DETERMINING THE
SHAKESPEARE CANON
DETERMINING
THE
SHAKESPEARE
CANON
ARDEN OF FAVERSHAM AND
A LOVER’S COMPLAINT
MACDONALD P. JACKSON
1
1
Great Clarendon Street, Oxford, ox2 6dp,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© MacDonald P. Jackson 2014
The moral rights of the author‌have been asserted
First Edition published in 2014
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2013957912
ISBN 978–0–19–870441–6
Printed and bound by
CPI Group (UK) Ltd, Croydon, cr0 4yy
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
Acknowledgements
This book incorporates, in substantially revised form, material previously published in the following journals: Archiv für das Studium der neuren Sprachen und Literaturen, Medieval and Renaissance Drama in England,
Memoria di Shakespeare, Modern Language Review, Notes and Queries, Papers of
the Bibliographical Society of America, The Shakespearean International Yearbook,
The Shakespeare Newsletter, Shakespeare Quarterly, and Shakespeare Studies.
The specific articles and notes are those listed in the Bibliography under my
name and with Arden of Faversham or A Lover’s Complaint in the title.
I have been helped by several scholars, who have commented on particular chapters in one or other of their versions, answered specific questions,
exchanged research findings with me, discussed principles of attribution,
solicited articles or reviews, or, whether directly or indirectly, offered encouragement: Jonathan Bate, Brian Boyd, Doug Bruster, John Burrow, Susan
Cerasano, Rosy Colombo, Hugh Craig, Katherine Duncan-Jones, Gabriel
Egan, Ward Elliott, Helen Hargest, Gil Harris, the late Charles H. Hobday,
Roger Holdsworth, Peter Holland, the late Trevor Howard-Hill, John Jowett,
John Kerrigan, Arthur Kinney, Roslyn L. Knutson, Tom Lockwood, David
McInnis, Tom Merriam, Will Sharpe, Tiffany Stern, Marina Tarlinskaja, and
Gary Taylor. Alan Lee of the University of Auckland’s Statistics Department
kindly gave advice about the probability tests I have used. PaulVincent standardized references to Shakespeare’s works, compiled the bulk of the Bibliography, and spotted typos in my manuscript before I submitted it to OUP.
I am grateful to all those mentioned and also, for many excellent suggestions,
to the anonymous readers for the Press and for the journals listed above.
My student-days’ interest in the authorship of Arden of Faversham was
reawakened when I was invited by Barry Gaines to a seminar on ‘The
Shakespeare Apocrypha’ that he chaired at the Shakespeare Association of
America conference in Atlanta in 1993; Christa Jansohn and Dieter Mehl
encouraged me to submit the consequent paper to Archiv. Then in 2002
Auckland graduate student Jayne Carroll completed a fine MA thesis that
viAcknowledgements
appeared to confirm my belief that Shakespeare had a share in the domestic tragedy. In 2005 I delivered at the School of Advanced Study at the
University of London the S. T. Lee Visiting Professorial Fellow’s lecture,
which reported on the findings now recorded in Chapter 1. I should not
have had that opportunity without the support of Brian Vickers, for which
I thank him. It has been disconcerting to find myself arguing against his
views on both Arden of Faversham and A Lover’s Complaint. In the introduction to his Shakespeare, ‘A Lover’s Complaint’, and John Davies of Hereford
(2007), Sir Brian avers that in writing his book claiming the poem for
Davies, ‘One personal problem I faced was that the two most convinced
proponents of Shakespeare’s authorship, Kenneth Muir (in 1964) and
MacDonald Jackson (in 1965) were scholars I had long admired.’ I must
here return the compliment. Sir Brian’s huge contribution to scholarship
on Shakespeare and his contemporaries commands admiration. But I think
he is wrong about Arden of Faversham and A Lover’s Complaint. His book
provided the stimulus for my undertaking almost all the research on the
Complaint described in mine.
I am indebted to Jacqueline Baker for putting my proposal for this book
before OUP’s Delegates, to Rachel Platt for getting publication under way,
to Jo North for her meticulous copy-editing, to Vijayalakshmi Kumar for
her work as project manager, to Daphne Lawless for compiling the index,
and to Nicole Jackson for supporting me in every possible way.
Of course, responsibility for the arguments and conclusions in this book
is mine alone.
Contents
List of Tables
References and Abbreviations
Introduction
ix
xi
1
PART ONE ARDEN OF FAVERSHAM
1. Shakespeare and the Quarrel Scene in Arden of Faversham
2. Reviewing Authorship Studies of Shakespeare and his
Contemporaries, and the Case of Arden of Faversham
3.Gentlemen, Arden of Faversham, and Shakespeare’s Early
Collaborations
4. Parallels and Poetry: Shakespeare, Kyd, and Arden of Faversham
5.Arden of Faversham: Counter-arguments and Conclusions
9
40
60
85
104
PART TWO A LOVER’S COMPLAINT
6.A Lover’s Complaint: Phrases and Collocations
7. Spellings in A Lover’s Complaint as Evidence of Authorship
8. Neologisms and ‘Non-Shakespearean’ Words in A Lover’s
Complaint
9.A Lover’s Complaint, Cymbeline, and the Shakespeare
Canon: Interpreting Shared Vocabulary
10. A Lover’s Complaint: Counter-arguments and Conclusions
184
207
Appendix 1: Literature Online Data for Chapter 1
Appendix 2: Literature Online Data for Chapter 2
Appendix 3: Literature Online Data for Chapter 3
219
237
240
129
141
169
viiiContents
Appendix 4: Literature Online Data for Chapter 6
Appendix 5: Control Test for Chapter 7
Bibliography
Index
245
250
252
265
List of Tables
1.1. Summary of LION links to Arden Quarrel Scene
21
1.2. Summary of LION links to Doctor Faustus samples
25
1.3. Summary of LION links for Shakespeare and Marlowe
26
1.4. Multiple links between Arden Quarrel Scene and Lucrece35
1.5. Multiple links between Arden Quarrel Scene and 2 Henry VI36
2.1. Summary of LION links to Arden’s dream, 6.6–31
55
2.2. Multiple links between Arden and Venus and Adonis57
3.1. Numbers of rare phrases and collocations linking Arden’s Quarrel
Scene to Shakespeare’s early collaborations
70
3.2. Summary of LION links to Arden, 14.1–76
73
3.3. Compound adjectives in Arden of Faversham77
7.1. Contrast between Davies and Shakespeare in use of selected forms
163
8.1. Types and tokens in Shakespeare poems
173
8.2. Latinate neologisms in selected Shakespeare works
174
8.3. All neologisms in selected Shakespeare works
176
8.4. Peculiar words in Shakespeare poems
177
8.5. Thisted–Effron scores
179
References and Abbreviations
References are first given in full before cue titles are introduced for a book
or article cited more than once. Complete details may also be found in
the Bibliography. In each of Chapters 2, 5, and 9, references to the most
frequently cited work are incorporated into the text. Quotations from
old-spelling editions and from Literature Online have been modernized,
unless spelling is relevant to the discussion or retention of a title-page spelling helps to locate a particular book. A few passing references to early modern texts are to signatures in original printings, which are not listed in the
Bibliography but may be found in Early English Books Online: STC numbers are given when this seems desirable. Information about all other texts
quoted or referred to and about the sources for dates of first performances of
plays is provided at the appropriate points. References to tables and appendixes in the present book capitalize the words ‘Table’ and ‘Appendix’; references to tables, appendixes, and figures in other books leave the words
uncapitalized. The following abbreviations are used:
CUP Cambridge University Press
F
Folio
LION Literature Online (the electronic database http://lion.chadwyck.co.uk)
MSR Malone Society Reprint
OED Oxford English Dictionary (compact edition, 1971)
OUP Oxford University Press
Q, Q2 quarto, second quarto
STC
Alfred W. Pollard and G. R. Redgrave et al., A Short-Title Catalogue of
Books Printed . . . 1475–1640, 2 vols., 2nd edn. (London, 1976)
UP
University Press
Wing Donald Goddard Wing et al., A Short-Title Catalogue of Books
Printed . . . 1641–1700, 4 vols., 2nd edn. (New York, 1972–98)
Introduction
I
n 1952 a correspondent to The Times (London), Sydney Cockerell, recalled
once sitting next to Algernon Charles Swinburne at a dining table at
Watts-Dunton’s Number 2 The Pines, immortalized by Max Beerbohm.
Cockerell ventured some trivial remark, which the aged poet, now very deaf,
mistook for a reference to Arden of Faversham. He was delighted at his fellow guest’s supposed erudition. ‘For the rest of the meal,’ wrote Cockerell, ‘I
listened entranced to an eloquent discourse on that remarkable play, which
he emphatically declared to have qualities of genius that could be found in
Shakespeare alone.’1
Swinburne had committed himself in print to the same verdict. He
ended an extravagant encomium by declaring that, in view of ‘the various
and marvellous gifts displayed for the first time on our stage’ by the anonymous author of Arden of Faversham, it seemed to him ‘simply logical and
reasonable’ to regard the play, ‘a young man’s work on the face of it, as the
possible work of no man’s youthful hand but Shakespeare’s’.2
In the nineteenth and early twentieth centuries Swinburne was by no
means alone in favouring such an ascription. The fine editor of Shakespeare,
Charles Knight, the great German scholar Nicolaus Delius, the literary historian W. J. Courthope, and the play’s French translator, François Victor
Hugo, upheld it with various degrees of conviction, while many other
commentators, including Henry Tyrrell, John Addington Symonds, A. H.
Bullen, and Alfred Mézières were inclined to believe that Shakespeare had
been involved with the script of Arden of Faversham as part-author, reviser,
or corrector. Even sceptics were apt to detect speeches and lines with ‘a
Shakespearean ring’ to them.3
1.Sydney Cockerell, The Times, 2 October 1952, 7. Beerbohm’s essay ‘No. 2. The Pines’ is
readily available at http://www.readbookonline.net/readOnline/2355.
2.A. C. Swinburne, A Study of Shakespeare (London: Chatto and Windus, 1879; repr.
Heinemann, 1918), 140–1. Swinburne assumed that the play was of single authorship.
3.The history of opinion about the authorship of Arden of Faversham is outlined by M. P.
Jackson, ‘Material for an Edition of Arden of Feversham’ (BLitt thesis, Oxford, 1963),
90–270, and by Jill Levenson, ‘Anonymous Plays: Arden of Faversham’, in The Predecessors
of Shakespeare: A Survey and Bibliography of Recent Studies in English Renaissance Drama, ed.
Terence P. Logan and Denzell S. Smith (Lincoln, NE: University of Nebraska Press, 1978),
2
Determining the Shakespeare Canon
Other scholars promoted the claims of Thomas Kyd or Christopher
Marlowe or both in collaboration. But by 1966 S. Schoenbaum, in his
sardonic survey of authorship studies in the field of early modern drama,
could announce that ‘The conjecturists appear to have given up on Arden of
Faversham’, a renunciation that he suggested might be due ‘in part to recognition of the law of diminishing returns’.4 Shortly afterwards, in 1973, M. L.
Wine, in editing the play for the Revels series, asked ‘Did Shakespeare
write Arden of Faversham?’ and added, ‘Today, if the matter of authorship
is raised at all, this question is the only one likely to arouse serious interest. The leading rival claimants of the past, Kyd, and to a lesser extent,
Marlowe, no longer compel any strong assent.’5 Though struck by some of
the play’s Shakespearean characteristics, and particularly its affinities with
the Henry VI trilogy, Wine concluded that it must remain anonymous.
Anthologists Russell A. Fraser and Norman Rabkin confidently asserted
that ‘no evidence supports the attribution’ to Shakespeare and ‘one can say
conclusively’ that he ‘did not write the play’.6
‘Did not’, like Hamlet’s ‘by and by’, ‘is easily said’. But this judgement
was premature. Over the last two decades evidence has accumulated that
Shakespeare, though not solely responsible for Arden of Faversham, contributed substantially to it.7 He was a co-author. Claims for Kyd’s authorship
have also been revived.8 These seem to me to be without foundation. The
first part of this book is devoted to demonstrating that Arden of Faversham,
like the anonymously published Edward III, is a play in which Shakespeare
collaborated early in his career, and should join the chronicle history in
editions of the Collected Works.
240–52. It is sketched by M. L. Wine in his edition, The Tragedy of Master Arden of Faversham
(London: Methuen, 1973), lxxxi–xcii. I quote Arden from Wine’s edition throughout,
except that for a few readings, duly noted, I have preferred the text of Arden of Feversham (sic)
in Minor Elizabethan Tragedies, ed. T. W. Craik (London: Dent, 1974), 203–78.
4.S. Schoenbaum, Internal Evidence and Elizabethan Dramatic Authorship: An Essay in Literary
History and Method (Evanston: Northwestern UP, 1966), 137.
5.Wine, Arden, lxxxi–lxxxii.
6. Russell A. Fraser and Norman Rabkin, eds., Drama of the English Renaissance I: The Tudor
Period (New York: Macmillan, 1976), 411. Like so many sceptics before them, Fraser and
Rabkin concede, however, that there are ‘passages’ in Arden ‘not unworthy of the author of
2 and 3 Henry VI’ (411).
7. This evidence is presented and discussed in the chapters that follow.
8. Brian Vickers, ‘Thomas Kyd, Secret Sharer’, The Times Literary Supplement, 18 April 2008,
13–15. The data on which this article draws were posted on the website of the London
Forum for Authorship Studies: http://ies.sas.ac.uk/events/seminars/LFAS, but have since
been withdrawn. My reasons for rejecting Vickers’s case will also become apparent in later
discussion.
Introduction
3
The extravagance of Swinburne’s praise of Arden of Faversham was
matched by the extravagance of his language in referring to A Lover’s
Complaint, published with Shakespeare’s Sonnets in 1609. Comparing this
329-line narrative poem with Venus and Adonis and The Rape of Lucrece, he
noted that it was ‘marked . . . throughout with every possible sign suggestive of a far later date and a far different inspiration’, adding that ‘it contains
two of the most exquisitely Shakespearean verses ever vouchsafed to us by
Shakespeare, and two of the most execrably euphuistic or dysphuistic lines
ever inflicted on us by man’.9 The dichotomy in Swinburne’s reactions is
duplicated in the history of scholarship and criticism on A Lover’s Complaint.
Admired by a mere handful of the commentators whose opinions were
summarized in Hyder Rollins’s Variorum edition of Shakespeare’s Poems
(1938), it was dismissed by most as spurious or as a youthful failure, until
in the 1960s two studies independently argued for its authenticity and a
seventeenth-century date of composition.10 These rescue attempts were
successful, in that the poem began to attract much illuminating criticism
and editors accepted it as integral to the design of the Sonnets quarto.
The new consensus has been disrupted, however, by Brian Vickers,
whose Shakespeare, ‘A Lover’s Complaint’, and John Davies of Hereford,
aims to prove that the true author was Davies, a famous calligrapher and
writing-master and a prolific poet.11 His book persuaded Jonathan Bate
and Eric Rasmussen to exclude A Lover’s Complaint from their RSC
Shakespeare Complete Works on the grounds that Vickers had ‘devastated’
the case for attributing it to Shakespeare and set forth a ‘strong’ one in
favour of Davies.12 They were not alone in this opinion. Reviewing the
year’s work in critical studies of Shakespeare, Michael Taylor affirmed: ‘It’s
hardly possible not to be convinced (swept away even) by the thoroughness
9.Swinburne, Shakespeare, 61–2.
10. A New Variorum of Shakespeare: The Poems, ed. Hyder Edward Rollins (Philadelphia and
London: Lippincott, 1938), 584–603. MacDonald P. Jackson, Shakespeare’s ‘A Lover’s
Complaint’: Its Date and Authenticity (Auckland: University of Auckland, Bulletin 72 English
Series 13, 1965); Kenneth Muir, ‘ “A Lover’s Complaint”: A Reconsideration’, in Shakespeare
1564–1964: A Collection of Modern Essays by Various Hands, ed. Edward A. Bloom (Providence,
RI: Brown UP, 1964), 154–66; reprinted in Muir’s Shakespeare the Professional and Related
Studies (London: Heinemann, 1973), 204–19.
11. Brian Vickers, Shakespeare, ‘A Lover’s Complaint’, and John Davies of Hereford
(Cambridge: CUP, 2007). Hereafter cited as ‘Vickers, Complaint’.
12.Jonathan Bate and Eric Rasmussen, eds., William Shakespeare: Complete Works, RSC
Shakespeare (Basingstoke: Macmillan, 2007), 2397.
4
Determining the Shakespeare Canon
and passion of Vickers’s argument. I’m happy to acknowledge myself a convert.’13 Walter Cohen, citing Vickers’s book, stated in the second edition of
The Norton Shakespeare: ‘Finally, it now appears that “A Lover’s Complaint”
was written not by Shakespeare but by John Davies.’14
In the second part of this book I set out reasons for thinking that the case
for Shakespeare’s authorship of A Lover’s Complaint has not been ‘devastated’; that to be ‘swept away’ by Vickers’s argument for Davies’s is to be
caught in a rip-tide of error; and that the poem was written not by John
Davies but by William Shakespeare.
The authorship of the Complaint is crucial to our sense of the trustworthiness of the Sonnets quarto and the extent to which its organization is
Shakespeare’s own. If Vickers is right, grounds for believing that, apart
from a sprinkling of typographical and scribal errors, the quarto preserves
authorial intentions are seriously undermined. The notion that its contents were deliberately arranged by Shakespeare as a tripartite structure, A
Lover’s Complaint being the third component, in which the Fair Friend and
Dark Lady opposites are synthesized, with Sonnets 153 and 154, variants on
the same mini-story, forming a bridge to the Complaint—this notion must
be a figment of critical over-ingenuity. But this book marshals evidence
that those who reject the quarto publisher Thomas Thorpe’s ascription of A
Lover’s Complaint to Shakespeare are wrong.
This volume draws on widely scattered essays and notes, some not easily accessible, that over a couple of decades I have contributed to academic
journals on these two problems of the Shakespeare canon.15 But the material has been revised, augmented, and reorganized into two parts—one
devoted to Arden of Faversham and one to A Lover’s Complaint—each of
which develops a coherent thesis in incremental steps. The results of different lines of investigation, reported upon in successive chapters, support
one another, so that, as the evidence accumulates, the solidity of the case
for regarding A Lover’s Complaint as Shakespeare’s and Arden of Faversham
as partly his should become apparent. The book contains several sections
based on fresh research that appreciably supplements and strengthens the
articles originally published. My analyses of Arden take into account an
13.Michael Taylor, ‘The Year’s Contributions to Shakespeare Studies: Critical Studies’,
Shakespeare Survey, 61 (2008), 359–83 at 369.
14.Walter Cohen, in The Norton Shakespeare: Second Edition, gen. ed. Stephen Greenblatt
(New York: Norton, 2008), 1944.
15.The journals are listed in the Acknowledgements and details of articles are given in the
Bibliography.
Introduction
5
important recent enquiry into the authorship of 3 Henry VI. Chapters 5 and
10 are almost entirely new.
It may be as well to set down from the outset what conclusions I believe
the evidence on Arden of Faversham sustains: Shakespeare was largely, if not
wholly, responsible for the writing of the central scenes, 4–9; he probably
also made at least some contributions to other parts of the play, but doubts
about the integrity of certain stretches of the quarto text render problematical any attempts to delineate these.
By now studies in attribution need no defence. Brian Vickers’s
Shakespeare, Co-Author (2002) did much to re-establish them in scholarly
estimation.16 His thorough historical survey and evaluation of the evidence
demonstrated that in at least six plays Shakespeare collaborated with other
playwrights, whose shares could be distinguished from his. Edward III
swells the number to seven, and subsequent research suggests that 2 and 3
Henry VI must also be added.17 This matters for our understanding of the
trajectory of Shakespeare’s career, of the early modern playwriting industry in which he was so heavily involved, and of features peculiar to the
co-authored plays. Shakespeare’s identified collaborators—Thomas Nashe,
George Peele, George Wilkins, Thomas Middleton, John Fletcher—had
their individual verse-writing styles.18
It is curious that an age of criticism that placed such emphasis on ‘materiality’ and ‘the body’ was so eager to deny the significance of the individual
‘author’. For the brain is part of the body, and it was within the brains
of particular biological beings that playscripts took shape. Shakespeare’s
brain was a unique physiological entity, through which the theatrical,
ideological, linguistic, political, social, and other conditions of his time
16. Brian Vickers, Shakespeare, Co-Author: A Historical Study of Five Collaborative Plays
(Oxford: OUP, 2002). Vickers’s case, adumbrated in the book, for adding 1 Henry VI as a
sixth co-authored play was made in ‘Incomplete Shakespeare: Or, Denying Coauthorship
in 1 Henry VI’, Shakespeare Quarterly, 58 (2007), 311–52.
17. Edward III, ed. J. J. M. Tobin, has been included in the second edition of The Riverside
Shakespeare, ed. G. Blakemore Evans (Boston and New York: Houghton Mifflin, 1997);
in the New Cambridge series, ed. Giorgio Melchiori (Cambridge: CUP, 1998); and in
William Shakespeare: The Complete Works, Second Edition, gen. eds. Stanley Wells and Gary
Taylor (Oxford: Clarendon Press, 2005). Unless indicated otherwise, I quote Shakespeare’s
works from the Riverside edition, because Marvin Spevack’s The Harvard Concordance to
Shakespeare (Cambridge, MA: Belknap Press of Harvard UP, 1973) is based on it, and the
concordance is indispensable for tracking down Literature Online quotations of Shakespeare’s
works. However, English spelling has been used for the title Love’s Labour’s Lost.
18.See MacDonald P. Jackson, ‘Collaboration’, in The Oxford Handbook of Shakespeare, ed.
Arthur F. Kinney (Oxford: OUP, 2012), 31–52.
6
Determining the Shakespeare Canon
and place were mediated.19 Attempts to determine the precise limits of the
Shakespeare canon remain worthwhile, more than a century after editor
Ronald Bayne asserted: ‘If Shakespeare wrote Arden it is the most interesting fact in his literary development.’20 This is an exaggeration, perhaps, and
Bayne did not allow for the possibility, by now a near-certainty, that Arden
of Faversham is the product of collaboration, but if the first surviving English
domestic tragedy is even partly of Shakespeare’s composition, accounts of
the earliest phase of his career as playwright are due for significant revision.
* * *
Since in this book I have frequent recourse to the electronic database Literature
Online (LION), it seems worth correcting a misapprehension given expression by Martin Wiggins in The Times Literary Supplement. Citing hearsay
complaints about LION’s inadequacies, he proposes that ‘a concrete test is
to search for “illeg.”, the notation used when the keyboarders found themselves unable to read their copy. In February 2008, such a search turned up
15,219 hits in 5,923 separate works. One can always verify the readings LION
produces, but there is no knowing whether other evidence has been overlooked because it was wrongly, or never, entered in the first place.’21 But the
results of Wiggins’s ‘concrete test’ exaggerate LION’s unreliability, because
‘illeg.’ often denotes not any failure of the keyboarders but an actual lacuna
or illegible reading in a manuscript, as noted by an expert transcriber. In early
modern plays most readings designated ‘illeg.’ are so designated in a Malone
Society or similar diplomatic reprint of a work for which the foundation text
is handwritten, not printed. LION keyboarders sensibly based their texts of
manuscript works on esteemed transcripts by qualified scholars. LION is far
from perfect. Sometimes a long ‘s’ is misread as an ‘f ’, for example, and tildes
substituting for ‘n’ or ‘m’ are too apt to be ignored rather than expanded. In
some texts keyboarders have failed to recognize other standard abbreviations,
and slightly damaged or poorly inked types are in places wrongly identified
and transcribed. But most of LION’s alleged inadequacies are those of users
who have not sufficiently familiarized themselves with its search functions
and their capabilities.
19. The point is developed in Mary Thomas Crane, Shakespeare’s Brain: Reading with Cognitive
Theory (Princeton: Princeton UP, 2001).
20. Ronald Bayne, ed., Arden of Feversham (London: Dent, 1897; repr. 1955), x.
21.Martin Wiggins, ‘Lord Mayor’s Showman’, review of The Works of John Webster, Volume
Three, ed. David Gunby, David Carnegie, and MacDonald P. Jackson, The Times Literary
Supplement, 4 April 2008, 26–7, at 27.
PART
ONE
Arden of Faversham
1
Shakespeare and the Quarrel
Scene in Arden of Faversham
I
‘Having produced Arden, I find it difficult to imagine that the intense scene
viii, entirely independent of Holinshed, was not written by Shakespeare,’
declared Keith Sturgess, who had also edited Arden of Faversham as one
of Three Elizabethan Domestic Tragedies.1 He was referring to the famous
Quarrel Scene between Alice Arden and her lover Mosby. In the early
Elizabethan drama, outside the Shakespeare canon, only the final scene of
Doctor Faustus matches it for its combination of sheer theatrical effectiveness, poetic merit, psychological depth, and emotional power; and, while
it is characteristic of Marlowe that his scene requires solo performance by
the hero who so dominates his play, the author of the finest scene in Arden
of Faversham places his two chief characters in an intensely dramatic conflict
that anatomizes the volatile relationship between them.
Scene 8 begins with a soliloquy in which Mosby expresses the anxieties
that plague his social climbing and his schemes to rise higher by murdering
Alice’s husband. He concludes that even Alice herself jeopardizes his security and must be eliminated. She enters holding a prayer-book and avowing
repentance of her adulterous affair and of conspiring to kill her husband.
A bitter quarrel ensues, in which the lovers trade insults, until Alice capitulates, pleading for Mosby’s forgiveness. Eventually, they kiss and make up.
The scene ends as Alice says, ‘Come, let us in to shun suspicion’ and Mosby
1.Keith Sturgess, review of The Tragedy of Master Arden of Faversham, ed. M. L. Wine, in
Durham University Journal, n.s. 36 (1974), 245; Keith Sturgess, ed., Three Elizabethan Domestic
Tragedies (Harmondsworth: Penguin Books, 1969).
10
Determining the Shakespeare Canon
replies, ‘Ay, to the gates of death to follow thee’ (8.166–7). The episode is
superbly shaped, mounting to a climax of acrimony, and then, when Alice
can no longer endure Mosby’s abuse, subsiding towards reconciliation.
The range of tones is extraordinary. Mosby’s opening monologue, which
has a two-part structure, foreshadows the trajectory of the scene as a whole.
At the mid-point of his speech, nostalgia for a ‘golden time’ when he ‘slept
secure’ gives way to recognition that he ‘cannot back | But needs must on
although to danger’s gate’ (11–22) and determination that Arden must die.
Like Richard of Gloucester and Macbeth, Mosby schemes to neutralize the
threats to his safety posed by witnesses and accomplices.2 With Alice he
has undergone a form of marriage, but, since she has supplanted Arden for
his sake, she may, he argues, tire of him and cannot be trusted: ‘’Tis fearful
sleeping in a serpent’s bed, | And I will cleanly rid my hands of her’ (42–3).
But seeing her approach, he resolves to ‘flatter her’ (44). Alice’s contrite
sighs prompt feigned solicitude. With typical egotism, he accuses her of
adopting ‘distressful looks’ in order to ‘wound a breast | Where lies a heart
that dies when thou art sad’, leading into the exchange:
Mosby. . . . It is not love that loves to anger love.
Alice. It is not love that loves to murder love.
Mosby. How mean you that?
Alice. Thou knowest how dearly Arden loved me.
Mosby.And then?
Alice. And then—conceal the rest, for ’tis too bad,
Lest that my words be carried with the wind
And published in the world to both our shames.
I pray thee, Mosby, let our springtime wither;
Our harvest else will yield but loathsome weeds.
Forget, I pray thee, what hath passed betwixt us,
For now I blush and tremble at the thoughts.
(56–69)
Mosby’s first speech here means ‘If you truly loved me you would not vex
me by looking miserable.’ The conscience-stricken Alice varies his wordplay to encapsulate the paradox that has been troubling her. Her ‘It is not
love that loves to murder love’ concedes the inherent contradictoriness of
2.His recognition that he cannot go back but ‘needs must on’ reminds one of Richard
Gloucester’s ‘But I am in | So far in blood that sin will pluck on sin’ (Richard III, 4.2.63–4)
and Macbeth’s ‘I am in blood | Stepp’d in so far that, should I wade no more, | Returning
were as tedious as go o’er’ (Macbeth, 3.4.135–7).
Shakespeare and the Quarrel Scene in
arden of faversham 11
killing, for love’s sake, a loving husband. But Mosby’s guilt at his own musings causes him, momentarily and irrationally, to misinterpret her riposte
as ‘If you loved me you would not wish to murder me.’ His ‘How mean you
that?’ is a startled reflex. Just before Alice arrived he had decided to do away
with her: to him, it is as though she has read his mind, as though his soliloquy—by convention a vocalization of inner thoughts for the audience’s ears
alone—has been overheard by an Alice waiting, as it were, in the wings.3
Mosby’s alarm abates as Alice elaborates, making her real meaning clear.
The portrayal of the workings of a guilty mind is so subtle that most actors
and directors fail to register the subtext here.
Alice’s shame at having earned ‘an odious strumpet’s name’ shifts into
accusations—it is Mosby, who has robbed her of her ‘honesty’—and into
class-conscious reviling of him as ‘mean artificer’ and ‘low-born’ (72–7).
Attuned to this upwardly mobile jobbing-tailor’s sensitivities, she taunts
him with reminders of his origins and then fawningly calls him ‘as gentle
as a king’ (140) when his rejection of her becomes intolerable. Earlier in the
quarrel she recognizes the truth of her friends’ admonitions that ‘Mosby
loves me not but for my wealth’ (108), but her infatuation with this brutal
Lothario induces a wilful blindness. Doubtless Mosby’s ambition is spiced
with lust, but his true motives emerge in his reaction to Alice’s initial complaints that he has sullied her good name. ‘Nay,’ he says, ‘. . . if you stand so
nicely at your fame, | Let me repent the credit I have lost’ (80–3). Here the
ambiguous ‘credit’ is pivotal, as Mosby’s thoughts slither from reputation to
money: he could, he protests, have married a maid ‘Whose dowry would
have weighed down all thy wealth’ (89). When Alice attempts to woo him
back, he resorts to heavy sarcasm:
O, no, I am a base artificer;
My wings are feathered for a lowly flight.
Mosby? Fie, no! not for a thousand pound.
Make love to you? Why, ’tis unpardonable;
We beggars must not breathe where gentles are.
(135–9)
3.There is a similar effect in Macbeth. After Duncan’s murder, Macbeth’s startled question
‘Whence is that knocking?’ (2.2.54), as the noise sounds ‘within’ (meaning offstage within
the tiring house), implies his guilty fear that the murder has no sooner been performed
than discovered. Arden of Faversham’s many foreshadowings of Macbeth have often been
noted: for some references see MacDonald P. Jackson, ‘Shakespearean Features of the Poetic
Style of Arden of Faversham’, Archiv für das Studium der neueren Sprachen und Literaturen, 230
(1993), 279–304, at 288, n. 17.
12
Determining the Shakespeare Canon
He had, as he confessed in his soliloquy, sought to climb to the top of the
social tree and nest ‘among the clouds’ (16). Here he picks up that imagery.
And, as John Addington Symonds observed, ‘The touch of not for a thousand
pound is rare. Alice never for a moment thought of money. It is the churl
who expresses the extreme of scorn by hyperboles of cash.’4
Mosby’s concern for status is central to this play, set at a time when the
dissolution of the monasteries enabled members of the rising bourgeoisie
such as Arden to become wealthy landowners and marry into aristocratic
families, while depriving poorer folk like the yeoman Greene and the
sailor Reede of a livelihood eked out on communal ground. Characters are
acutely aware of their positions within the social hierarchy: Lord Cheyne,
who enters the play as a virtual deus ex machina; the freeholder aptly named
Franklin;5 the tradesman and shopkeeper Bradshaw; the painter Clarke; the
servants Michael and Susan; and the ‘masterless men’ of the criminal underclass, Black Will and Shakebag, repatriated from the wars in Boulogne, to
continue killing and looting in civilian life.
The scene, intervening between Black Will and Shakebag’s decision to
assassinate Arden on Rainham Down and their typical bungling of the
attempt, is the turning point of the tragedy, sealing Arden’s fate. Mosby and
Alice entered separately, self-preoccupied and faltering in their resolve; they
exit together, their renewed commitment to murder decisive and irrevocable. It will indeed lead them and their associates, as well as their victim, to
‘the gates of death’ and beyond. This is Alice’s last chance of redemption.
The finality of her transformation from ‘Arden’s honest wife’ to Mosby’s
paramour is signalled in the kiss that seals ‘this new-made match’ (73, 150).
Her repudiation of biblical law is enacted in vivid stage images, as she tears
leaves from ‘The holy word that had converted [her]’ and vows to replace
them with Mosby’s love letters (117–20). Like Volpone’s worship of his
gold, which opens Jonson’s great satiric comedy, Alice’s sacrilege defines
her moral world. This gesture, too, crystallizes major themes: the perversion of religious symbols and sacraments and the corruption of language,
as oaths and bonds are broken and words become weapons. The Quarrel
Scene expands and adds complexity to an earlier brief altercation between
the adulterers and is immediately followed by a comic counterpart in a
4.John Addington Symonds, Shakespeare’s Predecessors in the English Drama (London: Smith
Elder, 1884, repr. 1900), 335.
5.Franklin is the character’s surname, but a franklin is a small landowner, of free but not
noble birth.
Shakespeare and the Quarrel Scene in
arden of faversham 13
dispute between the hired hitmen.6 It drives the plot forward, sharpens and
deepens characterization, and focuses the tragedy’s dominant concerns. It
is a remarkable piece of dramatic writing.
Could it be Shakespeare’s? While critical appreciation of the scene may
raise this question, for an answer we must turn to more objective modes of
analysis. But first some background information is needed.
II
Arden of Faversham was entered in the Stationers’ Register on 3 April 1592
and published anonymously in a quarto of the same year.7 During the
1580s and 1590s anonymous publication of playscripts from the commercial theatre was not the exception but the norm: in the period from 1594
to 1597 six plays of the accepted Shakespeare canon were printed without
his name on the title page; in 1598 a quarto of 1 Henry IV was also anonymous, but a quarto of Love’s Labour’s Lost ascribed it to ‘W. Shakespere’.8
Arden of Faversham was not, of course, included in the Shakespeare First
Folio of 1623. However, W. W. Greg demonstrated that the bookseller
Edward Archer intended to attribute Arden of Faversham to Shakespeare
in his play catalogue of 1656, his intentions being derailed by some accidental misalignment, of a kind that recurs later in Archer’s list.9 As Greg
noted, Archer’s blunders, ‘whether due to carelessness or ignorance, are so
many and so gross that very little reliance can be placed upon any particular
ascription he may make’, but he does show ‘occasional signs of rather unexpected knowledge’.10
The external evidence connecting Shakespeare with Arden of Faversham,
though slight, is thus a shade stronger than that connecting him with
Edward III, first printed in an anonymous quarto of 1596 and attributed
to Shakespeare by Richard Rogers and William Ley in their catalogue,
6. See 1.176–219, 9.1–37.
7.Wine, Arden, xix.
8. Data concerning the naming of playwrights on commercial title pages during the period
1576–1660 are provided by Alan B. Farmer and Zachary Lesser, ‘Vile Arts: The Marketing
of English Printed Drama, 1512–1660’, Research Opportunities in Renaissance Drama,
39 (2000), 77–165. For the Shakespearean details, see MacDonald P. Jackson, Defining
Shakespeare: ‘Pericles’ as Test Case (Oxford: OUP, 2003), 15.
9. W. W. Greg, ‘Shakespeare and Arden of Feversham’, Review of English Studies, 21 (1945), 134–6.
10. Greg, ‘Shakespeare and Arden of Feversham’, 135.
14
Determining the Shakespeare Canon
which, like Archer’s, appeared in 1656. Rogers and Ley rendered their testimony almost valueless by including Marlowe’s Edward II and Heywood’s
Edward IV as Shakespeare’s in a bracket of three histories.11 Independently
of Archer, the Faversham antiquary Edward Jacob proposed in 1770 that
Shakespeare had written Arden of Faversham; independently of Rogers and
Ley, Edward Capell proposed in 1760 that Shakespeare had written Edward
III.12 Both eighteenth-century proposals were vigorously debated in the
nineteenth and early twentieth centuries. Swinburne, while scornful of
the claim for Edward III, was an enthusiastic supporter of Jacob’s theory,13
and, as we have seen, many other commentators were inclined to believe
Shakespeare responsible for the whole play or granted him episodes,
speeches, or stray lines.
The domestic tragedy has yielded a much richer harvest of critical articles—feminist, Marxist, cultural materialist, new historicist, and traditional—than the history.14 But whereas Edward III has now gained
acceptance as at least partly Shakespearean, Arden of Faversham, which in the
nineteenth century seemed the more promising candidate for admission to
the canon, has not, though it is now knocking at the door. A convincing
case for attributing the Quarrel Scene to Shakespeare should help to nudge
Arden over the threshold, as one of his co-authored plays.
In the Oxford William Shakespeare: A Textual Companion, Gary Taylor
writes that the evaluation of internal evidence in Arden of Faversham ‘is
11. Greg described and assessed the various play catalogues in ‘Authorship Attributions in the
Early Play-Lists, 1656–1671’, Edinburgh Bibliographical Society Transactions, 2 (1938–45), 305–
29, esp. 308.
12. For Capell’s and Jacob’s attributions, see (as in Introduction, n. 3) Wine, Arden, lxxx i–xcii;
Levenson, ‘Anonymous Plays: Arden of Faversham’; Jackson, ‘Material for an Edition’, where
details about Edward III are also given (136–8); for Edward III see Introduction, n. 17; Stanley
Wells and Gary Taylor with John Jowett and William Montgomery, William
Shakespeare: A Textual Companion (Oxford: Clarendon Press, 1987), 136–37; and Riverside
Shakespeare, 1732–4.
13. Swinburne, Study of Shakespeare, 128–41, 231–75.
14. A substantial bibliography of key articles accompanies the edition of Arden in Simon Barker
and Hilary Hinds, eds., The Routledge Anthology of Renaissance Drama (London: Routledge,
2003), 78–9, to which should be added: Michael Neill, ‘ “This Gentle Gentleman”: Social
Change and the Language of Status in Arden of Faversham’, Medieval and Renaissance Drama in
England, 10 (1998), 73–97, reprinted in Putting History to the Question: Power, Politics and Society
in English Renaissance Drama (New York: Columbia UP, 2000), 49–72; Ros King, ‘Arden of
Faversham: The Moral of History and the Thrill of Performance’, in The Oxford Handbook of
Tudor Drama, ed. Thomas Betteridge and Greg Walker (Oxford: OUP, 2012), 635–52. The
play has appeared in two other recent large anthologies: Arthur F. Kinney, ed., An Anthology
of Plays and Entertainments (Oxford: Blackwell, 1999); Kathleen E. McLuskie and David
Bevington, eds., Plays on Women (Manchester: Manchester UP, 1999); see also Martin White,
ed., The Tragedy of Master Arden of Faversham (London: Ernest Benn, 1982), New Mermaid; and
rev. edn. introduced by Tom Lockwood (London: A and C Black, 2007).
Shakespeare and the Quarrel Scene in
arden of faversham 15
complicated—and perhaps permanently defeated—by uncertainty over the
text of the 1592 edition’.15 Alfred Hart thought that ‘the verse exhibits many
of the defects characteristic of a report’ and blamed actors for the many ‘harsh
or unmetrical’ lines.16 Jackson elaborated on the idea that the text suffers from
some form of memorial contamination, preferring to invoke a careless scribe,
and M. L. Wine, in his edition of the play for the Revels series, accepted this
view.17 Subsequent editors, pointing to ‘the general coherence of the play’,
have been more cautious, treating the text as tolerably ‘good’.18 The important
fact for our present purposes is that the basic textual integrity of the Quarrel
Scene has never been doubted: its few obvious corruptions are clearly due to
the misreading of Elizabethan handwriting.
There is no dispute over the play’s source. Arden of Faversham dramatizes
the lively and lengthy narrative of events leading up to Thomas Arden’s
slaying that was included in Raphael Holinshed’s famous Chronicles.19 Since
the tragedy seems influenced in places by copious marginalia added to the
1587 edition of Holinshed, it is unlikely to have been written more than
five years before publication. Plays were seldom published until at least a
year after they had begun their run on the stage. Apparent allusions to John
Lyly’s Endymion, performed at court on 2 February 1588, and to a scandalous rumour about Robert Dudley, Earl of Leicester, which the dramatist is unlikely to have perpetuated on stage before the earl’s death on 4
September 1588, together with borrowings from or by other plays, narrow
the probable chronological limits to 1588–91.20
Apart from Shakespeare, the main contenders for the authorship of
Arden of Faversham have been Thomas Kyd and Christopher Marlowe.
15. Textual Companion, 135.
16. Alfred Hart, Stolne and Surreptitious Copies: A Comparative Study of Shakespeare’s Bad Quartos
(Melbourne: Melbourne UP, 1942), 384.
17. Jackson, ‘Material for an Edition’, 1–37; Wine, Arden, xxiv–xxxiii.
18. The anonymous author of an introductory note to the facsimile, The Lamentable and True
Tragedy of M. Arden of Feversham (Menston: Scolar Press, 1971) expresses scepticism that
the text was based on ‘memorial reconstruction’ and mentions the play’s ‘general coherence’. Memorial reconstruction as a theory to account for the state of certain early modern
dramatic texts has since fallen out of favour, but even a sceptic such as Laurie E. Maguire,
Shakespearean Suspect Texts: The ‘Bad’ Quartos and their Contexts (Cambridge: CUP,
1996) judges it a likely explanation in some cases.
19. Sources are discussed by Wine, Arden, xxxv–xliii. There is a comprehensive account of the
historical events and their contexts by Patricia Hyde, Thomas Arden in Faversham: The Man
Behind the Myth (Faversham: The Faversham Society, 1996).
20. Wine, Arden, xliii–xlv; Jackson, ‘Material for an Edition’, 65–78. The likely glance at the
Leicester scandal comes in Shakebag’s boast that when ‘the widow Chambley’ refused to
allow him to take refuge in her house, he ‘spurned her down the stairs, | And broke her neck’
(15.8–9), which is what Robert Dudley was rumoured to have done to his wife Amy Robsart.
16
Determining the Shakespeare Canon
H. Dugdale Sykes, relying mainly on ‘verbal parallels’, developed the case
for Kyd and E. H. C. Oliphant proposed Marlowe as Kyd’s collaborator.21
Apart from Cornelia, a translation of Robert Garnier’s Cornélie, the one play
certainly by Kyd is The Spanish Tragedy, but he probably also wrote Soliman
and Perseda, which contains phrases also found in Arden of Faversham.22 So,
for example, ‘forge distressful looks’, ‘the railingest knave in Christendom’,
and ‘a greater sum of money | Than either thou or all thy kin are worth’ in
Arden of Faversham (8.56, 13.54, 9.17–18) were compared with ‘forge alluring looks’, ‘the braggingest knave in Christendom’, and ‘It was worth more
than thou and all thy kin are worth’ in Soliman and Perseda.23 Similar likenesses of phrasing were cited between Arden of Faversham and Marlowe’s
Edward II. The standard practice among early twentieth-century scholars
of attribution was to accumulate parallels between a disputed play and the
works, sometimes themselves dubiously ascribed, of their chosen candidate.
The weaknesses of such methods were exposed by Schoenbaum and others.24 Using them, different investigators of the same anonymous play could
reach quite different conclusions about its author, each making a superficially plausible case. The basic assumption was right—that playwrights
have individual habits as phrasemakers and tend to echo themselves more
often than they echo others—but the value of the parallels put forward
could not be reliably assessed, because the search for them had been haphazard and biased by the scholar’s preconceptions. While Soliman and
Perseda’s ‘the braggingest knave in Christendom’ runs to the same pattern
as Arden of Faversham’s ‘the railingest knave in Christendom’, so do ‘the
lying’st knave in Christendom’ in 2 Henry VI and The Taming of the Shrew,
‘the bluntest wooer in Christendom’ in 3 Henry VI, and ‘the prettiest Kate
in Christendom’ in The Taming of the Shrew.25 We need to know how rare
such formulas are and who among all dramatists within an appropriate time
frame used them.
21.See Jackson, ‘Shakespearean Features’, and ‘Material for an Edition’, 90–270, where
the cases for Kyd, made most fully by H. Dugdale Sykes, Sidelights on Shakespeare
(Stratford-upon-Avon: Shakespeare Head Press, 1919), 48–76, and Marlowe, made by E. H.
C. Oliphant, ‘Marlowe’s Hand in Arden of Feversham: A Problem for Critics’, New Criterion, 4
(1926), 76–93, are examined and rejected. See also Wine, Arden, lxxxi–xcii. Vickers’s more
recent arguments supporting Kyd’s candidacy are discussed in Chapters 2–5.
22.Kyd’s authorship of Soliman and Perseda is accepted by Lukas Erne, Beyond ‘The Spanish
Tragedy’: A Study of the Works of Thomas Kyd (Manchester and New York: Manchester UP),
157–202.
23. John J. Murray, ed., The Tragedye of Solyman and Perseda (New York: Garland, 1991), 1.3.201,
2.1.111, 1.4.72; I have modernized Murray’s spelling.
24. Schoenbaum, Internal Evidence.
25. 2 Henry VI, 2.1.123–4; Shrew, Ind.2.24; 3 Henry VI, 3.2.83; Shrew, 2.1.187.
Shakespeare and the Quarrel Scene in
arden of faversham 17
III
The advent of the Chadwyck-Healey electronic database Literature Online
(LION), by which almost the complete corpus of English drama can be
searched, affords a means of acquiring such knowledge and remedying the
defects of the old, discredited methodology.26 Parallels in imagery, wording, and association of ideas can be systematically collected, and the rival
claims of potential authors of the Quarrel Scene evaluated through both
quantitative and qualitative analyses.27
The mechanics of such Literature Online searching have been described
in detail elsewhere,28 but there have been recent improvements, on which
LION’s ‘Information Centre’ is informative; an online tutorial is included.
The database has the standard search functions: words and phrases can
be found, and so can instances of the proximity of one word or phrase to
another. The texts are in original spelling, but it is now possible to check a
box that instructs LION to bring up examples of a searched word or phrase
in almost all spellings in the database. Occasionally the programmers have
missed a possible spelling, nearly always because of a problem over homophones. Thus ‘hart’, for the database, is solely a male deer, and so ‘hart’ is not
included among spellings of ‘heart’, despite its being the most common of
all early modern spellings of the word. But since whenever a search is conducted all spellings that LION has searched are listed at the top of the search
results, it is easy enough to spot such omissions and perform another search
using the omitted spelling. Obviously the greater one’s familiarity with early
modern print culture the greater will be the accuracy of the results obtained.
Coverage of variant old spellings takes into account the difference between
26. Available online at http://lion.chadwyck.co.uk.
27.Jayne M. Carroll and MacDonald P. Jackson, ‘Shakespeare, Arden of Faversham, and
“Literature Online” ’, Shakespeare Newsletter, 54 (2004), 3–4, 6, reports the findings of
Carroll’s ‘The Mystery of Arden of Faversham: Investigating the Play’s Authorship through
Elements of Language and Imagery’ (MA thesis, University of Auckland, 2002), which
employed Literature Online searches. The present chapter is based on my own independent research, using different routines and modes of evaluation, and processing the Quarrel
Scene anew. Even so, Carroll’s careful work has alerted me to a few items I might otherwise
have missed.
28.MacDonald P. Jackson, ‘Determining Authorship: A New Technique’, Research
Opportunities in Renaissance Drama, 41 (2002), 1–14; Defining Shakespeare, 190–217. The
‘Help’ function provides a guide to use of the database. Gabriel Egan offers some useful
caveats at http://gabrielegan.com/publications/Egan2005a.htm.
18
Determining the Shakespeare Canon
early modern and current conventions in the printing of i/j and u/v, so that
searches for ‘ungoverned’ and ‘journey’ automatically include ‘vngouerned’
and ‘iourney’. Checking the box ‘Variant forms’, as well as ‘Variant spellings’,
extends the search to cognates, so that keyword ‘write’ also finds ‘writes’,
‘writeth’, ‘writing’, ‘written’, and ‘wrote’. Also available are a truncation symbol (namely *) and a wild-card symbol covering one letter (namely ?). Within
a phrase, truncation can be used only on the final word. The proximity operators have a default separation of ten words, but this can be extended.29
When searching for phrases it is prudent to use the proximity operators
‘NEAR’ or ‘FBY’ (‘followed by’). By doing so one avoids certain traps. Phrases
typed in full will not be found if they stretch across consecutive lines, and one
must be careful to place a phrase within double quotation marks if it contains
any of the operating words: ‘and’, ‘or’, ‘not’, or ‘near’.30
For the present study, the Quarrel Scene was methodically explored,
through Literature Online searches, for phrases and collocations that occur five
or fewer times in other plays first performed within the two decades 1580–
1600. One hundred and forty-three plays are available for inspection when
these limits are set, but the database has tagged some of them in error: searches
brought up nine plays that are dated before 1580 or after 1600 by Annals of
English Drama, which served as the default authority on dating and authorship for the present enquiry, except that for Shakespeare the Oxford Textual
Companion’s chronology was accepted.31 The few misclassified plays were
29. A line division appears to reduce this span by four words.
30. The words ‘and’, ‘or’, and ‘not’ are Boolean operators: ‘supremacy AND gloworm’ will find
examples in works containing both words; ‘supremacy OR gloworm’ will find examples
in works containing one or both of the two words: ‘supremacy NOT gloworm’ will find
examples of ‘supremacy’ in works that do not contain ‘gloworm’. The operators need not be
capitalized. LION searches do not distinguish between lower case and upper case.
31.Alfred Harbage, rev. Samuel Schoenbaum, rev. S. Waggonheim, Annals of English Drama
975–1700 (London and New York: Methuen, 1989). Plays outside the set chronological limits were Anon., Every Woman in Her Humour (1607); Anon., Thomas of Woodstock; Rowley
and others, The Birth of Merlin; Sidney, The Lady of May (1578); Woods, The Conflict of
Conscience (1572). The Birth of Merlin is dated 1608 in Annals, but was designated a new play
by the Master of Revels in 1622: see Jackson, Defining Shakespeare, 191 n. 3. Woodstock has
been misdated: it is a seventeenth-century play, probably written 1604–6; see MacDonald
P. Jackson, ‘Shakespeare’s Richard II and the Anonymous Thomas of Woodstock’, Medieval and
Renaissance Drama in England, 14 (2002), 17–65, and ‘The Date and Authorship of Thomas of
Woodstock: Evidence and its Interpretation’, Research Opportunities in Medieval and Renaissance
Drama, 46 (2007), 67–100. The investigation of Doctor Faustus, discussed in Section IV,
turned up four more misclassified plays: Anon., The Birth of Hercules (1604); Anon., Clyomon
and Clamydes (1570); Anon., Thorney Abbey (date unknown); Anon., The Trial of Chivalry
(1607). The Oxford Textual Companion’s attributions of plays associated with Shakespeare
were accepted, except that Titus Andronicus is here assigned to Shakespeare and Peele: the
reasons, judged persuasive in the second edition of the Oxford Complete Works, are outlined
Shakespeare and the Quarrel Scene in
arden of faversham 19
ignored. LION contains Shakespeare’s plays not only in their First Folio text,
but also in a quarto text, where one was published, and in the modern spelling edition of the nine-volume Cambridge edition of 1863–6.32 Parallels in
imagery and ideas were recorded only if passages had at least one prominent
word in common. Apart from compounds, single words linking the Quarrel
Scene to five or fewer places within the corpus were disregarded, unless supported by striking similarities of context. Phrases and collocations rare enough
to occur five or fewer times were retained even if one or more afforded slightly
fuller links to Arden of Faversham than did others. Links were not recorded
when collocated words were used in entirely different senses: thus ‘loathsome
weeds’ in line 67 of the Quarrel Scene provides a link to A Knack to Know a
Knave, where the ‘loathsome weeds’ are again plants, but not to Caesar and
Pompey, where ‘loathsome sable weeds’ are mourning clothes. No doubt an
element of subjectivity enters into decisions whether a parallel is close enough
to be recorded, and no doubt some relevant data have been accidentally overlooked, but searching was carried out thoroughly and without conscious bias.
The findings are presented in Appendix 1.
Before summarizing and evaluating them, I want to stress again that the
comprehensive collecting of rare verbal parallels between an anonymous
scene and the full range of theoretically possible authors—when a definition of ‘rare’ is established in advance—is a completely different process
from the uncontrolled accumulation of parallels between a disputed work
and an authorial candidate whom the scholar favours. The new procedure
allows equal scope for Shakespeare, Marlowe, Kyd, Greene, Peele, or any
other playwright, or nobody at all, to furnish the play or plays that share the
largest number of rare phrases and collocations with the Quarrel Scene, and
so to emerge as the strongest candidate for its authorship. And this technique of Literature Online searching has, in earlier investigations of plays
and passages involving Fletcher, Greene, Heywood, Marlowe, Middleton,
Nashe, Peele, William Rowley, Shakespeare, Shirley, Theobald, Webster,
and Wilkins proved successful in identifying the playwrights who are
by Vickers, Shakespeare, Co-Author, 148–243; also the Oxford editors’ change of mind over
Edward III (see n. 12 above) has been taken into account, as has recent work on 2 and 3 Henry
VI (discussed in Chapter 3) that has led to their being assigned to ‘Shakespeare and others’.
But in references to Edward III and 2 and 3 Henry VI ‘others’ is shorthand for ‘at least one
other’.
32. William George Clark, William Aldis Wright, and John Glover, eds., The Works of William
Shakespeare, 9 vols. (Cambridge: Macmillan, 1863–6). Shakespearean quarto texts in LION
include both ‘good’ and ‘bad’ quartos, but not mere reprints.
20
Determining the Shakespeare Canon
known, or strongly suspected on other grounds, to have written particular
dramatic scenes or passages.33 Each of these men, even such an inveterate borrower as Webster, had his own habits of expression—a tendency to
repeat certain ideas and images in peculiar combinations of words.
Especially analogous to the present investigation are Paul Vincent’s systematic LION searches of the whole of 1 Henry VI. These picked out 2.4,
4.2–5, and 4.7.1–32 as Shakespeare’s because in those parts of the play exclusive links to a restricted ‘early Shakespeare’ corpus—one without Titus
Andronicus, 2 Henry VI, and 3 Henry VI—appreciably outnumbered exclusive
links to any one of the approximately equal-sized corpora of playwrights
who had at one time or another been proposed as his co-authors. Vincent’s
results derived from LION agreed almost exactly with those of Gary
Taylor’s completely different mode of analysis and have since been endorsed,
with one minor divergence, by Brian Vickers.34 Moreover, Vincent’s control tests on passages from undisputed plays by Shakespeare, Nashe, Greene,
Marlowe, and Peele all succeeded in identifying the true author.
Of the 134 plays that were searched to furnish the data for the present
chapter’s Appendix 1 and that fall within the chronological limits 1580–
1600,35 thirty have four or more links to the Quarrel Scene in Arden of
Faversham. These plays are listed in Table 1.1.
Links to plays by Shakespeare are overwhelmingly predominant. It is
surely of further significance that four of the five plays with the most links
to the Quarrel Scene are his earliest, according to the Oxford chronology.
33.Beside the items cited in n. 28, see MacDonald P. Jackson, ‘Late Webster and his
Collaborators: How Many Playwrights Wrote A Cure for a Cuckold?’, Papers of the
Bibliographical Society of America, 95 (2001), 295–313; ‘A Lover’s Complaint Revisited’,
Shakespeare Studies, 32 (2004), 267–94; ‘Anything for a Quiet Life, IV.ii.1–44’, Notes and Queries,
251 (2006), 86–9; ‘John Webster, James Shirley, and the Melbourne Manuscript’, Medieval
and Renaissance Drama in England, 19 (2006), 21–44; ‘Looking for Shakespeare in Double
Falsehood: Stylistic Evidence’, The Quest for ‘Cardenio’: Shakespeare, Fletcher, Cervantes, and
the Lost Play, ed. David Carnegie and Gary Taylor (Oxford: OUP, 2012), 133–61; Gary
Taylor, ‘Middleton and Rowley—and Heywood: The Old Law and New Attribution
Technologies’, Papers of the Bibliographical Society of America, 96 (2002), 165–217; ‘A History
of The History of Cardenio’, Quest, 11–61; Taylor with John V. Nance, ‘Four Characters
in Search of a Subplot’, Quest, 192–213; Paul Vincent, When ‘harey’ Met Shakespeare: The
Genesis of ‘The First Part of Henry the Sixth’ (Saarbrücken: VDM Verlag Dr. Müller, 2008),
303–24, esp. control tests on passages of known authorship, 320–4.
34. Vincent, When ‘harey’ Met Shakespeare; Gary Taylor, ‘Shakespeare and Others: The
Composition of Henry the Sixth, Part One’, Medieval and Renaissance Drama in England, 7
(1995), 145–205; Brian Vickers, ‘Incomplete Shakespeare’. Taylor differed from Vincent in
giving Shakespeare 4.6, Vickers in denying Shakespeare 4.7.1–32.
35. It is possible Literature Online includes within the 1580–1600 limits a few plays besides the nine
listed in n. 31 that fall outside those limits according to Annals but that these yielded no links.
Shakespeare and the Quarrel Scene in
arden of faversham 21
Table 1.1 Summary of LION links to Arden Quarrel Scene
Author(s)
Title (probable date of first
performance)
Shakespeare and others
Shakespeare
Shakespeare and others
Shakespeare
Shakespeare
Shakespeare
Shakespeare
Shakespeare
Yarington
Greene
Shakespeare
Shakespeare and Peele
Shakespeare and others
Shakespeare and others
Shakespeare
Shakespeare
Shakespeare
Jonson
Greene
Anon.
3 Henry VI (1591)
The Two Gentlemen of Verona (1590–1)
2 Henry VI (1591)
The Taming of the Shrew (1590–1)
Romeo and Juliet (1595)
King John (1596)
Richard II (1595)
Richard III (1592–3)
Two Lamentable Tragedies (1594)
James IV (1590)
Love’s Labour’s Lost (1594–5)
Titus Andronicus (1592)
1 Henry VI (1592)
Edward III (1590)
A Midsummer Night’s Dream (1595)
The Merchant of Venice (1596–7)
2 Henry IV (1597–8)
Every Man in His Humour (1598)
Selimus (1592)
The Troublesome Reign of King John
(1591)
The Virtuous Octavia (1598)
Hamlet (1600–1)
Henry V (1598–9)
Julius Caesar (1599)
The Spanish Tragedy (1587)
Soliman and Perseda (1592)
Every Man out of His Humour (1599)
The Jew of Malta (1589)
Edward II (1592)
Love’s Metamorphosis (1590)
Brandon
Shakespeare
Shakespeare
Shakespeare
Kyd
Kyd
Jonson
Marlowe
Marlowe
Lyly
No. of links
22
12
12
11
11
10
10
9
8
8
7
7
6
6
5
5
5
5
5
5
5
4
4
4
4
4
4
4
4
4
Likenesses between Arden of Faversham and ‘Shakespeare’s first tetralogy, the
three Henry VI plays and Richard III’ have often been noted,36 but it is also
remarkable that whereas there are twenty-two links to 3 Henry VI and twelve
to 2 Henry VI, there are only six to 1 Henry VI, to which, as Taylor, Vincent,
and Vickers have shown, Shakespeare’s contribution was probably not large.
The authorial status of 2 and 3 Henry VI is briefy considered towards the
36. Wine, Arden, lxxxix.
22
Determining the Shakespeare Canon
end of this chapter and in Chapters 2 and 3 becomes crucial to the case for
Shakespeare’s part-authorship of Arden. The two non-Shakespearean plays
that occupy the highest positions on the table are Yarington’s Two Lamentable
Tragedies and Greene’s James IV. Yarington interweaves two ‘domestic tragedies’ based on recent crimes, as is Arden of Faversham, which, as the first extant
example of the genre, clearly served as Yarington’s model.37 Three of the
links between Two Lamentable Tragedies and the Quarrel Scene are with the
last few lines of the scene, after Bradshaw’s arrival, when the dialogue shifts
into the prosaic: ‘’tis almost supper time’, which the two plays share, is the
small change of domestic conversation. Yarington is unknown as a dramatist
apart from Two Lamentable Tragedies, which is a surprise to nobody who has
read them: lamentable they certainly are. James IV, arguably Greene’s best
play, is closely contemporary with Arden of Faversham, and the only other
play by Greene that belongs with the thirty affording four or more links to
the Quarrel Scene is Selimus, which is uncertainly ascribed. Anybody who
scrutinizes the links with 2 and 3 Henry VI, The Two Gentlemen of Verona, The
Taming of the Shrew, Romeo and Juliet, Two Lamentable Tragedies, and James IV
will see that those with the five Shakespeare plays are not only more numerous but of superior quality, including vivid and complex figurative modes of
speech, largely lacking from the two non-Shakespearean plays, where the
parallels with Arden of Faversham are mostly in wording only. And it is noteworthy that the eight Shakespeare plays that top the table cover a range of
genres: there are five English histories (one of which is also the personal tragedy of Richard II), two comedies, and a romantic tragedy.
The plays listed in Table 1.1 differ in length. When the figures are adjusted to
register links per 20,000 words, the picture changes slightly, without affecting the predominance of Shakespeare.38 The top places are occupied by 3
37. One of Yarington’s tragedies is based on a murder committed on 23 August 1594. See E. K.
Chambers, The Elizabethan Stage (Oxford: Clarendon Press, 1923), III, 518. I have accepted the
Annals dating and ascription of Two Lamentable Tragedies, but Andrew Gurr has pointed out to
me (email of 7 October 2007) that Chambers, on whose entry Annals relied, did not have ‘the
evidence that turned up later of Yarington being a registered scrivener in 1604’. This may have
modified his scepticism about Greg’s theory that Yarington was merely the copyist, late in
1599, of the play, which Greg identified with plays for which Henslowe paid Day, Haughton,
and Chettle in the period 1599–1601. Two Lamentable Tragedies was first published in 1601. In
style it strikes me as homogeneous and unlike the work of Day, Haughton, or Chettle.
38.I have drawn on Marvin Spevack’s A Complete and Systematic Concordance to the Works
of Shakespeare, 9 vols. (Hildesheim: Georg Olms, 1968–80) for the numbers of words in
Shakespeare’s plays; for other plays I have used computer counts by Ward E. Y. Elliott and
Robert J. Valenza, ‘And Then There Were None: Winnowing the Shakespeare Claimants’,
Computers and the Humanities, 30 (1996), 191–245, when available, or my own estimates based
on line totals and sample-derived averages of words per line.
Shakespeare and the Quarrel Scene in
arden of faversham 23
Henry VI (18.9), The Two Gentlemen of Verona (14.3), The Taming of the Shrew
(10.8), King John (9.8), 2 Henry VI (9.8), Richard II (9.2), Romeo and Juliet (8.2),
James IV (8.1), Two Lamentable Tragedies (7.4), Titus Andronicus (7.1), Love’s
Labour’s Lost (6.6), Richard III (6.4), A Midsummer Night’s Dream (6.2), and
Edward III (6.2). The length of Richard III takes it out of the top ten and the
gap between 3 Henry III and The Two Gentlemen of Verona is greatly reduced.
Might the author or co-author of Arden of Faversham have been, like
Viola de Lesseps in the movie Shakespeare in Love, a devout admirer of
the young Shakespeare’s works, with a recall of his dialogue that ensured
the repetition of his phrases, collocations, and images in his own composition? The short answer is no. None of Shakespeare’s plays was in print
in 1592, when Arden of Faversham was published. So any familiarity with
them would have to have been acquired through performances. But even
some mysterious actor-turned-playwright, who had taken Shakespearean
roles, would not serve the turn, since it is probable that no Shakespeare play
tabled above was written before Arden of Faversham, and it is virtually certain that several of those with many links were written after it. Moreover,
Shakespeare continues to be the dominant presence in the second half of
that thirty–play list: links persist in Shakespeare plays that according to
the Oxford chronology were written after 1595 and that even on an ‘early
start’ theory certainly postdate 1592, when Arden was undeniably in existence. Of the remaining Shakespeare plays that fall within the period 1580–
1600, The Comedy of Errors, Much Ado About Nothing, and The Merry Wives of
Windsor all have three links to the Quarrel Scene, along with half a dozen
non-Shakespearean plays, and 1 Henry IV has one; only As You Like It has
none.39 Besides, as will shortly appear, the Quarrel Scene displays highly
significant links to Shakespeare’s narrative poems, especially The Rape
of Lucrece, which can confidently be dated 1593–4. These considerations
seem decisive, quite apart from the extreme unlikelihood that some totally
unknown dramatist, however steeped in Shakespeare, could prove capable
of creating anything as good as the Quarrel Scene.
For these reasons of chronology, the results cannot be explained away as
a byproduct of Arden of Faversham’s textual transmission—that is, as created
by actors who introduced phrases from Shakespeare plays during a process
of memorial reconstruction. Besides, the links with Shakespeare are not
39. It is also worth noting that the extreme shortness of The Comedy of Errors means that when its
three links are adjusted for length (4.2 per 20,000 words) it takes its place on the list ahead of
2 Henry IV, Henry V, Julius Caesar, and Hamlet, as its relative earliness might have led one to
expect.
24
Determining the Shakespeare Canon
merely verbal and superficial but deeply imbedded in the Quarrel Scene’s
poetic and dramatic structure. And textually this scene is remarkably free
from signs of memorial error.
Perhaps Shakespeare, enthralled by the surviving play of an otherwise
unknown dramatist, absorbed the imagery and phrasing of the Quarrel
Scene so thoroughly that for a decade they affected his writing far more
than they affected anybody else’s? In Chapter 5, after other kinds of evidence have been advanced, this possibility will be given fuller consideration. But it seems a remote one. Even if he had been an actor in Arden of
Faversham, why should it be so much more influential than all the other
plays in which he acted? Shakespeare’s indebtedness to, and rivalry with,
Marlowe has been explored by many critics; but, although he was still
exorcizing this ‘familiar ghost’ in his last unaided play, The Tempest (1611),
nothing by Marlowe has the kind of verbal influence on Shakespeare’s plays
of the 1590s that we would have to posit in the case of the Quarrel Scene.40
In fact, Literature Online analysis of two control samples from Marlowe’s
Doctor Faustus may serve as a further test of the methodology applied to the
Quarrel Scene.
IV
The samples selected are two of Faustus’s finest and most famous speeches—
his rhapsodic apostrophe to Helen and his final soliloquy, as he waits to be
dragged by the devils to hell.41 Parallels, recorded in Part B of Appendix 1,
between these speeches in Doctor Faustus and plays of 1580–1600, were collected in the same way, and according to the same rules, as those with the
Quarrel Scene; phrases and collocations had to be rare enough to occur no
more than five times outside Doctor Faustus itself.
The eight plays listed in Table 1.2 have four or more links to the Doctor
Faustus samples. Five plays have three links: Edward II by Marlowe,
Englishmen for My Money by Haughton, King John and Hamlet by Shakespeare,
and 1 Henry VI by Shakespeare and others.
40. There is an excellent account of Marlowe’s influence on Shakespeare by Jonathan Bate, The
Genius of Shakespeare (London: Picador-Macmillan, 1997), 101–32.
41. The text used is Christopher Marlowe, Doctor Faustus, ed. John D. Jump (London: Methuen,
1962), 18.99–118 and 19.132–90. Marlowe’s authorship of these speeches has never been
doubted.
Shakespeare and the Quarrel Scene in
arden of faversham 25
Table 1.2 Summary of LION links to Doctor Faustus samples
Author(s)
Title (probable date of first
performance)
Marlowe and Nashe
Marlowe
Yarington
Anon.
Marlowe
Anon.
Greene and Lodge
Kyd
Dido Queen of Carthage (1586)
1 Tamburlaine (1587)
Two Lamentable Tragedies (1594)
Arden of Faversham (1588–91)
The Massacre at Paris (1593)
The Troublesome Reign of King John (1591)
A Looking Glass for London (1588)
Soliman and Perseda (1592)
No. of links
10
6
5
5
4
4
4
4
We would have no difficulty identifying the author of Faustus’s speeches
from these results. Two plays by Marlowe head the table, with a third in
the top eight and a fourth in the top thirteen, and there are links to each of
Marlowe’s other two plays: 2 Tamburlaine has two and The Jew of Malta one.
Marlowe is the dominant known presence in the inventory, a dominance
attained despite the fact that Dido Queen of Carthage, The Massacre at Paris,
and 1 Tamburlaine are (in that order) the three shortest plays of the eight
listed.42 Moreover, like Arden of Faversham’s parallels with Shakespeare,
Doctor Faustus’s with Marlowe tend to be of higher quality than those with
other dramatists. No non-Marlovian parallel—except that with Marston’s
Antonio’s Revenge at Doctor Faustus, 19.174–6, a deliberate echo—rivals those
with 2 Tamburlaine and Dido Queen of Carthage at 18.99–100, Dido Queen of
Carthage at 18.101 and 18.106–7, Edward II at 19.136–41, or 1 Tamburlaine at
19.157.
