complaint has been the subject of extensive attribution scrutiny, and while the majority of editions include it as a kind of epilogue to the Sonnets, the RSC Shakespeare (2007) takes the unusual decision to exclude it. This brief survey gives a sense of the areas of flux currently existing within the Shakespeare canon; while the bulk of the plays and poems are shared by all singlevolume Works and multi-volume series, each edition has its own subtle distinctions of inclusion and exclusion. The admission to a place within the bound pages of a collected works volume remains the gold standard of Shakespearean canonicity, with most editions now including a detailed justification for exclusions. However, a recent twist has seen the return of the separate canonical space previously marked by the apocrypha. Bate and Rasmussen’s Collaborative Plays (2013) includes ten plays absent from the RSC Complete Works volume, from near-canonical texts such as Sir Thomas More and Edward III, to plays more recently considered for canonicity such as Arden of Faversham and Double Falsehood, to plays with early attributions but no serious modern claim to Shakespearean authorship (Thomas Lord Cromwell, The London Prodigal). The compilation of all of these plays under the title Collaborative Plays is misleading, but the supplementary nature of the volume allows it to be the first to include the whole of Arden of Faversham and The Spanish Tragedy under a Shakespearean imprint.

NON-CANONICAL SHAKESPEARE The inclusion-exclusion paradigm modelled in the previous section is too often an unquestioned standard. Brian Vickers begins his Shakespeare, Co-Author with the assertion that ‘No issue in Shakespeare studies is more important than determining what he wrote. We cannot form any reliable impression of his work as a dramatist unless we can distinguish his authentic plays from those spuriously ascribed

THE SHAKESPEARE CANON

159

to him … and unless we can identify those parts of collaborative plays that were written by him’ (2002: 3). The underpinning principle is one of separation and exclusion, the authentic from the spurious, the authorized from the unauthorized. The canon – epitomized by the weighty tome of the single-volume Complete Works – acts as its own guarantor of value, with a legacy that can be traced back to the authorizing apparatus of the 1623 Folio. Such value cannot go unquestioned in an era in which the politics of canon formation as a form of cultural gatekeeping are under scrutiny. Although writing a quarter of a century ago, John Guillory’s important point about the significance of the noncanonical is relevant here: [T]he most surprising aspect of the current legitimation crisis is the fact that the ‘noncanonical’ is not what fails to appear in the classroom, but what, in the context of liberal pedagogy, signifies exclusion. The noncanonical is a newly constituted category of text production and reception, permitting certain authors and texts to be taught as noncanonical, to have the status of noncanonical works in the classroom. This effect is quite different from the effect of total absence, of nonrepresentation tout court. (1993: 9) The problem of Shakespeare’s heliocentrism, to return to Brandi Adams’s phrase, is exacerbated by the inclusion/exclusion paradigm, in which works gain attention and prestige as they orbit closer to Shakespeare, or even make it into the pages of the Complete Works. A small number of plays and poems absorb a disproportionate amount of critical attention driven by the goal of determining canonical position. Adams, noting the ‘fairly contained’ community that determines the way editions look – and by extension the constitution of the canon – argues that the latest generation of scholarly editions ‘perpetuate a hegemonic approach to Shakespeare and his plays’ that presents a ‘very particular historiographical and theoretical approach to the early modern world’ that fortifies exclusivity (2019). The very project of canonization is predicated on exclusion, and the effects of that exclusion – not least the ongoing bias of commercial investment and scholarly attention in favour of canonical Shakespeare at the expense of all else – cannot be ignored. In this concluding section, I consider some of the potential alternative forms that the Shakespeare canon has taken or could take, to think about some of the ways the canon might begin to represent a more diverse and heterogeneous view of Shakespeare and the context(s) of the works, focusing in turn on how the canon is organized, and how the canon approaches variant forms of the works.

Organization: genre and chronology The Folio’s division of the plays into three groupings based on genre – comedies, histories and tragedies – is one of the most lasting of its interventions. The order of the plays from The Tempest to Cymbeline (if not always the explicit division of genres) was retained by the vast majority of editions throughout the eighteenth and nineteenth centuries, and is still sometimes used (e.g. the RSC Shakespeare, 2007). As more works were added, they were appended at the end, retaining the impression of Pericles, the poems, and any supplemental volumes of disputed plays

160

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

as addenda. The fact that many plays sat uneasily within the genre categories to which they were assigned was not acted on in publishing terms for centuries. The Riverside Shakespeare (1974) added ‘romances’ and ‘poems’ as separate genres, but reorganized the works into chronological order within those categories, and in doing so rendered the generic divisions even more absolute, each genre a miniature canon in its own right. The problems of generic thinking are myriad. Jean Howard notes that the profession has been shaped by them: ‘A number of books continue to be written that focus only on Shakespearean tragedy, on the Shakespearean history play or on Shakespearean comedy. For many critics and students these categories have assumed the status of natural, not arbitrary, entities’ (1999: 297). The shape of the canon, that is, imposes pre-defined ways of thinking about the plays that place implicit restrictions on what might be admitted to the canon and how Shakespeare’s work relates to that of others; Richard Proudfoot’s comment that the city comedy The London Prodigal ‘isn’t what we think of as Shakespeare’s kind of play’ (1998: 152) evokes an assumed shared prior understanding of what we expect Shakespeare should look like. These predetermined grooves are already being productively scratched. Emma Whipday’s Shakespeare’s Domestic Tragedies argues that ‘Shakespeare’s engagement with the genre is significant not only for our understanding of his tragedies, but for our  conceptions of both the genre and the entire canon of early modern drama’ (2019: 17), suggesting that the category fundamentally elides distinctions and barriers of class and status. David Sterling Brown’s call for attention to Shakespeare’s ‘other race plays’ (2019), meanwhile, challenges the idea that Shakespeare does not write to particular subjects by applying a critical race framework across assumed canonical categories. New work of this kind involves reading past the apparatus of canon formation that divides up Shakespeare’s works. The other dominant paradigm of canon organization is chronology. Despite the interest in the order of Shakespeare’s plays inaugurated by Malone, the reorganization of the canon into chronological order is a relatively recent phenomenon, apparent in the Oxford (1986, 2005), the Norton (1997, 2008, 2016) and the New Oxford (2016). This arrangement ‘allows readers to trace the development of [Shakespeare’s] writing over time: it offers a kind of biography of his creativity’ (Taylor and Bourus 2016: 48). While this has the advantage of tracing a form of sequence of the author’s individual labour, it also views that history of theatre and literature through the prism of one figure, and risks the kind of qualitative teleology that assumes progression of quality. The New Oxford Critical Reference Edition offers an alternative chronology by first publication or inscription, and the micro-histories of James Shapiro (1599, 2005; and 1606, 2015) and David Bergeron (Shakespeare’s London 1613, 2017) suggest an alternative to the teleological authorial canon that might look at the works being produced simultaneously at a given point in time. This last also implies the intermingling of Shakespeare and other writers in a more thorough way than has hitherto been realized. While collected editions of Shakespeare, both single volume and multi-volume, now regularly accept that several of the works contain the work of multiple authors, the emphasis is on finding the other writers within Shakespeare; ‘Shakespeare’ remains the organizing

THE SHAKESPEARE CANON

161

principle of the canon. Shakespeare is rarely put in the canons of others; indeed, in the long history of anthologies of Renaissance drama, where choices of inclusion are subjective, Shakespeare is the only author to be consistently absent. Jeremy Lopez offers a prolegomenon for a new edition of thirty-seven plays that would include Macbeth, ‘because of its surprising, and perhaps entirely coincidental, resonances with a non-Shakespearean play (Northward Ho), which allows me to dramatize both the inevitability and the rewards of arriving at a formal and historical identity for “non-Shakespeare” by reading through Shakespeare’ (2014: 202). The idea here is that juxtaposing plays in surprising ways can reveal new connections, an argument I have made in relation to Robert Walker’s 1735 bound collection of the forty-three play canon in an apparently random order that allows the reader to make fresh associations between the plays (Kirwan 2015: 26–7). Elsewhere, other plays are escaping the Shakespeare canon: the Oxford Middleton (2007) included Macbeth and Measure for Measure; Emma Smith’s Five Revenge Tragedies (2012) puts Q1 Hamlet alongside plays by Thomas Kyd, Henry Chettle, John Marston and Thomas Middleton; and Greg Walker’s chronological anthology of Tudor Drama (2014) puts The Comedy of Errors and Titus Andronicus alongside plays by a range of late medieval and early modern authors, all of which offer fresh recanonization and contextualization of the plays.

Variant Shakespeares Sasha Roberts notes that by the eighteenth century, ‘the Sonnets had become something of an aesthetic and sexual embarrassment. How could the nation’s Bard write such passionate love poems to another man? What went wrong?’ (2007: 260). John Benson’s regendering of the addressee of several of the sonnets in Poems (1640) is only one early instance of an adaptation of Shakespeare’s works made to speak to the taste of the time. The persistence of Benson’s adaptation meant that several poems featuring a male addressee became ‘noncanonical’ for a time, in favour of versions that lent themselves more obviously to a heterosexual interpretation. Alexander Pope, infamously, relegated sections of the plays that he felt were spurious or displeasing to footnotes, but over time, ‘authenticity’ rather than taste became the dominant paradigm of determining which versions of Shakespeare were to be rendered ‘canonical’ and which excluded. At the most obvious level, the existence of many of Shakespeare’s works in multiple versions means that any edition offering just one version of the play had to make exclusions. Stephen Orgel argues that editions ‘aspire to completeness through their inclusion of textual variants’ (2007: 308) – most fully realized in the Variorum tradition – but in the case of more substantially variant texts, the principles of hierarchization developed by the New Bibliography (Maguire 1996: 21–71; see also John Jowett’s chapter in this volume) that cast certain quartos as ‘good’ and others as ‘bad’ meant that certain texts continued to be routinely excluded from the canon, or, at the very least, dismembered, their interesting variants included piecemeal in the textual apparatus of new editions. Plays at a further remove, such as The Taming of a Shrew, were almost entirely kept apart from the canon.

162

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

What Barbara Mowat calls ‘revisionist theorizing’ (2016: 62), with a long history stretching back to Capell’s reappraisal of the early quartos, offered one of the most disruptive challenges to the canon from the 1980s onwards. The inclusion of two versions of King Lear based on the quarto and folio in The Oxford Shakespeare (1986) made a bold statement that both versions were canonical. The inclusion of multiple texts of Hamlet in the Arden Shakespeare single-play edition (Thompson and Taylor 2006a; 2006b), and the ambitious inclusion of all substantially variant editions of the plays in the third edition of the Norton Shakespeare (2016) – resulting in the latter case in some fifty-nine complete play texts being included in the online version – offer material realization of an inclusive canon that attempts to sidestep value judgements about relative quality or precedence and instead allow the independent early versions of texts to stand on their own terms. This is the authorial canon in its most inclusive form, though even here, the publication of the 1603 and 1623 texts of Hamlet in a separate Arden volume and the exclusion of variant texts from the print version of the Norton continue to maintain a degree of hierarchization. The forthcoming Alternative Versions edition in The New Oxford Shakespeare promises to go further by encompassing A Shrew and other early versions of Shakespeare. While this paradigm is unlikely to remain dominant, the future of digital editions and resources (see Sonia Massai’s chapter in this volume) should help ensure that future exclusion of variant texts from any given canon will not result in the denial of access to those texts. Finally, there is the question of Shakespearean adaptations. Emma Depledge’s important work has demonstrated the significance of what she calls ‘alterations’ during the period of Shakespeare’s canonization. Her integration of book history and theatre history shows how the waning public appetite for ‘unaltered’ Shakespeare after around 1660 was answered by the success of adaptations by Thomas Shadwell, John Crowne, Nahum Tate, John Dryden, William Davenant and others in response to the Exclusion Crisis (1678–82). These adaptations ‘set a precedent for acknowledging [Shakespeare’s] role as an author source’ (2018: 170), intertwining Shakespeare’s authorial agency with the rewriting of Shakespeare for the stage. Indeed, for over a century, The Tempest and King Lear were likely to be encountered in their adapted Restoration versions, and when Robert Walker published his edition of King Lear in 1735, it was Tate’s text he used rather than Shakespeare’s (Dugas 2006: 216). Tate’s return to the earlier sources for his ‘happy ending’ is perhaps not so different from the 1986 Oxford Shakespeare’s inclusion of a text of Pericles that was ‘reconstructed’ through consultation of George Wilkins’ prose account. And the inclusion of Double Falsehood in the Arden 3 series (2010) admits an avowed eighteenth-century adaptation into this particular iteration of the Shakespeare canon. While there is no justification for replacing versions of Shakespeare based on the earliest printed texts with new or adapted versions from later centuries, the importance of adaptation to the development of Shakespeare’s reputation and his later cultural centrality might give pause for thought. Many of today’s booksellers offer copies of competing editions of Shakespeare, each containing substantially similar versions of the canonical plays and poems, but the adaptations of Shakespeare

THE SHAKESPEARE CANON

163

that were central to his canonization are (with the exception of Double Falsehood) largely inaccessible to the lay reader. Adaptation in all its forms offers a version of Shakespeare that looks very different. The anthology Adaptations of Shakespeare (Fischlin and Fortier 2000) places texts by Welcome Msomi, The Women’s Theatre Group, Djanet Sears and others alongside adaptations by Nahum Tate and Charles Marowitz; and Lukas Erne and Kareen Seidler’s edition of Der Bestrafte Brudermord and Romio und Julieta (2020) – published under the Arden Shakespeare imprint but not as part of the series – enables readers to access a version of Shakespeare experienced in continental Europe in the seventeenth century. Peter Holland’s work on ‘Theatre Editions’, meanwhile, traces a parallel history of published adaptation which ‘can confer a surprising longevity’ (2015: 247) on particular iterations. Indeed, the Arden Shakespeare currently offers customers looking for Coriolanus a choice between Peter Holland’s Arden 3 edition or a slimline volume based on the 2014 Donmar Warehouse playing text with a topless Tom Hiddleston on the front cover. The proliferation of theatre-based adaptations, commercial or gift texts, reworkings and responses, offers a Shakespeare print culture with its own claims to legitimacy and authenticity, rooted in diverse engagements between Shakespeare and editors and artists that occur outside of the ‘contained’ mainstream Shakespeare editing community. The isolation of Shakespeare’s works from those of others, or from the commercial print and theatrical cultures within which his works circulated, is unrepresentative of the fluidity of early modern theatre practice, and any one iteration of a canon, bound and fixed in time, makes static a protean, shifting collection of interconnected texts that have always been adapted and edited to suit the time. The heliocentrism of the traditional author-shaped model, in the era of digital texts and multiple points of access, is giving way to something that might better be understood as a multiverse, with Shakespeare a fixed point intersecting with a potentially infinite number of combinations of other organizing paradigms. In such a multiverse, the historically contingent and constantly changing shape of the Shakespeare canon may be a useful entry point to any number of otherwise neglected texts, adaptations and textual performances. While the drive to gather and celebrate Shakespeare’s authorial canon is unlikely to diminish, perhaps the need to exclude can.

NOTE 1.

With dates of first unambiguous attribution to Shakespeare: Hamlet (1603), 1 Henry IV (1599), 2 Henry IV (1600), 2 Henry VI and 3 Henry VI (as The Whole Contention, 1619), King Lear (1608), The London Prodigal (1605), Love’s Labour’s Lost (1598), The Merchant of Venice (1600), The Merry Wives of Windsor (1602), A Midsummer Night’s Dream (1600), Much Ado About Nothing (1600), Othello (1622), The Passionate Pilgrim (1599), Pericles (1609), The Rape of Lucrece (1616), Richard II (1598), Richard III (1598), Romeo and Juliet (undated, c. 1621–2 (Jowett 2007a: 184) or 1623 (Erne 2013b: 15)), Shakespeare’s Sonnets (1609), 1 Sir John Oldcastle (1619), Troilus and Cressida (1609), The Troublesome Reign of King John (1622), A Yorkshire Tragedy (1608) (Jowett 2007a: 174–86).

164

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

REFERENCES Acker, Faith (2017), ‘John Benson’s 1640 Poems and Its Literary Precedents’, in Emma Depledge and Peter Kirwan (eds), Canonising Shakespeare: Stationers and the Book Trade, 1640–1740, 89–106, Cambridge: Cambridge University Press. Adams, Brandi Kristine (2019), ‘Mediators of the Wor(l)d: Editors, Shakespeare, and Inclusion’, Before Shakespeare, 3 June. Available online: https://beforeshakespeare. com/2019/06/03/mediators-of-the-world-editors-shakespeare-and-inclusion/ (accessed 13 November 2020). Bate, Jonathan and Eric Rasmussen, eds (2013), William Shakespeare and Others: Collaborative Plays, Basingstoke: Palgrave Macmillan. Bergeron, David M. (2017), Shakespeare’s London 1613, Manchester: Manchester University Press. Brooke, C. F. Tucker, ed. (1908), The Shakespeare Apocrypha, Oxford: Clarendon Press. Brown, David Sterling (2019), ‘White Hands: Shakespeare’s “Other Race Plays”’, Shakespeare Association of America 47th annual meeting, Washington DC, 19 April. Cannan, Paul D. (2017), ‘The 1709/11 Editions of Shakespeare’s Poems’, in Emma Depledge and Peter Kirwan (eds), Canonising Shakespeare: Stationers and the Book Trade, 1640–1740, 171–86, Cambridge: Cambridge University Press. Capell, Edward, ed. (1760), Prolusions, London. Carnegie, David and Gary Taylor, eds (2012), The Quest for Cardenio, Oxford: Oxford University Press. Cheney, Patrick (2004), Shakespeare, National Poet-Playwright, Cambridge: Cambridge University Press. Craig, Hugh (2009), ‘The 1602 Additions to The Spanish Tragedy’, in Hugh Craig and Arthur F. Kinney (eds), Shakespeare, Computers, and the Mystery of Authorship, 162–80, Cambridge: Cambridge University Press. Dawson, Anthony B. and Gretchen E. Minton, eds (2008), Timon of Athens, London: Cengage. de Grazia, Margreta (1991a), Shakespeare Verbatim: The Reproduction of Authenticity and the 1790 Apparatus, Oxford: Clarendon Press. de Grazia, Margreta (1991b), ‘Shakespeare in Quotation Marks’, in Jean I. Marsden (ed.), The Appropriation of Shakespeare: Post-Renaissance Reconstructions of the Works and the Myth, 57–71, London: Wheatsheaf. Depledge, Emma (2018), Shakespeare’s Rise to Cultural Prominence: Politics, Print and Alteration, 1642–1700, Cambridge: Cambridge University Press. Dugas, Don-John (2006), Marketing the Bard: Shakespeare in Performance and Print, 1660–1740, Columbia: University of Missouri Press. Erne, Lukas (2013a), Shakespeare as Literary Dramatist, 2nd edn, Cambridge: Cambridge University Press. Erne, Lukas (2013b), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Erne, Lukas (2017), ‘Cupids Cabinet Unlock’t (1662), Ostensibly “By W. Shakespeare”, in Fact Partly by John Milton’, in Emma Depledge and Peter Kirwan (eds), Canonising Shakespeare, 107–29, Cambridge: Cambridge University Press.

THE SHAKESPEARE CANON

165

Erne, Lukas and Kareen Seidler, eds (2020), ‘Hamlet’ and ‘Romeo and Juliet’: ‘Der Bestrafte Brudermord’ and ‘Romio und Julieta’ in Translation: Early Modern German Shakespeare, London: Bloomsbury. Fischlin, Daniel and Mark Fortier, eds (2000), Adaptations of Shakespeare, London: Routledge. Guillory, John (1993), Cultural Capital: The Problem of Literary Canon Formation, Chicago: University of Chicago Press. Holland, Peter (2015), ‘Theatre Editions’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 233–48, Cambridge: Cambridge University Press. Hooks, Adam G. (2016a), Selling Shakespeare, Cambridge: Cambridge University Press. Hooks, Adam G. (2016b), ‘Afterword: The Folio as Fetish’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 185–96, Cambridge: Cambridge University Press. Howard, Jean E. (1999), ‘Shakespeare and Genre’, in David Scott Kastan (ed.), A Companion to Shakespeare, 297–310, Malden, MA: Blackwell. Hubbell, Lindley Williams (1977), A Note on the Shakespeare Apocrypha, Kobe: Ikuta. Hylton, Jeremy (n.d.), ‘The Complete Works of William Shakespeare’, MIT Shakespeare. Available online: http://shakespeare.mit.edu/ (accessed 13 November 2020). Jackson, MacDonald P. (2014), Determining the Shakespeare Canon: ‘Arden of Faversham’ and ‘A Lover’s Complaint’, Oxford: Oxford University Press. Jacob, Edward, ed. (1770), The Lamentable and True Tragedie of Mr. Arden of Feversham, in Kent, London. Jowett, John (2007a), Shakespeare and Text, Oxford: Oxford University Press. Jowett, John (2007b), ‘Shakespeare Supplemented’, in Douglas A. Brooks (ed.), The Shakespeare Apocrypha, 39–73, Lampeter: Edwin Mellen Press. Kinney, Arthur F. (2009), ‘Authoring Arden of Faversham’, in Hugh Craig and Arthur F. Kinney (eds), Shakespeare, Computers, and the Mystery of Authorship, 78–99, Cambridge: Cambridge University Press. Kirwan, Peter (2011), ‘The First Collected “Shakespeare Apocrypha”’, Shakespeare Quarterly, 62 (4): 266–75. Kirwan, Peter (2015), Shakespeare and the Idea of Apocrypha, Cambridge: Cambridge University Press. Kirwan, Peter (2016), ‘“Complete” Works: The Folio and All of Shakespeare’, in Emma Smith (ed.), The Cambridge Companion to Shakespeare’s First Folio, 86–102, Cambridge: Cambridge University Press. Knight, Charles, ed. (1841), The Pictorial Edition of the Works of Shakspere: Doubtful Plays, London. Lesser, Zachary (2015), Hamlet After Q1: An Uncanny History of the Shakespearean Text, Philadelphia: University of Pennsylvania Press. Lopez, Jeremy (2014), Constructing the Canon of Early Modern Drama, Cambridge: Cambridge University Press. Lyons, Tara L. (2012), ‘Serials, Spinoffs, and Histories: Selling “Shakespeare” in Collection Before the Folio’, Philological Quarterly, 91 (2): 185–220.

166

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Maguire, Laurie E. (1996), Shakespearean Suspect Texts: The ‘Bad’ Quartos and Their Contexts, Cambridge: Cambridge University Press. Malone, Edmund, ed. (1780), Supplement to the Edition of Shakespear’s Plays, 2 vols, London. Marcus, Leah S. (1996), Unediting the Renaissance: Shakespeare, Marlowe, Milton, London: Routledge. Meres, Francis (1598), Palladis Tamia, Wits Treasury, London. Milhous, Judith and Robert D. Hume (2015), The Publication of Plays in London 1660–1800, London: British Library. Mowat, Barbara A. (2016), ‘Facts, Theories, and Beliefs’, in Dympna Callaghan and Suzanne Gossett (eds), Shakespeare in Our Time: A Shakespeare Association of America Collection, 57–64, London: Bloomsbury. Orgel, Stephen (2007), ‘The Desire and Pursuit of the Whole’, Shakespeare Quarterly, 58 (3): 290–310. Potter, Lois, ed. (2015), The Two Noble Kinsmen, rev. edn, The Arden Shakespeare, London: Bloomsbury. Potter, Lois (2018), ‘Much Ado or Love’s Labour’s Won? – Does it Matter Which?’, in Deborah Cartmell and Peter J. Smith (eds), Much Ado About Nothing: A Critical Reader, 133–53, London: Bloomsbury. Proudfoot, Richard (1998), ‘Shakespeare’s Most Neglected Play’, in Laurie E. Maguire and Thomas L. Berger (eds), Textual Formations and Reformations, 149–57, Newark, DE: University of Delaware Press. Ravenscroft, Edward (1687), Titus Andronicus, or, The rape of Lavinia, London. Roberts, Sasha (2007), ‘Reception and Influence’, in Patrick Cheney (ed.), The Cambridge Companion to Shakespeare’s Poetry, 260–80, Cambridge: Cambridge University Press. Shakespeare, William (1623), Mr. William Shakespeares Comedies, Histories, & Tragedies, London. Shapiro, James (2005), 1599, London: Faber and Faber. Shapiro, James (2015), 1606, London: Faber and Faber. Smith, Emma, ed. (2012), Five Revenge Tragedies, London: Penguin. Smith, Emma (2015), The Making of Shakespeare’s First Folio, Oxford: Bodleian. Smith, Emma (2016), Shakespeare’s First Folio, Oxford: Oxford University Press. Stevenson, Warren (2008), Shakespeare’s Additions to Thomas Kyd’s ‘The Spanish Tragedy’, Lampeter: Edwin Mellen. Taylor, Gary and Terri Bourus (2016), ‘Why Read This Complete Works?’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: Modern Critical Edition, 45–58, Oxford: Oxford University Press. Taylor, Gary and Rory Loughnane (2017), ‘The Canon and Chronology of Shakespeare’s Works’, in Gary Taylor and Gabriel Egan (eds), The New Oxford Shakespeare Authorship Companion, 417–602, Oxford: Oxford University Press. Thompson, Ann and Neil Taylor, eds (2006a), Hamlet, The Arden Shakespeare, London: Thomson Learning. Thompson, Ann and Neil Taylor, eds (2006b), ‘Hamlet’: The Texts of 1603 and 1623, The Arden Shakespeare, London: Thomson Learning.

THE SHAKESPEARE CANON

167

Tyrell, Henry, ed. (1853), The Doubtful Plays of William Shakspere, London. Vickers, Brian (2002), Shakespeare, Co-Author, Oxford: Oxford University Press. Walker, Greg, ed. (2014), The Oxford Anthology of Tudor Drama, Oxford: Oxford University Press. Watts, Graham James (2015), Shakespeare’s Authentic Performance Texts, Jefferson, NC: McFarland & Company. Whipday, Emma (2019), Shakespeare’s Domestic Tragedies: Violence in the Early Modern Home, Cambridge: Cambridge University Press.

CHAPTER 2.7

Shakespeare’s editors from the eighteenth to the twenty-first century ANDREW MURPHY

In 1700, someone curious to read the works of Shakespeare could have gone to a book shop and picked up a copy of the Fourth Folio. The line of folios, stretching from the first edition of 1623, had been silently tweaked and corrected (and sometimes miscorrected) iteration by iteration, but the core text can be said to have remained substantially the same from edition to edition. So someone buying the Fourth Folio in 1700 was acquiring a book that had much the same look and feel – and essentially the same text – as the First Folio, issued just over three quarters of a century earlier (though it did contain the seven extra ‘apocryphal’ texts that had been added in the second issue of F3, in 1664). Our imagined reader of 1700 would likely have been struck by the fact that the layout and format of the volume was quite outdated by contemporary standards, the pages dense with cramped text, squeezed into two narrow columns. But, beyond the appearance of the volume, such a reader would also likely have been puzzled by a number of aspects of the text itself. Why, for instance, does Act 1, Scene 1 of Antony and Cleopatra last the entire length of the play? Where, exactly, do we find ourselves in the opening scene of A Midsummer Night’s Dream, which begins simply with the stage direction ‘Enter Theseus, Hippolita, with others’ (F4, Comedies, 130)? And just who are these others – and how, indeed, do the various characters in the play more generally relate to each other? In Romeo and Juliet, who are ‘Mo.’, ‘La.’ and ‘Lad.’? Could they possibly be the same person as the character variously called ‘Wife’ and ‘Old La.’? In The Tempest, when Caliban tells Trinculo ‘I’le get thee young Scamels from the Rock’ (F4, Comedies, 9), what exactly is he offering him? In Henry V, when the Hostess says of Falstaff that ‘his Nose was as sharp as a Pen, and a Table of green Fields’ (F4, Histories, 94), what does she mean – what exactly is ‘a Table of green Fields’? In answer to all these questions, the F4 text, in effect, responds in a similar manner to Othello’s Iago: ‘Demand me nothing. What you know, you know’ (5.2.300).

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

169

In essence, we can say that the editorial tradition initiated at the beginning of the eighteenth century represents an ongoing effort to answer these, and other important, questions. The fundamental impulse motivating the editorial enterprise has been, from the start, to make the text more intelligible and more user-friendly for readers. The person who can be said to bear primary responsibility for beginning this development is Jacob Tonson the Elder, whose family firm served as the dominant literary publishers in London for the greater part of the eighteenth century. After he acquired the majority stake in the rights to Shakespeare in 1709, Tonson registered the need to revivify this old textual property.1 He did this in a number of different ways. At the turn of the century, Tonson had collaborated with Cambridge University Press in producing a series of editions of classical authors. As Robert B. Hamm, Jr has noted, this work ‘introduced him to a precise and sophisticated typographical style that he would incorporate into his later projects’ (2004: 181). In collaboration with the printer John Watts, Tonson developed a house style which gave his publications a distinctive, modern feel – quite different, we might say, from the appearance of the Shakespeare folios. Tonson issued his first edition of Shakespeare in 1709. If our imagined purchaser of the Fourth Folio in 1700 had decided to add this new edition to her library, she would have noticed a number of significant changes in the new text. For one thing, it was smaller in size, Tonson having switched from the single volume folio format to six octavo volumes. So our reader could now more easily have carried a volume or two around with her – a significant change from the earlier unwieldy folio editions. Gone were the double columns, so the pages looked a little more spacious and less cluttered. Our reader would also have noticed that, by contrast with the folios, with their single portrait illustration of Shakespeare, the new edition was ‘Adorn’d with cuts’, meaning that it included forty-five engraved illustrations, helping readers to imagine what particular scenes from the plays might look like. It also included ‘an account of the life and writings of the author’ – in effect, the first substantial attempt to create a biography of Shakespeare. This biographical sketch was largely based on little more than anecdotes gathered in Stratford-on-Avon by the actor and theatre manager Thomas Betterton, but it did at least provide readers with some context for Shakespeare’s career and his work. The greatest innovation offered by Tonson in his first edition of Shakespeare was, however, undoubtedly his breaking with the tradition of the previous century whereby preparation of the text for publication was left in the hands of unknown agents who carried out modest interventions, aimed largely at correcting obvious errors (or perceived errors) in the text of the immediately previous edition. By contrast, Tonson’s text declared on its title page that it had been ‘Revis’d and corrected … by N. Rowe, Esq.’ Nicholas Rowe was a successful contemporary dramatist and, in taking on Tonson’s commission to prepare an edition of the playwright, he became the first editor of Shakespeare to be acknowledged on the title page of an edition. Rowe saw his role as editor as involving, primarily, making the text more navigable and more readable. He systematically broke all of the plays down into acts and scenes, following the conventional practice of his own day, so

170

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

that, for instance, the Folios’ eccentric single-act-single-scene Antony and Cleopatra became a regular five-act play with a total of thirty-one scenes. Rowe also added a dramatis personae list at the beginning of each play, naming all of the characters, and mapping the relations among them. Thus, in the case of A Midsummer Night’s Dream, for instance, we learn that Theseus is ‘Duke of Athens’ and Hippolita is ‘Princess of the Amazons, betrothed to Theseus’ (vol. 2, sig. B1v), so we understand the nature of their relationship and of their context before we even begin reading the play. Rowe also added indications of location to the plays. Again, in the case of A Midsummer Night’s Dream we are told: ‘SCENE Athens, and a Wood not far from it’ (vol. 2, sig. B1v). Speech prefixes are regularized and standardized throughout, so that a single prefix ‘La. Cap.’ indicates Lady Capulet in Romeo and Juliet, rather than the five confusingly different prefixes that we find in the folios. Rowe also modernized and standardized spelling and punctuation throughout, giving the text a more contemporary feel. A great deal of Rowe’s work proved to be enduring, and Judith Milhous and Robert D. Hume have observed that ‘much of the scene division, the character lists, and quite a lot of the settings and stage directions that we take for granted in reading and teaching Shakespeare were in fact created by Nicholas Rowe’ (2015: 235). Rowe thus made life easier for eighteenth-century readers of Shakespeare in a variety of different ways. In other respects, however, he left these readers to their own devices. Caliban still promises to fetch the mysterious ‘Young Scamels from the Rock’ (1709: 1.33; cf. The Tempest, 2.2.169), and Falstaff’s nose is still ‘as sharp as a Pen, and a Table of Green Fields’ (1709: 3.1315; cf. Henry V, 2.3.16–17). Even if these remaining unexplained elements in the text still left some readers puzzled, Rowe’s Shakespeare was, nevertheless, a commercial success, and the Tonson firm issued a second edition within a matter of months, with a further edition following in 1714, in three slightly different configurations. Rowe himself died four years later and when, at the beginning of the next decade, the Tonson firm decided to venture a further new edition of Shakespeare, they drafted in another well-known writer to succeed Rowe in undertaking the editorial work for them: Alexander Pope, the most brilliant poet of this era. Pope’s editorial sensibility was much the same as Rowe’s and he built on Rowe’s work by adding further scene divisions, additional indications of location and by generally further regularizing the text. Unsurprisingly, perhaps, he was particularly active in adjusting the metre, so that, as Thomas Lounsbury has noted, ‘words were inserted in the verse, words were thrown out, or the order of the words was changed’, adjustments which Lounsbury estimated ‘mounted into the thousands’ (1906: 108–9). Pope’s objective was, in essence, to make the plays conform as closely as possible to the aesthetic norms of his own time. Like Rowe, while making the plays easier to read in many respects, Pope also simply ignored some aspects of the text that would have been likely to have puzzled many readers. So, for instance, Caliban still offers to fetch the mysterious ‘scamels’ for Trinculo, except that now, owing (presumably) to a printer’s error, the line has become even more baffling, as he offers to bring ‘Young scamels from the ock’ (1725: 1.40). Pope does, however, tackle the question of the Hostess’s thorny ‘Table of green Fields’. He simply cuts the phrase from his text, adding a note at the bottom of the page, indicating that

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

171

This nonsense got into all the following editions by a pleasant mistake of the Stage-editors, who printed from the common piecemeal-written Parts in the Play-house. A Table was here directed to be brought in, (it being a scene in a tavern where they drink at parting) and this direction crept into the text from the margin. Greenfield was the name of the Property man in that time who furnish’d implements &c. for the actors. (1725: 3.423) Pope here offers an ingenious unravelling of the crux, but his reading lacks evidential support, since there is, in fact, no record of anyone by the name of Greenfield having served as ‘Property man’ in the theatre of the time. Pope’s comments here are, however, revealing in another way. Rowe concerned himself very little with the greater textual history of the plays or with the provenance of particular editions. He noted that the original manuscript copies of the texts were ‘lost, or, at least, are gone beyond any Inquiry I could make; so there was nothing left, but to compare the several Editions, and give the true Reading as well as I could from thence’ (1725, vol. 1, sigs A2r–v). But his comparing of ‘the several Editions’ would appear to have been little more than cursory. He registered, for instance, that F4 Hamlet lacked a scene present in the quarto line of texts (4.2, where Fortinbras appears with his army), so he added it to his edition. Likewise, the folios lacked the Prologue to Romeo and Juliet but Rowe came across it in an early quarto, so, again, he added it to his edition, though, curiously, he put it at the end of the play, rather than at the beginning. Other than this, Rowe troubled himself very little with textual matters. By contrast with Rowe, Pope had a very strong theory about the origins of the texts, as is indicated in his comments about the supposed ‘table of Greenfields’. As he says here, the ‘Stage-editors’ – by which he means John Heminges and Henry Condell, signatories to the preliminaries to the First Folio – ‘printed from the common piecemeal-written Parts in the Play-house’, so he imagined that F1 was based on texts cobbled together from the roles written out for individual actors. These texts were, he proposed in the introduction to his edition, adulterated by ‘Additions, Expunctions, Transpositions of scenes and lines, confusion of Characters and Persons, wrong application of Speeches, corruptions of innumerable Passages by  … Ignorance, and wrong Correction’ (1725: 1.xxi). So, in Pope’s view, the inherited text was corrupt and confused and required large-scale intervention to set it right. Believing this of the text, Pope felt licensed to intervene extensively as an editor. He cut from the text passages that he deemed to be corrupt – sometimes ‘degrading’ them to the bottom of the page, other times silently expunging them. In Pope’s view, the text was much the better for these excisions, and for the other alterations that he made to the plays. Just a year after Pope’s edition had appeared, a volume was published under the title Shakespeare Restor’d: or, A Specimen of the Many Errors as well Committed, or Unamended, by Mr Pope in his Late Edition of this Poet. The author was Lewis Theobald, a lawyer by training, who had been given a thorough grounding in Greek and Latin early in life and who had carved out a career as a minor literary figure and translator. His thinking about textual matters was greatly influenced by the work of

172

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

the classical scholar, Richard Bentley, who had, at this point, produced important editions of Horace and Terence. Theobald was damning in his assessment of Pope’s methods, observing that there were many places where he had substituted a fresh Reading, [when] there was no Occasion to depart from the Poet’s text; where he has maim’d the Author by an unadvis’d Degradation; where he has made a bad Choice in a Various Reading, and degraded the better Word; and where he, by mistaking the Gloss of any Word, has given a wrong Turn to the Poet’s Sense and Meaning. (1726: 134) Theobald’s critique caused something of a stir and when the Tonson firm were looking for their next editor of Shakespeare they entrusted the text to him – much to Pope’s annoyance.2 Theobald brought a very different mindset to the business of editing than Pope and Rowe had done. Whereas Pope – given his theory of how the texts had been compiled – felt entirely justified in cutting or changing anything that he did not understand or that did not match his own aesthetic criteria, Theobald, by contrast, sought, where possible, to see whether any obscurity or difficulty might be elucidated by examining its greater set of contexts. The Hostess’s ‘Table of green Fields’ in fact provides a very nice example of Theobald’s editorial method in action. In a note in his text, Theobald registers Pope’s suggestion regarding the imagined Mr Greenfield and the table he was to supply. Drawing on his own knowledge of theatre practice he dismisses the idea that a note regarding a stage property would be included in a script (cobbled together or otherwise) at this particular point in the text – rather, he argues, such a requirement would need to be signalled significantly earlier in the script, so that the property could be ready in advance of when it was actually needed. Theobald suggests that instead of the phrase being an intrusion in the text, it was more likely to have come about owing to a misreading on the part of the F1 compositor. Theobald had familiarized himself with the handwriting styles of Shakespeare’s own time and he felt that ‘Table of green Fields’ was likely to be a garbled phrase and that the misreading was, in fact, entirely characteristic of the kind of error produced by a compositor quickly reading a rushed early modern secretary hand, where particular letters, when written hurriedly, typically produced certain predictable confusions. Working backwards to what he styles the ‘Traces of the Letters’, Theobald suggests that the underlying correct reading here is likely to be ‘a’ [i.e. he] babled of green fields’, which is to say, that Falstaff’s dying delirium causes him to ‘speak indiscriminately’ of random things (1733: 4.31), such as babbling of green fields. Theobald’s argument for his reading has proved so compelling that virtually all subsequent editors – right up to the present day – have adopted it. Theobald’s way of proceeding here is typical of his approach more generally. Where Pope relies on unfounded (and uninformed) speculation, Theobald brings his knowledge of theatre practice into play in dismissing Pope’s reading, and brings an informed understanding derived from a close study of early modern handwriting to bear in making his own suggestion. Again and again in his edition Theobald attempts to elucidate the text on the basis of accumulated scholarly knowledge

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

173

and, by contrast with the Rowe and Pope editions, the pages of his text include extensive annotations, explaining outdated usage and suggesting considered solutions to particular cruces. Again, for instance, where Rowe and Pope simply ignored Caliban’s mysterious ‘scamels’, Theobald emends the word to ‘Shamois’. In a note at the bottom of the page, he registers that his chosen emendation means ‘young Kids’.3 But he also offers a number of other potential solutions to the crux: perhaps what is being referred to is the ‘Sea-mall, or Sea-mell, or Sea-mew … that Bird, which is call’d Larus cinereus minor’, or possibly it is a different bird called a ‘stannel’ (1733: 1.39). Again, Theobald draws on a range of knowledge in attempting to make sense of the text. In addition to deploying such knowledge, Theobald also pioneered the use of a ‘parallel passages’ approach to explicating difficult moments in the plays. As Peter Seary has noted, ‘because he had the plays by heart, he was familiar with recurring patterns of imagery’ (2006) and therefore was able frequently to elucidate one passage by reference to another elsewhere in the plays. This would become common practice for his successors. Theobald, in essence, set the standard for much of the editorial work that was to follow in relation to Shakespeare’s text, both in the eighteenth century and, indeed, beyond. For all that, however, there was one significant element of his work that would be regarded as falling short of modern editorial standards. We have seen that Pope had a strong theory of the provenance of Shakespeare’s printed texts (that the plays were cobbled together from actors’ parts), albeit a theory for which there was absolutely no supporting evidence. Theobald, by contrast – for all that he was influenced by the work of the textual scholar Richard Bentley – did not have a very strong sense of the importance of mapping textual lines of authority. Surprisingly, having criticized Pope’s first edition so much in Shakespeare Restor’d, Theobald actually used Pope’s second edition of 1728 as the base text for his own edition. His general approach to textual matters has been summarized by Simon Jarvis as follows: as much material as possible from early Quartos and the First Folio is to be gathered and used to correct a text whose basis is Pope’s … [D]ecisions as to whether the Quarto or Folio readings should have priority in disputed cases are often made on the basis of a variety of aesthetic or linguistic, rather than bibliographical, criteria. (1995: 101) The first editor to register clearly the significance of textual lines of descent in Shakespeare was Samuel Johnson, who published an edition with the Tonson firm in 1765. Considering the relationship among the four folios, Johnson observed that ‘the first is equivalent to all others, and … the rest only deviate from it by the printer’s negligence’ (vol. 1, sig. D1v). Here, Johnson broadly sketches out the starting point for what will become a fundamental editorial principle in literary editing: that printed texts can be arranged into a logical sequence and that the text presumed to be closest to the author’s own original has an authority which outweighs that of all other editions. This being the case, the duty of the editor is to establish which early text can be thought of as having the highest authority, with

174

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

that text serving as the copy text – i.e. the fundamental base text – for any new edition.4 In the case, specifically, of Shakespeare, where a pre-1623 quarto text of a given play exists, editors must decide whether the earlier text (or one of the earlier texts, in cases where there are multiple early quarto editions) can be considered more authoritative than the F1 text and, if so, they must make that text their copy text. While Johnson indicated that he grasped the underlying principle here, he did not, in fact, put it into practice in his own edition, instead following his predecessors in using a previously edited text as his own base text.5 Edward Capell, whose edition of Shakespeare appeared three years after Johnson’s first edition of the plays, also registered the fundamental principle of the logic of increasing deviation within a line of textual descent, observing specifically of the early quartos that ‘generally speaking, the more distant they are from the original, the more they abound in faults; ’till, in the end, the corruptions of the last copies become so excessive, as to make them of hardly any worth’ (1768: 1.13–14). Capell was the first editor of Shakespeare to build his text from the ground up, writing out his own text from scratch, rather than simply marking up a predecessor edition. Though he frequently selected readings from editions other than his base text, he can be said generally to have offered a text that was more faithful to a chosen original than any of his predecessors had done. Capell, like Theobald, was a lawyer by training, which may help to explain his grasp of the importance of differences among variant versions of the same text. A signal edition published at the end of the eighteenth century was produced by another lawyer-turned-gentleman-Shakespeare-scholar, Edmond Malone. Marcus Walsh has observed that Malone’s edition of 1790 ‘represents a step change in scholarship and method’, as his work initiated a new emphasis on primary documentary evidence, and a new determination in its pursuit; a new awareness of the historical context of Shakespeare’s drama; a rigorous and innovative investigation of biographical, chronological and sociological issues; and a focussed and selective interpretive method of real theoretical self-consciousness and cogency. (2012: 34) Walsh registers the extraordinary range of new materials that Malone brought to bear in editing the text and in attempting to map Shakespeare’s biography, and the theatrical, social and historical context of his work: parish records, the Stratford Corporation books, manuscript sources at the Chancery, the British Museum, the Stamp Office, the Remembrancer’s Office in the Exchequer and the Lord Chamberlain’s Office. As Walsh has noted, Malone’s ‘most significant discoveries included the Office-book of Sir Henry Herbert, Master of the Revels to James I and Charles I, and the account book of Philip Henslowe, proprietor of the Rose Theatre in Southwark’ (2012: 37). These discoveries, in themselves, helped to change the nature of editing and of textual and theatrical scholarship in relation to the entire early modern period, as, together, they offered a much richer understanding of the theatrical, literary and political culture of Shakespeare’s time – and of the mechanisms through which early modern culture operated.