Nashe’s contribution to Dido Queen of Carthage is generally agreed to
have been minimal. The fact that the largest numbers of links are with
Marlowe’s first two plays supports an early date for Doctor Faustus.43 It is
not surprising that among the non-Marlovian plays with the most links
to Faustus’s speeches are Arden of Faversham, Yarington’s Two Lamentable
4 2. In terms of links per 20,000 words The Massacre at Paris (15,930 words, 5.0 links per 20,000
words) exceeds Two Lamentable Tragedies (estimated length 21,600 words, 4.6 links per
20,000 words). The text of the undated octavo of The Massacre at Paris is poor.
43.For recent adjudication on the controversy over whether Doctor Faustus should be dated
about 1588–9 or about 1592, see David Bevington and Eric Rasmussen, eds., Doctor Faustus
A- and B-texts (1604, 1616) (Manchester: Manchester UP, 1993), 1–3. It should be added that
Thomas Merriam, ‘Marlowe and Nashe in Dido Queen of Carthage’, Notes and Queries, 245
(2000), 425–8, argues for assigning Nashe a larger share in that play than he has been generally granted.
26
Determining the Shakespeare Canon
Table 1.3 Summary of LION links for Shakespeare and Marlowe
Author (no. of plays)
Shakespeare (23)
Marlowe (6/7)
No. of links
Quarrel Scene
Faustus’s speeches
154
14
23
26
Tragedies, and A Looking Glass for London: the domestic tragedies have their
didactic elements, and Looking Glass belongs to the genre of ‘warning literature’, so naturally they share some language with the damned Faustus
facing his horrible doom. All the parallels with the domestic tragedies are
with Faustus’s final soliloquy, none with his apostrophe to Helen. The
influence of Marlowe on Lodge is well known.
Perhaps the most telling point is that the Doctor Faustus samples yield
twenty-six links to Marlowe’s six plays, but only twenty-three to the
twenty-three plays by Shakespeare that fall within the search limits.44 The
contrast with the Quarrel Scene in Arden of Faversham, which yields 154
links to Shakespeare and fourteen to Marlowe (in this case to the full corpus of seven plays) is worth highlighting in Table 1.3.
The control test on Doctor Faustus confirms the validity of the Literature
Online technique and of our interpretation of the data from Arden of
Faversham’s Quarrel Scene.
V
Literature Online also enables us to assess the rarity of some of the Quarrel
Scene’s complex congeries of words, ideas, and images that occur or recur
in the Shakespeare canon. The focus in this section is on a select few of
these. The evidence is merely supplementary to the main LION investigation. It is more vulnerable, because it has been compiled, not by impartially
searching for rare phrases and collocations within a pre-established range of
works by many different authors and discovering which plays yield the most
links to the Quarrel Scene, but by noting ostensibly significant features that
the Quarrel Scene has in common with early Shakespearean works and
then checking how rare within the LION database they are. Theoretically,
a believer in Kyd’s authorship of the Quarrel Scene might find comparable
4 4. The figure of twenty-three includes Edward III.
Shakespeare and the Quarrel Scene in
arden of faversham 27
rarities connecting it to his canon. But no scholar has come remotely near
doing so in practice.
In her pioneering study of Shakespeare’s imagery, Caroline Spurgeon
remarked: ‘I do not find, in all my search of the other dramatists, any single image of frosts and sharp winds nipping buds, which is so common
in Shakespeare.’ 45 Mosby’s words at 8.5–6 are thus good indicators of
Shakespeare’s hand. His anxiety, he says, ‘nips me as the bitter northeast
wind | Doth check the tender blossoms in the spring’. The basic image
shared with many Shakespearean passages is of untimely cold checking tender
new growth within a season noted for its changeable weather: as The Rape
of Lucrece puts it, ‘little frosts . . . sometime threat the spring’ (331). To parallels
that qualify for inclusion in the appended inventory we may add: ‘Unruly
blasts wait on the tender spring’ (The Rape of Lucrece, 869–70); ‘Rough winds
do shake the darling buds of May’ (Sonnets, 18.3); ‘Confounds thy fame, as
whirlwinds shake fair buds’ (The Taming of the Shrew, 5.2.140); ‘Berowne is
like an envious sneaping frost | That bites the first-born infants of the spring’
(Love’s Labour’s Lost, 1.1.100–1); ‘Death lies on her like an untimely frost |
Upon the sweetest flower of all the field’ (Romeo and Juliet, 4.5.28–9); ‘And
like the tyrannous breathing of the north | Shakes all our buds from growing’ (Cymbeline, 1.3.36–7). More intricate, but drawing on the same imagery,
is the simile in 2 Henry IV:
as in an early spring
We see th’ appearing buds, which to prove fruit
Hope gives not so much warrant, as despair
That frosts will bite them.
(1.3.38–41)
While superficially similar lines by Kyd contain words used by Shakespeare,
they are utterly different in character. The Spanish Tragedy has:
But in the harvest of my summer joys
Death’s winter nipp’d the blossoms of my bliss.
Had Proserpine no pity on thy youth?
But suffer’d thy fair crimson-colour’d spring
With wither’d winter to be blasted thus?
(1.1.12–13)
(3.13.147–9)46
45. Caroline F. E. Spurgeon, Shakespeare’s Imagery and What it Tells Us (Cambridge: CUP, 1935;
repr. Boston: Beacon Press, 1958), 91.
46. The three Kyd passages are here quoted from Thomas Kyd, The Spanish Tragedy, ed. Philip
Edwards (London: Methuen, 1959).
28
Determining the Shakespeare Canon
In each case Kyd is simply making the cliché connection of life with spring
or summer and death with winter, and the first passage, especially, is a muddle, with winter nipping blossoms while the fruits of summer are enjoyed.
The author of Arden of Faversham, 8.5–6, in contrast, grounds his imagery
in experience, writing, like Shakespeare, with a countryman’s eye on the
weather, and even the direction from which the wind blows.
Use of the image of the muddied fountain to indicate emotional turmoil
or moral taint is also habitual with Shakespeare. In the whole Literature
Online database (covering poetry, drama, and prose collections), no other
work furnishes such a close parallel to Alice’s lines as does The Taming of
the Shrew:
I deserve not Mosby’s muddy looks.
A fount once troubled is not thickened still;
Be clear again, I’ll ne’er more trouble thee.
(Arden of Faversham, 8.132–4)
A woman mov’d is like a fountain troubled,
Muddy, ill-seeming, thick, bereft of beauty.
(The Taming of the Shrew, 5.2.142–3)
No other play, poem, or prose work collocates ‘fount(ain)’, ‘troubled’,
‘muddy’, and ‘thick(ened)’ in any figurative way. William Clark’s poem ‘The
Grand Trial’ (1685) contains ‘what formerly was clear, | Now muddy, thick,
and troubled doth appear’, which even has the extra verbal link of ‘clear’, but
the poet is literally describing a liquor shaken in a crystal bottle. Searching
early modern drama of 1576–1642, one finds very few non-Shakespearean
instances of ‘thick’ or ‘muddy’ fountains or springs. Lyly’s Endymion offers
‘over mine eyes either a dark mist or upon the fountain a deep thickness’,
but the thickness appears to be an enveloping haze, rather than murkiness in
the water. Heywood has ‘Some difference she entreats your honours make
| ’Twixt crystal fountains and foul muddy springs’ in If You Know Not Me
(1604) and ‘Raise up the mud in clear springs when she drinks’ in Jupiter and Io
(1635), but the second of these lacks the figurative dimension.
Yet besides the Shrew parallel and the other examples of the image
recorded in the inventory of links in Appendix 1, Part A (under ‘I deserve
not Mosby’s muddy looks . . .’ from lines 132–4)—two in 2 Henry VI, one
in Titus Andronicus, and one in Richard II—and the three in The Rape of
Lucrece that are listed in Appendix 1, Part A, n. 22, there is a passage in
Shakespeare and the Quarrel Scene in
arden of faversham 29
Troilus and Cressida that bears a particular close resemblance to the Arden of
Faversham one: Achilles says, ‘My mind is troubled, like a fountain stirr’d,
| And I myself see not the bottom of it’, to which Thersites replies, ‘Would
the fountain of your mind were clear again’ (3.3.308–11). Literature Online
yields no other instance of an image involving ‘fount(ain)’, ‘troubled’, and
‘clear again’ with figurative application, as in Arden and Troilus.
In Hamlet the turbid water is implicit in the key adjectives: ‘the people muddied, | Thick and unwholesome in their thoughts and whispers’
(4.5.81–2); and there are traces of the image in the Sonnets—‘Roses have
thorns, and silver fountains mud’ (35.2), which also has the roses–thorns
link to the Quarrel Scene—and in The Winter’s Tale, where Leontes asks,
‘Dost think I am so muddy, so unsettled’ (1.2.325), the underlying image
again being of murky waters.
It must be emphasized that, although the quarto of Arden of Faversham
reads ‘a fence of trouble’ and Wine’s ‘fount once troubled’ is an emendation, first proposed by W. Headlam in 1903, the small portion of the argument for Shakespeare’s authorship of the Quarrel Scene that is based on its
use of the Shakespearean ‘muddied fountain’ image complex is not circular.47 Rather, the emendation is required in the interests of coherence, and
the parallels would strongly support it, whoever had written the pertinent
lines of Arden of Faversham. Whether we emend to ‘fount once troubled’,
‘fount of trouble’ (equivalent to the Latin fons doloris), or even ‘fountain
troubled’, a fountain, not a fence, is so obviously needed for the poetic sense
that it would be perverse of any editor to retain Q’s ‘fence’, which was evidently due to an easy misreading of Elizabethan handwriting.
Also typical of the young Shakespeare are Mosby’s ‘It is not love that loves
to anger love’ and Alice’s response, ‘It is not love that loves to murder love’
(8.58–9). Shakespeare is addicted to this ‘set[ting] the word itself against the
word’, as it is called in Richard II, 5.3.122. His use of ‘the reflexive conceit’ was
discussed by Frank O’Connor, who considered it ‘very curious, and very personal’.48 Its basic form involves repetition of a reflexive pronoun: ‘Narcissus
so himself himself forsook’ (Venus and Adonis, 161), ‘When he himself himself
confounds’ (The Rape of Lucrece, 160), ‘Thyself thyself misusest’ (Richard III,
47.W. Headlam, ‘Arden of Feversham: “Pathaires” ’, The Athenaeum, 26 December 1903. See
Wine’s commentary on 133, and Jackson, ‘Material for an Edition’, 48–51.
48. Frank O’Connor, Shakespeare’s Progress (New York: Collier Books, 1961), 49. ‘Frank
O’Connor’ is a pseudonym for Michael O’Donovan and the book is a revised version of his
The Road to Stratford (1948).
30
Determining the Shakespeare Canon
4.4.297),49 ‘Lest Rome herself be bane unto herself ’ (Titus Andronicus, 5.3.73),
‘Therefore thy later vows, against thy first, | Is in thyself rebellion to thyself ’
(King John, 3.2.288–9), ‘For having traffic with thyself alone, | Thou of thyself
thy sweet self dost deceive’ (Sonnet 4.9–10), ‘To pay ourselves what to ourselves
is debt’ (Hamlet, 3.2.193), ‘there is left us | Ourselves to end ourselves’ (Antony
and Cleopatra, 4.14.21–2), and so on. There are at least forty examples in the
Shakespeare canon, ten occurring in The Rape of Lucrece, but I have encountered
only two in non-Shakespearean plays of 1580–1600.50 Paradox and antithesis are
of the essence. As O’Connor explains, ‘Everyone and everything contains its
own opposite by which it is saved or destroyed.’51 More elaborate forms of the
conceit repeat nouns, since ‘Anything that is capable of being personified may
be presented as an antithesis of which the two parts are identical.’52 O’Connor
quotes, among other examples, ‘And time doth weary time with her complaining’, ‘There’s no such sport as sport by sport o’erthrown’, and ‘Light, seeking
light, doth light of light beguile’, where ‘light’ means successively intelligence,
knowledge, the eyes, and sight—study ruins one’s eyesight.53 But the noun
most often played upon in this way is ‘love’, as in the Shakespearean passages
noted in Apendix 1, Part 1, as forming links to Arden of Faversham, 8.58–9. For
example, the reluctant Adonis, wooed by Venus, asserts ‘My love to love is
love but to disgrace it’ (Venus and Adonis, 412); the Duchess of York tells her
husband ‘Love loving not itself, none other can’ (Richard II, 5.3.88); Richard of
Gloucester manipulates Anne with ‘This hand, which for thy love did kill thy
love, | Shall for thy love kill a far truer love’ (Richard III, 1.2.190–1). And there
is the quibbling on ‘love’ in The Two Gentlemen of Verona, 1.2.31–2, where Julia
says, ‘They do not love that do not show their love’ and Lucetta retorts, ‘O, they
love least that let men know their love’, an exchange prefaced by Julia’s line ‘Fire
that’s closest kept burns most of all’ (1.2.30), a Shakespearean image employed
by Alice Arden (8.48–9). The paradoxical play on ‘love’ by Alice and Mosby in
the Quarrel Scene does more than provide verbal links to Shakespeare’s plays: it
reflects a habitual turn of mind.
49. The Tragedy of King Richard III, ed. John Jowett (Oxford: OUP, 2000). The phrase comes
from the quarto (1597). Evans, Riverside Shakespeare, follows the First Folio (1623) and reads
‘Thyself is self-misus’d’ (4.4.374).
50. Peele’s The Arraignment of Paris has ‘And for myself myself must speaker be’ (Q 1584, D3r,
STC 19530); the anonymous The True Chronicle History of King Leir has ‘None but himself
hath dispossessed himself ’ (Q 1605, C2v, STC 15343).
51. O’Connor, Shakespeare’s Progress, 47.
52. O’Connor, Shakespeare’s Progress, 48.
53. O’Connor, Shakespeare’s Progress, 48; the quotations are from The Rape of Lucrece, 1570;
Love’s Labour’s Lost, 5.2.153; Love’s Labour’s Lost, 1.1.77.
Shakespeare and the Quarrel Scene in
arden of faversham 31
Alice’s entreaty to Mosby, as, after offending him, she seeks to win him
back, is strongly reminiscent of similar entreaties in two early Shakespeare
plays. Alice says:
Nay, hear me speak, Mosby, a word or two;
I’ll bite my tongue if it speak bitterly.
Look on me, Mosby, or I’ll kill myself;
Nothing shall hide me from thy stormy look.
. . .
Wilt thou not look? Is all thy love overwhelmed?
Wilt thou not hear? What malice stops thine ears?
Why speaks thou not? What silence ties thy tongue?
Thou hast been sighted as the eagle is,
And heard as quickly as the fearful hare,
And spoke as smoothly as an orator,
When I have bid thee hear or see or speak.
And art thou sensible in none of these?. . .
And I deserve not Mosby’s muddy looks.
(8.110–13, 123–30, 132)
In The Comedy of Errors, Adriana pleads with Antipholus of Syracuse,
whom she takes to be her husband, but who is actually his perplexed and
long-lost twin:
Ay, ay, Antipholus, look strange and frown,
Some other mistress hath thy sweet aspects:
I am not Adriana, nor thy wife.
The time was once, when thou unurg’d wouldst vow
That never words were music to thine ear,
That never object pleasing in thine eye,
That never touch well welcome to thy hand,
That never meat sweet-savor’d in the taste,
Unless I spake, or look’d, or touch’d, or carv’d to thee.
How comes it now, my husband, O, how comes it,
That thou art then estranged from thyself?
Thyself I call it, being strange to me,
That undividable incorporate,
Am better than thy dear self ’s better part.
Ah, do not tear thyself away from me.
(2.2.110–24)
Sight, sound, and speech are worked into similar rhetorical patterns that
emphasize mutual delight, though Adriana also mentions touch and
taste. Antipholus’s ‘frown’ has its counterpart in Mosby’s ‘stormy look’
and ‘muddy looks’. Adriana’s equation of the two ‘selves’ of husband and
32
Determining the Shakespeare Canon
wife, which she elaborates in the phrase ‘if we two be one’ (142), parallels Mosby’s thought about Alice a little earlier in the Quarrel Scene: ‘but
she’s myself, | And holy church rites makes us two but one’ (Arden, 8.37–
8). When Adriana urges Antipholus, ‘Ah, do not tear thyself away from
me’, she echoes Alice’s use of the rare phrase ‘tear away’ as she asserts that
she will ‘tear away the leaves’ of the prayer-book that had converted her
and replace them with Mosby’s love letters (115–20).54 The chaste Adriana
goes on to invite Antipholus to contemplate how upset he would be were
she ‘licentious’, ‘contaminate’ by ‘lust’, and argues that if he is unfaithful
she herself, husband and wife being one, is necessarily ‘possess’d with an
adulterate blot . . . strumpeted by thy contagion’ (130–46). The hypothetical in The Comedy of Errors thus echoes the reality in Arden of Faversham,
where Alice has ‘an odious strumpet’s name’ (72) and is reviled by Mosby
as ‘a wanton giglot’ (87).
Also close to Alice’s supplication is Queen Margaret’s to the King in 2
Henry VI:
What, dost thou turn away and hide thy face?
I am no loathsome beggar, look on me.
What? art thou like the adder waxen deaf?
Be poisonous too, and kill thy forlorn queen.
Is all thy comfort shut in Gloucester’s tomb?
Why then Dame Margaret was ne’er thy joy.
Erect his statuë and worship it,
And make my image but an alehouse sign.
Was I for this nigh wrack’d upon the sea,
And twice by awkward wind from England’s bank
Drove back again unto my native clime?
(3.2.74–84)
‘Look on me, Mosby’, begs Alice (Arden, 8.112), as her lover turns scornfully
aside; and the Queen makes the same desperate request to Henry: ‘look on
me’. Alice moves from ‘Wilt thou not look?’ to ‘Wilt thou not hear? What
malice stops thine ears?’, and Margaret from ‘What, dost thou turn away
and hide thy face? | I am no loathsome beggar, look on me’ to ‘What? art
thou like the adder waxen deaf?’ Alice talks of killing herself and Margaret
of being killed by Henry’s disdain. Margaret’s line ‘Was I for this nigh
54. As explained in Appendix 1, Part A, n. 20, ‘tear away’ appears only in Arden of Faversham,
The Two Gentlemen of Verona, and The Comedy of Errors among plays of 1580–1600.
Shakespeare and the Quarrel Scene in
arden of faversham 33
wrack’d upon the sea’ calls to mind the earlier occasion on which Alice
seeks a reconciliation with Mosby:
Have I for this given thee so many favours,
Incurred my husband’s hate, and—out alas!—
Made shipwreck of mine honour for thy sake?
And dost thou say, ‘Henceforward know me not’?
(1.187–90)
More significantly still, Alice’s lines have a close parallel in Love’s Labour’s
Lost. Alice asks, ‘Wilt thou not look? . . . Wilt thou not hear? . . . Why speaks
thou not?’ (8.123–5) and avers:
Thou hast been sighted as the eagle is,
And heard as quickly as the fearful hare,
And spoke as smoothly as an orator.
(8.126–8)
In Love’s Labour’s Lost Berowne declares:
A lover’s eyes will gaze an eagle blind.
A lover’s ear will hear the lowest sound
When the suspicious head of theft is stopp’d. . . .
And when Love speaks, the voice of all the gods
Make heaven drowsy with the harmony.
(4.3.331–3, 341–2)
As in Arden of Faversham, there is the progression from seeing to hearing to eloquent speaking (though here too, as in Errors, touch and taste
intervene). In Love’s Labour’s Lost and Arden of Faversham, the emblem for
sharp-sightedness is the eagle. In both passages love is thought of as honing
the senses. For Berowne, ‘Love’s feeling is more soft and sensible | Than are
the tender horns of cockled snails’ (334–5), and Alice uses ‘sensible’ (Arden,
8.130) with the same rare meaning of ‘capable of feeling or perception, sensitive’ (not the more usual ‘judicious, having good sense’): ‘And art thou
sensible in none of these?’, where the last four words mean ‘in none of these
faculties’.55 Both speakers imagine fear or suspicion inducing a keen alertness to sound, and contrast this hypersensitivity with ears being ‘stopped’.
Finally, in an essay published in 1993 I drew attention to a passage of
some 250 lines in The Rape of Lucrece and a passage of some 200 lines in
2 Henry VI, each of which has multiple connections with the 150 lines of
55. See the Oxford English Dictionary, Compact Edition (Oxford: Clarendon Press, 1971): sensible, A. adj. II.7–9.
34
Determining the Shakespeare Canon
the Quarrel Scene in Arden of Faversham: they share several images that
are rare outside the Shakespeare canon, other images that, although less
closely alike, quarry the same areas of experience in similar ways, and one
or two unusual words.56 Tables 1.4 and 1.5 reproduce the relevant data for
each passage. Nobody doubts that The Rape of Lucrece, first published in
1594, was composed in 1593–4, and is thus of later composition than Arden
of Faversham, which reached print in 1592. Moreover, as a narrative poem
Lucrece provides evidence supplementary to that in Appendix 1’s inventory
of Literature Online links with plays. Notes keyed to that inventory record
eleven links between Arden of Faversham and The Rape of Lucrece in rare
phrases and collocations, so the poem of 1855 lines yields results comparable to those for the plays with the largest numbers of links to the Quarrel
Scene. But Table 1.4 reinforces the association. One can say with some
confidence that no non-Shakespearean passage of 250 lines in any work
of literature in English shares with the Quarrel Scene so many images,
image-subjects, and key items of vocabulary as does The Rape of Lucrece,
829–1078. In fact a search of the entire Literature Online database (poetry,
drama, and prose) finds no other complete work that contains even the
three words ‘copesmate’, ‘rifled’, and ‘cannon’.57 In the Lucrece passage the
heroine is confronting her violator; in the Arden passage the heroine is
confronting her seducer.
The multiple links between the Quarrel Scene and 2 Henry VI, displayed
in Table 1.5, are less striking but nevertheless remarkable. The shared
images and ideas include: the destruction by inclement weather of blossoms
that are still in the bud or have just emerged; the dire consequences of not
weeding a garden; spring time; supplanting; stage acting (with reference
to murder); the dangers of a serpent or snake; a contrast between raven
(evil) and dove (good); and the muddiness or clarity of a spring or fountain
as a figure for contrasting emotional or moral states. Embedded in their
imagery both passages have the words ‘blossoms’, ‘weeds’, ‘supplant(ed)’,
‘raven’, ‘dove’, and ‘clear’. Literature Online finds no other play, let alone
56. Jackson, ‘Shakespearean Features’, 290–2, 297–9.
57.A search of Early English Books Online (EEBO)—eebo.chadwyck.com—finds all three
words in Mateo Alemán, The Rogue: or The Life of Guzman de Alfarche, trans. ‘from the
Spanish’ (1623); Marin Le Roy, sieur de Gomberville, The History of Polexander, trans.
William Browne (1647); François Rabelais, Gargantua and Pantagruel, trans. Thomas
Urquhart (1693–4); and Guy Miege, A New Dictionary of French and English (1677). These are
all very large works. When ‘slanderous’ is added to the other three words, only the dictionary joins Arden, 1.150, and Lucrece, 829–1078, in containing all four.
Shakespeare and the Quarrel Scene in
arden of faversham 35
Table 1.4 Multiple links between Arden Quarrel Scene and Lucrece
Arden of Faversham, 8.1–150
The Rape of Lucrece, 829–1078
And nips me as the bitter
northeast wind
Doth check the tender blossoms
in the spring (5–6)
Unruly blasts wait on the
tender spring (869)
Continual trouble of my
moody brain
Feebles my body . . .
And nips me . . .
My golden time was when I had
no gold (3–5, 11)
The aged man that coffers up
his gold
Is plagu’d with cramps and
gouts and painful fits (855–6)
To make my harvest nothing
but pure corn (25)
And useless barns the harvest
of his wits (859)
And for his pains I’ll hive
him up awhile
And, after, smother him to
have his wax;
Such bees as Greene must
never live to sting (26–8)
My honey lost, and I, a drone like bee,
. . . robb’d and ransack’d . . .
In thy weak hive a wand’ring
wasp hath crept (836–9)
’tis fearful sleeping in a
serpent’s bed (42)
The adder hisses where the
sweet birds sing (871)
like to a cannon’s burst
Discharged against a
ruinated wall (51–2)
To ruinate proud buildings (944)
discharged cannon (1043)
not Arden’s honest wife.
Ha, Mosby, ’tis thou hast
rifled me of that
And made me sland’rous to
all my kin (73–5)
sland’rous (1001)
. . . I was a loyal wife:
So am I now—O no, that
cannot be,
Of that true type hath
Tarquin rifled me (1048–50)
Even in my forehead is thy
name engraven (76)
Reproach is stamp’d in
Collatinus’ face (829)
That showed my heart a raven
for a dove (97)
raven’s wings (949)
The crow may bathe his coal-black
wings in mire,
And unperceiv’d fly with the filth
away
(Continued)
36
Determining the Shakespeare Canon
Table 1.4 (Continued)
Arden of Faversham, 8.1–150
The Rape of Lucrece, 829–1078
But if the like the snow-white swan
desire,
The stain upon his silver down
will stay (1009–12)
copesmate (104)
copesmate (925)
Thou hast been sighted as the
eagle is (126)
eagles gaz’d upon with every
eye (1015)
And I deserve not Mosby’s
muddy looks.
A fount once troubled is not
thickened still;
Be clear again, I’ll ne’er
more trouble thee (132–4)
Or toads infect fair founts
with venom mud (850)
As from a mountain spring
that feeds a dale,
Shall gush pure streams to purge
my impure tale (1077–8)
Flowers do sometimes spring
in fallow lands,
Weeds in gardens, roses grow
on thorns (142–3)
Unwholesome weeds take root
with precious flow’rs (870)
Table 1.5 Multiple links between Arden Quarrel Scene and 2 Henry VI
Arden of Faversham, 8.1–150
2 Henry VI, 3.1.31–229
And nips me as the bitter
northeast wind
Doth check the tender blossoms
in the spring (5–6)
Thus are my blossoms blasted
in the bud (89)
For Greene doth ear the land
and weed thee up
To make my harvest nothing
but pure corn (24–5)
You have supplanted Arden
for my sake
And will extirpen me to plant
another (40–1)
I pray thee, Mosby, let our
springtime wither;
Our harvest else will yield but
loathsome weeds (66–7)
Now ’tis the spring, and weeds
are shallow-rooted;
Suffer them now, and they’ll
o’ergrow the garden,
And choke the herbs for want of
husbandry.
. . . if better reasons can
supplant (31–7)
(Continued)
Shakespeare and the Quarrel Scene in
arden of faversham 37
Table 1.5 (Continued)
Arden of Faversham, 8.1–150
2 Henry VI, 3.1.31–229
Then is there Michael and
the painter too,
Chief actors to Arden’s
overthrow,
Who, when they shall see
me sit in Arden’s seat,
They will insult upon me
for my meed
Or fright me by detecting of
his end (29–33)
I know their complot is to have
my life,
And if my death might make this
island happy,
And prove the period of their
tyranny,
I would expend it with all
willingness.
But mine is made the prologue
to their play;
For thousands more, that yet
suspect no peril,
Will not conclude their plotted
tragedy (147–53)
’Tis fearful sleeping in a
serpent’s bed (42)
Or as the snake roll’d in a
flow’ring bank,
With shining checker’d slough,
doth sting a child (228–9)
That showed my heart a raven
for a dove (97)
Seems he a dove? his feathers
are but borrow’d,
For he’s disposed as the
hateful raven (75–6)
A fount once troubled is not
thickened still;
Be clear again, I’ll ne’er
more trouble thee (133–4)
The purest spring is not so
free from mud
As I am clear from treason to
my sovereign (101–2)
short passage, that contains all these words in these forms.58 2 Henry VI,
dated 1591 by the Oxford editors, was probably written after Arden of
Faversham, and The Rape of Lucrece undoubtedly was. If Shakespeare did not
write the Quarrel Scene, it infected him, and him alone among dramatists
of the day, like a contagious disease.
58. Since LION does not recognize the Arden quarto’s spelling ‘blosoms’, one must enter it
specifically, rather than rely on the ‘Variant spellings’ function. Also, Arden has the misprint
‘dowe’ for ‘dove’. Shakespeare’s The Tempest has, widely scattered, ‘blossom’, ‘raven’s’, ‘supplant’, ‘weeds’, ‘dove-drawn’, and ‘clear’, and with Arden’s ‘loathsome weeds’ (8.67) compare
The Tempest’s ‘weeds so loathly’ (4.1.21), in each case in connection with an illicit sexual
union. Two nineteenth-century plays have all the words in variant forms. In Otway’s The
Orphan (1680), which LION searches find, the ‘weeds’ are garments, not plants.
38
Determining the Shakespeare Canon
VI
Keith Sturgess was surely right. The Quarrel Scene in Arden of Faversham
must be Shakespeare’s. But there is solid evidence that this was not the only
scene that he contributed to the play. Chapter 2 summarizes independent
research that finds Shakespeare responsible for the bulk of the writing in
scenes 4–9, the middle section of the play, and the technique of Literature
Online searching confirms that Arden’s account of his dream in scene 6 is no
less Shakespearean than the Quarrel Scene. W. H. Clemen long ago observed
that in Shakespeare’s early histories and in Titus Andronicus ‘it is above all
the monologue that breeds imagery’ and that this imagery tends to derive
from ‘gardening or the processes of blossoming, ripening, etc.’.59 Others have
noted the prevalence in these plays of imagery of ‘birds and beasts of prey
and their victims’, of ‘the jungle, the chase and the slaughter-house’.60 Such
images dominate several of the longer speeches in Arden of Faversham. It is
most unlikely that Shakespeare wrote the whole play, as Swinburne supposed, but his contribution to it appears to have been quite substantial.
The omission of Arden of Faversham from the First Folio argues against
Shakespeare’s sole authorship but not against his part-authorship—witness the
collaborative Pericles, The Two Noble Kinsmen, and (almost certainly) Edward
III.61 The same may be said of the findings of Ward E. Y. Elliott and Robert
J. Valenza and their Claremont Shakespeare Clinic. Arden lies outside the
‘Shakespearean range’ on ten among the battery of computerized tests that they
devised to differentiate between Shakespeare and non-Shakespeare plays.62
59. Wolfgang H. Clemen, The Development of Shakespeare’s Imagery (London: Methuen, 1951),
42, 79; see also Spurgeon, Shakespeare’s Imagery, 215–24.
6 0. The Second Part of King Henry the Sixth, ed. Andrew S. Cairncross (London: Methuen, 1957;
Arden Shakespeare), liii–liv; see also Spurgeon, Shakespeare’s Imagery, 227–30.
61. Even the Folio’s testimony against Shakespeare’s sole authorship may not in itself be absolutely decisive. Copyright difficulties might have prevented the play’s inclusion: copyright
rested with Edward White’s widow from the time of his death in 1613 till 29 June 1624,
when, according to the Stationers’ Register, Edward Allde inherited the play (Wine, Arden,
xxi). The Folio syndicate may have been unable to negotiate terms with Mrs White, as they
did for the handful of Folio plays to which other publishers held the title. It is also possible
that John Heminges and Henry Condell, who, as Shakespeare’s fellow actor-sharers in the
King’s Men, collected plays for the First Folio, were less than fully conversant with the facts
about what Shakespeare had written before the Chamberlain’s (later the King’s) company
was formed in 1594.
62. The fullest account of their research, and one that makes adjustments to data presented in earlier articles, is ‘Oxford by the Numbers: What Are the Odds That the Earl of Oxford Could
Have Written Shakespeare’s Poems and Plays?’, Tennessee Law Review, 72 (2004), 323–453.
See especially the summary at 402. Arden’s failure in one test—rates of use of hyphenated
Shakespeare and the Quarrel Scene in
arden of faversham 39
No play of Shakespeare’s undisputed sole authorship had more than two such
‘rejections’. But Arden’s tally is lower than 1 Henry VI’s eleven. Since 1 Henry
VI is undoubtedly at least partly by Shakespeare, Elliott and Valenza’s tests
of whole plays do not rule out the possibility that Arden of Faversham is too,
and there is not much difference, in number of rejections, between Arden and
3 Henry VI, which collected eight, but which, though probably not wholly
Shakespeare’s, is mainly so.
In 1986 the editors of the Oxford Shakespeare: The Complete Works provisionally presented Titus Andronicus and 2 and 3 Henry VI as Shakespeare’s
alone. While acknowledging that, before a mid-twentieth-century reaction to ‘disintegration’ set in, Shakespeare’s sole authorship of each of those
three plays had been widely doubted, they left the question of possible collaboration open.63 In the second edition of The Complete Works (2005), Titus
Andronicus was attributed to Shakespeare ‘with George Peele’, while the
attributions of 2 and 3 Henry VI were unchanged. But since then Hugh
Craig, Arthur F. Kinney, and John Burrows, using several forms of ‘computational stylistics’, have given compelling reasons for adding 2 and 3 Henry
VI to 1 Henry VI as works in which Shakespeare’s is not the only hand.64
Even if we were to ignore all plays in which Shakespeare may have had
collaborators, the Literature Online findings would still overwhelmingly
support his authorship of the Quarrel Scene. But in fact it will be shown in
Chapter 3 that if the Craig–Kinney–Burrows allocations are even approximately right, analysis of the distribution of Arden’s LION links to the putative collaborations immeasurably strengthens the case.
compounds—is due to faulty data. See Jackson, Defining Shakespeare, 77, n. 74. Privately
Elliott informed me (17 October 2005) that he later tested the Quarrel Scene by his methods.
Thirteen tests were found suitable for scenes of this length. The Quarrel Scene passed all but
two. On this measure it is thus comparable to the four ‘Shakespeare’ scenes of Edward III—1.2,
2.1, 2.2, and 4.4 (the three scenes, 2, 3, and 12 in the second edition of the Oxford Complete
Works)—which each failed only one or (in the case of 4.4) two tests. See Ward E. Y. Elliott
and Robert J. Valenza, ‘Two Tough Nuts to Crack: Did Shakespeare Write the “Shakespeare”
Portions of Sir Thomas More and Edward III? Part I and Part II: Conclusion’, Literary and
Linguistic Computing, 25 (2010), 67–83 and 165–77. The burden of his 2005 email about the
Quarrel Scene is repeated in Ward Elliott, ‘Notes from the Claremont Shakespeare Clinic’,
Shakespeare Newsletter, 61 (2011/12), 105, 108–12, at 110, where scenes 4–7 of Arden are reported
as passing the Claremont tests for Shakespeare. See also Chapter 3, n. 65.
63. Textual Companion, 111–12.
64. Hugh Craig and Arthur F. Kinney, eds., Shakespeare, Computers, and the Mystery of Authorship
(Cambridge: CUP, 2009), 40–77, hereafter ‘Craig and Kinney’; Hugh Craig and John Burrows,
‘A Collaboration about a Collaboration: The Authorship of King Henry VI, Part Three’,
in Collaborative Research in the Digital Humanities, ed. Marilyn Deegan and Willard McCarty
(Farnham: Ashgate, 2012), 27–65. Craig and Kinney detected no sure signs of a second hand in
3 Henry VI, but the further investigation by Craig and Burrows did.The Craig–Kinney book is
the main subject of Chapter 2 below, and both studies are discussed in Chapter 3.
2
Reviewing Authorship
Studies of Shakespeare and his
Contemporaries, and the Case
of Arden of Faversham
I
A recent review essay by Brian Vickers on ‘Shakespeare and Authorship
Studies in the Twenty-First Century’ is a remarkably thorough and
well-documented survey of the field, describing traditional ‘reading-based’
approaches to problems of attribution before providing valuable guidance
to ‘linguistic-processing’ or ‘computer-assisted’ methods, besides placing
both kinds within their historical context.1 Vickers raises various theoretical issues, and points the reader to major articles, such as Patrick Juola’s
excellent ‘Authorship Attribution’, where further description and discussion can be found.2
Vickers’s essay is also a review of two books, the first by Warren Stevenson,
in which he argues that Shakespeare wrote the Additions printed in the
fourth edition (1602) of Kyd’s The Spanish Tragedy, and the second entitled
Shakespeare, Computers, and the Mystery of Authorship, edited by Hugh Craig
and Arthur F. Kinney, in which the same ascription is upheld, along with
1.Brian Vickers, ‘Shakespeare and Authorship Studies in the Twenty-First Century’,
Shakespeare Quarterly, 62 (2011), 106–42. In the present chapter simple page references to this
article are incorporated in the text.
2.Patrick Juola, ‘Authorship Attribution’, Foundations and Trends in Information Retrieval, 1
(2008), 233–334.
Authorship Studies of Shakespeare and his Contemporaries 41
many others.3 Vickers judges that Stevenson’s monograph, ‘an example of
traditional authorship studies’ (106) and largely reliant on the citation of
parallel passages, makes out a convincing case, but he is severely critical of
the Craig–Kinney methodology and findings. The present chapter argues
for a very much more positive assessment, largely through highlighting certain details of Craig and Kinney’s work that seem to me crucial but are not
mentioned in Vickers’s account. Results in authorship studies, like symptoms presented for medical diagnosis, may be difficult to interpret, so that
a ‘second opinion’ can be of value.4 Craig and Kinney devote a chapter to
Arden of Faversham. Does their research support the conclusion of Chapter 1
in the present book—that Shakespeare was a co-author of the play?
Craig is Director of the Centre for Literary and Linguistic Computing
at the University of Newcastle, Australia, where Kinney, Director of the
Massachusetts Center for Renaissance Studies in Amherst, spent a period
of research. Their book is the outcome of a collaborative enterprise, drawing on a team of co-workers from both institutions. Craig and Kinney fully
describe their methods, which employ two kinds of data: (a) lexical words
used more frequently by Shakespeare than by other playwrights and vice
versa, and (b) function words—‘and’, ‘of ’, ‘the’, and the like—subjected
to principal component analysis. The results are displayed in scatter-plots.
The figures for lexical words, counted as proportions of the total number of word-tokens within a given text, are plotted on two axes—for
words used more often and less often by Shakespeare than by others. The
function-word results are likewise plotted in two-dimensional space, as
first and second principal components, the first component (on the horizontal axis) accounting for the greatest amount of variance in the data, the
second (on the vertical axis) for the second greatest amount, the two components being uncorrelated.
Vickers rightly notes that the mechanical counting of ‘lexical words’
(those bearing semantic content) cannot measure the ways in which they
combine into grammatical and syntactical units, and so neglects a highly
3. Warren Stevenson, Shakespeare’s Additions to Thomas Kyd’s ‘The Spanish Tragedy’: A Fresh Look
at the Evidence Regarding the 1602 Additions (Lewiston, NY; Queenstown, Ont.; Lampeter,
Wales: Edwin Mellen Press, 2008); Craig and Kinney, though the chapter on Edmond Ironside
is by Philip Palmer and the chapters on Edward III and Hand D of Sir Thomas More are by
Timothy Irish Watt. Gary Taylor, reviewing these two books in Medieval and Renaissance
Drama in English, 34 (2010), 198–201, evaluates them very differently from Vickers.
4.In ‘A Second Opinion on “Shakespeare and Authorship Studies in the Twenty-First
Century” ’, Shakespeare Quarterly, 63 (2012), 355–92, John Burrows offers a detailed rebuttal
of Vickers’s strictures on Craig and Kinney’s book.
42
Determining the Shakespeare Canon
significant aspect of style. Craig and Kinney suggest that the analysis of
function words goes some way towards repairing this deficiency, since these
tend to have ‘grammatical force rather than meaning’ and ‘perform a wide
range of syntactic functions’.5 In Vickers’s view, to make this second claim
‘is to confuse syntax with grammar’ (117). But many of the high-frequency
words tested by the Craig–Kinney team and listed in an appendix can
indeed serve as syntactic markers.6 Consider two differently constructed
sentences that convey the same information. Here is the first: ‘As soon as we
guests had finished dinner, we said goodbye to our kind hosts and drove to
the theatre, where we saw a performance of Twelfth Night, which we greatly
enjoyed.’ And here is the second: ‘Straight after dinner, we guests, saying
goodbye to our kind hosts, drove to the theatre and saw a most enjoyable performance of Twelfth Night.’ The two sentences each contain two
examples of ‘to’ and one of ‘a’, ‘and’, ‘of ’, ‘our’, and ‘the’. But the first has
three more instances of ‘we’ than the second, and also contains ‘as’ (twice),
‘had’, ‘where’, and ‘which’, none of which are found in the second sentence,
which has instances of ‘after’ and ‘most’, both absent from the first. The two
types of sentence construction entail the use of different function words.
The first sentence uses a relative clause, introduced by ‘which’, whereas
the second does not. The first sentence uses the conjunction ‘and’ to link
co-ordinate clauses ‘we said . . . and drove’, whereas the second modifies ‘we
guests’ by using the present participle ‘saying’.
Vickers observes that the technique of principal component analysis (briefly explained in Craig and Kinney’s glossary of statistical terms)7
fails to account for all the variance between texts in their function-word
rates and that, being a data reduction method, it ‘cannot also be a suitable
method of classification’ (119). He quotes the words of Craig and Kinney’s
‘scholarly progenitor, John Burrows, who explicitly warned that “PCA
is not intrinsically a test of authorship but only of comparative resemblance” ’.8 Yet classification is itself based on degrees of comparative resemblance—on similarities and differences. If principal component analysis of
the function-word rates of twenty texts by Author A and twenty texts by
Author B reveals A’s texts to be sufficiently like one another and unlike B’s
5. Craig and Kinney, 20 and 12.
6. Craig and Kinney, 221–2.
7. Craig and Kinney, 225.
8.Vickers, ‘Authorship Studies’, 119. John Burrows, ‘Questions of Authorship: Attribution
and Beyond’, Computers and the Humanities, 37 (2003), 8.
Authorship Studies of Shakespeare and his Contemporaries 43
to separate the two authorial groups, then the two sets of texts, A’s and B’s,
have been correctly classified by this method.
Whatever the theoretical limitations of lexical-word tests and of PCA,
Craig and Kinney show that their versions of them work well in practice.
Vickers never mentions that their lexical-word test of 1,287 two-thousandword segments from plays of undisputed authorship correctly classifies
over 98 per cent as Shakespearean or non-Shakespearean: ‘over 98 per cent’
applies not only to the 1,287 total but also to both the 1,009 non-Shakespeare
segments and the 278 Shakespeare segments separately.9 And when
Coriolanus and Thomas Middleton’s Hengist are tested in this way, neither
having been used to draw up the lists of Shakespeare and non-Shakespeare
words, the thirteen Coriolanus segments fall, on a scatter-plot, within
Shakespeare territory, while nine of the ten Hengist segments are placed in
unambiguously non-Shakespeare territory and one appears within a region
shared by Shakespeare and non-Shakespeare segments.10 Vickers gives an
inaccurate report of the Coriolanus results, claiming that ‘nearly 30 per cent’
of seventeen Coriolanus segments were misclassified (120), because he confuses the figures for ‘Shakespeare versus non-Shakespeare’ with the figures for Coriolanus. He gives a similarly misleading account of the Hengist
results (120). He writes as though only absolute separation between all 278
Shakespearean and all 1,287 non-Shakespearean segments could produce
attributions of any value, which is not the case. It is true that not only does
one Hengist segment fall into territory shared by Shakespeare and Middleton
segments, but also one of the 278 Shakespeare segments is placed nearly as
far within non-Shakespeare territory as one of the ten Hengist segments.
But which authorial grouping—Shakespeare or non-Shakespeare—nine
of the ten Hengist segments belong to is perfectly clear.
Moreover principal component analysis of function words, applied only
to scenes of at least 1,500 words, confirms Vickers’s own allocations of
authorship—given in his invaluable Shakespeare, Co-Author—in the collaborative Titus Andronicus, Timon of Athens, Henry VIII (All Is True), and
The Two Noble Kinsmen: all the eighteen scenes long enough to qualify for
testing are correctly classified as by Shakespeare or his agreed collaborator,11 and yet Vickers says nothing about the Craig–Kinney results for the
9. Craig and Kinney, 15–23.
10. Craig and Kinney, 23–4.
11. Craig and Kinney, figures 2.4–2.7. Vickers, Shakespeare, Co-Author.
44
Determining the Shakespeare Canon
last three of those four collaborations. Also, lexical words confirm the identification, which again Vickers has supported, of Sir Thomas More Hand
D with Shakespeare. They also confirm the extreme improbability that
Shakespeare was concerned with Edmond Ironside, while disposing of several other candidates who might be put forward.12
Vickers acknowledges the existence of chapters on More and Ironside in
a footnote, where the chapter on Edward III is also mentioned, but these
contributions to the volume are summarily dismissed as offering ‘no new
attributions’.13 Yet in reaching verdicts that Vickers must surely consider
correct, do they not go some way towards vindicating the team’s methodology? In the analysis of two contrasting blocks of Edward III, using
6,000-word segments, the overall pattern of results from lexical-word tests
and function-word tests gives good discrimination between 1.2–2.2 (generally credited to Shakespeare) and 3.3–4.3 (generally not), and classifies
them according to expectations, once the analysis has been extended and
refined. Control lexical-word trials on King John, 1 Henry IV, Henry V, James
IV (Greene), Edward I (Peele), and Edward II (Marlowe) correctly classify all
twenty 6,000-word history-play segments. Figures 6.11 and 6.12, graphing
a statistically valid alternative means of processing the lexical-words data,
show that Shakespeare is the best of sixteen playwrights as a candidate for
the authorship of 1.2–2.2, but only the tenth best of the sixteen as a candidate
for the authorship of 3.1–4.3. It would be possible to quibble that figure 6.10,
which maps the results for lexical-word tests of 6,000-word segments in
pre-1600 plays, situates Edward III, 3.1–4.3, between the Shakespeare and
non-Shakespeare clusters. But this would be to highlight one ambiguous
result at the expense of the overall pattern of results for the two Edward
III blocks. Mathematically, there must inevitably be a slight tendency for
newly tested texts to be less securely placed within their correct authorial
groups than the texts employed to form those groups. This is because of
the ‘regression effect’, whereby discriminators are apt to lose some of their
original potency when applied to a new set of data. It was explained in a
classic study of the Federalist papers by statisticians Frederick Mosteller and
David L. Wallace.14
12. Craig and Kinney, 134–1 for More, 100–5 for Edmond Ironside.
13. Vickers, ‘Authorship Studies’, 121, n. 42. For Edward III, see Craig and Kinney, 116–33.
14. Frederick Mosteller and David L. Wallace, Inference and Disputed Authorship: The Federalist
(Reading, MA: Addison-Wesley, 1964), 200.
Authorship Studies of Shakespeare and his Contemporaries 45
In minimizing the Titus results Vickers exaggerates the significance of
the slight overlap between Peele and Shakespeare segments, ignoring the
telling distinction between where Peele’s Act 1 falls on the scatter-plot
(figure 2.4) and where the four other Titus segments fall. He complains
that ‘the opening scene of the play, represented by a grey triangle, is placed
simultaneously over a hollow circle, representing Peele segments, and a
shadowy diamond representing Shakespeare’ (120–1). This formulation
withholds the fact that the scatter-plot pinpoints the position of twenty-five
Peele segments and three hundred Shakespeare segments, and that the diamond representing one of these three hundred Shakespeare segments and
impinging on the black circle representing Titus Act 1 is the only one placed
as far (or, strictly speaking, almost as far) in Peele territory as the Titus Act 1
black circle.
Concentrating on Titus Andronicus, where the Craig–Kinney evidence,
though clearly supportive of Peele’s authorship of Act 1, is less decisive
than for other collaborations, Vickers ignores figures 2.5, 2.6, and 2.7, in
which the separation between Middleton and Shakespeare (figure 2.5) and
Fletcher and Shakespeare (figures 2.6 and 2.7) is near-perfect and the correct classification of all thirteen scenes unambiguous.
Vickers does not acknowledge Craig’s fair summarizing of the strengths
and limitations of his team’s tests and his remarks about the ‘right balance of
confidence and caution in interpreting the results’, nor his important point
that ‘the fallibility of the methods individually can be reduced’ by the use
on any particular case of both the lexical-word tests and the function-word
tests.15
Vickers professes that Craig’s analysis of the Additions to The Spanish
Tragedy yields a ‘disappointing result’ and that Craig is forced to ‘admit
defeat’, claiming that his ‘inability to make a confident attribution using
his own methodology shows the limitations of computational stylistics’
(126). Others will recognize that, for all Craig’s admirable candour and
caution, the case he makes out is strong, and that his methods (even without
Stevenson’s parallel-passages, which Vickers finds persuasive) demonstrate
that Shakespeare is much more likely to be the author of the Additions than
Webster, Dekker, or Jonson—the only other candidates who have seriously
been proposed. Vickers seems reluctant to concede that all authorship studies of internal evidence have their ‘limitations’ and simply make particular
15. Craig and Kinney, 38–9.
46
Determining the Shakespeare Canon
ascriptions more or less probable: no one type of testing can create absolute
certainty.
He highlights the inability of Craig’s methods to identify Nashe as author
of Henry VI, Act 1 (121–3). Yet Craig himself shows quite as full an appre
ciation as Vickers does of the fact that Nashe’s prose works and one single quasi-allegorical play afford no adequate basis for comparisons. Despite
these limitations the results of lexical-word tests shown in figures 3.6 and
3.7 are suggestive of Nashe’s possible authorship of Act 1, which is in both
figures closer than any of the other Acts to the Summer’s Last Will and
Testament segments. And Vickers avoids noting that the function-word test
points clearly to the mixed authorship of 1 Henry VI (figure 3.2), in which
he himself believes, or admitting how well both the function-word and
lexical-word tests separate the portions of the play generally accepted as
Shakespeare’s from the rest (table 3.1 and figure 3.5). The same lexical-word
test unambiguously dissociates all 2,000-word segments of 1 Henry VI from
Kyd, to whom Vickers attributes much of the play, and associates them with
plays by ‘others’ (figure 3.8).
Craig’s work on the three parts of Henry VI has been more thorough and
sophisticated than Vickers’s strictures suggest (124–6). Craig’s conclusion
that 2 Henry VI, like 1 Henry VI, is deviant from ‘Shakespeare norms’, for
example, is based on counts of both lexical words and function words and a
statistical analysis that places the combined results for 2,000-word segments
within one of four quadrants on a graph, the upper right quadrant containing the vast majority of fifty-eight segments from six early Shakespeare plays
(including Richard III, Richard II, and King John) and the lower left quadrant
containing only one (figure 3.1). As Craig explains, ‘To make sure they were
being treated exactly like the Henry VI segments, we withdrew all the segments of each play in turn from the process of finding Shakespeare markers’—a prudent precaution.16 Both figure 3.2, for 1 Henry VI, and figure 3.3,
for 2 Henry VI, show the majority of segments not falling into the upper
right quadrant and several falling into the lower left quadrant.
One of Vickers’s criticisms is, however, perfectly justified. Any findings
concerning Marlowe—and particularly Craig’s identification of his hand in
some scenes of 1 and 2 Henry VI—must be tentative, since, as Vickers justly
complains, Doctor Faustus ‘was undoubtedly coauthored’ (124) and (as he forbears mentioning) The Jew of Malta also contains writing by at least one other
16. Craig and Kinney, 45, n. 17.
Authorship Studies of Shakespeare and his Contemporaries 47
playwright, while the text of The Massacre at Paris is bad.17 Deducing Marlowe’s
lexical-word and function-word characteristics from all six Marlowe plays
(these three as well as 1 and 2 Tamburlaine and Edward II) is problematical.
However, Craig ends his chapter on the three parts of Henry VI by merely
proposing that ‘the evidence is strong enough to warrant Shakespeareans giving a hearing to the possibility that we should look to Marlowe rather than to
Shakespeare as the source of many of the Joan la Pucelle and Jack Cade scenes’.18
This—‘strong enough to warrant . . . giving a hearing to the possibility’—is
scarcely an extravagant claim. But it deserves respect.
Vickers is especially scathing about the chapter on Arden of Faversham,
a play that he has attributed entirely to Kyd on evidence that, it seems to
me, not only fails to support the ascription but tells strongly against it.19
Vickers accuses Kinney of failing ‘to disclose the remarkable fact that’ the
case for Kyd ‘was made over half a century by five scholars working independently’ (127). This charge against Kinney is groundless. Kinney lists,
with dates, the very five scholars (Crawford, Miksch, Sykes, Rubow, and
Carrère) whom Vickers lists, and adds the names of several other believers in Kyd’s claims.20 Vickers himself ‘fails to disclose’ that Kyd’s editor
Frederick S. Boas rejected the case for Kyd as utterly unconvincing, and
so did Kyd specialist Arthur Freeman in his biographical and critical study
Thomas Kyd: Facts and Problems.21 Nor does he note that the Revels editor
of Arden, M. L. Wine, judged that the ‘evidence’ for Kyd (Wine’s quotation
marks) ‘fails to persuade’.22 Vickers’s five scholars were doubtless ‘working
independently’ in the sense that they did not collaborate, but they did not
17.David J. Lake, ‘Three Seventeenth-Century Revisions: Thomas of Woodstock, The Jew of
Malta, and Faustus B’, Notes and Queries, 228 (1983), 133–43; Maguire, Shakespearean Suspect
Texts, 279–81.
18. Craig and Kinney, 77.
19.Vickers, ‘Thomas Kyd, Secret Sharer’; MacDonald P. Jackson, ‘New Research on the
Dramatic Canon of Thomas Kyd’, Research Opportunities in Medieval and Renaissance Drama,
47 (2008), 107–27. Vickers’s research fails to support his attributions to Kyd because 2 Henry
VI and The Taming of the Shrew each share more ‘unique trigrams’ with Arden of Faversham,
for example, than do any of the three Kyd plays, and in quality the Kyd parallels are not
superior. The argument that Vickers’s own evidence actually tells against his attributions is
more complex, but rests on the fact that larger proportions of the trigrams shared between
canonical Kyd plays, than of trigrams shared between canonical and newly attributed plays,
are unique; and the difference in the two sets of proportions is statistically significant. See
also Chapter 5, Section II.
20. Craig and Kinney, 84.
21. The Works of Thomas Kyd, ed. Frederick S. Boas (Oxford: Clarendon Press, 1901), lxxxix;
Arthur Freeman, Thomas Kyd: Facts and Problems (Oxford: Clarendon Press, 1967), 179–80.
22. Wine, Arden, lxxxvii.
48
Determining the Shakespeare Canon
all reach their views ‘independently’: a poorly argued ascription was ‘reiterated with some frequency’, as Freeman noted,23 and Sykes and Carrère were
influenced by Crawford.
Vickers claims that Kinney’s application of computational stylistics ‘yields
no useful results’ on Arden (127). No results useful to a case for Kyd’s authorship, certainly, but results that help confirm that Shakespeare contributed
to the play. After giving a good account of the play’s qualities, Kinney
reports the team’s findings. He first presents the results for the lexical words
(figure 4.1 and table 4.1). This maps 2,000-word segments by Shakespeare
and 2,000-word segments by other dramatists, along with the eighteen scenes
and epilogue from Arden. The scenes range from as few as 118 words to as
many as 5,230 and only two have more than 2,000 words, so their positions
on the map are avowedly unreliable, the aim being merely exploratory—to
see whether there is any warrant for supposing some section or sections of
the play may be Shakespearean. Six scenes fall on the Shakespeare side and
twelve on the non-Shakespeare side of a line drawn by bisecting the line joining the centroids of the Shakespeare and non-Shakespeare clusters. Vickers
points out that this leaves only 17.2 per cent of the play in Shakespeare territory and that the two longest scenes, 1 and 14, fall on the non-Shakespearean
side. He judges that other scenes are too short to be amenable to valid testing. He also notes that scene 8, which I had given reasons—set forth again in
Chapter 1 of the present volume—for attributing to Shakespeare, falls into
non-Shakespearean territory ‘on the non-Shakespeare side [of the bisector line], but at the lower end of the [non-Shakespeare] cluster’, in Kinney’s
words, the ‘lower end’ being that nearest to the Shakespeare cluster.24
Vickers overlooks, however, the fact that the scenes falling on the
Shakespeare side of the bisector line are 4, 5, 6, 7, 9, and 16. The last is easily
the shortest of these, consisting of a mere 179 words (or twenty-one lines),
and so its placement may reasonably be regarded as accidental. But (a) scenes
4–7 are consecutive, forming a block of 1,965 words, and (b) although Craig
and Kinney do not mention this, scenes 4–9 compose Act 3 in older editions
that divide the play into Acts and scenes, so that (c) five of the six scenes of
Act 3, amounting to 3,251 words, fall on figure 4.1 within Shakespeare territory. A division of labour by Acts, whether precisely or approximately, was
common among collaborating early modern dramatists.25 Scene 8, though
23. Freeman, Thomas Kyd, 179.
24.MacDonald P. Jackson, ‘Shakespeare and the Quarrel Scene in Arden of Faversham’,
Shakespeare Quarterly, 57 (2006), 249–93; Craig and Kinney, 94.
25. Vickers, Shakespeare, Co-Author, 27–34.
Authorship Studies of Shakespeare and his Contemporaries 49
falling on the non-Shakespeare side of the bisector line is not very far from
it, given that with 1,312 words it is appreciably smaller than the 2,000-word
segments used to discriminate between Shakespeare and non-Shakespeare
and so more subject to random variation. We might expect some genuinely
Shakespearean scenes of considerably fewer than 2,000 words to stray further into non-Shakespeare territory than any of the genuinely Shakespearean
2,000-word segments. Of course the reverse also applies: we might expect
some non-Shakespearean scenes of appreciably fewer than 2,000 words
to stray further into Shakespearean territory than any of the genuinely
non-Shakespearean 2,000-word segments. But the fact that five out of six
scenes in Act 3 test positive whereas only one (and that extremely short) out
of the remaining thirteen scenes (including the epilogue) tests positive can
hardly be coincidental. Fisher’s Exact Probability Test yields a probability
of lower than one in 100 of such a difference occurring by chance.26 If the
identification of some scenes in Arden as Shakespearean were fortuitious and
without significance, we would expect those scenes to be randomly distributed, not clustered in this way.
Further, scene 6 has (proportionally to the total number of words)
more of the ‘Shakespeare’ lexical words than any single one of the 1,067
two-thousand-word non-Shakespeare segments, while scene 5 has fewer
of the ‘non-Shakespeare’ words than all but one, or possibly two, of the
non-Shakespeare segments. These results for consecutive scenes are, at the
very least, highly suggestive.
The Craig–Kinney team, guided by the results of figure 4.1, then joined
scenes together so as to divide the play into larger sections made of contiguous scenes: scenes 1–3 (a large block of 7,892 words), scenes 4–7 (1,965
words), scenes 8–9 (2,598 words), and scenes 10–end (7,396 words). The
4–7 segment was placed well within Shakespeare territory, the other three
segments within non-Shakespeare territory, 8–9 being the closest of these
to the border. Vickers is silent about these important results, presented
in figure 4.2, giving the impression that Kinney moves straight from the
lexical-word analysis of mere scenes to the step described in my next paragraph. It is true that even when scene 8 is joined to scene 9 to create a block
of more than 2,000 words it remains placed on the non-Shakespeare side of
the bisector line, but some misclassified Shakespeare segments are close by.
26.Fisher’s test is described in most introductory books, such as Russell Langley, Practical
Statistics Simply Explained (New York: Dover Publications, 1971). I have used http://
research.microsoft.com/en-us/um/redmond/projects/MSCompBio/FisherExactTest/.
50
Determining the Shakespeare Canon
Moreover, the lexical test was one of two, the second producing unequivocally positive results for an attribution of scenes 8–9 to Shakespeare.
A function-word test in which the four Arden blocks were plotted
against twenty-seven Shakespeare plays and eighty-five well-attributed
single-author plays of the period 1580–1619 by other dramatists showed
Arden 4–7 far distant from the non-Shakespeare plays and pretty closely
associated with the Shakespearean ones (figure 4.3). Scenes 8–9 were
placed right in the middle of the Shakespeare group, but so was a single non-Shakespeare play. The other two Arden segments were placed
among the non-Shakespeare plays, but among those close to the
Shakespeare plays.
Kinney’s conclusion about Arden that ‘Shakespeare was one of the authors;
and his part is concentrated in the middle section of the play’ seems perfectly reasonable.27 It is true, as Vickers says, that Kinney is wrong, or only
half-right, to claim that the Craig–Kinney project ‘confirms MacDonald
P. Jackson’s recent attribution of scene viii to Shakespeare’.28 It yields no
clear verdict on that one scene, taken alone. But the overall pattern of the
results, including the performance of scenes 8–9 on function-word testing,
which classes them as quintessentially Shakespearean, does strongly suggest that the whole of Act 3 of Arden, including scene 8, was substantially
Shakespeare’s. And it accords well with the fact that it is mainly from material in the middle of the play that the strongest evidence cited in earlier
articles has, without foreknowledge of the Craig–Kinney findings, been
drawn. This is demonstrated in Chapter 3 of the present volume. Moreover,
as Kinney observes, the four groupings of scenes remain open to modification: any one of them could well contain writing by more than one author.
Michael’s soliloquy at the end of scene 3, for example, seems stylistically
indistinguishable from speeches of similar length within Act 3. The last
seventeen lines of scene 8, when the quarrel between Alice and Mosby is
over and Bradshaw arrives (151–67) are less Shakespearean on the LION
results than the monologue of lines 1–44 and dialogue of lines 45–150.
27. Craig and Kinney, 99.
28. Vickers, ‘Authorship Studies’, 129. Kinney does not use the exact words quoted by Vickers,
who has conflated Kinney’s mention of Jackson’s argument for Shakespeare’s authorship
of scene 8 (Craig and Kinney, 90) with Kinney’s statement that his research ‘confirms
MacDonald P. Jackson’s recent proposal’ (99). Vickers misleads when he says that ‘none
of the three tests place scene 8 on the “Shakespeare” side of the diagram’ (129). There was
only one test of scene 8 as a scene, and the scenes 8–9 segment falls right in the middle of the
Shakespeare plays in figure 4.3, which graphs results for tests of function words.
Authorship Studies of Shakespeare and his Contemporaries 51
The reason why scene 8—as part of the block amalgamating it with
scene 9—is thoroughly Shakespearean on the function-word test but not
on the lexical test may well be as follows. Craig and Kinney use original
early modern texts. They explain that they ‘standardize selected function
words to modern usage’, but for other words have developed software that
collects different spellings of the same word under a single head for counting.29 However, although the quarto text of the Quarrel Scene (scene 8) is
basically sound, it contains several very unusual spellings, some possibly
misprints: ‘gaile’ for ‘gale’, ‘erre’ for ‘ear’ (the verb meaning ‘to plough’),
‘mede’ for ‘meed’, ‘damne’ for ‘dam’, ‘demianor’ for ‘demeanour’, ‘slaundrous’ for ‘slanderous’ or ‘sland’rous’, ‘excirsimes’ for ‘exorcisms’, ‘dowe’
for ‘dove’, ‘gentiles’ for ‘gentles’. LION searches of the modern spellings of these words, with the ‘Variant spellings’ function activated, recognize none of the quarto variants and the Craig–Kinney software was
probably likewise flummoxed. Yet, if LION data for the period 1580–1619
can be trusted, ‘meed’, ‘demeanour’, ‘dove’, ‘gentles’, and ‘slanderous’ or
‘sland’rous’ are all words that appear in a larger proportion of Shakespeare’s
plays than of plays by his contemporaries. There are also several words and
phrases that most editors, rightly in my view, emend: ‘gentle stary’ (‘gentlest airy’), ‘heaue’ (‘hive’), ‘fence’ (‘fount’). The words ‘airy’ and ‘hive’
are also used more frequently by Shakespeare than by other playwrights of
his time. So is ‘gentlest’, but in this case, if the emendation is correct, the
quarto compositor simply mistook one Shakespearean word for another;
besides there are three other instances of ‘gentle’ in the scene, and Craig and
Kinney register only whether a lexical word is present in a block of text, not
the number of occurrences.
Computer software would have interpreted the Arden quarto’s ‘erre’ as
‘err’. But this is of less significance. Although Q’s ‘erre’ really stands for
the verb ‘ear’ meaning ‘plough’ and this is a Shakespearean rarity (as made
clear in the entry under ‘ear the land’ in Appendix 1, Part A and the associated note), the Craig–Kinney procedures—which distinguish concordance
words (or graphic units) rather than dictionary headwords—would, even
from a modernized text, have counted instances of the pertinent ‘ear’ along
with instances of its homograph, the organ of hearing.
Whether or not anomalous spellings affected Craig and Kinney’s lexical
tests of Arden’s scene 8, the multiplicity of evidence presented in Chapters 1–5
of the present book vindicates Kinney’s conclusion that the whole central
29. Craig and Kinney, xvii.
52
Determining the Shakespeare Canon
portion of the play covered by scenes 4–9 is substantially Shakespeare’s. It
is a pity that Craig and Kinney did not test Arden by using the Act divisions of older editions. Presumably they did not notice that scenes 4–9 constituted Act 3. It seems probable that Act 3 would have been classified as
Shakespearean on both the lexical-word and function-word tests.
In considering Kinney’s chapter on Arden, Vickers says nothing whatsoever about the additional function-word tests, of the single-author versus single-author type, which fail to associate any of the four blocks with
Marlowe (figure 4.4) and locate all four within Shakespeare territory rather
than Kyd territory (figures 4.5–6). The only small complication in this latter finding is that, when Soliman and Perseda is included in the Kyd canon,
one solitary Kyd segment out of twenty-two falls within the Shakespeare
group. As Kinney sums up, ‘there is no pattern of association’ even with a
Kyd grouping that includes Soliman and Perseda (widely attributed to Kyd on
persuasive internal evidence), as well as The Spanish Tragedy and Cornelia.30
Shakespeare, Computers, and the Mystery of Authorship strikes me as a good
book. More might have been done to investigate a possible chronological
element in the various findings. From a traditional literary-critical point of
view, Shakespeare’s mature dramatic verse seems more recognizable than
his earlier verse. I should expect to find more early than later Shakespeare
plays or segments in those areas on the scatter-plots where Shakespeare
and non-Shakespeare texts overlap. It would have been helpful to have
those texts that inhabit these shared regions identified and listed and the
co-ordinates or principal components for each tabulated, though that
information may have been too hard to extract. One wonders what particular ‘Kyd’ segment, for instance, so closely adjoins the Arden 8–9 and 4–7
segments within the Shakespeare-not-Kyd grouping in figure 4.6. A further point is worth making. It might have been wise randomly to divide
the initial 1,298 segments for lexical-word testing into two sets, the larger
being used to generate the discriminators, the other, constituting at least
10 percent of the segments, reserved to test their efficacy on new material.
This would have enabled a clearer estimate of the strength of ‘the regression
effect’.
30. Craig and Kinney, 99.
Authorship Studies of Shakespeare and his Contemporaries 53
II
Vickers makes some positive remarks about the accomplishments of traditional ‘reading-based’ attribution studies, with their ‘attentiveness to language and their direct engagement with the text, down to the minutest
details of usage’ (113). They can be alert to ‘the occasion of an utterance and
the uses to which it is put’ (113). Vickers is generous in his praise of such work
on Thomas Middleton’s canon, including my own. In connection with
Stevenson’s book, he affirms that, ‘if performed systematically’, the ‘citation of parallel passages is a legitimate procedure’, and is impressed by ‘the
series of striking convergences between the 1602 additions and Shakespeare
plays published before and after that date’ (107). En masse, Stevenson’s parallels are indeed impressive, and it is true, as Vickers notes, that several
are unique within the period. But can we be sure that a thorough search
of the Jonson canon, for example, would not also yield impressive parallels, including some unique ones? In collecting parallels ‘systematically’ we
need to employ a system that allows comparisons between rival candidates.
Vickers outlines a new technique for attribution that he has devised with
Marcus Dahl (138–42). They use plagiarism software to find sequences
of three words, or ‘trigrams’, that are shared between pairs of plays, consulting each matching trigram in context to determine whether there is
more extensive parallelism, and then checking an electronic database
of sixty-four plays performed in the London theatres within the period
1580–95, in order to compile a list of ‘unique’ matches. Vickers cited many
unique matches between Kyd plays (The Spanish Tragedy, Cornelia, and
Soliman and Perseda) and plays he wished to attribute to Kyd: Fair Em the
Miller’s Daughter, Arden of Faversham, King Leir, and the portions of 1 Henry
VI that are by neither Shakespeare nor Nashe.31 He assumed that a case for
his attributions had been made, considering that ‘a sufficient number of
unique shared collocations . . . gives us a strong authorship marker, eliminating other possible explanations, such as chance, imitation, or plagiarism’
(140). But how can we decide how many would be ‘sufficient’? Despite
the technological advance, which identifies trigrams objectively, the methodology is incomplete, because, like Stevenson’s amassing of parallels, it
creates results for only a single candidate. These prove nothing, because
31. Vickers, ‘Thomas Kyd, Secret Sharer’.
54
Determining the Shakespeare Canon
when plagiarism software is set to search for trigrams shared by Arden of
Faversham and 2 Henry VI, or by Arden of Faversham and The Taming of the
Shrew, and these are checked for uniqueness during the specified period,
each of the two Shakespeare plays turns out to afford considerably more
unique matches than any of the three canonical Kyd plays; and the quality of
the Shakespeare matches is at least as high, if not higher.32
One way to overcome this deficiency is to adopt ‘the cumbersome
method’, as Vickers calls it (139), of methodically working through the
anonymous or disputed text line by line, searching all the plays, within
a specific period, in the Literature Online database for phrases and collocations in the work under investigation that occur not more than five times
elsewhere. Conducting the enquiry in this way allows any play or group of
plays, by any dramatist, to yield the greatest number of matches with the
text whose authorship is uncertain. The performances of rival candidates
on such tests can be compared.33
In Chapter 1 this methodology has already been applied to the Quarrel
Scene, scene 8, in Arden of Faversham. But, as a complement to the Craig–
Kinney tests, which attribute scenes 4–7 unambiguously to Shakespeare,
we may examine Arden’s account, in scene 6, of his distressing nightmare.
The writer displays remarkable insight into the mind’s innermost workings. Arden’s unconscious sends him messages that on a conscious level
he has repressed. He dreams (6.6–31) that nets are set to catch deer, and
that while waiting for ‘the herd’s approach’, he falls asleep and is himself
trapped in the toils by a forester; another herdsman advances on him with
a broadsword crying, ‘Thou art the game we seek.’ As he declares to his
friend Franklin, when he awoke in terror, the dream remained so vivid that
‘I stood in doubt whether I waked or no.’ Beginning as hunter Arden ends
as prey, an accurate reflection of his dual role in the tragedy, as a predatory
landlord who is a target for killers: the menacing forester and the herdsman symbolize the hired assassins Black Will and Shakebag. Instinctively
32. Jackson, ‘New Research’. David Hoover, in his Hamburg Digital Humanities conference
paper, ‘The Rarer They Are, the More There Are, the Less They Matter’, has also exposed
the weakness in Vickers’s case: it is available at http://www.dh2012.uni-hamburg.de/
conference/programme/abstracts.
33.Inevitably, the searcher will miss a few parallels, because of oddities in some LION texts
or his or her own inadvertence, and a subjective element doubtless influences the choice of
items to report. But occasional errors and decisions with which another investigator might
disagree cannot be avoided in any such human undertaking, and the overall pattern of
results is what matters.
Authorship Studies of Shakespeare and his Contemporaries 55
Table 2.1 Summary of LION links to Arden’s dream, 6.6–31
Author(s)
Title (probable date of first
performance)
Shakespeare and others
Shakespeare
Shakespeare
Shakespeare
Shakespeare and others
Marlowe
Shakespeare
Shakespeare
Shakespeare
Shakespeare and Peele
Greene and Lodge
Jonson
3 Henry VI (1591)
Richard III (1592–3)
A Midsummer Night’s Dream (1595)
Love’s Labour’s Lost (1594–5)
Edward III (1590)
The Massacre at Paris (1593)
Romeo and Juliet (1595)
King John (1596)
Julius Caesar (1599)
Titus Andronicus (1592)
A Looking Glass for London (1590)
A Tale of a Tub (1596)
No. of links
6
3
3
3
3
3
2
2
2
2
2
2
Arden senses his vulnerability, and the manifest content of his nightmare
is apt to an aficionado of the chase, as Arden is revealed to be by an exchange
unobtrusively planted in later dialogue (10.6–8). The action of the whole
play is a protracted manhunt, which Arden’s dream emblematizes. The
Shakespearean ‘inset’ of the gamekeepers in 3 Henry VI, 3.1, serves a similar
function, with its deer-killing imagery. Arden’s speech has a psychological
subtlety equalling Clarence’s monologue about his dream in Richard III,
1.4. But what of its verbal texture?
As for the Quarrel Scene, the period set for LION searching was 1580–
1600 and the same rules governed procedures. The results are listed in full
in Appendix 2. Table 2.1 gives figures for the twelve plays that have two or
more links to the Arden passage.
In that list there are six plays wholly by Shakespeare and three of which
he is co-author, and only three by three other playwrights. The full catalogue of matches contains thirty-three to Shakespeare plays, thirty-four to
all the other dramatists’ plays combined.34 Twenty-five of the thirty-three
Shakespeare matches are to plays of 1590–5. The only non-Shakespeare
play with as many as three matches, Marlowe’s The Massacre at Paris, is
widely accepted as being preserved in a corrupt text, probably based on
memorial reconstruction, so that it is likely to contain actors’ importations
34.This figure includes matches with Shakespeare’s collaborations. These are discussed in
Chapter 3.
56
Determining the Shakespeare Canon
from other plays; moreover two of its matches are interdependent and to
the same Arden line about a toil being pitched to catch deer. In quality, the
better Shakespeare parallels are equal or surpass the non-Shakespeare ones.
The evidence from shared phrases and collocations supports Craig and
Kinney’s findings that Shakespeare contributed to Arden of Faversham. As
Vickers notes, the convergence of ‘different analytical methods on the same
candidate is always a significant outcome in attribution studies’ (123).
Phrases that do not qualify according to the rules would have furnished
additional links to Shakespeare. Arden’s locution ‘This night I dreamed’, for
example, is an unusual variant on the expected ‘Last night’, but Shakespeare
has no fewer than nine references, in plays of 1580–1600, to somebody
having dreamed ‘tonight’, this collocation of the past tense with ‘tonight’
occurring in only two non-Shakespeare plays of the same period; although
the LION search also brings up an example in The Merry Devil of Edmonton,
this is dated 1602 in Annals. The hunting scenes of 3 Henry VI (4.5) and
Love’s Labour’s Lost (4.1) contain many of the elements of Arden’s dream
within a short space: 3 Henry VI, 4.5.3–17, has ‘park’, ‘king’, ‘the game’,
‘stand’, and ‘deer’, while Love’s Labour’s Lost, 4.1.1–8, has ‘King’, ‘rising’,
‘hill’, ‘forester’, ‘bush’, and ‘stand’, with ‘deer’ following at 35.
Moreover, Shakespeare’s narrative poem Venus and Adonis, entered in
the Stationers’ Register on 18 April 1593, published in that year, and probably written after the closure of the theatres through plague in July1592,
contains at least two passages with multiple links to Arden’s speech. It is
natural that the hunting context common to the poem and Arden’s dream
should generate some similarities in the language used, but the series of
items listed in Table 2.2 are shared by no other passage of comparable
length in any play, poem, or prose work in the whole LION database.35
Complex networks of association in a single brain provide the most
35.This can be confirmed by checking LION’s ‘Variant spellings’ and ‘Variant forms’ slots,
setting the proximity function ‘NEAR’ at a range that covers the 95 lines of Venus and
Adonis 554–648 or, for the second set of links, the 32 lines of 868–99, and keying in a string
of some of the actual shared words (whatever their inflexion), thus: ‘lion NEAR.200 thorn*
NEAR.200 joint NEAR.200 trembled NEAR.200 bush’, or ‘for?ag* NEAR.800 impression NEAR.800 bent NEAR.800 thorn*’ for 554–648; ‘horn NEAR.400 quake NEAR.400
bush NEAR.400 fantasy’, or ‘lion NEAR.400 timorous NEAR.400 brake NEAR.400 surprise’ for 868–99. The last of these strings also brings up a nineteenth-century novel, but the
noun ‘brake’ does not appear: ‘Variant forms’ has merely found ‘broken’. Searches for the set
of correspondences with 554–648 were described in Jackson, ‘Quarrel Scene’, 272–3, where
‘NEAR.100’ is, however, a mistake for ‘NEAR.1000’. This is corrected in Chapter 1 above.
Authorship Studies of Shakespeare and his Contemporaries 57
Table 2.2 Multiple links between Arden and Venus and Adonis
Arden of Faversham, 6.7–30
Venus and Adonis, 554–648
rest
A toil was pitched to overthrow the deer
rounded ( = surrounded, encircled)
Another herdman came | With falchion
drawn, and bent it at my breast
With falchion drawn
my breast
I . . . trembled every joint
obscurèd in a little bush. . . the thorny
casements of the brake
lion
foraging
timorous
quakes and shivers though the cause
be gone
rest
the fleet-foot roe that’s tired with
chasing compass’d
a mortal butcher bent to kill
With javeling’s point
my breast
my joints did tremble
thorny brambles and embracing bushes
impression
lion
forage
fearful
My boding heart pants, beats, and takes
no rest, | But like an earthquake
shakes thee
impression
Arden of Faversham, 6.14–32
rounded (= surrounded, encircled)
deer
horn
trembled
obscurèd
bush
lion
timorous
brake
quakes
surprise
fantasy
Venus and Adonis, 868–99
embrace
doe fawn
horn
trembling
hid
bushes
lion
timorous
brake
quaking
surprise
fantasy
plausible explanation of these data. The correspondences can hardly be
due to Venus and Adonis’s having influenced a non-Shakespearean author of
Arden, since the play was published first—in 1592, after having been entered
in the Stationers’ Register on 3 April of that year. On the other hand, in
writing his poem Shakespeare can hardly have been influenced by having
acted in a wholly non-Shakespearean Arden, because the only two characters on stage in scene 6 are Arden and Franklin, while in the Quarrel
Scene (scene 8)—which shows a similar concatenation of words linking it
to 2 Henry VI and The Rape of Lucrece (published 1594) and in which LION
58
Determining the Shakespeare Canon
matches to Shakespeare are prevalent—only Mosby and Alice are present,
until Bradshaw makes a brief appearance at the end. Since actors learned
their roles solely from their parts, no mere member of the cast acting Arden
of Faversham would have had the opportunity thoroughly to familiarize
himself with both Arden’s dream and the scene 8 dialogue between Mosby
and Alice.36 Whether or not Shakespeare ever poached deer from a park
owned by Sir Thomas Lucy or somebody else, he had an intimate knowledge of rural Warwickshire and of hunting practices, so that the imagery of
Arden’s speech, Venus and Adonis, Love’s Labour’s Lost, 4.1, and 3 Henry VI,
4.5, came naturally to him.37
Among the items linking Arden of Faversham, 6.7–30, with Venus and
Adonis, 554–648, are ‘lion’, and forms of ‘forage’, and ‘tremble’. The association of a lion with foraging is also listed in the main LION search results
as peculiar to Henry V, Love’s Labour’s Lost, and Edward III among plays of
1580–1600. If we add ‘tremble’ and its cognates, the whole LION database
(poetry, drama, and prose) finds only three works, all of them plays, that
share close juxtaposition of all three elements: Arden (6.20–2), King John
(5.1.57–9), and Edward III (2.1.395–7). In Edward III they occur within one
of the ‘Countess scenes’, which are those that have been most confidently
attributed to Shakespeare.38 The King John lines did not qualify for the main
LION list of matches, because ‘Forage’ is not applied to the lion but to those
who are advised not to ‘seek the lion in his den’ (57), but mention of a lion
has clearly evoked the verb. So three of the elements in the cluster of words
and images linking Arden, 6.7–30, with Venus and Adonis, 554–648, themselves form a uniquely tight cluster found only in Arden, an early play by
Shakespeare (King John, 1596), and a scene by Shakespeare within an early
co-authored play (Edward III, 1590).
36. Simon Palfrey and Tiffany Stern, Shakespeare in Parts (Oxford: OUP, 2007).
37.The deer-stealing legend is discussed by S. Schoenbaum, William Shakespeare: A Compact
Documentary Life (Oxford: Clarendon Press, 1977), 107–9. As we shall see in Chapter 3,
Section III, the block of text in 3 Henry VI consisting of 4.2–8 has recently been categorized
as not by Shakespeare, but the methodology employed would have been unable to single
out from 4.2–8 a genuinely Shakespearean scene as short as the twenty-nine line 4.5, which
continues the deer-hunting theme of the indisputably Shakespearean 3.1.
38. The LION search used the range ‘lion NEAR.40 forage* NEAR.40 tremble’, with ‘Variant
spellings’ and ‘Variant forms’ activated. Truncation of ‘forage’ was necessary to pick up
Edward III’s ‘forragement’ (the quarto spelling). When the range was extended to ‘NEAR.50’,
an instance of the triple juxtaposition was found in John Kirke’s The Seven Champions of
Christendom (1635), written thirty years later than any of the other three plays. Extension to
‘NEAR.100’ yielded no further additions. The same searches of Early English Books Online
did not augment these findings.
Authorship Studies of Shakespeare and his Contemporaries 59
Vickers quotes approvingly Stevenson’s statement that ‘the strongest
proofs of common authorship are not single or isolated parallels, but clusters or interrelated groups of images and phrases which combine to form a
distinct pattern’.39 The links between Arden of Faversham, scene 6, and Venus
and Adonis are of this kind, and are at least as extensive as any that Stevenson
cites between the Spanish Tragedy Additions and Shakespeare’s works. As
evidence they are, however, subject to the same reservations as Stevenson’s
parallels and Vickers’s matching trigrams: some other author’s poem might
afford multiple links to Arden’s dream, were we to conduct a diligent
search—different links from those to Venus and Adonis but in combination
no less rare. Although, intuitively, this seems unlikely, it is possible. But
the results of the main LION searches of plays first performed within the
period 1580–1600 are not open to this criticism, because all the extant plays
of all the dramatists writing within that period have been examined in the
same way. The examination has been of those ‘recurrent verbal patterns’
and ‘phrasal repetends’ that Vickers, on the basis of ‘corpus linguistics’ theory, considers good authorial indicators. And the results point clearly to
Shakespeare’s authorship of Arden’s narrative of his dream in scene 6.
39. Vickers, ‘Authorship Studies’, 107. Vickers quotes from an earlier article: Warren Stevenson,
‘Shakespeare’s Hand in The Spanish Tragedy 1602’, Studies in English Literature: 1500–1900, 8
(1968), 314.
3
Gentlemen, Arden of Faversham,
and Shakespeare’s Early
Collaborations
I
‘This Figure, that thou here seest put, | It was for gentle Shakespeare
cut’, announced Ben Jonson in his epigram on the page facing the famous
Droeshout engraving in the First Folio of 1623. Shakespeare would doubtless have been pleased with the adjective, testimony not only to his demeanour but also to his social standing. Amiable, courteous, and kind (according
to Jonson), he was also a gentleman. He had worked hard to earn the title
for his father, John, and himself, having been granted a coat-of-arms from
the College of Heralds in 1596. In Will in the World, Stephen Greenblatt
suggests that Shakespeare’s whole career was driven, to a significant extent,
by an urge to recover the status his father had lost upon the collapse of the
family fortunes when William was thirteen.1
Upward mobility was possible even within the rigid early modern
English class system. But few among the masses climbed far up the social
ladder. Rousing his troops before the Battle of Agincourt, Shakespeare’s
Henry the Fifth promises:
. . . he to-day that sheds his blood with me
Shall be my brother; be he ne’er so vile
This day shall gentle his condition.
(4.3.61–3)
1.Stephen Greenblatt, Will in the World: How Shakespeare Became Shakespeare
(New York: Norton, 2004), passim.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
61
Common soldiers will, the king claims, be ennobled by their actions. But
when the victorious Henry reads the list of the English dead, the roll call
comprises ‘Edward the Duke of York, the Earl of Suffolk, | Sir Richard
Ketley, Davy Gam, esquire; | None else of name’ (4.8.103–5). Those below
the level of the gentry are simply nameless ‘other men’.
The Prologue to Henry V apologizes for the incapacity of the bare
London stage to do justice to Henry’s exploits:
But pardon, gentles all,
The flat unraised spirits that hath dar’d
On this unworthy scaffold to bring forth
So great an object.
(Pro. 8–11)
The appeal to ‘gentles all’ is pure flattery. The groundlings were no more
transformed by the appellation than Henry’s soldiers were elevated in society by their heroic deeds. Robin Goodfellow’s (Puck’s) plea, at the end of
A Midsummer Night’s Dream, that the audience overlook the actors’ faults
may have had a firmer basis in fact: ‘Gentles, do not reprehend. | If you
pardon, we will mend’ (5.1.429–30). The ‘Gentles’ whom Peter Quince
addresses in his prologue to the ‘Pyramus and Thisbe’ interlude (5.1.127)
include Theseus, Duke of Athens, his bride Hippolyta, and their wedding
guests, and it has often been supposed that A Midsummer Night’s Dream itself
was ‘designed to grace a wedding in a noble household’.2
Besides Henry V and A Midsummer Night’s Dream, there is—according to
LION—one, and only one, other English play of the period 1575–1600 in
which a prologue or epilogue asks ‘gentles’ or ‘gentlemen’ to ‘pardon’ any
perceived shortcomings. This is the earliest of the three, Arden of Faversham,
first published anonymously in 1592, having probably been composed
within the years 1588–91.3 At the end of the play, Franklin steps forward
to deliver an epilogue in which he recounts the fates of those implicated
in the murder of his friend Thomas Arden and to apologize, in a grossly
hypermetrical line, for the plainness with which the crime has been dramatized: ‘Gentlemen, we hope you’ll pardon this naked tragedy’ (Epi. 14).
Here again the playwright ingratiates himself with an original audience
unlikely to have contained a majority of theatre-goers entitled to call
themselves, in any formal sense, ‘gentlemen’. But Arden himself is twice
2. William Shakespeare, A Midsummer Night’s Dream, ed. Harold F. Brooks (London: Methuen,
1979), liii.
3.Wine, Arden, xlv.
62
Determining the Shakespeare Canon
labelled ‘gentle Arden’ (1.401, 4.21), and every character in the play—from
deus ex machina Lord Cheiny to the free-ranging criminals Black Will and
Shakebag—is acutely conscious of his or her status in a hierarchical society. Indeed, like Franklin in the epilogue, the dramatis personae confer titles
to curry favour; or they revile as ‘peasant’, ‘churl’, ‘knave’, and ‘villain’,
those to whom they wish to assert their superiority. When the well-born
Alice Arden quarrels with her lover Mosby, a jobbing-tailor who has risen
to be Lord Clifford’s steward, she abuses him as ‘[a]‌mean artificer, that
low-born name’ (8.77). Unable to bear his vituperation in response, she
seeks to mollify him, eliciting his wounded sarcasm: ‘Make love to you?
Why, ’tis unpardonable; | We beggars must not breathe where gentles are’
(8.138–9); to which she replies: ‘Sweet Mosby is as gentle as a king’, half
acknowledging that this is flattery by adding ‘And I too blind to judge him
otherwise’ (8.140–1). Alice’s besotted protestations are, in reality, no better able to ‘gentle’ Mosby’s ‘condition’ and make him king-like than King
Henry’s martial rhetoric is able to turn his subjects into his brothers.
Arguing that ‘a fully informed historicism cannot afford to overlook the
way in which texts, at the level of the most intricate verbal detail, are vehicles of historical meaning’, Michael Neill explored the subtle ways in which
social change is reflected in ‘the language of status’ by which the characters
manipulate one another in Arden of Faversham.4 The actual murder of Thomas
Arden in 1551 at the instigation of his adulterous wife—on which, by way
of Holinshed’s narrative, the play was based—became the focus of anxieties
surrounding ‘the politics of state and the politics of gender’.5 Political and
household government are among Arden of Faversham’s concerns. But the
struggle for land ‘as both an index of rank and a fungible commodity’ is, as
Neill shows, crucial.6 Sexual possession becomes a means to material possession and social prestige. ‘Gentle Alice’ is a pawn in the power-play between
the Renaissance ‘new man’ Arden and the ambitious Mosby. The previous
chapters gave grounds for supposing that a co-author of this play featuring
4.Michael Neill, ‘ “This Gentle Gentleman”: Social Change and the Language of Status in
Arden of Faversham’, Medieval and Renaissance Drama in England, 10 (1998): 73–97, at 73. My
references are to the journal, but the article was reprinted in Neill’s Putting History to the
Question: Power, Politics, and Society in English Renaissance Drama (New York: Columbia
UP, 2000).
5.Neill, 73, describing an essay by Catherine Belsey, ‘Alice Arden’s Crime’, Renaissance
Drama, n.s. 13 (1982), 83–102, revised in Belsey’s The Subject of Tragedy (London: Methuen,
1985), 129–48.
6.Neill, 75.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
63
‘gentle Arden’ and ‘gentle Alice’ was the ‘gentle Shakespeare’. Here I want
to consider how well independently derived data, mostly already available in
miscellaneous publications, comport with the determinations of Craig and
Kinney, in whose approach to problems of the Shakespeare canon the word
‘gentle’, curiously enough, played a significant role.
II
As we have seen, Thomas Kyd, Christopher Marlowe, and William
Shakespeare have all, at one time or another, been proposed as sole or
part-author of Arden of Faversham. In Chapter 1 I argued that Shakespeare
wrote the Quarrel Scene, scene 8, and in Chapter 2 I outlined the research
reported in the recent book edited by Hugh Craig and Arthur F. Kinney,
Shakespeare, Computers, and the Mystery of Authorship, judging that it represents
a significant advance in attempts to determine the limits of the Shakespeare
canon and that their evidence in support of the theory that Shakespeare
contributed to Arden is sound. Their methods of attribution are validated on
plays and sections of plays whose authors are known: before applying them
to doubtful texts, Craig and Kinney regularly assess the degree to which
their statistical tests correctly credit texts to the playwrights agreed to have
written them. The tests seldom yield perfect separation between authors,
but they classify texts of known authorship correctly in the vast majority of
cases. No single procedure for determining authorship on internal grounds
can be conclusive, but Kinney and Craig are forthright about the limitations of theirs and cautious in interpreting their results. They go a long way
towards confirming some previous theories and showing others to be most
improbable. The evidence is particularly compelling when, as is usually
the case, the two independent modes of testing—through function words
such as ‘and’, ‘thy’, ‘it’, and ‘very’, and through lexical words such as ‘gentle’,
‘answer’, ‘brave’, and ‘hopes’—clearly point to the same conclusions.
Note in that last sentence the ‘lexical word’ gentle. Craig explains that
this, ‘it turns out, is a favourite Shakespeare word’.7 Although neither unusual, nor neologistic, nor archaic in Shakespeare’s time, ‘gentle’ occurs
‘nearly twice as regularly in Shakespeare’s plays as in Early Modern English
dialogue generally’ being ‘a persistent preference, a minor but recurring
7. Craig and Kinney, 16.
64
Determining the Shakespeare Canon
thread in Shakespeare’s linguistic fabric’.8 Two-thousand-word segments
of Shakespearean dramatic text contain ‘gentle’ more often than not, the
opposite being the case in a control set of segments from other playwrights.
This does not, of course, mean that the presence of ‘gentle’ in the scene of
an anonymous play can, in itself, tell us anything much about the likely
author. But in combination large numbers of words favoured or eschewed
by Shakespeare can. It is reassuring that, for instance, the Craig–Kinney
findings are largely in line with orthodox opinion, as reflected in the
Oxford William Shakespeare: The Complete Works.9 Because Thomas Nashe
wrote only one single-author play, Craig and Kinney are unable to consolidate the recently accumulated evidence that Thomas Nashe is the author
of Act 1 of Henry VI, but ‘[t]‌he Nashe hypothesis survives’ their examination.10 And, after testing claims for Shakespeare’s involvement in Arden
of Faversham ‘in two quite independent but mutually reinforcing ways’,
Kinney sums up: ‘Given that the results broadly confirmed each other, we
can be confident in our conclusions: Arden of Faversham is a collaboration;
Shakespeare was one of the authors; and his part is concentrated in the
middle section of the play.’11 This portion is defined, in terms of modern
editors’ division into scenes but not Acts, as beginning at scene 4 and ending with scene 9. Older editors marked scenes 4 and 9 as the first and last
of Act 3.
Craig and Kinney concede that the shorter the blocks of Arden text tested
the less reliable the findings, and that the boundaries they set between
‘Shakespearean’ and ‘non-Shakespearean’ writing are provisional and
approximate. There could well be passages outside scenes 4–9 of Arden that
would qualify as Shakespearean by their criteria, and even within ‘the middle section of the play’ the Craig–Kinney tests do not preclude the presence
of some non-Shakespearean matter. But the fact that scenes 4–9 constitute
the traditional Act 3 is highly suggestive, since, as noted in the previous
chapter, playwrights’ shares in early modern dramatic collaborations often
took the form of large blocks such as Acts.12 Moreover, re-examination of the
distribution of various features previously recorded in work on the authorship of Arden and 2 Henry VI uncovers some remarkable correspondences
8. Craig and Kinney, 16.
9. Attributions are discussed in Wells and Taylor, Textual Companion; Edward III is admitted
into the second edition of Complete Works (2005).
10. Craig and Kinney, 55.
11. Craig and Kinney, 99.
12. Vickers, Shakespeare, Co-Author, 27–34.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
65
with the Craig–Kinney results. These and other correspondences, detected
in new research, are the central topic of this chapter.
First, however, we may consider some data derived from M. L. Wine’s
Revels edition of Arden of Faversham. In appendix 1 Wine lists ‘the more
interesting’ verbal ‘parallels’ that previous scholars had noted between
Arden and contemporary works.13 Within scenes 4–9, covering 23.6 per cent
of the play’s total number of lines, there are twenty-seven such references
to works by Shakespeare, thirteen to works by other authors. Within the
remainder of Arden, covering 76.4 per cent of the play’s total number of
lines, there are thirty-one references to works by Shakespeare, forty-six to
works by other authors. The probability of so great a disparity (27:13 versus
31:46) occurring by chance is one in 155.14 If there is no substance in the
Craig–Kinney claim, the proportion of parallels that are with Shakespeare
ought to be roughly the same in scenes 4–9 as elsewhere.
Other data show similar disparities. In an appendix to ‘Shakespearean
Features of the Poetic Style of Arden of Faversham’, published in 1993, I listed
items belonging to strands of imagery common to Arden and Shakespeare’s
early works.15 The images were of horticulture and unweeded gardens;
animals and victims; archery, riding, hunting, and bird snaring; words as
weapons; and acting and theatre. Fourteen of the thirty examples in Arden
came from scenes 4–9, which means that almost half were concentrated
within the ‘Shakespearean’ section of the play as defined by the Craig–
Kinney tests, a stretch of text containing fewer than a quarter of the play’s
lines. A further seven listed images fell within scene 3, all but two within its
final forty-six lines. Among those scenes of Arden not categorized by Craig
and Kinney as Shakespearean, scene 3 is—apart from the very short scene
11, of only 268 words and entirely in prose—the closest to the bisector
line separating, on lexical-word tests, the majority of 2,000-word segments
known to be by Shakespeare from the majority of 2,000-word segments
known to be by other dramatists. In fact it inhabits a region on Craig
and Kinney’s figure 4.1 where there appears to be some overlap between
13. Wine, Arden, 141–7. In his commentary Wine naturally glosses or otherwise illuminates
particular phrases by citing parallel usages in contemporary plays.
14.The website research.microsoft.com/en-us/redmond/projects/MSCompBio/FisherExact
Test/ calculates Fisher Exact Test probabilities for 2 × 2 contingency tables. For 27:13 versus 31:46 it yields p = 0.006445. Fisher’s test is explained in all introductions to statistics.
A good one for non-mathematicians is Russell Langley, Practical Statistics Simply Explained
(New York: Dover, 1971).The website obviates the need to consult tables of probability values.
15. Jackson, ‘Shakespearean Features’, 209–304.
66
Determining the Shakespeare Canon
Shakespeare and non-Shakespeare segments. It seems likely that scene 3
contains at least some Shakespearean writing. Further grounds for believing so are presented below.
In ‘Shakespearean Features’ I drew up two tables displaying whole complexes of images that Arden, scene 8.1–150, shares with The Rape of Lucrece,
lines 829–1078, and with 2 Henry VI, 3.1.31–229, and also itemized eleven
linked elements that Arden, scene 6.7–30, shares with Venus and Adonis, lines
554–648.16 The tables are incorporated in Chapter 1, while the itemized
links are tabulated in Chapter 2, along with a newly discovered series associating Arden, scene 6.14–32 with Venus and Adonis, lines 868–92. Scenes 6 and
8 both fall, of course, within the portion of Arden attributed to Shakespeare
by Craig and Kinney. In ‘Shakespearean Features’ I also analysed in detail
Arden’s speech at 4.1–20 as displaying a typically Shakespearean complexity
in the generation of imagery through a kind of wordplay.17 This speech, too,
turns out to belong within the section of Arden identified as Shakespeare’s
by Craig and Kinney’s computerized techniques. Analysis of the speech has
been included within Chapter 4.
It is Act 3, not only of Arden of Faversham but also of 2 Henry VI, in which
Craig and Kinney’s tests locate the surest evidence of Shakespeare’s hand,
and the image complex shared between the two plays falls within Act 3
of 2 Henry VI.18 Paul Vincent—analysing the distribution of O and Oh
spellings of the exclamation, along with a few linguistic variables—had
earlier advanced evidence isolating Act 3 of 2 Henry VI and connecting
it firmly to Shakespeare.19 It can hardly be coincidental, therefore, that in
my ‘Shakespearean Features’ list of separate images belonging to categories
favoured by Shakespeare and found in Arden eight of the nine examples
drawn from 2 Henry VI come from the history play’s Act 3.
There are further such correlations. In ‘Shakespeare, Arden of Faversham,
and “Literature Online” ’, Jayne Carroll and I summarized some research
using the resources of the LION electronic database.20 This involved a systematic search for rare links in phrasing between Arden and the works of
16.Jackson, ‘Shakespearean Features’, 294, n. 29, 297–9. The uniqueness of the links with
Venus and Adonis and The Rape of Lucrece was later established by LION searches, as reported
in Chapter 1.
17. Jackson, ‘Shakespearean Features’, 283–5. The analysis has been incorporated in Chapter 4.
18. Craig and Kinney, 68–77.
19.Paul Vincent, ‘Unsolved Mysteries in Henry the Sixth, Part Two’, Notes and Queries, 246
(2001), 270–4.
20. See Chapter 1, n. 27.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
67
the most prominent playwrights of the 1580s and early 1590s, but the methodology—devised when investigations employing LION were in their
infancy—differed from that employed in investigating the Quarrel Scene,
as described in Chapter 1. Corpora matched, so far as possible, for size and
generic mix were formed of plays and poems by Greene, Marlowe, Peele,
Shakespeare, and a miscellaneous group of dramatists from whom fewer
plays survive: Kyd, Lodge, Munday, and Nashe. The LION search functions were employed to check ten sample passages from Arden for phrases
and collocations shared with only one of the five authorial groups. As we
reported, ‘The passages in which links with Shakespeare clearly predominate are 4.1–20, 5.1–12, 6.5–31, and 8.1–150.’21 All fall within the central Act
3 of the play, selected by the Craig–Kinney tests as Shakespearean. All four
passages are again shown in this book to exhibit other Shakespearean characteristics. Arden’s lament at 4.1–20 has just been mentioned; 5.1–12 will
be discussed in Chapter 4, and 6.5–31 covers his account of his nightmare,
examined in Chapter 2.
Of course 8.1–150 covers most of the Quarrel Scene, and the more comprehensive LION searching—also carried out before Craig and Kinney’s
book appeared and reported on in my 2006 article on ‘Shakespeare and
the Quarrel Scene’ and here in Chapter 1—demonstrated that scene
8 as a whole contains far more verbal links to individual early plays by
Shakespeare than to individual plays by other dramatists. Plays wholly or
partly by Shakespeare overwhelmingly dominated the list of those furnishing four or more links. I pointed out that, whereas 3 Henry VI provided
twenty-two links and 2 Henry VI twelve, 1 Henry VI—of which, as Gary
Taylor argued, Shakespeare had written less than 15 per cent—provided
only six, two of which fell within the undoubtedly Shakespearean 2.4; that
all six rare phrases shared by Arden, scene 8, and Edward III fell within the
three ‘Countess scenes’ (1.2, 2.1, and 2.2), which had most confidently been
ascribed to Shakespeare; and that only one of seven phrases shared between
Arden and Titus Andronicus fell within the roughly one-third of the Roman
tragedy that had been ascribed to George Peele.22 I also commented upon
the distribution of rare phrases that Arden, scene 8, shared with 2 Henry VI,
noting that six of the twelve are with the history’s 3.1 and 3.2, which had
21. Carroll and Jackson, 3.
22. Taylor, ‘Shakespeare and Others’. Vincent’s When ‘harey’ Met Shakespeare was published too
late to be consulted by Craig and Kinney.
68
Determining the Shakespeare Canon
been classed by disintegrationist editor John Dover Wilson as among those
where Shakespeare’s hand was most ‘manifest throughout’.23
Craig and Kinney’s findings are in broad agreement with the assumptions made in 2006 in ‘Shakespeare and the Quarrel Scene’ about the division between Shakespeare’s and his collaborators’ writing in 1 Henry VI,
Titus Andronicus, and Edward III, and, as we have seen, they judge Act 3
of 2 Henry VI to be the section that can most confidently be ascribed to
Shakespeare. This means that twenty out of the thirty-one links to Arden,
scene 8, afforded by 1 and 2 Henry VI, Titus Andronicus, and Edward III are
with sections of these four plays that are identified by the Craig–Kinney
tests as Shakespeare’s, although these sections constitute only a little over a
third of the total number of lines.
III
The fact that of the three Henry VI plays, 3 Henry VI has easily the most links
to the Quarrel Scene had itself seemed suggestive, in the light of Craig and
Kinney’s inability to discern in it the writing of any author but Shakespeare.
However, in a painstaking recent analysis, undertaken in collaboration
with John Burrows, Craig has uncovered grounds for attributing almost
two-fifths of 3 Henry VI to a playwright who cannot yet be identified,
while validating a little over three-fifths as by Shakespeare.24 Employing
Craig’s database of all the surviving well-authenticated single-author
English plays of 1580–99, he and Burrows applied two separate tests—of
function words and of lexical words peculiar to a single authorial group—
to ‘rolling segments’ of 2,000 words of the play. These were sets ‘in which
each successive segment abandons the opening of its predecessor and
replaces it by advancing a further step into the text’.25 Thus the first segment consisted of the first 2,000 words, the second of words 201–2,200, the
third of words 401–2,400, and so on. This allowed likely boundaries to be
established, as fluctuations in degrees of closeness to Shakespearean norms
were observed. Allocations of scenes as predominantly Shakespeare’s and
predominantly not Shakespeare’s, made on this basis, were then checked
23. Jackson, ‘Shakespeare and the Quarrel Scene’, 259, 274–5; The Second Part of King Henry VI,
ed. John Dover Wilson (Cambridge: CUP, 1952), 49, 158.
24. Craig and Burrows, ‘A Collaboration about a Collaboration’.
25. Craig and Burrows, ‘A Collaboration about a Collaboration’, 39.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
69
by a further test—of ‘word-types that occur with some consistency in one
sample-batch but not in another comparable batch’.26 This kind of testing
produced a clear separation between the two groups of scenes. Those in the
non-Shakespearean group were 1.1, 1.2, 2.3, 3.3, 4.2–4.8, and 5.2.
Corroboration of Craig and Burrows’ determinations on 3 Henry VI comes
from an unexpected quarter. The most perceptive literary-critical analysis of style in Shakespeare’s early plays is Marco Mincoff’s Shakespeare: The
First Steps.27 Mincoff wrote within a period of reaction against ‘disintegration’, so that while noting characteristics of Nashe in the first Act of 1 Henry
VI and of Peele in the first Act of Titus Andronicus, he explained them by
assuming that Shakespeare, as actor turned novice dramatist, was susceptible to the influence of his peers. But in his discussion of Shakespeare’s progress in employing imagery and speech-patterning in 3 Henry VI Mincoff
cited, in illustration of the advances he detected, some thirty-nine passages,
all but three of which fall into scenes that Craig and Burrows attribute to
Shakespeare.28 Proportionally to the sizes of the putatively Shakespearean
and non-Shakespearean shares, we would expect (to the nearest whole
number) a 24:15 division, rather than a 36:3 one. The probability that a random distribution of the images and rhetorical figures selected by Mincoff
would result in such an imbalance is less than one in 6,000.29
The chief point of interest, however, is that only five of the Arden Quarrel
Scene’s twenty-two links to 3 Henry V are to non-Shakespeare scenes
(totalling 9,044 words), the remaining seventeen being to Shakespeare
scenes (totalling 14,439 words). In Table 3.1 the results for 3 Henry VI are
summarized along with those for the other four collaborations.30 In each
26. Craig and Burrows, ‘A Collaboration about a Collaboration’, 57.
27. Marco Mincoff, Shakespeare: The First Steps (Sofia: Bulgarian Academy of Sciences, 1976).
28. Mincoff, First Steps, 90–9. Mincoff describes one of the three that falls within a scene that
Craig and Burrows designate ‘not Shakespeare’ as ‘rather colourless’ (95). It is worth adding
that five of the six images from 3 Henry VI that I cited in ‘Shakespearean Features’, 299–304,
as illustrating that Arden and Shakespeare’s plays and poems shared particular strands of
imagery came from scenes of 3 Henry VI that Craig and Burrows allocate to Shakespeare.
29.If we use Yates’s chi-square to test the ‘goodness of fit’ of the actual figures (36:3) to the
expected figures (23.98:15.02, to give more precise figures than the whole numbers), we
get Yates’s chi-square = 14.37, 1.d.f., p < 0.00015. The chi-square test, Yates’s correction,
and degrees of freedom (d.f.) are also explained in Langley, Practical Statistics, and similar
textbooks. I have used the Vanderbilt University website: K. J. Preacher, ‘Calculation for
the chi-square test: An interactive calculation tool for chi-square tests of goodness of fit and
independence’, http://www.quantpsy.org/chisq/chisq.htm.
30. Line totals are based on A. S. Cairncross’s Arden editions of 1, 2, and 3 Henry VI (1962, 1957,
1964), J. C. Maxwell’s Arden edition of Titus Andronicus (1953), and on the Riverside edition for
Edward III. In the order in which, for each play, they are recorded in the inventory of
70
Determining the Shakespeare Canon
Table 3.1 Numbers of rare phrases and collocations linking Arden’s Quarrel Scene to
Shakespeare’s early collaborations
Shakespeare scenes
2 Henry VI
3 Henry VI
1 Henry VI
Titus
Andronicus
Edward III
TOTALS
Non-Shakespeare scenes
Links
Lines
Links
Lines
6
17
2
6
837
1,770
376
1,732
6
5
4
1
2,207
1,137
2,297
785
6
37
997
5,712
0
16
1,496
7,922
one of Shakespeare’s five early collaborative plays the LION links to Arden’s
Quarrel Scene are, proportionally to the amounts of text, more numerous
with scenes designated Shakespeare’s than with the remaining scenes. This
is despite the fact that Craig and Kinney do not claim that Shakespeare’s
part in 2 Henry VI was necessarily confined to Act 3.31
Further, the total number of links to Shakespeare scenes in all five plays
is thirty-seven out of fifty-three. In proportion to the amounts of text
categorized by Craig and Kinney’s tests as substantially Shakespearean or
non-Shakespearean, instead of a 37:16 division of links with Arden, scene
8, we should expect a 22:31 division, were the links spread randomly. Such
a disparity between actual and expected figures, which has a less than one
in 14,000 probability as a purely chance result, seems inexplicable if the
Quarrel Scene is not by Shakespeare.32
There is a further angle from which to look at these results. Converted to
rates for a play of 2,000 lines, the total Quarrel Scene links to Shakespearean
scenes of the five early collaborations work out, to the nearest whole
links in Appendix 1, Part A, the Riverside line references are (giving only the first line of
longer passages): 2 Henry VI: 5.3.13, 3.1. 89, 2.1.98, 1.3.102, 4.2.89, 3.2.260, 3.1.76, 1.1.230,
3.1.155, 3.1.101, 4.1.72, 3.1.32; 3 Henry VI: 4.6.13, 2.5.49, 3.2.127, 5.2.14, 2.2.31, 3.2.176,
2.6.65, 5.4.62, 2.3.47, 1.4.169, 4.7.58, 2.5.70, 2.2.139, 3.2.98, 1.4.57, 3.1.121, 2.5.130, 5.6.21,
1.4.106, 5.7.29, 3.2.57, 3.3.207; 1 Henry VI: 3.1.123, 2.1.52, 5.4.159, 2.4.33, 2.4.69, 4.1.136; Titus
Andronicus: 2.3.132, 2.4.37, 1.1.457, 3.1.131, 5.2.140, 5.2.171, 2.4.49; Edward III: 1.2.96, 1.2.79,
2.1.438, 2.2.109, 2.1.2, 2.1.370.
31. As Mincoff notes (First Steps, 89–92), passages in 5.2 and 5.3, such as Young Clifford’s monologue at 5.2.31–65 and York’s speech at 5.3.1–7, remind one of the mature Shakespeare’s
Henry IV.
32. With actual figures of 37:16 and expected figures of 22.2:30.8, Yates’s chi-square = 15.851, 1
d.f., p = 0.00006853.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
71
number, as thirteen and the rates of links to the non-Shakespearean scenes
as four. These contrasting rates correspond quite closely to the actual
numbers of links that, in Table 1.1, distinguish the early unaided plays by
Shakespeare from the plays of his contemporaries.
These findings about the distribution of Quarrel Scene links with early
Shakespeare collaborations are of the utmost importance. Possible authorial
divisions within them played no part whatsoever in the collection of LION
data, and when the searches were first carried out, it was assumed that 2 and
3 Henry VI were probably Shakespeare’s alone. So the LION evidence for
Shakespeare’s authorship of the Quarrel Scene cannot have been generated
through biased conduct of the searches or biased recording of their results.
Further, the soundness of the methodology seems guaranteed by its capacity
to distinguish, in terms of numbers of links to the Quarrel Scene, between
Shakespeare’s writing and that of other playwrights within the five early collaborative plays—at least so far as these two categories have been determined
by Craig and Kinney, using their own different techniques.
One additional detail affords extra confirmation. The control searches of
key passages from Marlowe’s Doctor Faustus, described in Chapter 1 and in
Appendix 1, Part B, yielded five links to Arden, two to 1 Henry VI, and one
each to 2 Henry VI and Titus Andronicus, and eight of these nine items fell
within portions of text not assigned to Shakespeare, Titus’s ‘come, come,
and’ being the sole exception.33 In striking contrast to the Quarrel Scene,
the passages from Doctor Faustus are overwhelmingly associated, on LION
testing, with Craig and Kinney’s non-Shakespearean portions of these plays.
Arden’s narrative of his dream at 6.6–31, on the other hand, has thirteen links to Shakespeare’s five early collaborations and nine of these are
to scenes attributed to Shakespeare.34 Two of the three links to Edward III
are to 3.5, which is included by Timothy Irish Watt in his research for Craig
and Kinney’s Shakespeare, Computers, and the Mystery of Authorship in his
6,000-word segment of the play’s non-Shakespearean matter, but which
Oliphant considered Shakespeare’s, at least in part.35 There are certainly
33. References for Doctor Faustus links to Arden and early Shakespeare collaborations (in the
order in which, for each play, they are recorded in Appendix 1, Part B) are as follows: 2
Henry VI: 4.10.5; 1 Henry VI: 5.3.25; 5.3.82; Titus Andronicus: 5.3.160; Arden: 18.10–11, 12.42,
13.17, 1.356.
34.References (for each play in the order in which phrases occur in Arden, 6.6–31): 2 Henry
VI: 4.7.85; 3 Henry VI: 4.5.17–19, 1.4.67, 1.4.12, 1.4.18, 5.6.13, 5.7.11; 1 Henry VI: 4.2.45; Titus
Andronicus: 3.1.89, 2.3.212; Edward III: 3.5.2, 3.5.108, 2.1.395–6.
35.Craig and Kinney, 116–33; E. H. C. Oliphant, ‘Problems of Authorship in Elizabethan
Dramatic Literature’, Modern Philology, 8 (1911), 411–59, at 421–2.
72
Determining the Shakespeare Canon
touches suggestive of Shakespeare’s hand, such as the ‘reflexive conceit’ in
‘But if himself himself redeem from thence’ (3.5.50) and the vivid and complex image of Prince Edward as a lion entangled in a net, while the French
‘like emmets on a bank, | Muster about him’ (3.5.28–32).
IV
Craig and Kinney’s differentiation of those portions of Arden of Faversham
most and least likely to be by Shakespeare makes a further LION test desirable. Scene 14 is not only the second longest in the play but one of the least
Shakespearean on the Craig–Kinney lexical tests, falling in their figure 4.1 in
the middle of the non-Shakespeare blocks of 2,000 words. LION searches of
14.1–76 (to the entrance of Mosby), carried out in the same way as searches
of scene 8, produced results recorded in Appendix 3. Plays sharing four or
more rare phrases or collocations with Arden 14.1–76 are listed in Table 3.2.
Links with Sir Thomas More are all to the original, unrevised script. The
Annals date for More has been accepted, though there is a strong possibility
that even the original play was composed later than 1595.36 The totals for
More, Every Man in His Humour, 1 Edward IV, and A Warning for Fair Women
would, in each case, be lower by one, were counts based on the presence
in a play of the phrase shared with the Arden passage, and the number of
instances disregarded. For instance, in More the phrase ‘our guests’ occurs
twice. Plays in Table 3.2 other than the four mentioned have only a single
instance of each shared phrase or collocation.
Tables 1.1 and 3.2 could hardly be less alike. Shakespeare utterly dominated Table 1.1, with eight of his plays at the head of the rank order according to numbers of links and twelve in the top fourteen. The total of
twenty-two for 3 Henry VI, which topped the list, was almost three times
greater than the total of eight for the two highest plays by dramatists other
than Shakespeare. Only two Shakespeare plays qualify for the thirteen of
Table 3.2 (a proportion of one in six and a half ), which is the number we
36.‘Deciphering a Date and Determining a Date: Anthony Munday’s John a Kent and John
a Cumber and the Original Version of Sir Thomas More’, Early Modern Literary Studies, 15
(2011), 2, 1–24; http://purl.org/emls/15-3/jackdate.htm; Sir Thomas More, ed. John Jowett
(London: Arden Shakespeare, 2011), 424–32.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
73
Table 3.2 Summary of LION links to Arden, 14.1–76
Author(s)
Title (probable date of first
performance)
Haughton
Shakespeare
Porter
Jonson
Anon.
Shakespeare
Heywood
Anon.
Chettle, Day, Haughton
Marlowe
Marston
Munday and Chettle
Jonson
Englishmen for My Money (1598)
As You Like It (1599–1600)
The Two Angry Women of Abingdon (1588)
Every Man in His Humour (1598)
A Knack to Know an Honest Man (1594)
The Merry Wives of Windsor (1597–8)
1 Edward IV (1599)
A Warning for Fair Women (1599)
The Blind Beggar of Bednal Green (1600)
Edward II (1592)
Jack Drum’s Entertainment (1600)
Sir Thomas More (1595)
A Tale of a Tub (1596)
No. of links
8
7
7
6
5
5
5
5
5
4
4
4
4
might expect purely by chance, since twenty-three of the 134 plays searched
(one in six) are by Shakespeare.
Shakespeare’s first four plays, according to the Oxford chronology, were
at the top of Table 1.1, and the twelve Shakespeare plays in the top fourteen
were all earlier than the two Shakespeare plays in Table 3.2, As You Like It
(1599–1600) and The Merry Wives of Windsor (1597–8). Curiously, As You
Like It was the only Shakespeare play of 1580–1600 to afford no links to the
Quarrel Scene. The items that it shares with Arden, 14.1–76 (‘word of a tapster’, ‘hence . . . here comes’, ‘they shook hands’, ‘railed on’, ‘And this night’,
‘And . . . welcome shall you be’, ‘I’ll stand to it’) are literal, colourless, and
purely verbal, whereas many of those that 2 and 3 Henry VI, The Taming of
the Shrew, and The Two Gentlemen of Verona share with the Quarrel Scene
involve parallel ideas and images, as well as words.
There is a further way in which the LION results for Arden scene 14
contrast sharply with those for the Quarrel Scene: five of the seven links to
Titus Andronicus and the three Parts of Henry VI (there being none to Edward
III) are to scenes not assigned to Shakespeare.37
37. 2 Henry VI: 4.7.121, 4.10.14; 3 Henry VI: 2.1.5, 3.2.42; 1 Henry VI: 2.2.8; Titus Andronicus:
1.1.121, 4.1.118.
74
Determining the Shakespeare Canon
Does Table 3.2 afford any clue to Shakespeare’s co-author in Arden of
Faversham? William Haughton is prominent, but the earliest record of him
as a playwright is an entry in Henslowe’s Diary for 5 November 1597, in
which he appears as ‘yonge’ Haughton.38 He worked on at least a score of
plays for the Admiral’s Men during 1597–1602. But a man who was ‘yonge’
in 1597 could hardly have co-authored a play that was in print by 1592.
Henslowe first mentions Henry Porter, also high in Table 3.2, on 16
December 1596, but in Porter’s case an earlier beginning as playwright
is possible.39 A pamphlet called Plaine Percivall (1590; STC 12914) by
Richard Harvey contains the words, ‘Qui mocchat, moccabitur quoth the
servingman of Abington’ (B4v). In the quarto of The Two Angry Women
of Abingdon, first published in 1599, the servant Nicholas Proverbs quotes
this dog-Latin tag. So it has been supposed that Harvey was alluding to
a version of Porter’s play, which Annals enters under the year 1588. In
her critical edition, however, Marianne Brish Evett argues that Porter’s
Nicholas and Harvey’s Percivall—a ‘plain fellow’ who mixes homely
wit and university learning—may have recycled the same popular jest,
or that Porter may have read Harvey’s pamphlet. 40 She gives good reasons for thinking that The Two Angry Women dates from about 1598,
shortly before the performance of its lost sequel, a Part 2 with the same
title, was performed in February 1599. E. H. C. Oliphant believed in
the early dating, but detected in the quarto of the extant The Two Angry
Women of Abingdon some uncertain signs that the script had undergone
alteration for an Admiral’s Men’s revival.41
The Two Angry Women is, as Oliphant averred, ‘a comedy of splendid
vivacity, with a breeziness and a swing that are very attractive’, adding: ‘In
his bluff heartiness Porter is the most English of the dramatists.’42 J. M.
Nosworthy observes that Two Angry Women ‘is native and racy because
38. Chambers, Elizabethan Stage, III, 334–6.
39. Chambers, Elizabethan Stage, III, 466–7. J. M. Nosworthy advocates the early dating in
‘Notes on Henry Porter’, Modern Language Review, 35 (1940), 517–21, and ‘Henry Porter’,
English, 6 (1942), 65–9.
40. Henry Porter’s The Two Angry Women of Abingdon: A Critical Edition, ed. Marianne Brish
Evett (New York and London: Garland, 1980), 11–26; Evett fully documents previous commentary on the play.
41. Elizabethan Dramatists Other than Shakespeare, ed. E. H. C. Oliphant (New York: PrenticeHall, 1931), 240–1.
4 2. Oliphant, Elizabethan Dramatists, 241.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
75
Porter refers every character and incident to the realities of everyday life
as he experienced it’. Remarking that ‘[t]‌here is no lack of local colour’
and that certain speeches ‘have something of the Falstaffian richness’, he
noted ‘the curious fact that The Two Angry Women stands rather outside
the general drift of Elizabethan dramatic endeavour, in company with
the seemingly different play of Arden of Feversham. Arden is the standard
representative of domestic tragedy verging on melodrama: The Two Angry
Women, the clearest absolute specimen of domestic comedy verging on
farce.’ He considered the ‘slight deviation from pure tragedy in the one and
pure comedy in the other’ to be ‘the inevitable consequence of the essential
domesticity of the plot and its treatment’.43 Characters spend much of their
time in Porter’s play (a ‘nocturnal’, like A Midsummer Night’s Dream) missing one another in the dark and stumbling into ditches, like Black Will
and Shakebag in the fog of Arden, scene 12. Conceivably, Porter might
repay investigation by Craig’s ‘computational statistics’ team as a possible co-author of Arden of Faversham. But the prevalence of rhyme in The
Two Angry Women and Porter’s ‘trick of distributing half-lines of rhyming
verse among his characters’ and sporadic use of tetrameters contribute to a
verse style unlike Arden’s, in which tetrameters are eschewed and rhyme is
sparse.44 Nor does Arden make The Two Angry Women’s uncommonly frequent use of ‘ye’. The relatively large number of phrases and collocations
that The Two Angry Women shares with Arden, scene 14.1–76, is probably
attributable to the ‘everyday life’ contexts, in which a demotic linguistic
register prevails, and the excessive length of Porter’s play, 3,037 lines by
Nosworthy’s count.
What we can say with full confidence is that the glaring contrast between
Arden scenes 8 and 14 in the nature of their LION data is in accord with the
Craig–Kinney testing that finds Shakespeare’s contribution to be concentrated in the middle of the play and scene 14 to fall outside the normal
Shakespeare parameters.45
43. Nosworthy, ‘Henry Porter’, 66–8 (for excerpts in this and the previous three sentences).
4 4. Nosworthy, ‘Henry Porter’, 67.
45. One more contrast should be noted. Whereas in Table 1.1, most links are with Shakespeare’s
earliest plays, composed in the early 1590s, as we should expect if Arden, scene 8 is
Shakespeare’s, in Table 3.2 most links are with plays, including two by Shakespeare, of the
late 1590s, which would be surprising were scene 14.1–76 Shakespeare’s. So not only does
Table 3.2 contain only a chance number of Shakespeare plays, but it lacks Table 1.1’s significance in terms of chronology.
76
Determining the Shakespeare Canon
V
The Craig–Kinney findings are further buttressed by some previously published data concerning compound adjectives in Arden of Faversham. Long ago
Alfred Hart pointed out that Shakespeare employed more adjectival compounds in his plays than his main early contemporaries employed in theirs and
that the disparity is most striking in the use of participial compound adjectives, especially those formed with a present participle or with a noun plus
participle.46 In ‘Compound Adjectives in Arden of Faversham’ I showed that
Arden is more like the early plays of Shakespeare than like those of Marlowe,
Greene, or Peele in this regard.47 But relating the compound adjectives in
Arden to their context, we find that their rate of use is very much higher in
scenes 4–9 than in the rest of the play, as can be seen in Table 3.3.
Moreover, those that fall outside scenes 4–9 are concentrated within lines
98–200 of scene 3 and, more intensely, lines 47–57 of scene 2. Scene 3 has
‘soft-mettled’ (98), ‘dry-sucked’ (111), ‘hunger-bitten’ (193) and ‘ill-intending’
(200), the last two involving noun plus participle. The other two compound
adjectives that are formed with noun plus participle and occur outside scenes
4–9 both appear within the tight cluster of compound adjectives at 2.47–57. In
this passage the goldsmith Bradshaw tells Black Will about a man who brought
to his shop some plate that turned out to have been stolen from Lord Cheyne.
He was ‘A lean-faced, writhen knave, | Hawk-nosed and very hollow-eyed’
and wore ‘A pair of threadbare velvet hose, seam rent’ (2.47–8, 56). Besides
‘lean-faced’, ‘hawk-nosed’, ‘hollow-eyed’, ‘threadbare’, and ‘seam-rent’ (which
is hyphenated in Craik’s edition), there is also ‘to-torn’. Bradshaw’s description
of the thief, whom Will identifies as one Jack Fitten, has a close parallel in the
description by Antipholus of Ephesus of Pinch in Shakespeare’s The Comedy of
Errors:
a hungry lean-fac’d villain,
A mere anatomy, a mountebank,
A threadbare juggler and a fortune-teller,
A needy, hollow-ey’d, sharp-looking wretch.
(5.1.238–41)
46. Alfred Hart, Shakespeare and the Homilies (Melbourne and London: Melbourne UP and
OUP, 1934), 236; for commentary that places the table in context, see Hart, 232–9, 254–5.
Hart notes that Marlowe, especially, used many compound adjectives in his poems, but,
like most other playwrights, eschewed them in plays.
47. MacDonald P. Jackson, ‘Compound Adjectives in Arden of Faversham’, Notes and Queries, 250
(2006), 51–5.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
77
Table 3.3 Compound adjectives in Arden of Faversham
Arden
Scenes 4–9
Rest of play
No. of lines
567
1,833
Numbers of compound adjectives
Total
Noun +
participle
Present
participle
18
24
6
4
5
4
Each passage contains the compound adjectives ‘lean-faced’, ‘threadbare’, and
‘hollow-eyed’, while ‘sharp-looking’ in Errors is virtually equivalent to the
slightly more specific ‘hawk-nosed’ in Arden. In the whole Literature Online
database there is no other passage in which ‘lean-faced’, ‘threadbare’, and
‘hollow-eyed’ are juxtaposed. The Oxford editors date The Comedy of Errors
1594 and no contemporary scholar considers it to be earlier than 3 April 1592,
when Arden was entered in the Stationers’ Register. Besides, Errors was not
in print before its inclusion in the Shakespeare First Folio of 1623. Even if the
current dating of Shakespeare’s farce is wrong, it is unlikely to have influenced
a non-Shakespearean author of Bradshaw’s speech. So as an alternative to the
theory that Shakespeare wrote both passages, we would again have to postulate his exceptional familiarity with Arden of Faversham, and this time with
an episode in which the only characters on stage are Black Will, Shakebag,
Bradshaw, and Greene.
LION searches for instances of Arden’s compound adjectives in all drama of
the period 1580–1600 established that five of the nine plays with three or more
are by Shakespeare, in whole or in part, with 2 Henry VI heading the list with
six, and that the twenty-three Shakespeare plays of the period use seventeen
of the Arden adjectives, while the over one hundred other plays use twenty.
But in the present context it is the disparity between scenes 4–9 and the rest of
the play that is crucial, since it reinforces Craig and Kinney’s conclusions: the
567 lines of scenes 4–9 contain nine of the seventeen words, while the 1,833
remaining lines contain eight, and three of those eight fall within Bradshaw’s
pen-portrait of Jack Fitten at 2.47, 48, and 56. So the distribution of compound
adjectives in Arden also supports Kinney’s verdict, on the evidence of ‘computational stylistics’, that it is within scenes 4–9 that Shakespeare’s contribution
is ‘concentrated’.
In another article whose publication preceded that of Craig and Kinney’s
book but which forms the basis of the present Chapter 4, I scrutinized some
verbal parallels between Arden of Faversham and Kyd’s Soliman and Perseda as a
78
Determining the Shakespeare Canon
step towards demonstrating that the domestic tragedy often evinces a poetic
imagination of a Shakespearean rather than a Kydian kind.48 Five of the eight
passages chosen for analysis (3.98–99, 3.100–103, 3.159–63, 5.6–9, 8.54–7, 8.88–
101 [especially 88–94], 9.38–45, 9.63–7) fall within scenes 4–9 and the three
outside this portion of the text all belong to the later part of scene 3.
It is conceivable that Craig and Kinney’s tests of lexical words and the
earlier tests of phrases and images that are described above are not completely independent in what they measure: certain kinds of imagery and
phraseology might, for example, be associated with certain kinds of lexical
words, regardless of authorship. But it is hard to imagine any appreciable
non-authorial association between choices of images, phrases, and collocations, on the one hand, and rates of usage of high-frequency function
words such as ‘and’, ‘thy’, ‘it’, and ‘very’, on the other. So the old evidence,
when revisited, confirms the new.
VI
Further confirmation comes from some locutions in Arden of Faversham that
point away from Shakespeare but that were probably not considered by
Craig and Kinney. These are absent from the middle of the play, congregating towards the beginning and end. For example, I long ago noted that
instances of the exclamation ‘tush’ and of ‘seeing’ to mean ‘considering
that’ or ‘since’—both rare in Shakespeare’s plays—were confined to the
earliest and latest scenes of Arden, where ‘tush’ occurs twelve times and
‘seeing’ ten.49
If we count ‘tush, tush’ as a single example of the exclamation, the
highest number of instances in any Shakespeare play is four in 1 Henry VI
(4.1.178, 5.3.89, 5.3.107, 5.5.10) and the next highest three in Much Ado About
Nothing (3.3.123, 5.1.58, 5.4.44), while the highest number of instances of
‘seeing’ with the sense ‘since’ is also three, in 2 Henry VI (3.1.270, 4.2.174,
and 4.7.73) and 3 Henry VI (1.1.218, 1.1.247, 1.2.26). The highest combined
total is four in 3 Henry VI (with ‘tush’ at 4.7.13), compared with Arden of
48.MacDonald P. Jackson, ‘Parallels and Poetry: Shakespeare, Kyd, and Arden of Faversham’,
Medieval and Renaissance Drama in England, 23 (2010), 17–33.
49. Jackson, ‘Material for an Edition’, 117–20. In Wine’s edition of Arden the words are located
as follows: ‘tush’: 1.103, 1.436, 3.89, 3.153, 3.207, 14.136, 14.140, 14. 217, 14.241, 14.276,
14.292, 14.297; ‘seeing’: 1.276, 1.409, 1.410, 1.478, 1.585, 3.115, 14.187, 14.200, 14.205, 18.37.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
79
Faversham’s twenty-one. ‘Seeing’ appears nine times in Marlowe’s Edward
II, while ‘tush’ appears eight times in Marlowe’s The Jew of Malta. Again,
it can hardly be coincidental that instances of these two locutions seldom
used by Shakespeare are concentrated within non-Shakespearean scenes of
the five early collaborations. ‘Tush’ occurs once in Edward III (3.2.32) but
neither expression occurs in Titus Andronicus. This means that eleven of the
twelve instances of ‘tush’ or ‘seeing’ (for ‘since’) in all five plays combined
fall outside scenes attributed to Shakespeare.
Less compelling, but nevertheless of interest, is the distribution of
the colloquialism ‘ay, but’ (invariably spelt ‘I but’) in Arden of Faversham.
It occurs nine times, and Thomas Merriam regarded this frequency as
favouring Kyd’s authorship of the play, because ‘ay, but’ appears six times
in The Spanish Tragedy and eight times in Soliman and Perseda.50 But none of
the nine instances in Arden falls within scenes 4–9.51 In his twenty-two early
plays up to 1599–1600 (ending with As You Like It), Shakespeare averages
one ‘ay, but’ per play: only two plays have more than two examples, The
Two Gentlemen of Verona with six and 3 Henry VI with eight.52
However, six of the eight in 3 Henry VI are in scenes identified as
Shakespearean by the Craig–Burrows tests,53 so that in this case there is not
the same distinction as in Arden between Shakespeare and non-Shakespeare
scenes. Also, there is one ‘ay, but’ in a Shakespearean scene of Titus
Andronicus (4.4.79), but the sole ‘ay, but’ in 1 Henry VI is in a scene not
assigned to Shakespeare (3.1.136), as are the three in 2 Henry VI (4.1.47,
4.1.113, 4.4.20). The colloquialism is absent from Edward III. So in the early
collaborative plays taken together (and excluding Arden), ‘ay, but’ is found
seven times in Shakespearean scenes, five in non-Shakespearean scenes,
which means that there is no tendency for Shakespeare’s co-authors to use
them at the greater rate. Nevertheless, the fact that all nine examples of ‘ay,
but’ in Arden are in scenes not assigned to Shakespeare by Craig and Kinney
50.Thomas Merriam, ‘Possible Light on a Kyd Canon’, Notes and Queries, 240 (1995), 340–1.
Merriam also counts the much less common ‘nay, but’, but, since this is absent from Arden,
I have adjusted Merriam’s totals by excluding instances of this variant from mine.
51. The Arden instances of ‘Ay, but’ occur at 1.22, 1.69, 1.144, 1.235, 1.238, 2.27, 11.29, 14.68,
14.338.
52. The seventeen later plays (from Hamlet to The Two Noble Kinsmen) have the same rate of use,
averaging exactly one ‘ay, but’ per play, but Twelfth Night scores four and Othello three.
53.‘Ay, but’ occurs in 3 Henry VI at 2.6.85, 3.1.42, 3.1.59, 3.2.48, 3.2.60, 4.3.16, 4.3.31, 5.5.75.
Since several plays of 1580–1600, including John Lyly’s Campaspe, Endymion, Gallatea, and
Midas, have five or more instances of ‘ay, but’, the presence of nine examples in Arden is a
very dubious pointer to Kyd’s authorship of even the scenes in which they are found.
80
Determining the Shakespeare Canon
is in accord with Shakespeare’s sparing use of the colloquialism, or complete avoidance of it, in nearly all his plays.
Also pertinent are some curious resemblances in phrasing between
Arden of Faversham and Marlowe’s The Jew of Malta (not published till 1633).
They would be unremarkable, except that they accumulate within a mere
forty-four lines of the latter (2.3.338–81). I list them in compressed form
in a note.54 The parallels may seem trivial, but being concentrated within
such a short stretch of Marlowe’s play, they cannot be purely coincidental. Besides, several of the shared phrases are much rarer than one might
have supposed. According to LION, among plays first performed within
the period 1580–96, Arden and The Jew of Malta are the only two in which
‘sure’ means betrothed (as in being ‘sure’ to Sue or Abigail), or that contain
‘I’ll make them (’em) friends’, ‘’Tis we (I) must do the deed’, or ‘cunningly
performed’. The only parallel to Arden’s ‘But, Michael, see you do it cunningly’ and The Jew of Malta’s ‘but do it cunningly’ is ‘And therefore will
I do it cunningly’ in Marlowe’s Edward II (1592), and this is less close to either
of the two than they are to each other, in that it lacks the introductory ‘but’.
Arden of Faversham and The Jew of Malta are closely contemporary, and
precedence cannot confidently be determined.55 Copy for the quarto of The
Jew of Malta was evidently a seventeenth-century transcript, which may
well have contained revisions to the play by Thomas Dekker.56 But the section of 2.3 that provides so many verbal parallels with Arden seems likely to
have been original, and Dekker is unlikely to have borrowed from Arden.
The significant point for the present purposes concerns the distribution
of the ten Arden phrases with parallels in forty-four consecutive lines of
The Jew of Malta: five in Arden, scene 1, four in scene 14, and one in scene
3. None of the ten belongs to scenes 4–9.
54. In each case I give the phrase from Arden first. Line numbers for The Jew of Malta are for 2.3
in Christopher Marlowe, The Jew of Malta, ed. David Bevington (Manchester: Manchester
UP, 1997): ‘Hath made report that he and Sue is sure’ (1.150): ‘I have made thee sure to
Abigail’ (338); ‘Revenge it on the proudest of you both’ (1.309), and ‘till we meet
next’ (1.408): ‘Revenge it on him when you meet him next’ (346); ‘what’s that to thee’
(3.142): ‘What’s that to thee’ (360); ‘I’ll make them friends’ (14.199): ‘I’ll make ’em friends’
(360); ‘How like you this’ (14.116): ‘how likest thou this?’ (367); ‘Ay, Alice, and it was cunningly performed’ (14.19): ‘True; and it shall be cunningly performed’ (370); ‘Why, so it
shall’ (1.428) and ‘’Tis we must do the deed’ (14.140): ‘Ay, so thou shalt; ’tis thou must do the
deed’ (372); ‘But, Michael, see you do it cunningly’ (1.164): ‘but do it cunningly’ (381).
55. Wine, Arden, xlv; Bevington, Jew of Malta, 1.
56.Lake, ‘Three Seventeenth-Century Revisions’ and private communication dated 10
January 2003 (on Dekker’s possible involvement).
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
81
This means that instances of four putatively non-Shakespearean features
of Arden of Faversham—‘tush’, ‘seeing’, ‘Ay, but’, and phrases shared with
The Jew of Malta, 2.3.338–81—are, taken together, distributed within Arden
as follows: seventeen in scene 1, one in scene 2, five in scene 3, one in
scene 11, sixteen in scene 14, and one in scene 18. The forty-one instances
cluster towards the beginning and end of Arden, and none occurs within
the middle portion identified by Craig and Kinney as Shakespearean. The
one anomaly in this pattern is that scene 3, which proved to be borderline on the Craig–Kinney tests and in which various signs of Shakespeare’s
hand have been noted, contains no fewer than five of the supposedly
non-Shakespearean markers. This overlap may indicate mixed authorship.
The extent to which independently observed Shakespearean and
non-Shakespearean features of Arden of Faversham fall into patterns matching the objective findings of the computerized Craig–Kinney tests of
favoured and non-favoured lexical words and function words strongly supports the conclusion that Shakespeare contributed substantially to the play,
having been largely responsible for at least scenes 4–9 or Act 3. The authorship of the rest of Arden remains problematical. Craig and Kinney, supplementing their main analysis with one-on-one tests pitting Shakespeare
against Marlowe and Shakespeare against Kyd, uncovered ‘no sustained
affinities between the Arden sections and the work of Marlowe or Kyd,
either in vocabulary or in function-word use’.57 Even the blocks of text
other than 4–9 proved, on the two-author comparisons, to be more closely
associated with Shakespeare than with either Marlowe or Kyd.