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

175

A revised version of Malone’s text was published in 1821, with work on the edition having been completed by James Boswell, as Malone himself had died in 1812. For the first long stretch of the nineteenth century, this edition, together with the final version of Samuel Johnson’s (which had been revised and reworked at various points by George Steevens and Isaac Reed), served as standard texts, reproduced in an extended series of derivative editions. Wholly new texts did still continue to appear in this period. The editors who produced them – such as Samuel Weller Singer, A. J. Valpy, Charles Knight and John Payne Collier – were, like their predecessors, gentlemen scholars, and they followed in the general editorial track that had been established in the best work of Theobald, Capell and Malone. But the recycling of older texts in derivative editions – with the source text sometimes acknowledged, sometimes not – received particular impetus from the fact that the readership for Shakespeare broadened very considerably over the course of the nineteenth century, as the educational franchise was extended and literacy rates increased dramatically. Publishers used various strategies to attempt to sell Shakespeare into the lower end of this market. Repackaging old texts helped to save on editorial costs, thus allowing for reduced prices, and texts were also oftentimes offered on an ‘instalment plan’, as in the case of an edition which Robert Tyas began issuing in 1839, under the title ‘Shakespeare for the People’, issued in weekly numbers at just 2d each. The most successful of these popular editions was that published by John Dicks, who, in March 1864, advertized ‘The People’s Edition of Shakspere’, ‘two complete plays in every number. one penny’ (1864a). By August of that year, Dicks was offering ‘The complete works of Shakspere, elegantly bound, containing thirty-seven illustrations and [a] portrait of the author’ for just two shillings (1864b). Two years later, he repackaged this volume in a paper cover and offered it for sale for just a shilling, claiming (in a somewhat odd formulation) that it was ‘The Cheapest Book ever published for One Shilling’ (1866). All told, Dicks claimed to have sold in the region of a million copies of his various editions (1868). If this is correct, then it seems likely that he gained more sales for his derivative Shakespeare than previous Shakespeare publishers had achieved for all of their editions combined. In one of the adverts for his text, Dicks declared: ‘Clergymen and schools liberally treated with for large quantities’ (1864b). His willingness to offer an educational discount on his already extraordinarily cheap text is an indication of how important the schools market became for publishers of Shakespeare over the course of the nineteenth century (and beyond). As early as 1822, the London publisher C. Rice issued a volume entitled The School-Shakespeare; or, plays and scenes from Shakespeare, illustrated for the use of schools, with glossarial notes, selected from the best annotators, edited by J. R. Pitman. In the case of many educational editions, editing consisted largely of annotating the texts and, crucially, expurgating them to make them suitable for juvenile readers.6 A representative example here is the case of the highly popular Pitt Press schools edition of Hamlet, where, at the beginning of the play-within-the-play, the editor, A. W. Verity, changes the dialogue between Hamlet and Ophelia, reducing F1’s

176

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

ham

I meane, my Head vpon your Lap? ophe

I my Lord. ham

Do you thinke I meant Country matters? ophe

I thinke nothing, my Lord. ham

That’s a faire thought to ly between Maids legs ophe

What is my Lord? ham

Nothing. ophe

You are merrie, my Lord? (Tragedies, 267; Hinman 1996: TLN 1968–75) to hamlet

I mean, my head upon your lap? ophelia

Ay, my lord … You are merry, my lord. (1911: 68) eliminating a great deal of sexually explicit by-play in the process, and thereby saving the blushes of several generations of school teachers. By the later decades of the nineteenth century, series aimed specifically at the school market were becoming increasingly common and some editors – such as John Hunter in the UK and Henry N. Hudson and William J. Rolfe in the US – built successful careers by serving as editors of dedicated educational Shakespeare series (and other literary texts). In 1868, Oxford University Press entered the schools market with its ‘Clarendon Shakespeare’, edited initially by W. G. Clark and William Aldis Wright (and subsequently by Wright alone). Clark and Wright were both based at Cambridge – though not, of course, in English, as it was not yet being formally taught as a subject at the university at this point (the Faculty of English was not established there until 1919). Clark and Wright had worked together as Shakespeare editors before undertaking the Clarendon texts, having produced the Cambridge Shakespeare of 1863–6. The Cambridge edition was significant for a number of reasons. As we have seen, up to this point, the editing of Shakespeare had been undertaken either by writers or by gentleman scholars. Now, for the first time, the work was carried out by editors based at a university, drawing on all of the expertise and resources available to them on campus, not least the extensive collection of early modern texts in the university library and in the libraries of the individual Cambridge colleges, particularly Trinity (they also drew on the resources of the Bodleian Library at

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

177

Oxford).7 Access to these resources meant that Clark and Wright were able to develop the clearest understanding yet of the earliest texts of the plays and how they might be related to each other in sequence. Their aim as editors was to pick the most appropriate early edition as their copy text and to stick to it as far as ever possible. Where the text was deemed to be faulty, it was to be altered from the subsequent editions, the reading which has the greatest weight of authority being chosen. When none of the early editions give a reading that can stand, recourse has been had to the later ones; conjectural emendations have been rarely mentioned, and never admitted into the text except when they appeared … to carry certain conviction of their truth with them. (Clark and Luard 1860: 4)8 In essence what they produced by this process was an eclectic ‘best text’ edition, drawing readings variously from the earliest texts, while remaining generally faithful to their underlying copy text. But their apparatus enabled readers to reconstruct the exact path they had followed so that, as they assert, by their plan ‘a reader will really have the same advantage as if he possessed all the early editions, and constantly referred to them for each individual passage’ (ibid.: 6). The Cambridge text – in nine hefty volumes – was highly influential, as the dominant scholarly edition of the later decades of the nineteenth century. But it also had a separate life as a popular text, as the publisher Alexander Macmillan spun it off into a compact single volume version under the title the ‘Globe Shakespeare’. While, at a retail price of 3s 6d, the Globe could not compete with Dicks’s shilling edition (and the other ultra-low-priced editions that it spawned, by publishers such as Routledge and Ward, Lock and Co.), nevertheless Macmillan’s text proved to be an immediate and enduring commercial success, with the first run of 20,000 copies selling out within a week and the edition achieving sales of almost a quarter of a million copies by the end of the first decade of the twentieth century.9 The Globe also had a scholarly impact, as its line numbering system became the standard for references to the plays, with the result that the text was much invoked in critical commentary. The text achieved an unexpected second life almost 130 years after it first appeared, when Grady Ward used an edition of the Globe to create the ‘Moby Shakespeare’, a digital text which, in various forms, has been far and away the most commonly disseminated Shakespeare complete works on the web.10 The number of hits on Moby-based online Shakespeare editions stretches far into the millions, effectively – and unexpectedly – making Clark and Wright’s text probably the most heavily used edition of all time.11 The move of Shakespeare editing into the academy with the Cambridge edition set the template for much of what was to follow in the second half of the nineteenth century and beyond. In 1899, for instance, the publisher A. M. M. Methuen initiated the Arden Shakespeare and he recruited Edward Dowden to serve as the (initial) general editor. Dowden was one of the very first professors of English, in the modern sense that he served his entire working life as a career academic, helping to establish English as a university subject at Trinity College Dublin. He was approached to oversee the Arden partly on the basis of his having written the

178

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

highly successful Shakspere: A Critical Study of His Mind and Art in 1875. The basic architecture of the Arden series was that it was overseen by a general editor, who recruited a team of (mostly) fellow academics to produce the individual volumes, which were each issued for sale as they were completed. This became the standard template for many scholarly editions in the twentieth century, with the traditional single editor complete works edition – which, as we have seen, had been the norm in the eighteenth century – becoming increasingly less common (though there were some notable exceptions). The Arden itself was highly successful, with a second series commencing publication in 1951, a third beginning in 1995, and a fourth series currently in preparation. In the early years of the twentieth century a small group of scholars emerged who made a significant and enduring contribution to the editing of Shakespeare. These were A. W. Pollard, R. B. McKerrow and W. W. Greg. Together they came to be styled the ‘New Bibliographers’ and they were notable both for creating resources which remain valuable to scholars to this day, and for offering a set of speculations regarding the nature of Shakespeare’s texts which shaped editorial practice for most of the twentieth century. Pollard worked at what is now the British Library (then still part of the British Museum), where he became Keeper of Printed Books. He was co-editor of The Library, the foremost journal for textual scholarship, and served as one of the early directors of the Short-Title Catalogue of Books Printed in England, Scotland, and Ireland, and of English Books Printed Abroad, 1475–1640 (STC). The STC has become a singularly important resource for understanding publishing in the early modern period and its most modern incarnation allows scholars to track the entire printing or publishing record of those agents who were involved in the issuing of Shakespeare’s plays in print. R. B. McKerrow produced Printers’ and Publishers’ Devices in England and Scotland, 1485–1640 (1913) and (with Frederic Sutherland Ferguson) Title-Page Borders Used in England and Scotland, 1485 to 1640 (1932), both of which made the business of identifying the origins of books lacking indication of printer considerably easier. W. W. Greg was a prolific editor and scholar, who published – among other resources – an edition of Henslowe’s Diary and Papers (1904–8), Records of the Court of the Stationers’ Company, 1576 to 1602 (with Eleanore Boswell, 1930), Dramatic Documents from the Elizabethan Playhouses (1931) and A Bibliography of the English Printed Drama to the Restoration (1939–59). All of these resources, taken together, fundamentally changed the way in which Shakespearean textual scholarship was undertaken, since they provided ready access to the resources needed to map the publishing and theatrical culture of the early modern period. Beyond producing such resources, much of the work undertaken by the New Bibliographers was of a technical nature. A good example is a pair of discoveries made by Greg through a careful consideration of material evidence. A 1637 play entitled The Elder Brother was published in two editions in the same year and bibliographers could not agree as to which of the two texts was the earlier. Greg registered that, in one edition, a loose space had dropped down and had made a mark on the paper, which was then subsequently misread as an apostrophe by the compositor of the other edition, making this latter edition, logically, the later of the

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

179

two. As F. P. Wilson has noted: ‘The economy of the proof shows the workman’s confidence in his tools’ (1959: 324). Likewise, Greg finally made sense of a group of early Shakespeare quartos with miscellaneous dates on their title pages, which turned up together in a number of different collections. By examining the paper stock on which the editions were printed, Greg (1908) was able to demonstrate that, the title page information notwithstanding, all of the texts had been printed in 1619, at the printshop of William Jaggard. Greg’s work on The Elder Brother and on the 1619 quartos was technical in nature and relied on material evidence. Such work encouraged scholars to attend more closely to the materiality of the printed page than had ever previously been the case. A good example of this kind of work is Charlton Hinman’s two volume study of F1, published in 1963, in which, by closely tracking the repeated use of damaged pieces of type and other material evidence, Hinman was able to offer a precise account of exactly how F1 had progressed through the Jaggard printshop. It was – and remains – an extraordinary bibliographical achievement. But the New Bibliographers, collectively, also published work that was much more speculative in nature, a good deal of it focussed specifically on problems relating to the provenance of the earliest Shakespeare texts. Scholars had, for instance, long been puzzled by the fact that some Shakespeare plays appear in attenuated early texts that are significantly different from their longer equivalents. Pollard (1909; 1917) argued that these editions were, in fact, garbled texts secured illegally by pirating publishers or printers eager to cash in on the popularity of Shakespeare’s plays. He labelled these texts ‘bad quartos’ and suggested that they could simply be discounted as textually insignificant. Greg (1910) built on Pollard’s theorizing to propose a mechanism for how the texts of the ‘bad quartos’ had been arrived at. In his view, they were likely to have been created through a process of ‘memorial reconstruction’ by actors who had played minor roles in the plays when they were staged in the theatre – this, he felt, accounted for the fact that, for particular stretches, the text of the attenuated versions sometimes coincided reasonably closely with that of their longer counterparts. Another problem that editors had traditionally grappled with had been trying to decide, when faced with a choice between an early quarto (a ‘good’ quarto, in Pollard’s parlance) and an F1 text, which of the two was likely to be closer to Shakespeare’s original. McKerrow (1935) offered a suggestion intended to make this easier. Focussing on the names of characters, he proposed that, in an original authorial manuscript, produced in the heat of composition, names might be more likely to be irregular – as the author might be thinking of the same character differently in different contexts. This, could, in fact, be advanced as an explanation of the variant speech prefixes we have seen assigned to Lady Capulet in Romeo and Juliet. Viewed from the perspective of McKerrow’s theory, might the variations in her name be a sign that the underlying manuscript was Shakespeare’s original and that he was, in the heat of composition, sometimes thinking of the character as an aristocratic lady (La., Lad.), sometimes as Juliet’s mother (Mo.), sometimes as Capulet’s wife (Wife), sometimes simply as an old woman (Old La.)? But if original authorial texts could be erratic, McKerrow argued that texts prepared to be used

180

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

in the theatre would need to be consistent. So he suggested that, given two early texts, one with erratic names, the other with uniform names, the former was more likely to be based on an authorial manuscript. Partly from this simple suggestion of McKerrow’s, the notion took hold that compositors habitually dealt in essence with one of two different types of play manuscripts: either authorial ‘foul papers’ or playhouse ‘prompt copies’. One further piece of work by the New Bibliographers was also significant. The British Museum held a manuscript of a collaboratively written play entitled Sir Thomas More. There had been speculation for some time that one particular section of the play – identified as the work of ‘Hand D’ – might have been written by Shakespeare. In 1923, Pollard edited a collection entitled Shakespeare’s Hand in the Play of Sir Thomas More. In the volume, Pollard himself contended that ‘the writing of the three pages is compatible with a development into the hand seen in Shakespeare’s considerably later extant signatures’ (v) and the other contributors also, for the most part, argued in favour of identifying Hand D as Shakespeare’s. The implications here were highly significant: if Pollard and his colleagues were correct, the More manuscript provided a clear, extended example of Shakespeare’s handwriting, revealing both his characteristic way of handling letterforms and also certain of his particular spelling preferences, some of them highly eccentric (such as spelling ‘silence’ as ‘scilens’). Taking all of this work together, we can say that what the New Bibliographers appeared to offer editors of Shakespeare was a more potent set of textual tools than had ever previously been available. The troubling ‘bad quartos’ could now simply be discounted, though viewing them as the product of ‘memorial reconstruction’ potentially provided some insight into early modern theatre practices – and their clearest passages could potentially be relied on for shedding light on obscurities in the longer, ‘good’, texts. McKerrow’s suggestion offered a ready way of distinguishing between authorial and theatrical manuscripts, making the choice between early texts easier. Hand D provided a ready guide to Shakespeare’s handwriting, so that Theobald’s technique of ‘Tracing the Letters’ could be more fully refined. With all of these tools at their disposal, editors could, in other words, proceed with greater confidence than ever before. Of the New Bibliographers, only McKerrow attempted actually to produce an edition of Shakespeare. In 1929, he agreed to edit a text for Oxford University Press. His plan for the edition indicated that he aimed for a far less eclectic text than Clark and Wright had produced in the Cambridge edition – and even the helpful additions that had been introduced by Rowe, Pope and their successors were to be removed from the text itself and relegated to the bottom of the page: Text. Follow as closely as reasonably possible one early text. Emendations only admitted into text when this is obviously corrupt and the emendations can be, to some extent, at least justified. Nothing admitted merely because it seems to be an improvement. Collations of all early texts, including in some cases Restoration quartos (as possibly embodying stage tradition), but no editor’s readings except

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

181

in places of real difficulties. Eighteenth-century editors’ stage-directions, scene divisions and indications of locality, to be given in footnotes only, not in text. (McKerrow 1929) McKerrow laboured at the edition for several years, bringing in the textual scholar Alice Walker along the way to assist him. A decade into the project, however, he died, with a considerable amount of work still unfinished. Walker carried on alone for a time, but eventually the edition ran into the sand, never to see the light of day. But though McKerrow’s own text was never brought to fruition, the set of theories and editorial strategies that he had worked out in conjunction with Pollard and Greg proved to be very highly influential. They were successfully put into practice by John Dover Wilson in an edition he undertook for Cambridge University Press, as a replacement for the Clark and Wright text. Wilson had been a protégé of Pollard’s and he advanced New Bibliographic thinking in various ways (some of them highly speculative). In all, it took forty-five years for him to bring the New Cambridge Shakespeare to a conclusion; he was eighty-five when the final volume appeared. His edition was the most considered attempt to apply New Bibliographic principles to the business of editing Shakespeare, but we can say that, in essence, all significant editions up to the closing decades of the twentieth century were undertaken within the parameters of New Bibliographic thinking. Repeatedly, twentieth- century editors invoked the notions of good and bad quartos, memorial reconstruction, foul papers and prompt copies, and repeatedly they drew conclusions by reference to the handwriting of More’s Hand D. We can see this influence at play in a highly successful single-editor British edition such as Peter Alexander’s (1951) and in an American edition such as G. Blakemore Evans’ Riverside (1974), which for many decades dominated the US college market.12 In his introduction to his edition, Evans described New Bibliography as ‘a fresh and comparatively “scientific” approach to the problems presented by the text’ (34). In the multiple-editor series, too, the New Bibliographic approach was dominant, providing the core editorial principles for Arden 2 and for more commercially oriented series such as the Pelican (1956–67) and the Signet (1963–8). In the closing decades of the twentieth century, the New Bibliographic approach finally faced some challenges. A new Oxford text, edited by Stanley Wells, Gary Taylor, John Jowett and William Montgomery, sought – in some respects at least – to tack away from the author-centric methodology characteristic of the New Bibliographers, privileging, instead, the theatrical lineage of Shakespeare’s texts. The editors were partly influenced here by theories of the ‘social text’ advanced at this time by Jerome J. McGann and D. F. McKenzie.13 Broadly speaking, when faced with a choice between an authorial text and a theatrical text – between, in essence, texts considered to originate in what the New Bibliographers’ considered ‘foul papers’ and a ‘promptbook’ – the Oxford editors opted to take the theatrical edition as their copy text, on the grounds that this was the version of the text closest to that intended for delivery to the theatrical audience of Shakespeare’s time. At the same time, however, they were also in the vanguard of a movement that argued

182

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

for seeing some pairs of early texts as offering both an authorial original and a later text that had been revised by Shakespeare himself. Thus, for instance, Wells and his colleagues provided their readers with two different King Lears: the History, based on the 1608 quarto, and the Tragedy, based on the text included in F1, with the latter being presented as Shakespeare’s own revision of the former. The editors subsequently indicated that, had sufficient space been available to them in their edition, they would have wished to have ‘produced multiple texts of several other plays’ (Wells and Taylor 1990: 17).14 Beyond the Oxford edition, the fundamental premises of much New Bibliographic thinking were also being interrogated at this time by a number of different textual scholars. The utility of the distinction between ‘good’ and ‘bad’ quartos was, for instance, questioned by Randall McLeod (Cloud 1982) and others, who argued that the so-called ‘bad’ texts had their own theatrical integrity. Furthermore, in a substantial and detailed study, looking at a wide range of texts (both Shakespearean and non-Shakespearean), Laurie E. Maguire (1996) argued that there was little convincing evidence to support the concept of memorial reconstruction as Greg had originally conceived it. William B. Long (1985) and Paul Werstine (1990; 2012) argued that surviving playscript manuscripts offered little support for the clear-cut distinctions between ‘foul papers’ and ‘promptbooks’ that the New Bibliographers had proposed. A certain degree of scepticism regarding Hand D was also expressed in this period, with T. H. Howard-Hill – who edited a new volume of essays on the manuscript in 1989 – concluding that, while a weighing of the arguments ‘tips the balance in favour of Shakespeare’, it is ‘an index of the complexity of the More problem – and possibly of the insufficiency of evidence good enough to bring issues to closure – that however plausible and well supported a case may be, opposing arguments rear themselves immediately’ (8, 6).15 With the certainties of New Bibliographic orthodoxy being called into question, a number of critics called for the ‘unediting’ of Shakespeare (and of early modern texts more generally). Such scholars promoted the idea that Shakespeare and his fellows were best studied in facsimile editions, on the grounds that this would bring readers closer to the text as it had circulated in the early modern period.16 But, of course, facsimile texts are not, themselves, ‘unedited’, since any form of textual reproduction always involves making choices of one sort or another, thereby offering what is always a ‘remediated’ text. As Diana Kichuk has noted (specifically of digital facsimiles): Remediation acts like a distorting lens or opaque veil through which the scholar ‘sees’ the mediated Early English book. Scholars may attempt to penetrate the opacity of the new host medium through imagination and scholarly grit, but in the end, the facsimile is not, and perhaps never can be ‘the real thing’ or equivalent of the original. (2007: 296) In any case, we might say, the notion of forgoing editing in an effort to put the reader back in contact with a supposed original amounts, in essence, to returning us to the world of our early eighteenth-century reader who can only sit, flummoxed,

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

183

when confronted with the one-scene Antony and Cleopatra; the confusion of ‘La.’, ‘Mo.’, ‘Lad.’, ‘Wife’ and ‘Old La.’ in Romeo and Juliet; Caliban’s ‘Scamels’; and the Hostess’s ‘Table of green Fields’. The instincts of the Tonsons and their editors in setting out to clear up these confusions in the text – and of those editors who sought to make sense of the interrelationships among the earliest texts – were entirely sound in this regard: editing was necessary in order to make the text intelligible for readers. As Lukas Erne has astutely observed of editors: ‘their decisions and interventions have an enabling effect, allowing … readers to engage with Shakespeare’s dramas with greater ease and insight’ (2008: 4). In recent years, then, editors have sought to salvage as much as possible from the New Bibliographic approach, without, at the same time, being blind to the shortcomings of some of the more speculative aspects of the theorizing of individual members of the movement. Spirited defences of the underlying principles of the approach have been offered by scholars such as Grace Ioppolo (2006) and John Jowett (2007) and, in truth, we can say that most new editions continue to be heavily reliant on the resources and insights offered by Greg and his colleagues in the early decades of the last century.17 At the same time, however, editors have also increasingly begun to take advantage of the affordances of the digital edition which, in a sense, can be said to allow for the presentation of texts with varying degrees of editorial intervention, at the reader’s own choice. Thus, for instance, someone coming to the Internet Shakespeare Editions’ Henry V can opt to read the play in a digital facsimile of the F1 text or of Q1; or she can read a modern or old spelling transcription of either of these texts; and she can read the text with notes or without notes. A simple mouse click hides or brings into view the editorial history of the text. Displayed on screen, that history can be grasped quickly and easily, as it is presented as a set of colour-coded historical choices. We can see, for instance, that the ISE editor has opted to follow Theobald in rendering the Hostess’s line as ‘a babbled of green fields’, but that the F1 original was ‘a Table of greene fields’ and that the line has sometimes also been rendered as ‘and a’ talked of green fields’ (Mardock n.d.).18 More experimental work, such as that being undertaken by Alan Galey in the context of the digital version of the MLA’s New Variorum edition, raises the possibility that the variants among the earliest editions might be made dynamically visible on the screen as the viewer reads the text. In one prototype of Galey’s (n.d.), words displace each other on the screen: Macbeth’s ‘weyard’ becoming ‘weird’ becoming ‘wayward’. In another, variations appear first in bright red, shifting through brown to black, before being displaced by the next variation, which runs through the same colour shift, before cycling through the variants again. The text, in these cases, becomes a living thing, displaying its own variant history in real time as one reads. We have come a long way, then, from the world of F4, where the reader was left very much to her own devices to puzzle out the text as best she could. In our own time, the reader is afforded much more help by editors, but the digital text now also offers the reader a greater level of autonomy, with the editorial structures – evolved and refined over many centuries – reconfigurable at the reader’s own choice. If the eighteenth century witnessed the emergence of the editor as a central figure, the

184

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

twenty-first century may be said to be the age of the empowered reader, free to make her own decision as to how much or little editorial help she feels she needs.

NOTES 1. On the ownership of the Shakespeare copyrights, see Dawson (1946). 2. See, for instance, Pope’s letter to Tonson the Elder, dated 14 November 1731, included in Bernard (2015: 261–2). 3. OED, taking ‘shamois’ as a variant of ‘chamois’, suggests that a more precise meaning is a ‘capriform antelope’ – ‘chamois, n.’, definition 1.a. 4. The seminal meditation on the notion of copy text is Greg’s ‘The Rationale of CopyText’ (1966). 5. In fact, he used two previously edited texts, a combination of the William Warburton edition of 1747 and the 1757 fourth edition of Theobald. 6. Expurgation was a common element of Shakespeare publishing in the nineteenth century. In 1807, Henrietta Bowdler – from whom we derive the term ‘bowdlerization’ – produced (anonymously) The Family Shakespeare, in which ‘the various beauties of this writer’ were ‘unmixed with any thing that can raise a blush on the cheek of modesty’ (1.vii). Many other such editions appeared over the course of the century. 7. Oxford University Press had, in fact, published an edition of Shakespeare in 1743–4, but it had been edited (and underwritten) by the gentleman-scholar Thomas Hanmer. 8. The specimen in Clark and Luard (1860) was issued before Wright joined the project. Luard served only briefly as a part of the editorial team. 9. On the first run selling out within a week, see Birmingham Daily Post, 13 March 1865, 2076, p. 5. On accumulated sales, see the Cambridge University Press prizing books 17/3 through 17/9, 17/14 and 17/16, which provide details of print-run figures (CUP archives). 10. On the Globe as the source for Moby, see the introductory materials to Johnson (n.d.). 11. To take just one example: in 2000, Matty Farrow’s Shakespeare complete works website, which used the Moby text, claimed to have had 8,000,000 hits in less than a decade – see Farrow (n.d.). 12. The first significant US edition of Shakespeare was that edited by White (1857–66). Of great significance also were the variorum texts edited initially by Horace H. Furness and then by Horace H. Furness, Jr and published by J. B. Lippincott of Philadelphia between 1871 and 1928. 13. See, for example, McGann (1983) and McKenzie (1986). 14. While the notion that several of Shakespeare’s plays exist in versions which bear the mark of authorial revision was accepted by a broad range of scholars in the 1980s,

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

185

the tide would turn in the following decade. See, in particular, Honigmann (1996) who – having been a pioneering early advocate of the theory – offered a reversal of his former position. 15. But see also the evidence in favour of Shakespeare’s identification as Hand D presented in Jackson (2007). 16. Leah Marcus (1996) can be taken as a representative figure here. 17. See also, more generally, Hunter (2007). 18. See https://internetshakespeare.uvic.ca/doc/H5_FM/complete/index.html (accessed 13 November 2020), with ‘inline variants’ toggled on.

REFERENCES Bernard, Stephen, ed. (2015), The Literary Correspondences of the Tonsons, Oxford: Oxford University Press. Bowdler, Henrietta and Thomas Bowdler, eds (1807), The Family Shakespeare, 4 vols, London: Hatchard. Capell, Edward, ed. (1768), Mr William Shakespeare his comedies, histories and tragedies, 10 vols, London: Tonson. Clark, W. G. and H. R. Luard (1860), The First Act of Shakespeare’s ‘King Richard II’. Intended as a Specimen of a New Edition of Shakespeare, Cambridge: Cambridge University Press. Cloud, Random (i.e. Randall McLeod) (1982), ‘The Marriage of Good and Bad Quartos’, Shakespeare Quarterly, 33 (4): 421–31. Dawson, Giles E. (1946), ‘The Copyright of Shakespeare’s Dramatic Works’, in C. T. Prouty (ed.), Studies in Honor of A. H. R. Fairchild, 11–35, Columbia, MI: University of Missouri Press. Dicks, John (1864a), ‘Advertisement’, Reynolds’s Newspaper, 27 March, 711, p. 4. Dicks, John (1864b), ‘Advertisement’, Reynolds’s Newspaper, 21 August, 732, p. 4. Dicks, John (1866), ‘Advertisement’, The Era, 21 October, 1465, p. 15. Dicks, John (1868), ‘The Shilling Shakspeares’, The Bookseller, 1 July, p. 451. Erne, Lukas (2008), Shakespeare’s Modern Collaborators, London: Continuum. Evans, G. Blakemore, ed. (1974), The Riverside Shakespeare, Boston, MA: Houghton Mifflin. Farrow, Matty (n.d.), The works of the Bard. Available online: https://web.archive. org/web/20000301170950/http://www.gh.cs.su.oz.au:80/~matty/Shakespeare/ Shakespeare.html (accessed 13 November 2020). Galey, Alan (n.d.), ‘Visualizing Variation: Animated variants’. Available online: http:// individual.utoronto.ca/alangaley/visualizingvariation/animated.html (accessed 13 November 2020). Greg, W. W. (1908), ‘On Certain False Dates in Shakespearian Quartos’, The Library, 2nd ser., 9: 113–31, 381–409. Greg, W. W., ed. (1910), Shakespeare’s Merry Wives of Windsor 1602, Oxford: Clarendon Press. Greg, W. W. (1966), ‘The Rationale of Copy-Text’, in J. C. Maxwell (ed.), W. W. Greg, Collected Papers, 374–91, Oxford: Clarendon Press.

186

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Hamm Jr., Robert B. (2004), ‘Rowe’s Shakespear (1709) and the Tonson House Style’, College Literature, 31 (3): 179–205. Hinman, Charlton (1963), The Printing and Proof-Reading of the First Folio of Shakespeare, 2 vols, Oxford: Clarendon Press. Hinman, Charlton, ed. (1996), The Norton Facsimile of the First Folio of Shakespeare, 2nd edn, New York: Norton. Honigmann, E. A. J. (1996), The Texts of ‘Othello’ and Shakespearian Revision, London: Routledge. Howard-Hill, T. H. (1989), ‘Introduction’, in T. H. Howard-Hill (ed.), Shakespeare and ‘Sir Thomas More’: Essays on the Play and its Shakespearian Interest, 1–9, Cambridge: Cambridge University Press. Hunter, Michael (2007), Editing Early Modern Texts: An Introduction to Principles and Practice, Basingstoke: Palgrave. Ioppolo, Grace (2006), Dramatists and their Manuscripts in the Age of Shakespeare, Jonson, Middleton and Heywood: Authorship, Authority and the Playhouse, Abingdon: Routledge. Jackson, MacDonald P. (2007), ‘Is “Hand D” of Sir Thomas More Shakespeare’s? Thomas Bayes and the Elliott-Valenza Authorship Tests’, Early Modern Literary Studies, 12 (3). Available online: https://extra.shu.ac.uk/emls/12-3/jackbaye.htm (accessed 13 November 2020). Jarvis, Simon (1995), Scholars and Gentlemen: Shakespearian Textual Criticism and Representations of Scholarly Labour, 1725–1765, Oxford: Clarendon Press. Johnson, Eric M. (n.d.), Open Source Shakespeare. Available online: http://www. opensourceshakespeare.org/ (accessed 13 November 2020). Johnson, Samuel, ed. (1765), The Plays of William Shakespeare, 8 vols, London: Tonson et al. Jowett, John (2007), Shakespeare and Text, Oxford: Oxford University Press. Kichuk, Diana (2007), ‘Metamorphosis: Remediation in Early English Books Online (EEBO)’, Literary and Linguistic Computing, 22 (3): 291–303. Long, William B. (1985), ‘“A bed / for woodstock”: A Warning for the Unwary’, Medieval and Renaissance Drama in England, 2: 91–118. Lounsbury, T. R. (1906), The Text of Shakespeare, New York: Scribner. Maguire, Laurie E. (1996), Shakespearean Suspect Texts: The ‘Bad’ Quartos and their Contexts, Cambridge: Cambridge University Press. Marcus, Leah (1996), Unediting the Renaissance: Shakespeare, Marlowe, Milton, London: Routledge. Mardock, James D., ed. (n.d.), Henry V, Internet Shakespeare Editions. Available online: https://internetshakespeare.uvic.ca/Foyer/plays/H5.html (accessed 13 November 2020). McGann, Jerome J. (1983), A Critique of Modern Textual Criticism, Chicago: University of Chicago Press. McKenzie, D. F. (1986), Bibliography and the Sociology of Texts, London: British Library. McKerrow, R. B. (1929), letter to Kenneth Sissam of Oxford University Press, 15 December, Oxford University Press archives, file CP/ED/000018. McKerrow, R. B. (1935), ‘A Suggestion Regarding Shakespeare’s Manuscripts’, Review of English Studies, 11: 459–65.

SHAKESPEARE’S EDITORS FROM EIGHTEENTH TO TWENTY-FIRST CENTURY

187

Milhous, Judith and Robert D. Hume (2015), The Publication of Plays in London 1660–1800: Playwrights, Publishers, and the Market, London: British Library. Pollard, A. W. (1909), Shakespeare’s Folios and Quartos: A Study in the Bibliography of Shakespeare’s Plays, 1594–1685, London: Methuen. Pollard, A. W. (1917), Shakespeare’s Fight with the Pirates, London: Alexander Moring. Pollard, A. W. (1923), ‘Preface’, in A. W. Pollard (ed.), Shakespeare’s Hand in the Play of Sir Thomas More, v–vi, Cambridge: Cambridge University Press. Pope, Alexander, ed. (1725), The Works of Shakspear, 6 vols, London: Tonson. Rowe, Nicholas, ed. (1709), The Works of Mr. William Shakespear, 6 vols, London: Tonson. Seary, Peter (2006), ‘Theobald, Lewis (bap. 1688, d. 1744), literary editor and writer’, ODNB. Available online: https://doi.org/10.1093/ref:odnb/27169 (accessed 13 November 2020). Shakespeare, William (1623), Comedies, Histories, & Tragedies (First Folio), London: Isaac Iaggard and Ed. Blount. Shakespeare, William (1685), Comedies, Histories and Tragedies (Fourth Folio), London: H. Herringman, E. Brewster and R. Bentley. Theobald, Lewis (1726), Shakespeare Restor’d: or, A Specimen of the Many Errors as well Committed, or Unamended, by Mr. Pope in his Late Edition of this Poet, London: R. Francklin et al. Theobald, Lewis, ed. (1733), The Works of Shakespeare, 7 vols, London: Tonson et al. Verity, A. W., ed. (1911), Hamlet, Pitt Press Shakespeare, Cambridge: Cambridge University Press. Walsh, Marcus (2012), ‘Editing and Publishing Shakespeare’, in Fiona Ritchie and Peter Sabor (eds), Shakespeare in the Eighteenth Century, 21–40, Cambridge: Cambridge University Press. Wells, Stanley and Gary Taylor (1990), ‘The Oxford Shakespeare Re-Viewed by the General Editors’, Analytical and Enumerative Bibliography, n.s., 4 (1): 6–20. Werstine, Paul (1990), ‘Narratives About Printed Shakespeare Texts: “Foul Papers” and “Bad” Quartos’, Shakespeare Quarterly, 41: 65–86. Werstine, Paul (2012), Early Modern Playhouse Manuscripts and the Editing of Shakespeare, Cambridge: Cambridge University Press. White, Richard Grant, ed. (1857–66), The Works of William Shakespeare, 12 vols, Boston, MA: Little, Brown. Wilson, F. P. (1959), ‘Sir Walter Wilson Greg 1875–1959’, Proceedings of the British Academy, 45: 307–34.

CHAPTER 2.8

The modern editing of Shakespeare: The text MARGARET JANE KIDNIE

As we enter the third decade of the twenty-first century, the reader seeking to buy a copy of Hamlet to read for pleasure, study or performance is presented with a wealth of choices. There are editions that print the play with ‘Additional Passages’ after the last scene, and editions that enclose portions of the dialogue within square brackets, diamond brackets or both; there are ‘enfolded’ editions that display variants within single lines; and there are multi-volume and multi-platform editions that offer not just Hamlet, but three-text and even four-text Hamlets. An increasingly small portion of the editions available for purchase provide a clear reading text without the option of a different Hamlet or more Hamlets somewhere else. In one way or another, editors of Hamlet increasingly prioritize a fragmented reading experience that might at first glance seem alienating or bewildering to anyone who just wants to read a play. It is a trend that is by no means limited to this single tragedy. Presentational decisions such as these are driven by interpretive analyses of the early printed editions to which modern editors are seeking to do justice. Hamlet, along with about half of Shakespeare’s canon, survives in multiple editions printed during or just after the author’s lifetime. And like a number of these plays that are extant in more than one edition – including such favourites as King Lear, A Midsummer Night’s Dream and Henry V – some of the earliest editions of Hamlet differ strikingly from each other in terms of the text they present as Shakespeare’s drama (Greg 1955: 316, 312). Major discrepancies across early editions of what is supposed to be the ‘same’ play present textual scholars with puzzles and editors with methodological challenges. One wonders, for example, how the texts printed in these early editions came to differ. Which one – if any of them – accurately, or most accurately, represents the play as Shakespeare wrote it? We have only three pages in manuscript that might be in Shakespeare’s hand (see the chapter by Shrank and Werstine in this volume), and none of Shakespeare’s drama published in print survives in manuscript. The analysis of the plays’ transformation as they journeyed to completion (however one defines ‘completion’) that characterizes textual study of a twentieth-century author such as James Joyce is therefore hampered in Shakespeare studies for lack of evidence. Shakespeare, who was a sharer with the Chamberlain’s Men (subsequently the King’s Men), built a career working closely with other professional theatrical artists.

THE MODERN EDITING: THE TEXT

189

Should one value as collaboration or corruption the potential contributions to his writing made by these other creators (influence that might have taken the form of inscription, performance and/or verbal suggestion)? To take a specific example, Shakespeare’s plays, like those of every playwright in early modern England, were subject to the censor’s approval. Do we value alterations to a play such as Richard II that were likely made to accommodate the censor as highly as changes made arguably for ‘aesthetic’ reasons, potentially at the behest of the players? This chapter seeks to navigate and provide a background to debates guiding editors’ preparation of text; drawing illustrative examples from Hamlet, I attend, in particular, to practices of conflation and single-text editing. A concluding section addresses practices of modernization, standardization and editorial layout.

ECLECTIC EDITING AND THE CHALLENGE OF REVISION Barbara Mowat (2001) provides a superb account of periods of seismic change in the history of Shakespeare textual studies up to 2000. Borrowing the concept of ‘paradigm shifts’ from Thomas Kuhn, Mowat argues that momentous change in the 1760s and 1909 was propelled either by new information or through the introduction of innovative conceptual approaches to existing information. The eighteenthcentury discovery that nine plays printed in the First Folio (F1) are reprints of earlier quartos exposed as ‘demonstrably false’ the widely held premise that F1 was printed from ‘original authorial manuscripts that replaced mutilated quarto printings’ (2001: 18). As Mowat explains, this realization helped textual scholars understand that Shakespeare’s plays typically deteriorate rather than improve over sequential reprints, and it inspired an editorial method, still foundational today, that places a priority on the earliest substantive witness, whether that is F (instead of the Fourth Folio of 1685) or an earlier quarto from which a text in F was printed. Alfred W. Pollard’s 1909 argument that Heminges and Condell’s First Folio disparagement of the quartos as ‘maimed’ and ‘stolen’ applies not to all of the early quartos but only to five of them (Q1 Hamlet among them) does not, by contrast, unearth new information but instead provides a reassessment of already available evidence (ibid.: 19). By characterizing just five quartos as ‘bad’, Pollard reinvigorated research into the remaining quartos, so laying the groundwork for landmark studies in what came to be known as the New Bibliography, a legacy we continue to work through today. With renewed faith in the integrity of the manuscripts behind the First Folio and most of the quartos, bibliographers and textual scholars, following Pollard’s lead, sought to discern the manuscripts that lay behind the extant early editions in an effort to establish modern editions that might bring readers as close as possible to the plays as written by Shakespeare. For R. B. McKerrow, if the author submitted ‘a good and legible fair copy’ of his play, then this manuscript could be used to guide performance; if the author’s copy, however, ‘contained revisions or corrections, or was not clearly written’, then it would be transcribed, and that transcription would instead serve as the book held by the company prompter (also called the bookkeeper), during performance (1931: 264).

190

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

W. W. Greg formalized this premise into a category distinction between authorial ‘foul papers’ and theatrical ‘promptbooks’ (1942: 22–48). An author’s foul papers ‘must have contained the text substantially in the form the author intended it to assume though in a shape too untidy to be used by the prompter’, but Greg supposed it would be ‘roughly written and show some loose ends’ (31, 47). Other distinctive markers for Greg of an underlying foul papers manuscript were ‘indefinite or permissive directions’ such as ‘Enter four or five citizens’ or ‘Either strikes him with a staff or casts a stone’ (36): a book-keeper, Greg insisted, would need clarity on such basic matters as the number of actors who enter a scene and their required props. Drawing on his study of surviving playhouse manuscripts from the period, he showed that playhouse functionaries would often duplicate directions and sometimes add anticipatory entrances as warnings to themselves of stage business to which they especially needed to attend (37–9). Greg supposed that actors’ names might potentially be introduced by authors, but noted that this claim ‘is unsupported by any evidence in the extant manuscripts, in which all actors’ names are added by the book-keeper’ (40). Greg warned scholars in 1942 that the available evidence concerning distinctive authorial and playhouse habits resists easy interpretation and should be handled with ‘caution’ (1942: 36, see also 37–8). His summary of the differences between foul papers and promptbooks thirteen years later in The Shakespeare First Folio was nonetheless adopted by editors as a kind of road map with which to approach study of Shakespeare’s early printed editions (1955: esp. 141–2). The accepted editorial practice, following Greg, was to interpret – on the basis of such features as stage directions, ghost characters, variant speech prefixes, actors’ names and unresolved textual tangles – whether foul papers or else a fair copy, perhaps readied for performance, lay behind a particular early edition. The evidence was not always clear, and one sometimes had to factor in intervening scribal transcripts and forms of mixed copy, but at least the work of analysis had a place to begin. Presumed false starts have supported editors’ inference that Q2 Hamlet, for example, was printed from Shakespeare’s foul papers. As Harold Jenkins explains in his Arden 2 edition, The two clearest cases occur in a passage of three lines [in one of the Player Queen’s speeches] which, as evidence of foul papers, might be thought decisive in itself:

For women feare too much, euen as they loue, And womens feare and loue hold quantitie, Eyther none, in neither ought, or in extremitie … (III.ii.162–3)

That the first line here is a reject is shown by its lack of a rhyme in the midst of regular couplets and by the rephrasing of the same thought in the line which follows. Then at the beginning of the third line the extra-metrical ‘Eyther none’, again immediately rephrased, is another evident false start. (1982: 41)

THE MODERN EDITING: THE TEXT

191

Inferences about the underlying copy enable strategies of emendation. An editor who seeks to present the play according to what she infers were the author’s final intentions, for example, might choose to emend these redundancies as errors, and Jenkins does just that. Eclecticism is the editorial model most closely associated with the New Bibliography; it was devised by Greg and set out in his immensely influential ‘Rationale of Copy-Text’, and Jenkins’s edition of Hamlet is an exemplary instance of the method. Pointing out that compositors frequently imposed their own spelling and punctuation on the texts they set into print, Greg proposed minimizing the effect of such intervention by basing any new edition on the printing considered to stand closest to the author and his habits of writing. Sometimes, however, there exists a later edition that was either authorially revised or is of comparable authority to the earliest edition (1950/1: 23). In these cases, rather than mechanically import into one’s copy-text ‘all the alterations of that edition, saving any which seem obvious blunders or misprints’, as recommended by McKerrow (25), Greg argued that in face of variant readings the editor must use her judgement to decide which alternative is likely authorial (28–9). The later variant would be accepted, Greg suggested, if it corrects a likely error in the earlier edition and is not itself a likely transmission error, or (in the case where both readings are acceptable) if it seems the result of authorial revision; this revision should be accepted ‘whether the editor himself considers it an improvement or not’ (32). The result is a conflated text that eclectically gathers together readings from multiple extant early editions. Greg accepts that this methodology will sacrifice ‘completely indifferent or manifestly inferior’ authorial changes, but considers that in the hands of ‘an editor of reasonable competence’ this procedure will produce better results than slavishly importing all of the alterations (32). The design of the eclectic edition is guided by the desire to recover the plays as Shakespeare finally intended them, dressed as far as possible in the author’s own peculiarities of spelling and punctuation. Critics have since objected, however, that eclectic editing produces texts that have no correspondence to any early edition that actually exists. The eclectic edition instead offers an ideal text grounded in the editor’s conception of the work as she thinks Shakespeare, given the chance, would have preferred to have seen it printed. As Clifford Leech readily acknowledges, ‘this composite text will be in some degree a thing of [the editor’s] imagining, based on calculated guesswork’ (1970: 65). Greg’s ‘Rationale’ was flexible enough to handle cases of substantial authorial revision (34–5), but there was strong resistance to this view of Shakespeare prior to the last quarter of the twentieth century. The test case for Shakespeare as a revising author was King Lear (Warren 1978; Urkowitz 1980; Taylor and Warren 1983), and the position quickly drew in other plays, Hamlet foremost among them. As the consensus of opinion shifted, so too did editorial methodologies evolve. If one is persuaded that two early texts are witnesses to separate artistic visions, then a modern edition that eclectically picks among variants between them will seem to do a disservice to the work as represented in each of those texts. The editors of the 1986 Oxford Shakespeare Complete Works (Wells et al.) therefore introduce the term

192

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

‘control text’, which describes the early edition on which their new edition of a play is based. It is distinct from Greg’s term copy-text ‘in that it implies that the [early] text is the predominant basis for the edition but not the only one’ (Jowett 2007: 194–5). The editor may introduce emendations along with spelling and punctuation variants from a different early edition, for example, while still aiming to preserve the integrity of the revised or unrevised version as witnessed in the control text. This methodology resulted in Oxford’s publication, in the same volume, of The History of King Lear (based on Q1608), and The Tragedy of King Lear (based on F1). The theory that Shakespeare was a revising author draws editors into philosophical and political debates about the nature of both authorship and dramatic art. How much divergence, and of what sort, is one willing to absorb before considering these texts are no longer, not quite, the same play (Kidnie 2005)? Q2 and F1 Hamlet are quite different, but in the mid-1980s the two texts were not considered different enough to warrant the same two-text treatment as Q-F1 Lear (Q1 Hamlet, characterized by Pollard as a ‘bad’ quarto, was and still is considered a textual outlier). Without the same supporting scholarship in place, ‘[i]t was not yet at all clear that the rewriting of Hamlet was as important for one’s interpretation of the play as the rewriting of King Lear’ (Wells and Taylor 1990: 17). The Oxford editors instead published F1 as Shakespeare’s revised version, with Q2-only material included immediately after as ‘Additional Passages’. Their edition of Hamlet nonetheless has a blended textual authority that is consonant with many aspects of Gregian eclecticism: whereas F1 provides the control text for variant words and passages (the text’s ‘substance’), Q2 provides the control text for spelling and punctuation (its ‘incidentals’) since they argue Q2 was printed from the author’s foul papers (Wells et al. 1987: 146). As the examples of King Lear and Hamlet as published in the 1986 Oxford edition show, the belief that variant early printed texts provide alternative versions of a play, and that valuable evidence is lost – or, more polemically, misrepresented – when editors eclectically conflate early texts into a single modern edition, can shape quite different editorial methodologies. The New Folger editors offer yet a third presentational option. Mowat and Paul Werstine base their edition of Hamlet on Q2, and, not unlike an eclectic editor such as Jenkins, they bring into that text as many F1-only passages as is possible. They signal their different source materials within the modern reading text, however, by placing Q2-only readings in [square brackets], F1-only readings in , and their own editorial interventions in ⌜half brackets⌝. As they explain, this is an edition that seeks to appeal to a variety of reading constituencies: ‘It is designed both for those who prefer the traditional text of Hamlet … and for those who prefer to regard the Second Quarto and First Folio as distinct versions of the play’ (1992: xlix). Synoptic editions prepared by Jesús Tronch-Pérez (2002) and Bernice W. Kliman (1996; 2004) follow through the logic of such efforts to ‘enfold’ Q2 and F.

AUTHORIAL AUTHORITY Evidence and arguments assembled to support the position that some of Shakespeare plays survive in alternative versions unravelled, seemingly overnight, accepted practices of editorial conflation (Wells 1991: 186). But even as a great deal changed,

THE MODERN EDITING: THE TEXT

193

much remained the same. Revisionist arguments tend to lay a strong emphasis on specifically authorial revision: the agent behind the revision of Hamlet has a face and a name, and it is Shakespeare’s. There is thus clear continuity with the foundational work of the New Bibliography, which places its faith in ‘the possibility of seeing Shakespeare plain’ (Williams 1971: 77). The textual evidence, however, might seem to admit of alternative interpretations. If the Q2-only passages in Hamlet, for example, are cuts that form ‘part of a logical and coherent process of revision designed to make a better acting version with a wide appeal’ (Hibbard 1987: 109), then might these cuts have been made by the company book-keeper? One of the book-keeper’s responsibilities was preparing scripts for performance, a task that not infrequently involved cutting the text (Werstine 2012: 148–99, 243). As Eric Rasmussen explains, a book-keeper with ‘philistine inclinations’ is ‘absolutely necessary’ to the story that a play such as Hamlet underwent specifically authorial revision between the Q2 manuscript and its transcript, since ‘an intelligent book-keeper would interfere with the case for attributing all of the cuts to Shakespeare’ (1997: 444). Beneath the hypothesis that Shakespeare revised Hamlet runs McKerrow’s longstanding theoretical premise that playwrights delivered to companies manuscripts that might need transcription before being readied for use in performance. In the specific case of Hamlet, the argument is usually, following the Oxford editors’ 1986 rationale, that Shakespeare took the opportunity to revise the tragedy himself as he transcribed his own foul papers. These assumptions about the orderliness of theatrical manuscripts have since been overturned (Long 1985a; 1985b; 1989; 1999; Werstine 2012). There is no necessary reason to assume that Shakespeare’s manuscript underwent a process of transcription before the play went into performance, and so no obvious moment when the author would have had the opportunity to revise his tragedy in its entirety before the book-keeper introduced to the manuscript whatever alterations he considered it required. The textual disparities between Q2 and F1 strongly suggest that Hamlet met with adjustments and likely revisions in the first two decades of the seventeenth century. But how, and by whom, remain questions to be asked. To summarize the situation at the turn of the millennium, Greg’s categories of knowledge (foul papers and promptbooks) and his assumptions about processes of textual production in the playhouse were found to run counter to the available archival evidence. But if Greg’s conception of authorial manuscripts (specifically, as he presented them as distinct from playbooks) are too rigid – a point that seems to have been conceded in the decade since the 2012 publication of Werstine’s Early Modern Playhouse Manuscripts – then how is editorial practice to proceed? I would argue that this is a challenge with which a new generation of textual scholars and editors is still wrestling.

SINGLE-TEXT EDITING The recovery of the manuscripts described by Heminges and Condell as Shakespeare’s ‘True Originall Copies’, freed of the contaminating influence of print mediation – an ambition vividly figured by Fredson Bowers as the effort to ‘strip the veil of

194

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

print from a text’ (1955: 87) – is now understood by many as a goal premised on a ‘fantasy’ (Kastan 1999: 69), on an ‘immaterial abstraction, an enabling figment of the bibliographic and critical imagination’ (de Grazia 1988: 71), as ‘a product not of reason but of desire’ (Werstine 1990: 81). One hesitates in the face of such critique to risk saying much at all about either the author or the manuscripts that once must have provided copy for the early editions that have come down to us. It is in this intellectual climate that one sees a turn towards forms of single-text editing. Instead of treating each of Shakespeare’s early editions as a witness to something else, something antecedent to it, to which we no longer have access, single-text editing insists on the integrity of that edition as ‘an independent entity’ (Thompson and Taylor 2006: 92). As an editorial methodology, single-text editing seeks to give (more) interpretive control to readers, affording them the opportunity to assess for themselves a moment in a play’s history or, where multiple printings are separately edited and made available for the reader’s consultation, moments in a play’s evolution. Suzanne Gossett and Gordon McMullan thus explain in the ‘General Textual Introduction’ to the third edition of the Norton Shakespeare that they ‘have chosen to edit the texts we actually have, not the play or the poem we do not, to accept uncertainty, and to exercise a certain skepticism toward earlier claims that sometimes made the editor seem a substitute for Shakespeare’ (2016: 85). They therefore encourage readers ‘to read, compare, and contrast the two (or in the case of Hamlet, three) early texts of each of the plays for which multiple texts exist’ (86). Ann Thompson and Neil Taylor, likewise adopting a single-text methodology for their Arden 3 edition of Hamlet, are comfortable with the possibility – even likelihood – that the early editions contain collaborative writing of some nature. Since internal and/or external evidence attests to the involvement of both Shakespeare and his company in all three of those editions, ‘the issue of whether or not “playhouse practice” or “actors’ interpolations” have contaminated the author’s text is not, for us, a major consideration’ (2006: 510). Single-text editing, as these statements indicate, allows an editor to step past theoretical debates about the author, historical questions about forms of collaborative writing, and ontological conceptions of ‘the work’ as it exists (or not) apart from the evidence of a play’s extant early editions. These questions remain of interest, at least to some, but they do not drive the single-text methodology, whose ‘more modest’ ambition is ‘to correct a small number of verbal errors that were inadvertently committed by early scribes or compositors’ (Taylor and Bourus 2016: 52). For Sonia Massai, reviewing the state of the field in 2016, single-text editing ‘represent[s] an important corrective’ to a variety of speculative hypotheses about Shakespeare’s lost manuscripts (70). The prospect of error, however, inevitably and necessarily broadens the editor’s scope of analysis beyond the single early printing. Any time a text is transmitted, whether by a scribe, an author, or a compositor, errors will be introduced, but the identification of error now is the product of editorial judgement. Eclectic editors working in the Gregian tradition have something of ‘an open invitation’ to correct their copy-text since any variant in a later printing might be deemed authoritative (Thompson and Taylor 2006: 513). Their emendations as a result can sometimes

THE MODERN EDITING: THE TEXT

195

seem heavily dependent on editorial taste and ‘literary preference’; Thompson and Taylor’s sampling of past editorial emendations to Hamlet is instructive in this regard (ibid.: 512–16). Single-text editors are typically more restrained in the number of emendations they introduce. Even minimal correction, however, depends on a theory of error and a strategy for dealing with it. This, in turn, necessarily returns single-text editors to consideration of the early edition with which they are working, the underlying document it seeks to represent, and how its transmission into print might have gone wrong (Jowett 2017c: lxii; Thompson and Taylor 2006: 507). Editors thus find themselves returned to the crossroads with which the previous section ended. Editing ‘the texts we actually have’ depends on thinking, even speculating, about ‘the play or the poem we do not’, or at least about its no longer extant manuscript(s). What, if anything, can one say about the copy from which the early editions were set? Recent editions of Hamlet are revealing in this regard. Norton 3 characterizes the manuscripts behind Q2 and F1 as authorial foul papers and theatrically modified promptbook (Dawson 2016: 1761–2), a model also found in Arden 3 which specifies, however, that F1’s revisions are likely authorial (Thompson and Taylor 2006: 509). The New Oxford Shakespeare, based on Q2, is explicit that the Q2 manuscript saw ‘some theatrical annotation’, but suggests these markings were probably light, the manuscript otherwise conforming to the kind of manuscript Greg describes as foul papers (i.e. ‘an authorial manuscript that gave a fully developed script of the play’ (Jowett 2017a: 1118)). Greg’s modelling of dramatic manuscripts perseveres, even in the context of single-text editing, perhaps because it has proven so enabling of editorial decision making. Holding with a paradigm, however, that demonstrably fails to correspond to available empirical evidence is a risky editorial strategy: the conceptual models we bring to a study of Shakespeare’s early printed texts inform what we think we find in that material. An example of the scholarly lens shaping what it sees can perhaps be seen in John Jowett’s discussion of copy texts in the New Oxford Shakespeare. Jowett, presenting Shakespeare as a revising author, takes issue with editorial methodologies as inspired by Greg: ‘eclectic editing was given a check in the 1980s, when textual theorists began to see the conflated text as an idealized construct that related insecurely at best to any historically existent or lost state of the text. Recent editorial theory has sought to purge the Shakespeare editorial tradition of its apparent addiction to hybrids of mixed quarto and Folio origin’ (2017b: xlii). The unspoken premise is that an early edition corresponds to its underlying manuscript sufficiently closely to reveal different versions. A modern conflated edition may not relate to any ‘historically existent’ text, but its relation to the shape of the ‘lost state’ of the text (my italics) is more than we know. What is at stake here can be set out with reference to Hamlet. Scholars are increasingly prepared to accept that a duplicated stage direction in the last scene of Q2 Hamlet signals the hand of a book-keeper in the underlying manuscript (Jowett 2017a: 1118). Playbooks from the period are frequently marked with cuts introduced in a variety of hands, but primarily those of book-keepers and authors. We lack clarity about how compositors dealt with any cuts (or additions) they found in the

196

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

manuscript they were setting into print. If one postulates that some or even many of the cuts one can trace between Q2 and F1 Hamlet were marked in the manuscript behind Q2 but ignored in the printing-house, then one has the opportunity to think about a lost manuscript that once contained texts, rather than a text (for parallel conceptual frameworks as they apply to The Second Maiden’s Tragedy and The Lady Mother, both plays that survive in manuscript, see Purkis (2016: 130–7) and Rasmussen (1997: 455–6)). By this logic, the Hamlet manuscript in Roberts’s printing-house in 1604 might have looked something like Q2, but simultaneously something like F1, depending on how one navigated the annotations. (The scribe(s) who prepared the manuscript behind F1 would likewise have been required to make decisions about how to reproduce their exemplar(s).) How secure are we in our knowledge that none of the F1-only passages was in the Q2 manuscript (Kidnie 2020)? A single-text modern edition would not self-evidently capture the complex textuality of such a manuscript any better than would, for example, a conflated, or partially conflated, edition. Each methodology would approximate something ‘true’ about that manuscript, without exhausting – or even indicating – its textual flexibility. Mowat concludes her 2001 discussion of paradigm shifts by noting the ‘puzzling’ sway ‘Greg’s “foul papers” / “promptbook” categories and characteristics’ holds on editorial practice (24). That influence is still apparent two decades on. Mowat tentatively offers that perhaps ‘the only way out of such a crisis as that in which the field of Shakespeare textual criticism currently finds itself would be the establishment of a new paradigm’ (25). James Purkis, however, describes the sensation of déjà vu evoked by Werstine’s evidence-based 2012 refutation of Greg’s identification of printer’s copy. Just over a hundred years earlier, Greg savaged Sidney Lee’s ‘suppositions about the characteristics of manuscripts’ along with his criteria for identifying prompt copy on precisely the same grounds, ‘leav[ing] the … distinction between the New Bibliographic “advance of scholarship and knowledge” and the “revolt” against it decidedly unsteady’ (Purkis 2016: 9). Our way through this editorial crisis, as Mowat describes it, may depend less on new paradigms than on accepting and working through what Purkis theorizes as our ‘inheritance’ from an earlier generation of scholarship (6) – on the rehabilitation, in other words, of an already available, evidence-based paradigm. Instead of avoiding consideration of Shakespeare’s manuscripts in order to focus on extant printings (‘the texts we actually have’), only to see discredited ideas about those lost manuscripts make a necessary return by way of copy-text choice and textual emendation, we might set out to reconsider the available evidence, weigh certain kinds of likelihood, and so attempt and test different inferences about printer’s copy. That Greg’s identifications of printer’s copy ‘fail to find support in empirical evidence’ (Werstine 2012: 9) is a salutary warning. It is not the case, however, that textual scholars can say nothing about Shakespeare’s manuscripts. The interpretations scholars come to advance will be ‘more constrained’ than those attempted by Greg (ibid.: 221), who sought to discern in the evidence conventions of manuscript regularization that he presumed should be there. And yet precisely by merit of such redefined parameters, textual scholars have an opportunity to rethink, and potentially re-vision, models of editorial methodology and textual error.

THE MODERN EDITING: THE TEXT

197

‘MODERN-DRESS’ EDITIONS The process of establishing a text extends for editors well beyond the choice of copy-text, and decisions about whether and how to conflate or emend. One might characterize a process of editorial mediation that includes modernized spelling and punctuation, standardized speech prefixes, verse-line layout and modifications to both stage directions and act-and-scene breaks as offering readers a form of figurative ‘modern-dress’ Shakespeare. This final section will engage with some of the complexities, and the issues of interpretation they raise, that underpin these features of modern editions. Orthographic conventions guiding early modern use of the round and long s, along with the letters i and j, and u and v, are usually fairly readily transferred to their modern equivalents (but see McLeod 1991: 250–3). Spelling, however, was non-standardized in early modern England, and modernizing editors, inevitably, flatten out nuances of usage. David Bevington has argued, for example, for the interpretive value of negromancy (for necromancy) because it makes visible etymological confusions that speak to cultural uncertainties about sorcery’s roots in either the black arts (negro-) or communication with the dead (necro-) (2004: 147–8). The now distinctive meanings of travel (to journey) and travail (to labour), to take a different, oft-cited example, were captured in early modern England by the single word travail, which could also be spelled travel. This word’s ambiguity of meaning – whether one reads old- or modern-spelling editions – can only be accessed by non-specialists today through commentaries. Forced to choose between spellings, modernizing editors prioritize one sense in the reading text, relegating the other to a note (the debate for and against old- and modern-spelling editions is set out in Giddens 2011: 148–53). The current editorial trend is towards full modernization, and Stanley Wells’s Re-Editing Shakespeare for the Modern Reader (1984) and Modernizing Shakespeare’s Spelling (1979) remain foundational discussions of it. Even with these guidelines to hand, however, and supported by historical usages provided in the Oxford English Dictionary, editors will wrestle with difficult decisions. The goal is to modernize spelling where a more familiar form of the same word is now standard, but to preserve distinct words even if they are archaic or obsolete. This distinction, however, admits grey areas. The obsolete mo (i.e. more in number, as distinct from more which meant greater in size), is usually modernized to more, which now embraces both senses. Proper names, on the other hand, can set an antiquated but extremely familiar usage against a policy of comprehensive modernization (is the battlefield ‘Agincourt’ or ‘Azincourt’?). Dialects and foreign languages raise especially difficult questions about the representation of racialized difference (Dimmock 2015). Metre becomes a factor where modernization would alter the verse line’s scansion: should one retain ‘shrieve’, for example (an obsolete spelling of ‘sheriff’), if modernization alters the verse line’s scansion? Faced with this problem in Heywood’s A Woman Killed with Kindness, I decided in the end to modernize, recognizing that neither option is ideal. Metre is likewise at play when deciding how and even whether to represent the sounded inflection of the past tense. Early modern texts consistently indicate an unvocalized -ed ending by eliding the unsounded e,

198

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

sometimes but not necessarily with an apostrophe (propos’d, proposd). Arden 2 uses an apostrophe to signal unsounded preterites (propos’d), while the New Cambridge Shakespeare instead signals a sounded preterite with an accent (proposèd). Neither of these modernized alternatives is necessarily intuitive for modern readers who might simply read past both apostrophes and accents. Although the early modern convention is especially helpful as an aid to verse scansion (to those who understand the system), one might decide that cues to scansion are better addressed for modern readers in the form of commentary notes. Wells has less to say about modernized punctuation than spelling, but the challenges here are immense. William H. Sherman commented in 2013 on the ‘minimal guidance and total power’ afforded modernizing editors when it comes to punctuation, ‘while readers and viewers, for their part, are almost entirely unaware of the silent performances that have profoundly shaped the Shakespeare they see and hear’ (para. 10). Early modern writers and compositors tend to point rhetorically, shaping pauses and breathing, whereas we tend to use punctuation today to mark grammatical phrases (Graham-White 1995: 67–83). The gap between these habits can occasion substantial editorial intervention. In some cases the sense of the passage is clearly altered, and such instances are typically recorded in the collation line as emendations. More often, however, an editor’s shaping of a speech is not precisely an emendation, and yet it powerfully informs interpretation, pacing and tone; Eugene Giddens’s comparison of six modern treatments of the same five-line passage in Pericles makes clearly visible the effect of editorial mediation (2011: 153– 8). Excellent discussions of editorial punctuation – and its potential pitfalls – are provided by Anthony Graham-White, Barbara Hodgdon (2003), Sherman (2013) and Michael Warren (1977). Presenting Shakespeare’s plays in ‘modern-dress’ extends to the layout of verse lines on the page. Early modern authors and compositors marked verse lines shared among multiple speakers, but not as systematically and visibly as effected by many modern editors. In the following passage, for example, the lineation of Hamlet’s speech implicitly signals that his opening part-line metrically completes Horatio’s part-line immediately preceding it (see Figure 2.8.1):

FIGURE 2.8.1  Hamlet (Q2, 1604–5), sig. D2r. Folger Shakespeare Library, call number: STC 22276. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License.

Modernizing editors, following George Steevens (1778), usually signal that a line is shared between multiple speakers by indenting the second part-line. These shared

THE MODERN EDITING: THE TEXT

199

lines, as Abigail Rokison has argued, can carry significant weight with close readers, and actors especially, since actors are trained not to pause when they pick up the second-half of a line (2009: 15). But determining how – and even whether – to link part-lines demands substantial editorial interpretation. The lineation of a passage such as the one below is a case in point (see Figure 2.8.2):

FIGURE 2.8.2  The Taming of the Shrew, from the First Folio (1623), sig. T3v. Folger Shakespeare Library, call number: STC 22273 Fo.1 no.19. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License.

Hodgdon notes that most twentieth-century editors, ‘prescrib[ing] … orality’, stage the four-line sequence about the mutton (running from Petruccio’s first part-line through his servants’ replies) as a single verse line (2003: 99). The scene is in verse, but is this exchange verse? It might be verse or prose as printed in F, but must be one or the other in modern editions that visually stagger shared verse lines. A sequence of three part-lines, to take a different example, might suggest a shared line, followed by a part-line, or else a part-line followed by a shared line. Alternatively, the middle line might complete the part-line before and be itself completed by the part-line after (Bevington 2005: 276), a metrical effect that is obscured by indentation. George T. Wright’s Shakespeare’s Metrical Art (1988), while not providing solutions to these thorny editorial questions, offers an essential and accessible resource on early modern versification. Act and scene breaks are another important feature of modern layout. The logic of scene breaks rests on the cleared stage: a scene ends when the stage is emptied of actors (on scenic continuity and discontinuity, see Hirsh 1981: 211). Understanding when a scene is cleared requires, however, an alertness to early modern staging practices that sometimes transfer action from one fictional space to another without the actors leaving the stage (Dessen 1984, esp. 84–104). Act breaks, by contrast, originated in the indoor theatres where pauses in the action allowed for the trimming of candles and created opportunities for additional music (Taylor 1993: 31). Play quartos printed between 1591 and 1616 suggest that dramatists writing for the public theatres (where performance was experienced as continuous action)

200

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

only eventually adopted act breaks, likely increasingly from 1607 (Jewkes 1958: 40, 55, 98–9). None of Shakespeare’s quartos prior to Othello (1622) contains act breaks, and although they appear in twenty-eight of thirty-six Folio plays, these are likely the work of other hands. Gary Taylor argues that the decade or so from 1608 (when the King’s Men began leasing Blackfriars) marked a crucial transitional period, and offers that folio act breaks might indicate playhouse practice (1993: 30–43). James Hirsh, who finds patterns of act placement that suggest just two individuals divided the plays (one of whom he identifies as the scribe, Ralph Crane), instead interprets them as printing-house accretions (2002: 229). It would be hard to overstate the impact of the Horatian five-act structure on the reading experience. The editorial tendency is to retain a familiar break even when scholars agree that the positioning of a new scene or act is illogical or they question if a play follows a scenic (or other) underlying artistic structure. Folio Dream implicitly cues Titania, for example, to lie asleep onstage throughout the break between the second and third acts, while a scene break in Romeo and Juliet before the so-called balcony scene (2.2), first introduced by Hanmer (1743–4), implies the stage is cleared despite Romeo’s continued onstage presence. The 1986 Oxford Complete Works broke with editorial convention where breaks seem especially poorly placed, and some editors have since continued this revisionist process. Ease of reference for students who bring different editions of the plays into the classroom remains, however, a powerful potential deterrent to editors who consider relocating act and/or scene breaks. Digital environments can offer editors opportunities not available in codex. Alan Dessen, for example, advocates the development of online platforms able to support alternative editorial presentations of key passages, so enabling readers to explore the ‘difference … division into acts and scenes make[s] to our understanding’ of a play (2015: 341). As twentieth-century performance studies continued to grow in importance, editorial methodologies designed to shape readers’ engagement with the plays as texts for performance came increasingly into focus. Wells (1984) is again foundational to this conversation, as is Suzanne Gossett’s exploration of ‘missing’ directions, a category she acknowledges is ‘slippery’ (2018: 159). Wells’s and Gossett’s guiding principle is that readers should be able to visualize business that would be obvious to spectators in a theatre – entrances and exits, in particular, are central to both of their studies. Middleton’s and Jonson’s dramaturgy, for example, with its sometimes complex stage groupings, can be difficult to follow for first-time readers, and guidance in the form of interpolated stage directions can help. If a property such as a letter or handkerchief is dropped, but its dropping is only mentioned later in the dialogue, then a clarifying editorial stage direction makes the drama more immediately accessible to a reading audience. This decision-making process is shaped by the editor’s subjective vision of the stage (Kidnie 2000; Hodgdon 2003: 100–2). Moreover, since interpolated directions are considered intrusive where dialogue makes stage business obvious (Gossett 2018: 142), editors are often adding directions ‘at precisely those points in the text where the action is most open to (variable) interpretation’ (Kidnie 2005: 111). Is Hamlet’s opening line – ‘A little more than kin and less than kind’ – an aside? Editors sometimes cue it as one, powerfully shaping perceptions of character and situation

THE MODERN EDITING: THE TEXT

201

in the tragedy’s opening scene, but shifting cultural standards of etiquette and decorum make a publicly insolent Hamlet seem today unremarkable. Actors (who number among editors’ readers) regularly find playable, but not necessarily intuitive, performance options, challenging assumptions about what is textually self-evident. The conventional layout of the edited page, it should be noted, significantly constrains editorial practice. Anthony Hammond offers a good example of the problem, explaining that although the timing of Ferdinand’s entrance to his sister’s closet in The Duchess of Malfi is ‘not self-evident’, an interpolated direction implicitly privileges one moment over others. He describes this feature of the printed text as ‘one of the curses of editing’ (1992: 95). In their different ways, the RSC Complete Works (Bate and Rasmussen 2007) and the New Oxford Shakespeare (Taylor et al. 2016) try to build flexibility into the page layout through use of the right margin for permissive and/or open-ended editorial directions (Bourus 2018: 173–9). The goal of such experimentation with alternative layouts and editorial registers is to encourage a more interactive readerly engagement with a scene’s theatrical potential (Kidnie 2000; 2004). Speech prefixes are also part of this conversation about editorial stagings since in the earliest texts different descriptors can cue a single character. The courtier who enters to Hamlet and Horatio in the last scene of Hamlet, for example, is described in Q2 as ‘a Courtier’ (his speech prefix is ‘Cour’), until the Lord who subsequently enters describes him in dialogue as ‘Ostricke’. This character’s speech prefix is thereafter ‘Ostr’ or ‘Ostrick’. While acknowledging that such variance is potentially ‘revealing about Shakespeare’s imaginative processes’, Wells recommends the introduction of uniform character names, bar situations in which a character’s status changes over the course of the play (1984: 63–6, esp. 64). Randall McLeod weighs quite differently this trade-off between textual historicity and ease of reading (1981: 48–52, and 1991). Valerie Wayne (1998) and Michael Warren (2004) offer important commentaries on editorially mediated and emended speech prefixes, critiquing perceptions of error and querying the sexual-textual politics of normalization. The weight of the editorial tradition can be immense. Scholars have both challenged questionable instances of editorial mediation as ossifications that in some cases track back hundreds of years to Nicholas Rowe (Hodgdon 2003; McLeod 1991; Warren 1977), while elsewhere conceding that, at least in the case of Gertrude (known as ‘Gertrad’ or ‘Gertrard’ in Q2), a play’s afterlife sometimes has the power to overwrite the earliest texts (Mowat and Werstine 1992: li). Wells stresses the importance of editors grappling with textual problems for themselves by preparing new editions not from modern editions but from the earliest textual witnesses (1984: 14). This way of working enables independent reconsideration of text and forms of modernization, and forestalls inadvertent reproduction of other editors’ errors and biases. Transmission inevitably brings with it the risk of new errors, however, and accuracy is a valuable, even essential, quality in an editor (Wells 2006: 46). An increased interest in bringing twenty-first century readers into the meaningmaking process, especially as editions gravitate towards online platforms, presents opportunities to rethink the editions we prepare, and the ways we continue to shape the production and reception of early modern text.

202

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

REFERENCES Bate, Jonathan and Eric Rasmussen, eds (2007), William Shakespeare: Complete Works, New York: Modern Library. Bevington, David (2004), ‘Modern Spelling: The Hard Choices’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 143–57, Cambridge: Cambridge University Press. Bevington, David, ed. (2005), Antony and Cleopatra, updated edn, Cambridge: Cambridge University Press. Bourus, Terri (2018), ‘Editing and Directing: Mise en scène, mise en page’, in Sarah Dustagheer and Gillian Woods (eds), Stage Directions and Shakespearean Theatre, 163–87, London: Bloomsbury. Bowers, Fredson (1955), On Editing Shakespeare and the Elizabethan Dramatists, Philadelphia: University of Pennsylvania Library. Dawson, Anthony B., ed. (2016), Hamlet, in Stephen Greenblatt, Walter Cohen, Suzanne Gossett, Jean E. Howard, Katharine Eisaman Maus and Gordon McMullan (gen. eds), The Norton Shakespeare, 3rd edn, 1760–853, New York: Norton. de Grazia, Margreta (1988), ‘The Essential Shakespeare and the Material Book’, Textual Practice, 2 (1): 69–86. Dessen, Alan C. (1984), Elizabethan Stage Conventions and Modern Interpreters, Cambridge: Cambridge University Press. Dessen, Alan C. (2015), ‘Divided Shakespeare: Configuring Acts and Scenes’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 332–41, Cambridge: Cambridge University Press. Dimmock, Matthew (2015), ‘Shakespeare’s Strange Tongues: Editors and the “Foreign” Voice in Shakespearean Drama’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 342–57, Cambridge: Cambridge University Press. Giddens, Eugene (2011), How to Read a Shakespearean Play Text, Cambridge: Cambridge University Press. Gossett, Suzanne (2018), ‘When is a Stage Direction Missing?’, in Sarah Dustagheer and Gillian Woods (eds), Stage Directions and Shakespearean Theatre, 141–62, London: Bloomsbury. Graham-White, Anthony (1995), Punctuation and Its Dramatic Value in Shakespearean Drama, Newark, DE: University of Delaware Press. Greg, W. W. (1942), The Editorial Problem in Shakespeare, 3rd edn, Oxford: Clarendon Press. Greg, W. W. (1950/51), ‘The Rationale of Copy-Text’, Studies in Bibliography, 3: 19–36. Greg, W. W. (1955), The Shakespeare First Folio, Oxford: Clarendon Press. Hammond, Anthony (1992), ‘Encounters of the Third Kind in Stage-Directions in Elizabethan and Jacobean Drama’, Studies in Philology, 89: 71–99. Hibbard, G. R., ed. (1987), Hamlet, Oxford: Oxford University Press. Hirsh, James E. (1981), The Structure of Shakespearean Scenes, New Haven, CT: Yale University Press. Hirsh, James (2002), ‘Act Divisions in the Shakespeare First Folio’, Papers of the Bibliographical Society of America, 96 (2): 219–56.

THE MODERN EDITING: THE TEXT

203

Hodgdon, Barbara (2003), ‘Who is Performing (in) These Text(s)?; or, Shrew-ing Around’, in Ann Thompson and Gordon McMullan (eds), In Arden: Editing Shakespeare, 95–108, London: Thomson Learning. Jenkins, Harold, ed. (1982), Hamlet, London: Routledge. Jewkes, Wilfred T. (1958), Act Division in Elizabethan and Jacobean Plays, 1583–1616, Hamden, CT: Shoe String Press. Jowett, John (2007), Shakespeare and Text, Oxford: Oxford University Press. Jowett, John, ed. (2017a), Hamlet, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, 1. 1114–228, Oxford: Oxford University Press. Jowett, John (2017b), ‘Shakespeare, Early Modern Textual Cultures, and This Edition: An Introduction’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, 1.xxiii–xliv, Oxford: Oxford University Press. Jowett, John (2017c), ‘Shakespeare and the Kingdom of Error’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, 1. xlix–lxiii, Oxford: Oxford University Press. Kastan, David Scott (1999), Shakespeare After Theory, New York: Routledge. Kidnie, Margaret Jane (2000), ‘Text, Performance, and the Editors: Staging Shakespeare’s Drama’, Shakespeare Quarterly, 51 (4): 456–73. Kidnie, Margaret Jane (2004), ‘The Staging of Shakespeare’s Drama in Print Editions’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 158–77, Cambridge: Cambridge University Press. Kidnie, Margaret Jane (2005), ‘Where is Hamlet? Text, Performance, and Adaptation’, in Barbara Hodgdon and W. B. Worthen (eds), A Companion to Shakespeare and Performance, 101–20, Oxford: Blackwell. Kidnie, Margaret Jane (forthcoming), ‘Playhouse Markings and the Revision of Hamlet’, Shakespeare Quarterly 71:2 (2020). Kliman, Bernice W., ed. (1996), Enfolded Hamlet. Available online: http://triggs.djvu.org/ global-language.com/ENFOLDED/enfolded.html (accessed 13 November 2020). Kliman, Bernice W., ed. (2004), The Enfolded Hamlets: Parallel Texts of and {Q2}, New York: AMS Press. Leech, Clifford (1970), ‘On Editing One’s First Play’, Studies in Bibliography, 23: 61–70. Long, William B. (1985a), ‘“A bed / for woodstock”: A Warning for the Unwary’, Medieval and Renaissance Drama in England, 2: 91–118. Long, William B. (1985b), ‘Stage-Directions: A Misinterpreted Factor in Determining Textual Provenance’, Text 2: 121–37. Long, William B. (1989), ‘John a Kent and John a Cumber: An Elizabethan Playbook and Its Implications’, in W. R. Elton and William B. Long (eds), Shakespeare and Dramatic Tradition: Essays in Honor of S. F. Johnson, 125–43, Newark, DE: University of Delaware Press. Long, William B. (1999), ‘“Precious Few”: English Manuscript Playbooks’, in David Scott Kastan (ed.), A Companion to Shakespeare, 414–33, Oxford: Blackwell.

204

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Massai, Sonia (2016), ‘What’s Next in Editing Shakespeare’, in Dympna Callaghan and Suzanne Gossett (eds), Shakespeare in Our Time, 68–72, London: Bloomsbury. McKerrow, R. B. (1931), ‘The Elizabethan Printer and Dramatic Manuscripts’, The Library, 4th ser., 12 (3): 253–75. McLeod, Randall (1981), ‘UN Editing Shak-speare’, Sub-Stance, (33–4): 26–55. McLeod, Randall [i.e. Random Cloud] (1991), ‘Information on Information’, TEXT: Transactions of the Society for Textual Scholarship, 5: 241–81. McMullan, Gordon and Suzanne Gossett (2016), ‘General Textual Introduction’, in Stephen Greenblatt, Walter Cohen, Suzanne Gossett, Jean E. Howard, Katharine Eisaman Maus and Gordon McMullan (gen. eds), The Norton Shakespeare, 3rd edn, 75–92, New York: Norton. Mowat, Barbara (2001), ‘The Reproduction of Shakespeare’s Texts’, in Margreta de Grazia and Stanley Wells (eds), The Cambridge Companion to Shakespeare, 13–29, Cambridge: University of Cambridge. Mowat, Barbara and Paul Werstine, eds (1992), Hamlet, New York: Washington Square Press. Pollard, Alfred W. (1909), Shakespeare Folios and Quartos: A Study in the Bibliography of Shakespeare’s Plays, 1594–1685, London: Methuen. Purkis, James (2016), Shakespeare and Manuscript Drama: Canon, Collaboration and Text, Cambridge: Cambridge University Press. Rasmussen, Eric (1997), ‘The Revision of Scripts’, in John D. Cox and David Scott Kastan (eds), A New History of Early English Drama, 441–60, New York: Columbia University Press. Rokison, Abigail (2009), Shakespearean Verse Speaking: Text and Theatre Practice, Cambridge: Cambridge University Press. Sherman, William H. (2013), ‘Punctuation as Configuration; or, How Many Sentences Are There in Sonnet 1?’, Early Modern Literary Studies, 21. Available online: https:// extra.shu.ac.uk/emls/si-21/04-Sherman_Punctuation%20as%20Configuration.htm (accessed 13 November 2020). Taylor, Gary (1993), ‘The Structure of Performance: Act-Intervals in the London Theatres, 1576-1642’, in Gary Taylor and John Jowett, Shakespeare Reshaped 1606– 1623, 3–50, Oxford: Clarendon Press. Taylor, Gary and Terri Bourus (2016), ‘Why Read This Complete Works?’, in Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan (gen. eds), The New Oxford Shakespeare: The Complete Works: Modern Critical Edition, 45–58, Oxford: Oxford University Press. Taylor, Gary and Michael Warren, eds (1983), The Division of the Kingdoms: Shakespeare’s Two Versions of ‘King Lear’, Oxford: Clarendon Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2016), The New Oxford Shakespeare: The Complete Works: Modern Critical Edition, Oxford: Oxford University Press. Thompson, Ann and Neil Taylor, eds (2006), Hamlet, London: Thomson Learning. Tronch-Pérez, Jesús, ed. (2002), A Synoptic ‘Hamlet’: A Critical-Synoptic Edition of the Second Quarto and First Folio Texts of ‘Hamlet’, València: Universitat de València.

THE MODERN EDITING: THE TEXT

205

Urkowitz, Steven (1980), Shakespeare’s Revision of ‘King Lear’, Princeton: Princeton University Press. Warren, Michael J. (1977), ‘Repunctuation as Interpretation’, English Literary Renaissance, 7: 155–69. Warren, Michael J. (1978), ‘Quarto and Folio King Lear and the Interpretation of Albany and Edgar’, in David Bevington and Jay L. Halio (eds), Shakespeare, Pattern of Excelling Nature, 95–107, Newark, DE: University of Delaware Press. Warren, Michael (2004), ‘The Perception of Error: The Editing and the Performance of the Opening of Coriolanus’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 127–42, Cambridge: Cambridge University Press. Wayne, Valerie (1998), ‘The Sexual Politics of Textual Transmission’, in Laurie E. Maguire and Thomas L. Berger (eds), Textual Formations and Reformations, 179–210, Newark, DE: University of Delaware Press. Wells, Stanley (1984), Re-Editing Shakespeare for the Modern Reader, Oxford: Clarendon Press. Wells, Stanley (May 1991), ‘Theatricalizing Shakespeare’s Texts’, New Theatre Quarterly, 7 (26): 184–6. Wells, Stanley (2006), ‘On Being a General Editor’, Shakespeare Survey, 59: 39–48. Wells, Stanley and Gary Taylor (1979), Modernizing Shakespeare’s Spelling with Three Studies in the Text of ‘Henry V’, Oxford: Clarendon Press. Wells, Stanley and Gary Taylor (1990), ‘The Oxford Shakespeare Re-Viewed by the General Editors’, Analytical and Enumerative Bibliography, n.s., 4 (1): 6–20. Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery, eds (1986), William Shakespeare, The Complete Works, 2nd edn (2005), Oxford: Clarendon Press. Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery (1987), William Shakespeare, A Textual Companion, Oxford: Oxford University Press. Werstine, Paul (1988), ‘The Textual Mystery of Hamlet’, Shakespeare Quarterly, 39 (1): 1–26. Werstine, Paul (1990), ‘Narratives about Printed Shakespeare Texts: “Foul Papers” and “Bad Quartos”’, Shakespeare Quarterly, 41 (1): 65–86. Werstine, Paul (2012), Early Modern Playhouse Manuscripts and the Editing of Shakespeare, Cambridge: Cambridge University Press. Williams, George Walton (1971), ‘On Editing Shakespeare: Annus Mirabilis’, Medieval and Renaissance Studies, 5: 61–79. Wright, George T. (1988), Shakespeare’s Metrical Art, Oakland, CA: University of California Press.

CHAPTER 2.9

The modern editing of Shakespeare: The apparatus SUZANNE GOSSETT

The apparatus of an edition of Shakespeare is designed to shape and facilitate the reader’s experience without intruding upon his or her engagement with the text. For most readers the apparatus, while not invisible, passes without notice. Yet the apparatus is crucial on a number of fronts, not least in forming an implicit claim for the scholarly status of an edition and indicating the anticipated level and interests of its reader. The apparatus provides the framework for the kind of editing the editor is doing. Decisions about what texts to include, in what order, how much introductory material, what kinds of notes, will have a marked effect in conveying the nature of the text at hand and the intellectual work done by the editor to prepare it. While not wholly determinative of an editor’s decisions, the apparatus nonetheless limits those decisions and is therefore more significant than many users of editions of Shakespeare’s works may realize. Furthermore, this apparatus is undergoing radical change in the age of the internet. Decisions about the apparatus of a modern Shakespeare edition are where scholarship and commerce run up against each other. No matter how learned the scholarship, if the resulting edition does not sell – has no readers, is not adopted by instructors, is not used by acting companies – the effort is largely wasted. How to arrange the materials in the volume, what to annotate, when to gloss, whether to collate previous editions, what kind of introduction to include; these decisions are implicitly based on hypotheses about the readership. An error here can mean an edition will languish in the library or the bookstore; a clever decision will mean sales and the full communication of the editors’ critical and textual analysis. Some publishers have multiple series aimed at different audiences: the Cambridge School Shakespeare, the Bloomsbury New Mermaids and the Manchester Student Revels editions are intended for A level or undergraduate courses, while the New Cambridge Shakespeare, the Bloomsbury Arden Shakespeare and Arden Early Modern Drama anticipate university and scholarly readers. In the United States the Folger Shakespeare editions are primarily aimed at high school or lower-level college courses, the Norton Critical Editions at more advanced students. Traditionally,

THE MODERN EDITING: THE APPARATUS

207

American students were more likely to be directed to use a single-volume collected edition; British students to use individual play volumes. Generally speaking, the requirements and apparatus of individual volumes vary from those of the collected editions, but they do so largely in the extent of the introduction and commentary notes. For a long time collected editions looked very similar. From the First Folio through the many eighteenth-century editions published successively by the house of Jacob Tonson, and well into the twentieth century, all collected editions separated the plays into genres, although there were minor tweaks to the Folio arrangements: Alexander Pope (1725) began the list of histories with King Lear, and Edmond Malone (1790) began it with Macbeth.1 The poems, absent from the First Folio, were placed in a separate section. Most twentieth-century American single-volume Shakespeares, for example the many successive editions edited by David Bevington (‘the Bevington’, seventh edition, 2014) and the first (1974) and second (1997) Riverside Shakespeare edited by G. Blakemore Evans (‘the Riverside’), commenced with a lengthy general introduction and inserted shorter critical essays before each play. In a typical arrangement, the Bevington added appendices on ‘Canon, Dates and Early Texts’, ‘Sources’ and ‘Shakespeare in Performance’ as well as abbreviated ‘Textual Notes’. The differences in apparatus between these college-level teaching editions were small. For instance, where the Riverside does not separate glosses from other line-by-line commentary, the Bevington has commentary on the page and a glossary for the entire volume at the back, while the Norton Shakespeare has glosses in the margin and commentary notes at the foot of the page. All of these editions, whether single or multiple volume, worked within an intellectual framework that did not vary radically from that of Shakespeare’s fellow actors John Heminges and Henry Condell, who claimed to be presenting ‘the Author’ and his ‘owne writings … absolute in their numbers, as he conceiued them’ (Shakespeare 1623: sig. πA3r). These editions seemed to presume that the text presented was as close as possible to the author’s intentions (either original or final), that the editor was authoritative and knew more than the reader and implicitly that the reader, like the editor and like Shakespeare himself, was white, Christian, English and male. Since 1986, and the appearance of the Oxford Shakespeare (general editors, Stanley Wells and Gary Taylor), publishers and editors have been more adventurous, trying, it seems, to find a formula that will reflect modern scholarly ideas about Shakespeare and text while also appealing to a variety of buyers. This chapter will trace some of the changes and then look at the ways in which the advent of the internet has allowed certain poststructural ideas to manifest themselves in recent editions.

ORGANIZATION AND ARRANGEMENT One striking change initiated by the Oxford text, and consequently present in the Norton Shakespeare, which for its first two editions used the Oxford text, was the presentation of the materials in chronological order. Although the Oxford Shakespeare arrived at the high moment of critical theory preaching ‘the death of

208

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

the author’, such a presentation still focused attention on the creative life of William Shakespeare the individual. What seemed radical then has become less surprising thirty years later. Strikingly, although some one-volume Shakespeares have retained the division by genre, the chronological arrangement has been adopted not only by Oxford and Norton but by both the Middleton Collected Works (Taylor and Lavagnino 2007) and the Cambridge Edition of the Works of Ben Jonson (Bevington, Butler and Donaldson 2012). Almost any arrangement, any apparatus, will have built-in difficulties. For the generic division of Shakespeare’s plays these began with the First Folio, which put Cymbeline among the tragedies and printed Troilus and Cressida between the histories and tragedies without listing it in the catalogue (the table of contents), probably because rights to the text were obtained late (see Blayney 1991: 17). Implicit disagreements about generic divisions persist: for example, the Bevington and Riverside, as well as the Norton Shakespeare generic volumes, include Troilus among the comedies, but the RSC Shakespeare places it among the tragedies. Similarly, the correct chronology continues to be contested, with scholarship on the order of Shakespeare’s plays going back at least to Malone’s ‘Attempt to ascertain the order in which the plays attributed to Shakespeare were written’ (1778: 1.269–346). Uncertainty is physically embodied in the changes between the Oxford Shakespeare (Wells and Taylor 1986) and its successor, The New Oxford Shakespeare (Taylor et  al. 2016; 2017), when the latter revises some of the order of the former, for instance postdating Hamlet to follow Troilus and Cressida and bringing back All’s Well That Ends Well to follow Measure for Measure rather than Antony and Cleopatra. Furthermore, both The New Oxford Shakespeare and the third edition of the Norton Shakespeare (the Norton 3, Greenblatt 2015a) metaphorically throw up their hands about the chronology of the individual poems, instead printing each collection at the date of the publication rather than trying to decide when an individual sonnet was written.2 Oddly enough, what may be the most convenient arrangement for readers, alphabetical order by title, is used only in the Arden Shakespeare Complete Works (Proudfoot, Thompson and Kastan 2011), although the Oxford Shakespeare and the digital edition of the Norton 3, like the Middleton Collected Works and the sevenvolume Cambridge Edition of the Works of Ben Jonson, all implicitly acknowledge the usefulness of such a list by including one alongside the regular table of contents. More fundamental questions remain before the editor of a large volume can begin to decide on elements of the apparatus. Is this a ‘Complete Works’, or, as in the Bedford Shakespeare, ‘the twenty-five most frequently taught plays’ (McDonald and Orlin 2015: ix) or, as in the Norton 3 Essential Plays, ‘twenty-one of the plays that are most often read and produced’ (Greenblatt 2016: xvii)? Are the Sonnets in or out? Such decisions no doubt reflect market research, both about the desire for less voluminous books to place in student backpacks and about which plays to include. The slimmer volumes may increase sales for the publisher, but they will affect the editor in practical ways: for example, in a Bedford introduction to the second tetralogy the editor can write about the banishment of Falstaff, but she is restrained from doing so extensively in the Norton Essential because it includes only the first part of Henry IV.

THE MODERN EDITING: THE APPARATUS

209

These reduced volumes, aimed no doubt at the classroom, move in opposition to current scholarship, which has worked to expand the Shakespeare canon, finding his hand not only in The Two Noble Kinsmen but also in such anonymous plays as Edward III, Sir Thomas More and Arden of Faversham, and in the additions to Thomas Kyd’s Spanish Tragedy. All or parts of these works appear in their chronological placement in The New Oxford Shakespeare, but historically they were more likely to be put in a supplementary volume or at the back. Malone, who did not think a line of Titus Andronicus was written by Shakespeare, banished it to his final volume, and the Norton 3 (which does not include Arden of Faversham or the additions to Spanish Tragedy) puts complete texts of King Edward III and Sir Thomas More in the electronic edition only. Also working against slim volumes has been the explosion of textual scholarship on, and critical attitudes towards, variant texts. The most radical innovation of the 1986 Oxford Shakespeare was printing two texts of King Lear, based on its acceptance of scholarship suggesting that one was a revision of the other, rather than each deriving imperfectly from a single original (see the essays in Taylor and Warren (1983)). It also used the Folio as the basis for its text of Hamlet, but printed additional passages from Q2 as a supplement. When Norton took over the Oxford text, it not only printed both texts of King Lear but also commissioned a conflated text, believing that many instructors accustomed to a single King Lear would desire such a text for the classroom. This was a traditional conflation, in which the editor, Barbara Lewalski, judged each variant, whether a single word, a line, or a passage, against her sense of what Shakespeare must originally have written or, possibly unconsciously, against a sense of what would be ‘better’. The tendency to swell the collected editions by including variant texts has only grown. The New Oxford Shakespeare prints only one text of Hamlet and of King Lear, Hamlet from the second quarto and King Lear from the first quarto, but promises that edited texts of Hamlet from the first quarto and the Folio and King Lear from the Folio will appear in the Alternative Versions (Taylor, Bourus and Egan forthcoming). Meanwhile the Norton 3, unshackled from the Oxford text, avoids traditional conflations but prints for Hamlet the second quarto text ‘augmented by passages from F of a line or more’ (Greenblatt 2015a: 1762) as well as the first quarto, and for Lear quarto, Folio and the Folio with additions from the quarto. The Norton digital edition includes four texts for Hamlet and three for King Lear. It seems unlikely that such multiple-text editions can continue to be printed, and as we shall see this is one place where the availability of internet resources may solve the publisher’s practical problem while inviting a different theoretical understanding of the text. Once major decisions about contents are made, frequently by the publisher, who may have run focus groups hoping to discover what teaching faculty desire, it is the editor who will determine the internal apparatus. He or she may, of course, voice a scholarly opinion about which plays to include, or how to treat collaborations, but the editor’s hand will be most obvious in the introduction, commentary notes and glosses, textual data and appendices. These decisions affect editing both in individual volumes and in collected editions.

210

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

INTRODUCTIONS AND APPENDICES For a very long time there were two approaches to introducing each Shakespeare play or the poems included in a collected works: either no introduction at all, leaving interpretation entirely up to the reader, or a substantial critical reading. The latter was more frequent in American editions than in, for example, the familiar Globe (1864, with many reprints) and Peter Alexander (1951) editions in Britain. The brief, one-page introductions in the 1986 Oxford Shakespeare may have been partly responsible for the failure of that edition to sell for US classroom use; the presence of a lengthy general introduction and ample interpretations of each play by notable scholars helped make the Norton edition of the same text a success. Where the standard introduction once offered a predictable passage through set topics (e.g. text, date, source, relation to other Shakespeare plays), newer volumes are less uniform. Furthermore, over time material has moved back and forth between the introduction and appendices. Some of the earlier Arden Shakespeares discuss a play’s source(s) in detail, with lengthy quotations, within the introduction, while others print the entire source or sources in an appendix. Appendices in the New Cambridge and the third generation of the Arden Shakespeare (Arden 3) vary, depending partly upon the editorial and critical issues presented by a particular text but also upon the individual editor’s interests. Thus, there are Appendices on ‘Sources’ and on ‘Appropriations’ in the Arden 3 Tempest (Vaughan and Vaughan 1999), on the ‘Authorship Debate’ in the Arden 3 All’s Well That Ends Well (Gossett and Wilcox 2019), on ‘Rhyme’ in the Arden 3 Romeo and Juliet (Weis 2012), on ‘Type Shortages’ in the Arden 3 Merchant of Venice (Drakakis 2010), on ‘Relineation’ in the New Cambridge Macbeth (Braunmuller 1997) and on ‘Music’ in the New Cambridge Cymbeline (Butler 2005) and in the Arden 3 Henry VIII (McMullan 2000). Occasionally a variant quarto or a source is printed in facsimile. In recent years Arden Shakespeares include casting charts, showing possible doublings, and in The New Oxford Shakespeare Critical Reference edition (Taylor et al. 2017) each play is preceded by a list of ‘Roles and Requirements’ (for stage properties and sound effects). A snapshot of methods of Shakespeare scholarship and assumptions about what buyers and readers might expect in a single volume can be found in the introductions to successive editions of Pericles over nearly a century, from 1907 to 2003.3 These begin with the first Arden, edited by Kenneth Deighton. Deighton discusses the text and the authorship controversy. He proposes no general interpretation but makes a few critical comments in connection with the question of authorship: for example he argues that Shakespeare’s hand can be found in the brothel scenes and that Lysimachus, like the Duke in Measure for Measure, is in disguise to ‘probe the plague-sores of the city of which he was governor’ (Deighton 1907: xxi). The Cambridge edition, edited by J. C. Maxwell in 1956, is similarly restrained on interpretation. After sections on ‘The Story’ (x–xii) and ‘The Problem of Authorship’ (xii–xxv), Maxwell’s brief pages on ‘What Shakespeare Made of it’ (xxv–xxix) discuss the play’s similarity to the other ‘romances’, the characterlessness of the central figure and the comedy of the brothel scenes. However, Maxwell does include a substantial section on the stage history. A few years later, F. D. Hoeniger in the

THE MODERN EDITING: THE APPARATUS

211

Arden 2 edition (1963), adapts what seem to have become the conventional topics of the series, taking up in turn sources, text, authorship, date, stage history and, finally, literary interpretation.4 In the latter section Hoeniger examines such topics as similarities between Pericles and the last plays. His interpretation is heavily tinged with spirituality, comparing Pericles to miracle plays and emphasizing the theme of patience and redemption. A considerable change is visible in Philip Edwards’s New Penguin Pericles of 1976. The ‘account of the text’ is removed to the back of the volume, and Edwards discusses the much-vexed question of the authorship of Pericles only at the end of his introduction. Instead, he writes a lengthy critical and interpretive essay, which among other things defends the drama as ‘a work of remarkable beauty and power’ (1976: 8). (Such defences of the play one edits have become the norm and are sometimes called for in the general guidelines to a series.) Edwards believes that the contest between Marina and Lysimachus is ‘the hub of the whole play’ (21), and he emphasizes the role of family throughout. In another striking change, Edwards uses the introduction for critical controversy, rejecting the interpretation of another editor, Hoeniger, by name. Instead, he asserts that in Pericles ‘there is no pattern, either of providence or redemption’ (30). Such open confrontation seems to have become more frequent in recent volumes. For example, the New Cambridge edition by Doreen DelVecchio and Antony Hammond (1998), although it contains the usual sections on date, sources, authorship, performance and ‘the play’, is chiefly concerned to offer a provocative and unconvincing attempt to argue that the play is entirely by Shakespeare and that its variations in style are intentional. Furthermore, in a less nuanced religious reading than that offered by Hoeniger, DelVecchio and Hammond end their interpretation with a section on resurrection and read the play, straightforwardly, as ‘a metonymy of the greatest Christian mystery’ (1998: 75). A different kind of controversy is implicit in Roger Warren’s 2003 edition in the Oxford Shakespeare multi-volume series (not to be confused with the 1986 Oxford Shakespeare Complete Works). The edition is based on the conjectural reconstruction of Pericles by Gary Taylor and MacDonald P. Jackson in the 1986 Oxford Shakespeare, and Warren defends their choices in his sections on text and authorship. Nevertheless, Warren foregrounds performance: the introduction contains a substantial section on the ‘modern revival’ of the play, and descriptions of performance moments are scattered throughout the commentary notes.

COMMENTARY, GLOSSES AND TEXTUAL NOTES The commentary may be the most difficult thing for an editor to judge. First of all, knowing that many readers will skip the introduction, the editor must tread a fine line between repeating information included in the introduction or not supplying it where it would be most immediately pertinent. The problem is frequently addressed by writing a short topical note in the commentary and adding a cross reference to fuller discussion in the introduction. Of course there is no way to know if the cross reference is followed up, that is, if the problem is solved.

212

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

The length and density of commentary notes varies greatly, depending partly on the publisher’s criteria and partly on the editor’s sense of the information required to clarify the text. One extreme view was articulated by F. L. Lucas editing the plays of John Webster: he explained that his edition had to serve two masters, ‘the general reader and the learned’, and this concern ‘has governed the arrangement of this work as a whole. For the general reader wants, above all, a clear text that he can read not only with confidence but with pleasure; and accordingly the notes have been banished from the foot of the page, where they are always hideous and often distracting, to the end of the book’ (1927: 1, viii). On the contrary, many would argue, a Shakespeare or Webster ‘clear text’ without helpful commentary notes may be not a pleasure but an unsolvable mystery, driving students to Cliff Notes or crib sheets and encouraging productions, like those promised at the Ashland Oregon Shakespeare festival, where the plays are ‘translated’ into current English. On the other hand, it is notable that where in the original and second Arden series, the textual notes came just below the text, with the commentary notes beneath, in the Arden 3 series the reader can look at a commentary note without encountering the compressed and intimidating formatting of the textual notes. Though these are still present, they are banished to the bottom of the page.5 In editions like the Norton 3, glosses, that is, short definitions of early modern English words, are separated from the longer commentary notes. Such verbal glosses present a greater challenge since Shakespeare has ‘gone global’, especially in the difference between British and current North American diction. A common British usage, like ‘remit’ for ‘the sphere of responsibility specified for a particular role’, means nothing relevant to a US student. There is also the question of whether a word needs repeatedly to be glossed: if the definition of ‘knave’ or ‘undone’ is explicated every time it appears, readers may become annoyed by the repeated interruption of their pleasurable reading, but they may be equally annoyed to find a note that says ‘cf. 4.3.271’, requiring them to hunt through their books in a way more disruptive than merely being reminded of the term’s meaning. Nevertheless, assistance with Shakespeare’s language is among the most valuable offerings of a good Shakespeare edition, alerting readers to changes in meaning that even an experienced instructor may not realize must be explained. I have had honours students assume that when Lysander complains that Demetrius ‘made love to Nedar’s daughter, Helena’ (A Midsummer Night’s Dream, 1.1.107), the young people have been to bed together, and that when Orsino assures Viola that once in ‘other habits’ she will be ‘Orsino’s mistress’ (Twelfth Night, 5.1.380–1), he does not intend to marry her. The commentary notes and the introduction are the sections of an edition that most vividly reflect changing times and attitudes, though these may affect even the glosses. For example, older Shakespeare editions, such as that of W. A. Neilson and Charles Hill (1942), widely used in the United States in the middle of the twentieth century, or the older Ardens widely used in Great Britain at the same period, either simply did not gloss such words as ‘prick’ or the sexual jokes of Falstaff and Thersites, or had, at most, a note pointing out ‘a bawdy quibble’ but not explaining it. So to Malvolio’s line as he finds the fake letter – ‘These be her very c’s, her u’s, and her t’s, and thus makes she her great P’s’ (2.5.86–7) – Charles T. Prouty (1958), who edited

THE MODERN EDITING: THE APPARATUS

213

Twelfth Night for the Pelican Shakespeare (general editor, Alfred Harbage), has no note at all, whereas Jonathan Crewe, who edited the play for the Complete Pelican Shakespeare (general editors, Stephen Orgel and A. R. Braunmuller) almost half a century later, has two notes, first to ‘C’s … T’s’ explaining that ‘“cut” is Elizabeth slang for women’s genitals’ and then to ‘great P’s’, with two glosses: ‘Capital P’s; also pun on her urination’ (2002: 458). The RSC Shakespeare is even more direct, noting that ‘“cut” was slang for “vagina,” but the joke may well be on “cunt”’ (Bate and Rasmussen 2007: 670). This is, however, one place where care and taste are required. To ignore all the jokes on ‘prick’ from Shakespeare and simply say the word equals ‘penis’ is to eliminate the joy of the hidden pun. But when the language of the glosses is too vulgar – e.g. glossing the conversation between the Ward and Sordido in Women Beware Women, ‘stoop = fuck’, ‘hole = cunt’, ‘pitch out = ejaculate’ – some students may be offended and miss the slippage between meanings.6 In writing commentary notes editors are constantly making implicit judgments about their readers’ cultural knowledge. After decades of teaching Shakespeare in the United States at a denominational university, I can attest that American students have little knowledge of the Bible and effectively none of English history. British colleagues report that their students are not much better informed. This means that an edition must supply notes where, perhaps, they were once considered unnecessary for the anticipated reader. Earlier editions, if they noted a biblical citation, often gave merely chapter and verse, but modern editors do well to include the actual words. Similarly, when Henry V prays the night before Agincourt ‘Not today, O Lord, / O not today, think not upon the fault / My father made in compassing the crown’ (Henry V, 4.1.289–91) the older Penguin has no note at all, but The New Oxford Shakespeare explains the ‘fault’ as ‘the deposition and murder of Richard II’ (Taylor et al. 2016: 1581) and the American Norton 3 explains that ‘Henry’s father usurped the throne from its rightful possessor, Richard II’ (Greenblatt 2015a: 1588). A further reason for not assuming knowledge in this case is that the required Shakespeare course is, in the United States at this writing, an endangered species, and students may be reading this play in isolation from the rest of the tetralogy. If they are not supplied with the necessary background, they tend to interpret Henry V as a story exclusively about military success, without understanding the political shadings, and Henry IV as merely a tale of adolescent angst and rebellion. Notes also may reflect current political attitudes – or offend students if they don’t. Writing about interventions within the text such as added stage directions, Margaret Jane Kidnie examines practices and assumptions that underpin an editor’s shaping of an early modern play for contemporary readers (see her chapter in this volume). Such shaping is especially true in the commentary. For example, Ann Thompson, editing The Taming of the Shrew for the New Cambridge Shakespeare in the early 1980s, explicitly chose to work against traditional interpretations of that play by writing ‘feminist’ notes. Hence in commentary to the first scene between Petruchio and Katherina, she points out that Petruchio ‘“feeds” Katherina several good lines’ and ‘seems to relish her performance as much as his own’ (1984: 87). Later, when he asserts that she is ‘not hot, but temperate as the morn’ (2.1.297), Thompson comments that ‘Petruchio is probably not making a serious question of her virginity,

214

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

though he is not the only Shakespearean hero to assume that a woman with spirit must be unchaste’ (92). A similar intention seems to lie behind the Norton 3 warning about Katherina’s long final speech, ‘Though Katherina’s final speech may suggest she has been tamed, many actors have delivered it in ways that undercut or ironize its meaning’ (Greenblatt 2015a: 414). Early modern racial attitudes can be even more problematic than sexual ones for contemporary readers and hence for editors. In older editions Portia’s reaction when Morocco fails the text of the caskets and departs – ‘A gentle riddance. … Let all of his complexion choose me so’ (The Merchant of Venice, 2.7.78–9) – usually passes without comment, but the Norton 3 addresses the embarrassing speech directly in a ‘performance comment’: ‘Portia’s final line is often cut from productions so as not to compromise her standing with audiences. Rescuing the play from its objectionable speeches is common in Merchant and can affect the story and its generic balance considerably’ (Greenblatt 2015a: 1360). Othello presents even greater challenges: the New Pelican carefully glosses Roderigo’s complaint about the good fortune of ‘the thick-lips’ as a ‘racist slur on Othello’s African heritage’ (Orgel and Braunmuller 2002: 1403). Treatment of Jews throughout The Merchant of Venice causes modern discomfort: The New Oxford Shakespeare has a marginal note to ‘Hath not a Jew eyes?’ (3.1.53) pointing out that ‘Shylock’s exposition on the shared nature of human experience … is an important moment in performance: it may visually elicit empathy or mockery from the listeners, or they may simply ignore him’ (Taylor et al. 2016: 1240). But even a casual line like Benedick’s promise that ‘If I do not love her I am a Jew’ (Much Ado About Nothing, 2.3.253), presented without comment in the second Riverside, is glossed in the Norton 3, ‘That is, lacking in Christian charity (an antiSemitic stereotype)’ (Greenblatt 2015a: 1428). The Bedford Shakespeare handles such matters in full page, sometimes illustrated, ‘Marginal Asides’. For example, Petruccio’s line, ‘She is my goods, my chattels’ (The Taming of the Shrew, 3.2.222) elicits a discussion of early modern patriarchy, and Shylock’s claim that ‘Suff’rance is the badge of all our tribe’ (The Merchant of Venice, 1.3.103) is glossed with a variety of statements on English attitudes towards the Jews in Shakespeare’s period. These insertions, whether in notes, asides or the introduction, reveal an editor’s sense that current readers will be uncomfortable with many elements of Shakespeare’s world, and that only with considerable explication can he become ‘for all time’. The Performance Comments in both the New Oxford Critical Edition and the Norton 3 raise the question of how much or how often commentary should record particular stagings. On the one hand, actors often assert that scripts serve them best without any stage directions, even the original ones; they may also prefer not to be told how a moment has traditionally been performed. Yet such notes can be useful in the classroom, where they can open up the action to students unused to imagining words translated into performance. While recording the way a scene has been played in a specific production has limited interest, discussion of conflicting stagings can be enlightening, and both the Norton 3 and the New Oxford use performance notes in this way. Of course, the individual volumes in the Arden or New Cambridge Shakespeare or Oxford Shakespeare editions have room for more extensive notes than a

THE MODERN EDITING: THE APPARATUS

215

collected edition, but in all cases editors must decide how interested their readers are in specific topics. For example, it has been traditional to record places where the OED cites a Shakespearean line for the first usage of a word. Given that the original OED’s limitations in this area are now recognized, should any such note indicate whether the information is from the unrevised or revised OED? Or is this kind of note just a linguistic offshoot of the ‘Shakespeare as a great man’ theory and not of interest to current readers? Similarly, should the notes indicate pronunciation, for example stressed or elided syllables? The assumption of such notes has been that the pronunciation is the received standard British English, but this too has been contested. Sometimes an older pronunciation, like perséver rather than persevere, is required for the rhyme. Should this be noted? Commentary notes are occasionally used, especially in the more expansive single volumes, to discuss or justify editorial decisions, such as whether to call the heroine of All’s Well That Ends Well Helena or Helen, whether to retain the Folio’s generic designations for characters (e.g. Clown) or to use specific names (e.g. Touchstone), whether Hamlet rejects his ‘sallied’, ‘solid’ or ‘sullied’ flesh, or where to insert the stage direction indicating that Romeo first kisses Juliet. Missing and supplied speech prefixes, resolution of a verbal crux, ambiguous phraseology, the choice of spelling for a word with two possible meanings (metal/mettle) are all of great interest to the editor who has struggled with them, but does Lucas’s ‘general reader’ want to interrupt the flow of the play to learn about them? The record of such decisions is usually buried in the ‘collation’ or ‘textual’ notes, which normally document, at a minimum, any place where the editor has altered the copy or base text. Full collation notes will instead record all or many of the decisions of editors through the centuries, starting with second quartos for some plays and the Second Folio for many others. Historically such notes have been found in individual volume editions such as those in the Variorum series rather than in the collected editions, partly due to constraints of space. In the single volume collections there is sometimes a compromise, with attributions made to a collective ‘Edd’ rather than to individually identified editors. The revival of interest in textual studies has led to a variety of methods for sharing the extensive textual knowledge that any editor of Shakespeare will acquire. The Riverside Shakespeare, an edition with an early interest in conveying textual information to students, includes after each play a ‘Note on the Text’ and a full collation. The RSC Shakespeare similarly includes rather briefer textual notes after each play, and the Complete Pelican Shakespeare has such notes before each play. The Bevington Shakespeare, instead, places the textual notes in an appendix. But the 1986 Oxford Shakespeare, appearing just at the beginning of the internet age, pioneered a completely new arrangement, offering the kind of clear page that Lucas imagined his general reader to desire, while placing extensive textual introductions and notes to each text in A Textual Companion (1987). The separation has only grown in the intervening years. The New Oxford Shakespeare includes modernized texts in the Modern Critical Edition and puts all the textual notes with the original-spelling texts in the two-volume Critical Reference Edition.