Interpretation of these results is complicated, however, by the uncertain
state of the text of the Arden quarto of 1592. Alfred Hart judged that ‘the
verse exhibits many of the defects of a report and over a hundred lines are
harsh or unmetrical’. He considered the metrical defects ‘undoubtedly the
work of actors’.58 As long ago as 1963 I elaborated on Hart’s assertion that
‘the text of Arden is poor’ and argued that it evidently suffered from some
form of memorial contamination, perhaps through the agency of a careless
scribe familiar with the play as acted.59 But a disproportionate number of
the lines I cited as probably corrupt came from the scenes before and after
the ‘Shakespearean’ middle section, scenes 4–9. It seems unlikely that any
57. Craig and Kinney, 99.
58. Hart, Stolen and Surreptitious Copies, 384.
59. Jackson, ‘Material for an Edition’, 12–37; Wine, Arden, xxiv–xxxiii, agreed that the quarto
showed signs of memorial contamination, and offered further evidence.
82
Determining the Shakespeare Canon
author would have Alice say (to quote from the Malone Society Reprint
of the quarto)60 ‘His company hath purchest me ill freends’ (line 2200) and
so shortly afterwards ‘Your company hath purchased me ill freends’ (2228),
or have Greene say ‘And let vs bethink vs on some other place’ (878)
closely followed by ‘Let us bethinke vs on some other place’ (910). Alice
says both ‘How now Adam, what is the newes with you’ (109) and ‘How
now Bradshaw whats the news with you’ (1419), both ‘I know none such,
what meane these questions’ (2407) and ‘I one of them, what meane such
questions’ (2437). Alice asks ‘How now Michaell, whether are you going?’
(146) and Greene asks ‘How now Michael whether are you going?’ (919).
Mosby exclaims ‘These knaues will neuer do it, let vs giue it ouer’ (1822)
and Greene ‘I think we shall neuer do it. | Let vs giue it ouer’ (2001–2).
Franklin urges Arden, ‘Why I pray you sir, let her go along with vs’ (1618)
and, shortly afterwards, ‘Why I pray you let vs go before’ (1639). These
suspicious repetitions, along with the vast majority of the dozens of metrically irregular lines that seem to have been corrupted through recollection
or anticipation of other lines in which the context is similar, cluster in the
early or late scenes, outside 4–9 (MSR lines 1006–596).61
Even the frequency of ‘ay, but’ in the early and late scenes of Arden of
Faversham could conceivably be due to actors or an actor-copyist. The Folio
text of 2 Henry VI contains two examples, whereas the quarto, The First Part
of the Contention (1594), contains eight, only one of which the Folio repeats.
Modern editors supplement their Folio-based texts with one quarto line
that begins with ‘Ay, but’. If we can accept the majority view, upheld by the
Oxford editors in their Textual Companion, that the Contention quarto is a
memorial reconstruction by actors of the text behind the Folio, then actors
6 0. Hugh Macdonald with D. Nichol Smith, eds., Arden of Feversham 1592 (Oxford: OUP for
Malone Society, 1940 [1947]).
61. In ‘Material for an Edition’ I discussed, besides the six substantial repetitions mentioned,
forty-two passages that struck me as corrupt beyond the mere omission or interpolation of
a word such as ‘Why’, ‘and’, or ‘that’ (15–30). Only four of these fall within scenes 4–9. The
proportion of the whole text covered by 4–9 would lead one to expect eleven or twelve of
the combined forty-eight to fall within those scenes, were the corrupt passages randomly
distributed. Of course, I may well have been too zealous in diagnosing textual corruption,
but the distribution, arrived at decades before Craig and Kinney’s research, is noteworthy.
A short list of twelve possible small omissions that make for metrical irregularity shows a
similar pattern, only two falling within scenes 4–9 (15), but in a list of eighteen possible
one-word interpolations that affect the metre the items are distributed more or less in proportion to the lengths of 4–9 and the rest of the play.
Gentlemen,
a r den of fav ersh a m ,
and Collaborations
83
repeatedly turned ‘But’ or ‘Why’ into ‘Ay, but’, evidently employing this
familiar colloquialism to ease transition from the previous speech.62
It is possible, therefore, that faulty transmission of the text of Arden has
somewhat obscured the authorial character of much of the play, while leaving the middle largely unscathed. Although this allows scope for supposing that Shakespeare wrote portions of Arden beyond scenes 4–9, it seems
almost certain that at least one other playwright was involved.63 The idea
of collaboration between Shakespeare and Marlowe stirs the imagination, but in the case of Arden it is, on the Craig–Kinney evidence, unlikely,
since the blocks of text outside the middle section emerge as more akin
to Shakespeare than to Marlowe—and also as more akin to Shakespeare
than to Kyd. Admittedly, the function-word graphs setting Shakespeare
against each of those playwrights in turn show some overlap between
Shakespeare and Marlowe and between Shakespeare and Kyd, but in the
simple Shakespeare-versus-Marlowe comparison—as distinct from the
trial pitting Shakespeare against all comers—the final Arden segment, from
scene 10 to the end, is the most Shakespearean and least Marlovian of all;
which means that, while Shakespeare’s responsibility for this section is
improbable, Marlowe’s is even more so. Kinney suggests that ‘the quest
to find Shakespeare’s partner or partners in the Arden enterprise must look
beyond these two’.64 Unless Shakespeare’s collaborators in his earliest plays,
both inside and outside the First Folio, can confidently be identified, it
remains conceivable that at least some of the material in them that is classed
as ‘non-Shakespearean’ was written by Shakespeare at a stage of his career
before a style recognizable by the Craig–Kinney tests had been formed.
VII
To return in conclusion to Michael Neill’s essay, the ‘language of status’
that he analyses is sprinkled over the whole play—beginning, middle, and
end. Eight of the twenty instances of ‘gentle’ do fall within scenes 4–9,
62. Textual Companion, 175–8.
63.An un-Shakespearean feature of the quarto text is its repeated use of the formula ‘Here
enters’ for entry directions and its persistent use of descriptive directions beginning with
‘Then’, discussed in Chapter 5.
64. Craig and Kinney, 99.
84
Determining the Shakespeare Canon
comprising less than a quarter of the text, but the complete vocabulary
of social class is quite evenly spread, and the preoccupations highlighted
by Neill’s close reading not only drive the action but are woven into the
fabric of the dialogue throughout. If, as seems almost certain, more than
one author participated in Arden of Faversham, collaboration must have been
close, with the co-authors sharing the same grim vision, though one enlivened by humour. The affinities with the Henry VI ‘trilogy’ have often been
noted. Bourgeois rivalries for possessions and prestige, played out within
home, township, and county in Arden of Faversham, expand in the histories
into struggles among the nobility for rule, influence, and power within a
kingdom. The probability that Shakespeare had a substantial share in the
composition of the English theatre’s first extant domestic tragedy is now so
high that Arden of Faversham seems no less worthy than Edward III of inclusion in the canon.65
65. In ‘Notes’ Elliott announces that the Claremont Shakespeare Clinic’s preliminary testing
of Arden of Faversham by their own methods but in the light of the Craig–Kinney results
suggests that scenes 4–7 ‘could be’ by Shakespeare. This pronouncement excludes 8–9, but
I do not myself believe that the Clinic’s data constitute compelling evidence against adding
these scenes to those of Shakespeare’s authorship. See Chapter 1, n. 62. The Clinic’s data
do identify the long scene 14, which all but ends the play, as especially non-Shakespearean
by their measures, a finding that is in line with my own observations recorded above. This
climactic scene in Arden was judged by Wine to be ‘one of the least satisfactory textually’
(xxx). Elliott’s account of my views is inaccurate—I believe that scenes 4–9, at least, are substantially Shakespeare’s—and he misrepresents Kinney’s evidence and conclusions. Kinney
does not find Shakespeare’s authorship of scenes 8 and 9 ‘improbable’: Elliott neglects the
function-word tests and Kinney’s summing up.
4
Parallels and
Poetry: Shakespeare, Kyd, and
Arden of Faversham
I
In Attributing Authorship: An Introduction, Harold Love asserts that ‘literary
quality is a genuine attribute of writing and one that can be recognised. As
such it will be one of the criteria drawn on in conferring or denying attribution.’1 This seems to me sensible. In reaction to the excesses of eighteenthand nineteenth-century scholars eager to foist onto some lesser dramatist
anything of which they disapproved within the plays of the Shakespeare
First Folio, there arose a distrust of any attempts at ‘disintegration’ based
on subjective assessment of merit. Shakespeare, it was insisted, could write
poorly and other playwrights could write well. This is undoubtedly true.
But Shakespeare, even at the beginning of his career as dramatist, was a better poet than Thomas Kyd, for example. The author of The Spanish Tragedy
was a brilliant pioneer of stagecraft and dramatic plot construction, with a
flair for the creation of striking theatrical moments and a sense of how to
shape action to a climax. The verbal medium he devised—an ‘elaborately
patterned rhetoric’,2 in which dialogue can become almost operatic—is
an effective enough instrument, capable of expressing intense emotion.
But it lacks the linguistic subtlety, the lively play of imagery, and the rich
metaphorical content that characterize Shakespeare’s dramatic verse. Even
without making a value judgement, we could nevertheless say that as poetic
1. Harold Love, Attributing Authorship: An Introduction (Cambridge: CUP, 2002), 94.
2.The Spanish Tragedy, ed. Philip Edwards (London: Methuen, 1959), xxvii.
86
Determining the Shakespeare Canon
dramatists Kyd and Shakespeare exhibit different kinds of imagination and
habits of mind.
This chapter aims to show how this crucial difference between Kyd and
Shakespeare helps to assess which of the two playwrights is the more likely
to have written certain passages in Arden of Faversham. That play was, as we
have seen, published in a quarto of 1592, probably having been first performed some time within the period 1588–91. Shakespeare’s earliest plays
are plausibly dated 1590–1.3 Recent scholarship tentatively assigns Kyd’s
The Spanish Tragedy to 1587–8.4 Kyd was also the undoubted author of a
translation from Richard Garnier’s French, Cornelia, composed not long
before Kyd’s death in August 1594.5 A strong case has been made, on internal evidence, for thinking that Soliman and Perseda is also his. Entered in
the Stationers’ Register on 20 November 1592, and published shortly afterwards in an undated large-paper quarto (rather than octavo), it was almost
certainly written after The Spanish Tragedy, but whether as early as 1588 or as
late as 1591 remains in dispute.6
Brian Vickers has revived the theory—popular in the late nineteenth and
early twentieth centuries—that Kyd was the author of Arden of Faversham7 and
in the previous three chapters of the present book a case has been made for
Shakespeare’s part-authorship. Let us, then, consider some passages in Arden
in relation to the contrasting styles of dramatic poetry of the rival candidates.
The most persistent feature of Shakespeare’s language is its concreteness—its
tendency to tie abstractions to physical phenomena, to express thoughts and
feelings through images of objects and actions. Shakespeare is so alert to multiple meanings that imagery is often generated by a kind of wordplay. So in
King John, Austria avers to Arthur that he will not return home
Till Angiers, and the right thou hast in France,
Together with that pale, that white-fac’d shore,
Whose foot spurns back the ocean’s roaring tides
3.Textual Companion, 69–133.
4.Erne, Beyond ‘The Spanish Tragedy’, 58–9.
5.Erne, Beyond ‘The Spanish Tragedy’, 203–16.
6. For the authorship and date of composition of Soliman and Perseda, see Erne, Beyond ‘The
Spanish Tragedy’, 157–67. Freeman, Thomas Kyd, inclined towards a date of 1591–2. In The
Tragedye of Solyman and Perseda, Murray thinks it ‘not unreasonable to conclude . . . that
Kyd reworked his play in its present surviving form around the year 1591’ (xvi). It is listed
under 1590 in Annals. Scholarship on the play was surveyed by Jill Levenson, ‘Anonymous
Plays: Soliman and Perseda’, in The Predecessors of Shakespeare: A Survey and Bibliography of
Recent Studies in English Renaissance Drama, ed. Terence P. Logan and Denzell S. Smith
(Lincoln, NE: University of Nebraska Press, 1978), 230–9.
7. Vickers, ‘Thomas Kyd, Secret Sharer’.
Shakespeare, Kyd, and
a r den of fav ersh a m And coops from other lands her islanders,
Even till that England hedg’d in with the main,
That water-walled bulwark, still secure
And confident from foreign purposes,
Even till that utmost corner of the west
Salute thee for her king.
87
(2.1.22–30)
Here ‘pale’, as a fence or enclosure, leads into ‘coops’ (meaning ‘encloses for
protection’), ‘hedg’d in’, and ‘water-walled bulwark’, but also, as denoting pallor, leads into ‘that white-fac’d shore’. The chalk cliff-face of the
south-eastern coast of England (or ‘Albion’) is thus personified, and so the
moribund metaphor in ‘the foot of a cliff’ can be revived and drawn into
the personification as a human foot ‘spurning back’ the tides.8
Even in Shakespeare’s very earliest plays this mode of operation, in
which punning and imagery merge, is in evidence. For instance, in 2 Henry
VI, Suffolk, trying to convince King Henry of the Duke of Gloucester’s
hypocrisy, says:
Smooth runs the water where the brook is deep,
And in his simple show he harbors treason.
The fox barks not when he would steal the lamb.
No, no, my sovereign, Gloucester is a man
Unsounded yet and full of deep deceit.
(3.1.53–7)
The water metaphor in the semi-proverbial first line leads to the verb ‘harbors’ in the next and gives concreteness and life to the otherwise almost
dead metaphors of the words ‘unsounded’ and ‘deep’ in the last line. Most
writers would refer to depths of deceit without evoking any image, but
Shakespeare makes the depths real by juxtaposing ‘unsounded’—literally
not measured with a plummet, as well as figuratively untried or unexamined. Even ‘The fox barks not when he would steal the lamb’, which seems
unrelated to the nautical imagery, is connected to it by an associational link
in ‘bark’: when Shakespeare uses the word ‘bark’ in any sense, other senses
are apt to be just below consciousness—the bark of a tree, for instance, and,
as here, a sailing vessel (alternatively spelt ‘barque’).
Arden of Faversham displays the same linguistic awareness. For example,
Black Will, thinking of Alice Arden’s promise of payment for acting as her
husband’s assassin, says:
8.This passage was discussed by W. H. Clemen, The Development of Shakespeare’s Imagery
(London: Methuen, 1951), 75. Throughout the present chapter my definitions of words are
taken from OED.
88
Determining the Shakespeare Canon
Why, this would steel soft-mettled cowardice,
With which Black Will was never tainted with.
(3.98–9)
‘Mettle’ is an abstract noun, deriving its meaning ‘courage’ or ‘spirit’ by
metaphor from ‘metal’, of which it is a variant. Here the verb ‘steel’ draws
attention to the original sense: even soft metal would be turned into hard
steel, as the financial reward will harden Will’s resolve. In her valuable book
Shakespeare’s Wordplay, M. M. Mahood noted that ‘although Shakespeare
frequently puns on metal and mettle, there are many places in the plays where
the two words coalesce into one significance’.9 They do so in the following
passage from 2 Henry IV, which affords a line remarkably similar to Black
Will’s. Morton says that report of Hotspur’s death:
took fire and heat away
From the best-temper’d courage in his troops,
For from his metal was his party steeled,
Which once in him abated, all the rest
Turn’d on themselves, like dull and heavy lead.
(1.1.114–18)
The fusion of abstract and concrete is complete and either spelling of the
keyword would serve. From many other instances of such wordplay may be
singled out Romeo’s complaint that Juliet has made him effeminate, ‘And in
my temper soft’ned valor’s steel’ (3.1.115), where ‘temper’ is doubly meaningful, as of course is ‘tempered’ in the 2 Henry IV passage. A search of the LION
electronic database reveals no play of 1580–96 that provides parallels as close
to Will’s ‘steel soft-mettled cowardice’ as do 2 Henry IV and Romeo and Juliet,
while the only other instance of ‘never tainted with’ falls within a scene of 1
Henry VI that has been generally accepted as Shakespeare’s (4.5.46), where it
again refers to cowardice.10
Very common in early modern drama and poetry are allusions to the
‘closet’ of the breast or heart. A closet is a small room, cupboard, or cabinet. As Marlowe employs the analogy in Edward II, it is stock poetic
diction:
My daily diet is heart-breaking sobs,
That almost rents the closet of my heart.
(5.3.21–2)11
9. M. M. Mahood, Shakespeare’s Wordplay (London: Methuen, 1957), 16.
10.In quoting from LION, I modernize spelling and punctuation. For the authenticity of 1
Henry VI, 4.5, see Brian Vickers, ‘Incomplete Shakespeare’, esp. 339, where the ascription of
4.5 to Shakespeare is pronounced ‘uncontroversial’.
11. Christopher Marlowe: The Complete Plays, ed. J. B. Steane (Harmondsworth: Penguin, 1969,
repr. 1975).
Shakespeare, Kyd, and
a r den of fav ersh a m 89
Taken as genuinely figurative language, this is a muddle. Sobs that are consumed (‘diet’) almost tear or split (‘rents’) a closet. Words are being used
loosely, without interacting to create any sensory stimulus to the imagination. In Arden of Faversham cliché is transformed. Will tells Michael:
I am the very man,
Marked in my birth-hour by the Destinies,
To give an end to Arden’s life on earth;
Thou but a member but to whet the knife
Whose edge must search the closet of his breast.
(3.159–63)
The verb ‘search’ here shows that the author is not using the word ‘closet’
in any automatic and unimaginative way. His line is genuinely metaphorical. The searching of a closet blends with the kind of probing that Cassius
commands from Pindarus: ‘with this good sword . . . search this bosom’
( Julius Caesar, 5.3.41–2). Shakespeare twice uses the variant ‘the closure of
my breast’ so as to bring out the full force of the metaphor.12
II
Analysis of a few representative parallels that Vickers cites between Arden
of Faversham and Soliman and Perseda reveals crucial differences in the way
the shared three-word sequences or ‘triples’ (now renamed ‘trigrams’)
are used—differences that point to dissimilar kinds of poetic imagination and hence to separate authors. Among the trigrams that Arden of
Faversham shares with Soliman and Perseda is ‘to everlasting night’, preceded in both plays by the verb ‘to send’. Soliman and Perseda has ‘to send
them down to everlasting night’ (5.2.110) and has earlier used ‘down to
everlasting night’ without the verb (1.1.26).13 Arden has ‘And Arden sent to
everlasting night’ (5.9).
But if we consult the contexts in which the phrase occurs, the verbal
parallel between the two plays appears less significant than the disparity
in poetic quality. In Arden, the hired assassin Shakebag, after a brief and
evocative tribute to the ‘sheeting darkness’ that facilitates such villainy as
he and his accomplice Black Will delight in, concludes that the ‘night’ to
12. Venus and Adonis, 782; Sonnet 48.
13. I quote Soliman and Perseda from Works, ed. Boas, but modernize his old spelling and
punctuation.
90
Determining the Shakespeare Canon
which he is about to consign Arden will be an ‘everlasting one’ (5.1–9). Not
only is it to be under cover of literal night that Arden is dispatched to figurative night, but woven into the fabric of Shakebag’s speech is an opposition
between time and eternity. As Shakebag waits for night (‘In which sweet
silence such as we triumph’, line 5) to drape the earth in ‘the black fold of
her cloudy robe’ (line 3), he says:
The lazy minutes linger on their time,
Loath to give due audit to the hour,
Till in the watch our purpose be complete
And Arden sent to everlasting night.
(5.6–9)
The imagery and thought here are echoed in The Rape of Lucrece. Rapist
Tarquin, experiencing minor delays as he approaches the sleeping Lucrece’s
bed-chamber, takes them
. . . as those bars which stop the hourly dial,
Who with a ling’ring stay his course doth let
Till every minute pays the hour his debt.
(327–9)
The ‘bars’ are the lines on the clock face that mark off the minutes, but they are
also thought of as obstacles. They both punctuate the dial and appear to halt
the movement of the hand, since early modern clocks ‘moved with regular
jolts rather than a smooth movement’.14 ‘Let’ means hinder. In both Arden and
Lucrece, time is felt to slow down so as to postpone the moment at which the
crime (anticipated as fulfilment) is committed. With ‘minutes’, ‘linger’, and
‘hour’ compare ‘ling’ring’, ‘minute’, and ‘hour’. In each case there is also a
commercial metaphor. In Arden the minutes are ‘Loath to give due audit to the
hour’—they are reluctant to render payment so as to square the final account
that is owed (‘due’). Similarly Tarquin is thwarted ‘Till every minute pays the
hour his debt’. ‘Audit’ is a word of which Shakespeare is fond. In Sonnet 126
it is used in connection with time. Although Nature seems to have power to
‘kill’ ‘wretched minutes’ and so prevent the Fair Youth from aging, yet Time’s
‘audit (though delay’d) answer’d must be, | And her quietus is to render thee’
(lines 11–12).
In Shakebag’s speech ‘the watch’ is a period of vigilance or ‘a time division of the night’, but the word also evokes a time-piece or even ‘the marks
of the minutes on a dial-plate’. The word adds to the images of time set
14. William Shakespeare, The Complete Sonnets and Poems (Oxford: OUP, 2002), ed. Colin
Burrow, commentary on Lucrece, 327.
Shakespeare, Kyd, and
a r den of fav ersh a m 91
against ‘everlasting night’. The idea of the minutes accumulating to make
up the hour at which the assassins’ ‘purpose be complete’ is akin to King
Henry’s desire to sit and ‘carve out dials quaintly, point by point’:
Thereby to see the minutes how they run:
How many makes the hour full complete.
( 3 Henry VI, 2.5.25–6)
Arden of Faversham is full of anticipations of Macbeth, and Shakebag’s ‘sheeting darkness’ is well on the way to Lady Macbeth’s ‘blanket of the dark’
(1.5.52). There are, as Fluellen might have said, bedclothes in both, with
their connotations of night-time. But ‘sheeting darkness’ also carries suggestions of winding-sheets. This is OED’s only example of the present participial adjective, meaning ‘swathing, enfolding’. Its earliest example of the
past participial adjective ‘sheeted’ is in Hamlet’s ‘the sheeted dead’ (1.1.115),
where the sheets are shrouds, and its earliest example of ‘sheet’ as a verb
is from Antony and Cleopatra, where ‘snow the pasture sheets’ (1.4.65). So
Shakebag’s speech displays a typically Shakespearean verbal inventiveness.
In Shakebag’s speech ‘Black night hath hid the pleasures of the day’ and
with its ‘sheeting darkness’ conceals the would-be assassins ‘with the black
fold of her cloudy robe’, so that they may send Arden to ‘everlasting night’.
Likewise, in Richard III (1.3.266–8), Queen Margaret harangues Gloucester
as butcher of her sun-like son (‘now in the shade of death’), whose beams
‘thy cloudy wrath | Hath in eternal darkness folded up’. Gloucester ‘turns
the sun to shade’. Neither Kyd nor Marlowe uses ‘fold’ as noun or verb in
connection with darkness. Shakespeare does so not only in Richard III but
also in Venus and Adonis, where ‘the merciless and pitchy night’ did ‘Fold
in’ Adonis, obscuring him from Venus’s sight (821–2), and in The Rape of
Lucrece, where Tarquin’s crime is ‘folded up in blind concealing night’ (675).
Finally, with regard to the context of the trigram ‘to everlasting night’,
when LION is set to search all authors living in the years from 1000 to
1700 for juxtapositions of ‘minute(s)’, ‘hour’, and inflections of the verb
‘to linger’, it finds only two instances besides those in Arden of Faversham
and The Rape of Lucrece—in the anonymous play Mucedorus and in a poem
called ‘The Hour-glass’ by Rowland Watkins in Flamma Sine Fumo (1662).
In Mucedorus (Q 1598, B1r; Wing W1076) the lingering is of a person and is
quite distinct from ‘each minute of an hour’. Watkins has:
The sand within the transitory glass
Doth haste, and so our silent minutes pass.
92
Determining the Shakespeare Canon
Consider how the ling’ring hour-glass sends
Sand after sand, until the stock it spends.
(Flamma Sine Fumo, 1662, E5v)
Here the sand in the glass ‘Doth haste’ and the hour-glass lingers. The ideas
are different from those in the Arden and Lucrece passages, and in any case
the poem was written about seventy years later.15
As poetry, the passage in which ‘to everlasting night’ occurs within
Soliman and Perseda lacks the interest of the Arden lines. Soliman, apostrophizing the dead Erastus, vows that he himself will kill the two janissaries
who strangled him; Soliman’s hand
shall help
To send them down to everlasting night,
To wait upon thee through eternal shade.
Thy soul shall not go mourning hence alone.
(5.2.109–12)
There is no complex interplay of images here, just a straightforward reference to the classical descent into Hades. This is true also of Kyd’s first use
of the phrase. In the chorus involving Love, Fortune, and Death that opens
the play, Death asserts: ‘I will not down to everlasting night | Till I have
moralized this tragedy’ (1.1.26–7).
Another trigram common to Arden of Faversham and Soliman and Perseda
is ‘with eager mood’. The dying Soliman entreats Perseda:
And, sweet Perseda, fly not Soliman
Whenas my gliding ghost shall follow thee
With eager mood thorough eternal night.
And now pale death sits on my panting soul
And with revenging ire doth tyrannize.
(5.4.149–53)
Here ‘eager mood’ takes its place in an assembly of nouns preceded by the
most obvious epithets: ‘sweet Perseda’, ‘gliding ghost’, ‘eternal night’, ‘pale
death’, ‘panting soul’, ‘revenging ire’. The excerpt is a tissue of inert expressions and other men’s inventions.
15. Shakespeare seems to have unconsciously recalled the Arden passage in Cymbeline, 5.5.51–3,
where the words ‘minute’, ‘ling’ring’, ‘time’, ‘purpos’d’, and ‘watching’ echo the vocabulary of Shakebag’s speech, and the construction ‘in which’ (plus noun) recurs.
Shakespeare, Kyd, and
a r den of fav ersh a m 93
There could hardly be a greater contrast than with the rich particularity
of the lines in which ‘with eager mood’ appears in Arden. Greene advises
Black Will and Shakebag as they prepare to attack their victim:
Well, take your fittest standings, and once more
Lime your twigs to catch this weary bird.
I’ll leave you, and at your dag’s discharge
Make towards, like the longing water-dog
That coucheth till the fowling-piece be off,
Then seizeth on the prey with eager mood.
Ah, might I see him stretching forth his limbs
As I have seen them beat their wings ere now.
(9.38–45)
The metaphorical hunting scene has been clearly visualized by the poet
and vividly presented. A ‘water-dog’ is ‘a dog trained to retrieve waterfowl’. Shakespeare refers to a ‘water-spaniel’ (‘a variety of spaniel, much
used for retrieving waterfowl’) in The Two Gentlemen of Verona (3.1.272–
3) and to ‘water-rugs’ (belonging to ‘a shaggy breed of water-dog’) in
Macbeth (3.1.93). The retriever crouches, belly to the ground, longing for
the moment when the bird is shot, and then dashes to seize the prey ‘with
eager mood’—the phrase assumes a certain concreteness in the context.
The eagerness is in the dog’s very nature.
The opening image of ‘smearing . . . twigs with a sticky substance
known as bird-lime to catch birds’16 is one to which Shakespeare frequently
returns. There are three such bird-snaring images in 2 Henry VI: ‘Madam,
myself have lim’d a bush for her, | And plac’d a choir of such enticing birds
| That she will light to listen to the lays, | And never mount to trouble
you again’ (1.3.88–91); ‘And York and impious Beauford, that false priest, |
Have all lim’d bushes to betray thy wings, | And fly thou how thou canst,
they’ll tangle thee’ (2.4.53–5); ‘Like lime twigs set to catch my winged soul’
(3.3.16). There are many other cases: ‘lay lime to tangle her desires’ in The
Two Gentlemen of Verona (3.2.68); ‘Birds never lim’d no secret bushes fear’
(The Rape of Lucrece, 88), whereas ‘The bird that hath been limed in a bush,
| With trembling wings misdoubteth every bush’ (3 Henry VI, 5.6.13–14);
‘they are lim’d with the twigs’ (All’s Well That Ends Well, 3.5.23–4); and so
on. But Kyd also uses the image in The Spanish Tragedy: ‘I set the trap, he
breaks the worthless twigs, | And sees not that wherewith the bird was
limed’ (3.4.41–2).
16. Wine, commentary on Arden, 9.39.
94
Determining the Shakespeare Canon
It is the detailed picture of the water-dog—which reads like the work of
a country man who has experienced hunting for waterfowl at first hand—
that most suggests Shakespeare. For anything comparable in the drama of
the late 1580s and early 1590s we would have to turn to Shakespeare, as in 1
Henry VI, 4.2.45–52:
How are we park’d and bounded in a pale,
A little herd of England’s timorous deer
Maz’d with a yelping kennel of French curs!
If we be English deer, be then in blood,
Not rascal-like, to fall down with a pinch,
But rather, moody-mad; and, desperate stags,
Turn on the bloody hounds with heads of steel,
And make the cowards stand aloof at bay.
The imagery in that speech, of hunting for deer, governs Arden’s narrative
of his dream at 6.6–31.
That Shakespeare had handled a ‘fowling-piece’ is suggested by Robin
Goodfellow’s (Puck’s) account, in A Midsummer Night’s Dream, of the ‘rude
mechanicals’’ reaction to Bottom’s metamorphosis into an ass. When they
catch sight of him:
As wild geese that the creeping fowler eye,
Or russet-pated choughs, many in sort
(Rising and cawing at the gun’s report),
Sever themselves and madly sweep the sky,
So, at his sight, away his fellows fly . . .
(3.2.20–4)
Again we have the fowl, the stalker’s stealthiness (water-dog ‘couching’,
fowler ‘creeping’), the firing of the gun, the beating of wings. In The
Merry Wives of Windsor, Page, Ford, Caius, and Evans are said to have gone
‘a-birding’ (3.3.230, 3.5.45, 3.5.128, 4.2.8) and ‘fowling-pieces’ in the quarto
(1602) become ‘birding-pieces’ in the First Folio (1623) (4.2.58).
Both Arden of Faversham and Soliman and Perseda contain references to ‘a
sudden qualm’. In Kyd’s play, Lucina asks ‘What ails you, madam, that your
colour changes?’ and Perseda replies ‘A sudden qualm’ (2.1.49–50). In Arden,
Franklin asks ‘What ails you, woman, to cry so suddenly?’ and Alice replies
‘Ah, neighbours, a sudden qualm came over my heart’ (14.301–2). Both question and answer share a three-word sequence. But in this case it is worth consulting every instance of the word ‘qualm’ in plays of 1580–96. Some plays use
‘qualm’ without offering anything much by way of parallelism with Arden
or Soliman and Perseda. Greene’s Friar Bacon and Friar Bungay has ‘a qualm did
Shakespeare, Kyd, and
a r den of fav ersh a m 95
cross his stomach then’ (1.18).17 Greene and Lodge’s A Looking Glass for London
and England has ‘to ease a woman when a qualm of kindness come too near her
stomach’ (1100–2). Lyly’s Sapho and Phao has ‘or else a woman’s qualm’ (3.1,
p. 86).
Other plays offer more. Alice’s ‘qualm came over my heart’ is paralleled in ‘a qualm that often cometh over my heart’ of Lyly’s Endymion (3.4,
p. 170) and in an exchange in Shakespeare’s Love’s Labour’s Lost: Katherine
remarks, ‘Lord Longaville said I came o’er his heart, | And trow you what he
call’d me?’, which elicits the Princess’s quip ‘Qualm, perhaps’ (5.2.278–9).
Evidently qualms are apt to come over the heart. In 2 Henry VI, Gloucester
says ‘Some sudden qualm hath struck me at the heart’ (1.1.54). In the ‘bad
quarto’ (1594) of the same play this becomes ‘Pardon, my lord, a sudden
qualm came over my heart’ (A3r).18 It is possible that a reporter corrupted
the text preserved in the Folio through recollection of Alice’s ‘a sudden
qualm came over my heart’. Whatever the reason, the quarto Contention
shares with Arden of Faversham this uninterrupted seven-word sequence.
But it is Gloucester’s use of the verb ‘struck’ that points to the most significant link with a ‘qualm’ in Arden. Gloucester is reading out an agreement that
Suffolk has engineered between King Henry and the French King Charles in
which lands held by the English are to be released to the French. He breaks
off, dropping the paper: ‘Pardon me, gracious lord. | Some sudden qualm
hath struck me at the heart | And dimm’d mine eyes that I can read no further’ (1.1.53–5). In Arden of Faversham, Franklin is urged by Arden to continue
his tale of an adulterous wife. Franklin attempts to excuse himself:
I assure you, sir, you task me much.
A heavy blood is gathered at my heart,
And on a sudden is my wind cut short
As hindereth the passage of my speech.
So fierce a qualm yet ne’er assailèd me.
(9.63–7)
17. This excerpt and those that follow are modernized from LION and may be found by keying
words into the search boxes, but references are to the following editions: Robert Greene,
Friar Bacon and Friar Bungay, ed. Daniel Seltzer (Lincoln, NE: University of Nebraska Press,
1963), Regents Renaissance Drama; Thomas Lodge and Robert Greene, A Looking-Glass for
London and England, ed. W. W. Greg (Oxford: OUP, 1932), Malone Society Reprint; Sapho
and Phao and Endymion, in The Plays of John Lyly, ed. Carter A. Daniel (Lewisberg: Bucknell
UP; London and Toronto: Associated University Presses, 1988); Robert Greene, The
Scottish History of James the Fourth, ed. Norman Sanders (London: Methuen, 1970). Daniel’s
edition lacks line numbers, so page references are given.
18. Wine, commentary on Arden, 14.301–2.
96
Determining the Shakespeare Canon
In both passages the qualm strikes or assails, is sudden, and affects the heart.
The attack stops Gloucester from continuing to read aloud and Franklin
from continuing to narrate. In one further mention of a qualm in the drama
of 1580–96, Greene writes ‘A sudden qualm | Assails my heart’ in James IV
(5.1.61–2). Here is the same verb that Franklin employs in Arden. I doubt
that we can possibly establish relationships of influence or agency among all
these early modern qualms, though the parallel between Franklin’s qualm
in Arden of Faversham and Gloucester’s in 2 Henry VI is particularly close. But
the crucial matter to which I want to draw attention is the Shakespearean
way in which ‘fierce’ and ‘assailèd’ interact in Franklin’s ‘So fierce a qualm
yet ne’er assailèd me’. 2 Henry VI has ‘struck’ and James IV even has ‘Assails’.
But the addition of the simple adjective ‘fierce’ brings out the latent metaphor in ‘assailèd’: the two words act upon each other to create a line with
real metaphorical life.
To examine one further parallel, Crawford and Sykes, arguing for Kyd’s
authorship of Arden of Faversham, compared ‘forge distressful looks’ in Arden
(8.56) with ‘forge alluring looks’ in Soliman and Perseda (2.1.117). Since
LION yields no further instances of the forging of looks in English drama
of the period 1576–1642, from the opening of The Theatre to the closing of
the theatres, there is almost certainly a connection between these two. But
studying them in context leads to the conclusion that identity of authorship
is most improbable.
In Arden of Faversham, Mosby, with characteristic egotism, accuses the
sighing Alice of feigning misery only to make him feel miserable too:
Ungentle Alice, thy sorrow is my sore;
Thou know’st it well, and ’tis thy policy
To forge distressful looks to wound a breast
Where lies a heart that dies when thou art sad.
(8.54–7)
The author of these lines in Arden is alive to the language he is using. The
implications of the latent metaphor in ‘forge’ (a blacksmith forges weapons)
are fully appreciated, and carried on in ‘wounds’ and ‘dies’. This strand of
imagery begins in the earlier lines of Mosby’s speech when he compares
Alice’s sighs to ‘a cannon’s burst | Discharged against a ruinated wall’—sighs
that break his ‘relenting heart in thousand pieces’ (51–3); and the word ‘policy’
contributes to the thread with a hint of military strategy. Shakespeare similarly exploits the underlying, primary sense of the verb ‘forge’ in Sonnet 137:
Why of eyes’ falsehood hast thou forgèd hooks,
Whereto the judgement of my heart is tied?
(lines 7–8)
Shakespeare, Kyd, and
a r den of fav ersh a m 97
The fashioning of hooks from metal (probably with the aid of beams flashing from the eyes) is implicit in this intricate image of false beguiling.
In Soliman and Perseda, Perseda exclaims:
Ah, how thine eyes can forge alluring looks
And feign deep oaths to wound poor silly maids.
(2.1.117–18)
Kyd (if it be he) uses ‘forge’ prosaically to mean ‘contrive’ with no interest in its derivation. The word ‘wound’ is present but fails to connect with
any other word to generate a metaphorical charge in the lines. How, one
may ask, can eyes feign oaths, let alone deep ones? The playwright responsible for Soliman and Perseda did not trouble himself with such questions.
The author of Arden of Faversham, 8.54–7, awakes the dormant metaphors
in words; the author of Soliman and Perseda bundles together his ‘alluring
looks’, ‘deep oaths’, and ‘silly maids’, and lets the metaphors sleep.
III
In illustration of the Shakespearean vitality of much of the language of
Arden of Faversham, I cited the way that the verb ‘steel’ brings out the ‘metal’
in ‘soft-mettled’ in Black Will’s avowal of his resolve to kill Arden. His
speech continues:
I tell thee, Greene, the forlorn traveller
Whose lips are glued with summer’s parching heat
Ne’er longed so much to see a running brook
As I to finish Arden’s tragedy.
(3.100–3)
Not only does the language have a vivid concreteness rare in Kyd’s plays,
but it contains three significant links to early Shakespeare. The phrase
‘summer’s parching heat’ is found in 2 Henry VI, which has ‘In winter’s cold
and summer’s parching heat’ (1.1.81). It appears in no other play of 1576–
1642, and the whole of LION English drama contains only one eighteenthand one nineteenth-century echo of the phrase.19 In the very next line ‘a
19. Shakespeare’s The Rape of Lucrece has the line ‘That knows not parching heat nor freezing
cold’ (1145). George Lillo echoes Shakespeare in The London Merchant (1731), with ‘the summer’s parching heat and winter’s cold’ (2.11.68–9) and Alfred Austin borrows ‘Summer’s
parching heat’ in his Savonarola (1881), 1.1, p. 15. References are to George Lillo, The London
Merchant, ed. William H. McBurney (Lincoln, NE: University of Nebraska Press, 1965) and
to Savonarola (London: Macmillan, 2nd edn. 1891).
98
Determining the Shakespeare Canon
running brook’ is shared with The Taming of the Shrew (Ind. 2.50) and with
no other play of 1576–1642. Indeed, the phrase appears in only one other
LION work—whether poetry, drama, or prose—before the nineteenth
century, and that is John Studley’s translation of Seneca’s Medea, a closet
drama of 1566. Thirdly, ‘Whose lips are glued’ has its counterpart in ‘That
glues my lips’ in 3 Henry VI (5.2.38). This image of glued lips constitutes a
third Shakespearean link that is unique in early modern drama. And the
three links—within two consecutive lines—are to three of Shakespeare’s
first four plays, according to the Oxford Textual Companion, which dates
them 1590–1.
Moreover, the Arden lines have close parallels in Venus and Adonis, written in 1592–3 and published in 1593. Venus eagerly anticipates a kiss from
Adonis: her ‘lips were ready’ but he ‘turns his lips another way’ (89–90). Then
‘Never did passenger in summer’s heat | More thirst for drink than she for
this good turn’ (91–2). In both Arden and Shakespeare’s poem the ‘lips’ of
the traveller (or ‘passenger’, which means traveller) thirst for drink in ‘summer’s . . . heat’. The other element in the Arden image is present in a later stanza
of Venus and Adonis, where Venus presses her ‘thirsty lips’ (543) on Adonis’s
mouth, until the pair, ‘Their lips together glued, fall to the earth’ (545).20
The Shakespearean linguistic alertness typified by the juxtaposition of
‘fierce’ and ‘assailed’ in relation to a ‘qualm’ thus appears within a speech
that not only has a Shakespearean poetic vigour but displays striking parallels with Shakespeare’s earliest works.
One more example of Shakespearean wordplay-cum-image-making in
Arden of Faversham seems worth noting, since it includes a submerged pun
found also in Macbeth, which, as mentioned above, Arden often anticipates.21
Quarrelling with Alice, her lover Mosby protests that, in carrying on an
affair with her:
I left the marriage of an honest maid
Whose dowry would have weighed down all thy wealth,
Whose beauty and demeanour far exceeded thee.
This certain good I lost for changing bad,
And wrapped my credit in thy company.
20. Furthermore, the word ‘parch’ (in any form) is found near ‘lips’ in no other LION play of
1576–1642 except King John: ‘parched lips’ (5.7.40).
21. Likenesses were discussed by Percy Allen in a chapter on ‘Arden of Feversham and Macbeth’
in his Shakespeare, Jonson, and Wilkins as Borrowers (London: C. Palmer, 1928), 19–20, and,
more fully, by Robert F. Fleissner, ‘ “The Secret’st Man of Blood”: Foreshadowings of
Macbeth in Arden of Faversham’, University of Dayton Review, 14 (1979–80), 7–13. See also
Jackson, ‘Shakespearean Features’, 279–304, at 288 n.17.
Shakespeare, Kyd, and
a r den of fav ersh a m I was bewitched—that is no theme of thine!—
And thou unhallowed hast enchanted me.
But I will break thy spells and exorcisms
And put another sight upon these eyes
That showed my heart a raven for a dove.
Thou art not fair—I viewed thee not till now;
Thou art not kind—till now I knew thee not.
And now the rain hath beaten off thy gilt
Thy worthless copper shows thee counterfeit.
99
(8.88–101)
Mosby’s character is here reflected in the imagery he uses. His concern with
the ‘dowry’ and the ‘wealth’ that he has lost reappears in puns on ‘changing’,
‘credit’, and ‘company’, and in the final image in which he compares Alice
to a counterfeit gold coin. ‘Credit’ is good reputation in general but also has
specific application to financial credit-worthiness or solvency; ‘company’ is
both society and a business concern. Later in the play the commercial sense
of ‘company’ draws out the metaphorical content of ‘purchase’: ‘Your company hath purchased me ill friends’ (14.209).22 Shakespeare several times
quibbles on ‘company’.23 The Two Gentlemen of Verona affords a good illustration. Valentine says: ‘Sir Thurio borrows his wit from your ladyship’s
looks, and spends what he borrows kindly in your company’ (2.4.38–40),
and the commercial references are continued in ‘bankrupt’, ‘exchequer’,
and ‘treasure’ (42–4). Shakespeare brings Arden’s ‘credit’ and ‘company’
together in The Merchant of Venice, where Tubal tells Shylock: ‘There came
divers of Antonio’s creditors in my company to Venice’ (3.1.113–14).
The opposing of dove and raven to contrast value and worthlessness or
to emphasize the deceptive nature of appearances is Shakespearean, as is the
image of the gilding of copper to symbolize falseness in love.24 But it is the
22.The process, as regards ‘purchase’, is similar to that in Julius Caesar, when Metellus recommends that the conspirators plotting Caesar’s assassination enlist the support of Cicero,
because ‘his silver hairs | Will purchase us a good opinion, | And buy men’s voices to commend our deeds’ (2.1.144–6). ‘Silver’, primarily an adjective of colour indicating the venerable Cicero’s age, also suggests the precious metal and so revives the moribund metaphor in
‘purchase’ (obtain) and ‘buy’ (secure). Shakespeare uses Alice’s expression ‘purchase friends’
in 2 Henry VI, 1.1.223 (following ‘cheap pennyworths’), Titus Andronicus, 2.3.275, and The
Rape of Lucrece, 963 (where it is followed by the line ‘lending him wit that to bad debtors
lends’).
23. For example, in 1 Henry IV, 2.2.10; Love’s Labour’s Lost, 5.2.224; Much Ado About Nothing,
3.3.59.
24. For raven and dove, see A Midsummer Night’s Dream, 2.2.114; Twelfth Night, 5.1.131; 2 Henry
VI, 3.1.75–6; Romeo and Juliet, 3.2.76; Pericles, 4.Cho.32 (in this last instance a ‘crow’ rather
than a ‘raven’). For gilding copper, see Troilus and Cressida, where Troilus vows fidelity to
Cressida: ‘Whilst some with cunning gild their copper crowns, | With truth and plainness
I do wear mine bare’ (4.4.105–6).
100
Determining the Shakespeare Canon
word ‘wrapt’ to which I want to draw attention. In Shakspeare’s Wordplay,
Mahood notes Banquo’s remarks about Macbeth’s strange behaviour when
he learns that he has been made Thane of Cawdor: ‘Look how our partner’s
rapt’ (1.3.142). As she says, ‘the secondary meaning of “wrapped” is shown
to be in the air by his next words’:
New honors come upon him.
Like our strange garments, cleave not to their mould
But with the aid of use.
(1.3.144–6)25
Mosby’s unconscious wordplay on ‘wrapped’ (meaning involved, implicated, invested)26 works in the other direction, the homophone ‘rapt’
connecting with the subsequent ‘bewitched’, ‘enchanted’, ‘spells’, and
‘exorcisms’.
Mahood quotes Coleridge’s appreciation of Shakespeare’s ‘never broken
chain of imagery, always vivid, and because unbroken, often minute’. As
she explains, illustrating the point by analysis of a passage in Romeo and
Juliet, ‘it remains unbroken because its images are linked by unconscious
wordplay’.27 This characteristic creative activity is extended over twenty
lines in Arden’s complaint to his friend about his wife’s behaviour. Threads
of imagery are interwoven by key words that carry several meanings, which
although not, of course, all pertinent to how the lines are to be understood,
stir associations in Shakespeare’s mind and aid progression and coherence:
No, Franklin, no. If fear or stormy threats,
If love of me or care of womanhood,
If fear of God or common speech of men,
Who mangle credit with their wounding words
And couch dishonour as dishonour buds,
Might ’join repentance in her wanton thoughts,
No question then but she would turn the leaf
And sorrow for her dissolution.
But she is rooted in her wickedness,
Perverse and stubborn, not to be reclaimed;
Good counsel is to her as rain to weeds,
And reprehension makes her vice to grow
25. Mahood, Shakespeare’s Wordplay, 165.
26. Had Shakespeare used any one of these words, it would have been with full consciousness of
its Latin derivation (the rolling up of cloth, folding, clothing). He chooses ‘wrapped’, rather
than a synonym from which the original physical sense has been lost.
27. Mahood, Shakespeare’s Wordplay, 13.
Shakespeare, Kyd, and
a r den of fav ersh a m As Hydra’s head that flourished 28 by decay.
Her faults, methink, are painted in my face
For every searching eye to overread;
And Mosby’s name, a scandal unto mine,
Is deeply trenchèd in my blushing brow.
Ah, Franklin, Franklin, when I think on this,
My heart’s grief rends my other powers
Worse than the conflict at the hour of death.
101
(4.1–20)
The verbs ‘couch’ (‘to cause to germinate, to sprout or spread’) and ‘buds’
begin a train of gardening imagery carried on in a pun on ‘leaf ’, taken up
in ‘rooted’, touched on in ‘reclaimed’ (to reclaim meaning not only ‘to win
back from error’ but also ‘to bring under cultivation’), brought to a climax
in the reference to ‘weeds’ that ‘grow’ and ‘flourish’ by ‘decay’, and making its final appearance in ‘trenchèd’.29 There are at least three other image
threads. The references to ‘speech’ and ‘words’ are continued in ‘couch’
(‘to express in words’) and ‘turn the leaf ’ (as of a book), and it is perhaps
an underlying pun on ‘flourish’ in the calligraphic sense (‘an ornament
of flowing curves in handwriting’) that, following upon ‘counsel’, starts
these associations off again and leads to ‘painted’ and ‘overread’. Another
sequence begins with ‘mangle’ and ‘wounding’ and perhaps ‘couch’ in
its sense ‘to lower a spear or lance to the point of attack’, is just hinted
at in ‘flourished’ (to flourish meaning ‘to wave a weapon’), is renewed in
‘trenchèd’ (a military trench), and concludes with ‘rends’, ‘conflict’, and
‘death’.30 ‘Head’, ‘face’, ‘eye’, ‘brow’, and ‘heart’ are similarly connected.
There are linkages between pairs of words. For instance, ‘stormy’ links
up with ‘rain’. ‘Dissolution’ is an especially apt word, since it means both
‘licentiousness’ and ‘undoing of the marriage bond’, and so provides a sort
of antonym to ‘join’, which impinges on our consciousness as we read
28. The quarto reads, impossibly, ‘perisht’. Some editors, including Wine, emend to ‘plenisht’,
but ‘flourished’ is superior, because it does not require any omitted words to be understood
(replenished itself, or was replenished) and it links up so neatly with the three main strands
of imagery that run through the passage. Shakespeare associates ‘flourish’ and ‘painted’ in
Love’s Labour’s Lost, 2.1.14, 4.3.234–5, and Richard III, 1.3.240, 4.4.82–3.
29.For the analogy of vice to weeds growing and flourishing ‘by decay’, compare Hamlet,
3.4.151–2, ‘And do not spread the compost on the weeds | To make them ranker’, where the
context is again of a woman’s sexual sin.
30.‘Flourished’ is used of a sword in Arden at 14.56. For numerous Shakespearean parallels to Arden’s complaint to Franklin, ‘And Mosby’s name, a scandal unto mine, | Is
deeply trenchèd in my blushing brow’, with similar multiple associations see Jackson,
‘Shakespearean Features’, 284–5. For instance, both the military and agricultural senses of
‘trench’ are operative in Sonnets, 2.1–2.
102
Determining the Shakespeare Canon
‘’join’ (meaning ‘enjoin’). The wordplay, if we can call it that, operates
largely below the level of consciousness, but the main image threads—of
horticulture, words and inscription, violence, and parts of the body—vivify and modify those abstractions that pertain to the emotional, the ethical,
and the theological: fear, love, care, credit, dishonour, repentance, wanton,
sorrow, dissolution, wickedness, perverse, stubborn, reprehension, vice,
faults, scandal, and grief; the mix of simple monosyllables and polysyllabic
Latinisms is itself in line with Shakespearean practice.
Arden’s speech begins with a rhetorical structure scarcely less elaborate
than can be found in the formal set pieces of Kyd’s The Spanish Tragedy.
The opening exclamation forms a type of ploche—‘No, Franklin, no’—
where the vocative intervenes between the repeated ‘no’. Anaphora, parison, and isocolon combine in the balanced ‘If ’ clauses that follow (‘love of
me . . . care of womanhood . . . fear of God’), and anadiplosis governs the fifth
line, where ‘dishonour’ is first the subject and then the object of a verb. But
the overall effect is very different from that of Hieronimo’s great laments.
The interplay of words and images fleshes out the skeleton of rhetoric, so
as to create a dramatic language that is less abstract and stylized than Kyd’s.
Hieronimo’s soliloquy in The Spanish Tragedy at 2.5.1–34, as he discovers
his murdered son, is a theatrical tour de force. It is constructed, in short units
of sense, by means of a torrent of questions, exclamations, and injunctions,
as the father apostrophizes the dead Horatio; by repetition and parallelism
(‘And here within this garden did she cry, | And in this garden must I rescue her’; ‘. . . my sweet son | . . . my son | . . . my son’; ‘O heavens, why . . . |
O earth, why . . . ’), aided by inversion (declarative ‘did she’, ‘must I’); and
finally, over the last ten lines, by resort to the formality of rhyme. Epithets
abound—‘throbbing heart’, ‘trembling fear’, ‘murd’rous spectacle’, ‘sweet
son’, ‘savage monster’, ‘harmless blood’, ‘bloody corpse’, ‘dark and deathful shades’, ‘sacred bower’, ‘wicked butcher’, ‘sweet boy’—but the language
is literal, with scarcely a hint of the figurative. And Hieronimo’s speech
moves to very different rhythms from Arden’s. The patterning of Kyd’s
dramatic verse, here and in Hieronimo’s even longer soliloquy at 3.2.1–
52, seems more overt, as though superimposed on the material. If Arden’s
speech creates an impression of more naturally following a train of thought
and feeling in a real-life situation, this is largely due to the figurative undercurrent that complicates the rhetorical framework.
The truly distinctive qualities of Arden’s speech arise from the author’s
peculiar linguistic awareness, his sensitivity to the multiple meanings of
Shakespeare, Kyd, and
a r den of fav ersh a m 103
words and the range of experience they evoke. As Mahood says of a passage
that she analyses from Henry IV, ‘This is a good example of Shakespeare’s
workmanlike verse, not heightened by striking metaphors or rhythms
and yet animated by a nerve-like intricacy of meaning.’31 No critic has
ever drawn attention to a speech in a sixteenth-century play outside the
Shakespeare canon that manifests such complex organization at the level
where wordplay and imagery blend. And the sources of imagery in Arden of
Faversham are those upon which, early in his career, Shakespeare repeatedly
draws.32
In the present chapter I have concentrated on a handful of phrases shared
between Arden of Faversham and Soliman and Perseda or other early modern
plays. Their treatment in Arden, as opposed to their treatment by Kyd and
others, exemplifies, in a small compass, the essence of the Shakespearean
process described by Coleridge and analysed by Mahood. Arden’s lament
to Franklin affords a longer and more intricate example. There can be no
doubt of the immense value of numerical approaches to questions of authorship. Objective counts of particular features can discriminate between
one playwright and another. But traditional literary-critical methods also
have their place. Vickers’s pioneering employment of plagiarism software
to identify sequences of words that anonymous plays share with Kyd has
considerable potential. But when his results for numbers of three-word
sequences that Arden of Faversham shares with Kyd but not with other plays
of 1580–95 are set against the results of a search of three-word sequences
similarly unique to Arden of Faversham and 2 Henry VI or The Taming of the
Shrew, it turns out that each of the two early Shakespeare plays yields many
more unique trigrams than does any play by Kyd. And analysis of the literary qualities of the contexts in which shared trigrams occur within Arden
of Faversham and Soliman and Perseda reinforces the verdict that Kyd is much
less likely than Shakespeare to have written the pertinent passages in Arden.
Swinburne had no head for statistics, but, as a poet himself, he recognized
that the play contained dramatic verse of Shakespearean distinction.
31. Mahood, Shakespeare’s Wordplay, 22–3.
32. Jackson, ‘Shakespearean Features’, 299–304.
5
Arden of Faversham:
Counter-arguments and
Conclusions
I
Martin Wiggins has argued that Arden of Faversham cannot have been
written by a ‘theatre professional’, least of all an actor-playwright such as
Shakespeare. The author ‘knew the theatre, but he did not really know
how it worked: as a practical stage-writer he was immensely inexperienced,
even naïve’.1
Wiggins notes, as one symptom of the author’s ignorance, that Alice
Arden’s is by far the most substantial role in the play, with more than twice
as many lines as boy actors of the 1580s and 1590s were normally expected
to master. This is an interesting point. In Wine’s edition Alice has 588 lines.
Shakespeare’s Rosalind has 721 in As You Like It (1599–1600) and Cleopatra
has 670 in Antony and Cleopatra (1606).2 Among earlier Shakespeare plays,
the longest female roles are those of Juliet in Romeo and Juliet (1595), with
541, Julia in The Two Gentlemen of Verona (1590–1), with 324, and the
Princess of France in Love’s Labour’s Lost (1594–5), with 290. Clearly the
demands imposed upon the boy playing Alice are not appreciably greater
than those imposed upon the boy playing Juliet, though the size of Alice’s
role would have set the precedent. I can think of no female roles of comparable length in non-Shakespearean plays roughly contemporary with
Arden, except, of course, plays performed by children’s companies, such as
1. Martin Wiggin, ed., A Woman Killed with Kindness and Other Domestic Plays (Oxford: OUP,
2008), 285. Further reference to this edition are incorporated into the text.
2. The counts are taken from Spevack’s Shakespeare concordance.
arden of faversham : Counter-arguments
and Conclusions 105
Marlowe and Nashe’s Dido Queen of Carthage, advertised on the title page
of the 1594 quarto as ‘Played by the Children of her Maiesties Chappell’.
Dido has well over 500 lines to speak, but little personality. In Arden of
Faversham, in contrast, ‘The character portrayal is the play’s outstanding
feature’, as Keith Sturgess writes, and Alice, often seen as a forerunner of
Lady Macbeth, has a complex vitality unmatched in the drama of any of
Shakespeare’s early contemporaries.3
But the distribution of Alice’s lines within the play is noteworthy. It is
in the middle portion of the play, scenes 4–9 (Act 3 in older editions) that
Craig and Kinney’s computational stylistics most surely detect the hand of
Shakespeare. In these scenes 67 out of 567 lines, or 11.8 per cent, are Alice’s.
In the rest of the play 521 of 1,833 lines, or 28.4 per cent, are Alice’s. The
percentage for 4–9 compares with 14.7 per cent for Adriana in The Comedy
of Errors, 14.2 per cent for Julia in The Two Gentlemen of Verona, 10.3 per cent
for the Princess of France in Love’s Labour’s Lost, and 8.2 for Katherina in The
Taming of the Shrew (1590–1). After the first long scene of Arden of Faversham,
Alice makes no appearance during the next six scenes, 2–7, covering 557
lines. Within the ‘Shakespearean’ 4–9 Alice is present only in the brilliant
Quarrel Scene (scene 8) with her lover Mosby. The discrepancy between
the number of lines assigned to Alice in scenes 4–9 and the number assigned
her in the rest of the play, divisions independently established by Craig and
Kinney’s research, seems most unlikely to have occurred by chance.
As Wiggins also points out, the quarto stage directions ‘are very distinctively worded’ (285). Every entrance except the first and ‘Enter Susan’ at 2274
(14.250.1) begins ‘Here enters’, no matter how many characters are coming
on stage, and all the many directions describing what happens begin ‘Then’.
Wiggins infers that ‘in his mind’s eye, the author imagined the action from
the audience’s point of view: he is aware of the doors set in the back wall of
the stage, and twice mentions them (at the heads of Scenes 7 and 12), but
he never thinks beyond them into the backstage area’ (285). The consistent
use of both the ‘Here enters’ and ‘Then’ formulae is unique. It undoubtedly
imparts a literary air. But it is not clear that the directions are authorial. The
descriptive directions may have been introduced by a reporter or by a scribe
preparing the script for publication. Wiggins is sceptical about the theory
that the quarto text bears signs of memorial contamination, regarding
apparent textual flaws as more probably ‘marks of authorial inexperience’
3.Sturgess, Domestic Tragedies, 23.
106
Determining the Shakespeare Canon
(xxxvii). But we have seen that a disproportionate number of passages
(44 of 48) diagnosed almost half a century ago as evincing signs of memorial error fall outside scenes 4–9, where Craig and Kinney find the surest
signs of Shakespeare. If the defects are not errors of transmission, then these
supposed tokens of authorial inexperience must have been concentrated in
the non-Shakespearean parts of the play.
Despite the oddity of the descriptive ‘Then’ directions, there is one other
play-text contemporary with Arden of Faversham that almost invariably
adopts the same formulation, and that is the large-paper quarto of Soliman
and Perseda, entered in the Stationers’ Register on 20 November 1592 and,
although undated, probably printed in that year. Like Arden of Faversham, it
was printed by Edward Allde for Edward White. In two places the Soliman
and Perseda directions even lapse into the past tense: ‘Then they play, and
when she hath lost her gold, Erastus pointed to her chaine, and then she sayd:’
(D3r; five words of Lucina’s speech follow); ‘Then he deliuered her the Chaine.’
(E1r). The compiler of those directions is not only seeing the action from
the audience’s point of view but, when they are in the past tense, recounting what was seen. It seems possible that the texts of both of White’s plays
had undergone similar vagaries of textual transmission. The stage directions may even have been overhauled with readers of the printed quarto
in mind. On the other hand, if Wiggins prefers to believe that the descriptive ‘Then’ directions in Arden of Faversham and Soliman and Perseda came
from the author, then he is driven to accept one of two alternatives: either
(a) the professional playwright Kyd, whose The Spanish Tragedy was one of
the London theatre’s outstanding successes, was responsible for those in
Soliman and Perseda, so that use of the same formula in Arden need not brand
its author as an amateur, or (b) the author of Soliman and Perseda was not
Kyd, as specialists have supposed.
No extant play matches Arden of Faversham in its regular use of ‘Here
enters’ rather than the normal ‘Enter’. The anonymous Fair Em the Miller’s
Daughter (Q undated, but acted by ‘the Lord Strange his servants’, who
played 1589–93) has ‘Here enters’ four times and ‘Here enter’ once (all of
one character only). The Pedlar’s Prophecy (Q 1595) has three times each
‘Here enters’, ‘Enters’, and the orthodox ‘Enter’, as well as the Latin ‘Hic
intra[t]‌Pater’. That anonymous play is dated as early as 1561 in Annals. The
True Tragedy of Richard the Third (Q 1594), also anonymous, has one entrance
beginning ‘Here enters’ and twenty-six with ‘Enters’, besides seventeen
prefixed normally with ‘Enter’. In Greene and Lodge’s A Looking Glass
for London and England (Q 1594) twenty-eight entries begin with ‘Enters’.
arden of faversham : Counter-arguments
and Conclusions 107
‘Here entreth’, with the quasi-archaic verb ending, is common in early
Tudor plays. LION detects 117 examples in seventeen plays dated 1580 or
earlier in Annals, the formula evidently having been normal in moral interludes of the 1560s. The latest, the anonymous Common Conditions, entered
in the Stationers’ Register in 1576 and dated to the same year in Annals, has
‘Here entreth’ twenty-seven times, along with ‘Here enter’ seven times.
Arden of Faversham’s ‘Here enters’ directions seem to draw on a tradition
already going out of fashion by 1592, but it is impossible to know whether
they originated with a reporter, scribe, editor, or author.4
‘At one point’, says Wiggins, the author of Arden ‘does not see the theatre
at all: “Then Shakebag falls into a ditch” (12.20.1). Elizabethan stages were not
ordinarily equipped with ditches: he is evidently thinking in terms of a
real location, a Kentish marsh’ (285). A stage trapdoor would probably have
been used for this episode, but the term ‘trapdoor’ or ‘trap’ is ‘used rarely’
in early modern dramatic texts: ‘usually references are not to a trapdoor but
to the space . . . in fictional terms’, such as ‘cave’, ‘gulf ’, ‘vault’, ‘well’, ‘pit’, or
‘hole’.5 Use of the phrase ‘falls into a ditch’ is consistent with normal usage
in play-texts of the time. Wiggins complains that ‘in terms of practical
staging’ the author has not thought through the problems associated with
Shakebag’s getting ‘almost drowned’ (12.21) and ‘berayed’ (12.55), meaning ‘spattered with mud’, no time having been provided for any changes of
costume (285–6). But a few brown streaks on the face would have sustained
the fiction, and even the words alone might have sufficed. The Elizabethan
theatre placed few limits on make-believe. If Shakebag said he was ‘almost
drowned’, and if Mosby asked him how he came to be ‘so berayed’, then he
was wet and muddy, so far as the audience was concerned. In fact even use
of a trapdoor would not have been essential. The actor playing Shakebag
could simply have fallen to the stage. The Ferryman’s ‘Who is this that’s
in the ditch?’ would have created a ditch. Soliman and Perseda makes much
greater demands on the spectators’ imagination. Soliman orders the Lord
Marshal to ‘hale’ two men ‘to the towers top, | And throw them headlong downe into the valley’. Stage directions follow: ‘Then the marshall
beares them to the tower top’ and ‘Then they are both tumbled downe.’ (H2r).
Elizabethan stages were not ordinarily equipped with towers and valleys.
4. ‘Here’, not followed by ‘enters’ (as it always is in Arden), occurs sporadically in several early
modern plays.
5. Alan C. Dessen and Leslie Thomson, A Dictionary of Stage Directions in English Drama, 1580–
1642 (Cambridge: CUP, 1999; pbk. 2000), 235.
108
Determining the Shakespeare Canon
Any attempt to ‘tumble’ two actors from an upper playing area—such as
‘the top’ that appears to be higher than ‘the walls’ or ‘upper stage’ or ‘gallery’ and serves as a ‘tower’ or ‘turret’ in 1 Henry VI, 3.2.23–30—would
have been extremely hazardous.
The stage direction in Soliman and Perseda about bearing the victims to
the top of the tower echoes the dialogue, and the same phenomenon appears
in Arden of Faversham. ‘First lay the body in the countinghouse’, orders Alice
at 14.249 and the immediately following stage direction reads: ‘Then they
lay the body in the countinghouse.’ The author presumably expected the discovery space to be used as the ‘countinghouse’: Arden’s body is revealed and
brought forth again at 14.327.1–2 and 332.1. A little later Alice’s ‘But first
convey the body to the fields’ (14.350) is followed by ‘Then they bear the body
into the fields.’ But the quarto text hereabouts leaves it entirely unclear which
characters carry away the body, exactly when they do so, and when they
return. This might indicate memorial confusion. Speaking of the stage
directions in the ‘bad quarto’ of Romeo and Juliet (1597), Greg remarked
that ‘it is noticeable how many are derived from the text’.6 Wine judged
scene 14 of Arden to be ‘one of the least satisfactory textually’.7 On the other
hand, John Jowett has demonstrated that many of the directions in the 1597
quarto of Romeo and Juliet originated in the printing-house as means of filling unanticipated space.8
Wiggins notes that ‘when the killers bundle [Arden’s] dead body away
for outdoor disposal’, they leave ‘Alice alone on stage with nothing to say
or do’ (286). This is the case if Mosby, Greene, Michael, and Susan all carry
out the body immediately after the instruction has been given and return
before anything else is said, as they do in most recent editions, including
Wine’s and Wiggins’s. But there is no such problem if the servants Susan
and Michael drag away the body unaided, returning after Mosby, Greene,
and Alice have each spoken two lines (14.351–6), as in Oliphant’s edition,
or even if, as in Hopkinson’s and White’s editions, the carrying out of the
body by all four conspirators is delayed till Mosby and Greene have delivered their admonitory farewells and the servants return after Alice’s two
lines, ‘Now let the judge and juries do their worst; | My house is clear, and
6.W. W. Greg, The Shakespeare First Folio: Its Bibliographical and Textual History
(Oxford: Clarendon Press, 1955), 227. Wine (xxxi–xxxii, n. 5) notes another instance in
Arden: ‘Here enters Adam of the Flower-de-Luce’ at 1.104.1, followed by ‘And here comes Adam
of the Flower-de-Luce.’
7.Wine, Arden, xxx.
8.John Jowett, ‘Henry Chettle and the First Quarto of Romeo and Juliet’, Papers of the
Bibliographical Society of America, 92 (1998), 53–74.
arden of faversham : Counter-arguments
and Conclusions 109
now I fear them not’ (355–6).9 White offers a good justification for taking
this latter option. An argument against Oliphant’s solution is that Franklin,
acting as ‘the first detective in English literature’,10 later points out the incriminating evidence that Arden was carried from the house to the field behind
the Abbey: ‘Backwards and forwards may you see | The print of many feet
within the snow’ (14.394–5). This seems more compatible with four sets of
footprints in each direction than with two. Hopkinson and White solve the
problem most convincingly, but it appears to be textual imperfection rather
than authorial incompetence that has created it.
Another point at which, according to Wiggins, ‘words fail and things
simply come to a standstill’ at a moment of tension occurs ‘in the opening scene when Arden begins to eat the poisoned broth’ (286). This is at
1.360–4, but the episode presents no obstacles to effective performance. As
Wiggins himself observes, ‘it is possible to create an intense, uncomfortable
focus on Arden as the other characters watch him start what two of them
hope will be his last meal on earth’. Yet Wiggins is confident that ‘it is a
tyro author who unwittingly puts such difficulties in his actors’ way’ (286).
Whether or not this is so, and whether or not memorial confusion has
rendered ambiguous the action surrounding the removal of Arden’s body
from inside his house to the Abbey grounds, all the passages that Wiggins
finds problematical occur outside scenes 4–9, which Craig and Kinney
attribute to Shakespeare.
Other editors have thought that the Arden playwright or playwrights
employed doors, discovery space, trapdoor, and a few props in knowledgeable ways, and that he or they took full advantage of the fluidity of
Elizabethan staging.11 Craik drew attention to ‘the way the action will shift,
in the course of a single scene, from outside to inside the house’, changes of
locale being indicated by the dialogue. He concluded that ‘the playwright
left minor points of staging to the actors’, but that ‘his own understanding
of the basic elements of his stage gave them the framework within which to
perform his play’.12 Scores of early modern playscripts leave a modern editor
9.Oliphant, Elizabethan Dramatists; A. F. Hopkinson, ed., Arden of Faversham (London: privately printed, 1890), rev. in Shakespeare’s Doubtful Plays (1898, 1907); White, Arden.