216

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

THE ELECTRONIC CHALLENGE IN PRACTICE AND THEORY It seems unlikely that many publishers will have the resources expended in producing the multiple volumes of The New Oxford Shakespeare – four at this writing, with two more promised – so it seems destined to be a scholar’s or a library’s reference work. Instead, in the age of the internet, publishers have turned to digital solutions. And here we encounter the major questions facing twenty-first-century editors and publishers of Shakespeare editions after several decades of tentative experimentation. The first issues are practical: What should an electronic edition look like? What should it offer? How, if at all, does the nature of the apparatus change between a print and an electronic edition? If there are both print and electronic versions, how should material be divided between them? How can or should electronic editions be supported? The more profound issues are theoretical and confront editors with the question of how to integrate poststructuralist challenges to the stability of the text, the ‘death of the author’, and the singular agency of the editor into the apparatus of their editions. Because, with the exception of the Internet Shakespeare Editions, to be discussed below, publishers have normally begun by creating digital additions to their print editions, the first practical question they face concerns the division of the material, both primary and secondary. As of 2020, several major editions have moved towards using the availability of a website to split the riches they offer. For example, where the Norton 1 and Norton 2 (Greenblatt 1997; 2008) included a general appendix on ‘Textual Variants’, the Norton 3 (Greenblatt 2015a), based on a freshly re-edited text, places fuller textual comments and Textual Variants for each text only in its Digital Edition (Greenblatt 2015b). The Norton 3 also offers newly edited versions of alternative texts, as well as Sir Thomas More and Edward III, only in the digital edition. The New Oxford Shakespeare similarly promises alternative versions and additions such as the entire Spanish Tragedy on its eventual website. The Bedford Shakespeare includes ‘Textual Variants’ along with such supplementary materials as suggestions for further reading on a companion website. It is not clear, however, how many readers will choose to access these materials. This is especially true for the scholarly materials. Textual notes become even more invisible if banished to the ether, where a print reader must actively choose to encounter them, rather than being placed at the back of a volume where a reader leafing through may happen upon them. More elaborate secondary material may also be isolated in this way. To take a case in point: the seven-volume Cambridge Works of Ben Jonson has placed on a website essays on the textual history of each work as well as essays on the performance history of each play or masque. This seems to reflect a prejudgement that such items will be of lesser interest to most users of the edition. Yet reception history is a vibrant field, one that the edition may be seen as implicitly devaluing, and the edition itself has stimulated interest in textual analysis. Both The New Oxford Shakespeare and the Norton 3 have put the material found in their print volumes also on the web; clearly other publishers intend to do the like.

THE MODERN EDITING: THE APPARATUS

217

It is uncertain which of the experimental formats appearing in the first decades of the twenty-first century will become the standard.7 Presumably the arrangements that readers, especially students, prefer, will prevail, but issues of cost in developing electronic platforms will play a large and unpredictable part in these decisions. An immediate question is what the apparatus of a Shakespeare edition will look like with full digital resources. The Norton 3, for example, allows students to turn commentary notes and glosses on and off, and if they are turned on, to click on any word with a dotted line beneath it and receive the relevant information in a popup box. The texts have one-click arrows that will cause certain passages to be read aloud, other arrows that will cause all the songs to be played, and yet further symbols that will take the reader directly from the edited text to a facsimile of the quarto or folio page which forms its base. On the other hand, the textual editors’ desire that all variant versions could be streamed side by side turned out not to be possible on the platform chosen by Norton; it offers instead samples of parallel scenes. To those accustomed to the limitations of print, texts like these seem almost miraculous. However, if students expect that they can obtain similar texts with similar functionality online for free, will they buy a scholarly edition from a commercial publisher? The Internet Shakespeare Editions (Best 1996–) apparently provides the most immediate competition. It is freely available on the web and promises ‘edited texts alongside facsimiles of the original printed documents and searchable diplomatic transcripts of those documents, available for comparison at a mouse click’. As in the Norton 3, with one click the reader may choose whether to show or hide annotations and textual variants. The user can print out just an actor’s part, with cues, or navigate to a facsimile. ‘Textual analysis tools’ will generate a variety of statistics, such as the number of stage directions or the average speech length in a scene. However, as of this writing, only a few of the ISE texts offer all of these potential resources. The project, which was created by Michael Best of the University of Victoria through an extraordinary level of will power and sheer hard work, is dependent on the labour of scholars either offered freely or subsidized by university positions or outside grants. Thus, it can make progress only as such support is or becomes available. Similarly, the digital interface for the Electronic New Variorum Shakespeare project, developed by Alan Galey, was a product of his scholarly time and effort, along with funding by the Social Sciences and Humanities Research Council of Canada.8 Yet, despite their constraints, these editions offer a financial challenge to publishers like Oxford, Cambridge, Norton or Bloomsbury, and it is hard to predict how much the availability of such free resources will impinge on the market for Shakespeare.9 In the past decade American instructors’ orders for large, single-volume editions of the Bard have decreased considerably. While the decline in orders may be partly due to the reduction in the numbers of English majors and the frequent abolition of a required Shakespeare course, it is also the case that the rising generation of students is accustomed to reading onscreen and, more important, accustomed to not paying for texts they find there. The situation for publishers will be even worse if students turn not to scholarly texts like the Internet Shakespeare Editions but to some of the free digital texts already available, which are ‘selected with no reference to scholarly criteria for textual accuracy, and keyed

218

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

in by volunteers concerned more with sharing out-of-copyright materials than with establishing their reliability’ (Galey 2014: 258). A good-faith misunderstanding among those who worked on the Norton 3 serves as an indicator of how electronic resources may challenge traditional assumptions of publishers and purchasers. The Norton 3 was conceived as a digital edition with the print volume as a partial spin-off. A great deal of original scholarly work, including fully edited texts, is found only in the digital edition. Each volume of the Norton 3 (whether complete or a genre volume, e.g. ‘Tragedies’) brings with it an electronic key giving access to the entirety of the edition’s digital resources. However, there was dismay among the editors and their students early on, when it emerged that W.W. Norton thought of the Norton 3 as a textbook, and the norm for textbooks is to give the purchaser digital access for one year only. Neither instructors nor students – who instead thought of the edition as a continuing resource – could imagine buying a scholarly product, whether bound in covers or opened on a website, whose material would evaporate after a year. They had all assumed that the electronic material would, like their books, be a permanent acquisition. In this case the publisher yielded the point. Such contretemps are sure to occur again as editions find their way from print to pixels. The theoretical implications of moving from print to the web are addressed directly by the Internet Shakespeare Editions when they assert that their editions strive ‘toward the goal, not of a definitive, authoritative play text, but of the “postmodern edition” … presenting plural texts that reject “totalizations of all kinds”’. Thus an Internet Shakespeare Edition offers the reader even more choices than does the Norton 3. Here, if readers choose to show textual variants, they have a choice of whether to access them in a pop-up box or within the line. If within the line, the variants appear in different colours, with their sources (e.g. F4, Capell, Taylor) included, as here at the most famous crux of All’s Well That Ends Well (the variant colours are here reproduced in grayscale): I see that men make ropes in such a scarxxx rope’s in such a scarre, F1make Hopes in such Affairs Rowe make hopes in such a scene Malone may cope’s in such a stir Tannenbaum may rope’s in such a snare Sisson Such an apparatus is potentially very attractive to someone interested in the history of Shakespeare’s text. It is also distracting, and visually challenges any concept of the ‘stability of the text’. The author becomes more obviously a creation of his previous editors, while the apparatus undermines the presumed authority of the current editor as readers are given agency to choose which of the solutions they wish to insert in ‘their’ edition of ‘Shakespeare’. A similar transfer of authority is gradually happening in editorial introductions. Throughout the twentieth century the differences between the critical introductions in collected and individual editions were primarily caused by different limitations on length. However, some collected editions have begun to vary the formula, apparently distrusting a ‘topdown’ interpretation predetermining the reader’s experience. Instead, in postmodern fashion, either by removing a preparatory critical interpretation or by presenting an assortment of conflicting views, these editions

THE MODERN EDITING: THE APPARATUS

219

encourage readers to form their own interpretations. Thus the Bedford Shakespeare opens each play with a ‘Preview’ giving readers ‘just what they need to know to enter the world of the play’ and an ‘Action’ that provides a ‘scene-by-scene summary’. Each play text is only then followed by a ‘view’ or extended analysis that also includes a section on the ‘Afterlife’ or performance history of the play. The editors of The New Oxford Shakespeare, rather than impose their ‘own favourite theories, methods or interpretations’, present a ‘tapas Shakespeare’ (Taylor et al. 2016: iv). In practice this means a ‘bricolage’ of quotations from many different authors, some going back as far as Addison and Dryden, others from present-day scholars. Among the single play editions the individual Folger Shakespeare volumes also seem to be moving this way. Their updated editions begin with a brief introduction to the play at hand and defer a single critical essay, significantly identified as ‘A Modern Perspective’, to the end of the volume. All of these arrangements indicate a wariness about providing students with ‘answers’ before they have read the Shakespeare text; instead they begin to pass interpretive agency from the editor to the reader. The electronic version of this approach will give readers even more control. The Norton 3 has a facility allowing students to write their own notes and eventually to assemble these notes into an essay. Here a key apparatus of the play, the commentary notes, is shared by the reader and the editor in what might be thought of as a ‘Wiki’. Taken together with the student’s adjusted text, these notes might create, for example, a post-colonial or feminist or queer Tempest, inserting illustrations and supporting scholarly criticism at will. Thus, after nearly a century of competition among recognizable categories of Shakespeare editions that implicitly descend from those of the early seventeenth century – the hardbound collected works recalling the Folio, the more or less scholarly single volumes recalling the little quartos, some of them ‘augmented and corrected’ although without notes – publishers and scholars are trying to find the best ways to bring the works of a late sixteenth-century author who wrote with a quill into the world of screens and interlocked websites, PowerPoint lectures and screen shots. The intellectual openness of the postmodern climate proposes that presentations of Shakespeare’s works accept the equal relevance of topics from globalization, race and feminism, to stylistics and textual analysis, yet editors can no longer assume that students and readers come from a narrowly focused Anglophone background and do not require notes, glosses and introductions to an unfamiliar world. Similarly, the traditional apparatus of the Shakespeare edition, whether meant for the classroom or armchair reading, is in the process of changing. The twenty-first century editions are taking on this challenge. No one knows what the final shape will be.

NOTES 1. In the 1821 Variorum edition James Boswell attempted to ‘carry out Malone’s plan to organize the plays in chronological order’ but yielded to objections that ‘the chronological succession of events in the ten Folio histories was … indispensable’. Thus ‘the order in which Shakespeare worked was disrupted by the order of the historical events on which he worked’ (Hooks 2016: 360–1).

220

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

2. Exceptionally, the Norton 3 prints The Passionate Pilgrim ‘according to the date by which Shakespeare is likely to have composed the five poems in it that are his’ (Greenblatt 2016: 871). Thus in the Norton 3 the collection appears after Love’s Labour’s Lost and before Richard II, while in the New Oxford Modern Critical Edition it appears after Much Ado and before Henry V. 3. See Marcus (2015) for analysis of introductions to The Tempest, and Levenson (2015) for more general discussion. 4. For comparison, see the introduction to the Arden 2 Othello, edited by M. R. Ridley (1958), which has sections on date, sources, text, the play and the characters, and the ‘double time scheme’, or the introduction to the Arden 2 Macbeth, edited by Kenneth Muir (1951), which has sections on text, date, interpolations, sources, stage history and ‘the Play’. 5. Howard Staunton had introduced the three-layer page, text, collation and commentary, in his Shakespeare editions of the mid-nineteenth century. Alan Galey (2014: 102) suggests the format was derived from Richard Bentley’s 1711 edition of Horace. 6. The examples are from Middleton (Loughrey and Taylor 1988). I have found offended students to be both male and female. 7. See the chapter by Sonia Massai on ‘Shakespeare and Digital Editions’. 8. For samples, see Galey (n.d.). The project is now hosted by Laura Mandell at the Center of Digital Humanities Research at Texas A&M, in association with Alan Galey. 9. Also threatening the usual financial arrangements, in which editors receive royalties and publishers receive profits, is the growing demand in the UK that scholarly materials – so far only journal articles, but possibly in the future monographs and editions – be ‘open access’.

REFERENCES Bate, Jonathan and Eric Rasmussen, eds (2007), William Shakespeare: Complete Works, The RSC Shakespeare, New York: Modern Library. Best, Michael, coordinating ed. (1996–), Internet Shakespeare Editions. Available online: https://internetshakespeare.uvic.ca/ (accessed 13 November 2020). Bevington, David, ed. (2014), The Complete Works of Shakespeare, 7th edn, New York: Pearson. Bevington, David, Martin Butler and Ian Donaldson, gen. eds (2012), The Cambridge Edition of the Works of Ben Jonson, 7 vols, Cambridge: Cambridge University Press. Blayney, Peter W. M. (1991), The First Folio of Shakespeare, Washington, DC: Folger Shakespeare Library. Braunmuller, A. R., ed. (1997), Macbeth, The New Cambridge Shakespeare, Cambridge: Cambridge University Press. Butler, Martin, ed. (2005), Cymbeline, The New Cambridge Shakespeare, Cambridge: Cambridge University Press. Crewe, Jonathan, ed. (2002), Twelfth Night, The Pelican Shakespeare, London: Penguin.

THE MODERN EDITING: THE APPARATUS

221

Deighton, Kenneth, ed. (1907), Pericles, The Arden Shakespeare, London: Methuen. DelVecchio, Doreen and Antony Hammond, eds (1998), Pericles, The New Cambridge Shakespeare, Cambridge: Cambridge University Press. Drakakis, John, ed. (2010), The Merchant of Venice, The Arden Shakespeare, London: Methuen. Edwards, Philip, ed. (1976), Pericles, New Penguin Shakespeare, London: Penguin. Evans, G. Blakemore, gen. ed. (1974), The Riverside Shakespeare, Boston, MA: Houghton Mifflin. Evans, G. Blakemore, gen. ed. (1997), The Riverside Shakespeare, 2nd edn, Boston, MA: Houghton Mifflin. Galey, Alan (2014), The Shakespearean Archive: Experiments in New Media from the Renaissance to Postmodernity, Cambridge: Cambridge University Press. Galey, Alan (n.d.), ‘Visualizing Variation’. Available online: http://www. visualizingvariation.ca/ (accessed 13 November 2020). Gossett, Suzanne and Helen Wilcox, eds (2019), All’s Well That Ends Well, The Arden Shakespeare, London: Bloomsbury. Greenblatt, Stephen, gen. ed. (1997), The Norton Shakespeare, New York: Norton. Greenblatt, Stephen, gen. ed. (2008), The Norton Shakespeare, 2nd edn, New York: Norton. Greenblatt, Stephen, gen. ed. (2015a), The Norton Shakespeare, 3rd edn, New York: Norton. Greenblatt, Stephen, gen. ed. (2015b), The Norton Shakespeare Digital Edition, New York: Norton. Available online: https://digital.wwnorton.com/3013 (accessed 13 November 2020). Greenblatt, Stephen, gen. ed. (2016), The Norton Shakespeare: Essential Plays, the Sonnets, New York: Norton. Hoeniger, F. D., ed. (1963), Pericles, The Arden Shakespeare, London: Methuen. Hooks, Adam (2016), ‘Making Histories; or, Shakespeare’s Ring’, in Heidi Brayman, Jesse M. Lander and Zachary Lesser (eds), The Book in History, The Book as History: New Intersections of the Material Text, 341–73, New Haven, CT: Yale University Press. Levenson, Jill L. (2015), ‘Framing Shakespeare: Introductions and commentary in critical editions of the plays’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 377–90, Cambridge: Cambridge University Press. Loughrey, Bryan and Neil Taylor, eds (1988), Five Plays by Middleton, London: Penguin. Lucas, F. L., ed. (1927), The Complete Works of John Webster, 4 vols, London: Chatto and Windus. Malone, Edmond (1778), ‘An attempt to ascertain the order in which the plays attributed to Shakespeare were written’, in Samuel Johnson and George Steevens (eds), The Plays of William Shakespeare, 10 vols, 1.269–346, London. Malone, Edmond, ed. (1790), The Plays and Poems of William Shakespeare, 10 vols, London. Marcus, Leah (2015), ‘A Man Who Needs No Introduction’, in Margaret Jane Kidnie and Sonia Massai (eds), Shakespeare and Textual Studies, 285–99, Cambridge: Cambridge University Press.

222

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Maxwell, J. C., ed. (1956), Pericles, The New Shakespeare, Cambridge: Cambridge University Press. McDonald, Russ and Lena Cowen Orlin, eds (2015), The Bedford Shakespeare, Boston, MA: Bedford St Martin’s. McMullan, Gordon, ed. (2000), King Henry VIII, The Arden Shakespeare, London: Thomson Learning. Neilson, William Allen and Charles Jarvis Hill, eds (1942), The Complete Plays and Poems of William Shakespeare, Cambridge, MA: Houghton Mifflin. Orgel, Stephen and A. R. Braunmuller, gen. eds (2002), The Complete Pelican Shakespeare, London: Penguin. Pope, Alexander, ed. (1725), The Works of Shakespear, 6 vols, London. Proudfoot, Richard, Ann Thompson and David Scott Kastan, gen. eds (2011), The Arden Shakespeare Complete Works, rev. edn, London: Thomson Learning. Prouty, Charles T., ed. (1958), Twelfth Night, The Pelican Shakespeare, Baltimore: Penguin. Shakespeare, William (1623), Mr. William Shakespeares Comedies, Histories & Tragedies, London. Taylor, Gary and John Lavagnino, gen. eds (2007), The Collected Works of Thomas Middleton, Oxford: Clarendon Press. Taylor, Gary and Michael Warren, eds (1983), The Division of the Kingdoms: Shakespeare’s Two Versions of ‘King Lear’, Oxford: Clarendon Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2016), The New Oxford Shakespeare: The Complete Works: Modern Critical Edition, Oxford: Oxford University Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2017), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, Oxford: Oxford University Press. Taylor, Gary, Terri Bourus and Gabriel Egan, gen. eds (forthcoming), The New Oxford Shakespeare: The Complete Alternative Versions, 2 vols, Oxford: Oxford University Press. Thompson, Ann, ed. (1984), The Taming of the Shrew, The New Cambridge Shakespeare, Cambridge: Cambridge University Press. Vaughan, Virginia Mason and Alden T. Vaughan, eds (1999), The Tempest, The Arden Shakespeare, London: Thomas Nelson. Warren, Roger, ed. (2003), Pericles, The Oxford Shakespeare, Oxford: Oxford University Press. Weis, René, ed. (2012), Romeo and Juliet, The Arden Shakespeare, London: Bloomsbury. Wells, Stanley and Gary Taylor, gen. eds (1986), The Oxford Shakespeare: The Complete Works, Oxford: Clarendon Press.

PART THREE

New directions

224

CHAPTER 3.1

Shakespeare and authorship attribution methodologies HUGH CRAIG

This chapter is a primer in Shakespeare authorship attribution.1 I present a series of examples to illustrate some of the key considerations which one should bear in mind in an attribution study. They have some technical aspects which will require patience on the reader’s part to work through but doing so will (I hope) help to acclimatize new and prospective attributionists to the constraints and opportunities of this practice. On the whole the questions I deal with are statistical rather than literary. We come to Shakespeare attribution because of an engagement with the content of his work and his contemporaries’, but the skills and mental habits we need for quantitative attribution are not literary, or at least not part of the usual literary training. Those most interested in Shakespeare tend to focus on resonant details and linger on individual instances, and seek for large intuitive insights, but for attribution the key is even, wide attention and a systematic method inoculated against bias. The quantitative part of attribution occurs at the juncture between statistics and language. It fits awkwardly with the documentary and historical side of determining authorship, and with evaluative, interpretive literary studies. A quantitative approach is, unavoidable, nevertheless, if we are to make reliable assignations of anonymous and disputed texts and establish levels of confidence about an attribution. My first two case studies concern the author effect on which attribution depends. I show that this effect is objectively present in Shakespearean works, and go on to discuss exceptions and limitations to consistent authorial self-resemblance. I then treat two other key concepts: the law of large numbers and the problem of overfitting. Finally, I discuss the practicalities of assembling a corpus and applying statistical procedures. Some scholars are sceptical about the idea that playwrights of Shakespeare’s era have distinctive styles which make their works recognizable as belonging to them and no one else. For some, the doubts are based on a belief that individual authorship was subservient to broader cultural forces in the early modern period (McMullan 2000: 6, 174, 193–5). For others, the problem is that variation within

226

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

authorial canons is too great (Rudman 2016). Near the beginning of his Oxford History of English Literature volume, English Drama 1586–1642, G. K. Hunter strikes a general cautionary note on the topic: From a seat in the stalls it is difficult to know why we should think [Ben Jonson’s play] The New Inn is by the same author as The Alchemist, why the Heywood who wrote A Woman Killed with Kindness is the same Heywood who wrote The Golden Age, why the author of The Merry Wives of Windsor is also the author of The Tempest. (1997: 3n.5) Authorship attribution has to start from a ground zero, which is the suspicion that in this period an author’s plays have only weak connections to each other. If we had no external guide, if this was a blind tasting, as it were, would we connect two unlabelled plays by the same author with each other? I first reframe the problem as the question of whether authors repeat characteristic distinctive phrases. Do they keep the same writing habits as time goes on, when they turn to different genres or try different topics? If not, there will be no basis to identify a mystery work as theirs. It will have no necessary connection to the works we know to be written by them. If so, we can go forward with some confidence that there is a degree of predictability about what an author writes, though this will not remove the constant problem of defining just how much predictability, and in what circumstances. I also shift the ground from an audience’s perceptions of plays performed to patterns in written texts detectable by a computer programme, and test the conundrum of authorial self-consistency versus authorial indeterminacy on a numerical basis. Are an author’s works more like each other than they are like other authors’ works, even if those other works share a genre, or an era in time, or a plot outline? We have to be careful that any test is as fair as possible. Either side of the argument has to have an equal chance to prevail. I will now present an attempt to do this. We take five playwrights of the time, including Shakespeare, and works by them with a date of first performance between 1580 and 1619, comfortably straddling Shakespeare’s career. For the dating of the plays we rely on Wiggins and Richardson’s Catalogue of British Drama (2012–). We make a random selection of nine plays from each, and then choose a further two plays from each writer as test plays, again using a random selection if there are more than two to choose from.2 We focus on repeated sequences of six words. Choosing a sequence length of six is an arbitrary choice, designed to yield repetitions which will be unusual enough to give us only small numbers to deal with, so that each instance can be examined without undue labour. As a matter of simple mathematics, as the sequence length gets longer, there are fewer repetitions. Working in single words yields relatively few different words, most of them frequently recurring, two-word sequences have more different sequences, and fewer repeats, and so on. We seek examples that are rare enough to be potentially characteristic of an author rather than belonging to the common parlance of the day. To determine

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

227

rarity, we require from the beginning that any sequence we include should not appear in either of two reference sets. One set consists of all the plays in the corpus which are dated 1580–1619 and attributed to single authors (other than our chosen five) in the Catalogue. The other is the full set of 1580–1619 plays by the four chosen authors other than the author set being used at the time to test authorial linkages. If a sequence satisfies the condition that it does not occur in the general single-author set, or in the appropriate four-author set, then we can be sure that it is tolerably rare, and we can attach some importance to any repetition. We have ten test plays, two from each of the five authors. We find how many rare six-word sequences are shared between a given test play and each of the five nine-play authorial sets. Each play thus has five scores, one for each authorial set. There may be no matching rare sequences, or one, or more. The results are shown in Table 3.1.1, with test plays as rows and authorial sets as columns. In the table, the cells showing the number of sequences shared by the test plays and the nine-play sets of their known authors are shaded. This number is always more than zero apart from one case, Chapman’s Conspiracy of Charles Duke of Byron, which does not share a rare sequence with the Chapman set. The unshaded cells show the number of sequences shared between the test plays and the authorial groups other than that of their known author. These are zero in every case. Table 3.1.1  Rare six-word sequences shared between ten plays and five authorial sets Test plays

Sets of nine plays as sources for matched 6-grams Chapman Fletcher Jonson Middleton Shakespeare

Chapman

Conspiracy of Charles Duke of Byron

0

0

0

0

0

Chapman

Revenge of Bussy D’Ambois

6

0

0

0

0

Fletcher

Monsieur Thomas

0

7

0

0

0

Fletcher

Mad Lover

0

6

0

0

0

Jonson

Bartholomew Fair

0

0

16

0

0

Jonson

Poetaster

0

0

14

0

0

Middleton

No Wit, No Help Like A Woman’s

0

0

0

23

0

Middleton

Puritan

0

0

0

15

0

Shakespeare Henry IV, Part Two

0

0

0

0

7

Shakespeare Julius Caesar

0

0

0

0

3

228

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Authors do repeat themselves, according to this test. When we seek out parallels between the test plays and their known authors, we find (nine times out of ten) that there are links, sometimes many of them. In the extreme case of Middleton’s No Wit, No Help Like a Woman’s, there are twenty-three. When, under exactly comparable circumstances, we seek out parallels with other authors, we find none.3 We might have anticipated that any shared rare sequences would be evenly distributed across same-author and other-author sets, because of the ebb and flow of dialogue in a theatre with common preoccupations and ways of making drama, but this is not the case. We might have thought that there would be no such shared sequences, because any new sequences in a given test play would be unique, given the wide range of possibilities available in a language like English, and because of the motivation of authors to provide something new for audiences. This is not true either. It turns out that authors have a tendency to resort to characteristic unusual phrasings even in a much later play, or in a play in a different genre. Early in Julius Caesar Caska says to Cassius, ‘Stand close awhile, for here comes one in haste’ (1.3.131). Those last six words also appear in Much Ado About Nothing: benedick

And how do you? beatrice

Very ill too. benedick

Serve God, love me, and mend. There will I leave you too, for here comes one in haste. (5.2.83–6) This six-word sequence sounds commonplace enough, but it does not appear anywhere in the fifty-one plays by the other four authors, or in the eighty-eight plays by other writers. Julius Caesar also shares a second rare sequence, ‘I thank you for your pains’ (2.4.115), with Cymbeline (1.6.202), Much Ado (2.3.240), Taming of the Shrew (3.2.183) and Twelfth Night (1.5.275), and a third, ‘and bring me word what he’ (2.4.47), with 1 Henry IV (5.1.109–10), giving it a score of three for Shakespeare links (see Table 3.1.1).4 If we expanded our set of plays, or changed the rules in other ways, we might find examples of these sequences in other authors. But under precisely the conditions specified, we do not. Given an equal chance to emerge in parent author sets and in others, links to the parent authors dominate. We also ensure comparability by following standard protocols in preparing the texts. In all of them, spelling is modernized, for instance, contractions are expanded in the same way, and the same grammatical functions are marked. Stage directions and other text not meant to be spoken or sung on stage are excluded. The analysis in Table 3.1.1 is limited to one feature, six-word sequences, and one way of collecting statistics about this feature, but there is no reason to think that the same pattern would not hold true for five-word sequences, four-word sequences or a different selection of plays or authors. It is supported by a large number of Shakespeare authorial studies (e.g. Jackson 2014, and Taylor and Egan 2017), which

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

229

typically begin with a test of the reliability of the scheme proposed to determine authorship, using test plays and play portions of known authorship. Another way to test the hypothesis of authorial distinctiveness is to examine the relative frequency of individual words, focusing on the very common ones like the and know. Here we are interested in marked and consistent differences in frequencies between canons. The obvious way to assess this kind of difference would be to observe whether averages of a word are higher or lower in one author compared to another, but this would not take into account the variation from work to work. If a frequency changes wildly through a corpus, then a high or low count in a mystery text does not tell us much. The ‘t-test’, first introduced by W. S. Gosset in 1908, takes account of this element of variation in counts by dividing the difference between the two averages by the combined standard deviations of the two sets of counts (Craig and GreatleyHirsch 2017: 50–2). ‘Standard deviations’ are measures of dispersion around the average. The higher the standard deviation, the greater the dispersion. Thus, the size of the final t-statistic is moderated by the amount of variation. The t-statistic has a predictable distribution, given the number of observations, so we can tell how often a given score would come about by chance alone. We take each of 100 very common words and test how many of them are consistently different in frequency in pairs of authors. We take the same five authors as before, but this time use all available works by them, since the t-test results are not affected by differences in the amount of observations in the two sets being compared. We adopt the threshold of a probability of 1 in a 100 that the result could have come about by chance. A t-test score at that level or higher is counted as a highly significant difference. There are ten two-way comparisons possible between the five authors. Table 3.1.2 shows the results. The shaded cells are comparisons of one playwright group with itself, or repeats of comparisons in the upper right portion of the chart. The comparisons of interest are in the unshaded cells. The lowest score recorded is seventeen, for the comparison between Chapman and Jonson. The expectation for random data is that one word out of the 100 tested would be significantly different, so even seventeen is an unexpectedly high score. The table shows strong differentiations between

Table 3.1.2  Numbers of significant differences in the frequencies of 100 very common words in comparisons among five authors Chapman Chapman Fletcher Jonson Middleton Shakespeare

Fletcher 35

Jonson

Middleton

Shakespeare

17

32

21

38

32

45

31

34 46

230

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Table 3.1.3  Numbers of significant differences in the frequencies of 100 very common words in comparisons among five random sets of plays. Random group A Random group A Random group B Random group C Random group D

Random group B 0

Random group C

Random group D

Random group E

2

1

0

0

0

0

1

0 2

Random group E

the authors. The highest counts are forty-five for the comparison of Fletcher and Shakespeare, and forty-six for the comparison of Middleton and Shakespeare. To check that these counts do not arise from chance local concentrations of plays of one genre, era or type, we can make up some random groups of plays and try the same test. To align with the five-author comparison set, we establish groups of different sizes, thirteen, sixteen, seventeen, nineteen and twenty-eight, to match the sizes of the Chapman, Fletcher, Jonson, Middleton and Shakespeare sets. Table 3.1.3 shows the results. In six of the comparisons, there are no significantly different words, in two of them there is one, and in two of them there are two. This indicates that the test of word frequency profiles in groups of plays is working as expected, and that the theoretical expectation of one significantly different word in randomly mixed sets is broadly confirmed in practice. Chance throws up the occasional significant difference, but no more than that. As already mentioned, the second largest number of significant differences in the author comparisons is in the comparison of Fletcher and Shakespeare (Table 3.1.2). These two playwrights collaborated on plays and are thus a natural pair for investigation. Of the forty-five words the largest t-test score and thus the most significant difference is for rates of use of in as a preposition (‘she is in the court’ rather than ‘she went in just now’). Figure 3.1.1 shows percentage counts for Shakespeare plays to the left and for Fletcher plays to the right. The highest percentage count of this word among the Fletcher plays, appearing as the lowest black bar, is in Rule a Wife and Have a Wife, 0.9 per cent, but this is still lower than the lowest count for a Shakespeare play, the highest grey bar, just under 1 per cent, in The Winter’s Tale. Thus, the frequency of this word provides a complete separation between Fletcher and Shakespeare plays. This is true also of one other word in the set of one hundred, that as a conjunction, where the lowest Shakespeare score is again higher than the highest Fletcher score. We can conclude that there are persistent internal consistencies in authorial canons. In both our examples, shared rare sequences and different rates of use of very common words, we gave the test of authorial consistency a chance to fail. The rare sequences could have appeared across the comparisons with no real pattern.

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

231

FIGURE 3.1.1  Percentage counts of in as a preposition in 28 plays by Shakespeare and 16 plays by Fletcher.

There could have been no significant differences in frequency between canons, or differences appearing equally in canons and random assemblages. Canons, when given the chance, behave like clusters of works with marked strands of similarity. This authorial distinctiveness and self-consistency is the foundation for work in attribution. It is never to be taken for granted, since it does not necessarily appear in every mode, genre, period, and there are questions, to which we turn below, about exceptions, and small samples and small canons. But the two studies above have shown that it is reasonable to start a Shakespeare attribution study with an assumption of persistent underlying authorial difference. Hath, the older form of has, is another common word which Shakespeare and Fletcher use at different rates, as the t-test confirms. Shakespeare’s average is much higher. There is one stark exception to the Fletcher pattern, however (see Figure 3.1.2).

232

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 3.1.2  Percentage counts of hath in 28 plays by Shakespeare and 16 plays by Fletcher.

Fletcher counts are generally much lower than Shakespeare’s, and two Fletcher plays have no instances at all, but The Faithful Shepherdess has forty-seven instances, 0.24 per cent of its dialogue. The next highest, The Mad Lover, has six instances, or 0.03 per cent. Fletcher avoids hath generally, in strong contrast to his sometime collaborator Shakespeare, but departs from his general practice to a marked degree on one occasion. (There is always the possibility that in this case, exceptionally, a scribe or compositor altered the forms they found in their copy, but equally there is nothing in the bibliography of Faithful Shepherdess that indicates this.) If the authorship of this play was disputed, and we were relying on this marker, it would seem to offer strong evidence that Shakespeare is a more likely author than Fletcher. Cyrus Hoy and Jonathan Hope both excluded Faithful Shepherdess from their Fletcher reference sets for attribution purposes (Hoy 1956; Hope 1994). At times authors can confound standard authorship methods and write unlike themselves throughout a work.

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

233

We have to reckon also with the question of smaller samples. The texts we are interested in are often acts, scenes or even parts of scenes rather than whole plays. These shorter samples are inherently more likely to diverge from a general authorial standard, according to the statistical law of large numbers. Larger samples give more reliable results because more data lies behind them. To illustrate the importance of sample size, we can take the case of characters’ spoken parts in plays. In Shakespeare’s core canon the number of words spoken by a character (excepting ‘mute’ characters) ranges from one – ‘Aye’ (the second senator in Cymbeline) or ‘Stand’ (the thieves in 1 Henry IV) – to Hamlet’s 11,328. Usually we express the frequency of a particular word as a percentage, to take account of variations in size like this, but this may disguise the difference in meaningfulness between the percentage score in a small character part and the percentage score in a large one. Figure 3.1.3 shows the percentages of the word the, generally the commonest word in plays, for the 578 characters with 100 words or more in the core Shakespeare canon.

FIGURE 3.1.3  Percentages of the word the in larger Shakespeare character parts, versus size of part.

234

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Hamlet is at the top, with more than 11,000 words, as already noted. The datapoints for smaller characters are at the bottom of the chart. Their scores are widely scattered across the horizontal axis, which represents the percentages of dialogue represented by the. Mamillius in The Winter’s Tale is at the left-hand extreme. He speaks 155 words without using the at all, so has a score of 0 per cent. At the other extreme is the Second Gentleman in Othello, who speaks 136 words, of which fifteen are the, a score of 11 per cent. The larger characters cluster much more closely around the overall Shakespeare average, which is 3.3 per cent (20,418 instances of the in 625,041 words in the twenty-eight plays). Hamlet, at the top, has 4 per cent, then in descending order come Iago, Richard III and Henry V, with 2.6 per cent, 3 per cent and 3.8 per cent, respectively. With the larger parts, one-off factors are balanced out and underlying consistent trends win out. The difference between the percentages of Hamlet’s and Iago’s parts for the chosen word, 4 per cent for Hamlet, and 2.6 per cent for Iago, are more meaningful than this same difference in characters with very small parts, where we could just about dismiss the variation as a chance effect. In authorship attribution, percentage use of words like the is often an important marker. Shakespeare’s average is 3.3 per cent, as already noted, whereas Fletcher’s is much lower, 2.4 per cent (8,072 instances in a canon of sixteen plays totalling 340,719 words). We would expect 621 instances of the in an average 19,000-word play by Shakespeare, but 450 instances in a play of the same length by Fletcher.5 Yet Shakespeare-like or Fletcher-like percentages mean little in small samples, because they may be simply the result of chance. In Figure 3.1.4 the scores for Fletcher characters for the are plotted and shown with the scores for Shakespeare characters repeated from Figure 3.1.3. As with the Shakespeare characters, and again following the law of large numbers, as the Fletcher character parts get larger, they disperse less widely around the overall Fletcher average. (One black circle spoils the neat Christmas tree shape made by the Fletcher entries, with a count of the much higher than the Fletcher average despite a relatively large size of part. This is the Satyr from Faithful Shepherdess, with 1,931 words in all, of which five and a half percent are the. Fletcher’s pastoral is once again anomalous.) The tendency for Shakespeare characters to use more the proportionally than Fletcher characters is clear overall, but most of the smaller characters, those below 2,000 words, are in shared territory, with more overlap than distinctiveness. Given the choice, one would always want a larger sample with its consequent greater reliability. A percentage score is more weighty for purposes like attribution when based on a large sample. The same rule applies to authorial canons in attribution. We can think of the process as one of prediction. We have a set of sole-authored works by a given author, and we aim to use the patterns found there to predict how an author would write in their next attempt in a similar text type. If, as in the case of Thomas Nashe, there is just one sole-authored play available, we are on shaky ground. This work, Summer’s Last Will and Testament, is a comedy, but we need a guide to Nashe’s writing more broadly, if we are investigating his possible contribution to the history play 1 Henry VI, or the tragedy Dido, Queen of Carthage. Christopher Marlowe also has a claim to Dido, but his canon of six plays for comparison quickly shrinks

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

235

FIGURE 3.1.4  Percentages of the word the in larger Shakespeare and Fletcher character parts, versus size of part.

on examination. Doctor Faustus is very likely a collaboration, and exists in two different early editions, one much shorter than the other; The Jew of Malta may well have been revised by another author; and the surviving version of The Massacre at Paris, has evidently suffered in transmission (see the entries for the various plays in Wiggins and Richardson 2012–). Thomas Kyd is another author who is often a person of interest in attributing anonymous and disputed plays, but just one wellattributed original play, The Spanish Tragedy, is available as the basis to judge the potential stylistic range and limits of his writing, though we may be tempted to relax the requirements and add a translation, Cornelia, and one further anonymous play, Soliman and Perseda, which is very likely to be his.

236

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

By contrast Shakespeare is a haven of safety with twenty-eight well-attributed single-author plays. This is larger than any other surviving early modern dramatic canon apart from that of James Shirley. The canons of Chapman, Fletcher, Jonson and Middleton, already discussed in connection with questions of authorial consistency, also qualify as beneficiaries of the law of large numbers. We can be more confident in attributing plays and play sections between any of these five than with Nashe, Marlowe and Kyd. So far we have concentrated on single variables like the very common words or, as with the six-word sequences, on a simple accumulation of instances in a category. Readers can check a particular concentration or dearth against their own perceptions of the language of a work. A given number of concrete occurrences is a secure and simple foundation, introducing the minimum of potentially sophisticating transformations, and only going as far as expressing the raw counts as proportions to take account of different sample sizes. There is also a battery of ‘multivariate’ classification procedures which use a number of variables together (Tabachnick and Fidell 2018). They combine the separate discriminatory power of single variables into a composite method with power to classify in more difficult cases. One example of a procedure of this kind is Linear Discriminant Analysis. This is based on a method invented by Sir Ronald Fisher and first announced in 1936. We concentrate here on its simplest form in which there are just two classes to be considered – for instance, Author A and Author B, or Author A and a composite set of other authors. Each of the variables is given a weighting, high or low, positive or negative, with the objective of building a composite function which does the best possible job of delivering scores which divide the two classes. The ideal outcome is a function on which all the texts in one class have higher scores, or all have lower scores, than all the texts in the other class. We can consider applying this method to a standard quantitative authorship attribution project which has a mystery text and a pair of candidate authors. The first step is to seek out the distinctive characteristics of the candidate authors, and the second is to compare these profiles of features with the patterns in the mystery text. When this process is automated, we can speak of making a ‘classifier’ and ‘training’ it on samples of text known to be by the candidate authors. The procedure works through the data it is given, to find variables on which the authors differ consistently and to compile them into a single test. Now a second group of considerations emerge. The purpose of the exercise is to make the correct assignation of the mystery text to one class or the other. To prepare for this we have trained the classifier on the available texts of known provenance and the available features, and our measure of success is performance with those. However, we have as yet no way of knowing how well this classifier will perform with a freshly introduced text. We may have a classifier which is perfectly adjusted to the training set but does not generalize well to unknowns. This is the problem of ‘overfitting’ – tailoring a classifier to a training set while ignoring the ultimate purpose of the exercise, which is dealing with new items. The usual approach to this problem of estimating reliability with freshly introduced samples is to use some of the training set as ‘test’ items. We take some

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

237

samples out of the training set at the very beginning, train the classifier without them and then use the classifier to assign the reserved samples to one class or the other. In this case we know the true class to which the sample belongs, so this is a good check on the reliability of the classifier. To demonstrate the workings of these ‘training’ and ‘test’ sets we can carry out a Linear Discriminant Analysis which does not focus on a mystery text, but rather concentrates on evaluating the reliability of the method. We use the core canon of twenty-eight Shakespeare plays as one class and a set of 138 single-author wellattributed plays by others, with dates of first performance 1580–1619, as the second. We divide all the plays into 2,000-word segments. This yields 301 Shakespeare segments and 1,253 non-Shakespeare ones. We have 1,000 word-variables available, percentages of 1,000 very common words, having counted instances of these words in all the segments. With the assistance of the statistics program SPSS we create a Linear Discriminant Function maximizing the distinction between the two classes (IBM Corp. 2017).6 This achieves a perfect separation of the Shakespeare and nonShakespeare classes. To provide a guide to the severity or otherwise of overfitting in a given case, SPSS offers one standard form of ‘test’ sampling, known as ‘cross-validation’. A single segment from the training sets is withheld at the beginning and classified at the end, the result (right or wrong) is noted, then this sample is replaced and another is reserved and classified, and so on, until all segments have been used in this way. The program reports that in the end 261 out of 301 Shakespeare test segments and 1,158 out of 1,253 non-Shakespeare test segments were correctly assigned, or in all 1,419 out of 1,554 segments, 91.3 per cent. This is much lower than the result for the training segments, which was 100 per cent, but we could regard it as more accurate as a prediction of what would happen with a mystery text, since it comes closer to emulating the circumstances of the classification of a newly introduced sample. These newly introduced samples are likely to have idiosyncrasies which are not taken account of in the process of building the classifier, whereas the anomalies in the members of the training sets (on both sides) are accommodated in arriving at the weightings in the particular function which emerges after the training process. This result is probably still flattering to the classifier, however, since while each segment is new to the particular classifier which is built without it, the remaining segments of the longer text from which the test segment comes are not. They remain in the relevant training set. We can expect considerable likeness within the segments of a play, for instance, much greater than the likeness between the segments of different plays, so some of the idiosyncrasies in the cross-validation segment will have been taken account of this way in the classifier. This would not be true of a mystery set of segments. For this reason, we perform a second test of the test, this time reserving a group of whole plays, with all their segments, at the beginning, and classifying them at the end once the classifier has been trained on the remaining plays and their segments. We choose a quarter of the Shakespeare plays and a quarter of the non-Shakespeare plays at random and make these the test set while the training set is the remaining three-quarters of the two original groups (Table 3.1.4). This time we find that 319 out of a total of 389 segments have been correctly assigned, or 82 per cent. This test set result is much lower than for the training set

238

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Table 3.1.4  Training and test sets Plays Shakespeare training Shakespeare test Non-Shakespeare training Non-Shakespeare test Total

Segments

21

230

7

71

104

935

34

318

166

1554

(100 per cent). Given this difference we can say the function is ‘overfitted’ to the training set, i.e. it is wonderfully well attuned to the variation in the training set, but so much so that its performance with a test set is relatively poor. We have the option of using fewer variables, in an attempt to moderate this overfitting aspect. Fewer variables means less opportunity to fit the classifier minutely to the training set. Figure 3.1.5 shows what happens when we use first only the 100 most common words, then the 200 most common words, and so on, moving in 100word jumps up to the set of 1,000 we began with. Working backwards from the result we have already discussed, using 1,000 words, shown here at the right-hand end of the chart, we can see that the performance of the classifier in the cross-validation by segment and in the whole-play test is much better with fewer words, peaking at 400 words for both (at 94.3 per cent and 92.3 per cent, respectively). At 100 words the performance of the classifier is a little better with the whole-plays test than with the cross-validation set, but thereafter the performance with the test set is lower, and by a generally increasing margin. At 400 words, presumably, there is the advantage of having more markers, and a better chance of including good individual markers, than with the earlier marker sets, but the problem of overfitting has not yet set in. The performance with the training set reaches 100 per cent with 600 words and this is maintained through the rest of the trials. The descent of the dashed line and the tramline after 400 words visualizes the overfitting that is taking place. As the procedure includes more words and chooses weightings for them to maximize success with the training set, the function performs less and less well with freshly introduced segments. There is no way of knowing how closely the pattern of Figure 3.1.5 would be repeated with a different authorial contrast, smaller or larger segments, a different feature set or a different statistical procedure. It would be risky to rely too much on the fact that in this case the peak is at 400 words. On the other hand, this demonstration does serve to illustrate some general relationships that will emerge regularly in work of this kind. More markers make for a better performance up to a point, and then they bring the risk of overfitting; cross-validation performance counts will generally be lower than training set results; and whole-text test results will generally be lower again. We can note in passing that at its best, in a whole-text test set, this system correctly assigns 2,000-word segments to Shakespeare, and away from him, ninetytwo times out of a hundred. This suggests we can be confident of detecting a powerful author effect in studies like this, whatever the sceptics may say, but we also have to remember that we can be sure of making some errors in the attributions.

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

239

FIGURE 3.1.5  Success rates for the classification of Shakespeare plays and non-Shakespeare plays using linear discriminant analysis and ten different word-variable sets.

My aim in this chapter has been to establish some principles to guide the scholar embarking on authorship attribution work. I hope it is now evident that they can have confidence that an author effect exists, but must be aware that this effect is not necessarily even, given the myriad forces making for variation in style. They can be encouraged by the power of multivariate tests, but know that this power brings its own problems. Once they have a method and a corpus which seems adequate, they must always return to the fundamental question: how often does the new classifier assign correctly samples as close as possible to the mystery text in size and type, but of known provenance? The saving grace of quantitative authorship attribution is that so much of it is testable. We do not have to rely on a priori axioms, ex cathedra pronouncements or common sense beliefs about the strength or otherwise of a given kind of evidence or given method, but can estimate this strength for ourselves and watch the fascinating contest between pattern and variation in literary language play out.

240

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

PRACTICALITIES In an ideal world, Shakespeare attribution work would be supported by open data and open tools, so that steady progress could be made as methods are compared and experimental design is refined, but this is some way off. Fortunately, some open-source, well-supported tools do exist. Stylo, for instance, is a package for the statistical computing environment R, which is tailor-made for stylometry (Eder, Rybicki and Kestemont 2016). It can take the neophyte researcher from raw text to statistical procedures. Good metadata is also available. The online DEEP: Database of Early English Playbooks is limited only by the exclusion of manuscript plays from its remit (Farmer and Lesser 2007). Researchers can download the entire set of items and attributes, or search the information online. Nine of ten volumes of the British Drama 1533–1642: A Catalogue have now been published (Wiggins and Richardson 2012–). This gives a wonderful amount of detail – down to the props used in a given play – and offers the most authoritative current judgement across the board of questions like authorship and date. Metadata is a critical aspect of authorship study. With its help, the attributionist can strengthen a corpus for a given purpose by excluding some works, like translations, plays written for reading rather than performance, and masques and entertainments, or can test to see whether a fancied association such as a frequency and date is statistically significant. It is disappointing to report that there is no well-edited comprehensive corpus of Shakespeare-era drama supported by an enduring institution and free to download. Almost all printed texts from the period now exist as digitized images, but machine-readable text requires transcription by mechanical means like Optical Character Recognition or human means like keyboarding, and then a further stage of editing. Quality edited machine-readable texts based on individual early editions or manuscripts are still in short supply. The situation for Shakespeare texts, as ever, is the exception. Internet Shakespeare Editions (Best 1996–) and the Oxford Text Archive (n.d.) both make available proofread texts based on individual early printed versions. For texts beyond Shakespeare, EEBO-TCP offers a vast array of titles, but they contain many gaps where keyboarders found the text hard to read (Early English Books Online-Text Creation Partnership 2015). Literature Online, or ‘LION’, has a large set of well-edited texts, but these are for searching rather than downloading, and they require a subscription for access (ProQuest 2019). New Shakespeare attributionists will therefore need to find texts wherever they can. They will most likely have to do some editing, to fill in gaps or to convert old spelling to standard modern and to expand contractions. They may also want to lemmatize, collecting different forms (such as cry, cries, cried and crying) under a single dictionary headword, and they may want to tag words for parts of speech so as to distinguish various homographs (such as that in ‘she said that she would’, ‘see that sword’, and ‘the book that I left’). Software exists to help with these tasks, such as VARD 2 for modernizing spelling (Baron 2013), but none of it is capable of a good result on early modern English text without human input, so there is considerable labour involved, and the larger the corpus and the higher the standards of accuracy, the more laborious it becomes.

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

241

Texts for counting should be prepared in a consistent way, and there is much to be said for separating the text proper from other material like prefaces, dedications, commendatory verse and footnotes, and, in the case of drama, from speaker prefixes and stage directions as well. Otherwise instances of all in footnotes may be inadvertently included in totals for a poem (Craig 2012b: 171n.58); a repeated speaker tag like ‘An.’ may inflate the totals for the word an in dramatic dialogue (Jackson 1999); or counts of exeunt may be mistakenly added to a list of authorial markers derived from a corpus mixing drama and non-dramatic materials (FreeburyJones and Dahl 2019: 5). The Text Encoding Initiative offers a comprehensive standard set of tags to encode this parsing of the text (Text Encoding Consortium 2019). Before plunging into a study, a researcher new to the field might consult some background work on authorship and attribution (Love 2002; Craig 2012a) and read some model studies (Vickers 2002; Jackson 2014; Taylor and Egan 2017). Shakespeare attribution is not for the faint-hearted. The required investment in text preparation and in learning methods is considerable. Any findings will be given intense and sometimes hostile scrutiny before and after publication, since the stakes are high and positions on the many contested questions are entrenched. On the other hand, the intersection of statistics and language, and its application to works of towering cultural prestige, makes for a heady mix, and I imagine this will continue to prove irresistible for those of a certain temperament, though sometimes against their better judgement.

NOTES 1.

Thanks to Ruth Lunney and Brett Greatley-Hirsch for very helpful comments on an earlier version of this chapter.

2.

The lists of plays are in the ‘Supplementary Materials’ for this chapter, to be found at http://hdl.handle.net/1959.13/1406580, where the reader will also find extra background data and metadata to fill out details of each of the tables and charts in the chapter.

3.

Strictly speaking, the playing field is not entirely level because the five chosen dramatists have different size canons in the corpus – Chapman thirteen, Fletcher eleven, Jonson twelve, Middleton fifteen and Shakespeare twenty-eight. This means that the comparison set for each playwright varies in size. This does not affect comparisons up and down the columns of Table 3.1.1, since the ten test plays in each column have exactly the same comparison set.

4.

We are counting the number of different sequences that are repeated, rather than tallying up all the instances of these sequences, so these sequences count as one link for the purposes of Table 3.1.1.

5.

In the whole-plays comparison between Shakespeare and Fletcher, the has a t-test result which corresponds to a probability of 0.0000005 of coming about by chance.

6.

In SPSS, go to ‘Analyze’, then ‘Classify’ and finally ‘Discriminant’.

242

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

REFERENCES Baron, Alistair (2013), Variant Detector (VARD) 2, Leicester: University of Leicester. Best, Michael, coordinating ed. (1996–), Internet Shakespeare Editions. Available online: https://internetshakespeare.uvic.ca/ (accessed 13 November 2020). Craig, Hugh (2012a), ‘Authorship’, in Arthur F. Kinney (ed.), The Oxford Handbook of Shakespeare, 15–30, Oxford: Oxford University Press. Craig, Hugh (2012b), ‘George Chapman, John Davies of Hereford, William Shakespeare, and A Lover’s Complaint’, Shakespeare Quarterly, 63 (2): 147–74. Craig, Hugh and Brett Greatley-Hirsch (2017), Style, Computers, and Early Modern Drama: Beyond Authorship, Cambridge: Cambridge University Press. Early English Books Online-Text Creation Partnership (2015), EEBO-TCP Phase 1, Oxford and Ann Arbor, MI. Eder, Maciej, Jan Rybicki and Mike Kestemont (2016), ‘Stylometry with R: A Package for Computational Text Analysis’, R Journal, 8 (1): 107–21. Farmer, Alan B. and Zachary Lesser (2007), DEEP: Database of Early English Playbooks. Available online: http://deep.sas.upenn.edu (accessed 13 November 2020). Fisher, Ronald A. (1936), ‘The Use of Multiple Measurements in Taxonomic Problems’, Annals of Eugenics, 7 (2): 179–88. Freebury-Jones, Darren and Marcus Dahl (2019), ‘Searching for Thomas Nashe in Dido, Queen of Carthage’, Digital Scholarship in the Humanities. Available online: https:// doi.org/10.1093/llc/fqz008 (accessed 13 November 2020). Gosset, William S. [writing as ‘Student’] (1908), ‘The Probable Error of a Mean’, Biometrika, 6 (1): 1–25. Hope, Jonathan (1994), The Authorship of Shakespeare’s Plays: A Socio-Linguistic Study, Cambridge: Cambridge University Press. Hoy, Cyrus (1956), ‘The Shares of Fletcher and his Collaborators in the Beaumont and Fletcher Canon (I)’, Studies in Bibliography, 8: 129–46. Hunter, George K. (1997), English Drama 1586–1642: The Age of Shakespeare, Oxford: Clarendon Press. IBM Corp. (2017), IBM SPSS Statistics for Windows, Version 25.0, Armonk, NY: IBM Corp. Jackson, MacDonald P. (1999), ‘Titus Andronicus and Electronic Databases: A Correction and a Warning’, Notes and Queries, 244 (NS 46.2): 209–10. Jackson, MacDonald P. (2014), Determining the Shakespeare Canon: ‘Arden of Faversham’ and ‘A Lover’s Complaint’, Oxford: Oxford University Press. Love, Harold (2002), Attributing Authorship: An Introduction, Cambridge: Cambridge University Press. McMullan, Gordon, ed. (2000), King Henry VIII (All is True), The Arden Shakespeare, London: Thomson Learning. Oxford Text Archive (n.d.). Available online: https://ota.ox.ac.uk (accessed 13 November 2020). ProQuest (2019), Literature Online, Ann Arbor, Michigan: ProQuest, Inc. Available online: https://about.proquest.com/products-services/literature_online.html (accessed 13 November 2020).

SHAKESPEARE AND AUTHORSHIP ATTRIBUTION METHODOLOGIES

243

Rudman, Joseph (2016), ‘Non-traditional Authorship Attribution Studies of William Shakespeare’s Canon: Some Caveats’, Journal of Early Modern Studies, 5: 307–28. Tabachnick, Barbara G. and Linda S. Fidell (2018), Using Multivariate Statistics, 7th edn, Harlow, Essex: Pearson Higher Education. Taylor, Gary and Gabriel Egan, eds (2017), The New Oxford Shakespeare: Authorship Companion, Oxford: Oxford University Press. Text Encoding Consortium (2019), TEI P5: Guidelines for Electronic Text Encoding and Interchange, 3.6.0. Available online: http://www.tei-c.org/Guidelines/P5/ (accessed 13 November 2020). Vickers, Brian (2002), Shakespeare, Co-Author: A Historical Study of the Five Collaborative Plays, Oxford: Oxford University Press. Wiggins, Martin and Catherine Richardson, eds (2012–), British Drama 1533–1642: A Catalogue, 10 vols, Oxford: Oxford University Press.

CHAPTER 3.2

Shakespeare and digital editions SONIA MASSAI

In 2010, Andrew Murphy reviewed the most distinctive features, achievements and limitations of three open-access internet editions of Shakespeare, which he believed were representative of how the digital medium had changed the editing of Shakespeare in the previous decade.1 In his concluding remarks, Murphy paid tribute to the scholarly value and innovative qualities of their best features: the concordance function in the Open Source Shakespeare (Johnson n.d.), the glossarial materials in Shakespeare’s Words (Crystal and Crystal n.d.) and the vision of The Internet Shakespeare Editions (Best 1996–), for presenting us with ‘what editors can now do with the electronic text’. Overall, Murphy felt that ‘[t]he best elements of these three sites [were] good indications of where new technologies [could] take us’ (414). My chapter evaluates how digital editing, and the digital editing of Shakespeare more specifically, have evolved since 2010, using Murphy as a starting point to gauge how far textual scholars and Shakespeare editors have come in the quest for what is generally known as a ‘born-digital’ edition of Shakespeare’s works. In this chapter I also consider Murphy’s less optimistic prediction that openaccess texts of Shakespeare’s works on the internet will continue to be ‘somewhat unsatisfactory in one way or another’ (413). Murphy was here alluding to the fact that open-access Shakespeare still depended on texts that were badly out of date. Most widely used among them was the Moby Shakespeare, which in turn derived from the Globe Shakespeare edition first published in 1864 (Clark and Wright), as part of the larger project that produced the first monumental edition of the Cambridge Shakespeare (Clark, Glover and Wright 1863–6). Murphy also predicted that we would have to wait and pay ‘for the latest incarnations of the big scholarly editions to be released in electronic forms’ (2010: 413) in order to have freshly edited texts of Shakespeare more widely available online. The scenarios outlined by Murphy have only partly come to pass. The New Oxford Shakespeare and the third edition of The Norton Shakespeare were both published in print and digital format between 2015 and 2017, although one instalment of The New Oxford Shakespeare is still forthcoming. Included in the price of the printed volumes of the Modern Critical Edition of The New Oxford Shakespeare and of The Norton Shakespeare are access codes to the online editions for individual users, although access to The New Oxford

SHAKESPEARE AND DIGITAL EDITIONS

245

Shakespeare Online expires twelve months after the first log in. However, Murphy had not foreseen a very encouraging development, namely the encoding of the text of The Folger Shakespeare Library Editions and their release as part of The Folger Digital Texts (Mowat et al. n.d.). Although The Folger Shakespeare Library Editions offer free access to the edited texts, but not to the editorial apparatus, the encoding allows search functionalities that enhance their usability. These new resources will be discussed in the first half of this chapter as examples of what the digital editing of Shakespeare has achieved since 2010. The second half of this chapter reflects on how theoretical thinking about the digital editing of Shakespeare has also changed over the last ten years. By considering examples of digital editing beyond Shakespeare, I will show how the scholarly community seems to have moved on from a state of generalized anxiety about the potentially boundless size of digital editions and the lack of a clear rationale for the digital editing of Shakespeare to a sustained interest in the productive interplay between print and the digital medium. Instead of worrying about what a genuinely born-digital edition will or should look like, about its potentially infinite capacity as a container of texts and resources, and the consequent lack of a coherent editorial perspective, editors of Shakespeare are starting to explore productive solidarities and continuities between print and digital media which can in turn enhance usability and versatility of both paper and online editions. From fearing that ‘technological possibilities are outstripping our editorial imagination’,2 scholars have come to think about print and digital technologies as stages in ‘the long continuum of mechanical production’ of textual artefacts and about the questions raised by modelling texts for (re)presentation in a digital environment ‘as a practice native to the arts and humanities, and not limited to digital computing’ (Galey and Niles 2017: 23, 25). The second half of this chapter therefore reflects on a current shift in attitudes towards digital editing and outlines desirable directions of travel for the foreseeable future. Before moving on to its two main parts, this chapter offers a brief historical overview of the growth of the critical thinking around digital editing within Shakespeare studies in order to contextualize the recent and foreseeable developments reviewed here.

*** Jerome McGann’s essay ‘The Rationale of Hypertext’ (1996) was a milestone in the advancement of critical thinking about the opportunities offered by the digital medium to textual scholarship. McGann argued that the digital medium served canonical literary works by Robert Burns, William Blake, Emily Dickinson or William Wordsworth better than the codex format. He polemically asked who would prefer a paper edition to an audial text of Burns’s ballads, or paper editions of Blake’s ‘Songs’ that generally do not include colour facsimiles of the relief etchings that illuminate his poems to a digital archive that can effectively represent both the textual and the visual elements of his works.

246

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

As intimated by its title, McGann’s essay challenged editors to revisit how they approached the editorial task. W. W. Greg’s ‘The Rationale of Copy-Text’, first published in 1950/1, had spelled out not only the imperative for editors to recover the version of the text that was closest to the author’s original intentions but had also formulated the principles of an influential methodology that limited subjective editorial choice. The title of McGann’s essay harked back to Greg’s rationale in order to challenge it. McGann’s approach to textual editing, also known as ‘the sociology of the text’, was founded on his belief that texts do not exist independently from the medium through which they are transmitted to their readers and that a multitude of agents other than the author, including annotating readers and editors who prepare the manuscript or typescript copy for the press, but also printers and publishers, contribute to its production, thus affecting the way in which texts signify. The advent of the digital medium, by allowing access to multiple versions of the same text, or even by displaying multiplicity within the same text by the visualization of animated variants, was encouraging editors to rethink the Gregian imperative that had shaped their approach to ancient and modern classics like Shakespeare up to the late twentieth century: The goals of classical scholarship and the material formalities of the book encouraged scholars to imagine and produce single-focus works – editions that organized themselves around what used to be called a ‘definitive’ text, the source and end and test of all the others. … Unlike traditional editions, ‘hyper’ editions need not organize their texts in relation to a central document, or some ideal reconstruction generated from different documents. (1996: 27–8) McGann in fact preferred to categorize the products of digital editing as critical archives, rather than editions, to emphasize not only their potentially infinite capacity but also all the functionalities and the multimediality that the codex cannot support. Shakespearean editors largely endorsed the enthusiasm with which McGann embraced the intellectual and practical opportunities offered by the new medium, though with some degree of caution and scepticism. David Scott Kastan agreed that the digital medium offered editors ‘an escape from the felt tyranny of the book’ (2001: 112), because the level of multiplicity and variation it supported exposed the elusive quality of authorial agency and intentions, and the ‘quixotic’ aspiration of editors who had traditionally set out to recover them (121). Others, though, sounded a note of warning, in stressing how the ability of the digital medium to display multiplicity does not make the editorial function redundant, but even more necessary than in the preparation of a paper edition. John Jowett, for example, warned that ‘it is easy to become entranced by the abstract possibilities of the electronic text … to the extent that the practical issues of editing seem about to vanish entirely. They will not. New editorial issues emerge, and they do not simply displace the old ones’ (1999: 79–80). I found myself agreeing with Jowett, when I first became interested in digital editing:

SHAKESPEARE AND DIGITAL EDITIONS

247

Multiplicity does not override the need to assess the relative textual authority of each available version. Although editors can now exploit the versatility of the electronic medium to present multiple textual authorities, they still need to make informed decisions on what complex textual evidence actually represents. (Massai 2004: 94) In the early noughties, editors also started to reflect on more pragmatic questions raised by the advent of digital editions. Michael Best, for example, reported concerns within the scholarly community about the ‘preservation of scholarly work in a medium that is still subject to rapid change and development’ (2007: 150). Even at the time of writing and in the wake of the publication of the major digital editions this chapter goes on to discuss, some editors hesitate to commit years of intense editorial labour to born-digital editions, because of their perceived impermanence. Best also addressed a related anxiety about the perceived lack of rigour associated with the medium: Academic standards are set by peer review, and it is inevitable that many academics will be skeptical of the quality of publication on the web: the happy proliferation of amateur sites … can be read as a signal that publication on the web is untried and untested by appropriate standards. Thus a further challenge to the development of scholarly editions on the web is that scholars working on them may lack adequate recognition in the currency of the profession. (2007: 151) This concern is thankfully on the wane: the commitment of publishers and scholars to digital editing in recent years has amply demonstrated that rigour and accuracy are not dependent on the medium but on the professionalism and credentials of those involved in digital projects. More recently, critical attention has focused on wider issues to do with the impact of digital editing on larger communities besides the editors themselves. Some of the essays collected in Christie Carson and Peter Kirwan’s Shakespeare and the Digital World (2014), for example, tackle the impact of digital editions on pedagogy and the current movement towards making refereed resources open access. Other scholars have honed in on how digital editing is changing our appreciation of specific features of Shakespeare’s works: Janelle Jenstad, Mark Kaethler and Jennifer RobertsSmith’s Shakespeare’s Language in Digital Media (2018), for example, shows what new questions can be asked and what new insights can be gained now that the Shakespearean lexicon can be accessed, visualized and studied electronically. The major digital editions I am going to discuss in the next section of this chapter are the product of the invigorating debate prompted by the rise of digital editing, which this brief overview can only begin to outline.3

*** Turning his notoriously exacting eye to the digital components of the New Oxford (Taylor et al. 2016–) and the Norton 3 (Greenblatt 2015b), Brian Vickers (2017) found key features worth praising in both editions:

248

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

I turned to the digital versions rather warily, as a man who has always loved his books. … [T]he New Oxford offers a page-for-page rendering of the print version, in a more legible text and with full quality illustrations. … [The Norton 3] has stolen a march on its rival by fully embracing the new possibilities of digital publication. … [T]his digital edition is going to be hard to beat. While I endorse Vickers’s appraisal of the main strengths of the digital versions of the New Oxford and the Norton 3, I believe that more can be said about the insights that both editions lend into how we are beginning to respond to the opportunities and challenges offered by the simultaneous publication of Shakespeare’s works in print and online. The New Oxford is vast. The print version develops across four large volumes. The Authorship Companion (Taylor and Egan 2017), a hefty collection of essays about canon and chronology, presents a review of current attribution methodologies as well as the evidence underlying the identification of Shakespeare’s co-authors, including Christopher Marlowe as co-author of the Folio version of all three Henry VI plays and Thomas Middleton as the author of the ‘fly-scene’ in Titus Andronicus and a handful of passages in All’s Well That Ends Well. The two-volume Critical Reference Edition (Taylor et al. 2017) for scholars offers detailed textual introductions to oldspelling versions of the texts of the plays and poems. Finally, the Modern Critical Editions (Taylor et al. 2016) comes with a lighter-touch apparatus meant to appeal to students, teachers, actors and directors and Shakespeare enthusiasts.4 Unusually, The New Oxford Shakespeare Online does not add to the material in the print version, but it does enhance access to it: instead of leafing through four large volumes, printed mostly in rather small print on thin, if soft and beautiful, paper, online users can switch between the different parts of the editions quickly and effortlessly. The digital texts of the plays and poems are flanked on the right-hand side of the text field by ‘Notes’, which, when clicked on, highlight the relevant line or word(s), and on the left-hand side by act and scene divisions, which allow users to navigate the primary texts. Accessing a play or a poem by selecting the ‘Critical Reference Edition’ tab on the top menu of the homepage opens a busier screen, where a fourth field offers additional resources called ‘Extras’ – mostly images of the printed pages preserved in multiple witnesses of the early modern edition used as source text. The layout is still elegant, however, and allows for a better reading experience than the printed volumes, whose sheer size and weight make them unwieldy. A little odd is the lack of active links in key sections of the digital edition. The introductions to the old-spelling versions of the plays and poems, for example, are headed by a list of cross references to the modernized texts, commentaries and reception histories in the Modern Critical Edition and to relevant sections in The Authorship Companion. What digital users might easily take to be active links are in fact simply handy reminders of where else useful information about any given play or poem can be found. Also unusual is the decision to attach a short expiry date to the access code that individuals are given on purchasing the printed volume of the Modern Critical Edition. While a twelve-month access code allows users to familiarize themselves with all the components of this monumental edition, those

SHAKESPEARE AND DIGITAL EDITIONS

249

who are not affiliated to large academic organizations that are able and willing to pay hefty institutional subscription fees lose access after a year to all textual notes and introductions, as well as to the rationale for inclusion or exclusion of individual texts from the canon. This policy, which might be justified from a marketing point of view, problematically categorizes readers of the printed Modern Critical Edition as predominantly interested in the texts of the plays, although the critical and performance traditions are also covered, if rather selectively, in this volume. It seems regrettable that, twelve months after they first log in, individual owners of the printed Modern Critical Edition lose access to the exceptional resources that the editors of The New Oxford Shakespeare have assembled. The Norton 3 adopts a more integrated approach to print and digital publication and fully embraces the opportunities offered by digital technologies to complement and expand the contents of its printed, single-volume edition. This digital edition offers additional resources that cannot be supported by print – most prominently, audio recordings of the sixty-six songs included in the plays from Ross Duffin’s Shakespeare’s Songbook and over eight hours of passages spoken by especially commissioned members of the Actors From The London Stage company. Even more crucially, the editors’ decision to publish online variant versions of the texts of the plays and poems included in the print edition is intellectually and logically justified by the ‘single-text editing’ rationale that informs the edition as a whole. Rejecting an older editorial method, initiated by editors of Shakespeare who chose to ‘conflate’ the variant readings in plays that have survived in multiple early modern printed witnesses, the Norton 3 opts instead for a ‘single-text’ approach, which the General Textual Editors, Gordon McMullan and Suzanne Gossett, sum up as follows: [W]e have in the case of plays that exist in significantly different texts printed the text that is most complete and apparently most finished … [without] claiming (as has been done in the past) to be able to determine and present the text that was Shakespeare’s ‘original’ version or his ‘final’ version, or the one that the company probably performed. … This often means the text in the First Folio …. But when … the Folio text is itself derived from a good quarto, we choose that earlier quarto as the base text from which our print edition is created. (Greenblatt 2015a: 85) While complete according to this principle, the print edition offers only edited and modernized versions of a single witness, even when there are multiple extant early modern printed incarnations of Shakespeare’s works, because, as the editors emphatically put it, they decided ‘to edit the text, not the work’ (ibid.).5 The digital edition logically complements the printed volume by offering users access to edited and modernized versions of the texts which, though falling short of what the editors regard as ‘the most complete and apparently most finished’ version of the works, still offer invaluable insights into the state of constant flux in which early modern plays existed – both on stage, where they were often updated and revised to suit revivals in a variety of venues and for different occasions, and on the page, when the most popular plays were prepared for the press in the process of being reprinted.

250

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

The digital edition also offers a wealth of edited documents to do with ‘Shakespeare and his Works’ and ‘The Theater Scene’. Some of these documents are well known and already freely available online, such as the front matter from the First Folio. Others, for example, an anonymous elegy written on the death of Richard Burbage or a ballad on the burning of the Globe, are probably unknown and inaccessible to most users and yet represent invaluable pedagogical resources and suggestive contextual evidence. Resources for instructors are downloadable as JPEGs or PowerPoint slides for classroom use from a separate web page on the Norton site (Instructor Resources n.d.). Last, but not least, while the digital edition includes no video clips of recent theatrical productions or adaptations for film and television, users are referred to a dedicated YouTube Norton Shakespeare channel, which gathers and curates video materials already available online (YouTube Norton Shakespeare n.d.). This is an elegant solution to the thorny problem of clearing copyright and paying hefty fees for video materials, which digital editions of Shakespeare could and should include but remain mostly out of bounds because of costs and licensing regulations. Another solution worth exploring would involve linking editions behind paywalls to wellestablished open-access resources that gather video materials, including live archives like the MIT ‘Global Shakespeare’ website (https://globalshakespeares.mit.edu) or even occasional, but institutionally funded and therefore stable platforms like ‘Performance Shakespeare 2016’ (http://performanceshakespeare2016.org). As Vickers points out, the digital edition of the Norton 3, far from merely complementing the print edition by offering additional resources online, really comes into its own when exploiting functionalities that are native to the digital medium. The digital edition can be searched via the table of contents or in its entirety, by selecting these two functions from a sidebar menu. The sidebar menu also includes a ‘Notebook’, which allows users to gather and save all passages they have highlighted or annotated. The Norton’s annotation tool, though, is a feature of the Norton Ebook Reader, access to which can only be purchased by users in the USA. Users elsewhere with an access code for the digital edition will be redirected to VitalSource, a platform that provides access to digital textbooks and course materials. When used in online mode, VitalSource supports all other functions, and includes all the audio files mentioned above and the possibility to compile flash cards.6 On both platforms, the text field takes up the central section of the screen, where wide margins enhance the reading experience. The presence of glosses and annotations is signalled by underlined text, while icons indicating the presence of images, audio files as well as longer Textual Notes and Performance Comments are discreetly added to the right-hand side of the text. All the content accompanying the text – including underlining alerting users to the presence of pop-up annotations and the symbols in the margin indicating the presence of notes or audio files – can be muted, so users can enjoy reading the text in a spacious and uncluttered format. The different ways in which the New Oxford and the Norton 3 use the digital medium – to complement or, more ambitiously, to enhance the experience of using the paper edition – shed light on interesting and opposing tensions generated by the publication of these two editions across two media. The Norton 3, for all its creative use of the digital medium, organizes its resources according to the same principles

SHAKESPEARE AND DIGITAL EDITIONS

251

as the best scholarly editions since the early eighteenth century. Users can undertake online the types of activities associated with reading and studying a scholarly edition of Shakespeare in print, including highlighting passages of interest and compiling pop-up annotations. They are also provided with a critical apparatus that offers an authoritative point of access to the Shakespearean canon. In other words, the Norton 3, although built and structured like a digital archive, retains the traditional apparatus of a scholarly print edition. As I explain elsewhere, a ‘critical edition is structured hierarchically and privileges the modern text over other textual alternatives, … [while] a critical archive provides accurate and searchable digital versions of the editions from which those textual alternatives derive’ (Massai 2004: 103). The Norton 3, while providing links to images of the early modern witness used as source text for the modern edition, relegates the early modern witness to the status of an additional resource. Of course, a critical archive also predetermines what kind of searches the user can perform, but an editor of a critical archive cannot control what pathways users opt to follow. The Norton 3 allows its users to find additional resources quickly and efficiently, but the links have a predetermined start and end point. Like editors of a scholarly edition in print, the editors of the Norton 3 decided what supplementary materials illuminate either a specific line or scene (when additional resources are linked to the texts of the plays and poems) or Shakespeare’s works as a whole (when additional resources are included in the appendices). The fairly linear quality of the user experience of the Norton 3 online replicates the mode of access supported by the codex format, which may explain why the Norton 3 appeals to a user like Brian Vickers, a man who, admittedly, ‘likes his books’. Conversely, while ostensibly transposing the printed volumes page by page online, thus prioritizing print over the digital medium, the New Oxford shows the impact of digital technologies more tangibly than the Norton 3. The Modern Critical Edition, for example, dispenses altogether with full introductions to the texts of individual plays. Readers of this printed volume are instead treated to a selection of images and bite-size quotes, which are arranged neither chronologically nor in any other immediately obvious order. The editors account for this choice as follows: Rather than impose our own favourite theories, methods, or interpretations, rather than wrap Shakespeare in a uniformity of our own making, we offer a representative verbal and visual montage, a constructed sampling of the expanding universe of responses to Shakespeare. You might think of this as tapas Shakespeare, offering a small taste of many different dishes, ‘infinite riches in a little room’. … Any single-volume Complete Works is a gateway edition; ours aspires to contain thousands of different gateways. (Taylor et al. 2016: iv) The ‘verbal and visual montage’ that prefaces the texts of the plays and poems in The Modern Critical Edition resembles the type of bite-size snippets of information produced by online searches. Of course, most of the quotes and images included in this edition are not freely available online and were carefully selected by experienced editors. But their format and layout, especially when compared to the types of

252

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

introductions that readers of scholarly editions of Shakespeare in print are familiar with, reproduce the experience of reading and researching Shakespeare online, which might start to explain why Vickers responded so strongly to this feature: ‘I have an idea for a new circle of hell, where all these thousands of quotations (“gateways”) are on a continuously playing tape, a cacophony of random opinions’ (2017). Also worth noting is how this type of ‘verbal and visual montage’ makes Vickers hark back nostalgically to the specially written introductions to one-volume editions, some of which have, according to him, ‘permanent value’ (2017). Despite being published in print by a leading academic press, Vickers attributes to The Modern Critical Edition a quality of randomness and impermanence that is still more readily associated with digital resources. All in all, the New Oxford departs from how the print and the digital medium have been used so far to represent Shakespeare editorially: while, at least in one important respect The Modern Critical Edition mimics how information is accessed online, the digital edition exploits none of the functionalities generally associated with online resources. This policy must stem from the fact that the target readership for The Modern Critical Edition is expected to include students and younger readers who are born-digital, while the digital edition reproduces the codex format as closely as possible because it is mostly aimed at scholars who instinctively trust print and gravitate towards it. As a result, this edition rethinks the conventions associated with print in light of the functionalities made available by digital technologies, and vice versa. Returning briefly to Murphy, it is satisfying to see not only how much digital editions of Shakespeare have advanced over the last ten years, but also that established scholarly editions, like the Norton 3, have not simply released edited texts in electronic form. Of course, the Norton 3 is a commercial enterprise, as opposed to the openaccess resources Murphy reviewed in 2010. However, Murphy’s assumption that ‘free Shakespeare on the Internet’ would always prove ‘unsatisfactory in one way or another’ (413) has been at least partly (and most fortunately) contradicted by the Folger Shakespeare Library’s commitment to making many of their resources, including the texts of the Folger Shakespeare Library Editions, publicly available on open-access digital platforms. The Folger Digital Texts present in an elegant, unfussy format the texts of the Folger Shakespeare Library Editions, as originally edited by Barbara Mowat and Paul Werstine. The site advertises the fact that users are granted access to ‘free, high-quality digital texts of Shakespeare’s plays … meticulously edited on the basis of current scholarship’. A brief synopsis of the play flanks the central text field, while several search boxes encourage users to navigate the text by ‘Quick Jumps’, by typing in act/scene/line divisions or page numbers, or by typing individual words or phrases in the ‘Search this Text’ box. The results of the word search appear directly underneath the ‘Search this Text’ box, embedded in the speech(es) or exchange(s) in which they occur. The search word or phrase is a highlighted hyperlink, which allows users to toggle between the list of results and the texts of the plays and poems. The functionalities added to the texts of the Folger Shakespeare Library editions are not as extensive as those made available by the Norton 3, and users still need

SHAKESPEARE AND DIGITAL EDITIONS

253

to buy the print editions if they wish to access the textual and critical apparatus produced by Werstine and Mowat. But the Folger has made this resource freely available, and the digital texts are eminently useable. Besides adding ‘sophisticated coding that works behind the scenes to make the plays easy to read, search, and index’ and laying ‘the groundwork for new features in the future’ (Mowat et al. n.d., ‘About Us’), the Folger has also made it possible for users to download the encoded texts, thus gifting the scholarly community with a formidable resource, which is available free of charge for non-commercial purposes.7 It is possible to detect in the care and technical sophistication that inform this digital edition the impetus, expertise and ethos that Eric Johnson brought with him when he moved to the Folger to head its Digital Team. In a paper that predates his move to the Folger and reads like a manifesto for the advancement of the digital editing of Shakespeare, Johnson was already envisaging the sort of edition made available by the digital release of the texts of the Folger Shakespeare Library Editions, but he was despairing that it would ever find an institutional sponsor enlightened enough to support it: Perhaps someday, a group of individuals will produce a modern, scholarly, free alternative to Moby Shakespeare. The deck is stacked against it, however. For one thing, the amount of labor involved in producing this critical edition of the text would be huge. … One would hope that some publisher somewhere would make its text, if not free, at least more widely available online. (Johnson n.d.) Like Murphy, Johnson was mostly right in the way he predicted the development of digital Shakespeare. The Internet Shakespeare Editions has progressed since Murphy reviewed it in 2010, but slowly, because, as Johnson explains, it relies on a group of individuals intent on producing a freshly edited version of Shakespeare’s works for free internet access. Despite generous subsidies from funding bodies in Canada and beyond, The Internet Shakespeare Editions exemplifies the hurdles faced by an initiative that quite rightly prides itself on being entirely non-commercial. But Johnson was also thankfully right when he hoped, and merely hoped at that stage, that ‘some publisher somewhere’ would step up and share its editions of Shakespeare with larger communities of users. The Folger evidently managed to negotiate an agreement with the series’ original publisher, Simon & Schuster, for the release of the texts as an open-access resource. My prediction is that more publishers will follow suit over the next ten years, releasing more exciting products, either as openaccess resources or behind paywalls, consolidating and building on the achievements reviewed in the first half of this chapter.

*** The second half of this chapter considers how our approaches to editing Shakespeare are not only effectively supported but also challenged in exciting and creative ways by the digital medium. While editions like the New Oxford and the Norton 3 show how the digital medium has started to impact on their dissemination, on the range

254

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

of content and functionalities that they offer, and, most interestingly, even on the organization of their apparatus in print, both editions are firmly anchored in the principles and type of editorial agency that have informed the evolution of the editorial tradition so far. Among the most powerful organizing principles that have shaped the editing of Shakespeare in print to date, the play is probably the most influential. The play constitutes both the fundamental unit of dramatic signification through which the canon is still accessed by both readers and online users and the organizing principle for the division of the editorial labour invested in the production of new editions. Similarly, chronology, a central preoccupation for Shakespeareans since Edmond Malone,8 has recently determined the order in which the plays are presented to readers (and, sometimes, to online users too, as in the Norton 3). The critical apparatus, though discreetly relegated to parallel windows or easily muted, retains online all the component parts and traditional functions accrued in print since the rise of the editorial tradition in the early eighteenth century. Most crucially, the editor, and by ‘the editor’ I mean both the general editors and the editors of individual plays within complete works, is still the ultimate guarantor and progenitor, along with Shakespeare (and his collaborators), of authoritative texts and their intended meaning. None of these categories, principles or models of editorial agency, which online editions have inherited from print and have by and large reproduced, pertains to the rise of Shakespeare in print. Plays in Shakespeare’s time existed as ‘an accumulation, or a meeting, of numerous separate parts’ (Palfrey and Stern 2007: 2), and not only actors’ parts but also other parts, such as prologues and epilogues, or letters and songs, which existed independently from the play or co-existed with it only provisionally, as ‘temporary visitors’ (Stern 2009: 151). A preoccupation with chronology also postdates the original conditions of theatrical and print production in Shakespeare’s time, stemming from a later tendency to think of ‘Shakespeare’ as coinciding with his lifework and his lifework as a canon, that is as a single, organic body of works (de Grazia 2014). It was only in the eighteenth century that biographies and other parts of the critical apparatus started to grow around the texts of Shakespeare’s works to illuminate and clarify Shakespeare’s authorial intentions. Around the same time, and through the establishment of the critical apparatus, editors took upon themselves the role of authoritative interpreters of Shakespeare, as opposed to the first anonymous editors of Shakespeare, who aimed to ‘perfect’ rather than ‘recover’ the meaning of his words (Massai 2007). In short, the material and intellectual contexts within which Shakespeare was first transmitted into print produced textual artefacts that are both materially and conceptually different from the types of printed editions modern readers of Shakespeare have become familiar with. Taking my cue from Alan Galey and Rebecca Niles’s analysis of ‘the long continuum’ (2017: 23) of the mechanical production and reproduction of Shakespeare over the last 400 years, I would therefore like to consider whether there might in fact be more meaningful continuities between textual cultures in Shakespeare’s time and our own time than between early modern playbooks and the modern scholarly editions through which they are still predominantly re-presented to modern readers.

SHAKESPEARE AND DIGITAL EDITIONS

255

Pioneering experiments with the digital editing of other early modern literary authors and their corpora have begun to highlight some of these meaningful continuities. An enlightening example is A Social Edition of the Devonshire Manuscript (BL MS Add 17,492), first published as a Wikibook in 2015 (Siemens et al. n.d.). What sets this edition apart from earlier print editions that had solely focused on the poems attributed to Thomas Wyatt is its editors’ commitment to ‘the socially mediated textual and extra-textual elements of the manuscript’ as a whole and their commitment to the digital medium as more immediately suited to the task at hand than print: We believe that the physical and social elements of the Devonshire Manuscript lend themselves to digital editing and publication processes that more readily represent these aspects than a print environment can. A Social Edition of the Devonshire Manuscript focuses on the editorial and scribal practices that inform the context and production of the Devonshire Manuscript. By shifting our own editorial process into an environment representative of the inherent sociality of texts, A Social Edition of the Devonshire Manuscript hearkens back to the multiauthor roots of the text itself. (Siemens et al. 2016) Most surprising about this statement is how the functionalities offered by a digital platform like Wikibooks are believed to help editors re-present the original conditions of material production and the paleographical features of a complex textual artefact like the Devonshire Manuscript more closely than print ever could. The homepage of this edition lists sections immediately recognizable to users of printed scholarly editions of early modern texts: acknowledgements are followed by ‘A Study of the Devonshire Manuscript’, which covers topics generally addressed by scholarly introductions included in single-volume editions of literary texts; the content of ‘The Devonshire Manuscript, British Library, Add. MS 17,492’, is in turn followed by two sections that function like Appendices, one devoted to explaining how the text of the manuscript and all other parts of the critical apparatus were marked up in TEI P5, the other alerting users to the existence of other related, external resources. The similarities with printed scholarly editions of early modern texts, however, stop here: clicking on ‘The Devonshire Manuscript, British Library, Add. MS 17,492’ opens a drop-down menu of active links to individual pages, and each page contains a wealth of resources, including a diplomatic transcription of the text and a facsimile of the page, which, when clicked on, takes the user to a media viewer which allows the image to be enlarged, downloaded, shared or embedded in another file. Both the short notes to the right of the transcription of the text and the longer commentary note underneath it have active links that take users to other relevant sections of the edition, including a list of paleographic features, which reproduces and explains common abbreviations, omissions and variant inscriptions of individual characters, and a list of the hands that transcribed or penned the poems included in the manuscript. The commentary note also alerts users to the presence of other annotating hands, which sometimes comment, sometimes edit the text of the poem. The multiple layers of transcription and annotation of the poems are therefore clearly represented in a digital environment that allows users to navigate in and out of individual texts, without losing sight of the original artefacts.

256

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Even more crucially, this edition mirrors the collaborative nature of its source text because it presents the labour undertaken by a large editorial team as process rather than product. The Wikibook format of this edition allows for three modes of use: users can ‘read’, ‘edit’ and ‘view’ the history of how each individual section of the edition has been prepared. The edition is therefore not only in progress but invites members of the scholarly community, as well as what the editorial team refer to as ‘citizen editors’, to continue the work done to date. As the editorial team explain: The discussion sections on each page, a feature unique to Wikimedia projects, promote conversation on various aspects of the poem at hand. In this way, the Wikibook edition extends the social context of the Devonshire Manuscript by providing a space for ongoing discussion, collaboration and negotiation. … We produced an edition in Wikibooks that is scholarly and peer reviewed in a traditional sense, but also enables citizen scholars to access, contribute and annotate material. (Siemens et al. 2016) The model of ‘open scholarship’ enabled by digital platforms like Wikibooks in turn forces us to reconsider the role of the scholarly editor: Technological advances potently shape how individuals and communities create new knowledge. As such, it behoves scholars to think through the affordances and implications of any collaborative publishing platform, space for social knowledge creation or multi-authored environment. Incorporating social media allowances and Web 2.0 practices into a scholarly edition recasts the primary editor as a facilitator, rather than progenitor, of textual knowledge creation. (Siemens et al. 2016) By attending to the productive continuities between early modern manuscript and late modern digital cultures, the editors of The Devonshire Manuscript, British Library, Add. MS 17,492 effectively show the benefits of rethinking not only the rationale that informs the critical apparatus in printed scholarly editions of early modern literary texts, but also the role of the editor. It is therefore imperative to wonder how the ‘open scholarship’ and the role of the ‘open editor’ supported by digital platforms now available to scholars and to users alike can help us think critically and creatively about how we choose to represent Shakespeare’s works editorially. What are the main features and the material conditions of production associated with early modern printed (play) books which could or should inform the digital editing of early modern (play)texts? Rather than replicating the critical apparatus of printed scholarly editions of early modern texts or attempting to establish what born-digital editions of early modern texts could or should look like, we might more productively focus on identifying significant continuities and solidarities between the early modern printed object and the digital affordances through which we can now represent it. In other words, what would a digital edition of Shakespeare look like, were it to be prepared according to the principles that have informed the editing of The Devonshire Manuscript, British Library, Add. MS 17,492?

SHAKESPEARE AND DIGITAL EDITIONS

257

I have explained elsewhere how attending to the processes through which early modern plays were collaboratively composed, disassembled and reassembled into provisional congregations of parts on stage and on the page, and then corrected, revised and ‘augmented’ for stage revivals or resubmission to the press, could help us decide how best to use the print and digital resources at our disposal (Massai 2018). My suggestion that Shakespeare could be profitably edited ‘in parts’, including and beyond the play as the main dramatic and editorial unit through which scholarly editions are organized, stemmed from the realization that there are productive continuities between how early modern playbooks came into being and how digital technologies organize content and the labour that goes into its organization. Chronology, which, as mentioned above, is both a concern and a construct that postdates Shakespeare’s own time and is only one of the possible organizing principles of a complete-works edition, especially when prepared and published as a digital archive, should similarly be used alongside alternative ways of arranging its contents. And how might one decide to restructure the critical and textual apparatus, in light of the types of expertise that the expanded apparatus supported by a digital environment requires? Answering these questions requires flexing ‘our editorial imagination’ (Galey and Niles 2017: 23). But rather than trying to imagine a born-digital edition that radically departs from how editors have reproduced Shakespeare for the printed page, it seems more profitable to think of ways in which a digital edition can enhance our appreciation of the features of the textual object it aims to reproduce. In this respect, it is important to stress how even a ground-breaking edition like that of The Devonshire Manuscript, British Library, Add. MS 17,492 exists not only as a digital resource that exploits the opportunities and functionalities made available by Web 2.0 but also as the product and producer of print artefacts. Not only does Wikibooks support a book format for the organization of digital materials, but it also functions as a complement to print: besides the online edition, the editorial team also produced a ‘static’ digital edition of the manuscript, which served as a control text ‘against which [the] international advisory group … could compare the Wikibook edition as it evolved’ and will serve as the basis for ‘a final … edition … for print and e-publication with Iter and Medieval and Renaissance Texts and Studies (MRTS)’ (Driscoll and Pierazzo 2016). My prediction for the foreseeable future is that the most exciting and productive developments in the digital editing of Shakespeare and early modern drama might likewise stem from a creative interplay between media rather than from a shift from print to digital technologies. The editions reviewed in this chapter suggest that thinking about the two media not in evolutionary terms but as profitably interacting with each other relieves editors from striving to reimagine an early modern textual artefact in ways that forget and override the conditions, contexts and practices that informed its material origins.

NOTES 1.

I am grateful to Sarah Rose Aquilina, Digital Media Editor for Literature at W. W. Norton & Company, and Rupert Mann, Head of the Editorial Digital Development

258

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

team for the Oxford Scholarly Editions Online (OSEO), who facilitated access to the online Shakespeare editions hosted by Norton and by Oxford University Press. Special thanks to Aquilina for giving me a live demonstration of the main features of the Norton 3 edition. 2.

Ryck Rylance, in conversation, at ‘What is a Text in the Digital Age I’ symposium, organized by the Institute of English Studies, at the School of Advanced Studies, University of London and the School of English at the University of Newcastle, and held at Senate House, University of London, on 1 July 2017.

3.

There is regrettably no space to consider how the digital editing of literary works by contemporaries of Shakespeare has enriched this critical debate. Laura Estill has recently written about the way in which the cultural capital accrued by Shakespeare grants more prominence to critical thinking and experimental editorial practices connected with the editorial reproduction of Shakespeare’s works for the digital medium, at the expense of both theoretical and practical work carried out by scholars who have a more sustained investment in works by his contemporaries (2019). It is in this spirit that I want to mention at the very least some of the most interesting, successful and thought-provoking projects which have made available digitally works by early modern authors beyond Shakespeare: The Cambridge Edition of the Works of Ben Jonson Online (Bevington et al. 2012), Richard Brome Online (Cave 2010), Digital Renaissance Editions (Greatley-Hirsch n.d.), and the Queen’s Men Editions (Ostovich et al. n.d.).

4.

Two further volumes, with The Complete Alternative Versions in original and in modern spelling, are still forthcoming at time of writing (Taylor et al. forthcoming). The plan for these volumes is to continue to explore the questions already considered in The Authorship Companion and The Critical Reference Source about editorial approaches, canon and chronology and book history. Like the volumes published so far, this final instalment of The New Oxford Shakespeare will be released by simultaneous print and digital publication.

5.

A notable exception is King Lear, which started to be presented to readers in paralleltext or multiple-text editions in the 1980s and is offered to readers of the Norton 3 both on facing pages for an effective, visual rendering of the variants in the 1608 Quarto and the 1623 Folio texts and as a ‘combined’ text, which acknowledges the cultural capital and influence accrued by the conflated text of the play. The use of the term ‘combined’, as opposed to ‘conflated’, its more traditional and established counterpart, is meant to alert readers to the fact that, even in the single text of this play, the editors have adopted a ‘“scars-and-stitches” solution’ (Greenblatt 2015a: xxvi), by using the Folio text as their base text and by indenting, numbering and printing in a slightly different font the quarto-only passages embedded in it. The same approach is used for Hamlet, for which the second quarto of 1604/5 is used as base text. While the first quarto of 1603 is also included in the print edition in its entirety, due to a recent increase in both scholarly and theatrical explorations of its uniqueness, the text of the second quarto includes Folio-only passages that look typographically extraneous to it.

SHAKESPEARE AND DIGITAL EDITIONS

259

6.

Sarah Rose Aquilina has confirmed that the functionalities of the Norton edition were set up and fine-tuned after usability tests revealed how target users were interacting with it and will be developed in light of feedback that the Norton Digital Media Team continue to receive, mostly from instructors and their students.

7.

One spectacular example of how this resource is already enhancing research in the field is the JSTOR Understanding Shakespeare Series (https://www.jstor.org/ understand/shakespeare), which links individual lines from the Folger Digital Texts to scholarship in JSTOR that quotes those lines, thus showing users, at a glance, what passages from Shakespeare have attracted most critical attention.

8.

Malone was the first to include a chronology in his edition of The Plays and Poems and the Malone-James Boswell revised edition of 1821 was the first to arrange the works chronologically, except for the English history plays, for which the royal order introduced by the First Folio was retained (de Grazia 1991; forthcoming).

REFERENCES Best, Michael, coordinating ed. (1996–), Internet Shakespeare Editions. Available online: https://internetshakespeare.uvic.ca/ (accessed 13 November 2020). Best, Michael (2007), ‘Shakespeare and the Electronic Text’, in Andrew Murphy (ed.), A Concise Companion to Shakespeare and the Text, 145–61, Oxford: Blackwell. Bevington, David, Martin Butler and Ian Donaldson, gen. eds (2012), The Cambridge Edition of the Works of Ben Jonson Online. Available online: https:// universitypublishingonline.org/cambridge/benjonson/ (accessed 13 November 2020). Carson, Christie and Peter Kirwan, eds (2014), Shakespeare and the Digital World: Redefining Scholarship and Practice, Cambridge: Cambridge University Press. Cave, Richard, ed. (2010), Richard Brome Online. Available online: https://www.dhi. ac.uk/brome/ (accessed 13 November 2020). Clark, W. G. and W. A. Wright, eds (1864), The Globe Shakespeare, Cambridge and London: Macmillan. Clark, W. G., John Glover and W. A. Wright, eds (1863–66), The Works of William Shakespeare, 9 vols, Cambridge and London: Macmillan. Crystal, David and Ben Crystal (n.d.), Shakespeare’s Words. Available online: https:// www.shakespeareswords.com/ (accessed 13 November 2020). de Grazia, Margreta (1991), Shakespeare Verbatim: The Reproduction of Authenticity and the 1790 Apparatus, Oxford: Clarendon Press. de Grazia, Margreta (2014), ‘Shakespeare’s Timeline’, Shakespeare Quarterly, 65: 379–98. de Grazia, Margreta (forthcoming), Four Shakespearean Period Pieces, Chicago: Chicago University Press. Driscoll, Matthew and Elena Pierazzo, eds (2016), Digital Scholarly Editing: Theories and Practices, Cambridge: Open Book Publishers. Available online: http://books. openedition.org/obp/3381 (accessed 13 November 2020). Estill, Laura (2019), ‘Digital Humanities’ Shakespeare Problem’, Humanities, 8 (1): 45. Available online: https://doi.org/10.3390/h8010045 (accessed 13 November 2020).

260

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Galey, Alan and Rebecca Niles (2017), ‘Moving Parts: Digital Modeling and the Infrastructures of Shakespeare Editing’, Shakespeare Quarterly, 68: 21–55. Greatley-Hirsch, Brett, coordinating ed. (n.d.), Digital Renaissance Editions. Available online: https://digitalrenaissance.uvic.ca/ (accessed 13 November 2020). Greenblatt, Stephen, gen. ed. (2015a), The Norton Shakespeare, 3rd edn, New York: Norton. Greenblatt, Stephen, gen. ed. (2015b), The Norton Shakespeare Digital Edition. Available online: https://digital.wwnorton.com/shakespeare3; https://bookshelf.vitalsource. com/#/ (accessed 13 November 2020). Instructor Resources: The Norton Shakespeare (n.d.). Available online: https://www. wwnorton.com/instructor-resources/9780393934991 (accessed 13 November 2020). Jenstad, Janelle, Mark Kaethler and Jennifer Roberts-Smith, eds (2018), Shakespeare’s Language in Digital Media: Old Words, New Tools, London: Routledge. Johnson, Eric M. (n.d.), Open Source Shakespeare. Available online: http://www. opensourceshakespeare.org/ (accessed 13 November 2020). Jowett, John (1999), ‘After Oxford: Recent Developments in Textual Studies’, in W. R. Elton and J. M. Mucciolo (eds), The Shakespearean International Yearbook, Vol. 1: Where are We Now in Shakespearean Studies, 65–86, Aldershot: Ashgate. Kastan, David Scott (2001), Shakespeare and the Book, Cambridge: Cambridge University Press. Massai, Sonia (2004), ‘Scholarly Editing and the Shift from Print to Electronic Cultures’, in Lukas Erne and Margaret Jane Kidnie (eds), Textual Performances: The Modern Reproduction of Shakespeare’s Drama, 94–108, Cambridge: Cambridge University Press. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Massai, Sonia (2018), ‘Editing Shakespeare in Parts’, Shakespeare Quarterly, 68: 56–79. McGann, Jerome J. (1996), ‘The Rationale of HyperText’, Text, 9: 11–32. Mowat, Barbara, Paul Werstine, Michael Poston and Rebecca Niles, eds (n.d.), Shakespeare’s Plays, Sonnets and Poems: Folger Digital Texts. Available online: www. folgerdigitaltexts.org (accessed 13 November 2020). Murphy, Andrew (2010), ‘Shakespeare Goes Digital: Three Open Internet Editions’, Shakespeare Quarterly, 61: 401–14. Ostovich, Helen, Peter Cockett and Andrew Griffin, gen. eds (n.d.), Queen’s Men Editions. Available online: https://qme.uvic.ca/ (accessed 13 November 2020). Palfrey, Simon and Tiffany Stern (2007), Shakespeare in Parts, Oxford: Oxford University Press. Siemens, Raymond, et al., eds (n.d.), A Social Edition of the Devonshire Manuscript (BL MS Add 17,492). Available online: https://en.wikibooks.org/wiki/The_Devonshire_ Manuscript (accessed 13 November 2020). Siemens, Ray, Constance Crompton, Daniel Powell, Alyssa Arbuckle, Maggie Shirley and Devonshire Manuscript Editorial Group (2016), ‘Building A Social Edition of the Devonshire Manuscript’, in Matthew Driscoll and Elena Pierazzo (eds), Digital Scholarly Editing: Theories and Practices. Available online: http://books.openedition. org/obp/3409 (accessed 13 November 2020).