10. Kenneth Tynan, He That Plays the King (London: Longmans, Green, and Co., 1950), 211.
11. White, Arden, asserts that comparison with Holinshed ‘illustrates the playwright’s awareness of the requirements of a play to be performed as opposed to a prose narrative to be
read, and his technical skill in meeting these demands is convincing evidence that whoever
wrote this play had direct, practical experience of the theatre’ (xiii).
12. Craik, Minor Elizabethan Tragedies, xix. Madeleine Doran, Endeavors of Art: A Study of Form
in Elizabethan Drama (Madison: University of Wisconsin Press, 1954, repr. 1964) remarks
110
Determining the Shakespeare Canon
pondering questions about precisely ‘who does what when’. The entries
in Arden at the head of scene 7—where Michael enters ‘at one door’ and
Greene, Will, and Shakebag enter ‘at another door’—and of scene 12—
where Will enters ‘at one door’ and Shakebag ‘at another’—show familiarity with the convention for indicating that characters arrive separately at
the new locale. At the beginning of scene 12, Black Will and Shakebag have
lost each other in the fog; at the beginning of scene 7, Greene, Will, and
Shakebag come across Michael, who has been hoping to avoid them.
There are some inconsistencies in the dramatization of the Arden story.
As Wiggins mentions (285, 299), in scene 8 Bradshaw hands Alice a letter
from Greene (157–9), in which he admits the failure of the London attempt
on Arden’s life that had not yet taken place when Greene had entrusted
the letter to Bradshaw (2.74–7). The Painter Clarke knows at 1.269–71,
but not at 1.620, that Alice is plotting to kill Arden, while the stratagem of
the poisoned crucifix, discussed with Clarke (1.609–34), is never followed
through. Mosby knows who Greene is at 1.295, but seems not to at 1.447. At
3.90–7 Black Will relishes the prospect of being paid by Mosby and Alice
for his services as assassin, but in the play as it stands he has not met them
or heard of their involvement, having been hired by Greene to revenge
Arden’s alleged wrong to him ‘about the Abbey land’ (2.93). Greene at
3.119–23 has information that, while possessed by the audience, has not
been imparted to him in the quarto dialogue. At 3.130 Michael knows
Black Will and Shakebag, whereas at line 35 of the same scene ‘Greene
points Michael out to the two cutthroats, who obviously do not know
him.’13 Anomalies of this kind pass unnoticed in the theatre and are common in early modern plays, especially those on which two or more dramatists collaborated. Some of the above glitches might plausibly be explained
away. Others might be due to memorial confusion. None can confidently
be blamed on a theatre novice’s incompetence. In any case, all involve the
opening scenes or a disjunction between them and 4–9.
Wiggins suggests that the Canterbury-born Christopher Marlowe ‘fits
the Arden author’s profile with uncanny exactitude’, though an ascription to
Marlowe ‘will not do’, since he is ruled out by being a ‘theatre professional’
that ‘In plays such as George-a-Greene, Arden of Feversham, Stukeley, and Faustus, all plays of
the early nineties, there are still traces of that medieval ease with which an actor traversed
miles in plain sight and in a minute or two’ (287). She adds that the most striking use of a
medieval ‘simultaneity’ is at the end of Richard III, where the ghosts of Richard’s victims
appear to Richard and Richmond in their tents set on the two sides of the stage.
13. Wine, Arden, 53.
arden of faversham : Counter-arguments
and Conclusions
111
(284–5). Wiggins believes that the anonymous playwright ‘knew north
Kent intimately: the marshy ground where it tends to get misty and unwary
men risk falling into bogs, the places where roads fork and fellow-travellers
with different destinations must leave one another, the fact that Faversham,
a few miles inland, nevertheless has ready access to the sea’. The play’s
‘action has an unostentatious geographical precision which is usually only
found in plays set in London’ (284). Oliphant had made the same connection
between Marlowe’s Kentish origins and the topographical detail in Arden.14
But, while it is perfectly possible that the author (or one of the authors) of
Arden had travelled between London and Faversham—and it would have
been prudent of a playwright intent on dramatizing Holinshed’s narrative
to have undertaken such ‘research’—specialist knowledge of Kent exhibited in the play could readily have been acquired by reading Holinshed’s
account of Arden’s murder and by the everyday contacts of a man based
in London. Holinshed himself was a Londoner who had been born into a
Cheshire family, but he garnered the lively details about Kentish villages
and terrain that gave his story about events leading up to Arden’s murder an
air of authenticity.
All but two of the places named in Arden are mentioned by Holinshed,
who provides information needed for the play’s accounts of the journey
from Faversham to London and back again. Admittedly, it was the playwright who, enlarging on Holinshed, created in scenes 11 and 12 a ‘foggy
mist’ (12.42) to explain Black Will’s failure to locate and dispatch Arden
on the route between Faversham and the ferry to the Isle of Sheppey. But a
writer would not need to have been born in Kent to be familiar with morning mist and know or guess that the coastal area across from Sheppey was
vulnerable to it. It may well have been Holinshed’s statement that Black
Will ‘mist the way’ that gave the playwright a cue for his fog: Shakebag
exclaims in the quarto, ‘See how the Sunne hath cleard the foggy mist,
| Now we have mist the marke of our intent’ (G2v). Holinshed makes no
mention of ‘Sittingburgh’ (2.61, 17.8), which the play rightly places on the
route between Faversham and Rochester. Moreover, in both Arden contexts there is an additional local detail, as Wine’s commentary records. In
Sittingburgh Will ‘broke the tapster’s head of the Lion with a cudgel-stick’
(2.61–2), and there was a Lion Inn in that village, while in the same soliloquy
in which Will tells us that he was almost arrested at Sittinburgh he declares
14. Oliphant, ‘Marlowe’s Hand’, 89; Elizabethan Dramatists, 297.
112
Determining the Shakespeare Canon
that he intends to take ‘some oyster-boat’ in order to board a ‘hoy’ (or small
coastal vessel) bound for Flushing (17.5–7). The Kentish coast was famous
for its oysters and the oyster grounds nearest to Faversham were especially
esteemed, as Londoners would, however, have known.15 In the same speech
Will boasts that he once robbed the Sittingburgh constable and his man ‘at
Gadshill’, a detail absent from Holinshed. But Gadshill, near Rochester, ‘was
a well-known resort of footpads and highwaymen’, as Wine (quoting Sugden)
notes, and it remains famous as the site where in 1 Henry IV, 2.2, Falstaff robs
and is robbed.
Several of the key place names are mangled in the quarto. Various early
modern sources, including Holinshed, and contemporary maps agree, with
minor variations, on the correct names.16 Q’s ‘Bolton’, where Michael’s elder
brother has a farm (1.172–3), should be Bocton or Boughton, a mile or two to
the south-east of Faversham. Q’s ‘Osbridge’, where, according to the Epilogue
Greene was hanged (Epi. 7), should be Ospring(e) or Osprenge, in the immediate vicinity of Faversham, to the south-west. Q’s ‘Sittin(g)burgh’ (2.61,
17.8) is really Sittingbourne, Sittingborne, Sittingburn, or Syttyngborne.
Q’s ‘Shorlow’ (9.144) is Shornelan in Holinshed and Shorland, Shoreland,
Shurland, or Shovland elsewhere, the last of these variants probably being due
to r/v misreading. Even Q’s ‘Raynum’ (7.18, 28; 9.56, 91) is unique for its ‘um’
ending: other sources have Raynam, Raynham, Rainham, and Reinam.
These manglings of the true Kentish names provide good reason for either
(a) doubting that Arden was written by a denizen of Kent, or (b) accepting
that the quarto text suffers from some significant corruption. Inaccuracy over
names is a feature of the Shakespearean ‘bad quartos’.17
The evidence suggests that one author of Arden may have travelled to
Faversham or spoken with people who had. The town was a common
15.‘Bluff oysters’, from the extreme south of the South Island of New Zealand, are famous
throughout the country.
16.The endpapers of Wine’s edition of Arden reproduce detail from Saxton’s Map of Kent,
Sussex, Surrey and Middlesex, 1575; White, Arden, xxiii, reproduces a map of Kent from
John Speed’s The Theatre of the Empire of Great Britain, 1611; Sturgess, Domestic Tragedies, 290,
reproduces the North Kent coast portion of Philip Symonson’s Map of Kent, 1596. Among
other sources are J. K. Wallenberg, The Place-Names of Kent (Uppsala: Lundequistska
Bokhandeln, 1934); a foldout map in William Lambarde, A Perambulation of Kent (London,
1570, 2nd edn. 1596), facing 70; Edward Jacob, The History of the Port and Town of Faversham
(London, 1774); the website www.hereshistorykent.org.uk; and, of course, Holinshed.
17.The quarto (1600) of Henry V, misrepresenting the Archbishop’s speech about Henry’s
claim to the throne of France, turns the Folio’s ‘the Flouds of Sala and of Elue’ into ‘the
flouds of Sabeck and of Elme’ and ‘th’Lady Lingare’ into ‘the Lady Inger’, besides spelling
‘Pharamond’ as ‘Faramount’ and ‘Faramont’, ‘Meisen’ as ‘Mesene’, and ‘King Pepins Title’
as ‘King Pippins title’.
arden of faversham : Counter-arguments
and Conclusions
113
destination for several theatre companies in the late 1580s and early 1590s,
including the Queen’s Men and Strange’s Men.18 Whatever Shakespeare’s earliest associations with the stage, he is likely to have mingled with actors who
had journeyed between London and Kent.19
II
As first mentioned in Chapter 1, Brian Vickers has revived claims for Kyd’s
authorship of Arden of Faversham, also attributing to him Fair Em the Miller’s
Daughter, The True Chronicle History of King Leir, and a substantial share of 1
Henry VI.20 His method is to use plagiarism software to make paired comparisons between each of these plays (or, in the case of 1 Henry VI, portion of a
play) and each of the plays belonging to the accepted Kyd canon—The Spanish
Tragedy, Soliman and Perseda, and Cornelia, the last having been translated from
Robert Garnier’s Cornélie. Shared sequences of three or more words detected
in this way are then checked for ‘uniqueness’ within a substantial corpus of
roughly contemporary plays. Each of the works claimed for Kyd exhibits a fair
number of three-word sequences (or ‘triples’ or ‘trigrams’) that are in these
terms uniquely shared with Kyd. Vickers regards this as good evidence that
Kyd was the author.
In a detailed critique, I exposed the flaws in this procedure, demonstrating that it yields considerably more trigrams that are peculiar to Arden of
Faversham and Shakespeare’s 2 Henry VI or The Taming of the Shrew than
to Arden of Faversham and any play accepted as Kyd’s.21 Moreover, the
18. Andrew Gurr, The Shakespearian Playing Companies (Oxford: Clarendon Press, 1996, repr.
2001), passim (see index).
19. One scrap of evidence suggests that scene 4 of Arden was written by a playwright not from
Kent but from Warwickshire. Michael imagines the ruffians Black Will and Shakebag with
daggers drawn to stab Arden: ‘Methinks I see them with their bolstered hair, | Staring and
grinning in thy gentle face’ (4.72–3). Although the quarto has ‘bolstred’, OED affords no
warrant for extending the normal sense of ‘supported, propped up’ to cover a ‘stiff, rigid,
bristly’ head of hair, as in Wine’s gloss. Sturgess (Domestic Tragedies) and later editors are
surely right to emend to ‘bolter’d’, meaning ‘matted with blood’ or ‘tangled in knots’. OED
cites the word under the spelling variant ‘balter’. According to Malone, the term was a
provincialism, used in Warwickshire (as noted by Kenneth Muir in his Arden edition of
Macbeth of 1951). It occurs in Macbeth, at a point where Macbeth, like Michael, is seeing an
apparition, which, like Michael’s, is grinning horribly: ‘For the blood-boltered Banquo
smiles upon me’ (4.1.123).
20. Vickers, ‘Thomas Kyd, Secret Sharer’.
21.Jackson, ‘New Research’. In ‘Identifying Shakespeare’s Additions to The Spanish Tragedy
(1602): A New(er) Approach’, Shakespeare, 8 (2012), 13–43, Vickers reports that he and
Dahl have improved their methodology by applying their full series of processes to the
114
Determining the Shakespeare Canon
percentage of shared trigrams that are unique is higher for each of the three
pairs of plays generally agreed to belong to the Kyd canon—The Spanish
Tragedy and Soliman and Perseda, The Spanish Tragedy and Cornelia, Soliman
and Perseda and Cornelia—than for any pair that includes one of the putatively
Kydian plays. This should not be the case if the plays that Vickers supposes to
be by Kyd really are his. These and other reasons for rejecting Vickers’s proposed redefinition of the Kyd canon are fully spelt out in my article.
III
In 2008 a paper by Marina Tarlinskaja, entitled ‘Kyd Canon’ was posted
on the London Forum for Authorship Studies website.22 In it she argued,
on metrical grounds, in favour of Vickers’s proposed expansion of the Kyd
canon. Her conclusions were far from warranted by the evidence she presented. The Forum’s website is under reconstruction and the paper, avowedly a report on ‘work in progress’, cannot currently be viewed. Should it
have swayed any scholars who consulted it, I give here the gist of a detailed
criticism that I made when it first appeared.23
Tarlinskaja is the foremost living analyst of English blank verse.24
Her work on A Lover’s Complaint will be assessed in Chapter 10, Section
I. Prominent in her analyses is the counting and tabulation of the percentages of stressed syllables within the ten different positions in the line.
According to the iambic paradigm, even-numbered syllables (which are
in ‘strong’ positions) receive stress whereas uneven-numbered syllables
(which are in ‘weak’ positions) do not. Employing pre-established linguistic criteria, Tarlinskaja determines how often the theoretical expectations
are actualized in practice. She also compiles data on the positions in the line
at which word boundaries and strong syntactic breaks occur. For her paper
canons of candidates other than the one favoured (34–5). But subjectivity in the identification of ‘matches’ has been increased by the decision to include ‘discontinuous’ trigrams,
where there is no shared three-word sequence, while LION reveals that many supposedly
‘uniquely Shakespearian matches’ are not unique at all. No results of trying out the ‘newer
approach’ on Arden of Faversham have yet been reported.
22.Quotations in this section are from Tarlinskaja’s website article, as posted, without page
numbering, in 2008.
23. The critique may be obtained in full from [email protected].
24.Of special interest to students of Shakespeare is her book Shakespeare’s Verse: Iambic
Pentameter and the Poet’s Idiosyncrasies (New York: Peter Lang, 1987).
arden of faversham : Counter-arguments
and Conclusions
115
on the Kyd canon she also gave tallies of miscellaneous features, such as
pleonastic ‘do’ and examples of ‘-ed’ and ‘-eth’ endings with syllabic value.
Tarlinskaja presented her data in four tables. For various metrical features,
she extrapolated ranges into which she believed plays by Kyd ought to fall
and awarded points to the four disputed plays and to eighteen roughly contemporary control plays or parts of plays by Greene, Marlowe, Nashe, Peele,
Shakespeare, or an unknown writer according to whether their scores did
in fact fall within these limits. However, her construction of Kydian ranges
was so subjective that for certain features canonical Kyd plays themselves
fell outside them, while some of the disputed plays fell within them. Thus
for one of her parameters, ‘unstressed syllable in position 10’, neither The
Spanish Tragedy, Soliman and Perseda, nor Cornelia received a point, but Arden
and Leir, which were outside Kyd’s actual range, were among those that
did receive a point. Several other of Tarlinskaja’s arbitrarily created ranges
excluded one or more of the genuinely canonical Kyd plays.
Such anomalies arose from two causes. The first is that Tarlinskaja often
assumed the correctness of the hypothesis that she was meant to be testing,
as when she wrote of ‘seven texts of the Kyd canon’, although in fact there
are only three. The second is that she made assumptions about Kyd’s development and about the effects of genre on her metrical results, and that,
while these assumptions were those of an expert whose intuitions derive
from the investigation of numerous texts over many years, they were rendered problematic by the limitations of scholarly attempts to establish an
accurate chronology of the plays of the period 1587–94, within which Kyd
was writing for the stage.
The only satisfactory way to use Tarlinskaja’s admirable metrical data for
the purposes of attribution is through mathematical measures of likeness,
such as chi-square fitting and principal component analysis. Application of
these objective modes of analysis demonstrated that, of all the plays analysed by Tarlinskaja, The Taming of the Shrew and Shakespeare’s share of
Titus Andronicus were metrically the most akin to Arden of Faversham. This
does not, of course, constitute good evidence of Shakespeare’s unaided
responsibility for the anonymous play, of which he was almost certainly no
more than a co-author. But we can conclude with some confidence that the
metrical characteristics of Arden of Faversham as a whole do not associate it
more closely with Kyd than with Shakespeare.
Tarlinskaja herself has steadily expanded the scope of her investigations
and reassessed her results. Her further research has persuaded her that some
116
Determining the Shakespeare Canon
of the proposed new attributions to Kyd receive little support from her data
and that certain scenes of Arden, including 4–8, share metrical features with
early Shakespeare.25
IV
One more general objection to attribution studies of early modern English
plays seems worth addressing here. In the course of a debate, in the correspondence columns of The Times Literary Supplement, about identification of Thomas Middleton as a collaborator with Shakespeare upon Timon
of Athens, Stephen Collins alleged that there was ‘a methodological problem with identifying a second author’ in Timon that had been ‘generally
overlooked’. Investigators had, he objected, ignored ‘the possibility of a
contributor who has left no identifiable work and whose name has been
lost’: non-Shakespearean portions of Timon ‘could be by Anonymous’.26
It is true that about half of the plays known by their titles to have been
written within the period 1576–1642 no longer survive, and there must
have been many more that have left no trace whatsoever.27 But this need not
undermine confidence in the results of properly conducted authorship studies. The evidence of Middleton’s hand in Timon of Athens is overwhelming.28
Divers objectively quantifiable Middleton markers serve in combination to
distinguish all his undoubted dramatic writing from all the extant dramatic
writing by other early modern English playwrights. These markers cluster
25. Email communications, July–August, 2013; esp. 12 August. Tarlinskaja’s latest findings will
be reported in an important book forthcoming from Ashgate.
26. Letters (always p. 6) in The Times Literary Supplement, by Brian Vickers 3, 17 August 2012;
Juliet Fleming, 10, 24 August; Stephen Collins, 31 August; Stanley Wells, 7 September.
Vickers and Wells had the advantage over their opponents of having read and fully understood the relevant scholarship.
27. I have made use of the Lost Plays Database (www.lostplays.org), for which the co-ordinating
editors are Roslyn L. Knutson and David McInnis, to whom I am grateful for helpful
answers to my queries.
28. A fully convincing case was made by R. V. Holdsworth, ‘Middleton and Shakespeare: The
Case for Middleton’s Hand in Timon of Athens’ (PhD thesis, University of Manchester,
1982). Brian Vickers had not been able to gain access to this when compiling his survey
in Shakespeare, Co-Author, 244–90, which summarizes the evidence of Jackson, Lake, and
many others. Recent editions of Timon of Athens in the Oxford (ed. John Jowett, 2004) and
Arden (eds. Anthony B. Dawson and Gretchen E. Minton, 2008) series accept that
Middleton was a co-author of the play.
arden of faversham : Counter-arguments
and Conclusions
117
in scenes and passages of Timon that early twentieth-century scholars had
picked out, largely on impressionistic grounds, as probably by Middleton.
Moreover, a comprehensive search for verbal parallels between Timon and
either Middleton or Shakespeare yields a preponderance of Middleton parallels with the putatively Middletonian scenes and passages and a preponderance of Shakespeare parallels with the putatively Shakespearean scenes
and passages. ‘Anonymous’ could have composed the Middleton scenes only
if, unlike the writers of all surviving plays, he was indistinguishable from
Middleton. Such an anonymous playwright would have to have formed
an association with the King’s Men around 1606, as had Middleton.29 And
despite having now sunk from sight, he would have to have been an experienced professional, capable of matching Middleton’s acknowledged brilliance in writing the kind of satirical comedy that so deftly exposes the
unctuous hypocrisy of Timon’s false friends in Act 3. The notion that there
existed such an ‘Anonymous’ is scarcely worth even entertaining.
The point with broader application concerns samples and parent populations. The extant plays of 1576–1642 constitute a very large sample (about
700) of all those that were written, and a large sample can, within a slight
margin of error, provide trustworthy information about the full population. If no extant play in which Middleton had no part contains a distinctively Middletonian set of quantifiable features, the probabilities are high
that no non-extant play ever did so either. In fact, if a lost play were to be
found without any author’s name attached but bearing the Middletonian
markers, we would be justified in attributing it to Middleton.
Similar reasoning holds with regard to Arden of Faversham. Might it be
the work of a playwright whose other plays have perished? The reasonable
answer is that the possibility is remote. The reliability of a sample depends
on its size and its representativeness. Its members should be randomly
selected. In other words, every member of the parent group ought to have
the same chance of being included in the sample. In the case of extant early
modern English plays constituting a sample of all those created, this condition is obviously not met. But the bias in Time’s winnowing of survivors
is such as to decrease, rather than increase, the chances that an otherwise
unknown playwright was responsible for the Shakespearean characteristics
of Arden of Faversham.
29. Thomas Middleton and Early Modern Textual Culture: A Companion to The Collected Works, gen.
eds. Gary Taylor and John Lavagnino (Oxford: OUP, 2007), 361.
118
Determining the Shakespeare Canon
Printed publication, which creates multiple copies, obviously boosts the
likelihood that the text of a play will survive, so the sample composing
all extant plays contains a larger proportion of playtexts that, like Arden,
reached print. These are, in turn, likely to have included a larger proportion of plays that had been theatrical successes, so that, as a body, extant
plays are more comparable to Arden than lost ones in this respect.
Admittedly, Love’s Labour’s Won is the only likely lost play by
Shakespeare, so that his oeuvre suffered very much less than average attrition. But the survival of so many of Shakespeare’s plays provides equal scope
for positive or negative results in the various kinds of authorship tests. For
instance, in LION searches for Arden phrases and collocations occurring
in not more than five other plays of a stipulated period, all plays of similar
length have, statistically speaking, the same chance of affording examples,
regardless of the sizes of the extant authorial corpora. If somebody other
than Shakespeare wrote the Quarrel Scene in Arden—somebody whose
output is not adequately represented in the surviving plays—one would not
expect eight plays by Shakespeare to each yield more links than any play
by another author, or 3 Henry VI to top the list with twenty-two, nearly
three times as many as any non-Shakespeare play. The results would surely
have been more ambiguous, muddled, and inconsistent. As we have seen,
when sample speeches from Marlowe’s Doctor Faustus were tested by LION
searching, Marlowe and Nashe’s Dido Queen of Carthage afforded the greatest number of links per play and Marlowe’s 1 Tamburlaine came second.
The relative smallness of Marlowe’s dramatic output did not prevent this
striking contrast with the results for the Quarrel Scene. Had Dido and 1
Tamburlaine been lost, the result for the Doctor Faustus passages would have
been less decisive, but the pertinent point is that it would not have erroneously supported an attribution to Shakespeare.
As for unknown playwrights, their number must be very small and
the number of those capable of writing anything resembling the central
scenes of Arden of Faversham much smaller. In Palladis Tamia (1598) Francis
Meres named twenty-eight English playwrights besides Shakespeare as
the ‘best for Tragedy’ and ‘the best for Comedy’.30 Most names are still
well known: Chapman, Chettle, Dekker, Drayton, Greene, Hathway,
Heywood, Jonson, Kyd, Lodge, Lyly, Marlowe, Nashe, Peele, Porter,
30.Francis Meres, ‘Comparative Discourse’, Palladis Tamia (1598), 279a–87a (2N7 r–2O7r);
reprinted in The Riverside Shakespeare (rev. 1997), 1970.
arden of faversham : Counter-arguments
and Conclusions
119
Shakespeare, and Wilson. Meres also pays dutiful deference to noblemen,
academics, clergymen, and other dignitaries responsible for translations,
closet dramas, university or Inns of Court plays (several of which were in
Latin), and masques and entertainments.31 These men—Thomas Sackville
(Lord Buckhurst), Thomas Legge, Richard Edes, George Ferrers, Edward
De Vere (Earl of Oxford), William Gager, Richard Edwards, and George
Gascoigne—are either impossible or next-to-impossible candidates for
the authorship of Arden of Faversham. Three of them had died well before
Arden was composed: Edwards in 1556, Gascoigne in 1578, and Ferrers in
1579. We have at least some dramatic material from all twenty-nine authors
except the politician Ferrers and the courtier De Vere. From Edes there
survive only manuscript fragments of a court entertainment of 1592. For
Michael Drayton and Richard Hathway there remain only indeterminate
shares in 1 Sir John Oldcastle (dated 1599, published 1600). The sole extant
dramatic work of the poet Watson—also included in Meres’s list of the ‘best
for Tragedy’—is a Latin translation (1581) of Sophocles’ Antigone.
Among those writers from whom nothing or very little survives,
Drayton and Hathway were both among collaborators in several lost plays
for the Admiral’s Men. Decades later than Meres, Thomas Heywood,
in his Hierarchy of the Blessed Angels (1635), named Watson in the company of Beaumont, Dekker, Fletcher, Ford, Greene, Kyd, Marlowe, May,
Middleton, Nashe, Shakespeare, and Webster.32 Heywood was joking about
how eminent writers were familiarly known as Frank, Tom, Jack, Robin,
Kit, and Will, but Watson is unlikely to have been admitted to the group of
playwrights solely because his name was Tom. Although Watson ‘thought
of himself primarily as a Latin writer’ and was also hailed as an English
Petrarch for his sequence of ‘love passions’, Hekatompathia (1582), and the
sixty-one sonnets of The Tears of Fancy (1593), devising ‘twenty fictions and
knaveries in a play’ was his ‘daily practice and his living’, according to his
employer William Cornwallis.33 Further, in The Knight’s Conjuring (1607)
Dekker pictures as seated in Elysium in the company of ‘learned Watson,
industrious Kyd, [and] ingenious Atchlow’ the Queen’s Men’s chief tragedian John Bentley, ‘a player moulded out of their pens’.34 Dekker evidently
believed that Watson wrote plays for the Queen’s Men, set up in 1583.
31. Statements about the writers and their works derive from Annals and the Lost Plays Database.
32. Heywood, Hierarchy (Q 1635), S1v.
33. Albert Chatterley, ‘Thomas Watson’, Oxford Dictionary of National Biography, online version.
34. Thomas Dekker, A Knight’s Conjuring (Q 1607), K4v–L1r.
120
Determining the Shakespeare Canon
Moreover, he implies that so did Thomas Achelley, to use the more
common spelling. Nashe extolled Achelley as a poet in his ‘Address to the
Gentlemen Students of Both Universities’ prefaced to Greene’s Menaphon
(1589), noting his ‘deep-witted scholarship’, and Meres groups him with
English poets comparable to the Italians: Roydon, Watson, Kyd, Greene,
and Peele. 35 Achelley wrote a twelve-line commendatory poem for
Watson’s Hekatompathia.36 There survives only one substantial work of his
in English—The Tragical History of Didaco and Violenta (1576), a long poem
adapted from William Painter’s The Palace of Pleasure and composed in heavily alliterative fourteeners, split after the eighth-syllable caesura.37 But ten
verse fragments that, in England’s Parnassus (1600), Robert Allott attributed
to Achelley have been accepted as his.38 Totalling seventeen lines, they are
all in iambic pentameters. Mostly sententious, in accord with Allott’s purposes, they include three rhymed couplets. It is not impossible that Allott
took some of them from plays, though all might equally well have come
from poems.
Drayton, Hathway, and even Watson and Achelley may have written, or
collaborated on, lost plays that are in some ways akin to Arden of Faversham.39
But Meres’s list (and to a lesser extent Heywood’s) does suggest that extant
plays and known dramatists comprise a sample that is unlikely to mislead
us about features that serve to identify the hand of Shakespeare and distinguish it from that of his contemporaries.
The crucial point, however, is the same for Arden and Shakespeare as for
Timon and Middleton. ‘Anonymous’, or somebody whose dramatic output
is lost or fragmentary, could have composed the middle scenes of Arden
only if, unlike the authors of all surviving plays, his writing was indistinguishable from Shakespeare’s in the many ways discussed in Chapters 1–4
and summarized below. This is a case for Occam’s razor. It is unnecessary to
postulate the existence in 1588–91 of such a Shakespeare clone.
35. Meres, Palladis Tamia, 282b (2O2v); Menaphon, STC 122762, A2v.
36. Arthur Freeman, ‘The Writings of Thomas Achelley’, The Library, 25 (1970), 40–2; and the
opening chapter, ‘Kyd’s Early Life and Circle’ of Freeman’s Thomas Kyd: Facts and Problems
(Oxford: Clarendon Press, 1967), 1–24, where Kyd, Watson, and Achelley are considered
together.
37. STC 1356.4.
38.They are listed by Freeman, ‘Writings’, 41–2, with references to Charles Crawford, ed.,
Englands Parnassus (Oxford: Clarendon Press, 1913).
39. However, in order to have had anything to do with Arden of Faversham, both Drayton and
Hathway would have had to begin playwriting several years before we have any record of
their doing so: Hathway’s first known (though lost) dramatic writing belongs to 1595 and
Drayton’s to 1597.
arden of faversham : Counter-arguments
and Conclusions 121
V Summary and Conclusion
The case for Shakespeare’s part-authorship of Arden of Faversham rests on the
convergence of diverse kinds of evidence upon the same conclusion. The
pivotal Quarrel Scene—the psychologically trenchant, theatrically riveting scene 8—has far more links in rare phrases and collocations to early
Shakespeare plays than to plays by other dramatists of the period 1580–
1600. The method of investigation that established this predominance
set all potential authors on an even footing: the searches might equally well
have found a majority of links with another playwright, but they did not.
The same technique of LION searching yields a preponderance of links
with early Shakespeare for Arden’s narrative of his dream (6.6–31). These
methodically accumulated data are reinforced by some remarkable parallels—involving whole clusters of shared words and images—between these
parts of Arden of Faversham and Shakespeare’s Venus and Adonis, The Rape
of Lucrece, 2 Henry VI, 3 Henry VI, and Love’s Labour’s Lost. Chronology
appears to rule out imitation of Shakespeare by some other author.
Can we, however, fully allay any suspicions that influence may have
operated in the other direction? Is it possible that, without having written
a word of Arden, Shakespeare was so impressed by such episodes as Alice’s
quarrel with Mosby and Arden’s account of his dream that, whether consciously or unconciously, he repeatedly echoed them? His mother’s maiden
name had been Arden, after all, and he could hardly have failed to take an
interest in a play starring villains whose sobriquets were Black Will and
Shakebag. Of course these are also reasons why Holinshed’s story of Arden’s
murder—in which the names were already present—might have caught his
attention as subject for a play of his own, at least as co-writer. But might an
Arden of Faversham entirely by some other playwright or playwrights have
captured his imagination, in ways that not even Marlowe’s Doctor Faustus
had done?
Unless the Oxford Textual Companion’s chronology is seriously wrong,
Arden’s appearance in print post-dated the composition of Shakespeare’s
earliest plays, so that the intimate familiarity with Arden postulated by any
theory that Shakespeare was heavily influenced by it could only have been
acquired through the theatre. Perhaps Shakespeare had absorbed the script
as a member of the cast? There are obstacles to accepting this hypothesis.
As explained in Chapter 2, Section II, Mosby is alone on stage for the first
122
Determining the Shakespeare Canon
forty-four lines of the Quarrel Scene and Alice is the sole character to join
him until the very end, while only Franklin and Arden are present for
Arden’s narrative of his dream in scene 6. Therefore Shakespeare could
not have played a role in both scenes. So we would have to suppose that he
imbibed an anonymous playwright’s words through hearing them spoken
by his colleagues while he waited in the tiring house or that he had assumed
some special responsibility as prompter or ‘bookholder’.
We do not know what company first staged Arden of Faversham. A. S.
Cairncross, believing that The First Part of the Contention (1594) and The
True Tragedy of Richard Duke of York (1595) had been reconstructed by
actor-reporters from their memories of 2 and 3 Henry VI as they were eventually printed in the First Folio (1623) and that Richard III (1597) and Romeo
and Juliet (1595) were likewise ‘bad quartos’, listed lines in which the actors
seemed to have misrepresented the true text by confusion with similar
lines in other plays in their repertoire. He concluded: ‘[I]‌t is . . . likely that
Pembroke’s, under whatever name or with whatever organization, existed
before 1592, probably as early as 1589, and that it was then Shakespeare’s
company, as it was, for a time at least, Kyd’s and Marlowe’s.’40 He deduced
that all three Parts of Henry VI, Richard III, Titus Andronicus, Edward III,
Soliman and Perseda, and Arden of Faversham had belonged to Pembroke’s
Men. Jackson, adding to Cairncross’s evidence, cited eleven instances in
which 1 Contention or Richard Duke of York printed lines in which recollections of Arden seemed to have caused deviations from the Folio wording.41
But Cairncross’s case rests on the assumption that the texts of 1 Contention,
Richard Duke of York, and the ‘bad quartos’ of Richard III and Romeo and
Juliet really are ‘memorial reconstructions’ by actors. The near-consensus
obtained in Cairncross’s day on this point has since dissolved.42
Both Shakespeare and Arden of Faversham may have belonged to Lord
Strange’s Men before Pembroke’s was formed, and Shakespeare may have
known the domestic tragedy before he wrote The Two Gentlemen of Verona
40. A. S. Cairncross, ‘Pembroke’s Men and Some Shakespearian Piracies’, Shakespeare Quarterly,
11 (1960), 335–49, at 349; also Cairncross’s Arden edition (1957) of 2 Henry VI, 182–5, and
Arden 3 Henry VI (1964), 180–4.
41. Jackson, ‘Material for an Edition’, 293–302. The relevant Arden lines (MSR numbering) are
19, 74, 186, 347, 470 (twice), 1042–3, 1111–13 and 1155, 1827, 2257–8, and 2331. The first six
are in scene 1, four others in scenes 3, 12, and 14 (two items). One rather doubtful instance
spans scenes 4 and 5.
4 2. Maguire, Shakespearean Suspect Texts, labels both The Contention and Richard Duke of York
‘not M[emorial] R[econstruction]’ (238, 320).
arden of faversham : Counter-arguments
and Conclusions 123
and The Taming of the Shrew and collaborated on 2 and 3 Henry VI.43 But it is
scarcely credible that the results exhibited in Table 1.1—where, among the
thirty plays of 1580–1600 that share at least four rare phrases and collocations
with the Quarrel Scene eighteen are, wholly or in part, by Shakespeare,
as are 150 of the 210 actual links—reflect nothing more than the influence on an impressionable young Shakespeare of a fine scene in another
man’s play. This is a huge disproportion, when Shakespeare was author or
co-author of only twenty-three of the 134 plays searched, about 16 per cent.
To explain away the figures in terms of influence would be to credit a single
scene by an unknown playwright with virtually inventing the distinctively
Shakespearean idiolect, with its many vivid images from nature.
But the coup de grâce to any such explanation is dealt by computational
stylistics. Systematic LION searching of scene 8 and of 6.6–31 is not the
only technique that discovers Shakespeare’s hand in Arden of Faversham.
Craig and Kinney’s computer-aided stylometric tests, validated on plays
of known authorship, classify the middle of Arden, corresponding to Act 3
in nineteenth- and early twentieth-century editions, as wholly or largely
Shakespeare’s. Both scene 6 and scene 8 fall within this section. Craig
and Kinney use two independent types of test—of lexical words and of
high-frequency function words—and these are in broad agreement over
Shakespeare’s share of Arden. A 2,000-word block formed of scenes 8 and
9 proved to be quintessentially Shakespearean in its rates of function-word
use. Frequencies of ‘and’, ‘all’, ‘but’, ‘most’, ‘very’, and so on are not easily
susceptible of imitation. The ability of Craig and Kinney’s methods to classify correctly the vast majority of texts agreed to be by Shakespeare or not
by Shakespeare gives credence to their findings.
Moreover, a retrospective analysis of a range of data noted before Craig
and Kinney undertook their investigations shows that stylistic and other
features associating Arden with Shakespeare (phrases, images, compound
adjectives) tend to congregate within the play’s central scenes, while pointers away from Shakespeare (non-Shakespearean locutions, parallels with
The Jew of Malta, metrical irregularities) congregate in earlier or later scenes.
43. Gurr, Shakespearian Playing Companies, 258–77, esp. 262. John Southworth, Shakespeare the
Player: A Life in the Theatre (Stroud, Gloucestershire: Sutton Publishing, 2000), argues that
Shakespeare began his acting career with Worcester’s Men; that when the company disbanded he, along with at least five of its sharers, including Edward Alleyn, transferred in
1585 to the Admiral’s Men; and that Arden of Faversham was written for them ‘by Shakespeare
with some unknown collaborator’ and performed in Faversham (20–50, esp. 48).
124
Determining the Shakespeare Canon
There are statistically significant correlations with the Craig–Kinney
results. Earlier studies have unwittingly, and quite independently, anticipated them. More recent research by Craig and Burrows on 3 Henry VI
reaches conclusions that further reinforce those derived from LION. Links
between Arden’s Quarrel Scene and 1, 2, and 3 Henry VI, Titus Andronicus,
and Edward III, increasingly acknowledged to be collaborations, are concentrated in Shakespeare’s shares, not in those of his co-authors.44 The
respective rates are thirteen per 2,000 words (thirty-seven in 5,712 lines)
and four per 2,000 words (sixteen in 7,923 lines). Statistically, we would
expect fewer than one of 16,000 random distributions of links to result in
such a disparity.
In contrast when LION searches are applied to the first section (lines
1–76) of Arden, scene 14, which, according to the Craig–Kinney computations, is markedly un-Shakespearean, the rare phrases and collocations
that this part of the text shares with other plays of 1580–1600 fall into a pattern confirmatory of the stylometric conclusions. Links to Shakespeare are
no more numerous than the proportion of plays in the period that are his
would lead us to predict, and most of the links to Shakespeare’s early collaborations are to scenes that have been independently categorized as more
likely to be by his co-authors. So again, computational stylistics and the
technique of LION searching are in accord.
Finally, a literary-critical approach to the dramatic verse of Arden’s central scenes uncovers a lively linguistic awareness, a penchant for the concrete and particular, and a poetic imagination that interweaves threads
of metaphor by means of creative punning. These are all characteristic of
Shakespeare, but uncharacteristic of Kyd and other playwrights of the 1590s.
It is true that a small minority of passages, images, or stylistic features
proffered in Chapters 3 and 4 as characteristically Shakespearean fall outside scenes 4–9 of Arden of Faversham or Shakespeare’s shares, as defined
by stylometric and other tests, of 1, 2, or 3 Henry VI, Titus Andronicus, or
Edward III. But this does not negate the importance of the correspondences
between the different kinds of data. Nobody has claimed that the boundaries between dialogue by Shakespeare and dialogue by his co-authors can
be determined with absolute precision or that any single piece of evidence
can be decisive. It is upon overall patterns that a case for attribution rests.
4 4.Of course, it is by no means yet universally ‘acknowledged’ that all five plays—Titus
Andronicus, Edward III, and 1, 2, and 3 Henry VI—are collaborations, but, as is made clear in
Chapters 1–3, the evidence is now strong.
arden of faversham : Counter-arguments
and Conclusions 125
Recognition that Shakespeare was responsible for the middle scenes
of Arden of Faversham will surely open up fresh fields for critical cultivation. Previously unnoticed patterns among Shakespeare’s early collaborations should emerge. It is in Act 3 of both Arden and 2 Henry VI that the
Craig–Kinney tests judge Shakespeare’s hand to be most evident, and Act
3 of 2 Henry VI has many of the features of Arden Act 3’s Quarrel Scene,
with the latter’s caustic accusations between the two adulterers and eventual renewal of their commitment to killing Alice’s husband. In Act 3 of 2
Henry VI, the murder of Humphrey, Duke of Gloucester is plotted, perpetrated, and defended against the ensuing recriminations. Queen Margaret,
her lover the Duke of Suffolk, and the King exchange long emotional
polemics expressive of their conflicting attitudes, while the Duke of York
soliloquizes about his ambitions for the throne, moving from ‘misdoubt
to resolution’ (3.1.332), as does Mosby in his soliloquy in which his covetousness of Arden’s status and land is exposed. At the centre of 3 Henry VI
(3.2) it is Shakespeare who handles King Edward IV’s seduction of Lady
Elizabeth Gray, and the scene in which he does so ends with the long soliloquy in which Richard Gloucester schemes for the crown. And in Edward III
Shakespeare wrote the gripping scenes 1.2–2.2, in which the Countess of
Salisbury repells the King’s assaults on her married chastity.
Ros King has noted that in Arden of Faversham itself it is within scenes 4–9
that characters, even Black Will and Shakebag, acquire ‘a more complex
range of emotions and motivations. The outer scenes maintain the providentialism in the second edition of Holinshed. . . . The central scenes, by
contrast, are much more interested in the psychology of the characters.’45
Likewise, it is Shakespeare who, initiating in Titus Andronicus, 2.3, the adulterous affair between Tamora and Aaron, breathes psychological life into
the automatons of Peele’s Act 1.
Adulterous liaisons (Alice and Mosby, Queen Margaret and Suffolk,
Tamora and Aaron), kings as seducers or would-be seducers (Edward III,
Edward IV), strongly portrayed women (the Countess of Salisbury, Lady
Gray), including three who are fiercely passionate and complicit in murder
(Alice Arden, Queen Margaret, Tamora), men revealing their ambitions
45. King, ‘Arden of Faversham’, 649. Another treatment of Arden that takes account of recent work
on the play’s authorship is Lucia Nigri’s ‘ “Speaking the truth”: On the Power of Words
in Arden of Faversham’, Memoria di Shakespeare, 8: On Authorship, ed. Rosy Colombo and
Daniela Guardamagna (Rome: Bulzoni, 2012), 373–84; Nigri shows that a Shakespearean
concern with the ‘metatheatrical’ and ‘the beguilements of language’ gives scenes iv–ix a
special coherence (384).
126
Determining the Shakespeare Canon
in soliloquies (Mosby, the Duke of York, Richard Gloucester)—all in
scenes of some psychological and emotional complexity: these are among
elements common to Shakespearean portions of the early collaborative
plays.46 It is not inconceivable that although scenes with such subject matter
called forth a stylistic register that enabled tests of authorship to distinguish
Shakespeare’s writing of the early 1590s from that of other playwrights,
less dramatic scenes of his did not. As A. S. Cairncross insisted, in respect
of the ‘varied and complicated mass of material’ in 1, 2, and 3 Henry VI,
Shakespeare would have been ‘artist enough to know that style, like treatment, is determined by matter and purpose’.47 Computational stylistics and
LION searching may perhaps be prone to misclassify as ‘not Shakespeare’
some early Shakespearean passages, especially prose passages, that are simply
not distinctively enough his to be separated from passages by his contemporaries. But although there is potential for ‘false negatives’ of this kind, ‘false
positives’ should be confined to borderline cases: when a stretch of playtext is classified by Craig’s computations or by methodical LION searches
as clearly belonging with Shakespearean texts and clearly distanced from
non-Shakespearean texts, the verdict may be trusted. Shakespeare’s authorial involvement in Arden of Faversham seems beyond reasonable doubt.
46. Some of these points are also made by John Jowett towards the end of ‘Disintegration, 1924’,
Shakespeare (forthcoming). I am grateful to Professor Jowett for sending me a copy of his
article in advance of its publication.
47. Cairncross, 2 Henry VI, lii.
PART
TWO
A Lover’s Complaint
6
A Lover’s Complaint: Phrases and
Collocations
I
A Lover’s Complaint followed the 154 sonnets published as ‘SHAKE-SPEARES
| SONNETS’ in the famous quarto entered in the Stationers’ Register on
20 May 1609 and printed by George Eld for Thomas Thorpe in that year.
The poem consists of forty-seven stanzas of rhyme royal (ababbcc), the
form used by Shakespeare in The Rape of Lucrece. It was given the heading
‘A Louers complaint. | by | William Shake-speare.’ Half a century ago
two independent studies helped reverse widespread disbelief in this ascription.1 Commentators have since sought to show not only that the quarto
preserves the poet’s own arrangement of his sonnets, but that he intended A
Lover’s Complaint to be the third movement in a sonata-like structure preceded by the sections devoted to Fair Friend and Dark Lady, and working
to resolve their contradictions, or at least to put the experience embodied in
them into a new perspective.2 The status of A Lover’s Complaint—whether
1.Muir, ‘ “A Lover’s Complaint”: A Reconsideration’; Jackson, Shakespeare’s ‘A Lover’s
Complaint’: Its Date and Authenticity.
2. Roger Warren, ‘“A Lover’s Complaint”, “All’s Well”, and The Sonnets’, Notes and Queries,
215 (1970), 130–2; John Kerrigan, Motives of Woe: Shakespeare and the ‘Female Complaint’
(Oxford: OUP, 1991); Jennifer Laws, ‘The Generic Complexities of A Lover’s Complaint
and its Relationship to the Sonnets in Shakespeare’s 1609 Volume’, AUMLA: Journal of
the Australasian Universities Language and Literature Association, 89 (1998), 79–97; MacDonald
P. Jackson, ‘Aspects of Organisation in Shakespeare’s Sonnets’, Parergon, 17 (1999), 109–34;
Ilona Bell, ‘ “That which thou hast done”: Shakespeare’s Sonnets and A Lover’s Complaint’,
in Shakespeare’s Sonnets: Critical Essays, ed. James Schiffer (New York and London: Garland,
1999), 455–74; various contributions to Critical Essays on Shakespeare’s ‘A Lover’s
Complaint’: Suffering Ecstasy, ed. Shirley Sharon Zisser (Aldershot: Ashgate, 2006) and A
Companion to Shakespeare’s Sonnets, ed. Michael Schoenfeldt (Oxford: Blackwell, 2007);
Katherine A. Craik, ‘Shakespeare’s A Lover’s Complaint and Early Modern Criminal
Confession’, Shakespeare Quarterly, 53 (2002), 437–59; Kenji Go, ‘Samuel Daniel’s The
130
Determining the Shakespeare Canon
it is Shakespearean or non-Shakespearean—is thus directly related to the
question of the authority of the quarto’s text and the order in which sonnets
are presented and numbered. A spurious A Lover’s Complaint would undermine trust in Thorpe’s volume; a Shakespearean A Lover’s Complaint tends
to authenticate it.
But even while criticism has been discovering a rationale for the authorial inclusion of A Lover’s Complaint within Shakespeare’s Sonnets, Brian
Vickers, Marina Tarlinskaja, and the Claremont-McKenna Shakespeare
Clinic, run by Ward E. Y. Elliott and Robert J. Valenza, have, reviving an
old scepticism, argued against Shakespeare’s responsibility for the poem.3
In ‘A Lover’s Complaint Revisited’ (2004), written before the most resolute
of these new challenges to the authenticity of the poem were published,
I provided reasons for trusting Thorpe’s attribution, stressing the clusters of
vocabulary, imagery, and ideas that the Complaint shared with the acknowledged Shakespeare canon.4 This chapter rehearses and boosts with new
evidence the argument advanced there.
Elliott and Valenza distinguish between ‘green-light’ or ‘smoking gun’
testing, in which the investigator finds ‘quirks’ that link a disputed work
with the Shakespeare canon and then seeks to establish the genuine rarity of these quirks within non-Shakespearean controls, and their own
‘red-light’ or ‘silver bullet’ testing, which begins by setting Shakespearean
Complaint of Rosamond and an Emblematic Reconsideration of A Lover’s Complaint’, Studies
in Philology, 104 (2007), 82–122; Margaret Healy, Shakespeare, Alchemy and the Creative
Imagination: The Sonnets and ‘A Lover’s Complaint’ (Cambridge: CUP, 2010). Among the
more significant editions are John Kerrigan, ed., The Sonnets and A Lover’s Complaint
(Harmondsworth: Penguin, 1986); Katherine Duncan-Jones, ed., Shakespeare’s Sonnets
(London: Nelson, 1997; rev. edn. London: Methuen Drama, 2010); John Rowe, ed.,
The Poems (Cambridge: CUP, 1992); Colin Burrow, ed., The Complete Sonnets and Poems
(Oxford: OUP, 2002). For further arguments for the authenticity of the 1609 quarto’s
ordering of sonnets, see MacDonald P. Jackson, ‘Shakespeare’s Sonnets: Rhyme and Reason
in the Dark Lady Series’, Notes and Queries, 244 (1999), 219–22; ‘The Distribution of
Pronouns in Shakespeare’s Sonnets’, AUMLA: Journal of the Australasian Universities Language
and Literature Association, 97 (2002), 22–38; Brian Boyd, Why Lyrics Last: Evolution, Cognition,
and Shakespeare’s Sonnets (Cambridge, MA: Harvard UP, 2012).
3.Vickers, Complaint; Ward E. Y. Elliott and Robert J. Valenza, ‘And Then’; ‘Glass Slippers
and Seven-Leagued Boots: C-prompted Doubt About Ascribing A Funeral Elegy and
A Lover’s Complaint to Shakespeare’, Shakespeare Quarterly, 48 (1997), 177–207; ‘Did
Shakespeare Write A Lover’s Complaint?: The Jackson Ascription Revisited’, in Words That
Count: Early Modern Authorship: Essays in Honor of MacDonald P. Jackson, ed. Brian Boyd
(Newark: University of Delaware Press, 2004), 117–40; Marina Tarlinskaja, ‘The Verse of A
Lover’s Complaint: Not Shakespeare’, in Words That Count, 141–58; ‘Who Did NOT Write A
Lover’s Complaint’, Shakespeare Yearbook, 15 (2005), 434–82.
4.MacDonald P. Jackson, ‘A Lover’s Complaint Revisited’, Shakespeare Studies, 32 (2004),
267–94.
a lov er’s compl ai n t : phrases
and collocations
131
boundaries outside which, for any one test, rates for at least 10 per cent
of non-Shakespearean controls fall, and goes on to demonstrate that the
non-Shakespearean works flunk significantly more tests than any of the
canonical ones. And they judge ‘silver bullet’ testing to be superior: ‘with
less-than-perfect identifiers, strong exclusionary evidence normally trumps
strong inclusionary evidence’.5 Elliott and Valenza are right in claiming
that their ‘silver bullet’ procedures more effectively safeguard against ‘false
positives’—against the identification as Shakespeare’s of material that is not
in fact his. But those procedures may be more apt to suffer from the obverse
flaw of sometimes registering ‘false negatives’. Besides, a recent analysis
demonstrates that when statistically desirable corrections are made to their
parameters for non-dramatic verse, A Lover’s Complaint performs better
than certain similar sized blocks of Shakespeare’s sonnets.6
Moreover, the accumulation of quirks can be systematized so as to yield
results that are immune from the strictures that may be directed at simple
citing of parallels, as in 2004 I attempted to show. Using Literature Online
I subjected samples of A Lover’s Complaint to comprehensive searches: the
first five stanzas, stanzas 20–4 from the middle of the poem, and the last
five stanzas. These fifteen stanzas were systematically checked for phrases
and collocations occurring five or fewer times in dramatic works of the
period 1590–1610. The total number of such works in the database that are
assigned to first performance dates during these years is 267, though several
of these are masques, entertainments, or pageants, rather than plays, and
three could be disregarded as falling outside the 1590–1610 limits but misclassified by LION.
Scholars who reject Shakespeare’s authorship of A Lover’s Complaint
either ignore the many connections that have been noted between the poem
and Shakespeare’s sonnets or seek to explain them by a non-Shakespearean
poet’s writing in deliberate response to them.7 Later chapters of this book
are designed to prove that such explanations will not serve, and in the
findings of ‘A Lover’s Complaint Revisited’ they were already confronted
with a solid obstacle. We would hardly expect the mind of an imitator of
5. Elliott and Valenza, ‘Glass Slippers’, 183.
6. MacDonald P. Jackson, ‘A Lover’s Complaint and the Claremont Shakespeare Clinic’, Early
Modern Literary Studies, 16.3 (2013), 1–12. This rebuttal of the Claremont Clinic’s case against
Shakespeare’s authorship of the Complaint is summarized in the ‘Counter-arguments’ section of the present book’s Chapter 10.
7. The commentaries on A Lover’s Complaint in the editions listed in n. 2 above note many of
the verbal connections between the poem and the Sonnets.
132
Determining the Shakespeare Canon
Shakespeare’s sonnets to be saturated with the phraseology and diction of
Shakespeare’s plays, but A Lover’s Complaint’s links, in phrases and collocations, with Shakespearean drama turned out greatly to outnumber its links
with the drama of other playwrights, even once the amount of searchable
text for various authors was taken into account.
Details of the procedures adopted are spelled out in the article, where
the relevant data are listed. Here a summary of the results will suffice. The
searches yielded fifty-three links to Shakespeare, eight to Marston, eight
to Chapman, six to Heywood, and six to Jonson. Of course the period
1590–1610 covers most of Shakespeare’s highly productive playwriting
career. Of his plays, only The Tempest and the collaborations with Fletcher,
Henry VIII and The Two Noble Kinsmen, fall outside this period, so that
thirty-five Shakespeare plays were searched. However, since the database
for 1590–1610 includes fourteen dramatic works by Chapman, eight by
Marston, fifteen by Heywood, and twenty-one by Jonson (though many
of Jonson’s are masques and entertainments), even when authorial links
were reckoned in proportion to the number of each playwright’s searched
plays, Shakespeare (with 1.5 links per play) was far ahead of the playwrights
whose scores on the raw figures came closest to his: Marston led the chase
with 0.8, while Chapman (0.6), Heywood (0.4), and Jonson (0.3) followed.8
These were the only playwrights to display five or more links to the fifteen stanzas of A Lover’s Complaint. Between them Chapman, Marston,
Heywood, and Jonson had a total of twenty-eight links from sixty dramatic works, compared to Shakespeare’s fifty-three from thirty-five. And
in fact all 231 non-Shakespearean works yielded only eighty-two links, just
slightly over 50 per cent more than Shakespeare’s thirty-five. Moreover,
only four of Shakespeare’s thirty-five plays failed to provide at least one
link, whereas 162 of the 231 non-Shakespearean plays provided no links.
Links to Shakespeare predominated in each of the three sets of five stanzas, whereas links to Chapman, for example, though quite frequent in the
opening stanzas, disappeared completely from the last five.
It is unlikely that some dramatist who wrote only one or two plays during the relevant period was responsible for A Lover’s Complaint, since even
in terms of links to single plays Shakespeare outstripped all rival candidates.
Only three non-Shakespearean plays had as many as three links: Heywood’s
The Golden Age, Middleton’s The Revenger’s Tragedy, and Peele’s The Old
8. The totals for numbers of plays by different authors are for plays of single authorship only,
except that Shakespeare’s collaborations were included in his total of 35.
a lov er’s compl ai n t : phrases
and collocations
133
Wives Tale. Shakespeare had two plays with four links, Othello and All’s Well
That Ends Well.
Almost as significant as the unequalled proportion of links to
Shakespeare’s plays was the chronological distribution of links.9 When
Shakespeare’s plays were arranged in chronological order, according
to the Oxford Shakespeare’s Textual Companion, and each of the links to
Shakespeare was marked against the appropriate play, those two plays with
the greatest number of links, namely Othello and All’s Well That Ends Well
with four each, could be seen to be consecutive, and they were immediately followed by two plays with three links each, Timon of Athens and King
Lear. So this tight group of four Shakespeare plays, dated 1603–4 to 1606,
afforded fourteen links, many more than any other group of four plays.
This neat result has since been complicated by the Oxford editors’ decision in the second edition of the Complete Works to date All’s Well 1606–7,
but the strong association with Shakespeare’s plays of the slightly extended
period, 1603–4 to 1607 remains: twenty of the fifty-five links are to the
seven-play series Othello, Lear, Timon, Macbeth, Antony and Cleopatra, All’s
Well, Pericles.
The only other Shakespeare plays with three links were Titus Andronicus
and The Merchant of Venice. An interesting point about the Titus Andronicus
links is that they were all to 3.2, the ‘fly scene’, which was absent from the
quarto of 1594 and first appeared in the First Folio of 1623, and so has usually been considered a later addition to the play.10
The LION searches thus supported the conclusions that Shakespeare was
the author of A Lover’s Complaint and that he most probably wrote it within
the period 1603–7.
Further analysis of the data in ‘A Lover’s Complaint Revisited’ strengthens this conclusion. Picking out from the list only the phrases and collocations peculiar to one or more plays by Shakespeare, we find twenty-eight,
with no fewer than seven plays containing two: Titus Andronicus, Love’s
Labour’s Lost, The Merchant of Venice, Hamlet, Othello, All’s Well That Ends
9. In these chronological computations the links to Middleton’s portion of Timon of Athens and
Wilkins’s of Pericles were included, giving 55 links altogether.
10. In ‘A Lover’s Complaint Revisited’, 289, I added that the poem’s links to non-Shakespeare
plays also peaked in the years 1604–7, but neglected to calculate links in terms of the numbers of searchable non-Shakespeare plays first performed in any given year. Once these calculations have been carried out, the association of the Complaint with seventeenth-century
non-Shakespeare plays persists but is reduced to statistical insignificance. Clearly, the
chronological relationship, in the matter of shared phrases and collocations, is most robust
between works by the same author.
134
Determining the Shakespeare Canon
Well, and Timon of Athens. Phrases and collocations peculiar to any one
other playwright number twenty-five, the only playwrights to afford
more than one being Heywood (with three) and Chapman (with two),
and the sole play containing two being Heywood’s The Golden Age
(1610), despite more than six and a half times more non-Shakespearean
than Shakespearean dramatic works falling into the period for which the
searches were carried out. Since the seven-play sequence from Measure
for Measure (1603–4) to All’s Well (1606–7, in the revised Oxford dating)
affords appreciably more links than any other, this sifted and refined evidence corroborates the more inclusive.
The results for A Lover’s Complaint phrases and collocations peculiar to a
single playwright, 1590–1610, may be summarized as follows, with figures
for items unique to just one play added:
Peculiar to Shakespeare:
28
Peculiar to another playwright:
26
Shakespeare plays with two or more such links: 7
Other plays with two or more such links: 2
Peculiar to a single Shakespeare play:
15
Peculiar to a single non-Shakespeare play:
22
No. of plays: 35
No. of plays: 231
No. of plays: 35
No. of plays: 231
No. of plays: 35
No. of plays: 231
The LION search for links between A Lover’s Complaint and plays, rather
than poems, was undertaken because (a) plays are of roughly similar sizes,
whereas poems range in length from brief lyrics to Chapman’s translation of Homer’s Iliad, (b) Shakespeare’s corpus of non-dramatic verse is
relatively small, and (c) there is no obvious reason why if somebody other
than Shakespeare wrote the Complaint it should be linked most strongly
in its phrasing with plays by Shakespeare. It must be stressed, however,
that this was not a cross-genre comparison open to the criticisms made in
Chapter 10 of the work on A Lover’s Complaint of Tarlinskaja and of Elliott
and Valenza. My investigation did not compare the poem with plays, on
rates of usage of particular features; it was designed merely to determine
which plays of 1590–1610 shared with the Complaint the greatest number of
rare phrases and collocations.
Moreover, as a check on the soundness of the procedure, it can be
applied to a couple of Shakespeare’s sonnets. The linked pair 109–10 was
chosen at random from those in the last twenty-five of the sub-sequence
centred on the Fair Youth, since numbers 101 to 125 (plus the twelve-line
envoy) are probably the last composed and most nearly contemporaneous
a lov er’s compl ai n t : phrases
and collocations
135
with the Complaint.11 The data from LION searches for phrases and collocations in Sonnets 109–110 that occur five or fewer times in drama of
1590–1610 are recorded in Appendix 4, Part A. There are twenty-seven
links to Shakespeare plays, fifty-one to plays by other dramatists. So 35 per
cent of the links are with the 13 per cent of LION plays 1590–1610 that are
Shakespeare’s. This result is comparable to that for A Lover’s Complaint,
where 39 per cent of the links are with Shakespeare. Sonnet 109 has an
appreciably higher proportion of links to Shakespeare than Sonnet 110.
Unsurprisingly, the intensity of linkage to Shakespearean plays may vary
over short stretches of Shakespearean non-dramatic verse.
Seven of the plays providing two or more links are among Shakespeare’s
thirty-five, five among the 251 by other playwrights. Eight of the twelve
items shared exclusively with a single dramatist or with a single collaborative play are Shakespeare items, and the pair of sonnets closely matches
the Complaint in the chronological distribution of links, in that sixteen of
Shakespeare’s twenty-seven are with his eleven plays of the period 1598–9 to
1605–6. The resemblance between the performances of A Lover’s Complaint
and Sonnets 109–110 on this kind of LION testing strongly suggests identity of authorship.
II
Noting the findings of ‘A Lover’s Complaint Revisited’, Brian Vickers asked
himself, ‘why must the author of the Complaint have been a dramatist? Why
not a poet?’ The question led to research that, he hoped, ‘settled the authorship of A Lover’s Complaint once and for all’ with an attribution to John
Davies of Hereford.12 However, no poem agreed to be by Davies has been
shown to display such affiliation to Shakespearean drama as does A Lover’s
Complaint, and the methods used in ‘A Lover’s Complaint Revisited’ can be
extended to test Vickers’s theory.
11.MacDonald P. Jackson, ‘Rhymes in Shakespeare’s Sonnets: Evidence of Date of Composition’, Notes and Queries, 244 (1999), 213–19; ‘Vocabulary and Chronology: The Case of
Shakespeare’s Sonnets’, Review of English Studies, 52 (2001), 59–75; ‘Dating Shakespeare’s
Sonnets: Some Old Evidence Revisited’, Notes and Queries, 247 (2002), 237–41.
12. Vickers, Complaint, 3–4, 6.
136
Determining the Shakespeare Canon
As backing for a review of Vickers’s book, John Jowett conducted a miniature version of LION searching, restricted to the first stanza of the poem
and to ‘comparing Shakespeare and Davies as putative authors and identifying features exclusive to, or clearly favored by, one or the other’.13 This produced a dozen affinities with Shakespeare, only one with Davies, and while
Davies used ‘tearing of ’ and Shakespeare did not, this ‘single lexical choice
favoring Davies’ was ‘immediately cancelled’ by the Shakespearean ‘of
papers’ in the whole Complaint phrase, since Shakespeare, but not Davies,
used ‘of paper’ and ‘tear his paper’.14
Extending the search to the first seven stanzas so strikingly substantiates these results that such intensive LION investigation of the rest of A
Lover’s Complaint seems unnecessary. Appendix 4, Part B lists phrases
and collocations found in the works of Shakespeare but not in those
of Davies and vice versa. Except for compound words, single items of
vocabulary do not qualify for inclusion unless falling within a context
of similar images or ideas. Shakespeare’s ‘a-twain’ was not regarded as a
compound, and a simple definite or indefinite article before a noun was
not regarded as sufficient to turn a lexical item into an eligible phrase: as
a spelling ‘a-twain’ is considered in Chapter 7, and as items of vocabulary both ‘a-twain’ and Davies’s ‘a maund’ are considered in Chapter 8.
When identical juxtaposed words are used with different meanings, they
are not accepted as establishing links.
Restricting the list to items with some exact verbal correspondence leaves
out some remarkable parallels between the opening stanzas of A Lover’s
Complaint and Shakespeare’s works, where there are the same multiple associations of ideas. For example, the woman who positions herself by the ‘weeping margent’ of a river, where she is ‘applying wet to wet’ (39–40), is like
Jaques who in As You Like It ‘Stood on th’extremest verge of the swift brook,
| Augmenting it with tears’ (2.1.42–3), or Romeo, who is seen ‘With tears
augmenting the fresh morning’s dew’ (Romeo and Juliet, 1.1.132). Moreover,
the woman in the Complaint’s shedding of tears into a river is likened to
‘monarch’s hands that let not bounty fall | Where want cries some, but where
excess begs all’ (41–2), while the melancholy Jaques moralizes:
First, for his weeping into the needless stream:
“Poor deer,” quoth he, “thou mak’st a testament
13. John Jowett, review of Vickers, Complaint, Shakespeare Quarterly, 60 (2009), 493–7, at 496.
14. Jowett, 496.
a lov er’s compl ai n t : phrases
and collocations
As worldlings do, giving the sum of more
To that which hath too much.”
137
(2.1.46–9)
The same conceit occurs in 3 Henry VI: ‘With tearful eyes add water to
the sea, | And give more strength to that which hath too much’ (5.4.8–9).
These parallels lack shared keywords but share the same concepts. King
Lear, 4.1.70–1, on the other hand, shares with the Complaint’s lines 41–2 not
only the socialistic thought about the distribution of wealth or resources,
but also the exact keyword ‘excess’, and so enters the list.
Likewise, the adjective ‘concave’ links the opening line of A Lover’s
Complaint with Julius Caesar, 1.1.45–7, where the shouts of the plebeians, as
they greeted Pompey, are said to echo resoundingly within Tiber’s riverbed, so this item is included in the inventory of links. It is Venus and Adonis,
however, that, while different in wording, provides the closest parallel to the
opening lines of A Lover’s Complaint. Venus has been deserted by Adonis:
And now she beats her heart, whereat it groans,
That all the neighbor caves, as seeming troubled,
Make verbal repetition of her moans;
Passion on passion deeply is redoubled:
“Ay me!” she cries, and twenty times, “Woe, woe!”
And twenty echoes twenty times cry so.
(829–34)
Marking the echoes, she ‘begins a wailing note, | And sings extemporally
a woeful ditty’ or ‘heavy anthem’, to which the ‘choir of echoes answer’
(835–40). A Lover’s Complaint compresses the ideas elaborated here. In each
poem caves resound with a lament that is musical as well as verbal. ‘Double
voice’ in A Lover’s Complaint picks up Venus and Adonis’s ‘redoubled’, which
the later poem recasts as ‘reworded’. Moreover, the adjectival use of ‘neighbor’ in ‘the neighbor caves’ begins the process which leads to ‘neighboring’
in All’s Well (4.1.16) and ‘sist’ring’ in A Lover’s Complaint (2).15 By the third
stanza the ‘fickle maid’ is ‘shrieking undistinguished woe’ (20), as Venus
does. (One might even add that in A Lover’s Complaint ‘woe’ rhymes with
‘high and low’, while in Venus and Adonis, 1139–40 ‘woe’ rhymes with ‘high
or low’.)
15. OED’s first example of ‘sister’ as a verb is from Pericles, used in the Gower chorus heading
Act 4. The Complaint’s ‘sist’ring’ has been coined on the analogy of ‘neighbouring’ (for
which OED’s first citation is from All’s Well ) but suggests a more intimate natural relationship between the hill with its ‘concave womb’ and the nearby vale: they are not just neighbours, but sisters, so that nature is not merely humanized but feminized to provide a setting
for the entry of the ‘fickle maid’.
138
Determining the Shakespeare Canon
The extended Venus and Adonis parallel might perhaps have warranted
inclusion on the strength of ‘redoubled’ or ‘double’, but I decided to leave it
out. Henry V affords a further close parallel of imagery and ideas:
He’ll call you to so hot an answer to it
That caves and womby vaultages of France
Shall chide your trespass and return your mock
In second accent of his ordinance.
(2.4.121–6)
Despite the ‘womby’or ‘womb’ connection in a context of caves and echoes,
this association between A Lover’s Complaint and the Shakespeare canon is
also simply noted here, rather than itemized for statistical purposes.
So also is a parallel with Richard III. The woman in the Complaint throws
‘favors . . . | Of amber, crystal, and of beaded jet’ into a river (36–8), along
with ‘many a ring of posied gold and bone, | Bidding them find their
sepulchres in mud’ (45–6). The image of ‘gold’, ‘bone’, and gemstones in
their muddy ‘sepulchres’ recalls Clarence’s dream ‘of gold . . . | Inestimable
stones, unvalued jewels’ on the sea bed, with ‘reflecting gems | That
woo’d the slimy bottom of the deep, | And mock’d the dead bones that
lay scatt’red by’ (1.4.26–8, 31–3). Although Richard III’s ‘gold’ and ‘bones’
furnish lexical links to the Complaint, in the play they are too far apart to
constitute an acceptable collocation.
Vickers notes that in The Holy Rood Davies collocates ‘eyes’, ‘tears’, ‘brine’,
and ‘season’, while the Complaint has ‘the brine | That seasoned woe had pelleted in tears’ (17–18) following ‘eyne’ (15).16 But in The Rape of Lucrece, we
find Lucrece ‘Seasoning the earth with show’rs of silver brine, | Mingling
my talk with tears, my grief with groans’ (796–7), and the Countess in All’s
Well That Ends Well comments on Helena’s ‘tears. ’Tis the best brine a maiden
can season her praise in’ (1.1.47–9). It is true that Davies includes ‘eyes’ and the
Complaint includes ‘eyne’, but ‘tears’ are obviously implicit in all three contexts, and Shakespeare’s weepers are betrayed or love-sick women (a ‘maid’
in the Complaint and a ‘maiden’ in All’s Well), whereas in Davies the tears
are Christ’s. Vickers observes that Davies elsewhere associates ‘woe’ with the
‘brine’ of tears, but the word ‘woe’ actually occurs in the Lucrece context
(790). So in this case I have not counted a hit for either poet.
But although some inclusions and exclusions are debatable and a few
valid items have doubtless been overlooked, the evidence accumulated in
16. Vickers, Complaint, 243–4.
a lov er’s compl ai n t : phrases
and collocations
139
Appendix 4, Part B is overwhelming. Links with non-Shakespeare scenes
in 1 Henry VI, 2 Henry VI, Titus Andronicus, Pericles, and The Two Noble
Kinsmen have been discarded: both links to 3 Henry VI are to Shakespeare
scenes, as outlined in Chapter 3, Section III above. Appendix 4, Part B
records seventy-seven links to Shakespeare, seven to Davies. Even though
Davies’s oeuvre is only about 40 per cent the size of Shakespeare’s, this is
a huge disparity. And there are sixteen links to Shakespeare’s poems, over
twice as many as to Davies’s, despite Davies’s output of non-dramatic verse
being eight times larger than Shakespeare’s. Venus and Adonis has ten links,
compared to the seven of the whole Davies corpus, with thirty-five times as
many lines. King Lear, generically so distant from A Lover’s Complaint, has
eight links.
To put it in this way is, however, slightly to exaggerate the case against
Davies. If, ignoring Shakespeare’s plays, we search for phrases and collocations found in the poems of Davies but not of Shakespeare, and vice
versa, Davies gains a few extra items that are not listed in Appendix 4,
Part B because they appear in at least one Shakespeare play. ‘If thought
can think’ in Humour’s Heaven on Earth creates a link to The Complaint’s
‘the thought might think’ (10); this is absent from the main inventory because The Merchant of Venice offers an equally close phrase, ‘the
thought | To think’ (1.1.36–7). Although ‘both high and low’ appears
in Shakespeare but not Davies (see list at line 21), there are no instances
of ‘high and low’ in Shakespeare’s poems, whereas Davies has seven—
in Humours Heaven, Microcosmos (thrice), The Scourge of Folly, and Wit’s
Pilgrimage (2). In Humour’s Heaven Davies has ‘as they did’, which is
absent from Shakespeare’s poems, but in the Complaint (23) ‘as’ means ‘as
if ’, which is not the sense in Davies, so that it must be excluded, according to my normal criteria. In the wrong sense it is found several times
in Shakespeare’s plays, and in the right sense it appears in a scene of
Edward III ‘possibly’ by Shakespeare: ‘eyes | Look on each other, as they
did attend | Each other’s words’ (4.5.9–11), where ‘as they did attend’
means ‘as if they did listen to’, when in fact they did not. The locution
‘every place’ (27) occurs three times in Davies’s poems (Humour’s Heaven,
Mirum in Modum, and Summa Totalis), but not in Shakespeare’s, though
it is used several times in the plays. Davies also has three parallels to the
Complaint’s ‘letters, sadly penn’d in blood’ (47): ‘command my pen to
make | Ink of thy blood’ (Wit’s Pilgrimage), ‘The spear the pen, his precious blood the ink’ (The Holy Rood), ‘And with this pen write . . . | In
140
Determining the Shakespeare Canon
lines that . . . flow like blood’ (The Scourge of Folly). These are excluded
from the main inventory because once Shakespearean drama is restored
to the picture the ideas, images, and wording are no less well matched
in ‘Turning . . . your ink to blood, | Your pens to lances’ in 1 Henry IV
(4.1.50–1).
This procedure of searching only poems would add a further fourteen
links to Davies’s tally, giving a total of twenty-one. But since Davies’s
poetic corpus is eight times the size of Shakespeare’s, the crucial point is
that, proportionally to the amount of non-dramatic verse available for
searching, links to Shakespeare’s poems would still be six times more
frequent than links to Davies’s. Moreover, Davies’s recalculated total is
swollen by his use six times of ‘high and low’ and three times of ‘every
place’—both, especially the latter, very common in the period in which he
was writing but missing from Shakespeare’s non-dramatic verse because
of its relatively small quantity. If we count Complaint phrases and collocations according to whether they are used in Shakespeare’s poetic corpus but
not Davies’s, and vice versa, irrespective of how many times they appear,
there are eleven items for Shakespeare, nine for Davies, which means that
their rate of occurrence is nearly ten times greater in Shakespeare than in
Davies. Whatever mode of reckoning we adopt, the affiliations of A Lover’s
Complaint’s idiolect are with Shakespeare, rather than with Davies.
7
Spellings in A Lover’s Complaint
as Evidence of Authorship
I
In his 1609 quarto of Shakespeare’s Sonnets publisher Thomas Thorpe
included A Lover’s Complaint and ascribed it unequivocally to ‘William
Shakespeare’. Over the last fifty years this ascription has been widely
accepted, but dissenting views culminated in the publication in 2007 of
Brian Vickers’s Shakespeare, ‘A Lover’s Complaint’, and John Davies of Hereford,
in which the poem is attributed to Davies. As made clear in Chapter 6,
I think that Vickers is wrong, that Thorpe was right, and that the poem
should be retained within the Shakespeare canon.
The evidence put forward in this chapter is of a kind employed by
J. Dover Wilson to strengthen the case for Shakespeare’s having composed
and penned Hand D’s three pages of the manuscript play Sir Thomas More.1
Wilson investigated ‘bibliographical links’ between Hand D’s pages and the
Shakespearean ‘good quartos’. Early modern authors differed in their spelling preferences and practices during a time when orthography was fluid.
Compositors tended to normalize many of the most eccentric spellings, but
perpetuated some. As an editor of Shakespeare, Wilson had noted certain
odd spellings—mainly old-fashioned and presumably authorial—surviving into quartos thought to have been printed directly from Shakespeare’s
‘foul papers’, which were in his own handwriting. Wilson showed that parallels to Hand D’s more unusual spellings could be found in the quartos and
1.J. Dover Wilson, ‘Bibliographical Links between the Three Pages and the Good
Quartos’, in Alfred W. Pollard, ed., Shakespeare’s Hand in ‘The Play of Sir Thomas More’
(Cambridge: CUP, 1923), 113–41.
142
Determining the Shakespeare Canon
that misprints in the quartos could be explained by Hand D’s letter forms.
Because Literature Online preserves the spelling of the early printed texts,
I have since been able to confirm that orthographical links between the
three pages and Shakespeare’s works are genuinely significant.2
In studying the spellings of A Lover’s Complaint, I fully acknowledge the
potential effects of scribes and compositors on early modern printed texts.
But for all the vagaries of their textual transmission, quartos of Thomas
Middleton’s plays, for example, are marked by a sprinkling of the idiosyncratic spellings that are also found in his Trinity MS holograph of A Game
at Chess: reuennewe (revenue), closse (close), dambd (damned), enuite (invite),
and so on.3 Thomas Heywood’s holographs exhibit a peculiar liking for Ey
as a spelling of ‘Ay’ and this infiltrates several of the quartos of his plays.4
Jonson favoured ’hem over the standard ’em as a contraction of ‘them’.5 What
is of prime importance, from a methodological point of view, is that such
evidence be gathered in a systematic manner that facilitates its evaluation.
The caveat has been sounded more than once in this study, because failure to heed it has resulted in the chasing of many a will-o’-the–wisp. Too
often in attribution studies scholars have searched for similarities between
a disputed work and the writings of their favoured candidate for its authorship, amassing a body of evidence that is superficially impressive but that
cannot be assessed because the search for similarities has been uncontrolled.
A more satisfactory approach is to search a predetermined range of texts,
by a variety of authors, for a carefully defined category of features that
they share with the disputed work, so that the investigation is not biased
in favour of any one authorial candidate. Searchable electronic databases
facilitate such a procedure.
2. MacDonald P. Jackson, ‘Is “Hand D” of Sir Thomas More Shakespeare’s? Thomas Bayes and
the Elliott–Valenza Authorship Tests’, Early Modern Literary Studies, 12.3 ( January, 2007),
1.1–36 <http://purl.oclc.org/emls/12-3/jackbaye.htm>, esp. 15–25. For further confirmation that Hand D’s three pages are by Shakespeare, see MacDonald P. Jackson, ‘The Date
and Authorship of Hand D’s Contribution to Sir Thomas More: Evidence from “Literature
Online” ’, Shakespeare Survey, 59 (2006), 69–78.
3.MacDonald P. Jackson, ‘Editing, Attribution Studies, and “Literature Online”: A New
Resource in Renaissance Drama’, Research Opportunities in Renaissance Drama, 37 (1998),
1–16, at 5.
4.MacDonald P. Jackson, ‘John Webster and Thomas Heywood in Appius and
Virginia: A Bibliographical Approach to the Problem of Authorship’, Studies in Bibliography,
38 (1985), 217–35, at 226. Heywood’s peculiarities are described in the introductions to
the Malone Society Reprints of The Captives, ed. Arthur Brown (1953), and The Escapes of
Jupiter, ed. Henry D. Janzen.
5. David J. Lake, The Canon of Thomas Middleton’s Plays (Cambridge: CUP, 1975), tables following 252: Band 3 (g).
spellings in
a lover ’ s complaint
as evidence of authorship 143
II
For the present study, my modus operandi was to enter into the LION
search box every unusual spelling in A Lover’s Complaint as it is printed
in the Sonnets quarto of 1609.6 For the first phase of this experiment only
drama of 1590–1614 was searched. A second phase examined poetry,
but the initial choice of drama was made because LION tags plays by
date of composition or first performance as given in Annals of English
Drama and plays are of roughly comparable sizes.7 Chronological limits
for poetry, in contrast, can be set in LION only by date of publication
or by the broad period in which the poet lived, and poetical works may
vary in length between epics and short lyrics, while the lapse of time
between composition and publication may be considerable. The number
of dramatic works available for LION inspection within 1590–1614 is
about 335, including, of course, Shakespeare’s thirty-nine.8 I recorded all
examples of a ‘rare’ spelling, the criterion for rarity being that it should
occur in the plays of no more than five dramatists. Almost always this
meant that it occurred in no more than five plays, but occasionally a
spelling turned up in more than five plays of which two or more were
by the same dramatist. Anonymous plays were treated as by different
dramatists. The record was restricted to spellings that were exactly the
same—the same graphical unit—as in A Lover’s Complaint. For substantives, plurals were thus distinguished from singulars, while verbs were
distinguished by inflexion. Spellings beginning with an upper-case letter were not differentiated from spellings beginning with a lower-case
one, and indeed I ignore capitals in citing separate words and spellings
throughout this chapter. I discounted words that are rare as words but
appear in the Complaint in a normal modern spelling or in a spelling that
6.‘Unusual’ may seem a loose term, but I checked all spellings that were not the standard
modern ones, unless my familiarity with sixteenth- and seventeenth-century printed texts
enabled me to be certain that a particular non-modern spelling was very common at the
time. I used the facsimile, William Shakespeare: Sonnets 1609 (Menston: Scolar Press, 1968).
7. There are a few minor discrepancies between LION and Annals, but for this study of spellings I accepted LION’s datings, and, with slight adjustments, its attributions of authorship.
8.This figure includes the thirty-six First Folio (1623) plays, plus Pericles, The Two Noble
Kinsmen, and Edward III, which, like the Folio’s 1, 2, and 3 Henry VI, Titus Andronicus, Timon
of Athens, and Henry VIII, are collaborations. Some of the non-Shakespearean ‘dramatic
works’ are masques or entertainments, but to avoid clumsiness I sometimes use ‘plays’ as an
inclusive term.
144
Determining the Shakespeare Canon
would be normal once allowance were made for the difference between
Elizabethan–Jacobean conventions governing u and v and present-day
conventions. Verb endings in -’d were regarded as ‘modern’, since they
are used in modern-spelling Shakespeare editions, such as The Riverside
Shakespeare, in which -ed endings after consonants are syllabic.
As noted in Chapter 1, LION includes more than one version of
most Shakespeare plays—from quartos, First Folio (1623), and the
modern-spelling old Cambridge edition of 1863–6. I recorded Lover’s
Complaint spellings that occurred in either a quarto or the Folio text of a
Shakespeare play, but statistical analysis of the results will, of course, take
into account this duplication of texts for several Shakespeare plays.
The results of these searches follow. I give A Lover’s Complaint’s ‘rare’
spelling (with the modern spelling in parentheses); a line reference; the plays
in which the spelling is found, indicating each play’s author or co-authors;
and the date of the published text on which LION draws. Although all
plays were, according to LION, composed or first performed within the
1590–1614 limits, many of LION’s source texts were not printed till after
1614. For plays in manuscript LION normally resorted to a facsimile or diplomatic reprint, but these have been designated ‘MS’ in the following list.
Spellings in A Lover’s Complaint (1609) that are rare in plays of 1590–1614:
doble (double), 3: Anon., Caesar and Pompey, or Caesar’s Revenge (1607);
Anon., Edmond Ironside (MS).
a twaine (a-twain), 6: Shakespeare, King Lear (1623).
peept (peeped), 14: Barry, Ram Alley (1611); Middleton, A Trick to Catch
the Old One (1607) twice; Shakespeare, 2 Henry IV (1600); Shakespeare,
Romeo and Juliet (1597).
lettice (lattice), 14: Chapman, All Fools (1605); Heywood, The Royal King
and the Loyal Subject (1637); Shakespeare, All’s Well That Ends Well
(1623).
charecters (characters), 16: Anon., The Wars of Cyrus (1594); Marston, The
Entertainment at Ashby (MSS).
laundring (laundering), 17: Jonson, The Alchemist (1616).
shriking (shrieking), 20: Brandon, The Virtuous Octavia (1598) twice;
Shakespeare, Henry V (1623); Shakespeare, Troilus and Cressida (1609).
ti’d (tied), 29: Anon., Alphonsus, Emperor of Germany (1594); Beaumont
and Fletcher, The Woman Hater (1679); Beaumont and Fletcher, Cupid’s
Revenge (1679); Fletcher, Bonduca (1647); Shakespeare and Fletcher,
The Two Noble Kinsmen (1679).
spellings in
a lover ’ s complaint
as evidence of authorship 145
ny (nigh), 57: Chettle and Munday, The Downfall of Robert Earl of
Huntingdon (1601); Middleton, The Second Maiden’s Tragedy (MS);
Shakespeare, A Midsummer Night’s Dream (1600).
greeuance (grievance), 67: Shakespeare, Othello (1622, 1623); Shakespeare,
Romeo and Juliet (1599, 1623).
parradise (paradise), 91: Shakespeare, Love’s Labour’s Lost (1598), twice.
ayds (aids), 117: Daniel, Cleopatra (1611).
forbidde (forbid), 150: Anon., Arden of Faversham (1592); Anon. Timon (MS)
twice; Dekker and Webster, Sir Thomas Wyatt (1607); Heywood, 1
Edward IV (1599); Shakespeare, A Midsummer Night’s Dream (1600);
Shakespeare, Richard II (1597).
sheelded (shielded), 151: Shakespeare, Cymbeline (1623).
perrils (perils), 158: Greene, Friar Bacon and Friar Bungay (1594).
patternes (patterns), 170: Anon., The Birth of Hercules (MS); Chapman,
Caesar and Pompey (1631); Dekker and Middleton, The Roaring Girl
(1611); ‘S. S.’, The Honest Lawyer (1616); Shakespeare, As You Like It
(1623); Shakespeare, Henry V (1623).
adulterat (adulterate), 175: Middleton, The Second Maiden’s Tragedy (MS).
teene (teen = grief, misfortune, misery), 192: Anon., The Stonyhurst
Pageants (MS); Fisher, Fuimus Troies (1633); Peele, Edward I (1593);
Shakespeare, Love’s Labour’s Lost (1598, 1623); Shakespeare, Richard
III (1597, 1623); Shakespeare, Romeo and Juliet (1597, 1599, 1623);
Shakespeare, The Tempest (1623).
encampt (encamped), 203: Shakespeare and others, 3 Henry VI (Richard
Duke of York) (1595).
tallents (talents), 204: Anon., 2 The Return from Parnassus, or The Scourge
of Simony (1603); Dekker, The Whore of Babylon (1607); Lodge, The
Wounds of Civil War (1594); Shakespeare, Cymbeline (1623); Shakespeare
and others, 3 Henry VI (Richard Duke of York) (1595); Shakespeare and
Fletcher, The Two Noble Kinsmen (1679).
receau’d (received), 206: Anon., Timon (MS); Chapman, All Fools (1605);
Field, Amends for Ladies (1618); Shakespeare, The Merchant of Venice (1600).
beseecht (beseeched), 207: Shakespeare, Hamlet (1604/5).
radience (radiance), 214: Shakespeare, All’s Well That Ends Well (1623);
King Lear (1608, 1623).
hewd (hued), 215: Chapman, The Mask of the Middle Temple and Lincoln’s
Inn (1613).
opall (opal), 215: Shakespeare, Twelfth Night (1623). subdew’d (subdued), 219: Alexander, Croesus (1637).
146
Determining the Shakespeare Canon
aulter (altar), 224: Anon., The Telltale (MS); Anon., The Troublesome Reign
of King John (1591); Anon., The Wars of Cyrus (1594).
hollowed (hallowed), 228: Heywood, The Rape of Lucrece (1608).
lunges (lungs), 228: Marston, The Dutch Courtesan (1605); Shakespeare,
Hamlet (1604/5); Shakespeare, Love’s Labour’s Lost (1598, 1623).
parcells (parcels), 231: Beaumont and Fletcher, The Captain (1647); Jonson,
The Alchemist (1612); Jonson, Cynthia’s Revels (1616); Shakespeare, As
You Like It (1623); Shakespeare, 2 Henry IV (1600).
mastring (mastering), 240: Mary Herbert, Antonius (1592); Middleton,
The Second Maiden’s Tragedy (MS).
giues (gyves), 242: Daniel, Philotas (1623); Dekker, Old Fortunatus (1600);
Dekker, The Whore of Babylon (1607); Shakespeare, Hamlet (1604/5);
Shakespeare, 1 Henry IV (1598); Shakespeare, Romeo and Juliet (1599).
filliall (filial), 270: Shakespeare, Hamlet (1604/5); Shakespeare, King Lear
(1623).
pangues (pangs), 272: Heywood, The Brazen Age (1613), five times; Percy,
The Fairy Pastoral (MS); Shakespeare and Wilkins, Pericles (1609);
Webster, The Duchess of Malfi (1623), twice.
shockes (shocks), 273: Chapman, Bussy D’Ambois (1607/8); Shakespeare,
Hamlet (1623).
battrie (battery), 277: Shakespeare and Wilkins, Pericles (1609).
leaueld (levelled), 282: Anon., Arden of Faversham (1592).
perticular (particular), 289: Chapman, Byron’s Conspiracy (1608);
Chapman, The Revenge of Bussy D’Ambois (1613); Dekker, Troia Nova
Triumphans (1612); Jonson, Bartholomew Fair (1640); Shakespeare,
Hamlet (1604/5); Shakespeare, 2 Henry IV (1600); Shakespeare, Othello
(1623); Shakespeare, Troilus and Cressida (1609).
invndation (inundation), 290: Marston, Jack Drum’s Entertainment (1601).
daft (doffed), 297: Shakespeare, Antony and Cleopatra (1623); Shakespeare,
1Henry IV (1598); Shakespeare, Much Ado About Nothing (1600, 1623).
gardes (guards, sb.), 298: Anon., Fair Em the Miller’s Daughter (1593);
Shakespeare, Love’s Labour’s Lost (1598); Shakespeare, Measure for Measure
(1623).9
diffrence (difference), 300: Heywood, The Fair Maid of the Exchange (1607);
Shakespeare, 2 Henry IV (1600).
straing (strange), 303: Shakespeare, Sir Thomas More Hand D (MS).
9. An instance of gardes in Richard II (1597) has been omitted, since the word is there a verb.
spellings in
a lover ’ s complaint
as evidence of authorship 147
preacht (preached), 315: Shakespeare and others, Edward III; Chapman,
Bussy D’Ambois (1607/8); Fletcher, The Noble Gentleman (1647);
Greene, George a Greene (1599).
spungie (spongy), 326: Greville, Alaham (1633); Peele, The Hunting of
Cupid (1591); Shakespeare, Macbeth (1623); Shakespeare, The Tempest
(1623); Shakespeare, Troilus and Cressida (1623).
Before I analyse this list, a few explanations are in order.10 Hyphened words
have, for search purposes, been broken into their separate components.
This is because in the course of a bibliographical study of Middleton’s The
Revenger’s Tragedy I found that George Eld’s compositors had an exceptional liking for hyphens, often using them quite arbitrarily.11 A Lover’s
Complaint contains some-times (22), him-selfe (116), and them-selues (117).
LION, for drama 1590–1614, yields thirty-eight examples of these forms
(eight some-times, ten him-selfe, and twenty them-selues), all but three of them
in eleven plays printed by Eld.12 Hyphenated downe-ward (284) occurs in two
Eld plays, namely Barry’s Ram Alley and Middleton’s The Revenger’s Tragedy,
and in two plays from the Jonson Folio of 1616, Catiline and Sejanus. Three
of the remaining hyphenated words in A Lover’s Complaint do not occur
within LION drama 1590–1614: out-brag’d (95), hart-wisht (314), fore-betrayed
(328). This leaves witch-craft (288), which is not rare, land-lord (140), found in
four plays, and by-past (158), an acceptable modern spelling), found only in
Alexander’s The Alexandraean Trilogy (1607). Several words that would be
hyphenated in modern texts are unjoined in the quarto printing of A Lover’s
Complaint. Also, on the subject of hyphens, LION finds shriking in Henry V
10. I excluded three instances of forbod (164) that appear in the anonymous A Warning for Fair
Women (1599) twice and in Wilson’s The Three Lords and Three Ladies of London (1590). This
form is quite common in poetry, doing duty for preterite (‘forbade’) or past participle (‘forbidden’). It occurs in Shakespeare’s The Rape of Lucrece (1594), 1648. But the instances in
drama all fall within the exclamation ‘Gods forbod’, where, according to OED the word is a
noun. I included sawne, which is, like forbod, a variant form rather than strictly a spelling: the
decision is inconsequential, since, among LION texts in the periods searched, it is unique
to A Lover’s Complaint. But LION does find it in Heywood’s translation of Ovid’s The Art of
Love (c.1625): ‘Much shamfull things haue in your sleep bin sawne’, where sawne again stands
for ‘seen’.
11. MacDonald P. Jackson, ‘Compositorial Practices in The Revenger’s Tragedy, 1607–08’, Papers
of the Bibliographical Society of America, 75 (1981), 157–70.
12. Barnes, The Devil’s Charter (1607); Barry, Ram Alley (1611); Chapman, Byron’s Conspiracy
(1608); Chapman, Byron’s Tragedy (1608); Chapman, The Mask of the Middle Temple
and Lincoln’s Inn (1613); Marston, Histriomastix (1610); Marston, What You Will (1607);
Middleton, The Revenger’s Tragedy (1607/8); Middleton, A Trick to Catch the Old One (1608);
Middleton, Your Five Gallants (1608); Shakespeare, Troilus and Cressida (1609).
148
Determining the Shakespeare Canon
(1623) as the second element of shrill-shriking, and hewd in Chapman’s Mask
(1613) as the second element of seueral-hewd. These were counted as legitimate additions to the record.
Past tense and past-participial endings pose problems. During the early
seventeenth century most spellings in -t for the phonetically ‘voiceless’
consonant ending [t]‌were giving way in print to -ed and -’d spellings—
preacht becoming less common than preached, for example.13 In A Lover’s
Complaint all endings after p (peept, encampt), sh (vndistinguisht, accomplisht, hart-wisht), x ( fixt, commixt), s (kist, promist), ck (crackt), and ch (toucht,
empleacht, beseecht, inricht, preacht)—all of which are phonetically [t]—have
the -t spelling. Although we would still use the [t] pronunciation, we
would now use the -ed or -’d spelling. The Complaint also uses burnt, which
remains an alternative to past participial burned. However, -t is not used
after a c that is phonetically [s]: peec’d, grac’d, forc’d. There is method here,
and the distinction between syllabic and non-syllabic endings in words that
in ordinary modern spelling would end -ed is carefully preserved, except
in the single case of vnconstraind, where a final -ed should be sounded as a
syllable.14 The important point for the present enquiry is that for some of
the verbs with -t endings in the Complaint, -t endings are very common
in LION plays of 1590–1614, occurring over a hundred times. Others are
rare, but this is sometimes because the verbs themselves are rare, whatever
the ending. I decided to include verbs with -t endings, when they qualified
as ‘rare’ because of their endings, so long as LION (for drama 1590–1614)
also yielded examples of the verb with one or more alternative endings.
This disqualified vndistinguisht, which occurred in four plays by different
authors, but never appeared as vndistinguished or vndistinguish’d.
The spelling ti’d is listed on the grounds that this use of the apostrophe
after a vowel is not modern practice: Riverside, for example, has tied. In the
other verbs listed with -’d endings, there are medial departures from modern spelling. No verbal ending in unapostrophized -d qualified as rare.
I have taken daft as a spelling of doffed. At Othello 4.2.175, the Folio of 1623
has Roderigo object to Iago ‘thou dafts me’, while the quarto of 1622 reads
‘thou dofftst me’ (that is, ‘you put me off’). The meanings of daff are the same
as those of doff, and OED has a cross-reference from the Shakespearean form
13. Alice Walker, Textual Problems of the First Folio (Cambridge: CUP, 1953), 153–6.
14.Past tense and past participle endings in Shakespearean quartos were discussed, along
with metrical spellings indicating elisions, by Hereward T. Price, ‘The First Quarto of
Titus Andronicus’, English Institute Essays 1947 (New York: Columbia UP, 1948; repr.
New York: AMS Press, 1965), 137–68.
spellings in
a lover ’ s complaint
as evidence of authorship 149
to the modern one. On the other hand, since eyne (15) remains the modern spelling of the archaic plural for ‘eyes’, instances have not been included,
although it occurs in the plays of only five playwrights, four of the plays being
Shakespeare’s.
The Complaint also employs metrical spellings, without apostrophes, to
indicate medial elision of a syllable. Most of these are either not present in
the drama of the period (sistring, emrald) or quite common (suffring, watrie,
battry), but four (laundring, mastring, battrie, diffrence) are used by fewer than
five playwrights.
The spelling invndation is included because v has been used internally.
Elizabethan–Jacobean printing practice distinguished v and u by position,
rather than phonetically, as now: v was used initially, u elsewhere. The
form invndation is exceptional in discounting the prefix in, since the period
affords many instances of inundation.
Inevitably, the inclusion of some of the spellings in the list is due as much
to the rarity of the words themselves as to the forms in which they appear.
But further phases of the investigation—checking poetry, for example—
should have a filtering effect on these. And, in any case, if the words are,
from a strictly lexical point of view, ones to which certain authors are more
partial than others, including them should not compromise the quest for
data serviceable for attribution.
Sorting examples of giues as a spelling of ‘gyves’ (meaning ‘fetters’) from
instances of the third person singular verb ‘gives’ was no easy matter. It
involved checking—sometimes in a context broader than the line provided
in the basic LION search results—762 items. Distinguishing between lettice
as ‘lattice’ and lettice as ‘lettuce’ or between daft as ‘doffed’ and daft as ‘foolish’, and identifying contexts where ny meant ‘nigh’ (rather than contributing to ‘hey no ny’) were less onerous undertakings.
Now that these preliminary explanations have been made, we can analyse the list. The following plays have three or more rare spelling links to A
Lover’s Complaint.15
Shakespeare
Shakespeare
Shakespeare
Hamlet (1604/5)
2 Henry IV (1600)
Love’s Labour’s Lost (1598)
5
4
4
15. I use the term ‘link’ merely as a shorthand way of indicating that a spelling is shared by A
Lover’s Complaint and another text. The term ‘hit’ might have served the same function.
Another detail of wording may also be worth explaining. To avoid clumsiness I sometimes
allude to spellings as being ‘used by’ such-and-such a writer: this is simply a shorthand way
of saying that it occurs within his or her works.
150
Determining the Shakespeare Canon
Shakespeare
Shakespeare
Middleton
King Lear (1623)
Romeo and Juliet (1599)
The Second Maiden’s Tragedy (MS)
3
3
3
Twelve Shakespeare texts have two links: All’s Well That Ends Well (1623),
As You Like It (1623), Cymbeline (1623), 1 Henry IV (1598), Henry V (1623), 3
Henry VI as Richard Duke of York (1595, with others), A Midsummer Night’s
Dream (1600), Othello (1623), Pericles (1609, with Wilkins), The Tempest (1623),
Troilus and Cressida (1609), The Two Noble Kinsmen (1679, with Fletcher).
Seven plays by other dramatists have two links: Anon., Arden of Faversham
(1592); Anon., Timon (MS); Anon., The Wars of Cyrus (1594); Chapman,
All Fools (1605); Chapman, Bussy D’Ambois (1607/8); Dekker, The Whore of
Babylon (1607); Jonson, The Alchemist (1616). Although I follow Annals in
designating Arden of Faversham anonymous, it is, of course, the burden of
Part One of this book that Shakespeare had a share in writing it.
A further ten Shakespeare plays, plus Hand D of Sir Thomas More, and
fifty-one non-Shakespearean plays register one link.
The four links to Pericles and The Two Noble Kinsmen are all to Shakespeare
scenes, though ti’d, found in the latter play, seems strongly associated
with the Beaumont and Fletcher Folios and the Pericles quarto is probably
printed from a memorially reconstructed text.16 The link to Edward III is
also to a Shakespeare scene featuring the Countess (2.1.356). One of the two
links to 3 Henry VI (the rare encampt) is to a Shakespeare scene, but though
it is recorded it has been discarded from computations, since it occurs
only in the doubtful 1595 quarto. The totals of links that I have given for
Shakespeare plays are to the texts with most links and completely disregard
links to others. For example, one of the spellings in Folio King Lear also
appears in the quarto (1608), but the quarto appearance is ignored. One
of Romeo and Juliet’s good quarto (1599) link-spellings also appears in the
doubtful quarto of 1597 and two appear in the Folio, while the 1597 quarto
has one spelling that is in neither of the other texts, but these appearances
outside Q 1599 (the foundation text for all modern editions) are ignored. In
counting links, I take the number of times a spelling occurs within a text
as irrelevant. Thus, although parradise occurs twice within Love’s Labour’s
Lost (1598), this constitutes one link: the presence of the spelling in the text,
rather than its recurrence, is the essential point.
16. Pericles: pangues, 3.1.13; battrie, 4.4.43; The Two Noble Kinsmen: ti’d, 1.3.42; tallents, 1.1.41.
spellings in
a lover ’ s complaint
as evidence of authorship 151
Shakespeare’s predominance in the results is glaringly obvious.
Moreover, the three play-texts that top the table with four or five links have
always been recognized as replete with spellings ‘referable to the author’,
as Greg put it in discussing Love’s Labour’s Lost (1598).17 The Oxford Textual
Companion endorses the orthodox view that Hamlet (1604/5) and 2 Henry
IV were set from holograph: ‘Wilson established that Q2 [Hamlet 1604/5]
was set from foul papers’; ‘Q [2 Henry IV 1600] is a good example of a text
printed directly from the author’s papers.’18 Love’s Labour’s Lost (1598) is
judged to have closely reprinted a lost earlier quarto that had been ‘set from
holograph (foul papers)’.19 Of the plays with three links Romeo and Juliet
(1599) is another typical ‘foul papers’ text.20 King Lear (1623)—which the
Oxford Shakespeare calls The Tragedy of King Lear and considers an authorial revision of the 1608 quarto’s The History of King Lear—is more problematical: the Textual Companion theorizes that copy was ‘Q2 (1619), annotated
from either holograph or scribal transcript of holograph’,21 but it seems to
me much more likely that, although Shakespeare began his revision on a
copy of Q2, he transcribed it, and his holograph, or a manuscript dependent
on it, served as printer’s copy.22
Presenting the results in terms of links to individual plays overcomes the
difficulty that some playwrights were far more productive during the set
period than others. It is hard to see why, if Shakespeare did not write A Lover’s
Complaint, three of his plays should each share more rare spellings with it than
17. Greg, Shakespeare First Folio, 220.
18.Wells and Taylor, Textual Companion, 398, 351; see also the ‘Summary of Control-Texts’,
145–7. The reference to Wilson is to J. Dover Wilson, The Manuscript of Shakespeare’s
‘Hamlet’ and the Problems of its Transmission, 2 vols. (Cambridge: CUP, 1934).
19. Textual Companion, 145, 270.
20. Textual Companion, 145, 288.
21. Textual Companion, 147.
22. The Second Maiden’s Tragedy, entitled The Lady’s Tragedy in Thomas Middleton: The Collected
Works, gen. eds. Gary Taylor and John Lavagnino (Oxford: Clarendon Press, 2007),
is the one anomaly among the six plays with three or more links to A Lover’s Complaint.
E. B. Everitt, The Young Shakespeare: Studies in Documentary Evidence: Anglistica 2
(Copenhagen: Rosenkilde and Bagger, 1954), 81–112, and Charles Hamilton, Cardenio, or,
The Second Maiden’s Tragedy (Lakewood, CO: Glenbridge, 1994), have both proposed that
the manuscript of The Second Maiden’s Tragedy is in Shakespeare’s handwriting, but have
convinced no expert palaeographers. Hamilton’s claim is part of a completely unconvincing case for identifying the play with the lost Cardenio by Shakespeare and Fletcher. Eric
Rasmussen, ‘Shakespeare’s Hand in The Second Maiden’s Tragedy’, Shakespeare Quarterly, 40
(1989), 1–26, suggested that fifty lines of the additional material inserted into the manuscript
on five separate slips of paper were written by Shakespeare, but his arguments were countered by MacDonald P. Jackson, ‘The Additions to The Second Maiden’s Tragedy: Shakespeare
or Middleton?’, Shakespeare Quarterly, 41 (1990), 402–5.
152
Determining the Shakespeare Canon
does any play by another playwright, and why plays by Shakespeare should
so dominate the list of those with two or more spelling links to the poem.
Seventeen of the thirty-nine Shakespeare plays have two or more links,
whereas only eight of the almost three hundred non-Shakespearean plays
have two or more links.
Thirty-one of the forty-five rare-in-drama spellings are employed in at
least one play by Shakespeare or in his share of the collaborative The Two
Noble Kinsmen, Pericles, or Edward III. Thirty-one is only three less than
the number of spellings employed by all non-Shakespearean plays combined, since ten of the spellings are peculiar to Shakespeare’s dramatic texts
(a twaine, greeuance, parradise, sheelded, beseecht, opall, radience, filliall, battrie,
straing). Otherwise, the dramatist whose plays use the largest number of
rare spellings is Chapman with seven, but only one of these (hewd meaning
‘hued’) is unique to him.
III
However, many of the spellings that are rare in drama may be fairly common in poetry, if only because words such as ‘teen’ and ‘nigh’, whatever
their spellings, belong to a ‘poetic diction’. The best way to determine
whether this is the case and to put poets and playwrights on an equal footing is to search a similar period within the LION poetry database, checking
all the A Lover’s Complaint spellings that were rare in drama or completely
absent from it, in order to compile a reduced list of spellings used by no
more than five writers, whether poets or playwrights. I determined on the
chronological limits 1593–1617 for ‘date of publication’, so that both John
Davies of Hereford’s and Shakespeare’s published verse volumes would
be among those searched. The total amounts of text in drama performed
1590–1614 and poetry published 1593–1617 are roughly equal.23
The Complaint spellings that fail to appear in LION drama 1590–1614
are: *plattid (plaited) 8; *sithed (scythed) 12; *gases (gazes) 26; flud (flood)
44; *greyned (grained) 64; satte (sat) 66; didde (did) 83, 127; *sawne (seen) 91;
23.The twenty-five year period for drama of 1590–1614 was chosen to cover Shakespeare’s
dramatic output, while the twenty-five year period for poetry was chosen to cover both
Davies’s and Shakespeare’s verse publications. Drama dates are for probable first performance (or of composition, if the play is not known to have been performed), whereas poetry
dates are necessarily for publication, which has to be later than composition.
spellings in
a lover ’ s complaint
as evidence of authorship 153
*mannad’g (manege) 112; addicions (additions) 118; forbod (forbade, forbidden) 164; *palyd (pallid) 198; *wepingly (weepingly) 207; emrald (emerald)
213; saphir (sapphire) 215; *manyfold (manifold) 216; *enpatrone (enpatron)
224; obaies (obeys) 229; subdewe (subdue) 248; *brynish (brinish) 284; *cautills
(cautels) 303. Asterisks mark those that cannot be found in LION poetry
1593–1617 either. The spellings flud and forbod each turn out to be used by
more than five poets, and so are discarded from the final combined drama
and poetry list of rare spellings.
I list below the spellings used by five or fewer writers, whether of
poems published 1593–1617 or plays first performed 1590–1614. Items from
the poetry database precede items from the drama database. Forenames
of authors are given only when two share a surname. The source of each
poetry spelling is identified by the volume from which the LION text was
copied, an individual poem within it being specified only in special cases.
Titles are modernized, except for a few proper names.
Spellings in A Lover’s Complaint (1609) found in five or fewer writers of
poems published 1593–1617 and plays first performed 1590–1614:
doble (double), 3: Weekes, Ballads and Madrigals in Five Voices (1598);
Anon., Caesar and Pompey, or Caesar’s Revenge (1607); Anon., Edmond
Ironside (MS).
a twaine (a-twain), 6: Shakespeare, King Lear (1623).
peept (peeped), 14: Fairfax, Godfrey of Bulloigne (1600); Niccols, The
Three Sisters’ Tears (1613); Barry, Ram Alley (1611); Middleton, A Trick
to Catch the Old One (1607) twice; Shakespeare, 2 Henry IV (1600);
Shakespeare, Romeo and Juliet (1597).
charecters (characters), 16: Garden, A Garden of Grave and Godly Flowers
(1609); Anon., The Wars of Cyrus (1594); Marston, The Entertainment at
Ashby (MSS).
laundring (laundering), 17: Jonson, The Alchemist (1616).
satte (sat), 66: Giles Fletcher, Licia (1593).
greeuance (grievance), 67: Chapman, The Whole Works of Homer (1616);
Leighton, The Tears and Lamentations of a Sorrowful Soul (1613); Ogle,
The Lamentations of Troy (1594); Shakespeare, Othello (1622, 1623);
Shakespeare, Romeo and Juliet (1599, 1623).
didde (did), 83 and 127: Shakespeare, Sonnets (1609) Sonnet 120.
parradise (paradise), 91: Chapman, Ovid’s Banquet of Sense (1595);
Shakespeare, Love’s Labour’s Lost (1598), twice.
154
Determining the Shakespeare Canon
ayds (aids), 117: Davies, Wit’s Bedlam (1617); Heywood, Troia Britannica
(1609); Warner, Albion’s England (1602); Warner, A Continuation of
Albion’s England (1606); Daniel, Cleopatra (1611).
addicions (additions), 118: William Rowley, ‘To his Friend John Taylor’ in
Taylor, The Nipping and Snipping of Abuses (1614).
forbidde (forbid), 150: Anon., Arden of Faversham (1592); Anon. Timon (MS)
twice; Dekker and Webster, Sir Thomas Wyatt (1607); Heywood, 1
Edward IV, (1599); Shakespeare, A Midsummer Night’s Dream (1600);
Shakespeare, Richard II (1597).
sheelded (shielded), 151: Shakespeare, Cymbeline (1623).
adulterat (adulterate), 175: Craig, Poetical Essays (1604); Egerton, England’s
Hope against Irish Hate (1600); Shakespeare, The Rape of Lucrece (1594);
Shakespeare, Sonnets (1609) Sonnet 121; Middleton, The Second
Maiden’s Tragedy (MS).
encampt (encamped), 203: Chapman, The Whole Works of Homer (1616);
Harrington, Orlando Furioso (1607); Shakespeare, 3 Henry VI (Richard
Duke of York) (1595).
beseecht (beseeched), 207: Shakespeare, Hamlet (1604/5).
emrald (emerald), 213: Baxter, Sir Philip Sidney’s Ourania (1606); Gorges,
Lucan’s Pharsalia (1614); Heywood, A Marriage Triumph (1613);
Peacham, The Period of Mourning (1613).
radience (radiance), 214: John Davies of Hereford, Microcosmos (1603);
Shakespeare, All’s Well That Ends Well (1623); Shakespeare, King Lear
(1608, 1623).
hewd (hued), 215: Chapman, The Mask of the Middle Temple and Lincoln’s
Inn (1613).
saphir (sapphire), 215: Heywood, A Marriage Triumph (1613); Marlowe and
Chapman, Hero and Leander (1598); Sabie, Adam’s Complaint (1596).
opall (opal), 215: Burel, To the Right High Lodowick Duke of Lennox (1595);
Niccols, The Cuckoo (1607); Weever, Faunus and Melliflora (1600);
Winter, The Third Day’s Creation (1604); Shakespeare, Twelfth Night
(1623).
hollowed (hallowed), 228: Churchyard, The Wonders of the Air (1602);
Heywood, The Rape of Lucrece (1608).
obaies (obeys), 229: Churchyard, A Sad and Solemn Funeral (1596); John
Davies of Hereford, Wit’s Pilgrimage (1605); Fairfax, Godfrey of Bulloigne
(1600), three times; Gordon, The Famous History (1615); Griffin,
Fidessa (1596).
spellings in
a lover ’ s complaint
as evidence of authorship 155
parcells (parcels), 231: Beaumont and John Fletcher, The Captain
(1647); Jonson, The Alchemist (1612); Jonson, Cynthia’s Revels (1616);
Shakespeare, As You Like It (1623); Shakespeare, 2 Henry IV (1600).
mastring (mastering), 240: Gorges, Lucan’s Pharsalia (1614); Warner,
Albion’s England (1602); Mary Herbert, Antonius (1592); Middleton,
The Second Maiden’s Tragedy (MS).
subdewe (subdue), 248: Warner, Albion’s England (1602), four times.
filliall (filial), 270: Shakespeare, Hamlet (1604/5); Shakespeare, King Lear
(1623).
leaueld (levelled), 282: Marlowe, Lucan’s First Book (1600); Anon., Arden of
Faversham (1592).
brynish (brinish), 284: Baxter, Sir Philip Sidney’s Ourania (1606).
invndation (inundation), 290: Copley, A Fig for Fortune (1596); Gordon,
The Famous History (1615); Heywood, Troia Britannica (1609); Marston,
Jack Drum’s Entertainment (1601).
daft (doffed), 297: Anon., The Passionate Pilgrim (1599) Poem 14; Weever,
The Mirror of Martyrs (1601); Winter, The Third Day’s Creation (1604);
Shakespeare, Antony and Cleopatra (1623); Shakespeare, 1 Henry IV
(1598); Shakespeare, Much Ado About Nothing (1600, 1623).
gardes (guards, sb.), 298: Anon., Fair Em the Miller’s Daughter (1593);
Shakespeare, Love’s Labour’s Lost (1598); Shakespeare, Measure for
Measure (1623).
straing (strange), 303: ‘J. C.’, Saint Mary Magdalene’s Conversion (1603);
Shakespeare, Sir Thomas More Hand D (MS).
spungie (spongy), 326: Chapman, Ovid’s Banquet of Sense (1595); Dymoke,
Caltha Poetarum (1599); Greville, Alaham (1633); Peele, The Hunting of
Cupid (1591); Shakespeare, Macbeth (1623); Shakespeare, The Tempest
(1623); Shakespeare, Troilus and Cressida (1623).
Shakespeare dominates this filtered list of rare spelling links to poetry
and drama texts no less clearly than the earlier list of links to drama alone.
He uses seventeen of the thirty-four surviving spellings (not counting
encampt in the 1595 quarto of 3 Henry VI ). There are twenty-eight links
to his works, seventy-five to all other poets and dramatists combined.24
24. This figure also ignores the link to the ‘bad’ 1597 quarto of Romeo and Juliet because it is not
the text by which the play is represented. As before, when a spelling is repeated within a
single work it is counted as just one link.
156
Determining the Shakespeare Canon
Yet Shakespeare’s canon covers only about 8 per cent of the amount of text
searched. We can no longer straightforwardly compare links to individual
works, since Chapman’s The Whole Works of Homer, for example, is enormous.
But it is notable that the same Shakespeare play-texts that were prominent in
the earlier list are prominent again: King Lear (1623) has three links, while 2
Henry IV (1600), Hamlet (1604/5), and Love’s Labour’s Lost (1598) have two each.
The only non-Shakespearean plays with as many as two links are Arden of
Faversham, Jonson’s The Alchemist, and Middleton’s The Second Maiden’s Tragedy
(The Lady’s Tragedy). In the filtered list, links to non-Shakespearean poetry
outnumber links to non-Shakespearean drama by fifty-one to twenty-three.
Yet there are twenty-five links to Shakespeare’s plays, more than to those of all
other playwrights combined.
John Davies of Hereford’s extensive poetic output affords only three
links, no more than afforded by Shakespeare’s poems, let alone his whole
canon. Davies’s Microcosmos, Wit’s Pilgrimage, and Wit’s Bedlam have one
link each, while there are two in Shakespeare’s Sonnets and one in his The
Rape of Lucrece.25 Yet Davies’s contribution to the poetry database is eight
times greater than Shakespeare’s, amounting to about five-twelfths the size
of Shakespeare’s complete works.26
The only authors whose total links outnumber those to Shakespeare’s
poems alone are Heywood with six (four from poetry), Chapman with five
(four from poetry), and Warner with four (all from poetry).27 Heywood’s
25. Vickers (Complaint, 228) claims that the Complaint and Davies’s Summa Totalis are the only
LION poems of 1580–1623 to share the spelling perticular, but LION finds it in Carew’s
Godfrey of Bulloigne (1594) and Chapman’s Eugenia (1614). For this reason, the spelling is
discarded from the drama-and-poetry list as occurring in the work of six writers (including Shakespeare in three separate plays). Presumably Vickers discounted the instances in
Carew’s and Chapman’s poems because they are nouns, not adjectives. If I were to discard them, Chapman’s Byron’s Conspiracy would also disappear from the list of drama links,
and the spelling would qualify for inclusion in the sifted drama-and-poetry list. Chapman,
Davies, Dekker, and Jonson would each gain one link, and Shakespeare would gain four—
to Hamlet (1604/5), 2 Henry IV (1600), Othello (1623), and Troilus and Cressida (1609). In the
drama-and-poetry results the good quartos of Hamlet and 2 Henry IV would be even further
ahead of all non-Shakespearean plays.
26. Estimates of the sizes of works, canons, and total text in the periods searched are based on
LION’s counts of the main high-frequency function words (the, and, to, and of ). These
give an approximate, but serviceable, guide to size in terms of the total number of words.
Shakespeare play-texts other than those serving as foundation text for most modern editions were subtracted from calculations. For Davies, we also have Vickers’s statement
that he wrote more than 42,000 lines of verse (Complaint, 202). The best line-counts for
Shakespeare’s plays are those of Alfred Hart, ‘The Number of Lines in Shakespeare’s Plays’,
Review of English Studies, 8 (1932), 19–28, since they are based on the old Cambridge edition,
in which prose and verse lines have an equal average number of words.
27.Chapman’s five exclude the link to Marlowe’s Hero and Leander (1598), which Chapman
‘completed’. The spelling saphir occurs within Marlowe’s section.
spellings in
a lover ’ s complaint
as evidence of authorship 157
searched corpus is about two-thirds the size of Shakespeare’s, his poetry
amounting to almost four times the size of Shakespeare’s. Chapman’s
searched corpus is only marginally smaller than Shakespeare’s, his poetry
amounting to almost twelve times the size of Shakespeare’s. Warner’s
poetry is more than three times the bulk of Shakespeare’s.
Once links to plays are taken into account, it is fairer to cite the number of spellings used by each author: Shakespeare seventeen, Heywood six,
Chapman five, Warner three.28 However the fact that the spellings occur
in no fewer than twenty-two separate Shakespeare works, differing in
their textual transmission, confirms their significance as evidence of his
hand: the author is the one agent common to all twenty-two.
Moreover, when we further reduce our list to the rarest spellings of
all—those employed by a single writer, and those employed by no more
than two writers—the predominance of Shakespeare becomes even more
striking. Shakespeare is the only author to use a twaine (King Lear), didde
(Sonnets), sheelded (Cymbeline), beseecht (Hamlet), and filliall (Hamlet, King
Lear). Other spellings to be used by a single author are laundring ( Jonson),
satte (Giles Fletcher), addicions (William Rowley), hewd for ‘hued’
(Chapman), subdewe (Warner), and brynish (Baxter). No poet or dramatist
is named in that list twice. Of spellings found in texts of two authors,
Shakespeare uses parradise (Love’s Labour’s Lost), radience (All’s Well That
Ends Well, King Lear), gardes sb. (Love’s Labour’s Lost, Measure for Measure),
and straing (Hand D of Sir Thomas More). Other users of two-author
spellings are parradise Chapman; radience Davies; hollowed for ‘hallowed’
Churchyard, Heywood; leaueld Marlowe, Anon. in Arden of Faversham;
gardes Anon. in Fair Em; straing ‘J. C.’.
So Shakespeare texts yield examples of nine of the seventeen rarest
spellings, with three of the seventeen occurring in two Shakespeare texts.
Chapman’s works have two of the seventeen, while no other writer’s works
have more than one.
It is possible to be even more selective. Seven of the seventeen Complaint
spellings are rare (used by five or fewer writers) within the whole LION
28. This is fairer because when a spelling occurs in more than one Shakespeare play it counts as
more than one link. If Warner’s long Albion’s England or Gordon’s long The Famous History
were divided into play-sized lengths, their four instances of subdewe and three of obaies,
respectively, may have been distributed in such a way that each counted as a link. These are
the only two poems thus affected. Neither of the rare spellings that link the Complaint with
Chapman’s The Whole Works of Homer occurs in that volume more than once.
158
Determining the Shakespeare Canon
database of over 350,000 works covering more than six centuries of English
poetry, drama, and prose. These are:
a twaine, used only in Shakespeare’s King Lear (1623).
sheelded, used in Shakespeare’s Cymbeline (1623), Anthony Munday’s
poem The Mirror of Mutability (1579), and George Whetstone’s prose
work An Heptameron of Civil Discourses (1582).
beseecht, used in Shakespeare’s Hamlet (1604/5), John Wilson’s play
Andronicus Comnenius (1664), and Richard Head’s prose work An
English Rogue, Part 1 (1665).
filliall, used in Shakespeare’s Hamlet (1604/5) and King Lear (1623),
Middleton and Rowley’s play The Old Law (1656), and John Reynolds’s
prose work The Triumphs of God’s Revenge (1623).
laundring, used only in Jonson’s The Alchemist (1616).
addicions, used in William Rowley’s commendatory poem to John
Taylor (1614) and (three times) in Thomas Warton’s The History
of English Poetry (1774–81), where he is, however, discussing the
‘Addicions’ in a work of 1413.
leaueld, used in the anonymous play Arden of Faversham (1592),
Marlowe’s poetic translation Lucan’s First Book (1600), ‘J. C.’’s poem
A Poor Knight his Palace (1579), and Sir Philip Sidney’s The Countess of
Pembroke’s Arcadia (1593).
So four of the seven spellings appear within Shakespeare’s works, filliall
in two plays. The works of no other writer have more than one. Hamlet
(1604/5) and King Lear (1623) contain two each. This is of particular interest because in an excellent essay on ‘The Spelling and Punctuation of
Shakespeare’s Time’ for the Oxford Shakespeare original-spelling edition,
Vivian Salmon points out that all three occurrences of the word ‘summit’ in
the Shakespeare canon, two in Hamlet (1604/5) and one in King Lear (1623),
take the form somnet, ‘an anomalous spelling not recorded elsewhere by the
Oxford English Dictionary and—as yet—not noted by scholars in any other
works of the period’.29 Somnet was evidently a Shakespearean spelling that
these two texts preserved, and Hamlet’s and King Lear’s filliall, King Lear’s a
29.Vivian Salmon, in Stanley Wells and Gary Taylor, gen. eds., William Shakespeare: The
Complete Works: Original-Spelling Edition (Oxford: Clarendon Press, 1986), liii. LION yields
no further instances of somnet.
spellings in
a lover ’ s complaint
as evidence of authorship 159
twaine, and Hamlet’s beseecht were doubtless also Shakespearean spellings,
shared in these cases with A Lover’s Complaint.
IV
If we now take a few steps back, I think we can further strengthen the
bond between A Lover’s Complaint and Shakespeare. Dover Wilson pointed
out that straing in Hand D’s pages of Sir Thomas More is supported by six
instances of straingers, and that the misprint ‘straying’ for ‘strange’ in Love’s
Labour’s Lost (1598), 5.2.763, can be neatly explained by the presence of straing
(as in A Lover’s Complaint) in the compositor’s copy.30 Wilson also thought
that King Lear (1608), like A Lover’s Complaint, contained one example of
the spelling addicions (1.1.136), but what looks like a c is probably a bent t.31
However, he cited addicions as a near match to Hand D’s adicion, listing such
similar formations from the Shakespeare good quartos as condicions, deuocion, oblacion, and peticioner.32 The Complaint’s rare hollowed for ‘hallowed’
would usually have been normalized by compositors, but evidence that it
is a Shakespearean spelling exists in Hollowmas for ‘Hallowmas’ in Richard
II (1597, 1623), 5.1.80. Somewhat similar is the case of doble (double), which
does not appear in canonical Shakespeare texts; nor does the near-match
dooble, which is also rare, appearing in the works of only ten authors in
the extended period 1580–1640. But ‘old Dooble’ survives as a name in 2
Henry IV (1600), 3.2.40, 52, rather in the way that the almost unique Hand
D spelling scilens survives eighteen times in the Justice ‘Scilens’ of the same
quarto.33 Also, Hand D’s pages afford analogous usages to the Complaint’s
rare invndation, with its medial v: advauntage, prevayle, especially.34
Again, while hewd (hued) turns out to be rare, hew (common in early
modern texts) is found not only in the Complaint but also in five sonnets
30. Wilson, ‘Bibliographical Links’, 127–8.
31. Wilson, ‘Bibliographical Links’, 135. Peter Blayney identified the letter as a damaged t in his
The Texts of ‘King Lear’ and their Origins, I (Cambridge: CUP, 1982), 511: it is Blayney’s ‘t1’
at B2v27.
32. Wilson, ‘Bibliographical Links’, 135–6.
33. For scilens/Scilens see Wilson, ‘Bibliographical Links’, 128–9. For its genuine rarity, see
Jackson, ‘Is “Hand D” of Sir Thomas More Shakespeare’s?’, 16 and n. 25. Early English Books
Online yields two further instances: in A Book of Prayers (1546) STC 3326.5 and The Holy
Bible (1568) STC 2099. But the spelling is not found in drama outside ‘Hand D’ and the 1600
quarto of 2 Henry IV.
34. See Greg’s transcript of Hand D’s pages, in Shakespeare’s Hand, 228–45, lines 71 and 81.
160
Determining the Shakespeare Canon
(20, 67, 82, 98, 104), three times in Venus and Adonis (1594), and a few
times in the plays, notably in ‘Cuckow-budds of yellow hew’ in Love’s
Labour’s Lost (1598), 5.2.896, while use of the plain -d ending after w is also
Shakespearean.35 Not only are the Complaint’s specific adulterat and straing
matched in Shakespearean texts, but omission of the final -e in such words
(as also in the Complaint’s: mannad’g) is a well-known feature of Hand D
and the quartos.36 The list of spellings rare in drama, 1590–1614, linked the
Complaint’s giues (gyves) to good quartos of Hamlet, 1 Henry IV, and Romeo
and Juliet; that giues was Shakespeare’s spelling is further suggested by the
appearance of giue (gyve) as a verb in Othello (1623), where the quarto (1622)
has ‘catch’ (2.1.170). More common, but not modern, Complaint spellings—
such as heare for ‘here’ (54, 197)—are amply illustrated in good Shakespeare
quartos.37 Finally, since John Davies of Hereford never uses the Complaint’s
shriking (shrieking), it may be worth adding that the instances in Henry V
(1623), as part of shrill-shriking, and Troilus and Cressida (1609) are supported
by the same spelling in The Phoenix and Turtle (5).
In addition, at least two textual errors in A Lover’s Complaint are readily explicable as misreadings of Shakespearean spellings. In ‘All ayds
them-selues made fairer by their place, | Can for addicions’ (117–18) editors customarily emend ‘Can’ to ‘Came’. Precisely the same error occurs
in Macbeth (1623), 1.3.98, suggesting that a copy spelling Cam led to simple
minim misreading (n for m). Wilson noted absence of final e after m in com
(come) in Hand D and three Shakespeare quartos, and in other words,
including nam (name), nam’s meaning ‘name is’ in King Lear (1608).38 The
other example involves the Complaint’s nun who to avoid temptation
wished to be enur’d (251). Editors rightly emend to ‘immured’, a word
used in the appropriate sense (‘imprisoned’, ‘walled in’) in Sonnet 84. The
error seems to have arisen from a Shakespearean spelling emured, which
appears in Love’s Labour’s Lost (1598), 3.1.124. In Troilus and Cressida (1623)
Troy’s walls are ‘strong emures’ (Pro. 8).
35.Compare, for example, showd in Titus Andronicus (1594), 2.3.98, and vnderualewd in The
Merchant of Venice (1600), 1.1.165.
36. Wilson, ‘Bibliographical Links’, 133–4.
37. Wilson, ‘Bibliographical Links’, 138, under xx.
38. Wilson, ‘Bibliographical Links’, 134.
spellings in
a lover ’ s complaint
as evidence of authorship 161
V
The last three paragraphs present subsidiary evidence of relatively minor
importance. But the main investigation, in its two phases, has produced
data that cannot, I think, be reconciled with John Davies’s authorship of
A Lover’s Complaint and overwhelmingly support Thorpe’s attribution to
Shakespeare. The poem’s rare spellings associate it more closely with texts by
Shakespeare than with texts by any other early modern poet or playwright,
and indeed by any other writer in any genre whose works are incorporated
within the LION database. This can only be due to common authorship.
At least, I can think of no plausible alternative explanation.39 The relevant
Shakespeare texts issued from many different printing-houses, and those
with most links to the Complaint have long been recognized as good sources
of probable Shakespeare spellings. The 1609 quarto of Shakespeare’s Sonnets,
in which A Lover’s Complaint was printed, was set by two compositors, who
had different habits of punctuation and different spelling preferences over a
few common words.40 But these individual differences prevented neither of
them from perpetuating some unusual copy spellings. A scribal transcript
may, or may not, have intervened between author and compositors, but this
possibility does not lessen the significance of the Complaint’s orthographical links to the Shakespeare canon.41 It could only do so if the scribe were
Shakespeare himself, copying another author’s poem. It is unthinkable that
John Davies of Hereford, a famous writing-master, having composed A
Lover’s Complaint, would have allowed Shakespeare to copy the poem so
that it could be included in Thorpe’s quarto of Shakespeare’s Sonnets under
the heading ‘A Louers complaint. | by | William Shake-speare’ or that
Shakespeare would have abetted such a deception.
The LION database is not perfect, and no doubt I have been guilty of a
few oversights here and there. But it would take a huge amount of error to
alter the overall picture. The investigation has proceeded according to rules
39. One explanation that might conceivably be suggested is eliminated in Appendix 5.
40. MacDonald P. Jackson, ‘Punctuation and the Compositors of Shakespeare’s Sonnets 1609’,
The Library 5th series, 30 (1975), 1–24; and ‘New Work on the Compositors of Shakespeare’s
Sonnets (1609)’, Shakespeare Newsletter, 48 (1998), 31–4.
41.Possible indications that printer’s copy for the actual sonnets was in two hands are summarized by Jackson, ‘Punctuation’, 13. But this evidence has no bearing on the question of
whether A Lover’s Complaint was set from holograph or a transcript.
162
Determining the Shakespeare Canon
and in a methodical manner that prohibits selection of links to Shakespeare
at the expense of links to other writers. John Davies’s voluminous works
contain several of the Complaint spellings that were found to be ‘rare’ in
drama, but all but two of these turned out not to be rare once poetry was also
searched. Although in the filtered drama-and-poetry list there are more than
twice as many links to non-Shakespearean poems as to non-Shakespearean
plays, there are only three links to Davies’s very large poetic corpus.
It is notable that all but one of the twelve instances of emrald, saphir, and
opall in the filtered list come from poems. Obviously this is mainly because
the jewels themselves, however spelled, are more often mentioned in poems
than in plays. As I have conceded, there is inevitably a lexical component
to this study, as well as the orthographical. But concentration on spellings adds an important dimension, because spellings are not readily subject
to imitation: nobody is likely to suggest that writing-master John Davies
‘imitated’ Shakespeare’s spellings—a twaine, didde, sheelded, beseecht, filliall,
and the like.
The notion that Davies imitated Shakespeare’s vocabulary and imagery
strikes me as equally improbable, but whether or not John Davies of
Hereford was ever a fervent imitator of Shakespeare—and Vickers’s evidence that Davies, in his acknowledged poems, echoed Shakespeare is
remarkably sparse42—spelling is sub-stylistic and not a feature of literary
texts that traditionally invites imitation, or even one that permits it when a
dramatic script is simply heard in performance. It is incredible that Davies
should have used in A Lover’s Complaint sixteen rare spellings that occur
in Shakespeare’s work but not in his own undoubted works, and only one
that occurs in his works but not Shakespeare’s. This new evidence strongly
reinforces the findings of Chapter 5.
VI
As a coda to this study, I set forth a further obstacle to believing that John
Davies of Hereford, rather than William Shakespeare, wrote A Lover’s
Complaint. It mainly concerns spellings of two common words and variant
4 2. Vickers, Complaint, 50–4, cites a few dubious echoes of Shakespeare’s sonnets, but, so far
as I can see, he cites no evidence whatsoever that in his acknowledged poems Davies borrowed from Shakespeare’s plays.
spellings in
a lover ’ s complaint
as evidence of authorship 163
Table 7.1 Contrast between Davies and Shakespeare in use of selected forms
Davies
heau’n
heauen
heau’ns
heauens
pow’r
power
powre
sith
since
321
57
123
25
302
52
125
1158
90
Shakespeare
0.849
0.151
0.831
0.169
0.630
0.109
0.261
0.928
0.072
7
642
5
189
2
301
42
24
467
0.011
0.989
0.026
0.974
0.006
0.872
0.122
0.049
0.951
Note: In Table 7.1 raw figures are followed by proportions: 321 instances of heau’n constitute 0.849 of
the total for both heau’n and heauen.
forms of another. To take the variant forms first, anybody reading Davies’s
poems as first printed must notice his extraordinary liking for sith as an
alternative to since. LION counts 1,158 instances of sith in his poems, but
only ninety of since. Sith thus occurs in Davies’s poems at a rate of once
every thirty-six lines. Shakespeare’s texts yield only twenty-four instances
of sith, but 467 of since. It is clear also that Davies favoured apostrophes
to indicate metrical elision in various words. I shall concentrate on two
that one encounters repeatedly. LION finds 321 instances of heau’n and
123 of heau’ns (as plural or possessive) in Davies’s poems, compared with
fifty-seven of heauen and twenty-five of heauens. Davies also has a strong
preference for pow’r (302 times) over power (52 times) or powre (125). The
apostrophized spellings of either of these two words are very rare within
Shakespearean texts. Table 7.1 presents the contrasting figures, with the
proportions of variant spellings or (in the case of sith/since) alternative forms
used within each writer’s corpus.43
43. Figures in this section derive from LION, with those for Shakespeare taken from the foundation texts (quarto or Folio) for The Riverside Shakespeare. Marvin Spevack, A Complete
and Systematic Concordance to the Works of Shakespeare, 9 vols. (Hildesheim: Georg Olms,
1968–80) was used as an aid to eliminating instances of ere as a contraction of ‘ever’ and of
checking other figures. I sorted out Davies’s examples of ere by viewing the contexts. LION
tallies for Davies are not perfect, because, as Vickers notes (Complaint, 287, n. 27), the copy
that the database used for Summa Totalis repeats twenty-eight stanzas, and when a relevant
word or spelling occurs within a title LION counts it, even if the title (as is very occasionally
the case) is also the poem’s first line. But minor inaccuracies cannot have appreciably altered
the proportions of different forms and spellings. It should also be recorded that Microcosmos
prints (besides the usual forms) heav’n or heav’ns (with internal v) seventy-three times, and
heaven or heavens twenty-two times.
164
Determining the Shakespeare Canon
The contrasts between Davies’s texts and Shakespeare’s texts have obviously not been created by compositors. A printer might very occasionally
have modernized the archaic sith to since: two instances of sith in Q2 (1604/5)
of Hamlet become since in F (1623), while one instance of sith in F Othello is
since in Q (1622). But the numerous examples of sith, spread over Davies’s
whole poetic canon—in books with a variety of printers and set by many
different compositors—are clearly his. It is no less clear that Shakespeare
seldom used this archaism. Similarly, it must be an authorial practice, not
shared by Shakespeare, that is reflected in the recurrence of heau’n, heau’ns,
and pow’r throughout Davies’s verse. Vickers’s book prints two plates showing Davies’s skill as a writing-master.44 The first reproduces the final page
of the Penshurst manuscript in which Davies copied out the Sidney Psalms.
There pow’r duly appears in the second line of Psalm 150. The other plate
shows a dedicatory poem of Davies’s own composition: the seventh line
contains his characteristic sith. These are Davies’s forms, not the creations
of transcribers or printers. The LION counts include the wording of marginal notes, and a large proportion of the spellings without apostrophe fall
within them. So within the verse itself the contrast with Shakespeare is
even a little greater than the table indicates.
But if we base calculations on the table alone, any one instance of the
word ‘heaven’ (as it is in modern spelling) is 6.550 times more likely to
appear as heauen in a Shakespeare text than in a Davies text (0.989 ÷ 0.151);
the spelling heauens (for ‘heavens’ or ‘heaven’s’) is 5.763 times more likely;
power (for ‘power’) is 8.000 times more likely; powre (for ‘power’) is 0.467
times as likely or less than half as likely; and the choice of since over sith is
13.208 times more likely.
A Lover’s Complaint has one instance of heauen (215), one of heauens (13),
two of power (74, 146), one of powre (260), and one of since (224). The other
options are never taken. Multiplying the probabilities, we find that the
combination of choices made in A Lover’s Complaint is just under 15,000
times more likely in a Shakespeare text than in a Davies text. Of course, in
literary texts, spellings and variant forms are not randomly distributed, so
that the calculation is offered simply as a rough guide. But what can be said
with some assurance—because I have checked—is that no Davies poem,
or portion of a Davies poem, whatever its length (be it one line or several
4 4. Vickers, Complaint, 18, 21.
spellings in
a lover ’ s complaint
as evidence of authorship 165
thousand lines), contains even single instances of heauen/heauens, power/powre,
and since (all three) but none of heau’n/heau’ns, pow’r, or sith.45
Vickers notes that William Browne of Tavistock’s pastoral collection,
The Shepherd’s Pipe (1614), prints an eclogue by Davies that was absent from
Alexander B. Grosart’s edition of Davies’s works.46 This 258-line poem is
crammed with rustic dialect and Spenserian archaisms. In ‘Look how breme
winter chamfers Earth’s bleake face; | So, corbed Eld accoyes youths surquedry’ (121–2)—in which breme winter, chamfers, corbed, accoyes, and surquedry are
all borrowed from Spenser’s February eclogue—Davies engages with Spenser
in a very different spirit from the author of A Lover’s Complaint, it seems to me.
But my reason for mentioning Davies’s eclogue in The Shepherd’s Pipe is that,
were we attempting to decide whether it was by Davies or Shakespeare, we
could base our conclusion on the fact that sith occurs six times, heau’ns twice,
and heau’n once, and that Shakespeare’s preferred variants do not appear at all.
(The word ‘power’ does not occur in any spelling.)
The eclogue also contains two examples of yer for ‘ere’ (meaning ‘before’)
and two of it’s (‘it is’). The archaic spelling yer is a Davies indicator, sprinkled
over several volumes and occurring thirty times altogether. Shakespeare never
uses it, though ere appears 386 times in his work. Further, the eclogue contains
two instances of it’s meaning ‘it is’. Davies texts employ it’s (whether with or
without the apostrophe) at well over twice the rate of tis (whether with or
without the apostrophe), whereas Shakespeare’s texts employ tis or ’tis, as a
contraction of ‘it is’, over forty times more frequently than its or it’s. So it is
reassuring that A Lover’s Complaint contains ere twice (5, 131) and tis once (70),
but not Davies’s indicators.47
45. The means of checking was to first enter ‘heauen AND power AND since’ into the search
box with Davies’s name as author. This produced only three poems that contained all three
of these forms: the ‘Preface’ to Microcosmos, in ‘Honour and Devotion unto . . . [King] James’,
‘Humour’s Heaven on Earth’, and ‘The Muses Tears for . . . Henry Prince of Wales’. The
contexts in which hits occurred were then read to see whether all three forms ever occurred
without at least one of Davies’s favourite alternatives intervening. The same checks were
carried out with ‘heauens’ replacing ‘heauen’ and ‘powre’ replacing ‘power’.