SHAKESPEARE AND DIGITAL EDITIONS

261

Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Taylor, Gary and Gabriel Egan, eds (2017), The New Oxford Shakespeare Authorship Companion, Oxford: Oxford University Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2016), The New Oxford Shakespeare: The Complete Works: Modern Critical Edition, Oxford: Oxford University Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, eds (2017), The New Oxford Shakespeare: The Complete Works: Critical Reference Edition, 2 vols, Oxford: Oxford University Press. Taylor, Gary, John Jowett, Terri Bourus and Gabriel Egan, gen. eds (2016–), The New Oxford Shakespeare (Digital). Available online: https://www.oxfordscholarlyeditions. com/nos/ (accessed 13 November 2020). Taylor, Gary, Terri Bourus and Gabriel Egan, gen. eds (forthcoming), The New Oxford Shakespeare: Complete Alternative Versions, 2 vols, Oxford: Oxford University Press. Vickers, Brian (2017), ‘Too too solid: Brian Vickers on two attempts to collect the complete Shakespeare’, The Times Literary Supplement, 19 April. Available online: https://www.the-tls.co.uk/articles/public/norton-new-oxford-shakespeare/ (accessed 13 November 2020). YouTube Norton Shakespeare (n.d.). Available online: https://www.youtube.com/channel/ UCbtVyjUXL0-ANtMljuf3OXA (accessed 13 November 2020).

262

PART FOUR

Material for further research

264

CHAPTER 4.1

Chronology ALAN B. FARMER

This chronology provides an overview of significant moments in the history of Shakespearean textual criticism. There are several critical principles that underlie its selection of publications. For the early modern period, the chronology focuses on the earliest publications of Shakespeare, highlighting several aspects of the printing of his plays and poems: the popularity of Shakespeare in print; the initial publication of Shakespeare’s plays without his name followed by the increasing prominence of Shakespeare as an author; the prevalence with which Shakespeare was advertised as having revised his plays; the occasional printing of apocryphal plays and poems that are not now considered to have been written by Shakespeare; and, perhaps most important, the printing of Shakespeare’s plays in different versions, especially in the First Folio. For the eighteenth century, this chronology emphasizes the gradual development of modern editorial procedures, especially as editors sought to make sense of the different versions of Shakespeare’s plays. The nineteenth and twentieth centuries are represented both by significant editions of Shakespeare’s plays and by foundational studies in book history and bibliography, in theatre history and in editorial theory. Several studies have been particularly useful in compiling this chronology, and readers wishing to learn more about this topic are encouraged to consult them. For eighteenth-century editions of Shakespeare, see Margreta de Grazia, Shakespeare Verbatim: The Reproduction of Authenticity and the 1790 Apparatus (Oxford: Clarendon, 1991). On the editing of Shakespeare in the nineteenth century, see Zachary Lesser, Hamlet After Q1: An Uncanny History of the Shakespearean Text (Philadelphia: University of Pennsylvania Press, 2015). The editing of Shakespeare in the twentieth century is analysed in detail by Gabriel Egan, The Struggle for Shakespeare’s Text: Twentieth-Century Editorial Theory and Practice (Cambridge: Cambridge University Press, 2010). Finally, for a history of printed editions of Shakespeare from the sixteenth century to the twentieth century, see Andrew Murphy, Shakespeare in Print: A History and Chronology of Shakespeare Publishing (Cambridge: Cambridge University Press, 2003). 1593 1st edition of Venus and Adonis (STC 22354) Printed in sixteen editions by 1636, plus another edition in 1675. Although lacking an author attribution on its title page, this edition includes a dedication

266

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

by Shakespeare to Henry Wriothesley, Earl of Southampton, one of only two dedications written in Shakespeare’s own voice. 1594 1st edition of Titus Andronicus (STC 22328) Printed in three quarto editions by 1611, all without an author attribution to Shakespeare, and afterward in the Shakespeare folios. 1st edition of The First Part of the Contention of the Two Famous Houses of York and Lancaster (STC 26099) Printed in three quarto editions by 1619, and thereafter in the Shakespeare folios, in which it is retitled The Second Part of Henry the Sixth. The first two editions do not name an author, but the third edition advertises the play as ‘Written by William Shakespeare, Gent.’ (1619). 1st edition of The Rape of Lucrece (STC 22345) Printed in eight editions by 1632, plus a further one in 1655. This edition lacks a title-page author attribution, but it again includes a dedication by Shakespeare to the Earl of Southampton. 1595 1st edition of The True Tragedy of Richard Duke of York (STC 21006) Printed in three octavo and quarto editions by 1619, and afterward in the Shakespeare folios, in which it is retitled The Third Part of Henry the Sixth. The first two editions do not name an author, but the third edition advertises the play as ‘Written by William Shakespeare, Gent.’ (1619). 1596 1st edition of Edward the Third (STC 7501) Printed in a second quarto edition in 1599. Both editions were published without an authorship attribution, but the play has recently been considered by more and more scholars to have been collaboratively written by Shakespeare and another dramatist. (See Figure 4.1.1.) 1597 1st edition of Richard the Second (STC 22307) Printed in six quarto editions by 1634, as well as in the Shakespeare folios. This edition does not name an author, but the second edition, printed in 1598, adds the attribution, ‘By William Shake-speare.’ 1st edition of Richard the Third (STC 22314) Printed in eight quarto editions by 1634, and in the Shakespeare folios. This edition does not name an author, but the second edition, printed in 1598, claims the play is ‘By William Shake-speare.’ 1st edition of Romeo and Juliet (STC 22322) Printed in five quarto editions by 1637, and in the Shakespeare folios. The first three editions do not include an author attribution, but the second issue of the fourth edition (1623), claims it was ‘Written by W. Shake-speare.’ This

CHRONOLOGY

267

FIGURE 4.1.1  Title page of King Edward the Third (Q1, 1596). Folger Shakespeare Library, call number: STC 7501. Used by permission of the Folger Shakespeare Library under a Creative Commons Attribution-ShareAlike 4.0 International License.

edition would later be described as a ‘bad quarto’ (1909) because it contains a much different and much shorter text than those in the later quartos (1599) and in the folios. 1598 1st and 2nd editions of Henry the Fourth (STC 22279a–22280) Printed in nine quarto editions by 1639, and in another quarto in 1700, as well as in the Shakespeare folios, in which it is retitled The First Part of Henry the Fourth. The first edition of this play exists in a fragment of only four leaves while the title page of the second edition does not name an author. 1st extant edition of Love’s Labour’s Lost (STC 22294) Reprinted in 1631, as well as in the Shakespeare folios, but there may have been an earlier edition that is now lost. The title page for this edition advertises

268

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

both that it was ‘Newly corrected and augmented By W. Shakespere’, and that it was performed at court, ‘As it was presented before her Highnes this last Christmas.’ This edition, along with the reprints of Richard the Second and Richard the Third published in 1598, marks the first time Shakespeare is named as an author in a printed playbook, and the first time he is identified as having revised a play. (See Figure 4.1.2.) 1599 1st and 2nd editions of The Passionate Pilgrim (STC 22341.5–22342) Printed in three editions by 1612. Although the first extant title page of this collection claims it was ‘By W. Shakespeare’, the volume actually contains poems by several authors. The first edition is known only from a fragment of eleven leaves and does not include its title page; its publication has been dated to around 1599, which is when the second edition was printed. 2nd edition of Romeo and Juliet (STC 22323) This edition contains a text that substantially differs from the first edition, including over 800 new lines (1597). It was advertised on its title page as ‘Newly corrected, augmented, and amended’, which was repeated in all the subsequent quartos, but it does not name an author. (See Figure 4.1.3.)

FIGURE 4.1.2  Title page of Love’s Labour’s Lost (Q1, 1598). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

CHRONOLOGY

269

FIGURE 4.1.3  Title page of Romeo and Juliet (Q2, 1599). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

3rd edition of Henry the Fourth, Part One (STC 22281) The title page of this edition both attributes the play to Shakespeare and claims the play has been revised: ‘Newly corrected by W. Shake-speare.’ The text of the play, however, only adds a few new stage directions and speech prefixes. All the subsequent quartos would continue to advertise the play as ‘newly corrected’ by Shakespeare. 1600 1st edition of Henry the Fifth (STC 22289) Printed in three quarto editions by 1619, all without an author attribution, and thereafter in the Shakespeare folios. This edition would later be described as a ‘bad quarto’ (1909) because it contains a version of the play that is about 1,600 lines shorter than that in the Shakespeare folios. 1st edition of Henry the Fourth, Part Two (STC 22288–22288a) This playbook was not reprinted, but the play was included in the Shakespeare folios. This edition exists in two states. It was initially printed without the scene identified in modern editions as 3.1, but, as the edition was still being printed, new leaves were added containing the scene. (See Figure 4.1.4.) 1st edition of Much Ado About Nothing (STC 22304) Not reprinted in quarto, this play was next published in the Shakespeare folios. (See Figure 4.1.5.)

270

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 4.1.4  Title page of Henry the Fourth, Part Two (Q1, 1600). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

FIGURE 4.1.5  Title page of Much Ado About Nothing (Q1, 1600). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

CHRONOLOGY

271

1st edition of A Midsummer Night’s Dream (STC 22302) Reprinted in 1619, and afterward in the Shakespeare folios. 1st edition of The Merchant of Venice (STC 22296) Printed in three quarto editions by 1637, as well as in the Shakespeare folios. Copies of the third edition, originally published in 1637, were reissued in 1652. 1601 1st edition of ‘Let the bird of loudest lay’ (in STC 5119). This untitled poem of sixty-seven lines was printed in the final section of a book of poetry by Robert Chester, Love’s Martyr: or, Rosalin’s Complaint. The poem is attributed in the edition to ‘William Shake-speare’, and it is grouped among several poems on the subject of ‘The Phoenix and the Turtle’, which is how the poem is sometimes identified. This edition of Chester was reissued under a new title in 1611, and the poem itself was reprinted in Shakespeare’s Poems (1640). 1602 1st edition of The Merry Wives of Windsor (STC 22299) Printed in three quarto editions by 1630, and in the Shakespeare folios. This edition would later be described as a ‘bad quarto’ (1909) because it contains a version of the play that is around 1,100 lines shorter than that in the Shakespeare folios. 3rd edition of Richard the Third (STC 22316) The title page of this edition claims the play has been ‘Newly augmented, By William Shakespeare’, but the text of the play adds only several new stage directions. The next four quartos also advertise the play as ‘newly augmented’ by Shakespeare. 1603 1st edition of Hamlet (STC 22275) Printed in five quarto editions by 1637, and in the Shakespeare folios. Another four quartos would be printed from 1676 to 1695. This edition would later be described as a ‘bad quarto’ (1909) because the text of the play is only about half as long as that in the second edition of the play (1604) and in the folios. 1604 2nd edition of Hamlet (STC 22276) This edition contains a text that substantially differs from that in the first edition (1603), including over 1,500 new lines. It was advertised on its title page as ‘Newly imprinted and enlarged to almost as much againe as it was, according to the true and perfect Coppie.’ The imprint on the title page of this edition was initially dated ‘1604’, but during the printing of the edition, the year was changed to ‘1605’. The next three quarto editions also advertise the play as ‘newly imprinted and enlarged’.

272

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

1605 1st edition of The London Prodigal (STC 22333) Although this play is not now considered to have been written by Shakespeare, this edition claims the play is ‘By William Shakespeare’. The play would later be added to the Third Folio (1663–4). 1608 1st edition of King Lear (STC 22292) Reprinted in 1619, and in a further quarto edition in 1655, as well as in the Shakespeare folios. 1st edition of A Yorkshire Tragedy (STC 22340) Although this play is not now considered to have been written by Shakespeare, this edition advertises the play as ‘Written by W. Shakspeare.’ The play would later be added to the Third Folio (1663–4). 4th edition of Richard the Second (STC 22310–22311) This edition contains a new scene that is not in earlier quartos and that is not mentioned on the original title page of this edition. A second title page was thus printed calling attention to this scene: ‘With new additions of the Parliament Sceane, and the deposing of King Richard.’ The next two editions of the play also advertise these ‘new additions’. 1609 1st edition of Troilus and Cressida (STC 22331–22332) Not reprinted in quarto, this play was reprinted in the Shakespeare folios. There are two issues of this edition. The title page for the earlier issue claims the play was ‘acted by the Kings Maiesties seruants at the Globe’ (see Figure 4.1.6). The title page for the latter issue does not mention the play’s performance and adds an address to readers from the publishers, Richard Bonian and Henry Walley. 1st and 2nd editions of Pericles (STC 22334–22335) Printed in six quarto editions by 1635, the play was not included in the first two Shakespeare folios, only being added to the Third Folio (1663–4). This edition would later be described as a ‘bad quarto’ (1909) with an obviously ‘corrupt’ text. 1st edition of Sonnets (STC 22353) This version would not be reprinted until the late eighteenth century, but the sonnets were printed in altered forms in Shakespeare’s Poems (1640). 1611 2nd edition of The Troublesome Reign of King John, Parts 1 and 2 (STC 14646) First published in 1591 without naming an author, these plays were attributed to ‘W. Sh.’ in this quarto, and in 1622 to ‘W. Shakespeare’ (see Figure   4.1.7). Shakespeare probably used these plays as a source when writing his own play of King John, which was printed in the First Folio (1623).

CHRONOLOGY

273

FIGURE 4.1.6  Title page of Troilus and Cressida (Q1, 1609). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

FIGURE 4.1.7  Title page of The Troublesome Reign of John, King of England (Q3, 1622), attributed to ‘W. Shakespeare’. Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

274

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

1612 3rd edition of The Passionate Pilgrim (STC 22343) The title page of this edition claims it is ‘By W. Shakespere’ adding that it has been ‘newly corrected and augmented’ with ‘newly added … Loue-Epistles’. These ‘Loue-Epistles’ are excerpts from a poem by Thomas Heywood, a contemporary author, who alleged that Shakespeare was ‘much offended’ with the printer of the edition, William Jaggard, who ‘presumed to make so bold with his name’ (An Apology for Actors (London, 1612), sig. G4v). The title page was subsequently reprinted without Shakespeare’s name. 1616 6th edition of The Rape of Lucrece (STC 22350) The title page of this edition is the first to advertise the poem as ‘By Mr. William Shakespeare’, adding the claim that the poem has been ‘Newly Reuised’. 1619 Publication of the ‘Pavier Quartos’ Ten plays attributed to Shakespeare were intended to be published in a quarto collection, but the printing of the collection was stopped, probably because of an order by the Lord Chamberlain, William Herbert, in May 1619. The ten plays ended up being published separately (except for 2 and 3 Henry the Sixth, which share a joint title page), though the ten quartos are sometimes found bound together. The planned collection included two plays not by Shakespeare, A Yorkshire Tragedy and Sir John Oldcastle, Part 1, both of which would later be added to the Third Folio (1663–4). Four of the plays are falsely dated ‘1600’ (Henry the Fifth, A Midsummer Night’s Dream, The Merchant of Venice and Sir John Oldcastle, Part 1); one is falsely dated ‘1608’ (King Lear); two are undated (2 and 3 Henry the Sixth); and three are dated ‘1619’ (The Merry Wives of Windsor, Pericles and A Yorkshire Tragedy). Thomas Pavier has usually been considered to be the publisher of these editions (1908), but some recent scholars have argued it was the printer, William Jaggard, who published them and, as a result, that they should be referred to as the ‘Jaggard Quartos’. 1622 1st edition of Othello (STC 22305) Reprinted in 1630, in a further quarto edition in 1655, and in the Shakespeare folios. The title page of the 1655 playbook calls that edition ‘The fourth Edition’, which could indicate that there is a missing quarto edition of the play. (See Figure 4.1.8.) 1623 1st edition of Shakespeare’s Comedies, Histories, & Tragedies (First Folio) (STC 22273) This collection of thirty-six plays by Shakespeare would be reprinted three times in the seventeenth century. According to the edition’s address ‘To the great Variety of Readers’, the plays were ‘collected’ by the deceased author’s

CHRONOLOGY

275

FIGURE 4.1.8  Title page of Othello (Q1, 1622). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

‘Friends’, in particular William Heminges and Henry Condell, two of Shakespeare’s fellow actors in the King’s Men (sig. πA3r). The edition was published by the printers William and Isaac Jaggard and the booksellers Edward Blount, William Aspley and John Smethwick. Eighteen of the plays had never before been published. The other eighteen plays were printed in versions that differ from the texts in earlier quartos, sometimes in versions that had been lightly revised based on theatrical or authorial manuscripts and sometimes in versions that are radically different. The Folio does not include four plays earlier ascribed to Shakespeare: The London Prodigal, A Yorkshire Tragedy, Pericles and 1 Sir John Oldcastle. (See Figure 4.1.9.) 1630 2nd quarto edition of Othello (STC 22306) This edition of Othello conflates the texts of the first quarto (1622) and First Folio, the first time such a conflated text was printed. 1631 1st quarto edition of The Taming of the Shrew (STC 22327) This edition reprints the text of the play in the First Folio. The publisher of the playbook, John Smethwick, owned the right to copy for the anonymous

276

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 4.1.9  Title page of the First Folio (1623). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

The Taming of a Shrew (published three times from 1594 to 1607), suggesting he may have considered The Shrew and A Shrew to be alternate versions of the same play. 1632 2nd edition Shakespeare’s Comedies, Histories, & Tragedies (Second Folio) (STC 22274–222274e.5) A reprint of the First Folio, this edition includes over 1,600 changes to the texts of the plays. 1634 1st edition of The Two Noble Kinsmen (STC 11075) The title page of this edition names John Fletcher and William Shakespeare as co-authors, the only playbook printed before 1660 to link Shakespeare to another author. The play would not be added to any of the later Shakespeare folios, but it would be printed in the 1679 folio collection of plays by Francis Beaumont and John Fletcher. (See Figure 4.1.10.)

CHRONOLOGY

277

FIGURE 4.1.10  Title page of The Two Noble Kinsmen (Q1, 1634). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

1640 1st edition of Shakespeare’s Poems (STC 22344) This volume includes most of Shakespeare’s sonnets and lyric poems, but also poems by several other authors. Benson added titles to Shakespeare’s poems and often combined two to five sonnets in order to create larger composite poems. This version would be the basis for editions of Shakespeare’s poems until 1780. (See Figure 4.1.11.) 1663–4 3rd edition of Shakespeare’s Comedies, Histories, & Tragedies (Third Folio) (Wing S2913–S2914) This folio was first issued in 1663 as a reprint of the Second Folio, but it was reissued in 1664 with seven new plays. Four had earlier been printed in quartos attributed to Shakespeare: Pericles, The London Prodigal, 1 Sir John Oldcastle and A Yorkshire Tragedy. The other three had been printed in quartos naming ‘W. S.’ as author: Locrine, Thomas Lord Cromwell and The Puritan Widow. (See Figure 4.1.12.)

278

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

FIGURE 4.1.11  Title page of Poems (O1, 1640). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

FIGURE 4.1.12  Title page of the Third Folio, Second Issue (1664). Fondation Martin Bodmer, Cologny, Geneva, Switzerland. Digitized by the Bodmer Lab, University of Geneva.

CHRONOLOGY

279

1685 4th edition of Shakespeare’s Comedies, Histories, & Tragedies (Fourth Folio) (Wing S2915–S2917) This edition includes the seven plays added to the 1664 reissue of the Third Folio. 1709 Nicholas Rowe, ed., The Works of Mr. William Shakespear, 6 vols A poet and a dramatist, Nicholas Rowe was commissioned by the publisher Jacob Tonson to produce an edited collection of Shakespeare’s plays, including the seven plays added to the Third Folio but excluding the poems. Attempting to ‘redeem’ the plays from ‘the Injuries of former Impressions’ (1: sig. A2r), Rowe based his text on the Fourth Folio while also adding engravings, act and scene divisions, a list of characters for every play and the first biography of Shakespeare. 1723–5 Alexander Pope, ed., The Works of Shakespeare, 6 vols This edition was put together by the poet Alexander Pope, who based his text on Rowe’s edition but excluded the seven plays added in 1664. Pope’s editorial interests were primarily aesthetic, leading him to make numerous changes to the plays in order to regularize their metre and to cut entire passages that he believed were stylistically inferior because they had been added by actors or corrupted by someone else. Pope’s edition would be revised by William Warburton in 1747. 1733 Lewis Theobald, ed., The Works of Shakespeare, 7 vols Theobald based his text on the preceding Tonson edition, edited by Pope, even though Theobald had vociferously criticized Pope’s edition after it was published. Theobald claimed he had read many of the historical and literary sources used by Shakespeare, plus ‘above 800 old English plays’, so as to understand ‘the obsolete and uncommon Phrases’ (1: lxviii) used by Shakespeare. This research led Theobald to include longer and more frequent annotations than Rowe or Pope had. 1765 Samuel Johnson, ed., The Plays of William Shakespeare, 8 vols Samuel Johnson deployed the lexicographical knowledge he amassed in compiling his Dictionary of the English Language (1755) to restore ‘the primitive diction wherever it could for any reason be preferred’ (1: lxii) in Shakespeare’s plays. Although critical of earlier editions, Johnson included prefaces written by previous editors and collations of some of their emendations. Jonson’s edition would be revised by George Steevens in 1773.

280

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

1767–8 Edward Capell, ed., Mr. William Shakespeare his Comedies, Histories, and Tragedies, 10 vols Unlike his predecessors, Capell edited the plays based on his conviction that some of the quartos possessed greater textual authority than the folios did. The text of this edition was printed free from annotations, the notes being published in later volumes, though he did use a system of signs to indicate significant textual details in his edition. 1790 Edmond Malone, ed., The Plays and Poems of William Shakespeare, 10 vols The most ambitious of the eighteenth-century editions of Shakespeare, Malone included for the first time the 1609 sonnets in a complete works of Shakespeare, a new biography of Shakespeare based on documentary sources, a full chronology of the plays, a history of the early modern English stage and other supplementary materials. With the assistance of James Boswell, Malone’s edition would be revised and expanded to twenty-one volumes in 1821. 1838–43 Charles Knight, ed., The Pictorial Edition of the Works of Shakspere, 8 vols Following the discovery of Q1 Hamlet in 1823, Knight’s popular edition put forward the argument that the Folio versions of Shakespeare’s play are the product of authorial revision and most fully represent the author’s final intentions. 1842–4 John Payne Collier, ed., The Works of William Shakespeare, 8 vols Building upon an article published in 1831, Collier made the novel argument in this edition that the texts of the first editions of Hamlet (1603) and Romeo and Juliet (1597) were in fact later versions of the plays that originated in imperfect manuscripts that were ‘prepared from defective short-hand notes’ (6: 369) of the plays in performance. 1864 George William Clark and William Aldis Wright, eds, The Globe Edition. The Works of William Shakespeare Using the text of the Cambridge Shakespeare edited by Clark, Wright and John Glover (1863–6), this one-volume edition became the best-selling Shakespeare edition in the nineteenth century. Its act, scene and line numbers would become standard for other editions of the plays and poems. 1872 George William Clark and William Aldis Wright, eds, Hamlet, The Clarendon Shakespeare This edition of Hamlet made the innovative claim that the first edition of the play includes passages from earlier versions of the play not written by

CHRONOLOGY

281

Shakespeare. This theory would soon lead other scholars to seek to ascribe portions of Shakespeare’s plays to diverse other authors. 1902 Sidney Lee, Shakespeares Comedies, Histories, & Tragedies: A Census of Extant Copies Lee’s census catalogued all the copies then known of the First Folio (158 in total). 1908 W. W. Greg, ‘On Certain False Dates in Shakespearian Quartos [Part I]’, The Library, n.s. 34 (1908). Part I: 113–31; Parts II–III: 381–409 One of the most important early articles in the New Bibliography, Greg’s essay used the tools of analytical bibliography to determine that, despite the dates on their title pages, all the ‘Pavier Quartos’ were printed in 1619. 1909 A. W. Pollard, Shakespeare Folios and Quartos: A Study in the Bibliography of Shakespeare’s Plays, 1594–1685 Pollard coined the influential term ‘bad quarto’ in this monograph. He argued that ‘the diuerse stolne, and surreptitious copies’ referred to by Heminges and Condell in the First Folio (sig. πA3r) did not concern all the earlier quartos of Shakespeare’s plays, but only five of them: Romeo and Juliet (1597), Henry the Fifth (1600), The Merry Wives of Windsor (1602), Hamlet (1603) and Pericles (1609). Pollard contended these ‘bad quartos’ had been ‘pirated by shorthand writers, or men with quick memories’ (2). 1910 W. W. Greg, ed., Shakespeare’s Merry Wives of Windsor, 1602 In this edition of the ‘bad quarto’ of The Merry Wives of Windsor, Greg developed a new theory to explain the origins of its text: memorial reconstruction by actors, most centrally the Host. 1916 Edward Maunde Thompson, Shakespeare’s Handwriting: A Study A prominent paleographer, Thompson argued that Shakespeare was one of several authors (Hand D) who contributed to Sir Thomas More, a play manuscript owned by the British Museum (now the British Library). For scholars who agree with Thompson’s theory, this manuscript represents the only surviving dramatic manuscript with evidence of Shakespeare’s authorship. Henrietta C. Bartlett and Alfred W. Pollard, A Census of Shakespeare’s Plays in Quarto, 1594–1709 This census describes all the known copies of quartos of Shakespeare’s plays, excluding 2 Henry the Sixth, 3 Henry the Sixth and Pericles. Bartlett would publish a revised and extended edition of the Census in 1939, and in 2018, Zachary Lesser and Adam G. Hooks would publish an online census of Shakespeare’s plays and poems, excluding folios (shakespearecensus.org).

282

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

1917 A. W. Pollard, Shakespeare’s Fight with the Pirates and the Problem of the Transmission of His Text Extending Greg’s theory of memorial reconstruction (1910), Pollard argued the stationers who published the ‘bad quartos’ were guilty of piracy, in part because none of the five bad quartos was ‘entered on the Stationers’ Register by its publishers’ (49) (but see 1997). 1931 W. W. Greg, Dramatic Documents from the Elizabethan Playhouses: Stage Plots, Actors’ Parts, Prompt Books, 2 vols In this edition of documents that originated in the early modern London theatre, Greg describes his theory of ‘foul papers’, ‘fair copies’ and ‘prompt books’. 1939 R. B. McKerrow, Prolegomena for the Oxford Shakespeare: A Study in Editorial Method This work outlines the New Bibliographical editorial theories and procedures McKerrow planned to use for his Oxford edition of Shakespeare, which was never completed. McKerrow died in 1940, and the project was then taken over by Alice Walker. 1939–59 W. W. Greg, A Bibliography of the English Printed Drama to the Restoration, 4 vols This work provides descriptive bibliographies of almost every extant edition of a printed play, either written in English or printed in Britain, from the early sixteenth century to 1660. 1950–1 W. W. Greg, ‘A Rationale of Copy-Text’, Studies in Bibliography 3 (1950/1): 19–36 In what is probably the most important essay of editorial theory published during the twentieth century, Greg made an influential, but also controversial, distinction between ‘substantive’ readings and ‘accidentals’. 1951–82 The Arden Shakespeare, Second Series The first series of Arden Shakespeare editions was published in the early twentieth century, but the volumes in the second Arden series of Shakespeare’s plays and poems emerged during the twentieth century as the pre-eminent scholarly editions of Shakespeare.

CHRONOLOGY

283

1963 Charlton Hinman, The Printing and Proof-Reading of the First Folio of Shakespeare, 2 vols A landmark in analytical bibliography, Hinman’s study is the most comprehensive examination of the printing of the First Folio. 1968 Charlton Hinman, The Norton Facsimile of the First Folio of Shakespeare Rather than reproduce an individual copy of the First Folio, Hinman sought to reconstruct the ‘ideal copy’ of the First Folio, with ‘the latest or most fully corrected state of the text’ (xxii) on each page. 1982 Peter W. M. Blayney, The Texts of ‘King Lear’ and their Origins: Volume 1: Nicholas Okes and the First Quarto This monograph is the most detailed bibliographical analysis of the printing of a Shakespeare quarto. 1986 Stanley Wells and Gary Taylor, gen. eds, William Shakespeare: The Complete Works (Oxford Shakespeare) This complete works includes both old-spelling and modernized editions. Rather than attempting to reconstruct the ‘fair copies’ of Shakespeare’s manuscripts, the editors sought to produce editions of the plays as they were first performed, causing them to change the titles of some plays and to alter the names of some characters, most notoriously Falstaff in Henry the Fourth, Part One who became ‘Oldcastle’. The Oxford Shakespeare also printed separate quarto and folio versions of King Lear because the editors contended the play was revised by Shakespeare. 1993 MIT Shakespeare, shakespeare.mit.edu/ This site published the first online edition of the complete works of Shakespeare, which used as its text the ‘Moby Shakespeare’, a plaintext digital transcription of the plays made publicly available by Grady Ward in 1992 that was derived from the Globe Shakespeare (1864). 1995–2020 The Arden Shakespeare, Third Series Already established as the pre-eminent series for editions of Shakespeare, the Arden Third Series shows an increased attention to performance and to the socalled ‘bad quartos’, which culminated in publishing three versions of Hamlet based on the first quarto, the second quarto and the First Folio.

284

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

1997 Peter W. M. Blayney, ‘The Publication of Playbooks’, in A New History of Early English Drama, ed. John D. Cox and David Scott Kastan, 383–422 This essay describes in detail the steps involved in the publication of an early modern playbook, but more important, it demonstrates that the absence of an entry in the Stationers’ Register was unconnected to piracy. 1999–present Early English Books Online (EEBO), https://quod.lib.umich.edu/e/eebogroup/ Originally published online by Chadwyck-Healey, this site allows users to access digital reproductions of microfilmed copies of early modern books that were published for the English book trade from 1473 to 1700. Starting in 2000, the EEBO Text Creation Partnership (EEBO-TCP) began to make available fulltext transcriptions of the texts of some of these books (almost 35,000 by 2019). 2002–3 Anthony James West, The Shakespeare First Folio: The History of the Book, 2 vols A revision of Sidney Lee’s census (1902), these volumes provide detailed descriptions of 228 known copies of the First Folio as well as additional historical information about its sales, prices and distribution. 2003 Lukas Erne, Shakespeare as Literary Dramatist This book proposes a new theory to explain the multiple versions of Shakespeare’s plays, suggesting that Shakespeare wrote ‘literary’ versions of his plays to be read and that these ‘literary’ versions were then revised to produce shorter texts for performance. 2013 Paul Werstine, Early Modern Playhouse Manuscripts and the Editing of Shakespeare Building on a series of articles published starting in the early 1980s, this study offers a revisionary analysis of the concepts of ‘foul papers’, ‘fair copies’ and ‘prompt books’. 2016 Richard Dutton, Shakespeare: Court Dramatist Dutton puts forth a new theory to explain the different versions of Shakespeare’s plays, arguing they ‘were frequently and specifically revised for presentation at the courts of Elizabeth I and James I’ (1). 2016–17 Gary Taylor, John Jowett, Terri Bourus and Gabriel Egan, gen. eds, The New Oxford Shakespeare: Modern Critical Edition, Critical Reference Edition and Authorship Companion, 4 vols

CHRONOLOGY

285

This edition makes the most ambitious use to date of attribution studies to investigate the authorship of Shakespeare’s plays. It suggests several new attributions, including that Christopher Marlowe collaborated in the writing of the three Henry the Sixth plays and that Thomas Middleton co-authored or revised five plays. It also claims that Shakespeare wrote at least portions of several plays not traditionally ascribed to him: Arden of Faversham, Edward the Third, additions to The Spanish Tragedy and a lost version of Sejanus coauthored with Benjamin Jonson.

CHAPTER 4.2

A–Z of key terms and concepts ERIC RASMUSSEN AND IAN H. DE JONG

ACCIDENTALS AND SUBSTANTIVES The two different types of textual variant. Variants that affect the meaning or poetry of a passage, such as sullied/solid in Hamlet 1.2, qualify as substantive, a term introduced by W. W. Greg (1950/1: 19–22). In contrast, variants involving matters such as punctuation, spelling and capitalization have traditionally been considered accidental. In that same line from Hamlet 1.2, for example, the First Folio inserts a comma after ‘flesh’, but the first and second quartos do not. Recent scholarship has reappraised just how ‘accidental’ such variants are, suggesting that semiotic significance may reside even in these apparently arbitrary details.

ACT AND SCENE DIVISIONS Paratextual insertions that define breaks in dramatic action, and sometimes signal shifts in place or time. Act and scene divisions do not appear consistently in the printed drama until well after Shakespeare’s death. Those in the first few plays of the First Folio were probably added by the scribe Ralph Crane. Most act and scene divisions today are inserted by editors.

ALLOWANCE Before any book could be printed, it had to be examined by either a state official or an ecclesiastical dignitary. English law mandated this pre-publication scrutiny in order to exert a measure of censorial control over potentially incendiary books. The Stationers’ Company would sometimes refuse to license texts which had not been allowed. According to Peter Blayney, the fee for allowance could have been as little as seven shillings or as much as two pounds (1997: 397).

ANONYMOUS PUBLICATION A practice of omission motivated by any number of noble and ignoble intentions. Anonymity could serve to protect an author of particularly incendiary material, as

A–Z OF KEY TERMS AND CONCEPTS

287

in the case of Thomas Middleton’s political satire A Game at Chesse. Alternatively, it could signal ignorance on the part of the printer (and, likely, the general public) as to a play’s original author; the anonymous Mucedorus, early modern England’s most popular play in print, may belong in this category. The evidence in Philip Henslowe’s accounts, as well as in the Stationers’ Register, suggests that theatre professionals kept very sharp track of who wrote what, and how much and for what pay. The absence of an authorial name on a quarto’s title page does not necessarily imply a similar anonymity in theatrical and print commercial spheres.

ASSEMBLED TEXT A textual version supposedly created by the cobbling together of playhouse documents such as actors’ part-rolls (the source for speakers’ lines) and/or the plot (the source for stage directions). The assembled-text hypothesis, first promulgated by twentiethcentury New Bibliographers, explains certain patterns of textual obscurity, such as scarcity of stage directions or convoluted narrative development. Muriel Bradbrook suggested that The Two Gentlemen of Verona, The Merry Wives of Windsor and The Winter’s Tale in the First Folio were based upon assembled texts (1962: 135).

AUTHENTICITY A key question in the establishment of copy-text, authors’ canons and chronologies. Some early modern printers were careless, and a few deliberately disingenuous, in matters of attribution, date and place of origin. This mixture of accidental and purposeful misrepresentation clouds questions of authenticity. The rise of forgery further confuses the issue: John Payne Collier manufactured early annotations, and William-Henry Ireland ‘discovered’ a new play of Shakespeare’s. Experts in manuscript history compare letterforms, ink colour and orthography to pronounce handwritten texts authentic or not; experts in print history rely on their knowledge of typefont design, printers’ mechanical habits and minute characteristics of early modern paper to determine the authenticity of printed texts.

AUTHORITY Widely considered to be synonymous with allowance, this term most clearly reflects the political nature and goals of the allowance process (Blayney 1997: 396–7; Weiss 2007: 199; Bland 2013: 187–8). Works which were allowed were granted ‘authority’ by the state or ecclesiastical authorities. The Stationers’ Register occasionally notes the name of the authorizing dignitary.

AUTHORSHIP Long after Roland Barthes pronounced the author to be dead, the task of identifying the authors of early modern drama – particularly of the plays attributed to Shakespeare – continues to occupy popular attention and scholarly energy. Modern data analysis and database technologies have allowed for a more granular

288

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

sense than ever of which works, and which sections of works, may be attributed to Shakespeare, and which may be attributed elsewhere. Simultaneously with this turn toward stylometry, a consensus has developed that early modern dramatic creativity was highly collaborative. Authors have been recognized as part of a network of dramatic revisers, print-house correctors, actors, censors and theatrical impresarios responsible for shaping early modern plays. Some scholars argue that William Shakespeare never existed, or had no hand in the composition of the plays attributed to him. Sir Francis Bacon and Edward de Vere, 17th Earl of Oxford, have been proposed as alternative authors. The anti-Stratfordian movement has gained a degree of popular support, but is largely dismissed by the academic establishment.

AUTOGRAPH A manuscript written in the hand of the author, as opposed to that of a scribe. The Trinity autograph manuscript of A Game at Chesse was created by the play’s author Thomas Middleton himself, whereas the Bodleian scribal manuscript of the play is in the hand of Ralph Crane. Much of Ben Jonson’s poetry survives in autograph manuscript. Since the Hand D part of The Booke of Sir Thomas More has not been conclusively proven to be Shakespeare’s, no confirmed Shakespearean autograph survives beyond some signatures.

‘BAD QUARTOS’ Some of the extant early quartos of Shakespeare’s plays differ significantly from their subsequent reprints. Often, these earlier quartos seem garbled, or at least confusing, and many lack key elements that appear in later printings. For example, in the first quarto of Hamlet, the play’s most famous lines read, ‘To be, or not to be, I there’s the point / To Die, to sleepe, is that all? I all’ (sig. D4v). Early twentieth-century critics termed these shorter, earlier quartos ‘bad’, as contrasted with later, longer, more familiar ‘good’ quartos. This binary opposition had the benefit of explaining the reference, in the First Folio’s address ‘To the great Variety of Readers’, to ‘the diuerse stolne, and surreptitious copies, maimed, and deformed by the frauds and stealthes of iniurious impostors’ with which pre-Folio readers had been ‘abus’d’ (sig. πA3r). The origin of the ‘bad’ quartos’ corruption, so went the argument, could be traced to piracy by unscrupulous printers, who would pay spectators to copy down plays (‘memorial reconstruction’) or buy playscripts from impecunious actors. While some early quartos of Shakespeare’s works undoubtedly contain more errors than later editions do, modern scholarship has largely rejected the ‘bad’/‘good’ dichotomy in favour of more nuanced investigations into the reasons for variance among early quarto texts.

BASE TEXT In editorial parlance, this term is often used interchangeably with ‘copy-text’ and ‘control text’; the three mean roughly the same thing. The base text is the single early textual witness upon which a later edition is based. Most modern editors

A–Z OF KEY TERMS AND CONCEPTS

289

of Shakespeare choose early quarto or folio texts as their base texts. However, as Andrew Murphy notes, eighteenth-century editors like Lewis Theobald built their editions not on Elizabethan or Jacobean texts but on recent edited texts: Alexander Pope’s edition served as the base text for Theobald (2003: 73).

BEST TEXT Like copy-text, the best-text editorial paradigm attempts to minimize editorial intrusion. When multiple texts are available, the best text is deemed to be that which requires the least, and the least intrusive, programme of emendation. G. Thomas Tanselle sums up best-text editorial practice as ‘altering [the base text] only at the places that seemed obviously erroneous’ (1994: 1). Best-text editing emerged from the New Bibliography; it differs from more subjective models which had held sway in the nineteenth and early twentieth centuries.

BIBLIOGRAPHY The study of books as material and expressive objects. As a discipline, bibliography grew out of antiquarian bibliophilia, a field once dominated by scholars of leisure. It became a profession as a result of the early twentieth-century theoretical shift called the New Bibliography, which studied the material prehistories of books’ verbal semiosis in order to shed light on their textual history. Modern bibliographers deploy sophisticated digital tools for the reproduction, organization and analysis of early modern books.

BINDINGS The study of bindings has illuminated the history of early modern books in ways previously unimaginable. Early bindings were made of vellum, leather or skin, either stretched over boards or limp. Often, they were decorated using methods appropriate to their material (vellum might be illuminated, or leather might be tooled). Evidence suggests that early modern retail bookshops stocked more unbound than bound volumes. Some thin books were simply stitched together (see stabstitching). Customers who bought unbound books could have them specially bound; alternatively, they could bind several slender books together (see sammelbände).

BLOCKING ENTRY A purported business practice in early modern publishing. According to W. W. Greg and other New Bibliographers, some publishers registered, or ‘entered’, their rights to publish plays in the Register of the Stationers’ Company in order to prevent other publishers from doing so. Greg supports this hypothesis with the example of James Roberts, who registered his right to publish Troilus and Cressida in 1603, then assigned that right to Richard Bonian and Henry Walley, who printed the play in 1609 (1956: 112–22).

290

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

BOOK Attempts to define the limits of the book are notoriously difficult. Most books – but not all – are, or have been at some point, bound. Most books – but not all – are composed of semi-flexible sheets of writing support (paper or parchment, most often). Most books – but not all – are produced with mechanical pressure by an ink-distributing technology, whether pen, brush, block or movable type. Most books – but not all – convey meaning through verbal, visual and material rhetoric. Most books – but not all – exist at some point after manufacture as discrete textual units. Most books – but not all – possess internally consistent aesthetic, narrative or rhetorical structures.

BOOK-KEEPER The ‘book of the play’ in the early modern period referred to the manuscript used in the playhouse. The book-keeper, later known as the prompter, attended to dialogue, standing just offstage to remind forgetful actors of their lines. According to Tiffany Stern, book-keepers had some measure of authority, with assistants and other supernumeraries helping them discharge their role (2009: 219–27).

BOOKSELLERS One of three primary roles in the early modern English book trade (the others being printers and publishers). These roles could be combined; booksellers who were also publishers often sold wholesale and sometimes maintained resale shops as well. Another subset of booksellers confined themselves primarily to retail. Title-page descriptions of shop-locations usually refer to wholesale outlets, not to retail ones; for this reason, scholars know more about bookseller-publishers than about retailonly booksellers. Though both types of shops could be found throughout the City and its environs, they were concentrated in the yard of St Paul’s Church and in Fleet Street.

BOOK TRADE An umbrella term encompassing the many facets of the commercial market in printed books. Key to the book trade were the manufacturers of raw materials – paper, ink, type, press machinery and appurtenances, and binding. The backbone of the book trade, and the most studied element today, were publishers and printers, who created the printed books to be sold. The operations of wholesale and retail booksellers are more difficult to sketch, given the paucity of documentary evidence.

CANCEL Errors discovered after printing, such as flipped, missing or wrong letters, eye-skip or duplication, were frequently corrected in the printing house. One simple way

A–Z OF KEY TERMS AND CONCEPTS

291

to make these corrections was to cancel the erroneous leaf. Corrected leaves were printed, erroneous leaves were cut out and the corrected leaves were pasted in. Scholars often know of the existence of cancellanda from the presence of uncorrected states in certain copies; early modern English printers, it seems, caught some errors in the midst of printing, after a number of copies had already been organized into quires. One example of a cancelled page occurs in the First Folio. The publishers had planned to place Troilus and Cressida after Romeo and Juliet, but rights issues held things up, so they placed Timon there instead. The printers, however, had already begun printing Troilus on the back of the final leaf of Romeo and Juliet. The printers marked both of the offending pages with an inked X through their text, but certain copies of the First Folio retain the cancelled pages.

CAPITALIZATION Early playbooks, whether printed singly in quarto or collected in folio, include words capitalized apparently at random. Take, for example, Lucio’s line early in Measure for Measure: ‘Thou conclud’st like the Sanctimonious Pirat, that went to sea with the ten Commandements, but scrap’d one out of the Table’ (sig. F1v). Scholars disagree on whether capitalization was intentional. Most believe either that the printed text accurately represents erratic capitalization in manuscript, or that compositors introduced erratic capitalization for (still obscure) reasons of their own. Some, however, have proposed that capitalization, particularly in the First Folio, records points of vocal emphasis in the early modern theatre, and thus offers a unique window into performance history as well as giving clues to actors (Tucker 2002: 232–6).

CASES In the early modern print shop, the type was separated by character and size to facilitate the work of the compositor. Majuscules (A–Z) and minuscules (a–z) were stored in separate shallow boxes, each box containing compartments for each letter, number and symbol. The case containing majuscules was often stacked above the case of minuscules; from this arrangement we get our differentiation between ‘upper case’ and ‘lower case’ letters.

CATCHWORD Many early modern printed pages include a single word in the lower right-hand corner (or, in technical terms, near the fore-edge foot). This word, the catchword, signals the first word on the next page; for example, the first page of Othello in the First Folio (sig. ff3v) concludes with the catchword ‘Your’, and the next page (sig. ff4r) begins ‘Your heart is burst’. Catchwords were primarily for production-side, print-house use; they are not particularly helpful in reading. However, as Sarah Werner points out, when a catchword is not the same as the first word on the next page, this can indicate some disruption – either during the process of printing, with

292

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

the next page differing from what was forecast, or during the process of assembly, post-printing (2019: 172).

CENSORSHIP Both Elizabeth I and James I were concerned about the capacity of print generally, and drama particularly, to foment sedition and civil unrest. The office of the Master of the Revels upheld governmental interests by censoring material deemed potentially incendiary; this practice involved ‘allowing’ or ‘disallowing’ new plays. Allowed plays could be presented as they were; disallowed plays might be prohibited completely, or blocked from performance until required revisions were made. The penalties for evading this process could be great: Ben Jonson and George Chapman were imprisoned for their part-authorship of Eastward Hoe!, a comedy offensive to James I. The manuscript of The Book of Sir Thomas More contains the Master of the Revels’ marginal comments demanding specific elisions in the text.

CHAIN LINE In the process of handmaking paper, a wet slurry of pulped rags was spread in a mould with solid raised edges and a wire-screen face; the moisture would drip out through the screen and the remaining solid mixture would dry into paper. The screen was made of two gauges of wire: thinner ones, running parallel to the long edge of the mould, and thicker ones, running parallel to the short edge of the mould. The thicker wires left noticeable impressions in the paper; these are known as chain lines. Chain lines are not normally visible to the naked eye, but backlighting a piece of handmade paper can reveal not only the chain lines, but sometimes even which side of the paper was against the screen (Gaskell 1995: 61). This identification of chain lines, and specifically their direction relative to the spine, can be useful in determining a book’s format. Because of the way bookmakers folded printed sheets, a folio page will have chain lines parallel to the spine; in a quarto, chain lines will be perpendicular to the spine; in an octavo, chain lines will run parallel to the spine.

COLLATION Early printers, after impressing sheets and assembling them in quires, would arrange them in order according to their signature, double-checking to ensure that they appeared in the correct sequence. We refer to this as ‘collation’, and to a formulaic summary of signatures as a ‘collational formula’. Today, when textual scholars compare texts in different editions or variant states of the same edition, this process is also called ‘collation’. Modern collation, often done for editorial purposes, reveals the variants, accidental and substantive, between texts.

COLOPHON Before details of manufacture (date, location, printer, publisher, etc.) regularly appeared on title pages, such details were placed at the end of a book in a colophon.

A–Z OF KEY TERMS AND CONCEPTS

293

Sarah Werner traces this placement to manuscript tradition, in which scribes would include information about themselves after finishing the task of copying (2019: 85). Colophons appear most frequently in books printed before 1500, although the practice had not entirely disappeared by the turn of the seventeenth century. Colophons and title-page details were not mutually exclusive; the First Folio, for instance, has both (though its title page lists location, printer and date, while its colophon lists publishers and date).

COMMONPLACE BOOK Some early modern readers and playgoers wrote down noteworthy fragments of material that they had read or heard. Much of what they recorded was sententious or proverbial – commonplaces, in contemporary parlance. Commonplace books are rich with historical evidence, shedding light on individual readers’ habits of mind, the cultural permeation of different types of literature and dates of composition for well-known plays and poems. Some commonplace books were printed, but this was primarily a manuscript tradition. It was defined, paradoxically, by both deliberation and immediacy. Some commonplace books contain pages of carefully copied poetry, perhaps an opportunistic appropriation of loaned verses. Other evidence points to the mobility of the commonplace book: when Hamlet exclaims ‘My tables! Meet it is I set it down’, his ‘tables’ are perhaps a portable technology for commonplacing (1.5.107).

COMPOSITORIAL ANALYSIS The bibliographical practice of determining the compositors in a print shop, their work habits and the division of labour between them (usually on individual large projects). Careful scrutiny of the distribution of unique type, error patterns and idiosyncratic spellings allowed scholars in the twentieth century to posit at least five different compositors who worked on the First Folio (Hinman 1963). Subsequent studies, including analysis of historical documents of print shop business, have further illuminated the professional characteristics of early modern compositors (Blayney 1982).

COMPOSITORS Workers in the print shop who set up type to get it ready for impression. They would stand in front of a type-case, perhaps with copy hung up in front of them, and slide type into a composing stick (a tool which held a few lines of set-up type). When the stick was full, those lines would be slid into a page-sized frame called a galley, and the compositor would recommence filling the stick. Compositors set type by hand; some elements of type and type-case design suggest that expert compositors set by touch. In order to maximize profits, printers expected compositors to work quickly; this, coupled with a potential propensity for drunkenness, led to variable accuracy in compositorial work (Gaskell 1995: 48).

294

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

CONCURRENT PRINTING The workflow of an early modern print shop was complex, designed to minimize downtime. Since composition and presswork did not proceed at exactly equal rates, and since print shops did not have unlimited stocks of type, there was often a need to stagger the production of multiple texts. Because of this approach, single books might take longer to print, but more books could be printed in a shorter amount of time, and with less wasteful expenditure of labour, than individual, serial printing would allow (Gaskell 1995: 164). We know, for example, that the First Folio of Shakespeare was printed concurrently with a few other large folio texts in the Jaggards’s print shop over the course of nearly two years.

CONFLATION A practice in editing of amalgamating multiple variant texts into one new text. This practice most frequently occurs when the relative authority of early texts is in dispute. For example, generations of editors conflated Q and F King Lear, each of which contains material absent from the other, and each of which has its own claim to textual authority. The practice of conflation developed from twentiethcentury textual critics’ theory that authorial texts lay hidden behind the ‘veil of print’, requiring discovery by astute editors. Conflation helped remove that veil by reassembling the original plays – so went the narrative. The best-text approach stands diametrically opposite the practice of conflation.

CONTROL TEXT Another term interchangeable with ‘copy-text’ or ‘base text’, popularized by Wells and Taylor’s William Shakespeare: A Textual Companion (1987: 155). The control text from which an editor shapes their edited text exists in the editorial and historical context of other texts; the control text is judged to be of primary, but not sole, authority. That is, while editors make most of their decisions based on the textual features of the control text, they will certainly at some point consult other textual witnesses to solve textual cruxes or fill in textual gaps in the control text.