46. Vickers, Complaint, 70. Browne’s volume (The Shepheards Pipe in its original spelling) is
STC 3917.
47.Again, examples of it’s and its meaning ‘it is’ had to be sorted from examples of the possessive; and yer also required checking. The ‘bad’ octavo of 3 Henry VI, entitled The True
Tragedy of Richard Duke of York (1595), has ‘yer night’ where the authoritative Folio text has
‘ere night’ at 5.4.69. King Edward’s speech, in which the phrase appears, is abbreviated
and paraphrased in the octavo, and yer was probably introduced by somebody other than
Shakespeare.
166
Determining the Shakespeare Canon
Vickers has also discovered six short uncollected poems by Davies.48 The
first has none of the words discussed in this section, none of the six has any
form of the word ‘power’, and none has since, ere, or tis, But the second has
it’s. The third has sith and heaun, which lacks the apostrophe but marks the
metrical elision: heaun appears a dozen times in Davies’s works, never in
Shakespeare’s. The fourth has heau’n, as does the fifth. Only the sixth contains heauen, but it also has Davies’s yer. The contrast between the newly
added Davies material and A Lover’s Complaint is striking.
One further scrap of evidence may be mentioned.49 Three times within
the Complaint a word beginning with h is preceded by the indefinite article: a hill (line 1), a hell (288), a heart (309). Like the writer of the poem,
Shakespeare never uses an before any of these three nouns.50 Within his
works we can find a hill seven times, a hell nine times, and a heart thirty-six
times.51 Davies, in contrast, has a liking for an before words beginning with
h.52 For the three nouns that occur in the complaint, his figures are:
hell:
hill:
heart:
an 25
an 1
an 9
a 5
a 1
a 1
proportion of a = 0.167
proportion of a = 0.500
proportion of a = 0.100
Multiplying those proportions (0.167 × 0.500 × 0.100), gives the probability
of Davies’s using a hill, a hell, and a heart once each, but never using his preferred an, namely 0.00835. This is a one in 120 chance. Shakespeare’s known
practices create a 100 per cent chance of his preceding each of the three
nouns by a rather than an. So on these grounds alone Shakespeare is 120 times
more likely than Davies to have written the Complaint’s a hill, a hell, and a
heart. Such a calculation cannot be taken at face value, because, to repeat the
obvious proviso, a poet’s linguistic usages are not randomly distributed.53
48. Vickers, Complaint, 278–81.
49. First noted in MacDonald P. Jackson, ‘Shakespeare or Davies? A Clue to the Authorship of
“A Lover’s Complaint” ’, Notes and Queries, 254 (2009), 62–3.
50. Again, this was checked by means of Spevack’s concordance.
51. These figures include compounds such as ‘hell-broth’ and ‘a hell-hound’.
52. I have searched Davies’s work through LION. Davies’s normal spelling of ‘heart’ is ‘hart’;
I have excluded one instance where ‘an hart’ is a stag.
53. In his very last collection of poems, Wit’s Bedlam (1617), Davies avoided ‘an’ before ‘h’ and
used ‘a hill’, ‘a hell’ (twice), and ‘a heart’. But this was a volume of epigrams, in which he
must have decided, at the end of his career, to change his habitual usage. He would surely
have retained it for a formal love complaint published, like his The Holy Rood and Humour’s
Heaven on Earth, in 1609. In fact, if we discount Wit’s Pilgrimage, Davies never used ‘a heart’
or ‘a hill’. Before 1616 his ratios are hell 25:3; hill 1:0, heart 10:0 in favour of ‘an’.
spellings in
a lover ’ s complaint
as evidence of authorship 167
But once more the evidence indicates that Shakespeare, rather than Davies,
wrote A Lover’s Complaint.
It is probably necessary to stress again that the results reported in this
chapter cannot reasonably be dismissed by vague appeals to the undoubted
fact that compositors altered some, even many, authorial spellings. Only a
gigantic conspiracy among printing-house workers could have produced
the disproportionate number of links that were found between A Lover’s
Complaint and Shakespearean texts in the matter of rare spellings. And in
Appendix 5 I show that when a passage of Davies’s verse, of the Complaint’s
length, is subjected to the same systematic testing it has more links in rare
spellings to Davies’s works than to anybody else’s. The positive evidence
for Shakespeare’s authorship of the Complaint is presented in the first five
sections. This sixth section consolidates the case against Davies’s authorship, adducing evidence that is consistent with Shakespeare’s. I have noted
that the Sonnets quarto was set by two compositors, who differed in their
spellings of some words. ‘Power’ happens to be one of them: Compositor
A preferred to set flowre and powre, Compositor B to set flower and power.54
But this in no way invalidates the supplementary contra-Davies argument.
From the overall figures for Shakespearean texts it seems probable that
Shakespeare himself preferred the spelling power. That is the spelling on the
only occasion the word appears within Hand D’s pages of Sir Thomas More,
pages that I take to have been composed and penned by Shakespeare.55
Compositor B probably perpetuated the power spelling of his manuscript
copy and compositor A was probably responsible for the Complaint’s one
instance of powre. But even if it also derived from his manuscript copy, the
significant point is that neither compositor set Davies’s preferred pow’r, as
dozens of compositors repeatedly did when confronted with manuscripts
of his poems.
When type-setting The Revenger’s Tragedy (1607/8), George Eld’s compositors were not so high-handed in their treatment of their copy as to
obscure identifying features of Middleton’s orthography.56 But were we to
suppose that in dealing with A Lover’s Complaint they completely obliterated all traces of Davies’s characteristic heau’n, heau’ns, pow’r, sith, yer, it’s,
and an before h, we would still have to explain how they contrived at the
54. Jackson, ‘Punctuation’, 5.
55. Greg’s transcript in Shakespeare’s Hand, line 99.
56. Jackson, ‘Compositorial Practices’, 168–70.
168
Determining the Shakespeare Canon
same time to introduce such remarkable spelling links to Shakespearean
texts as a twaine, didde, sheelded, beseecht, and filliall. That they should have
transformed in this way a poem that shares more rare phrases and collocations with Shakespeare’s plays than with those of any other dramatist,
that exhibits some arresting lexical links to the Shakespeare canon, that
has the same Chinese-boxes narrative structure as Shakespeare’s The
Phoenix and Turtle, and that was published under the full name of William
Shakespeare—strikes me as incredible.57 It is surely much more reasonable
to conclude that Thomas Thorpe was telling the truth.
57.See Chapter 6 for phrases and collocations, and Jackson, ‘A Lover’s Complaint Revisited’,
277–8 for the Chinese-boxes structure. Lexical evidence is presented in Chapter 8.
8
Neologisms and
‘Non-Shakespearean’ Words in
A Lover’s Complaint
S
cholars arguing against Shakespeare’s authorship of a particular piece of writing sometimes advance as evidence its use of several
‘non-Shakespearean’ words, by which they mean words that do not appear
within Shakespeare’s undisputed works. For example, in 1954 Warren
D. Smith put forward the theory that the choruses in Henry V were not by
Shakespeare, but had been interpolated into the First Folio text by somebody else. He buttressed his case with the claim that the choruses contain
twenty-eight words ‘which appear nowhere else in Shakespeare’; he also
saw significance in the fact ‘that two or three words seem to be employed
by the choruses in senses other than Shakespeare gives them elsewhere’.1
Smith convinced no one. But Brian Vickers employs both these kinds
of argument in denying Shakespeare A Lover’s Complaint and reassigning
it to John Davies of Hereford. Much earlier, J. W. Mackail had enlisted the
evidence of ‘non-Shakespearean’ words in formulating his case against the
authenticity of A Lover’s Complaint.2
Mackail wrote decades before Alfred Hart published his painstaking
investigations into Shakespeare’s vocabulary.3 Hart counted the number of
different words in every Shakespeare play and in the poems and produced
1. Warren D. Smith, ‘The Henry V Choruses in the First Folio’, Journal of English and Germanic
Philology, 53 (1954), 38–57, at 55 and 56.
2. J. W. Mackail, ‘A Lover’s Complaint’, Essays and Studies, 3 (1912), 51–70.
3.Alfred Hart, ‘Vocabularies of Shakespeare’s Plays’ and ‘The Growth of Shakespeare’s
Vocabulary’, Review of English Studies, 19 (1943), 128–40 and 242–54. Also important was
Hart’s earlier ‘Shakespeare and the Vocabulary of The Two Noble Kinsmen’, Review of English
Studies, 10 (1934), 274–87.
170
Determining the Shakespeare Canon
tables showing how Shakespeare’s total vocabulary increased with each
new work. For every work he gave tallies for words that were peculiar to
it, that it shared with one other Shakespeare work, with two others, with
three others, and so on. He also provided, for a small selection of plays,
counts of ‘new’ words, meaning words for which a Shakespeare play or
poem was the source of OED’s first citation, though he recognized that
several of these words may have been in circulation beforehand and used
the term ‘new’ simply ‘for the sake of brevity’.4 He added remarks about the
frequency and nature of Shakespeare’s ‘new’ words.
Hart spelled out in some detail his working definition of a ‘word’. He was
concerned not with mere graphic units, as identifiable by computers, but
with parts of speech and with meanings. His ‘method of enumeration was,
in the main, based on the principles adopted by the editors of The Oxford
English Dictionary’.5 For his purposes, a word was, with minor exceptions
that were carefully explained, an OED headword. Thus identical graphic
units with completely distinct senses or functions were distinguished, but
different inflexions of the same verb came under a single heading, and
whether a noun was singular or plural was immaterial. Hart compiled his
counts from Alexander Schmidt’s excellent Shakespeare Lexicon, but corrected a few obvious errors and misunderstandings.6
Hart showed that, as one might have expected, every Shakespeare play
or long poem contains many words that do not appear elsewhere in the
canon, the percentages of ‘peculiar’ words among the total number of different words ranging from 3.2 in Julius Caesar to 10.4 in King Lear.7 He found
that Fletcher’s share of the collaborative The Two Noble Kinsmen contained
a smaller proportion of peculiar words than Shakespeare’s and that plays
by Kyd and Marlowe were similarly apt to compare unfavourably with
Shakespeare in this respect.8 About Shakespeare’s ‘new’ words (affording
OED’s first citations) the data he offered were more patchy. But he did note
that Shakespeare’s share of The Two Noble Kinsmen contained twenty-seven
new words and Fletcher’s only seven, while also countering J. M.
4. Hart, ‘Kinsmen’, 278.
5. Hart, ‘Vocabularies’, 129.
6.Alexander Schmidt, Shakespeare Lexicon, 2 vols., rev. Gregor Sarrazin (Berlin: Georg
Reimer, 1902).
7. Hart, ‘Vocabularies’, 132. There are a few small differences between ‘Vocabularies’, 132, and
‘Growth’, 249, over the number of words in certain plays. Both articles have notes regretting that the author was unable to read proofs. I have used the figures in ‘Growth’ for my
calculations, since it seems likely that the later article corrected mistakes in the earlier one.
8. Hart, ‘Kinsmen’, 275–8; ‘Vocabularies’, 138.
Neologisms and ‘Non-Shakespearean’ Words
171
Robertson’s claim that Chapman, his candidate for authorship of A Lover’s
Complaint, ‘is the supreme neologist among the poets of his time’: having
made a thorough search of Chapman’s twelve plays, he cited figures that
proved Shakespeare’s rate of occurrence of new words to be far higher than
Chapman’s.9
In an analysis of the vocabulary of A Lover’s Complaint that I published in
1965, I used Hart’s findings to rebut Mackail.10 It seemed to me that once the
prop of ‘too many non-Shakespearean words’ was removed, Mackail’s argument collapsed—that the proportions of peculiar and new words in the poem
were compatible with Shakespeare’s authorship and that its ‘coinages’ were of
a Shakespearean kind. I showed that the Complaint’s links in rare words were
overwhelmingly with Shakespeare’s seventeenth-century plays and deduced
that the poem was a work of Shakespeare’s maturity, not of his youth.
Vickers agrees about the date, but rejects my other conclusions.11 He
rightly points out that there has been significant research in lexicography
since Hart’s day; that Shakespeare’s works were processed more carefully
and comprehensively for OED than those of many of his contemporaries;
and that Hart examined only a very small sample of authors apart from
Shakespeare. He argues that, on the one hand, the extent of Shakespeare’s
contribution to the English lexicon has been overestimated and, on the
other hand, that A Lover’s Complaint contains too many new words to be
Shakespeare’s. He also seeks to demonstrate that the Complaint’s coinages
are typical of the period, rather than simply of Shakespeare.
Vickers understandably admires the work of two scholars, Jürgen
Schäfer and Bryan Garner, and draws on their findings.12 He produces a
9. Hart, ‘Kinsmen’, 278–9.
10. Jackson, Shakespeare’s ‘A Lover’s Complaint’.
11. Vickers does not devote any particular section of his book to discussion of the poem’s date
of composition, but at various points places this later than (a) the publication of Holland’s
translation of Pliny (1601) (Vickers, Complaint, 235–6), (b) the publication of the 1604/5
quarto of Hamlet (207–8), (c) the first performance of Antony and Cleopatra (1606) (168),
(d) the publication of the 1608 quarto of King Lear (208), and even (e) the first performance
of Cymbeline (1609) (213). The last three suggestions are offered only as possible explanations
of apparent links between the Shakespeare plays and the Complaint. In any case, Vickers
appears to endorse a ‘Jacobean’ date (208) ‘between about 1603 and 1609’ (213).
12. Jürgen Schäfer, Documentation in the OED: Shakespeare and Nashe as Test Cases (Oxford:
Clarendon Press, 1980); Shakespeares Stil: Germanisches und romanisches Vokabular
(Frankfurt: Athenäum, 1973); Early Modern English Lexicography, 2 vols. (Oxford: Clarendon
Press, 1989). In this chapter my figures and calculations from Schäfer are taken from
Documentation. Bryan Garner, ‘Shakespeare’s Latinate Neologisms’, Shakespeare Studies, 15
(1982), 149–70, repr. in Vivian Salmon and Edwina Burness, eds., A Reader in the Language
of Shakespearean Drama (Amsterdam and Philadelphia: John Benjamins, 1987), 207–28; my
references are to this reprint.
172
Determining the Shakespeare Canon
table showing the results of Garner’s meticulously created inventory of
‘Shakespeare’s Latinate neologisms’: Vickers calculates that they occur at
a rate of one every 29.9 lines in A Lover’s Complaint, one every 231.9 lines
in The Rape of Lucrece, one every 398.0 lines in Venus and Adonis, one every
430.8 lines in the Sonnets, and one every 187.3 lines in the whole canon of
plays.13 This may seem devastating to any claim that Shakespeare wrote A
Lover’s Complaint. But the figures are misleading, for the following reasons:
(a) The raw counts derived from Garner’s list are inaccurate. The numbers of neologisms in each row of Vickers’s table should be 12, 12, 3, 9, 584
(not 11, 8, 3, 9, 604).14 Garner lists 624 neologism altogether, including 4 in
The Phoenix and Turtle, ignored by Vickers. This is a trivial matter. It is hard
to get such counts exactly right.
(b) The table compares a 329-line poem with much longer poems and
with 113,136 lines of drama, without taking any account of the great variation between plays, let alone 329-line sections of Shakespearean verse.
(c) The table gives no figures for The Phoenix and Turtle, which is the
Shakespearean poem closest in date to A Lover’s Complaint. Both are
seventeenth-century works, whereas Venus and Adonis was published in
1593 and The Rape of Lucrece in 1594. Admittedly The Phoenix and Turtle is
short, but so is the Complaint compared to Lucrece, for example.
(d) The most important point—one that I myself overlooked in 1965—is
that the number of lines per neologism roughly indicates the proportion of
neologisms to a text’s length in terms of ‘tokens’ or total number of words
(iambic pentameter lines, and even prose lines as printed in some editions,
being of similar word-length), but the proportion ought properly to be of
neologisms per ‘types’ or different words. This is because the ratio of types
to tokens decreases as text-length increases, and yet neologisms will be
drawn overwhelmingly from individual types (different words).
Consider Shakespeare’s ‘Epitaph on Himself ’:
Good friend, for Jesus’ sake forbear
To dig the dust enclosèd here.
13. Vickers, Complaint, 151.
14. One mistake is Garner’s: he accidentally attributes ‘fount’ to A Lover’s Complaint instead of
to The Rape of Lucrece.
173
Neologisms and ‘Non-Shakespearean’ Words
Table 8.1 Types and tokens in Shakespeare poems
Tokens
The Phoenix and
Turtle
A Lover’s Complaint
Venus and Adonis
The Rape of Lucrece
Sonnets
All non-dramatic
poetry
Types
Ratio
352
217
1.62
2,563
9,730
14,548
17,520
47,824
1,089
2,574
3,536
3,239
6,797
2.35
3.78
4.11
5.41
7.04
Note: Table 8.1 shows, for each work or body of work, the number of tokens (total words), types
(different words), and ratio of tokens to types.
Blessed be the man that spares these stones,
And cursed be he that moves my bones.15
The poem contains twenty-eight words (tokens) and only three (‘the’, ‘be’,
‘that’) are repeated, so that there are twenty-five different words (types),
giving a ratio of 1.12 tokens to every type. Taking counts of tokens and types
from Marvin Spevack’s concordance, we find that in Shakespeare’s other
poems and the disputed A Lover’s Complaint the ratio increases with length
(Table 8.1).
However, Spevack’s concordance distinguishes words as graphic units,
so that ‘find’, ‘finding’, and ‘found’, for example, are given separate entries,
as are ‘man’, ‘man’s’, and ‘men’. Both Garner and Schäfer, like Hart before
them, enumerate words according to OED principles. So Hart’s figures for
the numbers of different words (types) in each Shakespeare work are the
appropriate ones to use.16 Table 8.2 shows the Shakespeare plays or poems
with the five highest and five lowest rates of Garner’s Latinate neologisms
per 1,000 Hart types. Hamlet (1600–1) and Troilus and Cressida (1602) afford
rates slightly higher than that for A Lover’s Complaint, on which composition may well have begun at about the same time.17 Shakespeare is generally agreed to have written The Phoenix and Turtle shortly before its
publication in 1601, and although it is not long enough for its proportion
15. Wells and Taylor, Complete Works, 783.
16. Hart, ‘Growth’, 249.
17.A. K. Hieatt, T. G. Bishop, and E. A. Nicholson, ‘Shakespeare’s Rare Words: “Lover’s
Complaint”, Cymbeline, and Sonnets’, Notes and Queries, 232 (1987), 219–24, argue that
Shakespeare began the poem around 1600–3 but did not finish working on it until he prepared it for inclusion in the 1609 Sonnets quarto.
174
Determining the Shakespeare Canon
Table 8.2 Latinate neologisms in selected Shakespeare works
Neologisms
The Phoenix and
Turtle
Troilus and Cressida
Hamlet
A Lover’s Complaint
Love’s Labour’s Lost
Pericles
Henry VIII
The Winter’s Tale
3 Henry VI
Venus and Adonis
Word types
Proportion
4
213
18.8
48
54
12
35
6
6
6
4
3
3,360
3,882
952
2,872
2,442
2,659
2,965
2,790
2,096
14.3
13.9
12.6
12.2
2.5
2.3
2.0
1.4
1.4
Note: Columns from left to right show (a) title of work, (b) number of Latinate neologisms according
to Garner, (c) number of different word-types according to Hart, (d) proportion of Latinate
neologisms per 1,000 types.
of neologisms to be very reliable, it is the Complaint’s nearest contemporary
among Shakespeare’s undisputed poems, and it heads the table. Moreover,
Hart would probably have added the word ‘distinct’ as a noun (in the plural
in The Phoenix and Turtle, line 27), since it derives from the Latin distinctus,
past participle of the verb distinguere, and OED cites this as the first instance
of substantival use. It is notable that among the plays with fewest Latinate
neologisms are the two collaborations, Pericles and Henry VIII: besides, the
Shakespeare–Fletcher tragicomedy, The Two Noble Kinsmen, has one fewer,
five, but I have ignored it because Hart did not include the play in his main
tables. As explained in Chapter 3, Section III, Shakespeare is unlikely to
have been the sole author of 3 Henry VI.
There is nothing un-Shakespearean, then, about the number of Latinate
neologisms in A Lover’s Complaint. Even if we reckon in terms of the number of lines to each Latinate neologism, stretches of Hamlet and Troilus and
Cressida match A Lover’s Complaint. The Complaint has one per 25.3 lines (13
in 329 lines). The first 303 lines of Hamlet (up to ‘Exeunt all but Hamlet’ after
1.2.128) have one per 25.3 lines (12 in 303 lines). The first 214 lines of Troilus
and Cressida, 1.3 (up to the entry of Aeneas) have one per 19.5 (11 in 214
lines). In my 1965 study I had mentioned Troilus and Cressida’s Latinisms as
matching the Complaint’s. Vickers objects that the play ‘is hardly an acceptable precedent for the vocabulary of a love poem’.18 I am not sure why: the
18. Vickers, Complaint, 146.
Neologisms and ‘Non-Shakespearean’ Words
175
Complaint and Troilus and Cressida both deal with deceitful lovers, and there
seems no a priori reason why Shakespeare should not use Latinisms in a
narrative poem composed by him in the seventeenth century. In any case,
The Phoenix and Turtle is a poem about love, and it has an even higher rate
of Latinate neologisms than the Complaint, while the twelve in the first 303
lines of Hamlet fall within two strikingly contrasted scenes—on the battlements of Elsinore on a cold night and within the glitter of Claudius’s court.
Garner benefited from Jürgen Schäfer’s important investigations into
OED’s inevitable shortcomings. After conducting supplementary searches
of a range of pertinent texts, Schäfer was able to publish a list, correcting
OED, of words that were first used in print by Shakespeare.19 Vickers writes
that ‘Given Schäfer’s extensive researches, some weight must be given to
his conclusion that the ratio of neologisms in A Lover’s Complaint, nine per
1,000 words, would represent an absolute peak for Shakespeare—if indeed
he wrote it.’20 But if, as with Garner’s figures, we compute the proportion
of Schäfer’s neologisms among different words (types) for each Shakespeare
play and poem, we obtain the results shown in Table 8.3 for the five highest
and five lowest works. The proportion remains above twenty for a further
six works: The Merry Wives of Windsor 26.5, 1 Henry IV 24.8, Romeo and Juliet
24.3, Macbeth 22.6, Troilus and Cressida 22.0, Othello 20.9, 2 Henry IV 20.8,
Venus and Adonis 20.0. The Phoenix and Turtle comes slightly lower with
18.8, but would have attained 23.5 had ‘distinct’ been added. Again the collaborations Henry VIII and Pericles and the probable collaboration 3 Henry
VI are among the five lowest.
The Complaint’s rate of neologisms is very similar to Hamlet’s and below
that for Love’s Labour’s Lost. The comedy is avowedly ‘a great feast of languages’ (5.1.36–7). But there is another factor contributing to the high positions of Love’s Labour’s Lost, A Midsummer Night’s Dream, and probably A
Lover’s Complaint on the table: the dates of composition assigned by OED
to all three are too early. Love’s Labour’s Lost is dated 1588, A Midsummer
Night’s Dream is dated 1590, and A Lover’s Complaint 1597. The Oxford
Shakespeare’s Textual Companion assigns Love’s Labour’s Lost to 1594–5, A
Midsummer Night’s Dream to 1595, and A Lover’s Complaint to 1603–4. The
poem may have been begun a little earlier, as we have seen, but it may have
been given its finishing touches not long before publication in 1609. The
19. Schäfer, Documentation, 87–136. My figures from ‘Schäfer’ all derive from this list.
20. Vickers, Complaint, 148.
176
Determining the Shakespeare Canon
Table 8.3 All neologisms in selected Shakespeare works
Love’s Labour’s Lost
A Lover’s Complaint
Hamlet
A Midsummer
Night’s Dream
King Lear
Julius Caesar
Much Ado About
Nothing
3 Henry VI
Pericles
Henry VIII
Neologisms
Words
Proportion
127
31
115
65
2,872
952
3,882
2,363
44.2
32.6
29.6
27.5
89
19
20
3,339
2,218
2,396
26.7
8.6
8.3
23
20
14
2,790
2,442
2,659
8.2
8.2
5.3
Note: Columns from left to right show (a) title of work, (b) number of neologisms according to
Schäfer, (c) number of different word-types according to Hart, (d) proportion of neologisms per
1,000 types.
Complaint and the two comedies have probably, therefore, been credited
with some first citations that are not rightfully theirs. Troilus and Cressida,
on the other hand, for which the OED dating is 1606 and the Oxford
Shakespeare dating 1602, should probably have been credited with a few
more. In any case, all things considered, A Lover’s Complaint cannot be said
to have too many new words to be Shakespeare’s.
What about words that do not occur in the acknowledged canon? Are
there too many ‘non-Shakespearean’ words? Hart himself tabulated the
percentage of each play’s and narrative poem’s peculiar words among its
different words or types, and the range is from Julius Caesar’s 3.2 to King
Lear’s 10.4. He gave no figures for either A Lover’s Complaint or The Phoenix
and Turtle, but they are short enough for the total number of different words
in each and the number of words not found elsewhere in the canon to be
quite quickly determined according to Hart’s criteria.21 Table 8.4 shows the
21. In Shakespeare’s ‘A Lover’s Complaint’, 12 and n. 10, I misunderstood Hart’s statement in
‘Vocabularies’, 131, as meaning that he had excluded from his counts of words peculiar
to a single Shakespeare work all compounds unless one of the hyphenated elements was
itself ‘peculiar’. But it is now clear to me that he included compounds in his tallies. So
my figure of 49 Complaint words not used in undisputed Shakespeare works is augmented
by the compounds signalled as ‘not used’ in my list of 1965 (8–12). The total of 61 is in
almost exact agreement with that of Donald W. Foster, Elegy by W. S.: A Study in Attribution
(Newark: University of Delaware Press, 1989), 94. Hart’s tables lump A Lover’s Complaint
and The Phoenix and Turtle together with the Sonnets and (probably) The Passionate Pilgrim.
177
Neologisms and ‘Non-Shakespearean’ Words
Table 8.4 Peculiar words in Shakespeare poems
Venus and Adonis
The Rape of Lucrece
The Phoenix and
Turtle
A Lover’s Complaint
Peculiar words
Word-types
Percentage
101
167
10
2,096
2,826
213
4.8
5.9
4.7
61
952
6.4
Columns from left to right show (a) title of work, (b) number of peculiar words, i.e. words found
only in the one Shakespeare (or disputed) work, according to Hart in first two rows and Jackson
in second two rows, (c) number of word-types according to Hart in first two rows and Jackson in
second two rows, (d) percentage of peculiar words.
results for Shakespeare’s poems. Seventeen of the thirty-seven plays have
percentages of peculiar words (found only in the one Shakespearean work)
of 6.4 or higher. These are King Lear 10.4, Hamlet 10.2, The Merry Wives
of Windsor 9.1, Troilus and Cressida 9.0, 1 Henry IV 8.9, Love’s Labour’s Lost
8.8, The Tempest 7.9, Antony and Cleopatra 7.8, Henry V 7.8, Coriolanus 7.8, 2
Henry IV 7.8, Othello 7.4, The Winter’s Tale 7.4, Macbeth 7.4, Romeo and Juliet
6.8, A Midsummer Night’s Dream 6.8, and Cymbeline 6.7. A Lover’s Complaint is
completely normal in this regard. Notably, the high-scoring plays include
none of the eight earliest: they are mainly works of Shakespeare’s maturity,
like the Complaint.
Vickers also mentions the computerized tests of Ward E. Y. Elliott and
Robert J. Valenza, in which the eighty-eight peculiar words detected in A
Lover’s Complaint put it just outside the upper range for 95 per cent of their
Shakespearean samples.22 Elliott and Valenza’s programs, like Spevack’s
concordance, employ the ‘graphic units’ definition of a ‘word’, which
explains why their figure for the Complaint is much higher than mine. In
1996 and 1999 their computer counted 2,579 tokens in A Lover’s Complaint,23
which is close to Spevack’s 2,563; Spevack lists 1,089 different words or
types (compared with the 952 derived from OED principles). In an email
sent to me on 17 July 2003, Elliott reported that a recent count of Complaint
words not found in canonical works put the total for peculiar words at
22. Vickers, Complaint, 152. Elliott and Valenza, ‘Did Shakespeare?’, 124.
23.Elliott and Valenza, ‘And Then’, appendix 4; ‘The Professor Doth Protest Too Much,
Methinks: Problems with the Foster “Response” ’, Computers and the Humanities, 32 (1998),
425–88, appendix 4.
178
Determining the Shakespeare Canon
eighty-six and so ‘just inside our Shakespeare profile’, but a further recount
restored the figure of eighty-eight.
But Elliott and Valenza deal with such words through the ‘novel, fancy,
high-tech tests’ that were devised by Ronald Thisted and Bradley Efron
and are difficult to explain in ‘nontechnical terms’.24 In Thisted–Efron tests,
words peculiar to a particular text are, rather confusingly, called ‘new words’,
because the statistical procedure was originally designed to determine how
many non-concordance words ‘would appear in a hypothetical, newly discovered Shakespeare poem or play of a given length’.25 The number of words
peculiar to a given text is subtracted from the number of words ‘expected’
according to Thisted–Efron calculations and so is transformed into a score
that is negative for texts with large proportions of peculiar words and positive for texts with small proportions. For whole plays there is a fair correlation
between a rank-ordering from (a) Elliott and Valenza’s largest minus figure to
their largest plus figure and (b) the percentage of Hart’s peculiar words (found
in a single Shakespeare work) among his total of different words or types.26
In ‘Are the Thisted–Efron Authorship Tests Valid?’ Valenza tested Romeo
and Juliet, Othello, and King Lear against Shakespeare’s early, middle, and
late dramatic corpora. He concluded that ‘the new words test always shows
greatest consistency of a given play with other plays of the same period and,
moreover, shows good consistency of each work with the entire corpus
of plays except for the case of King Lear, which manifests many more new
words than predicted’.27 Extending the tests to Venus and Adonis, The Rape
of Lucrece, and the Sonnets, he discovered ‘that poems are largely inconsistent with plays and possibly with other poetry by the same author’.28 He
suggested that not only was the ‘base lexicon’ for poems too small, but that
perhaps ‘the constraints which apply to the composition of poetry distort
the processes by which one chooses one’s words to such an extent that the
idealized assumptions of the Thisted–Efron tests diverge too far from reality to yield meaningful results’.29 He also concluded that ‘even in the case
2 4. Elliott and Valenza, ‘Glass Slippers’, 191.
25.Elliott and Valenza, ‘Did Shakespeare?’, 119. See also ‘And Then’, 197–8 and ‘Glass
Slippers’, 192.
26. Thisted–Efron ‘new word’ scores for whole plays are tabulated in Elliott and Valenza’s ‘And
Then’, appendix 1 (S) and appendix 1 (A), and in ‘Oxford by the Numbers’, 409 and 411.
27.Robert J. Valenza, ‘Are the Thisted–Efron Authorship Tests Valid?’, Computers and the
Humanities, 25 (1991), 27–46, at 38.
28. Valenza, ‘Thisted–Efron Authorship Tests’, 45.
29. Valenza, ‘Thisted–Efron Authorship Tests’, 38.
Neologisms and ‘Non-Shakespearean’ Words
179
Table 8.5 Thisted–Effron scores
Hamlet Block 2
The Tempest Block 3
Troilus and Cressida Block 1
Antony and Cleopatra Block 4
The Winter’s Tale Block 1
Venus and Adonis Block 1
Hamlet Block 6
Troilus and Cressida Block 5
1 Henry IV Block 2
Antony and Cleopatra Block 4
King Lear Block 3
Macbeth Block 1
Troilus and Cressida Block 2
Othello Block 1
Hamlet Block 1
Venus and Adonis Block 2
A Lover’s Complaint
–44
–40
–39
–38
–33
–32
–32
–32
–32
–30
–29
–28
–28
–27
–26
–26
–33
Note: Table 8.5 shows the lowest Thisted–Efron scores for
3,000-word blocks, as reported by Elliott and Valenza. The
lower the score, the greater the excess of observed over
expected ‘new words’, where ‘new’ means not used elsewhere
in the Shakespeare canon.
of plays versus plays with ample sample and base lexicon sizes we have the
possibility of statistical outliers’.30
In their later studies Elliott and Valenza refer to Valenza’s analysis but
seem no longer bothered by the serious reservations he expressed. From
fourteen Shakespeare plays they created eighty-two 3,000-word blocks
of text and from the two early narrative poems and the Sonnets a further
twenty-seven.31 They also tested A Lover’s Complaint, classifying it among
poems by ‘other poets’.32 Table 8.5 lists the Shakespeare blocks that produced
the largest minus figures for Thisted–Efron ‘new words’, which equate to
the greatest proportion of words peculiar to that text. The largest positive
score was forty-one for Romeo and Juliet Block 2.
Elliott and Valenza set their ‘Shakespeare profile’ at the range of –32 to
+21, which makes the Complaint’s score of –33 a ‘rejection’ by the barest of
margins. But the Complaint’s score sits well with the scores for those top
30. Valenza, ‘Thisted–Efron Authorship Tests’, 45.
31. Elliott and Valenza, ‘Oxford by the Numbers’, 427–8, 431.
32. Elliott and Valenza, ‘Oxford by the Numbers’, 424.
180
Determining the Shakespeare Canon
sixteen blocks, which average 32.25 and which all, except the two from Venus
and Adonis, come from plays of Shakespeare’s maturity, when the Complaint
was composed. Elliott and Valenza possess unpublished T–E ‘new words’
scores for every 3,000-word block of Shakespearean drama and poetry.33
Eighteen of these (not counting A Lover’s Complaint) have scores of –30 or
lower (which translates to ‘higher’ in terms of the number of peculiar words).
Presumably text was divided into blocks that do not exactly correspond to
the samples in the published study. Block numbers are in parentheses: Love’s
Labour’s Lost (10) –48, The Winter’s Tale (10) –46, The Merry Wives of Windsor
(2) –45, Love’s Labour’s Lost (6) –41, Hamlet (18) –38, King Lear (10) –38, 2 Henry
IV (8) –36, Antony and Cleopatra (10) –35, Coriolanus (6) –35, Romeo and Juliet
(6) –34, King Lear (6) –32, Venus and Adonis (2) –32, Romeo and Juliet (4) –31,
Macbeth (8) –31, King Lear (14) –30, The Merry Wives of Windsor (4) –30, The
Merry Wives of Windsor (8) –30, 2 Henry IV (10) –30. The mean value of all the
3,000-word blocks in The Merry Wives of Windsor is as low as –29. Clearly,
A Lover’s Complaint has, by Elliott and Valenza’s method of reckoning, a
proportion of peculiar words that would place it among those 3,000-word
blocks of Shakespearean text in which such words are used most liberally.
But this would hardly be surprising in a seventeenth-century poem in which
Shakespeare strove, as I take it, to adapt the condensed dramatic verse of his
maturity to the conventions of Spenserian complaint.
The discrepancy between the results for A Lover’s Complaint using Hart’s
definition of words peculiar to a single work and the T–E results, which treat
words as graphic units, is nevertheless puzzling. In Hart’s terms, the Complaint
is about average in respect of its percentage of peculiar words. In T–E terms
its score would set it among the highest-scoring Shakespeare blocks tested.
A partial explanation may perhaps be found in the fact that, although Elliott
and Valenza’s Shakespeare samples were of 3,000 words (tokens), A Lover’s
Complaint contains a total of only 2,579 words, according to their counts
(2,563 according to Spevack’s). Of course, all Elliott and Valenza’s results
were duly converted into rates per 3,000 words. But the smaller the samples,
the more subject they are to random variation, and although the Complaint
is about the same size as the Shakespearean blocks, it is smaller. In a situation
where a score of –32 counts as falling within the prescribed ‘Shakespearean’
range and a score of –33 as falling outside it, the relative smallness of the
33.I am grateful to Ward Elliott for discussing with me his procedures and findings and for
sending me the relevant files, from which the references to T–E new word figures that follow are taken.
Neologisms and ‘Non-Shakespearean’ Words
181
Complaint may matter. The Complaint’s –33 T–E ‘new words’ score cannot—
even if we ignore the serious caveats of Valenza’s theoretical paper—be taken as
a decisive ‘rejection’.
There are, in any case, some unexplained oddities about Elliott and Valenza’s
results for their Thisted–Efron ‘new words’ tests. Three of Shakespeare’s collaborations have relatively low percentages of Hart’s peculiar words (not used
elsewhere in the canon), falling well below the Shakespearean average of 6.3: 1
Henry VI (4.8), Timon of Athens (5.6), and Henry VIII (4.8). Yet their T–E scores
rank them at the top of the list of plays with the greatest surplus of observed
over expected new words—where ‘new’ means ‘peculiar’ or ‘not used elsewhere’. Similarly, Hart gave the percentages of non-concordance words for
each of Marlowe’s plays and for Kyd’s The Spanish Tragedy.34 These were all
low, ranging from 3.1 to 4.7, and yet on the T–E tests all but Edward II have too
many new words to fit the Shakespeare profiles, and Edward II would fall near
the top of the Shakespeare range.
Hart was meticulous in counting words according to his own criteria,
closely matching OED’s. Mechanical counts of graphic units evidently measure something besides true vocabulary richness. A glance at some of Elliott and
Valenza’s 3,000-word non-Shakespearean blocks reveals why they have such
large minus figures for T–E ‘new words’, or, to put it another way, such large
excesses of observed over predicted non-concordance words. The opening of
2 Tamburlaine, for instance, is chock full of proper names, with many lines like
‘Sclavonians, Almains, Rutters, Muffs, and Danes’ and ‘Illyrians, Thracians,
and Bithynians’. Each of these names counts as a word for Elliott and Valenza,
though not, of course, for Hart. Locrine begins with a line of Latin, soon refers to
‘Pandrassus’, ‘the Molossians’, ‘Goffarius’, ‘Lestrigon’, ‘Corineus’, ‘Gathelous’,
‘Gogmagog’, ‘Samotheus’, and the like, and uses ‘-th’ endings in verbs such
as ‘threat’neth’, ‘sparketh’, and ‘possesseth’. The Spanish Tragedy has smatterings of Latin and Italian, including fourteen consecutive lines of Latin at one
point. Elliott and Valenza’s computer program treats all graphic units indiscriminately. This need not, of course, mean that the test is invalid as a means of
distinguishing between Shakespearean and non-Shakespearean texts. But its
classification of A Lover’s Complaint is, in any case, borderline.35
34. Hart, ‘Vocabularies’, 138.
35. In Chapter 10 I summarize some criticisms of Elliott and Valenza’s statistical methodology, as
applied to poems, that I have mounted in ‘A Lover’s Complaint and the Claremont Shakespeare
Clinic’, Early Modern Literary Studies, 16.3 (2013), 1–12. Some of my reservations were foreshadowed by Thomas Merriam in ‘Untangling the Derivatives: Points for Clarification in the
Findings of the Shakespeare Clinic’, Literary and Linguistic Computing, 24 (2009), 403–16.
182
Determining the Shakespeare Canon
The main reason for discrepancies between Hart and Elliott and Valenza
appears, however, to be that Elliott and Valenza’s T–E test calculates proportions of peculiar (or ‘unique’ or, in their own terms, ‘new’) words to
tokens, not types. An independent study confirms that, in relation to separate types, A Lover’s Complaint’s number of words not found in the undisputed Shakespeare canon is perfectly normal. This study is described on the
Northwestern University website ‘Wordhoard’, which automates various
kinds of text analysis.36 One worked example answers the question ‘How
many words are unique to each Shakespeare work?’ The ‘graphic units’ definition of ‘words’ is employed and the investigators are concerned with different words or types. The authors, working from a modernized edition, call
them ‘distinct spellings’. They filter out proper names. The results are very
much in line with my findings derived from Hart, though naturally figures
are higher because of Hart’s use of dictionary words rather than computerized concordance words. The percentages of ‘unique’ among ‘distinct’
graphic units range from 5 per cent in Julius Caesar and The Two Gentlemen of
Verona to 15 per cent in Henry V. It seems clear that Henry V’s total has been
swollen by the inclusion of speeches in French. Similarly, the total of 14
per cent for The Merry Wives of Windsor presumably includes Caius’s French
dialect and some stray bits of Latin. Otherwise the highest percentages of
‘unique’ words occur in plays singled out by Hart’s data: Love’s Labour’s Lost
with 12 per cent, and Hamlet, King Lear, and Troilus and Cressida with 11 per
cent. A Lover’s Complaint has 1,106 ‘distinct’ words, of which eighty-seven,
or 8 per cent, are ‘unique’. The figure 1,106 is close to Spevack’s 1,089 figure (as in Table 1.1); eighty-seven is almost identical to Elliott and Valenza’s
eighty-eight, reinstated from eighty-six; and 8 per cent is the average number of unique words in the Shakespeare works tabulated, and happens to
be shared by Venus and Adonis and The Rape of Lucrece. A Lover’s Complaint
could not be more centrally Shakespearean in these terms.
To return briefly to Latinate neologisms, Vickers argues that those in
A Lover’s Complaint are of a kind more characteristic of Davies than of
Shakespeare.37 But the available evidence suggests the opposite. Garner,
whose research Vickers so highly commends, cites the following
Latinate neologisms in A Lover’s Complaint: ‘acture’, ‘annexion(s)’, ‘enpatron’, ‘encrimson(’d)’, ‘enswathe(d)’, ‘extincture’, ‘fluxive’, ‘impleach(’d)’,
36. http://wordhoard.northwestern.edu/userman/scripting-example.html.
37. Vickers, Complaint, 145–8.
Neologisms and ‘Non-Shakespearean’ Words
183
‘invised’, ‘phraseless’, ‘supplicant’, ‘unexperient’. Three of these words have
the prefix ‘en-’, while one begins with ‘im-’ and one with ‘in-’. Garner’s list
contains twenty-eight further Shakespeare neologisms in ‘en-’, fifteen in
‘im-’, and thirty-four in ‘in-’. Two of the Latinate neologisms in A Lover’s
Complaint are formations in ‘-ure’. Garner lists a further twenty-four. He
lists eighteen formations in ‘-ive’ in addition to the Complaint’s ‘fluxive’.
Garner lists eleven Shakespeare neologisms in ‘-ant’ and 40 in ‘-ent’, apart
from the Complaint’s ‘supplicant’ and ‘unexperient’. Besides ‘phraseless’ in
the Complaint, he cites thirty-three Shakespearean Latinate neologisms in
‘-less’, and besides ‘annexion(s)’ in the Complaint he cites eleven in ‘-ion’.
Troilus and Cressida has no fewer than twenty-eight neologisms of the nine
types found in the Complaint, while Hamlet has twenty-four. Vickers furnishes no information about Davies’s rates of coinage of words in ‘-ure’,
‘-ive’, ‘-im’, or ‘-ion’, but he does print tables in which asterisks mark
Davies’s neologisms in ‘en-’ and ‘in-’, ‘-ent’ and ‘-ant’, and ‘-less’ (142, 147,
148). There are thirteen altogether, fewer than in Hamlet, which has eighteen, or Troilus and Cressida, which has fourteen. This despite the fact that,
according to Vickers (202), Davies’s oeuvre amounts to over 42,000 lines.38
Vickers rightly notes that Shakespeare’s writing career coincided with ‘a
remarkably fruitful period of linguistic expansion’,39 so that many authors of
the time were liberal coiners of words. As Vickers, drawing on the work of
Elliott and Valenza and others, makes clear, the large number of words in A
Lover’s Complaint that are absent from the Shakespeare concordance and the
large number of words for which the poem provides the first OED citation
do not—as, following Hart, I had supposed in 1965—constitute positive
evidence of Shakespeare’s authorship. But nor do they constitute negative
evidence against the attribution, as Vickers supposes.40 And, as I demonstrate in the next chapter, specific items of the vocabulary of A Lover’s
Complaint point unmistakably to the correctness of Thorpe’s attribution.
38. Vickers, Complaint, 202. Garner’s list of Shakespeare’s Latinate neologisms contains two
close counterparts to those in A Lover’s Complaint: ‘acture’ is matched by ‘enacture(s)’ in
Hamlet and ‘annexion(s)’ by ‘annexment’ in Hamlet. Schäfer’s list of words first used in print
by Shakespeare adds ‘crimson’ as a verb in Julius Caesar as a near match to the Complaint’s
‘encrimson(’d)’ and ‘pleached’ in Much Ado About Nothing as a near match to ‘impleach(’d)’.
Vickers nowhere shows that Davies ever wrote a passage of similar length to A Lover’s
Complaint that matches its rate for Latinate neologisms or for neologisms of all kinds.
39. Vickers, Complaint, 151.
40.Vickers shows that Davies, like Shakespeare, coined a good many compound adjectives
with present or past participles (136–7) and 18 new words beginning ‘un-’ (134). Even so,
18 is only one-ninth as many as the 162 new ‘un-’ words that Schäfer lists as first used by
Shakespeare, despite Davies’s canon being almost two-fifths the size of Shakespeare’s.
9
A Lover’s Complaint, Cymbeline,
and the Shakespeare
Canon: Interpreting Shared
Vocabulary
I
In the previous chapter I countered some of the evidence that has been
adduced as disproving Shakespeare’s authorship of A Lover’s Complaint.
I showed that there is nothing un-Shakespearean about the number of words
in the poem that are rare or otherwise not found within the Shakespeare
canon. In this chapter I want to outline the positive case that can be made
from examination of the vocabulary of A Lover’s Complaint and its relation
to canonical works.
In his book ascribing A Lover’s Complaint to John Davies of Hereford,
Brian Vickers makes little attempt to explore links between the poem and
Shakespeare’s undoubted works, but those that he discusses are explained
as due to (a) coincidence, ‘Shakespeare and Davies sharing a Jacobean
vocabulary’,1 or (b) Davies’s having imitated or echoed Shakespeare, or
(c) Shakespeare’s having been influenced by Davies’s A Lover’s Complaint,
which he read after Thorpe had fraudulently included it with Shakespeare’s
Sonnets (213). Thus, citing several triple rhymes shared by A Lover’s
Complaint and The Rape of Lucrece, Vickers argues not that Shakespeare was
recycling rhymes from the only other work of his in which triple rhymes
1.Vickers, Complaint, 204. Hereafter in this chapter, simple page-references to Vickers’s book
are incorporated into my text.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 185
were repeatedly needed for the rhyme royal stanzas that the poems have in
common, but that we can detect instead ‘the methods of an imitator writing down rhymes in his notebook for future re-use’ (198). Any similarities
between A Lover’s Complaint and Davies’s works, on the other hand, are apt
to be seen as evidence that Davies wrote the poem.
Here, however, my initial focus is on one set of verbal connections
between A Lover’s Complaint and Cymbeline. They were pointed out in 1987
by A. K. Hieatt, T. G. Bishop, and E. A. Nicholson, who offered a very
different explanation of them from that which Vickers prefers.2 Some background information will clarify my own discussion.
Well over a hundred years ago the fine German scholar Gregor Sarrazin,
working from Alexander Schmidt’s Shakespeare Lexicon, listed for every
play, the two narrative poems, the Sonnets, and A Lover’s Complaint all the
words that occur only twice or thrice in Shakespeare’s oeuvre and gave
references for these other occurrences.3 His tables demonstrated a strong
association between chronological proximity of plays or poems and the
numbers of such rare words that they shared. A Lover’s Complaint exhibited
an overwhelming preponderance of vocabulary links with Shakespeare’s
seventeenth-century plays.4 In 1965 I showed that words used in A Lover’s
Complaint and not more than five times in Shakespeare’s canonical works
followed the same pattern.5 Eliot Slater’s more refined statistical analysis yielded similar results. He confirmed Sarrazin’s and my findings on
A Lover’s Complaint and, after compiling a card index of all words that
Shakespeare used ten or fewer times, recorded a tendency for words considerably less ‘rare’ than Sarrazin’s ‘dislegomena’ and ‘trislegomena’ to cluster
chronologically.6
2. Hieatt et al., ‘Rare Words’.
3.G. Sarrazin, ‘Wortechos bei Shakespeare’, Shakespeare Jahrbuch, 33 (1897), 121–65, and
34 (1898), 119–69. ‘Words’ in Sarrazin’s tables are essentially different OED headwords
or entries in Schmidt’s Lexicon. I have used Schmidt, Shakespeare Lexicon. However, my
own analysis includes one or two rare usages or inflexions, such as plural leisures. My list
in Shakespeare’s ‘A Lover’s Complaint’: Its Date and Authenticity of words occurring no more
than five times in Shakespeare’s works followed Schmidt in not distinguishing participial
adjectives from the verbs from which they derived, but here I do make the distinction, as
does OED.
4.Sarrazin’s tables in ‘Wortechos’, 34 (1898), 132–3, display nine links between A Lover’s
Complaint and Shakespeare’s first seventeen works and fifty-five to his next twenty-two.
These figures exclude links to the Sonnets.
5.Jackson, Shakespeare’s ‘A Lover’s Complaint’, 8–14.
6.Eliot Slater, ‘Shakespeare: Word Links between Poems and Plays’, Notes and Queries,
220 (1975), 157–63; The Problem of ‘The Reign of King Edward III’: A Statistical Approach
(Cambridge: CUP, 1988): see esp. the tables on 58–96.
186
Determining the Shakespeare Canon
The fact that A Lover’s Complaint’s links in vocabulary are so predominantly with his seventeenth-century plays is good evidence that if it is by
Shakespeare it is a work of his maturity not of his youth, but in itself it is
not, as Slater supposed, good evidence that Shakespeare was indeed the
poem’s author. Rare-word links between Shakespeare and works by other
writers might display the same tendency to congregate in Shakespeare
plays of about the same period of composition. New words were entering
the language each year, so that poems or plays written in 1600, let us say,
potentially shared items of vocabulary that they could not have shared with
poems or plays written around 1590. Conversely, some elements of literary
English current around 1590 were passing out of use by 1600. Slater found
that a disproportionate number of the rare-word links between the anonymous Edward III (published in 1596 and perhaps first performed as early
as 1590) and the Shakespeare canon were with the three parts of Henry VI
and other early Shakespeare plays, and concluded that Shakespeare was sole
author of Edward III. But M. W. A. Smith and Hugh Calvert showed that
Thomas Kyd’s The Spanish Tragedy and Robert Greene’s James IV exhibit
the same concentration of word-links with Shakespeare’s earliest plays.7 In
these two cases, the distribution of links to Shakespeare’s dramatic canon
was governed by chronology rather than authorship.
However, other aspects of the vocabulary linkages yield good evidence
for attribution. Hieatt, Bishop, and Nicholson, following Slater, argue
that Shakespeare began A Lover’s Complaint around 1600–3 and worked
on it again in preparation for its inclusion in Thorpe’s 1609 quarto. As
they point out, the excess of observed over expected rare-word links with
Shakespearean plays, though considerable for some composed earlier in the
seventeenth century, is greatest for Cymbeline, dated 1610 in the Oxford
Textual Companion.8 They examine the contexts of these words within
poem and play. The words, all of which occur no more than five times
within Shakespeare’s dramatic canon, are: gyves sb., physic v., amplify v.,
7. M. W. A. Smith and Hugh Calvert, ‘Word-Links as a General Indicator of Chronology of
Composition’, Notes and Queries, 234 (1989), 338–41.
8. Wells and Taylor, Textual Companion, 131–2. There was a minor flaw in the way that Slater
calculated figures ‘expected’ on a random distribution of links, but this does not seriously
affect his conclusions. See Jackson, Defining Shakespeare, 41. Correctly calculated figures
would still give Cymbeline the most significant excess of actual over expected links, with
All’s Well That Ends Well coming next.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 187
blazon v., ruby sb., outwardly adv., tempter sb., aptness sb., commix v., spongy a.,
slackly adv., feat a., rudeness sb., usury sb., and pervert v.
Dismissing the claim that the co-occurrence of these words in both
A Lover’s Complaint and Cymbeline points to common authorship and
Shakespeare’s involvement with the poem not long before its publication,
Vickers asserts that two other explanations are possible: either the fifteen
words ‘were circulating in general usage in London between 1603 and 1609’
or, ‘since the Sonnets were published in about June 1609, and the composition
of Cymbeline is usually dated to 1609–10, it may well be that Shakespeare had
read A Lover’s Complaint and recalled it while writing the play’ (213). Vickers
realizes that in this instance the option of casting John Davies of Hereford as
Shakespearean imitator is precluded by the probable date of Cymbeline’s first
performance and the undoubted date, namely 1623, of its first appearance in
print. Let us consider Vickers’s alternative explanations.
Hieatt, Bishop, and Nicholson show that the Complaint and Cymbeline contexts in which at least eight of the fifteen words appear are strikingly similar
in one respect or another. For example, gyves denotes ‘in both Cymbeline and
Complaint imprisoning devices desired by the prisoner’: Posthumus speaks of
‘gyves | Desired more than constrained’ (5.4.14–15, my italics, here and elsewhere in these quoted extracts), and the nun described in the Complaint is
in ‘unconstrained gyves’ (242).9 Both uses of the verb physic relate to love: ‘it
doth physic love’ (3.2.34), ‘love to physic your cold breast’ (259). Each instance
of outwardly applies to a symbolic ornament connected with seduction. The
word aptness ‘relates in both works to shifts of amorous technique according to need’.10 And so on. In discussing slackly they overlooked the detail
that in the Complaint some of the Fickle Maid’s hair, not quite breaking its
‘bondage’, is ‘slackly braided in loose negligence’ (34–5), while in Cymbeline the
king’s children were ‘slackly guarded’ due to ‘negligence’ (1.1.64–6). The shared
associations in these cases make Vickers’s first explanation in terms of sheer
coincidence very improbable. The whole LION electronic database, comprising over 350,000 works of English drama, poetry, and prose, yields only
two works in which gyves are collocated with the words constrained or unconstrained—A Lover’s Complaint and Cymbeline.11
9. Hieatt et al., ‘Rare Words’, 221.
10. Hieatt et al., ‘Rare Words’, 222.
11.Checking collocations with ‘gyves’ necessitates also searching ‘giues’, a common early
modern spelling of the noun.
188
Determining the Shakespeare Canon
Moreover, at least two words might reasonably be added to the Hieatt
list: the past-participial adjective seared and the word outward used as a noun
(in the plural in A Lover’s Complaint).12 Evans in the Riverside edition refuses
to emend the First Folio’s ‘fear’d’ to ‘seared’ in Cymbeline at 2.4.6, but confusion between ‘f ’ and long ‘s’ is one of the most frequent of all errors in
early modern texts and the imagery requires the emendation, adopted without comment in the Oxford Collected Works and fully defended by Martin
Butler in his New Cambridge Shakespeare edition of the play.13 The Folio
contains the same mistake in Measure for Measure at 2.4.9. The word, meaning ‘dried up, parched, withered’, occurs in A Lover’s Complaint within
the initial description of the central character: ‘spite of heaven’s fell rage, |
Some beauty peep’d through lattice of sear’d age’ (13–14). The broadly contextual link with Cymbeline is that shortly after Posthumus has spoken of his
‘seared hopes’ and within the same conversation, he says of Imogen, ‘let her
beauty | Look thorough a casement’ (2.4.33–4). The whole of LION yields
only one other instance of ‘beauty’ looking or peeping through a lattice or
casement, and that is in an eighteenth-century play doubtless influenced
by Cymbeline: in Henry Brooke’s tragedy The Earl of Westmorland (1789),
Rowena says of her husband that ‘beauty from each limb, | As through
a summer casement, look’d abroad, | And found no rival’. But the idea
also turns up in Thomas Shipman’s poem ‘Beauty’s Periphrasis’ from
Carolina: or, Loyal Poems (1683): ‘She looks as Beauty prisoner was | And
peeping through a double grate.’ In literature before this date there is no
parallel so close as afforded by Sonnet 3’s image: ‘So thou through windows
of thine age shalt see, | Despite of wrinkles, this thy golden time’ (11–12),
where, as in A Lover’s Complaint, beauty is to be glimpsed through the lattice or window of ‘age’, and this ‘despite’ (or ‘spite of ’) the ravages of time.
The noun outward is used in both A Lover’s Complaint and Cymbeline in
praise of a man’s exceptionally attractive physical appearance, or ‘ fair parts’
(A Lover’s Complaint, 83). In the poem the woman’s seducer is ‘one by nature’s
outwards so commended | That maidens’ eyes stuck over all his face’ (80–1),
while in the play the First Gentleman reports of Posthumus, ‘I do not think |
So fair an outward and such stuff within | Endows a man but he’ (1.1.22–4).
These two additional items bring the number of ‘rare-in-Shakespeare’
words shared by A Lover’s Complaint and Cymbeline to seventeen,
12. Seared ppl.a is given a separate headword in OED; outward is OED B sb. Schmidt gives
outward(s) as a substantive (‘external form, exterior’) its own separate entry.
13. Cymbeline, ed. Martin Butler (Cambridge: CUP, 2005), 131, 2.4.6 n.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 189
reinforcing the conclusion that, in view of the similar contexts in which so
many of them are set, their presence in both poem and play is due to something more than coincidental use of words ‘circulating in general usage in
London between 1603 and 1609’ (Vickers, 213).
A search of Literature Online that extends the chronological limits five
years before and after Vickers’s dates and encompasses works of any genre
published in the period 1598–1614 and all plays, masques, and entertainments (some 230 altogether) composed or first performed within these
same years reveals that four of the seventeen words were seldom or never
used in literature outside the Shakespeare canon: physic v., slackly adv., seared
ppl.a, outward sb. Outside of Cymbeline and A Lover’s Complaint, physic is
used as a verb only in Shakespeare’s As You Like It, Troilus and Cressida,
Macbeth, The Winter’s Tale, and Addition III of Sir Thomas More, and in
John Norden’s poem The Labyrinth of Man’s Life (1614). Slackly, shared by A
Lover’s Complaint and Cymbeline, is used elsewhere only in the anonymous
Stonyhurst Pageants, dated 1617 in Annals of English Drama and so falling outside the set 1598–1614 limits.14
Again, apart from the Complaint and Cymbeline, the only works to use
seared as a participial adjective are Shakespeare’s Measure for Measure and the
play The Valiant Welshman (by ‘R. A.’), dated 1612 in Annals and published
in 1615. And outward is a noun only in the Complaint, Cymbeline, Troilus and
Cressida, Sonnet 69, Sonnet 125, and one non-Shakespearean work. In all the
Shakespearean examples of outward the word is clearly a noun. LION yields
a small handful of non-Shakespearean examples of adjectival outward with a
following noun understood, such as ‘Fitting their outward to their inward
hue’ in Francis Rous’s Thule, or Virtue’s History (1598). But there is also one
somewhat enigmatic instance of substantival use in Samuel Rowlands’s
The Betraying of Christ (1598). Rowlands includes a series of lines describing
Judas and beginning with successive letters of the alphabet, A to Z. The line
for ‘X’ reads: ‘X pian the outward, inward, not at all.’ This becomes more
intelligible with the comma after ‘inward’ deleted and the old abbreviation
(normally ‘Xpian’) expanded to ‘Christian’: Judas was Christian in outward
show but not inwardly. Although outward is here a noun, it is less distinct from
the kind of elliptical ‘outward . . . inward’ plus noun formation than are the
Shakespearean examples: ‘beauty’s outward’ in Troilus and Cressida (3.2.158),
14. LION’s datings derive from Annals, but, as mentioned in earlier chapters, one or two plays
may sometimes be included erroneously in a period set for searching.
190
Determining the Shakespeare Canon
‘Thy outward thus with outward praise is crown’d’ in Sonnet 69.5, ‘With my
extern the outward honoring’ in Sonnet 125.2; in this last example most editors sensibly emend ‘the’ to ‘thy’.
So within Vickers’s specified narrower range, about 1603–9, LION yields
no non-Shakespearean instances of any of these four words, whereas, if, on
the strength of Vickers’s ‘about’, we allow slackly in Cymbeline, Shakespeare
uses them all within that period.
Vickers lists ten of the fifteen Hieatt words shared by A Lover’s Complaint
and Cymbeline that occur in Davies’s writings.15 Of these, the instance of
physic v. comes within Wit’s Bedlam (1617), which falls outside Vickers’s
‘between about 1603 and 1609’ and also outside the wider limits of my
LION search. Vickers has forgotten ruby, which truly is common in the
period and which Davies uses in five of his volumes. Neither outward sb. nor
seared ppl.a. is used by Davies.16 So his canon contains examples of eleven of
the seventeen words, and excludes three of the four rarest (slackly adv., outward sb., and seared v.); also absent are outwardly, aptness, and rudeness.
Shakespeare, of course, employs all seventeen, since all are found in
Cymbeline. But what we are attempting to determine is whether it is more
likely (a) that A Lover’s Complaint is by Davies and that Shakespeare, in writing
Cymbeline, was influenced by reading the poem when it was published with
his own sonnets, or (b) that Shakespeare wrote A Lover’s Complaint, as well
as Cymbeline. The seventeen words have already been identified as linking
poem and play. We may now examine their incidence in other Shakespearean
works. As Hieatt, Bishop, and Nicholson record, all except commix and slackly
are to be found in at least one, and usually in several, Shakespeare plays,
besides Cymbeline. But the significant feature is their chronological distribution. The nineteen plays up to and including Henry V (1598–9) have seven
instances, whereas the nineteen plays from Julius Caesar to The Two Noble
Kinsmen have twenty-nine.17 This is pertinent to dating the poem rather than
15. Vickers, Complaint, 217. He states that Davies uses ‘all but three’ of the fifteen Hieatt words
but counts ‘apt’ as though it were aptness and ‘slack’ as though it were slackly.
16. Davies used ‘seared’ as a verb (Vickers, Complaint, 243) but not as a participial adjective.
17.I have again relied on the chronology in the Oxford Textual Companion, 109–34. Hieatt
et al. (222) say that feat occurs in The Two Noble Kinsmen, but two instances of the word
in that play (3.1.45, 5.1.43) are of the modern noun meaning ‘exploit, action, skill’, not
of the archaic adjective meaning ‘dexterous, neat, trim’ (with adverbial force in A Lover’s
Complaint). A third instance in Kinsmen is also a noun, but probably a misprint for ‘seat’: see
MacDonald P. Jackson, ‘Arcite’s Horsemanship: A Reading in The Two Noble Kinsmen,
II.v.13’, Notes and Queries, 254 (2009), 605–7. The instance of gyves in The Two Noble Kinsmen
falls within a Shakespeare scene (3.1.72). Not counted in the figures given is the one link to
Addition III of Sir Thomas More.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 191
to determining its authorship. More importantly, Measure for Measure (1603)
has six, namely gyves, ruby, tempter, usury, pervert, and seared. Troilus and Cressida
(1602) has five, namely physic, spongy, rudeness, outwards, and outwardly. Macbeth
(with physic, outwardly, and spongy) and Coriolanus (with amplify, aptness, and
usury) both have three. Measure for Measure follows Troilus and Cressida in the
Oxford Shakespeare’s chronological order of plays, so that two consecutively
written plays (1602–3), covering about six hundred lines, contain eleven of
the seventeen Cymbeline–A Lover’s Complaint words, as many as the whole
Davies canon, amounting to over 42,000 lines.18 In the whole of LION,
four works contain spongy, outwardly, and physic as a verb: Troilus and Cressida,
Macbeth, Cymbeline, and A Lover’s Complaint.
Davies never used outwardly, aptness, slackly, rudeness, outward sb., or seared
ppl.a. Before the 1609 publication of A Lover’s Complaint, Shakespeare had
used outwardly in Troilus and Cressida (1602), Macbeth (1606), and possibly
The Winter’s Tale (1609), depending on its exact date of composition; aptness in Coriolanus (1608); rudeness in Julius Caesar (1599), Twelfth Night (1601),
and Troilus and Cressida (1602); outward sb. in Troilus and Cressida (1602) and
Sonnets 69 and 125 (1593–1603); and seared in Measure for Measure (1603). Of
course, he used all these words in Cymbeline. By 1609 Davies had not used
physic as a verb, his sole example coming from Wit’s Bedlam (1617), whereas
Shakespeare had used it in As You Like It (1599–1600), Troilus and Cressida
(1602), Sir Thomas More, Addition III (1603–4), Macbeth (1606), and possibly
The Winter’s Tale (1609).19
In the light of all the preceding data it seems highly improbable that in
Cymbeline Shakespeare echoed a Complaint by John Davies of Hereford. It
seems much more likely that in Cymbeline Shakespeare recycled some of
the words he himself had used not only in A Lover’s Complaint but in various other works, and placed several of them in strikingly similar contexts
because they formed part of a network of associations in his mind.
Further scrutiny of the Complaint contexts of some of the seventeen
words confirms this conclusion. We have already said something about
18. This figure is from Vickers, Complaint, 202. The concentration of rare Complaint–Cymbeline
words in Troilus and Cressida and Measure for Measure is consistent with Hieatt, Bishop, and
Nicholson’s belief that Shakespeare began writing A Lover’s Complaint around 1600–3 and
prepared it as late as 1609 for inclusion in the Sonnets quarto.
19. Davies’s only use of feat was in The Muse’s Sacrifice (1612). Shakespeare anticipated him with
the example in Cymbeline (1610). Davies’s uses of the verb pervert in Humour’s Heaven on
Earth (1609) and The Holy Rood (1609) were also anticipated by Shakespeare—in Measure for
Measure (1603) and All’s Well That Ends Well (1604–5).
192
Determining the Shakespeare Canon
seared. A Lover’s Complaint tells us in regard to the ‘fickle maid’ that ‘spite of
heaven’s fell rage, | Some beauty peep’d through lattice of sear’d age’ (13–
14). The word seared thus occurs within lines that have a close, and almost
unique, parallel in Cymbeline’s ‘let her beauty | Look through a casement’,
besides a striking similarity to Sonnet 3’s: ‘So thou through windows of
thine age shalt see, | Despite of wrinkles, this thy golden time’, which
offers the same idea of glimpsing beauty through the lattice or windows
of withered or wrinkled age, and in similar wording: through . . . age . . . spite
of/Despite of. In A Lover’s Complaint, the criss-crossed strips of lattice-work
themselves evoke the sonnet’s ‘wrinkles’. It seems probable that Shakespeare
three times used variations on the same image, not that in Cymbeline he was
reclaiming an image that Davies had borrowed from one of his sonnets.
Moreover, the metaphorical ‘peeping through’ of the Complaint can be
paralleled in Shakespeare, but not in any other LION work of 1598–1614.
In other authors, cases of literal peeping through (often through ‘crannies’)
are common enough, and the sun, morning, heaven, or light may peep
through some space or other. But only in Shakespeare can we find a figurative peeping through comparable to beauty’s in the Complaint: ‘I’ll force |
The wine peep through their scars’ in Antony and Cleopatra (3.13.189–90);
‘your youth, | And the true blood which peeps fairly through’t’ in The
Winter’s Tale (4.4.147–8), where ‘true blood’ means both ‘virtuous passion’
and ‘noble lineage’; ‘I can see his pride | Peep through each part of him’ in
Henry VIII (1.1.68–9). None of these plays was published until 1623, and yet
the composition of Antony and Cleopatra certainly predated the writing of
Cymbeline and the publication of A Lover’s Complaint. Davies has only the
one example of the commonplace ‘Heaven’s bright eye . . . Peeps through
the purple window of the east’ in Summa Totalis (1607).20
The rare noun outward was seen to have similar associations in the
Complaint and Cymbeline. Its context in the Complaint also connects with
other Shakespearean works. In the poem, the Youth is said to be
O, one by nature’s outwards so commended
That maidens’ eyes stuck over all his face.
20. Vickers discusses ‘Some beauty peeped through lattice of seared age’ (255), but cites no passage in which Davies has beauty figuratively peeping or looking through a lattice, casement,
or window. ‘Peeping through’ is juxtaposed to ‘lattice’ not only in A Lover’s Complaint but
also in 2 Henry IV, 2.2.80–3, and in only one other LION work of 1580–1640—Thomas
Brewer’s prose Merry Devil of Edmonton (1631), where ‘Smug . . . stood peeping through his
lattice.’
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 193
Love lack’d a dwelling and made him her place;
And when in his fair parts she did abide
She was new lodg’d and newly deified.
(80–4)
Sonnet 69’s use of outward as a noun begins:
Those parts of thee that the world’s eye doth view
Want nothing that the thought of hearts can mend;
All tongues (the voice of souls) give thee that due,
Utt’ring bare truth even so as foes commend.
Thy outward thus with outward praise is crown’d.