COPY, CAST-OFF Most projects printed from manuscript copy, or reprinted in a different format from print copy, required a process of estimating how much textual content could fit on each page, and usually marking on the copy where page breaks were to occur. The accuracy of the casting-off process was of serious importance to the integrity of the text, as well as to print-house workflow and even the financial well-being of the publisher. Sloppy casting-off could lead to unnecessary white space in the middle of books; it could, worse, force a book to spill over onto extra sheets.

A–Z OF KEY TERMS AND CONCEPTS

295

COPY, CONTINUOUS The theory that authorial manuscripts were used as primary theatrical documents (the book of the play), and that those manuscripts continued to accrete revisions, additions and alterations throughout the plays’ lives on stage. This hypothesis supported twentieth-century hopes of recovering ‘original’ texts from behind the veil of print.

COPYRIGHT This term, referring to ‘the exclusive right given by law for a certain term of years to an author … to print, publish, and sell copies of his [sic] original work’ (OED 1989), was not in use during Shakespeare’s lifetime. But the concept of owning rights to a play is pertinent to the somewhat complicated symbiosis between theatre and print house in early modern England. Individual companies of players (such as the King’s Men or the Admiral’s Men) often had exclusive rights to perform some plays – usually, when a company impresario had paid for their composition, and the Master of the Revels had allowed their performance. But an exclusive right to perform did not necessarily imply a legal right to print. Publishers had to register their ownership of plays with the Stationers’ Company, which had broad power to penalize financially those who printed works owned by others.

COPY-TEXT The text on which a modern critical edition of a work is based. In some cases, the choice of copy-text is simple: an edition of The Comedy of Errors, which was first printed in the First Folio, would naturally build upon that Folio printing as the earliest extant text of the play. But ‘earliest’ does not always mean ‘best’; sometimes, editors must make qualitative judgements between multiple early versions. Such is the case for Richard III, which exists in a 1597 quarto as well as in the First Folio. Each of these versions has a defensible claim to textual authority; but because the latter is longer and arguably has fewer errors, many modern editors choose the Folio Richard III as copy-text, although it appeared twenty-six years after the first quarto.

DELETION A practice of excision most visible in manuscript culture, though deletion also occurred, less obviously, in the process of printing. In manuscripts, deletion might involve removal of words from the writing support, through pasting blank paper in, scraping ink off or covering words over. Alternatively, it could involve strikingthrough words or lines to be cut, or even marking sections to be cut with large marginal strokes. The Booke of Sir Thomas Moore contains examples of the latter two practices, partially in response to Edmund Tilney’s demands for revision. Deletion in print is a different matter: it most often refers to unintentional omissions of letters, words or occasionally even lines (due to eye-skip).

296

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

DERELICT PLAY A play which, for a variety of reasons, was somewhat available to publishers looking to add to their catalogue. The Stationers’ Company assigned owners exclusive publication rights for any works ‘entered’ in its Register. If a publisher died without disposing of his rights (for instance, in a will), those rights reverted to the Company. In practice, however, the Company asserted such rights only in the case of works that had achieved substantial popularity; other publishers could print less popular, ‘derelict’ works as if they owned them, despite not having established ownership through official channels. This explains why certain plays and works appear in print without ever having been ‘entered to’ their publisher.

DEVICE A printer’s device is a woodcut or copper-plate block with a picture or design that served as the printer’s trademark, generally appearing on title pages or on the final leaf of a book. Devices might feature the printer’s crest or the sign that hung outside their shop. Thomas Pavier’s device punned on his name by representing the tools of the paving trade.

DRAMATIS PERSONAE The list of characters, including names and dramatic roles, appearing at the beginning or end of early modern plays. Dramatis personae lists are a staple of modern editions but were only sometimes included in early modern printed drama. The First Folio includes some lists of dramatis personae, often following their respective plays; these lists have titles that frequently use the word ‘actor’ instead of ‘character’ (e.g. ‘Names of the Actors’ at the end of The Tempest, sig. B4r). They sometimes name characters who are unnamed in the dramatic text, such as the Duke in Measure for Measure, who is identified as ‘Vincentio’ in the Folio’s list of ‘The names of all the Actors’ (sig. G6v). They also occasionally offer evaluative judgements of characters, such as the description of The Tempest’s Caliban as ‘a saluage and deformed slaue’ (sig. B4r). Throughout the period, female characters were almost invariably listed below male characters.

EDITING The practice of reproducing an early modern text for contemporary consumption. According to some scholars, editing and its ancestors have constituted Shakespeare since the earliest quartos of his work (Massai 2007: 1–38; Rasmussen 1998). Nicholas Rowe was the first named editor of Shakespeare, in 1709; during the late eighteenth century, Edmond Malone applied principles of Enlightenment rationalism to editorial practice. The nineteenth century saw the rise of a commercially successful ‘Illustrated Shakespeare’, among other popular-Shakespeare initiatives; the twentieth century’s major development was an ongoing attempt to minimize interpretive editorial interference. Since the turn of the millennium, editors have

A–Z OF KEY TERMS AND CONCEPTS

297

been exploring the possibilities of digital media and open-access publishing. Editing, throughout its history, has been fundamentally shaped by the desires of its intended readership. Those who bought single-play quartos differed from those who bought the First Folio (we presume); those who read a pedagogically focused text, such as an edition in the Folger Shakespeare Library series, differ from those who study the old-spelling Critical Reference Edition volumes of The New Oxford Shakespeare. This divergence in readership demands a divergence in editorial apparatus; The New Oxford Shakespeare includes tools excluded from the Folger paperback editions, and vice versa. Nevertheless, similar editorial practices and commitments undergird the great variety of editions, including interests in authenticity, textual transmission and responsiveness to readers’ needs.

EDITION This term has one meaning when applied to early modern texts, and a somewhat different meaning when applied to modern scholarship. In the former sense, ‘edition’ refers to an individual, discrete print run of a text; to be considered an edition, the text must in some way acknowledge or announce its distinctiveness from previous editions. Some early modern dramatic playbooks refer explicitly to their novelty: see, for instance, title-page claims to be ‘newly corrected and amended’. Many early texts, however, are more circumspect, and their status as new editions must be deduced from new dates, new printers or publishers, or other markers. Finally, there are a few texts, such as the Pavier Quartos, that obfuscate their new-edition status. Clouding this definition are local variations, minor and major, within press runs. Press variants destabilize the notion of uniform, monolithic ‘editions’. In modern scholarship, an edition is a curated re-presentation of an early modern text. Certain modern editions, such as those in the Arden Shakespeare series, have become, if not definitive, at least industry standard within Shakespeare studies.

EMENDATION An element of editing which has long been debated. Some of the earliest named editors (such as Alexander Pope) indulged in aesthetic emendation, which excised or demoted passages deemed poor in quality by the editor and elevated ‘shining passages’ for their beauties. Subsequent editors moved away from this practice, though its effects are still felt. Most modern editors emend when the syntax, sense or vocabulary of a passage seems absolutely unclear. A well-known example of valuable editorial emendation occurs in Folio Henry V, as Mistress Quickly recounts Falstaff’s death: ‘for his Nose was as sharpe as a Pen, and a Table of greene fields’ (sig. h4r). Lewis Theobald conjectured compositorial misreading, emending ‘and a Table of greene fields’ to ‘and ’a babbl’d of green fields’ (1733: 4.30).

ENTRANCE A key step for publishers in the long process of securing the early modern equivalent of copyright. After a publisher had secured a copy of the text to be printed, they

298

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

would usually pay the Stationers’ Company to have that text ‘entered to’ them. This process gave publishers a legal claim to ownership of a text, allowing them recourse to legal action (through the Stationers’ Company) if another publisher infringed on their right. A sizable minority of printed texts were never entered in the Register of the Stationers’ Company; scholars estimate rates of non-entry as between 30 per cent and 50 per cent (Bell 1994: 50–4). Non-entry was a gamble; it saved publishers money, lowering their overhead and financial risk, but left them vulnerable to the depredations of piratical competitors.

ENTRANCES AND EXITS Dramatic authors had various ways of demarcating actors’ entrances and exits. Some playwrights followed the classical practice of providing entrances and exits as group movements: thus, at the beginning of a new scene, all the participating characters are listed as entering together, even if one of these characters is supposed to enter mid-scene and be surprised. At the end of a scene, everyone leaves, including the messenger who has done nothing since his two-line speech at the scene’s beginning. A few scribes also preferred massed entrances and exits. The alternative, entrance and exit in narrative order, is more practical, but runs the risk of overlooking minor characters.

FACSIMILE EDITION As the study of early modern printed texts professionalized, a demand emerged for exact facsimile reproductions. Type facsimiles came first, attempting to reprint in exact duplicate the appearance, contents and errors of early texts. Next were photographic facsimiles, the best-known being Charlton Hinman’s impressive First Folio, which assembles page images from multiple folios in the Folger Shakespeare Library’s collection. (This means that the Folio reproduced in Hinman’s facsimile does not actually exist as a base-version, since his ‘ideal’ version is concocted from the most accurate pages in the Folger.) Most recently, the Early English Books Online (EEBO) project has digitized microfilms made in the 1960s, allowing its subscribers unprecedented access to early modern books.

FAIR COPY The term used to describe an early modern manuscript that has been prepared with an eye to its mise-en-page. In manuscript studies, fair copies have often been contrasted with foul papers. The latter were authorial in origin and focused on content rather than presentation; as a result, foul papers might be quite messy, as their name suggests. A fair copy, by contrast, would be created after the author or authors had finished creating the text; it might be prepared for use in the playhouse, or made for presentation to an influential patron. Most of the fair copy manuscripts that survive were apparently created by professional scriveners. This contrast

A–Z OF KEY TERMS AND CONCEPTS

299

between fair and foul undergirds the critical narrative in manuscript studies. And yet, recent scholars have argued that early modern dramatic creativity tended not to consider plays finished, in the way that a modern-day book manuscript might be finished and sent to the publisher. Rather, flexible early modern attitudes towards books’ perfectibility meant that fair copies represent iterations – stops along the route – not final versions (Massai 2007: 1–38).

FOLIO (FORMAT) Every format used in the hand-press period emerged from the number of folds applied to a sheet of paper; paper folded once yielded two leaves and four pages. Since sheets came in different sizes, it is not quite accurate to say that the folio was the ‘largest’ format in terms of page dimensions, since quartos on large paper could be wider and/or taller than folios on small paper. Generally, however, folios were larger than other formats – and frequently thicker, too. As more page-units were impressed on a sheet, more white space would be taken up by margins. As a result, more type would fit into the four pages of a folio sheet than into the eight pages of a quarto sheet; the folio was thus the most economical option for printing particularly long texts (Galbraith 2010: 48–9). Some scholarship ascribes a level of cultural prestige to the folio format. Although certain folios did have an aura of respectability, any innate prestige attached to the format as a whole remains conjectural.

FORMAT G. Thomas Tanselle defines format as ‘a designation of the number of page-units … that the producers … decided upon to fill each side of a sheet of paper or vellum of the selected size’ (2000: 112–13). Folds and page-units are connected: if a printer imposed eight page-units on a sheet, he expected to fold it twice and make it a quarto.

FORME A cluster of set type ready for printing. A forme would print the whole of one side of a sheet of paper. Thus, a forme for a sheet in quarto would consist of four pageunits, oriented correctly for folding. Maintaining an efficient workflow in the print shop often required printing by formes, in which ‘the first side of the sheet is printed while the second side is being set’ (Werner 2019: 9). Printing by formes necessarily required that the copy be cast-off. See Copy, cast-off.

FOUL CASE After a forme had been printed, the type set in it would be distributed back into its cases. Type was a valuable print shop asset, and most shops could only afford to

300

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

keep a certain quantity on hand. Distribution therefore had to happen regularly, to maintain compositorial workflow. Type could end up in the wrong case – sometimes because sorts looked similar (capital I and miniscule l, miniscule e and miniscule a, miniscule n and miniscule u, long s and f, for instance), or sometimes due to spillage. These ‘foul cases’ could in turn cause compositorial error. By recognizing such errors, editors can sometimes resolve textual cruxes.

FOUL PAPERS This term referred in the period to an author’s rough draft, showing her or his second thoughts and revisions. Twentieth-century textual critics believed that certain early printed versions of Shakespeare’s plays used his foul papers as printer’s copy and thus provided a direct link to the playwright and a window into his creative process. More recent scholars have observed that many of the supposed characteristics of ‘foul papers’ appear in playhouse manuscripts as well (Werstine 2012).

GALLEY ‘[A] two- or three-sided wooden tray about the size of a page’ (Werner 2019: 15). As the compositor’s stick filled up, he gradually emptied it into a galley; when the galley was full, it was set aside until the rest of that forme could be set. Gaskell notes that the variance in papers’ sheet-size, as well as between the page-sizes of different formats, required an assortment of galley-sizes to accommodate the variety of pagedimensions (1995: 49–50).

GENETIC EDITION A representation of a text which emphasizes its evolutions over time. As opposed to the best-text approach, genetic editing privileges rough drafts, revisions and elisions as much as it does a good-quality ‘final’ textual version. Genetic editions work best when multiple versions of a text may be dated and sequenced, as (for example) in the case of James Joyce’s notebooks. For this reason, the potentialities of Shakespearean genetic editions are limited. However, in cases where a clear line of revision has been or can be argued, the genetic approach can be usefully deployed.

HANDWRITING While some letterforms in early modern handwriting more or less resemble modern letterforms, most are unfamiliar. In one widespread style in early modern England, called secretary hand, lowercase d and e looked similar, as did r, a and u. Textual errors in Shakespeare’s plays suggest that early modern compositors found the era’s handwriting difficult to decipher. In the handwritten pages of Sir Thomas More, which some believe to be in Shakespeare’s hand, the writer tended to close up his ‘u’s’ so that they look similar to ‘a’s’. Thus in Q2 Hamlet, we find that the queen appears as ‘Gertrad’ and that ‘devil’ (which Shakespeare tended to spell ‘deule’) appears as ‘deale’.

A–Z OF KEY TERMS AND CONCEPTS

301

IMPOSITION The term used for arranging set-up pages of type in the forme. For a quarto, the second and seventh pages would be oriented with their feet on the long, bottom edge of the forme; the third and sixth pages would be oriented with their feet on the long, top edge of the forme. The second page would thus be head to head with the third page, and the seventh page would be head to head with the sixth page. Then, when it came time to print the other side of the sheet, containing the first, fourth, fifth and eighth pages, their orientation would be similarly complex. After imposition, the forme was locked into a wooden frame called a chase and was ready for impression.

IMPRESSION This term has two meanings. The first, more immediate, refers to the act of pulling the press, transferring ink from type onto paper; ‘an impression’ in this case is one sheet which has had one side printed. More frequently, ‘impression’ is used in bibliographical description, to refer to all of an edition’s copies to be printed around the same time, from the same formes of type (Werner 2019: 87). Thus, copies of an edition calling itself ‘the first impression’ would be quite similar in their typography (though the presence of corrected and uncorrected states might prevent these copies from being completely identical). Copies of a second impression could aim to reproduce the exact line-by-line layout of the first impression (as was done in the Shakespeare Second Folio), but the necessity of resetting type would mean that the second would differ from the first impression. According to Gaskell, ‘[i]n the hand-press period … the edition and the impression were generally the same thing’; not until later technological innovations, after the hand-press was phased out, could there be multiple impressions of the same edition (Gaskell 1995: 314–15).

IMPRINT Many early modern title pages took a formulaic shape: the title was placed near the top of the page, the authorial attribution appeared somewhere in the middle and details about printing (including the people involved in the book’s manufacture and the place and year of origin) were located on the bottom. These details are collectively referred to as a book’s imprint. But, as Sarah Werner notes, any of the standard components could be falsified, and the division of manufactory labour is sometimes unclear (2019: 85–6). Usually, books were printed ‘by’ printers ‘for’ publishers. Many imprints also include a place of sale: ‘at the spred Eagle in Paules Church-yeard, ouer against the great North doore’ for the first quarto of Troilus and Cressida.

INCIDENTALS W. W. Greg uses the word ‘accidentals’ to describe those textual elements which  ‘affec[t] mainly [the text’s] formal presentation’ (1950/1: 21). William

302

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Shakespeare: A Textual Companion, part of the consistently innovative Complete Oxford Shakespeare, prefers ‘incidentals’, which it defines as ‘non-substantive’ (1987: 155). ‘Incidental’ perhaps carries less connotative baggage than ‘accidental’, particularly regarding compositorial intention and capability; yet it has not gained wide purchase in discourses of textual criticism. See accidentals.

INTERPOLATIONS Passages or textual elements that appear to have been inserted after a text’s original composition by an agent other than the text’s original author. Most interpolated text either is passed off as part of the original composition or is justified as an improvement. Developing consensus about the collaborative nature of creative agency in early modern drama complicates scholarly identifications of interpolation. For instance, are the sections of Measure for Measure that appear to be written by Thomas Middleton interpolation, revision, or the results of collaboration?

ISSUE An issue differs from an edition in that the majority of the type for the book was not reset. The 1609 quarto of Troilus and Cressida was originally issued with a title page mentioning its performance by the King’s Men; then it was reissued with a title page making no mention of performance, as well as a preface disavowing any performance history for the play.

ITALICS A popular font-family invented in late medieval Venice, italics were intended to mimic letterforms from Italian manuscript. In English printing, italic type slowly gained a foothold during the sixteenth century, though never coming close to replacing blackletter type. When, in the early 1590s, printers shifted somewhat abruptly toward printing in roman fonts, italic appeared somewhat more frequently, particularly in paratextual matters or to create typographical variation – on title pages, for instance. Early printed texts of Shakespeare often (though not exclusively) use italics for entrances and exits, stage directions, speech prefixes, proper names, quoted material and printed songs: see, for example, Feste’s song at the end of Twelfth Night in the First Folio, which is reproduced in italics.

LEMMA The base text, or copy-text, from which variants in other texts diverge. In the standard form of a collation note, the early text from which an edited text is drawn serves as the lemma, which is followed by variant readings in other texts, earlier and later.

A–Z OF KEY TERMS AND CONCEPTS

303

LICENSING After a play had been reviewed by governmental authorities, its publishers would take it to the officers of the Stationers’ Company, who would review the text in question and determine whether it unduly threatened any other publishers’ market share or vendible properties. If this review was favourable, the publisher would pay a licensing fee of sixpence for official permission to print the text. License did not constitute entrance; after getting permission to print a play, a publisher still had to register a claim to it. This two-factor process filled the coffers of the Stationers’ Company, reminded publishers of its power and – in theory at least – regulated predatory printing practices (Blayney 1997: 398–400).

LIGATURE A piece of type on which two or more letters are connected. Ligatures apparently descended from scribal habits designed to increase writing speed by eliminating pen lifts. They served a related purpose in early hand-press printing: by representing frequent combinations of letters, ligatures could speed up type-setting. Ligatures in Shakespearean printed texts include the familiar (such as fi, ft, fl or ff) and the archaic (such as long-s ligatures with i, t or l, or decorative ct).

MANUSCRIPT PLAY At some point, every printed play was a manuscript play. Authors, alone or in collaboration, wrote by hand, and those handwritten pages were circulated, perused, censored, licensed and reproduced. This circuit was taxing on the material artefact; the few surviving manuscripts show signs of hard use. Most of the manuscript plays still extant are scribal reproductions of plays’ original handwritten versions. In some cases, such as that of John of Bordeaux (Alnwick Castle MS 507), manuscripts were produced for use in the theatre (Werstine 2012: 250). In other cases, they were created as gifts for patrons or other interested parties; such seems to be the origin of the Bonduca manuscript (BL Add MS 36758; Werstine 2012: 329–30). A relatively large number of play manuscripts created for theatrical use survive, but it is unclear to what extent their specific features may be generalized to apply to dramatic manuscript culture as a whole.

MARGINALIA Handwritten marginalia in printed books are immensely valuable for histories of reading and culture. Recording as they do the often-intimate details of book owners’ use-patterns, manuscript marginalia offer startling glimpses of early readers’ habits. For instance, marginalia in the recently discovered First Folio in Saint-Omer, France, reveals that its Catholic owners performed certain plays with all-male companies, using the Folio as a promptbook (cf. Cottegnies and Cordonnier 2016: 1–21).

304

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

MASSED ENTRY A stage direction that lists all the characters who are to appear in a scene at the scene’s beginning – regardless of when the action calls for them to enter (Dessen 1996: 119). This practice can pose narrative problems if the plot calls for a character in a scene to be unaware of what happens earlier in that scene. A few of the plays in the Shakespeare First Folio feature massed entries: The Two Gentlemen of Verona, The Winter’s Tale and The Merry Wives of Windsor. According to T. H. HowardHill, the professional scrivener Ralph Crane – who created the manuscripts from which these plays were printed – was apparently the source of these massed entries (1972: 20–1, 79–80).

MASTER OF THE REVELS In the early years of Elizabeth’s reign, the court officer called the Master of the Revels was primarily responsible for organizing entertainments for the monarch, but the role was expanded dramatically in 1581 to include much wider control over theatrical activities in London (Dutton 1997: 294). Before a play could be premiered on the stage, the Master of the Revels had to read and approve it. Between 1606 and 1612, the Master of the Revels – Sir George Buc at the time – also held the exclusive right to approve plays for print (variously called ‘allowance’ or ‘authority’); after 1612, the position lost exclusivity but retained the right of approval (Blayney 1997: 397). Despite the breadth of the powers granted to the Master of the Revels, the relationship between this official and the playing companies evolved into one ‘neatly symbiotic, and mutually profitable’ (Dutton 2018: 67). The Master of the Revels needed players to provide entertainments; the players needed a governmental advocate against constant pressure from London’s city government. ‘[N]either wanted trouble’ (Stern 2009: 236).

MEMORIAL RECONSTRUCTION The originary hypothesis which proposes that early print editions of dramatic texts were assembled from the memories of those who participated in, or attended, theatrical performances of those plays. Also known as the reported text hypothesis, it has been broadly applied to a wide variety of texts considered ‘bad’; garbling, repetition or deletion of words, lines or speeches have been marshalled as evidence of memorial reconstruction. For instance, the first quarto of Hamlet (printed 1603) contains certain textual peculiarities which led some critics to argue that the copytext was reconstructed from actors’ memories of their part-roles. Modern textual criticism posits textual origins in memorial reconstruction much more cautiously than did the New Bibliographers, if at all – indeed, Zachary Lesser condemns it as obsolete, ‘rest[ing] on the … foundation of retrograde bibliography’ (2014: 50). In a monograph on Shakespearean Suspect Texts, Laurie Maguire questions whether ‘the [historical] phenomenon of memorial reconstruction’ even existed (1996: 5–9), concluding that while ‘a strong case can be made for memorial reconstruction’

A–Z OF KEY TERMS AND CONCEPTS

305

for four early modern dramatic texts, none are ‘unquestionably’ the product of memorial reconstruction (324).

MISLINEATION The practice, accidental or purposeful, of compositorial setting of verse lines as prose, or of introducing extra-metrical or anti-metrical line breaks. If accidental, mislineation could be due to unclear copy, which did not visually communicate the differences between prose and verse passages, or to compositorial misreading of copy. Purposeful mislineation, by contrast, allowed compositors to manipulate page space. More type fit in a prose line than in a verse line, so mislining verse as prose condenses text; introducing extra line breaks has the opposite effect, stringing out multiple lines of verse to fill a page.

MISPRINT An instance where the text as printed is demonstrably incorrect (as opposed to simply doubtful). Misprints could occur when a compositor took the wrong letter from a foul case, flipped a letter, skipped or repeated a line of copy, repeated or omitted a word, misread copy or faithfully set nonsense text. The peril of calling a crux a misprint lies in interpretively closing off the crux’s potential for meaning, as is the case in Othello where the quarto reads ‘Indian’ and the folio ‘Iudean’. One of these may be a misprint, but both readings can be defended.

MODERNIZATION One of the more hotly debated editorial practices, modernization involves ‘translating’ early modern orthography (spelling and punctuation conventions) into modern orthography. Proponents of modernization claim, justly, that modernized texts are easier for the non-specialist to read than old-spelling texts, which reproduce early modern orthographic details. Detractors claim – also justly – that modernization erases early modern verbal texture by flattening out details. A classic example of such flattening occurs with the word ‘travail’. Early modern readers could read this word as synonymous with ‘travel’ or with ‘struggle’; it is considered to be a variant spelling of the former. But modernizing ‘travail’ to ‘travel’ eliminates the rich double entendre.

OCTAVO Usually smaller than quarto and folio, larger than duodecimo, octavos printed sixteen page-units to the sheet, which was folded three times. Aldus Manutius, who helped to invent italics in sixteenth-century Venice, pioneered the marketing of easily portable octavo books for popular consumption. Playtexts were not printed as frequently in the octavo format as in the quarto format. The Database of Early English Playbooks (DEEP) lists ninety plays in octavo, as opposed to 812 in quarto,

306

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

printed between 1576 and 1642. Only two early dramatic texts of Shakespeare’s plays were printed in octavo: The True Tragedy of Richard Duke of York (3 Henry VI) in 1595 and Pericles in 1635.

PAGINATION A reference technology in print and manuscript books which allows and encourages non-linear reading. In the print shop, a space for pagination would be left in a skeleton, but page numbers would have to be reset for every new forme, and there are frequent errors in pagination. In the First Folio, the page numbers in Hamlet jump from 156 to 257, whereas Troilus and Cressida is left unpaginated.

PARALLEL TEXTS A textual arrangement which places variant texts side by side on the page, in order to illuminate the nature and extent of their variance. This arrangement represents with particular force the dramatic ramifications of omitted and interpolated text; a parallel-text edition of King Lear, comparing Q1 and F1, represents through white space the absence from Q1 of the Fool’s prophecy (Orgel 2002: 1529).

PERFECTING The term for printing a sheet’s second forme. After the first forme was printed on blank paper, those sheets were piled in a heap – printed on one side, still blank on the other. Because printers would moisten paper before printing, they could not leave the heap for too long, lest it begin to dry out, contract and require remoistening. So, after the first forme had been printed – one side of hundreds of sheets – pressmen would invert the heap and begin printing the second forme right away. This was particularly crucial in quartos and smaller formats, because if the second forme was printed awry, pages might not align with those on the other side of the forme. In later centuries, print shops with multiple presses would print one forme on one press and the second forme on the second press, but this does not seem to have been the norm in early modern print shops (Gaskell 1995: 131–3; Werner 2019: 59).

PLAYBOOK This term has two distinct meanings. One is synonymous with the ‘book of the play’ – the manuscript text which governed performance in the early modern theatre. Scholars long used the anachronistic term ‘promptbook’ to refer to this playhouse document, but modern usage prefers ‘playbook’. The other meaning of this term refers to stand-alone dramatic texts. Collections of plays, such as the Shakespeare First Folio and Ben Jonson’s Workes, are not usually called ‘playbooks’. Playbooks were often printed in quarto and octavo; most were slender. Evidence in some extant playbooks suggests that they were sold unbound but roughly stitched together, for later binding or compilation by their owners (Dane and Gillespie 2010: 25–45; Pratt 2015: 305). The relative ephemerality – and thus social position – of single-play

A–Z OF KEY TERMS AND CONCEPTS

307

playbooks has been much debated; some historians argue that playbooks worked in early modern English culture as mass market paperbacks do today, while others contend that playbooks were created to be building blocks for more permanent buyer-assembled collections (sammelbände). These debates connect to critical controversy about the financial opportunities or liabilities represented by playbooks. Peter Blayney argues that the printing of playbooks required serious financial outlay and thus constituted a risky proposition (1997: 410–16). Alan Farmer and Zachary Lesser disagree, maintaining that reprint rates demonstrate playbooks’ popularity, and therefore lower financial risk (2005: 1–32).

PLOT Tiffany Stern differentiates between the plot-scenario and the backstage-plot as different textual artefacts of the early modern English theatre. The former refers to the broad narrative framework, created first, then fleshed out with dialogue and concrete action. The latter refers to a type of document which likely hung offstage and allowed a prompter or ‘book-holder’ to direct backstage movement during performance. As Stern notes, most backstage-plots (whose alternate spelling, ‘plat’, connotes the careful order of a map), functioned largely as call-sheets, but also recorded cast lists. The relationship of the backstage-plot to other theatrical documents is unclear; since they were handwritten, this relationship likely varied according to their creators and the companies for whom they were created (2009: 1–35, 201–31).

PRESSMAN An employee in a print shop responsible for impressing inked type onto paper. Gaskell notes that while compositors would take stints at the press, ‘few pressmen could set type efficiently’; in other words, press-work required less skill than type-setting (1995: 172). Nevertheless, pressmen had to have some level of skill, particularly since their remit was most responsible for the mass production of the physical book. Pressmen had to prepare paper and ink the night before printing would take place; then, on the day of printing, they would carefully position formes of type on the press. Once all was made ready, a flurry of action: two pressmen working together would ink the type, set the paper in the press, pull the press, move the paper/type assemblage, pull the press a second time (to print the second half of the forme), move the paper/type assemblage out and remove the paper – in perhaps less than twenty seconds. Then they would repeat the process, again and again, producing as much as 250 sheets an hour. Presswork was hard work.

PRINTER The owner and (often) manager of a print shop in which books were manufactured; most printers were men, but a sizeable minority were women, often widows of printers. Some printers also acted as publishers, but many confined themselves to material manufacture. Printers provided the mechanisms for printing: they owned

308

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

presses and stocks of type, along with the ink balls used to apply the ink to type. They also supervised the labour necessary for operating print mechanisms: they hired compositors to set type and pressmen to pull the press. Printers assumed some responsibility for the financial efficiency and typographical accuracy of the finished product. They were invested in proofreading and correction, as well, since shoddy work could tarnish a printer’s reputation. Regardless of the size of their shop, printers had their hands full.

PROOFREADING The early modern English print shop required some measure of proofreading; examples of cancels demonstrate this. But the slow development of the English print industry meant that most shops were small, and thus could not afford to keep a corrector on staff. As a result, these duties often fell to other print shop personnel, such as a compositor, the overseer (the master-printer’s deputy) or the printer. Frequently, a proof-sheet would be pulled and proofread at the beginning of imposing a forme, though pressmen would continue printing while the proofreading took place. Even if errors were found in the proof-sheet, uncorrected sheets would be retained; correction would occur, and printing would continue. It is for this reason that no two copies of the First Folio are typographical duplicates: corrected sheets mingle with uncorrected sheets, such that every copy has a unique makeup.

PUBLISHER In early modern English printing, the publisher was ‘the prime mover’ of book publication (Blayney 1997: 391). Whereas the printer shepherded books through their material manufacture, the publisher handled the more interpersonal elements of books’ genesis. Acquiring copy, licensing the text and entering it to the Stationers’ Register were duties of the publisher; publishers often hired or partnered with printers, when they did not undertake both publishing and printing duties themselves. The publisher probably supplied the printer with paper. Then, the text having been printed, the publisher sold the print run wholesale to retail booksellers throughout London, England and Europe. If the book sold well, the publisher might make back their investment in a year or two; if it sold poorly, the book might be a financial loss.

PUNCTUATION A common account of early modern dramatic punctuation holds that it did rhetorical, rather than syntactic, work: that is, that punctuation represented the breaths, pauses and stops of vocalized speech rather than subordinating or coordinating elements of sentence structure. Certainly, to modern eyes, familiar with standardized syntactic punctuation, the pointing in early modern drama seems haphazard, unsystematic. This appearance is misleading, however, as William Sherman points out: early modern dramatists used punctuation deliberately for structural effect, and early modern compositors were capable of accurately setting complex, precise punctuation (2016: 303–23). However, many modern editors emend early texts’ serial colons

A–Z OF KEY TERMS AND CONCEPTS

309

and repetitive commas, modernizing them to fit twenty-first-century standards. This practice reduces the estrangement effect caused by unfamiliar punctuation habits, but in doing so it effaces the rich artistic potential of what was, after all, a cohesive system.

QUARTO Sheets for books printed in quarto format were folded twice, yielding a total of four leaves and eight page-units per sheet. A quarto leaf was thus about half the area of a folio leaf; though scholars tend to class quartos with octavos and duodecimos as ‘small formats’, quarto books are not necessarily small. Most early plays by Shakespeare, Jonson, Marlowe and others were first printed in quarto, and ‘playbook’ tacitly connotes ‘quarto’ to modern scholars. Recent scholarship has argued that quartos were not necessarily associated with ephemerality or disposability (Farmer 2016: 87–92).

QUIRE A group of sheets printed together, sharing a signature and folded together in the final assembly of the book. Folio books often contained quires of three sheets, folded with one on the outside, one intermediate and one inside. The outside sheet, signed A1 on its first-leaf recto and usually unsigned on its second-leaf recto, would print the quire’s first and second pages on its first leaf and the quire’s eleventh and twelfth pages on its second leaf. The intermediate sheet, signed A2 on its first-leaf recto and usually unsigned on its second-leaf recto, would print the quire’s third and fourth pages on its first leaf and the quire’s ninth and tenth pages on its second leaf. The inside sheet, signed A3 on its first-leaf recto and sometimes unsigned (or signed A4) on its second-leaf recto, would print the quire’s fifth and sixth pages on its first leaf and the quire’s seventh and eighth pages on its second leaf. Quarto and octavo quires usually were made up of only one or two sheets. In studies of compositor workload, stints are often measured in terms of quires set (e.g. Compositor C set most of AA and all of BB-EE).

RECTO/VERSO Terms that refer to the front page and back page of a leaf, respectively. These terms, often abbreviated as ‘r’ and ‘v’, are of particular value for textual and bibliographical description, since many early modern books are not paginated. Signatures mesh particularly well with recto/verso distinctions; together, they allow referential specificity in texts which can resist concrete reference.

REISSUE A publication, with a new title page, of sheets that had been previously sold under a different title page. At times, printers found themselves with a stock of printed copies bearing preliminary matter which was erroneous or outdated. The fourth quarto edition of Richard II provides a salutary example: when it was published

310

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

in 1608 by Mathew Law, its title page mentioned the Chamberlain’s Men – who had five years previously become the King’s Men. At some point thereafter, the same edition was reissued with a new, more detailed title page, which included an up-to-date reference to the King’s Men and specific mention of the ‘new additions’ incorporated into the text. The new and improved title page may have driven sales, or it may not; it was another seven years before the play was reprinted.

REPORTED TEXT Some playtexts as printed bear signs of being memorially reconstructed, or reported. For decades, scholars read the First Folio’s preliminary reference to ‘the frauds and stealthes of iniurious impostors’ (sig. πA3r) as a condemnation of audience members hearing, memorizing (or jotting down in shorthand) and then selling a play to a printer. More recently, consensus has settled on actors as originators of reported texts. The process of memorial reconstruction is now seen as one of recovery, when neither theatrical nor authorial manuscripts were available, instead of larceny (Blayney 1997: 393–94). While plays printed from reported texts do provide access to drama which might otherwise have been unavailable, the unreliability of human memory renders them often garbled, shortened or confusing – the traits which identify the reported text also make it more difficult to follow. See ‘Bad quarto’ and Memorial reconstruction.

REVISION Early modern authorship was characterized by collaboration and revision. No text was seen as definitive; every text was available for further perfection. The pecuniary benefits of the ‘new and improved’ editions drove this culture of permissive revision, but those benefits only accrued to the text’s publisher: revising authors, their labour often unacknowledged, worked for a flat fee, paid either by a publisher or (in drama) by a theatrical impresario. Remarkably, though the fact of pervasive revision is well known, the processes and extents of revision in early modern theatre are less clear. Major additions, as to The Spanish Tragedy, for instance, were probably written by a playwright other than the original author, possibly by Shakespeare. Smaller-scale alterations, as between the quarto and folio versions of King Lear, could either be authorial or theatrical in origin.

SAMMELBÄNDE Bound-together collections of works, usually brief, either print, manuscript or both. Selection of a sammelbänd’s contents, a practice resonant with medieval practices of compilatio, could be done by a bookseller, but was more usually the product of early modern book-buyers’ collecting instincts (Knight 2015: 77–95). Today, the relatively few intact sammelbände which survive, as well as those texts which used to be compiled (and are listed as such in catalogues), provide invaluable insight on what early modern booksellers and book-buyers considered contextually appropriate.

A–Z OF KEY TERMS AND CONCEPTS

311

SHORTHAND Any of several systems of abbreviated writing, most often used for rapidly recording ideas or information, or recording live speech. There were systems of shorthand used in Elizabethan England for recording sermons, and some have argued that shorthand was used for recording plays as well. Theatrical impresario Philip Henslowe used shorthand to record financial transactions connected to the Admiral’s Men; his shorthand notation ‘ne’, marked against certain plays in his diary, has been variously interpreted to mean ‘new’, ‘newly improved’, ‘new to the Admiral’s Men’ and ‘première performance’.

SIGNATURE A printed mark, usually in the lower right-hand corner of the recto, varying between quires; it allowed printers to identify where, in which quire, a printed sheet belonged. Signatures are the easiest way for scholars to see the length and limits of a quire, because each quire has its own signature; they are also handy units of length in calculating compositors’ stints. For longer works, textual preliminaries are often unsigned (as in the case of the First Folio of Shakespeare), because preliminaries would be imposed at the end of the production process. Whether preliminaries are signed or unsigned, signatures usually run A to Z, ‘omitting either I or J, U or V, and W’, then begin again with AA (Werner 2019: 97). Not every leaf is signed; in a three-sheet quire, the first three leaves would be signed A1, A2 and A3, the fourth leaf might be signed A4, and the fifth and sixth leaves would likely remain unsigned.

SKELETON While text obviously varied from page to page, certain typographical elements seldom, if ever, varied across the length of a book: rules, running heads, some ornaments and so on. In order to maintain a consistent page appearance, these elements would be retained in the forme after its text-type was stripped; so, too, were furniture used for spacing and orienting page numbers, signatures, catchwords and text-block. Anything which was reusable from previous formes was referred to as the skeleton; it became a semi-permanent template into which new text could be set, from which sheets could be printed, and from which already-impressed type could be removed. Since skeletons would be broken down if printing were to cease for any significant length of time, tracking changes in skeleton appearance can shed some light on the prehistory of books. As Gaskell cautions, though, ‘it is unwise to read too much significance into’ skeleton changes (1995: 110).

SPEECH PREFIX The abbreviation used in dramatic texts for assigning speeches to different characters, sometimes called a ‘speech heading’. The speech prefixes for a single character could vary greatly. Famously, Lady Capulet in Q2 Romeo and Juliet appears as Capulets Wife, Wife, Old Lady and Mother. Evidence in the The Booke of Sir Thomas More

312

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

manuscript reveals that precise speech prefixes were sometimes post-authorial additions, where ‘Hand D’, possibly Shakespeare, wrote simply ‘Other’ and a specific character’s name was later inserted by a theatrical scribe. On occasion, a speech prefix can foster uncertainty: in King Lear, does Cor. refer to Cornwall or Cordelia? Speech prefixes can provide valuable information about casting and performance history. We know that Will Kempe premiered the part of Dogberry in Much Ado About Nothing because actor-specific speech prefixes in manuscript copy (referring to Dogberry as ‘Kemp.’) were reproduced in print.

STAB-STITCHING An inexpensive, quick method of binding, in which strong thread was run through (usually three) needle holes in a small number of sheets. Given the ease with which a slender book could be stab-stitched, it seems to have happened in retail bookshops to prepare texts for binding. Because slender books used less paper than thick books, with a proportionally lower price point, some scholars have connected cheapness and ephemerality with stab-stitching; however, recent research suggests that this theory is too reductive (Pratt 2015). Many different kinds of books were stab-stitched, for different reasons. For example, buying a stab-stitched book, or having an unbound book stab-stitched, allowed frugal book-buyers to avoid more expensive individual bindings and save money for a sammelbänd. Few stab-stitched books remain in their original form today; most have been rebound, and tiny holes in the margin are all that marks a text’s stab-stitched prehistory.

STAGE DIRECTIONS Paratextual annotations in dramatic texts marking onstage action, descriptively or prescriptively (or some combination of the two). As with the speech prefix, the mechanical simplicity of the stage direction belies its ultimate ambiguity. Stage directions’ origins are uncertain; some must have emerged from the pen of the author(s), but others must have been produced by theatrical performance. A number of different conventional formulae govern stage directions’ syntax, vocabulary, tense and even placement on the page (Syme 2008: 148–9).

STATE Whereas different issues of the same text diverge for specific purposes, the variance between different states is, in Sarah Werner’s words, ‘those changes that the printer or publisher might not wish to call attention to’ (2019: 87). States are often differentiated in chronological terms; later states diverge from earlier states, corrected from uncorrected, fuller text from sparser text. Variant states might, for example, emerge as follows: in the midst of printing the second sheet of a three-sheet folio gathering, after 100 sheets have been printed, someone in the shop notices a flipped-letter error, u for n. The process is halted, the forme unlocked, the piece of type flipped and printing resumes. Fifty sheets later, someone notices another error: duplicated the. After the forme has been (re)corrected, the rest of the sheets

A–Z OF KEY TERMS AND CONCEPTS

313

are printed, and the stack now includes three states: uncorrected, once-corrected and twice-corrected. The repetition of this process across dozens or hundreds of formes printed yields a large number of potential permutations; for simplicity’s sake, scholars refer to sheets’ state, rather than books’ state.

STATIONERS’ COMPANY AND REGISTER Royally chartered in 1557, the Stationers’ Company was the guild governing publishers, printers and booksellers in early modern England. It exercised authority over the size and output of the print industry, operating as a vector of political control over the threat of social tumult posed by printing (Werner 2019: 25). It also maintained a stock of titles which could be assigned to poverty-stricken printers, to ameliorate their distress. In addition to restricting rates and types of print production, the Stationers’ Company offered legal protection to publishers who registered their ownership of texts – entrance of a text, in exchange for a fee paid to the Company, guaranteed the publisher-owner’s right to that text. The Register remains an important document for understanding the history of early modern English drama: when a publisher acquired the text of a play, they usually entered it in the Stationers’ Register, in order to protect what could be a profitable textual commodity.

TAGGING When applied to the study of the early modern dramatic text, this term can mean various things. One sense arises straightforwardly from analogy with modern textual practices: tagging as graffiti. For a graffitist, tagging involves the display of one’s individual, unique signature ‘in as many places as possible’; Jason Scott-Warren likens this modern habit to early modern book-owners’ and book-readers’ habit of ostentatiously signing their books (2010: 365–6). ‘Tagging’ can also refer to a key process in the creation of digital editions: the marking, through coding, of textual elements as corresponding to literary or dramatic genre markers. Sonia Massai uses the example of ‘opening sequences, or “incipits”’, across Shakespearean drama; tagging the opening scenes of King John and King Lear as incipits – an interpretive move in its own right – would invite comparison of those scenes, from two plays infrequently juxtaposed (2017: 69–72). As more mainstream editions migrate online or develop online components, robust digital tagging like that Massai describes will open unexpected new avenues for critical analysis.

TEXTUAL CRITICISM That branch of scholarly inquiry concerned with excavating the context, origin and history of texts. Textual criticism may engage with manuscript texts, manuscript annotations, typography, orthography, textual variance, compositorial analysis, textual dating and more. Careful consideration of the material and verbal facets of texts, framed by diverse documentary evidence, can illuminate the prehistory, and explain the afterlives, of those texts. When it is responsibly executed, textual criticism

314

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

provides invaluable insight into conditions of early modern print production, predispositions of readers in the early modern period and beyond, and cultural attitudes toward creativity and collaboration. In the last three decades, a growing consensus in textual criticism has emphasized the individuality of textual states and copies; in rejecting earlier notions of ‘ideal texts’, modern textual criticism revels in the generative mutability of early modern texts.

TITLE PAGE A paratextual feature that developed throughout the sixteenth century, and by the Elizabethan period had reached a recognizable form. Most title pages include, as a bare minimum, a title for the work to which they are attached; many list the printer’s and/or publisher’s name and provide a location at which the books can be purchased, and the date of publication. As authorial authority increased throughout the seventeenth century, title pages more frequently included an author’s name; texts in translation or with scholarly apparatus might list translators or annotators on a title page. Title pages sometimes, though not always, boasted eye-catching ornamentation: a woodcut referencing textual content (such as the one on the title page of the 1615 quarto edition of The Spanish Tragedy), a printer’s or publisher’s device (like Thomas Pavier’s punning emblem) or abstract typographical ornaments. The appearance of the title page served a variety of rhetorical ends: in addition to inviting the attention of browsers in retail bookshops, ‘[a]dvertisements for books consisted of title-pages separately printed and hung up on the posts of the city’ (Stern 2009: 55).

TRANSCRIPT A copy of a text that reproduces its verbal sense and orthographic idiosyncrasies but clarifies illegibility. Many textual scholars propose that ‘fair copy’ transcripts of the original manuscripts of Shakespeare’s plays may lie behind the printed versions.

TYPE ANALYSIS Comparative scrutiny of the incidence of individual type-sorts in a printed text. Type was cast from soft metal and used for a long time before being melted down and recast; these factors, coupled with the speed of print-shop work and the force with which the press was pulled, led to type damage. Usually, the extent of this damage was minimal; bent descenders or ascenders, chipped serifs or warped bodies might leave a sort legible, but unique. By tracking unique damaged sorts, scholars have been able to determine whether books were printed seriatim or by formes (Hinman 1963: 2.313–18; Blayney 1982: 89–150).

TYPOGRAPHY The design, incidence and meaning-making impressions made by inked type on paper. The makers of early modern English books used a variety of methods to create

A–Z OF KEY TERMS AND CONCEPTS

315

meaning; the word is the most obvious, and the image is fairly straightforward, but type font, size and disposition on the page could also communicate concrete concepts about the nature of the text and how it was to be used. The study of typography also involves characters other than letters – punctuation marks, ornaments, numbers and symbols – as well as print elements such as rules. One challenge of studying typography is distinguishing between what Antony Hammond calls ‘signal’ and ‘noise’ (1988: 207–10). Signal transmits meaning, originating from author, publisher or compositor; noise risks obscuring meaning. Differentiating between symbol and noise can be difficult. For example, in Hamlet’s line ‘Not so my Lord, I am too much i’th’ Sun’, it is unclear whether the First Folio capitalizes ‘Lord’ and ‘Sun’ deliberately (sig. nn5v). Some might see the capital letters as typographical signal, recording early modern stage habits of emphasis (‘Lord’ and ‘Sun’ being singled out), while others might see them as typographical noise, the result of individual compositors’ type-setting decisions. Sometimes, typographical detail carries intentional meaning; sometimes, it generates meaning unintentionally; sometimes, it is simply a vehicle for the text.

UNEDITING The practice of attending to and valourizing early modern textual texture in editing – things like syntax, orthography, punctuation, variant states and obscure phrases. As popularized by Leah Marcus and Randall McLeod, unediting distances itself from earlier, more idealistic models of textual editing. To unedit, as Marcus describes it, is to reconsider the potentiality for meaning latent in passages or patterns which have traditionally been labelled cruxes; it is to reflect with generosity, rather than dismissal, on the validity of meaning made outside of modern rational editorial thought (1996: 1–37). In practice, unediting takes aim at editorial practices like emendation and modernization, arguing instead for the generative capacity of early modern textual conventions. Critics of unediting argue that its ‘text-centred fundamentalism’ defines early modern material texts as static, while they were in fact waypoints in trajectories of improvement and collaboration (Massai 2007: 204).

VARIANT A difference between one text and another. In order for a variant to exist, the text must exist in multiple surviving states; thus, we can talk of textual variants in Folio Titus Andronicus (since it existed previously in quarto), but not in Folio Macbeth (of which no prior printed edition survives). The origin and nature of textual variants is at the very heart of textual criticism: much critical and editorial energy is spent determining where variants came from, how they affect meaning, whether they occurred as emendation or error, to what extent they improve, debase or change the text, and what clues they provide about the textual and performative prehistory of early modern plays.

316

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

VARIORUM EDITION An edition that aims ‘to provide a detailed history of critical commentary together with an exhaustive study of the text’ (Knowles 2003: 1). A variorum collects and curates relevant scholarly perspectives on a text or group of texts. Condensing more than four centuries of criticism into a wieldy volume is a difficult proposition, requiring decades of work, an intimate knowledge of the critical history of the text(s) being edited and a keen sense of which insights deserve to be reproduced. The edition that results from this process aims to archive everything worthwhile that has been said about a text or texts since their creation. The first variorum editions of Shakespeare were produced in the late eighteenth century; a century later, H.  H. Furness compiled variorum editions of single plays, which he called the New Variorum. The Modern Language Association supported a New Variorum project for decades, and they are now published under the auspices of the Center of Digital Humanities Research at Texas A&M University.

VERSIONING An editorial mode which presents widely divergent texts of a work as discrete from each other. As opposed to conflation and best-text editing, versioning emphasizes the structural integrity and relative independence of widely variant texts. Where the lemma/variant relationship is clear, versioning is less necessary (though potentially still informative). When textual authority is undefined, however, versioning is essential: texts like Hamlet and King Lear, which exist in forms divergent in largeand small-scale ways, profit immensely from this practice. The development of digital editions has increased the viability of versioning as an editorial practice. When space no longer limits editorial inclusions, edited texts for each of multiple variants can be produced, benefiting scholarship and study.

WATERMARK When paper was made by hand, the wire screen upon which wet slurry dried into sheets of paper was frequently decorated with simple ornaments woven in wire into the vertical and horizontal screen-wires. Ornaments’ designs were not unique to individual manufacturers – paper marked with a pot motif or a crown motif might have come from any number of manufacturers (Hunter 1943: 262–4). But while designs were generic, damage was not: the wire woven into the screens to form the ornament shape often broke as moulds were used and reused. Attending to the incidence and frequency of uniquely damaged watermarks allows a degree of specificity in determining the sources of stocks of paper. It also allows scholars to track paper use-patterns within print shops and further illuminate early modern print practices.

REFERENCES Bell, Maureen (1994), ‘Entrance in the Stationers’ Register’, The Library 6th ser., 16 (1): 50–4.

A–Z OF KEY TERMS AND CONCEPTS

317

Bland, Mark (2013), A Guide to Early Printed Books and Manuscripts, Chichester: WileyBlackwell. Blayney, Peter W. M. (1982), The Texts of ‘King Lear’ and their Origins: Nicholas Okes and the First Quarto, Cambridge: Cambridge University Press. Blayney, Peter W. M. (1997), ‘The Publication of Playbooks’, in John D. Cox and David Scott Kastan (eds), A New History of Early English Drama, 383–422, New York: Columbia University Press. Bradbrook, M. C. (1962), Elizabethan Stage Conditions: A Study of Their Place in the Interpretation of Shakespeare’s Plays, Hamden, CO: Archon. Cottegnies, Line and Rémy Cordonnier (2016), ‘New Discoveries about the Saint-Omer Folio: The Signification of the PS Stamp’, Études Épistémè, 29: 1–21. Dane, Joseph A. and Alexandra Gillespie (2010), ‘The Myth of the Cheap Quarto’, in John N. King (ed.), Tudor Books and Readers: Materiality and the Construction of Meaning, 46–67, Cambridge: Cambridge University Press. Dessen, Alan C. (1996), ‘Massed Entries and Theatrical Options in The Winter’s Tale’, Medieval and Renaissance Drama in England, 8: 119–27. Dutton, Richard (1997), ‘Censorship’, in John D. Cox and David Scott Kastan (eds), A New History of Early English Drama, 287–304, New York: Columbia University Press. Dutton, Richard (2018), ‘The Limits of a Censor’s Authority: The Case of the Masters of the Revels’, in Sophie Chiari (ed.), Freedom and Censorship in Early Modern English Literature, 66–78, New York: Routledge. Farmer, Alan B. (2016), ‘Playbooks and the Question of Ephemerality’, in Heidi Brayman, Jesse M. Lander and Zachary Lesser (eds), The Book in History, the Book as History: New Intersections of the Material Text, 87–125, New Haven, CT: Beinecke Rare Book and Manuscript Library. Farmer, Alan B. and Zachary Lesser (2005), ‘The Popularity of Playbooks Revisited’, Shakespeare Quarterly, 56 (1): 1–32. Galbraith, Steven K. (2010), ‘English Literary Folios 1593–1623: Studying Shifts in Format’, in John N. King (ed.), Tudor Books and Readers: Materiality and the Construction of Meaning, 46–67, Cambridge: Cambridge University Press. Gaskell, Philip (1995), A New Introduction to Bibliography, New Castle, DE: Oak Knoll. Greg, W. W. (1950/51), ‘The Rationale of Copy-Text’, Studies in Bibliography, 3: 19–36. Greg, W. W. (1956), Some Aspects and Problems of London Publishing Between 1550 and 1650, Oxford: Clarendon Press. Hammond, Antony (1988), ‘The Noisy Comma: Searching for the Signal in Renaissance Dramatic Texts’, in Randall McLeod (ed.), Crisis in Editing: Texts of the English Renaissance, 203–50, New York: AMS Press. Hinman, Charlton (1963), The Printing and Proofreading of the First Folio of Shakespeare, 2 vols, Oxford: Clarendon Press. Howard-Hill, T. H. (1972), Ralph Crane and Some Shakespeare First Folio Comedies, Charlottesville, VA: University Press of Virginia. Hunter, Dard (1943), Papermaking: The History and Technique of an Ancient Craft, New York: A. A. Knopf. Knight, Jeffrey Todd (2015), ‘Organizing Manuscript and Print: From Compilatio to Compilation’, in Michael Johnston and Michael Van Dussen (eds), The Medieval Manuscript Book: Cultural Approaches, 77–95, Cambridge: Cambridge University Press.

318

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Knowles, Richard (2003), Shakespeare Variorum Handbook: A Manual of Editorial Practice, New York: Modern Language Association. Lesser, Zachary (2014), Hamlet after Q1: An Uncanny History of the Shakespearean Text, Philadelphia: University of Pennsylvania Press. Maguire, Laurie E. (1996), Shakespearean Suspect Texts: The ‘Bad’ Quartos and Their Contexts, Cambridge: Cambridge University Press. Marcus, Leah S. (1996), Unediting the Renaissance: Shakespeare, Marlowe, Milton, New York: Routledge. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Massai, Sonia (2017), ‘Editing Shakespeare in Parts’, Shakespeare Quarterly, 68 (1): 56–79. Murphy, Andrew (2003), Shakespeare in Print: A History and Chronology of Shakespeare Publishing, Cambridge: Cambridge University Press. Orgel, Stephen, ed. (2002), King Lear: The 1608 Quarto and 1623 Folio Texts, in A. R. Braunmuller and Stephen Orgel (eds), The New Pelican Shakespeare, 1480–572, New York: Penguin. Pratt, Aaron T. (2015), ‘Stab-Stitching and the Status of Early English Playbooks as Literature’, The Library 7th ser., 16 (3): 304–28. Rasmussen, Eric (1998), ‘Anonymity and the Erasure of Shakespeare’s First EighteenthCentury Editor’, in Joanna Gondris (ed.), Reading Readings: Essays on Shakespeare Editing in the Eighteenth Century, 318–23, Madison, NJ: Fairleigh Dickinson University Press. Scott-Warren, Jason (2010), ‘Reading Graffiti in the Early Modern Book’, The Huntington Library Quarterly, 72 (3): 363–81. Shakespeare, William (1623), Mr. William Shakespeares Comedies, Histories, & Tragedies, London: Isaac Jaggard, Edward Blount, John Smithweeke and William Aspley. Sherman, William H. (2016), ‘Early Modern Punctuation and Modern Editions: Shakespeare’s Serial Colon’, in Heidi Brayman, Jesse M. Lander and Zachary Lesser (eds), The Book in History, the Book as History: New Intersections of the Material Text, 303–23, New Haven, CT: Beinecke Rare Book and Manuscript Library. Stern, Tiffany (2009), Documents of Performance in Early Modern England, Cambridge: Cambridge University Press. Syme, Holger (2008), ‘Unediting the Margin: Jonson, Marston, and the Theatrical Page’, English Literary Renaissance, 38 (1): 142–71. Tanselle, G. Thomas (1994), ‘Editing without a Copy-Text’, Studies in Bibliography, 47: 1–22. Tanselle, G. Thomas (2000), ‘The Concept of Format’, Studies in Bibliography, 53: 67–115. Theobald, Lewis, ed. (1733), The Works of Shakespeare, 7 vols, London: Bettesworth, Hitch, Tonson, Clay, Feales and Wellington. Tucker, Patrick (2002), Secrets of Acting Shakespeare: The Original Approach, New York: Routledge.

A–Z OF KEY TERMS AND CONCEPTS

319

Weiss, Adrian (2007), ‘Casting Compositors, Foul Cases, and Skeletons: Printing in Middleton’s Age’, in Gary Taylor and John Lavagnino (eds), Thomas Middleton and Early Modern Textual Culture: A Companion to the Collected Works, 195–225, Oxford: Clarendon Press. Wells, Stanley and Gary Taylor, with John Jowett and William Montgomery (1987), William Shakespeare: A Textual Companion, Oxford: Clarendon Press. Werner, Sarah (2019), Studying Early Printed Books 1450–1800: A Practical Guide, Chichester: John Wiley & Sons. Werstine, Paul (2012), Early Modern Playhouse Manuscripts and the Editing of Shakespeare, Cambridge: Cambridge University Press.

CHAPTER 4.3

Annotated bibliography JEAN-CHRISTOPHE MAYER

Shakespeare and textual studies is a vast scholarly field, and an annotated bibliography like the present one is inevitably partial, in both senses of that word. Other scholars would no doubt omit some of the items that have been included below and substitute others that have not. The subscription-only online database The World Shakespeare Bibliography Online, www.worldshakesbib.org, provides the most comprehensive record of Shakespeare scholarship for the period from 1960. For a detailed bibliography of Shakespeare and textual studies up to the end of the twentieth century, see T. H. Howard-Hill’s, Shakespearian Bibliography and Textual Criticism: A Bibliography, 2nd edn (Signal Mountain, TN: Summertown, 2000). The present bibliography is divided into eight sections: I. Shakespeare and textual studies: General II. Shakespeare’s early books III. Shakespeare’s early texts IV. Shakespeare’s early bibliographic and textual reception V. Authorship and the Shakespeare canon VI. Editing Shakespeare VII. Modern Shakespeare editions VIII. Beyond Shakespeare: Bibliography, book history and the book trade The first section, ‘Shakespeare and textual studies: General’, lists items that span several of the main fields of Shakespeare textual studies, including collections that do so. The second section, ‘Shakespeare’s early books’, deals with the latesixteenth- and seventeenth-century editions of Shakespeare’s plays and poems, in particular their publication, the agents (printers, publishers, booksellers) involved therein and their bibliographic make-up, including their paratext. The third section, ‘Shakespeare’s early texts’, contains items about the make-up and provenance of Shakespeare’s texts, including the perceived nature of the lost manuscripts on which the printed texts are based, the New Bibliography and reactions to it, variant early texts (of King Lear, Othello, etc.), the ‘bad quartos’ (of Hamlet, Henry V, etc.) and theories of their origins, dramatic documents made for use in the playhouse, and factors that had, or may have had, an impact on at least some of Shakespeare’s play texts (e.g. censorship or authorial revision). The fourth section, ‘Shakespeare’s early bibliographic and textual reception’, lists

ANNOTATED BIBLIOGRAPHY

321

scholarship devoted to how early modern (and, occasionally, eighteenth-century) readers, actors, collectors and book-owners engaged with Shakespeare’s texts and books, including by annotating, excerpting, commonplacing and anthologizing the text, by producing promptbooks and by assembling and binding the books. The fifth section, ‘Authorship and the Shakespeare canon’, is concerned with what is (and was) considered to be by Shakespeare, and why. It includes items about modern and historical thinking about the Shakespeare canon, Shakespearean authorship, collaboration and co-authorship, and authorship attribution studies. The sixth section, ‘Editing Shakespeare’, lists publications that are concerned with the editorial mediation of Shakespeare’s texts, mostly from a modern but also from a historical perspective. The seventh section, ‘Modern Shakespeare editions’, lists a selection of relatively recent editions, especially those that incorporate and have a bearing on important textual scholarship. The eighth section, ‘Beyond Shakespeare: Bibliography, book history and the book trade’, includes items that do not focus on Shakespeare but are nonetheless important reading for students of Shakespeare and textual studies. Many items do not fit neatly into one of the eight sections, but have a claim to being included in at least one other. In those cases, we list the number assigned to the items at the end of the sections in which they are not included.

I SHAKESPEARE AND TEXTUAL STUDIES: GENERAL 1. Callaghan, Dympna and Suzanne Gossett, eds (2016), Shakespeare in Our Time: A Shakespeare Association of America Collection, London: Bloomsbury. Chapters by (a) Barbara A. Mowat (‘Facts, Theories, and Beliefs’, 57–64), (b) Lukas Erne (‘What We Owe to Editors’, 64–8), (c) Sonia Massai (‘What’s Next in Editing Shakespeare’, 68–72), (d) Gary Taylor (‘Collaboration 2016’, 141–9), (e) Laurie Maguire (‘The Value of Stage Directions’, 149–53), (f) Adam G. Hooks (‘The Author Being Dead’, 153–7). 2. Cox, John D. and David Scott Kastan, eds (1997), A New History of Early English Drama, New York: Columbia University Press. Chapters by (a) Heidi Brayman Hackel (‘“Rowme” of Its Own: Printed Drama in Early Libraries’, 113–30), (b) Barbara Mowat (‘The Theater and Literary Culture’, 213–30), (c) Jeffrey Masten (‘Authorship and Collaboration’, 357–82), (d) Peter W. M. Blayney (‘The Publication of Playbooks’, 383–422), (e) Eric Rasmussen (‘The Revision of Scripts’, 441–60), (f) Paul Werstine, (‘Plays in Manuscript’, 481–97). 3. Dutton, Richard and Jean Howard, eds (2003), A Companion to Shakespeare’s Works, 4 vols, Oxford: Blackwell. Chapters by (a) Sasha Roberts (‘Reading Shakespeare’s Tragedies of Love: Romeo and Juliet, Othello, and Antony and Cleopatra in Early Modern England’, 1.108–33), (b) Graham Holderness (‘Text and Tragedy’, 1.158–77), (c) Cyndia Susan Clegg (‘Censorship and the Problems with History in Shakespeare’s England’, 2.48–69), (d) James Siemon (‘“The Power of Hope?” An Early Modern Reader of Richard III’, 2.361–78), (e) John Jowett (‘Varieties of Collaboration in Shakespeare’s Problem Plays and Late Plays’, 4.106–28).

322

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

4. Giddens, Eugene (2011), How to Read a Shakespearean Play Text, Cambridge: Cambridge University Press. Focuses on both early texts and modern editions and shows how their make-up can affect reading and performance. 5. Jowett, John (2019), Shakespeare and Text: Revised Edition, Oxford Shakespeare Topics, Oxford: Oxford University Press. Offers comprehensive analyses of theatrical manuscripts, early printed editions, the modernization and editing of the text, digital editing as well as reading habits. 6. Kastan, David Scott, ed. (1999), A Companion to Shakespeare, Oxford: Blackwell. Chapters by (a) Heidi Brayman Hackel (‘“The Great Variety of Readers” and Early Modern Reading Practices’, 139–57), (b) Thomas L. Berger (‘Shakespeare in Print, 1593–1640’, 395–413), (c) William B. Long (‘“Precious Few”: English Manuscript Playbooks’, 414–33), (d) Laurie Maguire (‘The Craft of Printing (1600)’, 434–49), (e) Mark Bland (‘The London Book Trade in 1600’, 450–63), (f) Cyndia Susan Clegg (‘Liberty, Licence, and Authority: Press Censorship and Shakespeare’, 464–85). 7. Kastan, David Scott (2001), Shakespeare and the Book, Cambridge: Cambridge University Press. Explains how Shakespeare’s plays were transformed from performance texts into books, with an emphasis on the construction of Shakespeare by successive editions and electronic media. 8. Kidnie, Margaret Jane and Sonia Massai, eds (2015), Shakespeare and Textual Studies, Cambridge: Cambridge University Press. Chapters by (a) Heather Hirschfeld (‘Playwriting in Shakespeare’s Time: Authorship, Collaboration, and Attribution’, 13–26), (b) Paul Werstine (‘Ralph Crane and Edward Knight: Professional Scribe and King’s Men’s Bookkeeper’, 27–38), (c) James Purkis (‘Shakespeare’s “Strayng” Manuscripts’, 39– 54), (d) Sonia Massai (‘The Mixed Fortunes of Shakespeare in Print’, 57–68), (e) Helen Smith (‘“To London all”?: Mapping Shakespeare in Print, 1593–1598’, 69–86), (f) Alan B. Farmer (‘Shakespeare as Leading Playwright in Print, 1598–1608/9’, 87–104), (g) Zachary Lesser and Peter Stallybrass (‘Shakespeare between Pamphlet and Book, 1608–1619’, 105–33), (h) Emma Smith (‘The Canonization of Shakespeare in Print, 1623’, 134–46), (i) Laura Estill (‘Commonplacing Readers’, 149–62), (j) Jean-Christophe Mayer (‘Annotating and Transcribing for the Theatre: Shakespeare’s Early Modern Reader-Revisers at Work’, 163–76), (k) Jeffrey Todd Knight (‘Shakespeare and the Collection: Reading beyond Readers’ Marks, 177–95), (l) Alan Galey (‘Encoding as Editing as Reading’, 196–211), (m) W. B. Worthen (‘Shax the App’, 212–29), (n) Peter Holland (‘Theatre Editions’, 233–48), (o) Keir Elam (‘Editing Shakespeare by Pictures: Illustrated Editions’, 249–68), (p) Andrew Murphy (‘Format and Readerships’, 269–84), (q) Leah S. Marcus (‘A Man Who Needs No Introduction’, 285–99), (r) Lukas Erne (‘Emendation and the Editorial Reconfiguration of Shakespeare’, 300–13), (s) John Jowett (‘Full Pricks and Great P’s: Spellings, Punctuation, Accidentals’, 317–31), (t) Alan C. Dessen (‘Divided Shakespeare: Configuring Acts and Scenes’, 332–41), (u) Matthew Dimmock (‘Shakespeare’s Strange Tongues: Editors and the “Foreign”

ANNOTATED BIBLIOGRAPHY

9.

10.

11.

12.

13.

323

Voice in Shakespearean Drama’, 342–57), (v) Tiffany Stern (‘Before the Beginning; After the End: When Did Plays Start and Stop’, 358–74), (w) Jill L. Levenson (‘Framing Shakespeare: Introductions and Commentary in Critical Editions of the Plays’, 377–90), (x) Eric Rasmussen (‘Editorial Memory: The Origin and Evolution of Collation Notes’, 391–7), (y) David Weinberger (‘Shakespeare as Network’, 398–414). Kinney, Arthur F., ed. (2012), The Oxford Handbook of Shakespeare, Oxford: Oxford University Press. Chapters by (a) Hugh Craig (‘Authorship’, 15–30), (b) MacDonald P. Jackson (‘Collaboration’, 31–52), (c) Arthur F. Marotti and Laura Estill (‘Manuscript Circulation’, 53–70), (d) Ann Thompson (‘Quarto and Folio’, 71–84), (e) Grace Ioppolo (‘Revision’, 85–99), (f) Adam G. Hooks (‘Book Trade’, 126–42), (g) Sonia Massai (‘Early Readers’, 143–61). Murphy, Andrew, ed. (2000), The Renaissance Text: Theory, Editing, Textuality, Manchester: Manchester University Press. Includes four chapters that deal importantly with Shakespeare, by (a) Michael Steppat (‘(Un)Editing and Textual Theory: Positioning the Reader’, 73–90), (b) Peter Stallybrass (‘Naming, Renaming and Unnaming in the Shakespearean Quartos and Folio’, 108–34), (c) Laurie Maguire (‘Composition/ Decomposition: Singular Shakespeare and the Death of the Author’, 135–53), (d) Andrew Murphy (‘Texts and Textualities: A Shakespearean History’, 191–210). Murphy, Andrew, ed. (2007), A Concise Companion to Shakespeare and the Text, Oxford: Blackwell. Chapters by (a) Helen Smith (‘The Publishing Trade in Shakespeare’s Time’, 17–34), (b) Peter Stallybrass and Roger Chartier (‘Reading and Authorship: The Circulation of Shakespeare 1590–1619’, 35–56), (c) Thomas L. Berger (‘Shakespeare Writ Small: Early Single Editions of Shakespeare’s Plays’, 57–70), (d) Anthony James West, (‘The Life of the First Folio in the Seventeenth and Eighteenth Centuries’, 71–90), (e) Andrew Murphy (‘The Birth of the Editor’, 93–108), (f) Paul Werstine (‘The Science of Editing’, 109–27), (g) Leah S. Marcus (‘Editing Shakespeare in a Postmodern Age’, 128–44), (h) Michael Best (‘Shakespeare and the Electronic Text’, 145–61), (i) David Bevington (‘Working with the Text: Editing in Practice’, 165–84), (j) Sonia Massai (‘Working with the Texts: Differential Readings’, 185–203), (k) Neil Rhodes (‘Mapping Shakespeare’s Contexts: Doing Things with Databases’, 204–20). Smith, Bruce R., ed. (2016), The Cambridge Guide to the Worlds of Shakespeare: Shakespeare’s World, 1500–1660, Cambridge: Cambridge University Press. Chapters by (a) Mark Bland (‘The London Book Trade’, 335–40), (b) Carter R. Hailey (‘Paper’, 347–54), (c) Jean-Christophe Mayer (‘Reading Practices’, 354–60), (d) Alan Stewart (‘Manuscript Culture’, 360–66), (e) Adam Hooks (‘The First Folio’, 366–73). Smith, Emma (2016), ed., The Cambridge Companion to Shakespeare’s First Folio, Cambridge: Cambridge University Press. Chapters by (a) Tara L. Lyons (‘Shakespeare in Print Before 1623’, 1–17), (b) Eric Rasmussen

324

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

(‘Publishing the First Folio’, 18–29), (c) B. D. R. Higgins (‘Printing the First Folio’, 30–47), (d) Chris Laoutaris (‘The Prefatorial Material’, 48–67), (e) Gabriel Egan (‘The Provenance of the Folio Texts’, 68–85), (f) Peter Kirwan (‘“Complete” Works: The Folio and All of Shakespeare’, 86–102), (g) Jean-Christophe Mayer (‘Early Buyers and Readers’, 103– 19), (h) Edmund G. C. King (‘Editors’, 120–36), (i) Steven K. Galbraith (‘Collectors’, 137–54), (j) Emma Smith (‘Reading the First Folio’, 155–69), (k) Sarah Werner (‘Digital First Folios’, 170–84), (l) Adam G. Hooks (‘Afterword: the Folio as Fetish’, 185–96).

II SHAKESPEARE’S EARLY BOOKS 14. Blayney, Peter W. M. (1991), The First Folio of Shakespeare, Washington, DC: Folger Library Publications. A Folger Library exhibition catalogue in which the author explains how the First Folio (1623) was made and received by its early readers. 15. Blayney, Peter W. M. (2005), ‘The Alleged Popularity of Playbooks’, Shakespeare Quarterly 56 (1): 33–50. Responds to an earlier article (see §26) by arguing for the relative unpopularity of playbooks in the early modern book trade, including Shakespeare’s. (See also §27.) 16. Blayney, Peter W. M. (2016), ‘The Publication of Shakespeare’, in Peter W. M. Blayney, Alan Galey and Marjorie Rubright (eds), ‘So Long Lives This’: A Celebration of Shakespeare’s Life and Works 1616–2016, 17–26, Toronto, ON: University of Toronto Press. Focuses on the publication history of the First Folio and on William Jaggard’s printing house that produced the volume. 17. Connor, Francis X. (2014), Literary Folios and Ideas of the Book in Early Modern England, New York: Palgrave Macmillan. Compares Shakespeare’s First Folio (1623) to folios by other playwrights and poets, arguing that Shakespeare’s volume was ‘theatrical’ (see esp. 121–66). 18. Depledge, Emma (2018), Shakespeare’s Rise to Cultural Prominence: Print, Politics and Alteration, 1642–1700, Cambridge: Cambridge University Press. Sees Britain’s Exclusion Crisis (1678–82) as a fundamental moment in the history of Shakespeare in print in particular. The political crisis over matters of succession to the throne spawned numerous reprints, stage revivals and adaptations of Shakespeare’s work, which, Depledge argues, strongly anchored him as a national author. 19. Depledge, Emma, and Peter Kirwan, eds (2017), Canonising Shakespeare: Stationers and the Book Trade 1640–1740, Cambridge: Cambridge University Press. Fourteen chapters exploring the publication and editorial reproduction of Shakespeare’s plays and poems between 1640 and 1740. 20. Dugas, Don-John (2006), Marketing the Bard: Shakespeare in Performance and Print, 1660–1740, Columbia, MI: University of Missouri Press. Argues that commercial factors explain why numerous Shakespeare plays began reappearing in the early eighteenth century

ANNOTATED BIBLIOGRAPHY

21.

22.

23.

24.

25.

26.

27.

325

and why he became such a renowned author both in print and onstage by 1740. Investigates the theatre business, focusing on the possession and adaptation of Shakespeare’s plays by the Tonson publishing cartel. The latter used different marketing strategies (including editions of variable sizes and quality), which partly brought prices down and potentially opened up the market to new readers. See esp. chapters 2, 3 and 4. Erne, Lukas (2013), Shakespeare as Literary Dramatist, 2nd edn, Cambridge: Cambridge University Press. Argues that Shakespeare, apart from being a playwright who wrote theatrical texts for the stage, was also a literary dramatist who produced reading texts for the page. Examining the evidence from early published playbooks, the author argues that Shakespeare wrote many of his plays with a readership in mind. See esp. ‘Part I – Publication’, 53–152. Erne, Lukas (2013), Shakespeare and the Book Trade, Cambridge: Cambridge University Press. Offers an in-depth study of Shakespeare’s quartos published between 1594 and 1622 which reveals the extent of his early textual presence both as a printing house commodity and as a literary author with an already relatively important readership. According to Erne, this was in accord with what Shakespeare had wished for his work. Erne, Lukas and Tamsin Badcoe (2014), ‘Shakespeare and the Popularity of Poetry Books in Print, 1583–1622’, Review of English Studies, 65: 33–57. Focuses on Venus and Adonis, Rape of Lucrece and Sonnets to reveal that Venus and Adonis was the best-selling poetry book of its time. Farmer, Alan B. (2016), ‘Playbooks and the Question of Ephemerality’, in Heidi Brayman, Jesse M. Lander, and Zachary Lesser (eds), The Book in History, The Book as History: New Intersections of the Material Text, 87–125, New Haven, CT: Beinecke Rare Book and Manuscript Library. Uses large-scale quantitative data to determine if playbooks (including Shakespeare’s) were considered as ephemera by early modern publishers, booksellers and readers. Concludes that ‘the bibliographical ephemerality of playbooks is an idea more grounded in assumptions than in facts’ (116). Farmer, Alan B. and Zachary Lesser (2000), ‘Vile Arts: The Marketing of English Printed Drama, 1512–1660’, Research Opportunities in Renaissance Drama, 39: 77–165. Examines the marketing of drama (including Shakespeare’s) through the use of title-page attributions of author, authorial status, theatrical venue, theatrical company and the presence of Latin on the title page. Farmer, Alan B. and Zachary Lesser (2005), ‘The Popularity of Playbooks Revisited’, Shakespeare Quarterly, 56 (1): 1–32. Offers a comprehensive review of the critical literature on the publication and alleged popularity of playbooks, including Shakespeare’s (esp. by Blayney – see §2d), and argues against the contention that plays were an insignificant and unsuccessful portion of the market for books in early modern England. Farmer, Alan B. and Zachary Lesser (2005), ‘Structures of Popularity in the Early Modern Book Trade’, Shakespeare Quarterly, 56 (2): 206–13.

326

28.

29.

30.

31.

32.

33.

34.

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Responding to an earlier argument made by Blayney on the unpopularity of playbooks, including Shakespeare’s (see §15), argues that in order to understand popularity in the book trade, we need to consider the total number of editions and frequency of reprinting, as well as market share and profitability. Farmer, Alan B. and Zachary Lesser, eds (2007–), DEEP: Database of Early English Playbooks, http://deep.sas.upenn.edu/ (free; accessed 13 November 2020). A database with detailed information about publication and paratext of every playbook printed in England, Scotland and Ireland from the beginning of printing to 1660. Farmer Alan B. and Zachary Lesser (2013), ‘What Is Print Popularity? A Map of the Elizabethan Book Trade’, in Andy Kesson and Emma Smith (eds), The Elizabethan Top Ten: Defining Print Popularity in Early Modern England, 19–54, Aldershot: Ashgate. Because of the high rates of illiteracy during the period, ‘popularity’ is construed as ‘economic performance’ of early printed books (20). Through a comprehensive statistical analysis of the market share of books, finds that the late Elizabethan period was the time when poetry and plays became the second largest category of books (behind religious works) in terms of ‘popularity’. Notes that in the 1590s, the literary canons of such authors as Shakespeare, but also Sidney, Spenser, Marlowe, Chapman, Daniel and others, was being formed. Ferguson, W. Craig (1968), Valentine Simmes: Printer to Drayton, Shakespeare, Chapman, Greene, Dekker, Middleton, Daniel, Jonson, Marlowe, Marston, Heywood, and other Elizabethans, Charlottesville, VA: Bibliographical Society of the University of Virginia. Contains information on the output of printer Valentine Simmes (fl. 1585–1622), who produced eight Shakespearean quartos. Giddens, Eugene (2017), ‘Shakespeare’s Texts and Editions’, in Jill L. Levenson and Robert Ormsby (eds), The Shakespearean World, 465–80, London: Routledge. A scholarly account of Shakespeare’s printed texts, from the ‘bad’ quartos and First Folio to modern academic editions. Gurr, Andrew (2015), ‘Shakespeare’s Lack of Care for his Plays’, Memoria Di Shakespeare: A Journal of Shakespearean Studies, 2: 161–76. Puts forward evidence based on the theatrical history of Shakespeare’s company that the playwright was never really concerned about the printing of his plays. Hailey, Carter R. (2005–6), ‘The Shakespearian Pavier Quartos Revisited’, Studies in Bibliography, 57: 151–96. Re-examines the watermarks of a group of Shakespearean quartos, the Pavier Quartos, in order to shed light on the order and method of printing of these playbooks. Hooks, Adam G. (2014), ‘First Folios: Jonson and Shakespeare’, in Robert DeMaria, Jr, Heesok Chang and Samantha Zacher (eds), A Companion to British Literature: Volume II: Early Modern Literature 1450–1660, 281–93, Chichester: Wiley-Blackwell. Explores the intersecting textual networks between the two writers’ first folios.

ANNOTATED BIBLIOGRAPHY

327

35. Hooks, Adam G. (2016), Selling Shakespeare: Biography, Bibliography, and the Book Trade, Cambridge: Cambridge University Press. Shows how Shakespeare’s career in print was profoundly shaped by his publishers, booksellers and readers. 36. Hooks, Adam G. and Zachary Lesser (2018–), Shakespeare Census, https://shakespearecensus.org/ (free; accessed 13 November 2020). Locates and offers analytical descriptions of all extant copies of editions of Shakespeare’s works until 1700. Does not include the folios. 37. Howard–Hill, T. H. (1974), Ralph Crane and Some Shakespeare First Folio Comedies, Charlottesville, VA: University Press of Virginia. A comprehensive study of professional scribe Ralph Crane, and of the textual relation between his transcripts and the printing of five comedies in Shakespeare’s First Folio. 38. Lesser, Zachary (2004), Renaissance Drama and the Politics of Publication: Readings in the English Book Trade, Cambridge: Cambridge University Press. Chapter 5 (157–225) explores Thomas Walkley’s professional life as a publisher and analyses his edition of Othello (1622) as a commodity published at an opportune time: when rumours spread that James I was going to marry Charles to the Infanta of ‘Moorish’ Spain. 39. Lesser, Zachary (forthcoming), Ghosts, Holes, Rips and Scrapes: Shakespeare in 1619, Bibliography in the Longue Durée, Philadelphia, PA: University of Pennsylvania Press. A re-evaluation of the ‘Pavier Quartos’, the earliest attempt to print Shakespeare’s collected works in 1619. Bibliographic analysis of more than 300 copies reveals evidence – notably ‘ghosts’ (faint impressions resulting from two books bound in adjacent position in a sammelbänd), ‘holes’ (from early modern stab-stitching) and ‘rips and scrapes’ (post-production alterations of title pages) – that complicates what we know about their origins and their subsequent lives across the centuries. 40. Lidster, Amy (2018), ‘At the Sign of the Angel: The Influence of Andrew Wise on Shakespeare in Print’, Shakespeare Survey, 71: 242–54. Shows how Andrew Wise (fl. 1589–1603), Shakespeare’s most productive early publisher of his plays, established and maintained Shakespeare’s lasting popularity as a print author through a mixture of networking and marketing strategies. 41. Lyons, Tara L. (2013), ‘Serials, Spinoffs, and Histories: Selling “Shakespeare” in Collection before the Folio’, Philological Quarterly, 91 (2): 185–220. Considers the early modern seriality of plays as a driving force in the construction of the Shakespeare canon. 42. Massai, Sonia (2007), Shakespeare and the Rise of the Editor, Cambridge: Cambridge University Press. Offers in-depth analyses of how early Shakespearean texts were prepared for the press. Focuses mainly on the Wise Quartos (1597–1602), the Pavier Quartos (1619), the First Folio (1623) and the Fourth Folio (1685). Argues that this work of preparation

328

43.

44.

45.

46.

47.

48.

49.

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

could involve occasional manuscript annotating by a number of nonauthorial agents. Pratt, Aaron T. (2015), ‘Stab–Stitching and the Status of Early English Playbooks as Literature’, The Library, 16: 304–28. Shows that most short quartos (including Shakespeare’s) were bound by stab-stitching. This tended to be normal practice for quartos whatever their genre or contents. Thus, contrary to common opinion, physical features of playbooks give no sure indication of their literary worth. Roberts, Sasha (1997), ‘Editing Sexuality, Narrative, and Authorship: The Altered Texts of Shakespeare’s Lucrece’, in Cedric C. Brown and Arthur F. Marotti (eds), Texts and Cultural Change in Early Modern England, 124–52, Basingstoke: Macmillan. Examines the publishing history of Rape of Lucrece and remarks that this history sets the role of the editor, the representation of sexuality in the poem, as well as the presentation of narrative in a new light. The successive editions of the poem end up canonizing Shakespeare and English poetry for the seventeenth-century reading public. Rutter, Carol Chillington (2015), ‘Schoolfriend, Publisher, and Printer Richard Field’, in Paul Edmondson and Stanley Wells (eds), The Shakespeare Circle: An Alternative Bibliography, 161–73, Cambridge: Cambridge University Press. Focuses on the likely acquaintance of William Shakespeare and Richard Field in Stratford and describes Field’s ensuing career which led him to become a printer, who also printed Shakespeare’s Venus and Adonis and Rape of Lucrece. Smith, Emma (2016), Shakespeare’s First Folio: Four Centuries of an Iconic Book, Oxford: Oxford University Press. Offers a comprehensive history of the First Folio focusing on its owners, readers, editors, printers and on those directors or actors who use it as a source-text. Also shows how the folio has been differently contextualized through time. Smith, Emma (2016), The Making of Shakespeare’s First Folio, Oxford: Bodleian Library. Explains how the First Folio was created, designed and received. Sheds light on who supported the edition in the printing and publishing world at the time. Stern, Tiffany (2008), ‘Watching as Reading: The Audience and Written Text in Shakespeare’s Playhouse’, in Laurie Maguire (ed.), How to Do Things with Shakespeare: New Approaches, New Essays, 136–59, Oxford: Blackwell. Taking Shakespeare’s plays as an example, shows how theatres often relied on the association between play and book, as books, pamphlets and commonplace books were available and written upon within the theatres and during performances. Straznicky, Marta (2013), ed., Shakespeare’s Stationers: Studies in Cultural Biography, Philadelphia, PA: University of Pennsylvania Press. A collection of nine essays that shows how stationers (publishers, printers and booksellers) constructed Shakespeare as an author in print. Contributors also look at the ways in which marketing practices intersected with early modern culture and politics.

ANNOTATED BIBLIOGRAPHY

329

50. Williams, George Walton (1985), The Craft of Printing and the Publication of Shakespeare’s Works, Washington, DC: The Folger Shakespeare Library. Gives a historical overview of the art of printing and looks into the printing and publishing of Shakespeare’s works. 51. Yamada, Akihiro (1994), Thomas Creede: Printer to Shakespeare and his Contemporaries, Tokyo: Meisei University Press. Investigates the career and methods of Thomas Creede as printer, as well as the treatment of errors in setting in his shop. See also: 1f, 2b,d, 4–5, 6b,d–e, 7, 8d–h, 9d,f, 11a–d,k, 12a–b,e, 13a–d,l, 56–7, 65–6, 86, 93–4, 106–7, 112, 123, 128, 133, 140, 153, 158–9, 200, 213.

III SHAKESPEARE’S EARLY TEXTS 52. Clayton, Thomas (1992), ed., The ‘Hamlet’ First Published (Q1, 1603): Origins, Form, Intertextualities, Newark, DE: University of Delaware Press. Contains twelve chapters that cover such aspects as the differences between Q1 and the First Folio, the print production of Q1, the chronology of Shakespeare’s three texts of Hamlet, Q1’s relation to performance, Q1 as a ‘Bad Quarto’ and editing Hamlet in the light of Q1. 53. de Grazia, Margreta and Peter Stallybrass (1993), ‘The Materiality of the Shakespearean Text’, Shakespeare Quarterly, 44: 255–83. Adopts a material approach to early texts by exploring their typography, spelling, line and scene divisions, title pages and textual cruxes. Points out that the multiplicity of Shakespeare’s early texts is an invitation to reconsider the ‘category of a work by Shakespeare’ (255). 54. Dutton, Richard (2016), Shakespeare, Court Dramatist, Oxford: Oxford University Press. Argues that many, if not all, of Shakespeare’s plays were shaped and transformed because they were performed for the court. The first part of the book is devoted to court patronage of Shakespeare’s plays, and the second focuses mainly on six plays which exist in shorter and longer (more sophisticated) texts: 2 and 3 Henry VI, Romeo and Juliet, Henry V, Hamlet and The Merry Wives of Windsor. 55. Erne, Lukas (2007), ‘Shakespeare for Readers’, in Diane Henderson (ed.), Alternative Shakespeares 3, 78–94, London: Routledge. Focusing on the elements only an early reader of Shakespeare’s texts could have access to, insists that these elements can give us access to ‘the readerly specificity of Shakespeare’s early modern playtexts’ (93). As early plays give us little indication of how they were performed, argues that examining readerly elements such as speech prefixes or stage directions may give us some idea of Shakespeare’s choices. 56. Galey, Alan (2014), The Shakespearean Archive: Experiments in New Media from the Renaissance to Postmodernity, Cambridge: Cambridge University Press. Offers a history of Shakespearean texts and their relation to archival technologies over the past four centuries. Remarks that

330

57.

58.

59.

60.

61.

62.

63.

64.

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

Shakespeare’s texts are often used to test new technologies: new media experiments and theories of archiving and computing. Devotes chapters to the archive, the book, photography, sound, information and data. Greg, W. W. (1955), The Shakespeare First Folio: Its Bibliographical and Textual History, Oxford: Clarendon Press. A comprehensive account of the genesis of the First Folio. Gurr, Andrew (1999), ‘Maximal and Minimal Texts: Shakespeare v. The Globe’, Shakespeare Survey, 52: 68–87. Considers what types of texts Elizabethan companies possessed. Argues that ‘minimal texts’ (meant for the stage and not for the Master of Revels) can give us an idea of the length of performances. Hart, Alfred (1942), Stolne and Surreptitious Copies: A Comparative Study of Shakespeare’s Bad Quartos, Melbourne: Melbourne University Press. Offers an in-depth study of the ‘bad’ quartos. Explores in particular their respective lengths, vocabulary, plots, style, verse structure and rhymed lines, and considers their inter-play borrowings, as well as the issue of play-reporting. Hinman, Charlton (1963), The Printing and Proofreading of the First Folio of Shakespeare, 2 vols, Oxford: Clarendon Press. The first volume is concerned with bibliographical evidence, compositor characteristics and press variants, while the second offers an exhaustive study of the printing of the First Folio grounded on the collation of copies owned by the Folger Shakespeare Library. Honigmann, E. A. J. (1965), The Stability of Shakespeare’s Text, London: Arnold. Contends that for plays like Troilus and Cressida, Othello or King Lear, a finalized text never existed. Instead, the two versions of each of these plays were considered as complete by Shakespeare, although still open to revision. Honigman, E. A. J. (1996), The Texts of ‘Othello’ and Shakespearian Revision, London: Routledge. Studies the printers of the first quarto and First Folio texts of Othello, the respective punctuation of the two versions, their lineation and scansion, as well as the responsibility of Ralph Crane in preparing copy for the First Folio. Concludes that modern editors should probably conflate the quarto and folio texts. Ioppolo, Grace (2006), Dramatists and Their Manuscripts in the Age of Shakespeare, Jonson, Middleton, and Heywood: Authorship, Authority, and the Playhouse, London: Routledge. Argues for a circular, rather than linear, process in the way plays were disseminated, as authors could return to their texts during production and performance, even after the plays had been authorized by the censor. Offers, as examples of this practice, Hand D in Sir Thomas More and the first and second quartos of Romeo and Juliet. Jones, John (1995), Shakespeare at Work, Oxford: Clarendon Press. Examines Shakespeare’s textual revisions in plays such as Troilus and Cressida, Sir Thomas More, Hamlet, King Lear and Othello. Shows that the revisions had some effect on characterization and scene structure.

ANNOTATED BIBLIOGRAPHY

331

65. Jowett, John (2017), ‘Exit Manuscripts: The Archive of Theatre and the Archive of Print’, Shakespeare Survey, 70: 113–22. Proposes to distinguish between ‘the archives of theatre and of print’ (122) and shows, using examples from extant versions of Hamlet, Romeo and Juliet, Troilus and Cressida, and the Henriad, that there could well have existed what Jowett calls ‘the exit manuscript that moves from one archive to the other’ (122), which complicates the relation between manuscript sources and print. Concludes that, as a result, not every single printed Shakespeare quarto was systematically based on an annotated playbook. 66. Lesser, Zachary (2015), Hamlet after Q1: An Uncanny History of the Shakespearean Text, Philadelphia, PA: University of Pennsylvania Press. Shows how much the first quarto of Hamlet and its discovery (in 1823) had a profound impact not only on textual studies, but also on how we regard Shakespeare as an author. 67. Maguire, Laurie E. (1996), Shakespearean Suspect Texts: The ‘Bad’ Quartos and Their Contexts, Cambridge: Cambridge University Press. Offers a comprehensive analysis of the forty-one allegedly ‘bad’ quartos of plays from Shakespeare’s time. Contests the theories of the New Bibliography by reconsidering the notion of memorially reconstructed texts. Twelve Shakespeare-related plays are discussed at length. 68. Marotti, Arthur (1990), ‘Shakespeare’s Sonnets as Literary Property’, in Elizabeth D. Harvey and Katharine Eisaman Maus (eds), Soliciting Interpretation: Literary Theory and Seventeenth-Century English Poetry, 143–73, Chicago, IL: The University of Chicago Press. Looks at the textual history of the Sonnets in manuscript and print until the end of the seventeenth century and concludes that their textual instability stems from the overlapping systems of manuscript and print transmission of texts. 69. Menzer, Paul (2008), The Hamlets: Cues, Qs, and Remembered Texts, Newark, NJ: University of Delaware Press. Focuses on the cues that would have shown in the actors’ parts of the first and second quarto and First Folio texts of Hamlet. Argues that these cues were crucial for the transcription, the casting of the parts and for the transmission of the text. Does not believe that Q1 of Hamlet is a memorial reconstruction, but a distinct Hamlet play probably destined to be printed. 70. Palfrey, Simon, and Tiffany Stern (2007), Shakespeare in Parts, Oxford: Oxford University Press. Through substantial analysis of such plays as Merchant of Venice, King Lear and The Tempest, argues for the importance of transcribed ‘parts’, given individually to actors with the words they had to speak and a short cue. Explores Shakespeare’s use of ‘parts’ and ‘cues’ and points to the playwright’s particular talent for ‘engineering dramatic affect’ (12). 71. Petersen, Lene B. (2010), Shakespeare’s Errant Texts: Textual Form and Linguistic Style in Shakespearean ‘Bad’ Quartos and Co-Authored Plays, Cambridge: Cambridge University Press. Considers the manifold interconnections between oral/memorial and authorial composition, and how these elements can be traced in the extant playtexts of the period.

332

72.

73.

74.

75.

76.

77.

78.

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

The book’s main case studies are devoted to Hamlet, Romeo and Juliet and Titus Andronicus. Pollard, Alfred W. (1909), Shakespeare Folios and Quartos: A Study in the Bibliography of Shakespeare’s Plays, 1594–1685, London: Methuen. Makes the now-famous distinction between ‘good’ and ‘bad’ quartos, showing, however, that few of the quartos were really ‘bad’. The study claims to throw light on the circulation of print and manuscript source texts and the part they played in the making of the First Folio. Pollard, Alfred W. (1920), Shakespeare’s Fight with the Pirates and the Problems of the Transmission of his Text, Cambridge: Cambridge University Press. Contests the idea that the early quartos are of little bibliographical value because of the errors they contain. Emphasizes the efforts made to impede print piracy in early modern England and argues that the quartos are much closer to Shakespeare’s manuscripts than previously thought. Purkis, James (2016), Shakespeare and Manuscript Drama: Canon, Collaboration and Text, Cambridge: Cambridge University Press. Shows how Shakespeare wrote his plays and how they were altered during the performance process. Offers a detailed study of frequently unnoticed data in manuscripts used in early modern theatres. Concludes that Shakespeare took an active part in the whole collaborative culture of theatrical manuscripts. Stern, Tiffany (2013), ‘Sermons, Plays, and Note-Takers: Hamlet Q1 as a “Noted” Text’, Shakespeare Survey, 66: 1–23. Considers that Hamlet Q1 may be the work of audience reconstruction (through noting), rather than that of an actor-pirate. Taylor, Gary and Michael Warren, eds (1983), The Division of the Kingdoms: Shakespeare’s Two Versions of ‘King Lear’, Oxford: Clarendon Press. A collection which explores the idea that the quarto (1608) and Folio (1623) versions of the play are widely different texts that should not be conflated (as they had been since the eighteenth century). Contains eleven essays comparing the texts and considering such subjects as censorship of the text, the changing role of the King, authorial revision and adaptation, the editing of the First Folio text, the fluctuations between the two texts, the date and authorship of the F1 version, the influence of editorial tradition, omissions in the Folio text and the diminishing role of Kent. Taylor, Gary and John Jowett (1993), Shakespeare Reshaped 1606–1623, Oxford: Clarendon Press. Offers essays on the introduction of act divisions, the suppression of blasphemous elements and the posthumous introduction of new material in Measure for Measure. Shows how censorship, theatrical innovations and posthumous adaptation altered Shakespeare’s early texts. With appendixes on 2 Henry IV and on Middleton’s hand in Measure for Measure. Urkowitz, Steven (1980), Shakespeare’s Revision of ‘King Lear’, Princeton, NJ: Princeton University Press. Carries out a comparison of

ANNOTATED BIBLIOGRAPHY

79.

80.

81.

82.

333

the first quarto and First Folio texts of King Lear and considers that the variants come from authorial revision. Argues that modern composite texts obscure the effects of intended theatrical revision. Vickers, Brian (2016), The One ‘King Lear’, Cambridge, MA: Harvard University Press. Argues against the theory that the quarto (1608) and the Folio (1623) versions of King Lear represent two different authorial versions. Contends that the cuts in the quarto were carried out by the printer who had underestimated the amount of paper required. As for the Folio, Vickers believes that the cuts are not Shakespeare’s: they were probably made by the theatre company to speed up the action. Werstine, Paul (1990), ‘Narratives about Printed Shakespeare Texts: “Foul Papers” and “Bad” Quartos’, Shakespeare Quarterly, 41: 65–86. A critique of twentieth-century thinking about the so-called ‘bad’ quartos. Werstine, Paul (1999), ‘A Century of “Bad” Shakespeare Quartos’, Shakespeare Quarterly, 50: 310–33. A sceptical account of ‘memorial reconstruction’ as an explanation for the ‘bad’ quartos. Werstine, Paul (2017), ‘Authorial Revision in the Tragedies’, in Michael Neill and David Schalkwyk (eds), The Oxford Handbook of Shakespearean Tragedy, 301–15, Oxford: Oxford University Press. Points out that Shakespeare’s revisions of his tragedies have substantial effects on their interpretation. Compares early versions to later ones.

See also: 1e, 2e–f, 3b–c,e, 4–5, 6c,f, 8b–c,v, 9c,e, 10b–d, 11j, 12d, 13e,j, 21, 37, 112, 122, 126, 128, 133, 135, 143–4, 149, 151, 155–7, 165, 167, 212–13.

IV SHAKESPEARE’S EARLY BIBLIOGRAPHIC AND TEXTUAL RECEPTION 83. Beal, Peter (2013), Catalogue of English Literary Manuscripts 1450–1700 (CELM), https://celm-ms.org.uk/ (free; accessed 13 November 2020). Provides a searchable catalogue of literary manuscripts by 237 British authors of the sixteenth and seventeenth centuries. It references numerous Shakespearean extracts. 84. Bourne, Claire M. L. (2017), ‘Marking Shakespeare’, Shakespeare 13 (4): 367–86. Offers an overview of critical literature devoted to the analysis of readers’ marks. Argues that this literature sometimes misrepresents the early modern reception of Shakespeare’s texts. Examines readers’ marks in the First Folio to demonstrate how early modern readership can be better understood. 85. Bourne, Claire M. L. (2019), ‘Vide Supplementum: Early Modern Collation as Play-Reading in the First Folio’, in Katherine Acheson (ed.), Early Modern English Marginalia, 195–233, London: Routledge. Compares the annotations by two writers in a copy of the Shakespeare First Folio with quartos of Hamlet and Romeo and Juliet printed in 1637. Finds that one of the two marginalists emended the text of the Folio by

334

86.

87.

88.

89.

90.

91.

92.

THE ARDEN RESEARCH HANDBOOK OF SHAKESPEARE AND TEXTUAL STUDIES

taking into account textual variants present in the later quartos, thus reading the Folio as a textual editor. Chartier, Roger (2017), ‘Binding and Unbinding: The Seven Publishing Lives of William Shakespeare’, Cahiers Élisabéthains 93.1: 90–106. Offers a comprehensive analysis of seven types of early Shakespeare publications (1623–1700s): pamphlets, bound books, commonplace books, quartos, folios, complete works and anthologies of best passages by Shakespeare. Cummings, Brian (2012), ‘Shakespeare and Inquisition’, Shakespeare Survey, 65: 306–22. Examines the Valladolid copy of the Second Folio (now held by the Folger Shakespeare Library), which bears the certificate of a Jesuit censor and many censorship marks, and from which Measure for Measure was excised. Estill, Laura (2015), Dramatic Extracts in Seventeenth-Century English Manuscripts: Watching, Reading, Changing Plays, Newark, DE: University of Delaware Press. Examines early modern play readers and playgoers who copied dramatic extracts (selections from plays and masques) into their commonplace books, verse miscellanies, diaries and songbooks. Also studies the lasting attractiveness of such plays as Shakespeare’s Tempest. Estill, Laura (2019), ‘Shakespearean Extracts, Manuscript Cataloguing, and the Misrepresentation of the Archive’, in Tiffany Stern (ed.), Rethinking Theatrical Documents in Shakespeare’s England, 175–92, London: Bloomsbury. Points out that Shakespearean extracts in catalogues of pre-1600 manuscripts exceed those of other playwrights. Considers that is the result of the Shakespeare-centred habits of cataloguers and scholars alike and that a reception history relying on such catalogues runs the risk of misrepresenting the overall reception of early modern drama. Estill, Laura and Beatrice Montedoro, eds (2015), DEx: A Database of Dramatic Extracts, https://dex.itercommunity.org/ (free; accessed 13 November 2020). A database of extracts from plays published before the closure of the theatres in 1642. These extracts appear in manuscripts that were copied before 1700. The database currently contains 1,487 passages extracted from Shakespeare’s plays. Evans, G. Blakemore, ed. (1960–96), Shakespearean Prompt-Books of the Seventeenth Century, 8 vols, Charlottesville, VA: Bibliographical Society of the University of Virginia. Offers studies of stage texts used in various seventeenth-century performances of Shakespeare’s plays. The different manuscript hands that annotate the promptbooks (often Shakespearean seventeenth-century folios) are identified. Available online: http://bsuva. org/bsuva/promptbook/ (free; accessed 13 November 2020). Fehrenbach, R. J. and E. S. Leedham-Green, eds (1992–2009), Private Libraries in Renaissance England, vols 1–9, Tempe, AZ: Arizona Center for Medieval and Renaissance Studies. Transcribed and annotated booklists produced between the beginning of the sixteenth century and the mid-seventeenth century. A number of early Shakespearean editions are mentioned. The free database, PLRE.Folger: Private Libraries in Renaissance

ANNOTATED BIBLIOGRAPHY

93.

94.

95.

96.

97.

98.

99.

335

England, available online: https://plre.folger.edu/ (accessed 13 November 2020), complements but does not replace the printed volumes. Knight, Jeffrey Todd (2009), ‘Making Shakespeare’s Books: Assembly and Intertextuality in the Archives’, Shakespeare Quarterly, 60 (3): 304–40. Studies early copies of Shakespeare’s works, such as the Bodleian volume that binds together Venus and Adonis, Rape of Lucrece and Sonnets. Argues that more attention should be paid to the practices of collectors, compilers, conservators and curators, who necessarily have an im