(1–5)
Here, as in A Lover’s Complaint and Cymbeline, the noun outward relates to
praise of a man’s exceptional physical attractiveness, with the sonnet and
the poem sharing ‘parts’ and ‘commend’. Like A Lover’s Complaint, the sonnet goes on to suggest that the inward man, as manifested in his deeds,
‘matcheth not thy show’ (13). The Complaint stanza’s ‘maidens’ eyes stuck
over all his face’ may seem grotesque, but ‘stuck over’ probably has little more metaphorical content than our ‘all eyes were fixed upon him’.
Maidens’ eyes lingered over him. There are good Shakespearean parallels: Timon speaks of ‘The mouths, the tongues, the eyes, and hearts of
men . . . | That numberless upon me stuck’ (Timon of Athens, 4.3.261–3), and
the Duke in Measure for Measure exclaims: ‘O place and greatness! millions
of false eyes | Are stuck upon thee’ (4.1.59–60). The Complaint stanza’s idea
of ‘love’ (or ‘Love’) using the beautiful Youth as ‘a dwelling’ recalls a passage in Venus and Adonis where ‘Love’, imagining his own burial in the
hollows of Adonis’s dimpled cheeks, knows well that ‘if there he came to
lie, | Why there Love liv’d, and there he could not die’ (245–6). And of the
Young Man of Sonnet 93 it is said that ‘heaven in thy creation did decree |
That in thy face sweet love should ever dwell’ (9–10).21
The Complaint stanza thus has notable parallels with a sonnet published
in the same volume as itself, a narrative poem published in 1593 and therefore available for Davies to borrow from, and two plays that did not reach
print till 1623, one of which, Timon of Athens, had probably not been staged
by 1609.
21.Vickers (245) quotes Davies’s ‘Love, leave thy lodge (my heart) and enter hers’ in Wit’s
Pilgrimage (1605) and ‘Thus interchang’d we either’s form impart | To other’s liking by the
love we have, | And make the heart the lodge it to receive’ in Microcosmos (1603). These variations on the commonplace exchanges of hearts or the love within them seem very different
from the figure in the Complaint, Venus and Adonis, and Sonnet 93 whereby a Young Man’s
beautiful face or ‘fair parts’ is the dwelling place of a personified Love.
194
Determining the Shakespeare Canon
It must be emphasized that I am not offering these parallels as in
themselves proving that Shakespeare wrote A Lover’s Complaint. Rather,
I have cited them as further evidence that can help us determine the
most plausible explanation for the presence in both the Complaint and
Cymbeline of the rare noun outwards. The fact that outwards occurs as a
noun in a stanza of the Complaint that affords such Shakespearean parallels surely tells against the idea that the Complaint was written by Davies,
and that Shakespeare, reading it for the first time in 1609, echoed Davies
in Cymbeline, taking over from him a word that Shakespeare himself
had used in Troilus and Cressida and in two sonnets but that Davies never
used in his acknowledged canon.
Among the Hieatt words is feat. In A Lover’s Complaint, letters are ‘With
sleided silk feat and affectedly | Enswath’d and seal’d to curious secrecy’
(48–9). The whole LION database yields only one other instance of ‘sleided
silk’—in the Gower chorus of Pericles that, in unmistakably Shakespearean
terms, describes the accomplishments of ‘absolute Marina’ (4.Cho.31).22
Shakespeare could not have appropriated this phrase after encountering it
for the first time in the Complaint as printed in the Sonnets quarto of 1609,
since Pericles had been first performed sometime within the period 1606–8.
Davies could have borrowed the phrase from Pericles only if he had heard
it at a performance of the play or read it in the 1609 quarto of Pericles after
the unknown month of the play’s publication but before publication of the
Sonnets quarto, which had been entered in the Stationers’ Register on 20
May 1609. It seems far more likely that Shakespeare himself was responsible
for both uses of the term, especially since his lodging with the Mountjoys
on the corner of Silver and Muggle Streets, within the approximate period
1602–6, was located above a shop where sleave-silk (as OED calls it) was
worked.23 Again, the word common to A Lover’s Complaint and Cymbeline is
found in the poem in association with a phrase that affords a unique parallel
to a different Shakespearean play.
22. Moreover, the First Folio (1623) text of Troilus and Cressida has the variant ‘Sleyd silke’ (that
is, ‘sleid silk’), where the 1609 quarto has ‘sleiue silke’ (that is, ‘sleave-silk’) at 5.1.31. OED
found no instances other than those in the Complaint and Pericles of the irregular form sleided (spelled ‘sleded’ in Pericles). Macbeth’s ‘ravell’d sleave of care’ (2.2.34, ‘Sleeue’ in F) is
also silk.
23. Charles Nicoll, The Lodger: Shakespeare on Silver Street (London: Allen Lane, 2007),
17–18, 160–7.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 195
II
The links in vocabulary between A Lover’s Complaint and Cymbeline, to
which Hieatt, Bishop, and Nicholson drew special attention, are important
because the poem was published before Cymbeline was available for imitation, even by a playgoer hearing it in performance, let alone a reader, the
play’s first appearance in print having been in the First Folio of 1623. But of
course there are also several very rare words in the Complaint that, although
not found in Cymbeline, do occur within other Shakespeare plays: the verbal
substantive appertainings and the adjective credent (as distinct from the Latin
third person plural verb), for example. LION yields no non-Shakespearean
instances of either word before the nineteenth century. But appertainings occurs in the Complaint (115) and in the quarto of Troilus and Cressida
(1609) at 2.3.80 (where F 1623 has ‘appertainment’), while credent occurs in
the Complaint (279), Hamlet (1.3.30), Measure for Measure (4.4.26), and The
Winter’s Tale (1.2.142). Since The Winter’s Tale was composed in 1609, or
possibly 1610, credent was thus used by Shakespeare in plays written both
before and after the probable composition date and certain publication
date of A Lover’s Complaint. All the extant evidence points to its having
belonged to Shakespeare’s wordstock, but not to Davies’s. Indeed, before
the nineteenth century it appears to have been Shakespeare’s alone.
Another rare word, pelleted, occurs in the Complaint (18) and in Antony and
Cleopatra (3.13.165), and, as Kenneth Muir noted, ‘In the poem, the tears are
the pellets of sorrow; in the play, Cleopatra’s frozen tears, turned to hail, are
the pellets of her grief.’24 LION’s only other pre-nineteenth-century uses
of the word are in civic pageants by Heywood, Munday, and Middleton,
and in every case it is used in a technical, heraldic sense (‘marked or charged
with [heraldic] pellets’), especially with reference to ‘pelleted lions’, often
‘golden’. OED places its citations from the Complaint and Antony and
Cleopatra under a separate headword: they are the only citations in which
the verb pellet means ‘form or shape into pellets’ and in each case it is water
that is so formed—as teardrops, as hailstones. Since Antony and Cleopatra
was first printed in the First Folio (of 1623), Vickers has to again suggest
that either Davies ‘may well have seen it in the theatre’ or to invoke pure
coincidence (168).
24. Muir, ‘ “A Lover’s Complaint”: A Reconsideration’, 159.
196
Determining the Shakespeare Canon
The Complaint stanza containing pelleted is another with many connections to Shakespeare’s works:
Oft did she heave her napkin to her eyne,
Which on it had conceited characters,
Laund’ring the silken figures in the brine
That seasoned woe had pelleted in tears,
And often reading what contents it bears;
As often shriking undistinguish’d woe
In clamors of all size, both high and low.
(15–21)
Vickers finds much to deprecate here (167–8). But there are many striking
Shakespearean parallels, whatever one’s critical judgement on the way the
images are developed. The use of a napkin to wipe away tears may seem
commonplace, but among LION plays of 1590–1610 I found it only in Titus
Andronicus (twice) and 3 Henry VI.25 Davies never mentions a napkin and the
sole handkerchief (‘handkercher’) in his verse does not wipe away tears. In
the Complaint, the woman’s napkin, embroidered with ‘conceited characters’ and ‘silken figures’, is reminiscent of Desdemona’s, which had ‘magic
in the web’ (3.4.69), and was sewn with ‘silk’ (3.4.73) embroidery that
included ‘strawberries’ (3.3.435).26 Vickers reviles as ‘laboured’ the ‘spelling
out of the chain of associations’ in ‘Laund’ring . . . brine . . . seasoned . . . pelleted in tears’ (168). But Shakespeare, like many of his contemporaries,
repeatedly thinks of tears in terms of their saltiness, and ‘tears’, ‘brine’, and
‘season’ are brought together in All’s Well That Ends Well: ‘tears. | ’Tis the
best brine a maiden can season her praise in’ (1.1.47–9).27 The metaphorical content of Shakespeare’s verse is customarily produced or enhanced by
the kind of submerged punning that leads from the brine of tears to salt as
a seasoning: the associations in the Complaint lines are not ‘spelt out’ at all.
25. Jackson, ‘A Lover’s Complaint Revisited’, 283; and see Appendix 4, Part B below. All three
instances are in scenes attributed to Shakespeare: Titus Andronicus, 3.1.140 and 146; 3 Henry
VI, 2.1.61–2.
26. Napkin and handkerchief were interchangeable terms, as in Othello, where ‘handkerchief ’
is the usual term but ‘napkin’ appears at 3.3.287, 290, and 321. Othello asks for Desdemona’s
handkerchief in order to wipe away ‘a salt and sorry rheum’ (3.4.51). In Richard III, Queen
Elizabeth mockingly tells Richard to present her daughter, whom he wishes to marry, with
a handkerchief with which to ‘wipe her weeping eyes’ (4.4.278). And in All’s Well That Ends
Well, Lafeu, declaring, ‘Mine eyes smell onions, I shall weep anon’, requests of Parolles,
‘lend me a handkercher’ (5.3.319–22).
27.A quick check of entries for ‘salt’ in Spevack’s Shakespeare concordance reveals at least
sixteen references to salt tears, while ‘salt’ and ‘season’ are juxtaposed again in Much Ado
About Nothing, 4.1.142, and Troilus and Cressida, 1.2.255, and tears form a ‘brine-pit’ in Titus
Andronicus, 3.1.129.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 197
Romeo and Juliet adds washing or laundering (though of the cheeks, not a
napkin), when Friar Laurence, learning of Romeo’s new love, exclaims:
Jesu Maria, what a deal of brine
Hath wash’d thy sallow cheeks for Rosaline!
How much salt water thrown away in waste,
To season love, that of it doth not taste!
(2.3.69–3)
In Twelfth Night, Valentine reports to Orsino of Olivia that
like a cloistress she will veiled walk,
And water once a day her chamber round
With eye-offending brine; all this to season
A brother’s dead love . . .
(1.1.27–30)
In The Rape of Lucrece, the heroine ‘must sit and pine, | Seasoning the earth
with show’rs of silver brine, | Mingling my talk with tears, my grief with
groans’ (795–7).28
In the Complaint, ‘seasoned woe’ is of long duration, established, matured
over time (like seasoned timber), but the wordplay helps create ‘pelleted in
tears’. The same punning is present in The Merchant of Venice in Portia’s
‘How many things by season season’d are | To their right praise and true
perfection!’ (5.1.107–8): as M. M. Mahood explained, in ‘season’ there is
‘a play on the meanings “time” and “spice”, which is extended to “seasoned” ’.29 In fact, the same wordplay seems present in the Twelfth Night passage: Olivia preserves in the brine of tears the memory of her dead brother,
maintaining the piquancy of her sense of loss, but there is a suggestion also
of her allowing her grief to ripen and mature.
Vickers, retailing eighteenth-century editor George Steevens’s assertion that pellet was ‘the ancient culinary term for a forced meat ball’, finds
such an extension of the image ‘from the kitchen’ that is touched on in
‘seasoned’ to be an offence against ‘the Renaissance concept of decorum’
(167–8). OED lends no support whatsoever to Steevens’s gloss. If the sense
claimed by Steevens did indeed exist and was known to the poet, it may
have operated as an unconscious association. But the sense of pellet relevant
for a reader is OED’s ‘globe, ball, or spherical body, usually of small size’.
28. Vickers (243–4) cites a passage in Davies’s The Holy Rood (1609) that collocates ‘tears’, ‘brine’,
and ‘season’. The parallel is good, but no better than the Shakespearean ones, which Vickers
ignores, and which I quote simply as contributing to the various Shakespearean ideas and
images surrounding the very rare word pelleted.
29. The Merchant of Venice, ed. M. M. Mahood (Cambridge: CUP, 1987), 159, 5.1.107 n.
198
Determining the Shakespeare Canon
The woman’s ‘seasoned woe’ has produced globular tears. Vickers believes
that ‘This poet has been thinking too literally about tears and their shape’
(168). To me, it is through the workings of a Shakespearean visual imagination that teardrops are so economically brought into focus. The pellet as
‘shot, projectile, gunstone’ is also pertinent, as John Kerrigan explained. As
he noted, Shakespeare’s use of pelleted in Antony and Cleopatra also ‘associates tears with a destructive bombardment. Accused of being cold-hearted
towards Antony, Cleopatra protests that, if she is so, heaven should “engender hail” from her tears and in a “pelleted storm” destroy herself, her heirs,
and nation’ (3.13.158–67).30
As for ‘Laund’ring’, this first use of the verb known to OED is a natural extension of the washing with tears that, while common enough in
early modern texts, turns up with special frequency in Shakespeare.31
‘Laund’ring’ refreshes the commonplace, particularly in association with
the napkin or handkerchief.
To move to the stanza’s lines 19–21, reading the contents of something is far
from unusual, but in the study described in Chapter 6 I found the verb and
the noun juxtaposed only in three plays of 1590–1610, one of them being
King Lear.32 Vickers regards the woman’s ‘shrieking undistinguished woe’
as ‘excessive, melodramatic’, points out that the shrieks in Shakespeare are
‘mostly associated with night, omens, and unnatural behaviour’, and argues
that Davies was under the influence of Spenser (60–1). But the melodramatics
are no blemish. The ‘I’ of the poem’s opening stanza is a detached spectator to
the woman’s histrionics and an eavesdropper on the story of her seduction as
she relates it to a ‘reverend man’. Even that initial ‘I’ cannot be equated with
the poet, whose sympathy towards the main character in this little drama is
complicated by his sense of the ridiculous. As John Kerrigan has written, ‘In
Venus and Adonis Shakespeare is drily sceptical about the kind of lamenting
amplitude which features so largely in A Lover’s Complaint.’ The Complaint
has its share of ‘the ingrained tonal wit which comes from derivativeness
in the genre’. The poem’s ‘knowingness’ about its ‘Spenserian trappings’
imparts an air of the ‘urbane and droll’.33
The self-conscious drollness extends into the stanza’s final ‘clamors
of all size, both high and low’ (21). Vickers objects: ‘The collocation of
30. Kerrigan, Sonnets, 398, line 18 n.
31. LION, or scrutiny of a Spevack’s concordance, reveals at least twenty Shakespearean
instances of washing associated with tears.
32. Jackson, ‘A Lover’s Complaint Revisited’, 283; see also Appendix 4, Part B.
33. Kerrigan, Motives of Woe, 65–6.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 199
“clamours” and “size” leads us to expect such qualifying epithets as “great”
and “small”, which makes either “size” or “high and low” an inappropriate
word choice’ (219). Yet, among LION dramatists of 1590–1610, Shakespeare
is the only one to use either ‘size’ or ‘both high and low’ in connection
with sounds.34 In The Winter’s Tale, a servant reports that Autolycus ‘hath
songs for man or woman, of all sizes’ (4.4.191–2). The Arden 2 editor, J. H.
P. Pafford, glossed ‘sizes’ as ‘lengths, kinds’, citing OED, 12, and Schmidt
3.35 The songs are thus both short and long and for male (low) and female
(high) voices. Likewise, the complaining woman’s ‘clamours’ or wailings
vary in pitch and duration. The Complaint’s ‘both high and low’ (which
also includes ‘loud and soft’), is not ‘an unmotivated piece of padding’, as
Vickers supposes: it adds a further descriptive element. In Twelfth Night the
‘true love’ of Feste’s song ‘can sing both high and low’ (2.3.41)—again both
pitch and volume seem relevant. As OED makes clear, ‘size’ had a wider
range of meanings than Vickers recognizes, but even in its ordinary modern sense, it is not at odds with ‘high or low’, which need not be taken as
in apposition. As Katherine Duncan-Jones remarks, the line has a tinge of
‘mockery or burlesque’ about it, but this does not make it inept.36
Vickers cites six examples of ‘high or low’ in Davies’s poetry (220).
But the phrase is used seven times in Shakespeare’s plays, and, as in the
Complaint and Twelfth Night, in The Merry Wives of Windsor (2.1.113) we
find the full phrase ‘both high and low’ (my italics), which Davies never
uses.37 Nor does he ever apply ‘high and low’ to sounds. The web of
Shakespearean connections with the stanza in which ‘pelleted’ is placed
renders incredible Vickers’s offered explanation of the word’s occurrence,
with the meaning ‘formed into (non-heraldic, water-based) pellets’, in
both A Lover’s Complaint and Antony and Cleopatra but in no other LION
work. Coincidence is inconceivable. And even if we were to suppose that
Davies, like Sir Andrew Aguecheek collecting unfamiliar words that tickle
his fancy, not only jotted down ‘pelleted’ after attending a performance of
the play but also remembered its context, we would still have to postulate a
degree of intimacy with other Shakespearean works (some not published by
34. Jackson, ‘A Lover’s Complaint Revisited’, 283; see also Appendix 4, Part B.
35. The Winter’s Tale, ed. J. H. P. Pafford (London: Methuen, 1963), 100, 4.4.193 n.
36. Duncan-Jones, Sonnets, 433, line 21 n.
37. LION readily traces Shakespearean examples of the collocation ‘high and low’, which was
given a separate entry in John Bartlett’s Complete Concordance (London: Macmillan, 1894).
In The Merry Wives it refers to social standing, which Burrow considered of possible subsidiary relevance to the Complaint context (Sonnets and Poems, 696, line 21 n.).
200
Determining the Shakespeare Canon
1609) of which there is no evidence in his acknowledged verse—or none, at
least, that Vickers supplies.
One further example will reveal how untenable is the position
that Vickers adopts. The seduced woman of A Lover’s Complaint is seen
‘Storming her world with sorrow’s wind and rain’ (7), which furnishes
OED’s earliest citation of storm as a transitive verb.38 It has long been known
that there is an analogous use in a speech of King Lear that is preserved only
in the 1608 quarto. In the Oxford Complete Works edition of The History of
King Lear a Gentleman reports that the King ‘Strives in his little world of
man to outstorm | The to-and-fro conflicting wind and rain’ (sc. 8.9–10).
The parallel is most unlikely to be coincidental, because it not only covers the newly minted verb and the reference to the microcosm but also
encompasses the phrase ‘wind and rain’.39 Vickers claims that ‘since Lear was
published in 1608, the Complaint’s author could easily have imitated those
lines’ (208). But had Davies read King Lear in the 1608 quarto he would have
encountered not outstorm but outscorne. Steevens was able to cite A Lover’s
Complaint’s line, ‘Storming her world with sorrow’s wind and rain’, in support of an emendation to Lear that has rightly been followed by almost
all editors, though not by Evans in The Riverside Shakespeare (3.1.10–11).
Lacking such an analogue, Davies would not have recognized the authentic
verb behind the misreading, especially when ‘outstorm’ had never before
appeared in print.40
In the course of the above discussion, I have mentioned the following eight words that occur in A Lover’s Complaint and in the undisputed
Shakespeare canon but are very rare in other LION works of 1598–1614,
whether first published during that period or, in the case of plays, first performed then: appertainings vbl. sb., credent a., outward sb., pelleted (pellet v.),
physic v., seared ppl.a., slackly, sleided ppl.a. There are nineteen instances of
these words in Shakespeare’s acknowledged works, only three in other
LION works of 1598–1614, by three different authors, Davies not being
38.Vickers retains the 1609 quarto’s ‘sorrowes, wind and raine’, which editors have rightly
modernized as ‘sorrow’s wind and rain’: the woman creates a storm in her little world with
the wind and rain (sighs and tears) of her sorrow.
39. LION detects ‘wind and rain’ in only a handful of other plays first performed, or poems
first published, between 1598 and 1614, including Shakespeare’s As You Like It (written
1599–1600, first printed 1623) but also Davies’s The Scourge of Folly (1611). Of course ‘the
wind and the rain’ occurs in the song with which Feste ends Twelfth Night (written 1601,
published 1623).
40.The transitive use of storming in the Complaint and of outstorm in King Lear is quite distinct from Davies’s intransitive use of the verb (Vickers, Complaint, 242) on three occasions
between 1611 and 1617.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 201
one of them.41 Pelleted, meaning ‘formed into pellets’, and sleided do not
appear outside Shakespeare at all, while appertainings and credent are not
found outside his work till the nineteenth century. One might add to the
list. Within the period 1598–1614, a-twain is used only in the Complaint
(6), King Lear (2.2.74), and Othello (Q 1622, 5.2.206; ‘in twain’ in F 1623);
plural leisures appears only in the Complaint (193), The Merchant of Venice
(1.1.68), Timon of Athens (2.2.128), and the anonymous Nobody and Somebody
(1606); the Complaint’s reworded (reword v.) is used only in Hamlet (3.4.143);
laugher occurs only in the Complaint (124) and Julius Caesar (1.2.72);42 origin
appears once in the Complaint (222), twice in Hamlet (1.4.26, 3.1.177), once
in King Lear (4.2.32), and once, in the spelling origen, in a Shakespearean
scene of The Two Noble Kinsmen (5.4.61); disciplined (discipline v.) is used only
in the Complaint (261), Troilus and Cressida (2.3.244), Coriolanus (2.1.126), and
Arthur Gorges’s Lucan’s Pharsalia (1614); dialogued (dialogue v.) is found in the
period only in the Complaint (132), Timon of Athens (2.2.51), and William
Warner’s Albion’s England (1602). When these seven words are added to the
eight, we have fifteen words that are used thirty-two times by Shakespeare
within the period 1598–1614 and six times by other authors, all different
and none of them Davies.
I have concentrated on the period 1598–1614, because it adds five years
to both the upper and lower limits of the 1603–9 in which Vickers suggests that all the rare words common to A Lover’s Complaint and Cymbeline
‘were circulating in general usage in London’ (213), and because it serves to
test Vickers’s further suggestion that vocabulary links that I cited in 1965
between the Complaint and Shakespeare’s maturer plays ‘may only show
that the poem reflects word usages of [c.1601–8]’ (145). On the contrary,
many of them appear to have been used within the period exclusively, or
almost exclusively, by Shakespeare.
The evidence of the fifteen words, with their thirty-two appearances in
Shakespeare and mere six in all other authors of the period, heavily outweighs the evidence that in his section on ‘Diction: Rare Words’ Vickers
advances as connecting the Complaint to Davies (214–17). He discusses the
following words: lover with reference to a woman, platted, maund, affectedly,
fancy with reference to a person, forbod, and spongy. Vickers concedes that
while Davies uses lover of a woman once, Shakespeare does so five times,
41. But Davies does, as we have seen, use physic as a verb in Wit’s Bedlam (1617).
4 2. The Riverside Shakespeare does not, however, follow most editors in emending F’s ‘a common laughter’, which seems to me unidiomatic.
202
Determining the Shakespeare Canon
so we can summarily dismiss this usage as evidence of Davies’s rather than
Shakespeare’s authorship. The form forbod (for ‘forbidden’ or ‘forbade’) is
common in LION poetry published in the period 1598–1614, occurring
eleven times, apart from the single instance in Davies’s Wit’s Pilgimage
(1605). Besides, it is used by Shakespeare, although outside the period—
within The Rape of Lucrece (line 1648), though most editors modernize to
‘forbade’. Maund occurs ten times in LION poetry of 1598–1614, besides the
instance that Vickers notes in Davies’s Microcosmos (1603).
Spongy (sometimes spelled with medial ‘u’ and sometimes with ‘-ie’
ending) is even more common. Davies employs it only once, whereas
Shakespeare employs it in Troilus and Cressida (2.2.12), Macbeth (1.7.71),
Cymbeline (4.2.349), and The Tempest (4.1.65). There are twelve further
examples in LION works of the set period. Vickers makes much of the
fact that the Complaint has the phrase ‘spongy lungs’ (326, ‘spungie’ in the
original spelling), and that, although none of the four instances of spongy in
Shakespeare’s plays associate it with the lungs, Davies tells us in Microcosmos:
The lungs therefore are spongy, soft, and light,
That air might enter, and from them depart,
Which guard the heart (on left side and the right)
From bord’ring bones, that else annoy it might.
(455–8)
But Davies’s prosaic retailing of physiological information is strikingly
different from the Complaint’s turning of the lungs’ sponginess to poetic
use: ‘O, that sad breath his spungy lungs bestowed’ (326), exclaims the
woman of her seducer. ‘His lungs, like a sponge, can squeeze out the sighs
stored in them’, as Colin Burrow explains.43 Vickers judges the epithet
‘spongy’ to be ‘incongruous’ (216). Burrow shows that it is not.44 Moreover,
John Jowett has noted the same idea in Troilus and Cressida, where Hector
declares that, though fearless of the Greeks, he is prudent: ‘There is no lady
| More spungy to suck in the sense of fear, | More ready to cry out “Who
knows what follows?” ’ (2.2.11–12). ‘The image is of inhalation into the
lungs and crying out on exhalation’, as Jowett observes.45 Shakespeare is apt,
also, to place ‘lungs’ in negative contexts. In the Complaint ‘that infected
moisture of his eye’ (323) is the first of the brief catalogue that includes the
43. Burrow, Sonnets and Poems, 717, line 326 n.
4 4.So also does Duncan-Jones, who glosses spongy as ‘soft, absorbent’, adding ‘presumably
here suggesting a treacherous capacity to generate false breath, or words’ (Sonnets, 316, line
326 n.).
45. John Jowett, review of Vickers, Complaint, in Shakespeare Quarterly, 60 (2009), 495.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 203
‘spongy lungs’, and Pericles has the phrase ‘belch’d on by infected lungs’
in one of the brothel scenes (4.6.169), while Troilus and Cressida collocates
‘raw eyes’ and ‘whissing lungs’ (5.1.20–1). Vickers agrees with me that
Shakespeare had dipped into Microcosmos, so he may well have read Davies’s
doggedly informative lines and turned fact to poetic account.46 But among
the quite numerous examples of ‘spongy’ in 1598–1614 is Thomas Winter’s
allusion to ‘our spongy lungs’ in The Second Day of the First Week (1603). In
any case, ‘spongy’ is not a rare word in the period.
Nor is platted as rare as any of the fifteen Shakespeare words we have
considered, since it occurs as a participial adjective in Robert Chester’s
Love’s Martyr (1601), Arthur Gorges’s Lucan’s Pharsalia (1614), Joseph Hall’s
Virgidemiarum (1598), and the King James Bible (1611), besides several times
as a verb. In Humour’s Heaven on Earth (1609), Chronos has a ‘platted’ beard.
Shakespeare does not use the participial adjective, but in Mercutio’s Queen
Mab speech in Romeo and Juliet, Mab ‘plats the manes of horses in the night’
(1.4.89). Platted (or ‘plaited’) mane and platted beard seem about equally
close to the Complaint’s ‘platted hive of straw’ (8).
This leaves (a) affectedly and (b) the noun fancy applied to a person. Vickers
states that the only two instances of affectedly in the LION database for poetry
and drama between 1590 and 1620 are in A Lover’s Complaint and in Davies’s
Select Second Husband for Sir Thomas Overbury’s Wife (1616) (215). But this is
wrong, because affectedly occurs in Cyril Tourneur’s The Atheist’s Tragedy,
published in a quarto of 1611. And of course Davies’s use of the word in 1616
falls outside the defined period, 1598–1614. Davies instructs the husband to
‘Teach not thy wife to speak . . . affectedly’, and Puttenham had used affectedly in anatomizing faults in speech and writing in his The Art of English
Poetry (1589). It also occurs twice in George Whetstone’s Heptameron (1582).
These instances are earlier than our period, of course. In the Complaint, letters are ‘With sleided silk feat and affectedly | Enswath’d and seal’d to curious secrecy’ (48–9). The primary meaning here hardly seems to be ‘with
affectation’; rather, the enswathing and sealing have been done ‘with loving
care’. Davies is closer to Puttenham than to the Complaint.
In A Lover’s Complaint the lamenting woman is called ‘this afflicted
fancy’ (61). Vickers points out that Davies three times uses the noun with
reference to a person—a tubby, over-dressed lecher in Humour’s Heaven
46. MacDonald P. Jackson, ‘Shakespeare’s Sonnet cxi and John Davies of Hereford’s Microcosmos
(1603)’, Modern Language Review, 102 (2007), 1–10; Vickers, Complaint, 41–6.
204
Determining the Shakespeare Canon
(1609), the effeminate Glaucus in an epigram in The Scourge of Folly (1611),
and male lovers (of women) in Microcosmos. In his undisputed works
Shakespeare does not apply the word in this way. Editors differ in their
understanding of fancy in the Complaint. Katherine Duncan-Jones glosses
afflicted fancy as ‘unhappy apparition’ or ‘victim of delusion’, appealing to
OED senses 2 and 3.47 Others have accepted Schmidt’s explanation that
here and in the plural wounded fancies at line 197 the noun is a kind of
metonym from the very common Shakespearean meaning ‘love’, that it
is a case of ‘the abstract for the concrete’, so that this afflicted fancy means
something like ‘this woman who is distressed by love-sickness’. This tendency for the abstract to shade into the concrete makes it very difficult to
assess the degree of likeness between Vickers’s three Davies examples and
the Complaint examples, and no less difficult to decide how many of the
scores of instances of fancy or fancies in the period 1598–1614 come close to
matching either Davies or the Complaint. In one of Richard Alison’s songs
in An Hour’s Recreation in Music (1606) the singer laments, ‘I weep and she’s
a-dancing’, and ‘O cruel, cruel, cruel fancy’ seems directed at the desired
‘she’, not simply the state of mind that possesses him and makes him dote
on her. Thomas Greaves has ‘Farewell sweet Flora, sweet fancy adieu’
in a song (1604). In John Dowland’s ‘To catch young fancies in the nest’
the fancies may be persons (1603). When in Love’s Metamorphosis (1601)
Thomas Lyly declares that ‘a lady’s heart, though it harbour many fancies,
should embrace but one love’, both ‘fancies’ and ‘love’ seem to be at once
the emotion and its object. Robert Armin’s The Two Maids of More-Clacke
affords ‘I must have fancies, playfellows’ and goes on to specify pet animals and things—the objects of his fancy. And in Nicholas Breton’s An
Old Man’s Lesson (1605) ‘they that will be fools to give money for fancies’
are disparaged, where the fancies are again the objects of desire. Davies’s
human fancies are all male, and the first two are more ‘fantastic’ than
lovesick, whereas those in the Complaint are female. Too many ambiguities surround this particular Davies–Complaint link for it to carry much
weight.
Even if, despite these doubts, we were to accept fancy as a lexical link to
the Complaint of the same order as the Shakespearean ones, there would still
be fifteen items of the Complaint’s rare vocabulary shared with Shakespeare
and only one with Davies. Allowing Davies’s physic v. in Wit’s Pilgrimage
47. Duncan-Jones, Sonnets, 435, line 61 n.
a lover ’ s complaint , cymbeline , and
the Shakespeare Canon 205
(1617) and affectedly in Second Husband (1616)—although (unlike all the
Shakespeare items) they do not fall within the period 1598–1614—would
bring the Davies total to a mere three. Yet, since performance or publication within the period 1598–1614 cuts out ten of Shakespeare’s plays and
his two early narrative poems, Davies’s total searched output is more than
half the size of Shakespeare’s and has the advantage of being all in the same
genre as the Complaint—poetry, not drama.48
The findings of this analysis may now be summarized. Neither of Brian
Vickers’s explanations of why A Lover’s Complaint and Cymbeline have so
many rare-in-Shakespeare words in common is credible. Several of the
words were not ‘circulating in general usage in London between 1603 and
1609’ (213). In fact, four of them—physic v., slackly, seared ppl.a., outward
sb.—were used during these years by no writer except Shakespeare, with
his nine instances in works besides Cymbeline. Nor is it in the least likely
that Shakespeare, in writing Cymbeline, was influenced by a Complaint of
which Davies was the author, borrowing from it rare words that he himself
had used in other plays, but that Davies never used. The view of Hieatt,
Bishop, and Nicholson—that Shakespeare wrote A Lover’s Complaint
and was still working on it not long before its publication—best fits the
data. Moreover, there are rare words that do not recur in Cymbeline but
do appear in other Shakespearean works, so that altogether at least fifteen
words link the Complaint to Shakespearean plays first published and/or performed 1598–1614 or to sonnets in the quarto entitled Shakespeare’s Sonnets
(1609) and to not more than one non-Shakespearean work.49 These fifteen words are used thirty-two times by Shakespeare within this period
and only six times by all the other writers of poetry, drama, or prose, and
never by John Davies of Hereford. Commenting on the list of Complaint
48. Vickers notes that Davies used the collocation ‘fell rage’ and was ‘very fond’ of the adjective
‘fell’, using it thirty-seven times (219). But Shakespeare used it forty-three times. Davies
has a ‘conjuring, proud, remorseless priest | Rend, in fell rage . . . pompous vestures’ (The
Holy Rood, 1609). The Complaint’s phrase about the woman’s retaining some beauty in ‘spite
of heaven’s fell rage’ is really more akin to ‘time’s fell hand’, as agent of decay, in Sonnet
64. Within the period 1598–1614 ‘fell rage’ occurs in Edward Fairfax’s Godfrey of Bulloigne
(1600) and John Marston’s Antonio and Mellida (1602), and LION detects forty-two examples
of the collocation in the whole database.
49.One might reasonably add to this total. OED’s first example of sistering ppl.a. is from A
Lover’s Complaint (2), while its first citation for the verb sister is from Shakespeare’s
Pericles: Marina’s art ‘sisters the natural roses’ (5.Cho.7). But OED puts the verb and the
present-participial adjective under separate headwords. No other examples of verb or participial adjective appear in LION 1598–1614.
206
Determining the Shakespeare Canon
words that I published in 1965 as ‘rare’ in the Shakespeare canon, occurring
not more than five times, Vickers writes: ‘Neither in this study, nor in the
intervening years, did Jackson perform the “negative check” by examining
the occurrence of these words in other Jacobean authors.’50 The Literature
Online electronic database facilitates such checking. The results confirm
my earlier conclusions that the author of A Lover’s Complaint was William
Shakespeare.
50. Vickers, Complaint, 206; his reference is to Jackson, Shakespeare’s ‘A Lover’s Complaint’.
10
A Lover’s Complaint:
Counter-arguments and
Conclusions
I
Marina Tarlinskaja has devoted two articles to arguing that the metrical features of A Lover’s Complaint are incompatible with Shakespeare’s authorship
of the poem at any stage of his career.1 She judges that A Lover’s Complaint
is an early Elizabethan poem by some unknown imitator of Spenser. The
main weakness in her argument is that we have no way of knowing what
metrical characteristics we should expect to find in rhyme-royal stanzas
of a narrative poem by Shakespeare that was written in the first decade
of the seventeenth century, the period to which modern editors—rightly,
as I have attempted to show—assign it. Tarlinskaja extrapolates from the
development of Shakespeare’s dramatic blank verse. But Shakespeare’s last
written sonnets, numbers 100–26, dating from after 1600, exhibit a predominantly 4 + 6 hemistich segmentation, despite his plays having by then
shifted to a predominantly 6 + 4 segmentation.2 Different genres have different metrical constraints. Even the rhymed couplets of All’s Well That
Ends Well are more like the verse of the Complaint and metrically much
less ‘advanced’, in terms of Shakespeare’s progress, than the play’s blank
verse. The development of Jonson’s non-dramatic verse does not run
1. Tarlinskaja, ‘Verse of A Lover’s Complaint’; ‘Who Did NOT?’
2. For the dating of Sonnets 100–126, see Jackson, ‘Vocabulary and Chronology’; ‘Rhymes in
Shakespeare’s Sonnets’; ‘Dating Shakespeare’s Sonnets’.
208
Determining the Shakespeare Canon
parallel to the development of his dramatic verse, so it is rash to assume that
Shakespeare’s would have done so.
Tarlinskaja does not examine a range of longish non-dramatic rhymed
iambic pentameter poems of the first decade of the seventeenth century, in
order to determine whether the peculiarities of A Lover’s Complaint associate it exclusively with the time of Spenser. Had she done so, she would have
found several poems of the period displaying the liking for the 5 + 5 segmentation (as well as for 4 + 6) that she notes in the Complaint and considers so anomalous for mature Shakespeare verse. Further, Tarlinskaja’s own
figures for ‘utterance junctures’ in Shakespearean drama show a peak after
the fifth syllable in most plays of the period 1595–1600, while for Troilus
and Cressida, roughly contemporary with A Lover’s Complaint, her recorded
percentages of junctures after the fourth, fifth, and sixth syllables are 31.3,
26.7, and 22.1, forming a somewhat similar pattern to that which she notes
in the ‘syntactic boundaries’ of the poem.3
Tarlinskaja considers enclitic phrases (which she carefully defines) especially ‘indicative of a poet’s metrical style’,4 but by tabulating her counts of
these as numbers per 1,000 lines she makes A Lover’s Complaint appear more
abnormal than it is.5 A table showing the Complaint with 12.0 enclitics per
1,000 lines and Troilus and Cressida, for example, with 33.0 per 1,000 lines
might more fairly have shown the Complaint’s 329 lines having 4 enclitics
altogether and Troilus with a rate per 329 lines of 11. Since 11 is an overall
rate for the whole play, it seems likely that there are 329-line sections of
Troilus that have as few, or almost as few, enclitics as the Complaint.
3.Tarlinskaja, Shakespeare’s Verse, 137–8. In her articles Tarlinskaja offers figures for ‘strong
syntactic breaks’ in Troilus and Cressida of 15.7, 12.4, and 14.7 after the fourth, fifth, and sixth
syllables, but these are percentages of the total number of lines, not of the total number of
strong syntactic breaks. This latter mode of reckoning had been used for ‘utterance junctures’—the percentage of the total number of these that fell within the particular positions.
‘Utterance junctures’ are points at which a verse line is split between two or more speakers
and so constitute only a proportion of all ‘strong syntactic breaks’, but they are clear markers of caesuras. Tarlinskaja, ‘Verse of A Lover’s Complaint’, 153, summarizes some data for
‘strong breaks after position 6 as a percentage of breaks after 4’. This is a mathematically
inappropriate manœuvre, which exaggerates differences. Breaks after position 6 should be
calculated as percentages of all breaks after 4 and 6. So 37.9 for A Lover’s Complaint, 63.2 for
The Rape of Lucrece, 51.6 for Romeo and Juliet, and 93.6 for Troilus and Cressida ought to be,
respectively, 27.5, 38.7, 34.0, and 48.4. It is notable that the proportion of breaks after syllable
4 is much greater over the first dozen stanzas of A Lover’s Complaint than over the rest of the
poem. This might suggest a shift, in the course of composition, from deliberate imitation
towards a rhythm more akin to Shakespeare’s own.
4. Tarlinskaja, ‘Who Did NOT?’, 355.
5.Tarlinskaja, ‘Verse of A Lover’s Complaint’, 156. The table is unaccountably missing from
‘Who Did NOT?’
a lov er ’s compl ai n t : counter-arguments
and conclusions 209
Such details of Tarlinskaja’s analysis might be debated at some length.
But the crucial points are that, since the only poem using the same stanza as
A Lover’s Complaint and indisputably by Shakespeare is The Rape of Lucrece,
composed in 1593, we have no means of gauging what the metrical features
of the Complaint, composed a decade later, ought to be, and that Tarlinskaja
does not even provide enough information for us to gain a clear view of the
development into the seventeenth century of non-dramatic verse generally.
Tarlinskaja’s case, in her two essays, against the Complaint’s having been
written by John Davies of Hereford is, however, strong. Davies, unlike
Shakespeare, produced enough non-dramatic verse covering a wide chronological range to provide a solid basis for conclusions about his versification.
II
Under the direction of Ward E. Y. Elliot and Robert J. Valenza, the
Claremont Shakespeare Clinic has spent two decades investigating the
authorship of plays and poems associated with Shakespeare.6 Their project
began as an attempt to check whether any anti-Stratfordian claimant to the
title of ‘the true author’ of the works attributed to ‘the man from Stratford’
wrote in a distinctively ‘Shakespearean’ style. They soon discovered that
none did. They went on to consider problems of interest to mainstream
Shakespeare scholarship and to make valuable contributions to their solution. Their findings have been largely in accord with orthodox scholarly
opinion, which they have in turn helped to form. But among their conclusions is that A Lover’s Complaint is probably not by Shakespeare.
The Clinic’s tests for Shakespearean authenticity have been developed
and augmented piecemeal over the years. In dealing first with drama, Elliott
and Valenza’s procedure was gradually to accumulate data for various features in twenty-nine Shakespeare plays of his uncontested sole authorship,
to standardize for a play length of 20,000 words, to determine the range of
use over these core plays, and to trim the upper and lower limits of ‘outliers’
so as to create Shakespearean ‘profiles’ within which all or nearly all core
Shakespeare plays fell but a substantial number of non-Shakespeare plays
did not. Later, from the uncontested poems—Venus and Adonis, The Rape of
Lucrece, and the Sonnets—and fourteen core plays, fourteen approximately
6. The fullest and most recent published account of their work is Elliott and Valenza, ‘Oxford
by the Numbers’. But the Clinic’s publications specifically concerned with A Lover’s
Complaint are Elliott and Valenza, ‘Glass Slippers’ and ‘Did Shakespeare?’
210
Determining the Shakespeare Canon
3,000-word poem blocks and eighty-two dramatic verse blocks were
formed, and profiles for poems derived.7 The profiles were determined by
‘handfitting’. Their upper and lower limits corresponded fairly closely with
the upper and lower limits of the actual ranges for poem blocks, but with
some ad hoc trimming or extension suggested by the data from the ‘dramatic verse’ blocks.
Subjecting A Lover’s Complaint to the Clinic’s battery of fourteen tests
for Shakespearean authorship of poems, Elliott and Valenza judged it most
unlikely to be by Shakespeare. From the raw counts of various features,
figures were calculated per exactly 3,000 words or per 1,000 words or per
20,000 words, or they were standardized in some other way. A Lover’s
Complaint failed four tests for 3,000-word blocks, whereas only two of the
fourteen poem blocks undoubtedly by Shakespeare—one from Venus and
Adonis and one from the Sonnets—failed even a single test. (It is no more
than a confusing coincidence that there were fourteen tests applied to
fourteen blocks.) Elliott and Valenza calculate what they call the ‘discrete
composite probability’ and the ‘continuous composite probability’ that A
Lover’s Complaint should fail so many tests, and fail them to such an extent,
if it were by Shakespeare. But, as they themselves concede, these ‘are not
indicators of the absolute, actual probability that Shakespeare wrote the
block in question’. Rather the scores ‘permit comparison of the block in
question . . . with an actual Shakespeare block at the edge of his range’.8 A
Lover’s Complaint emerges as ‘hundreds of times’ less ‘Shakespearean’ than
even the worst performing actual Shakespeare block.9
The conclusion that A Lover’s Complaint is not by Shakespeare is, however, not warranted by this evidence, as I have demonstrated in a recent
article.10 The poem’s poor showing relative to the Shakespeare blocks was
largely determined by the way in which the tests were devised. The Clinic’s
procedure made it inevitable that nearly all fourteen 3,000-word poem
blocks would pass nearly all of their fourteen tests, and, since the tests were
7.The Phoenix and Turtle was considered too short to be used.
8. Elliott and Valenza, ‘Oxford by the Numbers’, 351. Merriam, ‘Untangling the Derivatives’,
includes an astute analysis of the implications of the fact that the Clinic’s ‘probabilities’ are
‘not probabilities as commonly understood’ (2–3). Merriam makes several valuable suggestions towards refinement of the Clinic’s methodology. He points out that a strong chronological element in the data needs to be taken more fully into account and that the Clinic’s
‘probabilities’ wrongly assume that all tests are independent of one another. He demonstrates the advantages of principal component analysis.
9. Elliott and Valenza, ‘Oxford by the Numbers’, 426.
10. Jackson, ‘Claremont Shakespeare Clinic’.
a lov er ’s compl ai n t : counter-arguments
and conclusions 211
especially chosen because of their capacity to fail a fair proportion of the
eighty-six non-Shakespearean poem blocks chosen for comparison, they
inevitably did so. But from the way the testing was set up we cannot know
how we should expect a poem such as A Lover’s Complaint to perform, supposing it is by Shakespeare. The problem is that the determining of the
profiles and the judging of how Shakespeare poem blocks perform on them
are dependent on the same data. We need a ‘calibrating’ set of Shakespeare
poem blocks independent of those from which the profiles were derived.
Yet even all fourteen blocks are few from which to generate reliable profiles, and in ‘Claremont Shakespeare Clinic’ I proved that it was illegitimate to use 3,000-word blocks of play verse either as supplementary means
of determining ‘handfitted’ profiles for poems or as ‘set-asides’ from which
to assess how well the Shakespearean profiles for 3,000-word poem blocks
would discriminate between Shakespeare and non-Shakespeare poems not
used to generate those profiles. So, suppose we had only Venus and Adonis
and The Rape of Lucrece from which to derive Shakespearean poem profiles
for 3,000-word blocks, and wished to determine whether the Sonnets were
by Shakespeare: how would blocks from the Sonnets perform?
In ‘Claremont Shakespeare Clinic’ I showed that when from the eight
narrative poem blocks (three from Venus and five from Lucrece) profiles
were determined in a strict mathematical way rather than by ad hoc ‘handfitting’, figures for A Lover’s Complaint fell outside them four times, while
three of the six Sonnets blocks fell outside them four, five, and six times.
So the Complaint performed as well as or better than half of the Sonnets
blocks.
Being long narrative poems, Venus and Adonis and The Rape of Lucrece,
published in 1593 and 1594 respectively, are different in style and purpose
from the Sonnets, published in 1609 and almost certainly including both early
and late work. Each fourteen-line sonnet is both complete in itself and part
of a sequence, and the three quatrains followed by a couplet differ from the
narrative-poem stanzas in the constraints they impose on metrical, lexical,
and syntactical choices. That Shakespeare’s Sonnets are not homogeneous
with his narrative poems on the measures used by the Claremont Clinic was
readily proved by statistical analysis. Although A Lover’s Complaint shares
its stanza form with The Rape of Lucrece, as Spenserian complaint wrought
in the style of Shakespeare’s mature dramatic verse, it is sui generis among
Shakespeare’s works, and so might be expected to differ in relation to the
Claremont test profiles from both the narrative poems and the Sonnets.
212
Determining the Shakespeare Canon
In ‘Claremont Shakespeare Clinic’ I further demonstrated that even
when 3,000-word Shakespeare poem block profiles are derived from all
fourteen poem blocks A Lover’s Complaint fails only two of fourteen tests
if profiles are determined according to approved mathematical rules. This
result is scarcely inferior to the results for four of the fourteen Shakespeare
poem blocks that generated the profiles and is in marked contrast to the
result for the decidedly non-Shakespearean Funeral Elegy, which under the
new profiles fails seven of the Clinic’s tests, instead of six. I showed also that
one of the two tests failed by A Lover’s Complaint is thoroughly unsatisfactory and should be abandoned.
Elliott and Valenza and their Claremont helpers have been assiduous,
resourceful, and innovative in their researches into questions surrounding the Shakespeare canon. They have remained open-minded, refining
their methods in response to criticism. Their articles have been written
with clarity and flair and in an unfailingly equable and courteous tone,
despite the provocations of anti-Stratfordians and other combatants. Elliott
and Valenza’s data do not, however, support a verdict against Shakespeare’s
authorship of A Lover’s Complaint.
III
Although Tarlinskaja’s prosodic analysis tended to undermine his theory
that John Davies of Hereford wrote A Lover’s Complaint, Brian Vickers
invoked it as research that ‘decisively rejected Shakespeare’s authorship’.11
But the explanations Vickers gives for the poem’s failing to fit with Davies’s
practices serve as more plausible counters to Tarlinskaja’s arguments against
Shakespeare: that Tarlinskaja’s studies, while ‘extremely important for
those working on the drama of the period’, are, for reasons that Vickers
spells out, ‘of less value for students of poetry’; and that evidently ‘the author
of A Lover’s Complaint was familiar with Spenser’s work and imitated some
features of his style’, producing a stylistic ‘hybrid’.12 There is no reason why
Shakespeare’s verse should not have been modified by his undertaking a
narrative in the tradition of Spenserian complaint.
11. Vickers, Complaint, 195.
12. Vickers, Complaint, 194–6.
a lov er ’s compl ai n t : counter-arguments
and conclusions 213
Vickers also mentions Elliott and Valenza, whose grounds for denying
Shakespeare A Lover’s Complaint do not stand up to scrutiny. Most of Vickers’s
case—based on vocabulary, phraseology, rhetorical figures, associations of
ideas, and the like—is addressed in my preceding chapters. Some details are
considered in reviews by John Jowett and myself.13 Vickers sees links between
the poem and Davies where others would see links with Shakespeare. The
curious logic by which he accounts for triple rhymes, as well as rhyme pairs,
shared by A Lover’s Complaint and The Rape of Lucrece has already been mentioned. He ignores the obvious explanation that Shakespeare was recycling
rhymes from the only other work of his in which triple rhymes were repeatedly needed for the rhyme-royal stanzas that the poems have in common,
supposing instead that we can detect ‘the methods of an imitator writing
down rhymes in his notebook for future re-use’.14 This seems like special
pleading. Vickers’s list of rhymes in the Complaint that appear in Davies’s
acknowledged oeuvre but not Shakespeare’s proves nothing, because Davies
was so much more prolific as a rhymester than Shakespeare. On this kind of
evidence Davies would emerge as a better candidate than Shakespeare for
the authorship of the A Lover’s Complaint only if the number of Complaint
rhymes shared with Davies were shown to be larger than the number shared
with Shakespeare, when the raw figures had been converted to percentages
of the total numbers of rhymes in each canon.
Vickers’s ‘list of rhymes in the Complaint that do not appear in
Shakespeare’, with those that ‘recur in John Davies’s poetry’ marked
with an asterisk, is, in any case, inaccurate and misleading.15 Strangely, it
includes rhymes that he himself records in inconspicuous footnotes as used
by Shakespeare, such as art/part, contains/remains, ride/tide, and strong/tongue.
There are several others that, while marked as used by Davies, are also found
within the Shakespeare canon.16 Vickers notes that age/rage appears twice in
1 Henry VI, though in a scene (4.6) that is probably not Shakespeare’s, but
he neglects to mention that it is also found twice in the Sonnets and once
13. Jowett, Shakespeare Quarterly, 60 (2009); Jackson, review in Review of English Studies, 58
(2007), 723–5.
14. Vickers, Complaint, 198.
15. Vickers, Complaint, table 6.4 on 258, with discussion of the table, 257–64.
16.I have relied on the inventory of Shakespeare’s rhymes in Helge Kökeritz, Shakespeare’s
Pronunciation (New Haven: Yale UP, 1953), 399–495, and cite all rhymes in modern spelling.
Vickers used this source (197) but his consultation of Kökeritz while compiling table 6.4
seems to have been over-hasty.
214
Determining the Shakespeare Canon
in The Rape of Lucrece. Listed and asterisked are art/heart, which is used at
least a dozen times by Shakespeare; receive/leave, which appears in All’s Well
That Ends Well, The Merchant of Venice, and Troilus and Cressida; and say/obey,
prominent in the final speech of King Lear. Unto/woo, similarly listed and
asterisked, twice rhymes in Venus and Adonis: unto him/woo him, unto her/woo
her; in contrast, Davies, as Vickers himself states, never rhymed unto with
woo, though he did ‘at least a dozen times’ rhyme woo with variants of to
(‘two’, ‘too’).17 Among rhymes that Vickers marks with an asterisk are the
Complaint’s sequences assigned/find/mind and fame/inflame/shame. He does
not claim that Davies ever employed the triple-rhyme sets, only that he
employed some of their components. But Shakespeare also rhymes some
members of the sets—for example, find with mind (twelve times, including
finds/minds) and fame with shame (twice in The Rape of Lucrece and once in
Much Ado About Nothing).
Vickers points out that the triple rhyme wind/find/mind occurs in both
A Lover’s Complaint and Davies’s Humour’s Heaven on Earth, and calls this ‘a
finding which far surpasses the bounds of coincidence’.18 But this is one of
the most common of all triple rhymes in early modern poetry. LION turns
up twenty instances in poetry of the period 1593–1617.19 It is used three
times in Anthony Nixon’s The Christian Navy (1602): at lines 15–19, 113–17,
and 540–4. Vickers’s treatment of rhymes is gravely flawed.
Vickers’s conviction that A Lover’s Complaint is not Shakespeare’s derives
in large measure from his critical judgement upon it. He considers it marred
by ‘clumsiness and lack of invention’, often vague, muddled in its imagery
and grammar, and lumbered with too many strange words and banal
metrical fillers.20 His disdain places him at odds with such editors as John
Kerrigan, Katherine Duncan-Jones, and Colin Burrow, all sensitive readers of poetry. Evidently, estimations of the poem may differ and de gustibus
non est disputandum. But, as Jowett declares, ‘Vickers has . . . assembled a heap
17. Vickers, Complaint, 264.
18. Vickers, Complaint, 262.
19.Searches for ‘find NEAR.100 mind NEAR.100 wind’, with ‘Variant spellings’ checked,
brought up potential cases, which were then examined in context.
20. Vickers, Complaint, 2. As a sign of poverty of invention in the author of the Complaint
Vickers adduces a tendency ‘to repeat a rhyme used not long before’ (200), but the Sonnets
include many such repetitions of precisely the rhymes he lists: art/heart occurs in Sonnets
22, 24, 41, 125, 131, and 139; make(s)/take(s) in 81 and 91; eyes/lies in 1, 2, 17, 24, 46, 137, and
153; find/mind in 27, 77, 92 (and finds/minds in 116); while grace/place, though restricted to one
appearance in the Sonnets (number 79), is nevertheless a Shakespeare favourite, occurring
eight times in his other works.
a lov er ’s compl ai n t : counter-arguments
and conclusions 215
of potential evidence . . . but he has not subjected it to rigorous screening
and evaluation.’21 Despite his use of LION, his methodology is essentially
that of those nineteenth- and early twentieth-century scholars who, fixing
on a candidate for the authorship of a disputed work, amassed data to buttress their views, whereas what are required for convincing demonstration
are, as made clear in Chapter 1 on the Quarrel Scene in Arden of Faversham,
modes of operation that treat rival candidates impartially and uniformly and
allow pointers to any one of them to predominate. No counter-arguments
so far offered form a convincing threat to Thorpe’s ascription of A Lover’s
Complaint to Shakespeare.
IV Summary and Conclusion
The internal evidence supporting Thorpe may briefly be summarized.
LION searches of the first and last five stanzas of A Lover’s Complaint, conducted in the same way as searches of Arden of Faversham, find many more
links in phrases and collocations with plays by Shakespeare than with plays
by other dramatists of the period 1590–1610. The distribution of links is
consistent with a date of composition for the poem of about 1603–7. There
seems no good reason why any poet other than Shakespeare himself should
so much more frequently echo and anticipate the phrasing, imagery, and
ideas of Shakespearean than of non-Shakespearean drama. A control test of
Shakespeare’s Sonnets 111 and 112 produces results closely matching those
for the Complaint.
If some non-dramatic poet did have such close affinities with Shakespeare,
he cannot have been John Davies of Hereford. A search for phrases and collocations that the first five stanzas of the Complaint share with Shakespeare’s
or Davies’s canon, but not with both, discovers that links to Shakespeare
are, proportionally to the size of the canons, far more numerous than links
to Davies, even when computations are restricted to non-dramatic verse.
A study of the Complaint’s vocabulary tells the same story. At least fifteen
of the Complaint’s rare words appear in Shakespeare’s acknowledged canon
of 1598–1614—mainly in markedly similar contexts—but in no more than
a single non-Shakespearean work. Altogether, they occur thirty-two times
in undoubted plays and poems written by Shakespeare during those years,
21. Jowett, Shakespeare Quarterly, 60 (2009), 496.
216
Determining the Shakespeare Canon
but only six times in poetry, drama, or prose by all other writers during
the same period, and never by Davies. In this case too, the chronological
distribution of these rare words over Shakespeare’s works militates against
an explanation in terms of Davies having borrowed from Shakespeare
or Shakespeare having in 1609 read and borrowed from a Complaint
that he had not composed. Neologisms are of a Shakespearean kind
and words in the Complaint that are not used by him elsewhere occur at
Shakespearean rates.
But the most compelling evidence of all derives from spellings, which in
Shakespeare’s time varied according to personal preference. Thirty-four
spellings in A Lover’s Complaint are found in five or fewer writers of
poems published 1593–1617 and plays first performed 1590–1614. The
dates enclose both Shakespeare’s and Davies’s writings. Shakespeare
uses seventeen of the rare spellings, and there are twenty-eight links
to his works, seventy-five to all other poets and dramatists combined,
despite the fact that Shakespeare’s canon covers only about 8 per cent of
the amount of text searched. John Davies of Hereford’s extensive poetic
output affords only three links, no more than afforded by Shakespeare’s
poems, which form a non-dramatic corpus an eighth the size of Davies’s.
Filtering the list of thirty-five spellings so as to retain only those
employed by one or two writers strengthens the connection with
Shakespeare, especially with printed texts believed to have been printed
from his autograph, and further filtering strengthens it even more. Seven
spellings are rare within the whole LION database covering more than
six centuries of English poetry, drama, and prose. Four of the seven
appear within Shakespeare’s works, whereas none appears more than
once in the work of any other writer.
A control exercise, performed on the first 329 lines of Davies’s Humour’s
Heaven on Earth, published in 1609, the same year as A Lover’s Complaint,
finds significantly more links in spelling with Davies’s oeuvre than with
Shakespeare’s, especially in proportion to their overall sizes. The contrast
with the results for the Complaint is stark. Moreover, Davies’s authorship
of A Lover’s Complaint is decisively ruled out by its failure to sport any of
the Davies markers, heau’n or heau’ns, pow’r, sith, yer, and it’s. No Davies
poem, or portion of a Davies poem, contains even single instances of the
alternative heauen or heauens, power or power, and since (all three) but none
of Davies’s favourite forms. The Complaint has six instances of the more
common forms, which Shakespeare overwhelmingly prefers.
a lov er ’s compl ai n t : counter-arguments
and conclusions 217
My own research, as summarized above, has received a degree of backing from an analysis by Hugh Craig, in which techniques of computational stylistics similar to those he had applied to Arden of Faversham were
employed in order to assess affinities between the language of A Lover’s
Complaint and the linguistic habits of Chapman, Davies, and Shakespeare.22
Chapman’s responsibility for the Complaint had been advocated almost a century ago by J. M. Roberston.23 Relying on a large corpus of poems and plays
by those three writers and several of their contemporaries, Craig used two
quantitative measures: (a) words peculiar to one author, on the one hand, and
words that he avoided, on the other, and (b) consecutive word pairs appearing
far more frequently in one author’s work than elsewhere. As in Shakespeare,
Computers, and the Mystery of Authorship, the ability of the methods to discriminate among candidates and to attribute texts correctly was assessed on samples of known authorship. The verse of Chapman and Davies tested as utterly
unlike that of A Lover’s Complaint. Tests for Shakespeare were less conclusive,
one way or the other, but revealed some marked affinities between his works
and the Complaint. Craig concludes that, judging solely from the results of
his testing, the poem ‘cannot be by Chapman or Davies, and could be by
Shakespeare’.
A mantra for Elliott and Valenza has been that negative evidence trumps
positive evidence: ‘fitting the tiny slipper does not prove you are Cinderella
nearly as conclusively as not fitting the tiny slipper proves you are not
Cinderella. If you are a size four, you could just as well be a false-positive
Little Miss Muffett or Tiny Tim; but, if you are a size ten, your claim to be
Cinderella is in trouble.’24
The logic is irresistible, but the analogy with Cinderella’s glass slipper is
not particularly apt to authorship studies. Glass slippers are physical objects
that feet either fit or do not fit, but the measurement of style is much less
absolute. A size-ten foot cannot fit into a size-four glass slipper. But that is
a statement of a different order from saying that a particular play or scene
cannot be by Shakespeare because it has fewer ‘enclitics’ and more ‘new
words’, say, than you infer that a Shakespeare poem would have. And the
22.Hugh Craig, ‘George Chapman, John Davies of Hereford, William Shakespeare, and A
Lover’s Complaint’, Shakespeare Quarterly, 63 (2012), 147–74. I am grateful to Hugh Craig for
sending me a pre-publication copy of his paper.
23. J. M. Robertson, Shakespeare and Chapman (London: Unwin, 1917), 7–95, and The Problems
of the Shakespeare Sonnets (London: Routledge, 1926), 109–17.
24. Elliott and Valenza, ‘Oxford by the Numbers’, 337.
218
Determining the Shakespeare Canon
distinction between negative and positive evidence is less sharp than Elliott
and Valenza suppose. Positive evidence in favour of Author A may be negative evidence against Authors B, C, D, . . . Z. Whatever methodology one
adopts, comparisons between Shakespeare and non-Shakespeare texts
remain of the essence, and there are limits on how many of the latter can
be investigated—limits imposed not only by time constraints but by the
fact that so much early modern literature has perished. The most pertinent
distinction is between evidence that survives critical scrutiny and evidence
that does not. The negative case against Thorpe’s ascription of A Lover’s
Complaint to Shakespeare is weak and the positive evidence defending it is
strong.
Appendix 1: Literature Online Data for
Chapter 1
part a
Below are recorded Quarrel Scene phrases and collocations that are found,
through Literature Online searches, not more than five times in drama of the period
1580–1600. Lines from the Quarrel Scene of Arden of Faversham are quoted (in
italics) from Wine’s edition, with verse line-divisions marked and with the line
number(s) in parentheses following the quotation. Other plays are quoted from
Literature Online as modernized by myself, except that excerpts from Shakespeare
match the Riverside edition; in these citations the beginnings of lines are marked
only by capitalization, and act, scene, and line references are not provided, since
these are not available in the database. An em-dash is used to indicate a change of
speaker. When only the Arden of Faversham phrase is quoted, the other play’s verbal
agreement with that quotation is exact.
The assignment of plays to authors is in some cases doubtful. When words and
phrases appear to be listed two or more times, this is because the citations cover
different rare links. Thus in the first item, ‘disturbèd thoughts’, this exact locution is found both in the Quarrel Scene and The Troublesome Reign of King John. In
the second item, where ‘disturbèd thoughts’ appears again as part of the Quarrel
Scene locution, it does so as a reinforcement of the broader link with two plays in
which mental turmoil or discontent ‘drives’ somebody into solitude, while the
actual phrase ‘troubled mind’ is used in expressing this idea in both the Quarrel
Scene and Romeo and Juliet. Other phrases or collocations within the opening lines
of the Quarrel Scene provide further rare links.
Disturbèd thoughts (1): Anon., The Troublesome Reign of King John; the ‘trouble’
of Mosby’s ‘moody brain’, which ‘feebles’ his ‘body’, is similar to the way that the
‘disturbed thoughts’ ‘Confound my wits, and dull my senses’ in Troublesome Reign.
Disturbèd thoughts drives me from company | . . . Continual trouble of my moody
brain | . . . troubled mind is stuffed with discontent (1–10): ‘A discontented humour drave
me thence’, Peele, The Battle of Alcazar; ‘A troubled mind drive me to walk abroad’,
Shakespeare, Romeo and Juliet; these are the only instances of negative mental states
‘driving’ somebody into solitude.
220Appendix 1
trouble of my . . . brain (3): ‘troubled brain’, Hughes and others, The Misfortunes of
Arthur.1
moody brain (3): ‘moody thoughts’, Shakespeare and others, 3 Henry VI; the only
example of moody brain, mind, or thoughts.
moody . . . discontent (3–10): ‘moody discontented’, Shakespeare and others, 1
Henry VI; Shakespeare, Richard III; there are no other collocations of ‘moody’ and
‘discontent(ed)’ within the space of sixty words.
Feebles my body (4): ‘feeble body’, Shakespeare and others, 2 Henry VI.
nips me as the bitter northeast wind | Doth check the tender blossoms in the spring
(5–6): ‘nipped the blossoms of our budding spring’, Anon., The Pilgrimage to
Parnassus; ‘If frosts . . . Nip not the . . . blossoms’, Shakespeare, Love’s Labour’s Lost;
two other examples juxtaposing ‘blossoms’ and the verb ‘nip’ (in Kyd’s The Spanish
Tragedy and King Leir) relate specifically to ‘winter’ cold, as does Greene’s ‘nipping
winter frosts’ in James IV, cited by Wine, 72 n. 5–6; since the point of the Arden
image is the premature destruction of budding spring blossoms, the following passages are more closely parallel: ‘Cold news for me . . . Thus are my blossoms blasted
in the bud’, Shakespeare and others, 2 Henry VI; ‘a May blossom with pernicious
winds . . . sullied’, Shakespeare and others, Edward III; ‘This bitter wind must nip
somebody’s spring’, Anon., The Troublesome Reign of King John; ‘Which is a flower
in spring may soon be nipped | With the least frost of cold adversity’, Chettle,
Dekker, and Haughton, Patient Grissil; these are all counted in the calculations
because their points of verbal contact with Arden differ.
the bitter northeast wind (5): ‘the blasting north-east wind’, Shakespeare and others, Edward III; ‘the northeast wind . . . blew bitterly against our faces’, Shakespeare,
Richard II.
check the tender blossoms (6): ‘check the blossoms of delight’, Anon., Mucedorus.
tender blossoms (6): Marlowe, 1 Tamburlaine.
Well fares the man, howe’er his cates do taste, | That tables not with foul suspicion; | And
he but pines amongst his delicates | Whose troubled mind is stuffed with discontent. | My
golden time was when I had no gold; | Though then I wanted, yet I slept secure (7–12): ‘His
wonted sleep under a fresh tree’s shade, All which secure and sweetly he enjoys,
Is far beyond a prince’s delicates—His viands sparkling in a golden cup, His
body couched in a curious bed, When care, mistrust, and treason waits on him’,
Shakespeare and others, 3 Henry VI; ‘where care lodges, sleep will never lie; But
where unbruised youth with unstuff’d brain Doth couch his limbs, there golden
sleep doth reign’, Shakespeare, Romeo and Juliet; the 3 Henry VI passage shares with
the Arden one the specific words ‘sleep’, ‘secure’, ‘delicates’, and ‘golden’ and the
ideas of eating and sleeping without mistrust; the Romeo and Juliet passage shares
1.Compare ‘her troubled brain’ and ‘the brain being troubled’ in Venus and Adonis, 1040
and 1068. Rare phrases and collocations that the Quarrel Scene shares with Shakespeare’s
narrative poems or sonnets are relegated to notes, because they have not been collected
through systematic searches of poetry written within a predetermined period, in the manner employed for plays.
Appendix 1
221
‘unstuffed’ of the mind or brain, ‘golden’, and the idea of sleeping (‘slept’ and
‘sleep’) without care. These are easily the closest parallels to Arden, though ‘sleep/
slept/sleepest secure’ and ‘troubled mind’ are common.2
howe’er his cates do taste (7): ‘cates . . . taste’, Anon., The Thracian Wonder;
Chapman, The Blind Beggar of Alexandria; ‘taste . . . cates’, Anon., The Wisdom of
Doctor Dodypoll (twice).
foul suspicion (8): Brandon, The Virtuous Octavia.
golden time (11): Brandon, The Virtuous Octavia; Shakespeare and others, 3 Henry
VI; ‘golden times’, Shakespeare, 2 Henry IV.3
My daily toil begat me night’s repose (13): ‘Winding up days with toil, and nights
with sleep’, Shakespeare, Henry V, in a similar context of the advantages of the
simple life.
My night’s repose made daylight fresh to me (14): ‘the days be fresh’, Greville,
Alaham; ‘Goodly day toward, and a fresh morning’, Jonson, Every Man in His
Humour; ‘the freshest summer’s day’, Shakespeare and others, Edward III; ‘fresh
days’, Shakespeare, A Midsummer Night’s Dream.
night’s repose (14): the conjunction of ‘night’ and the noun ‘repose’ occurs only
in ‘Good night, and good repose’, Shakespeare, Julius Caesar; ‘Sport and repose
lock from me day and night’, Shakespeare, Hamlet; the verb ‘repose’ is frequently
juxtaposed to ‘night’.
But, since I climbed the top bough of the tree | And sought to build my nest among the
clouds, | Each gentlest airy gale doth shake my bed | And makes me dread my downfall to the
earth (15–18): ‘They that stand high have many blasts that shake them, And if they
fall, they dash themselves to pieces . . . but I was born so high, Our aery buildeth
in the cedar’s top And dallies with the wind . . . cloudy . . . aery’s nest’, Shakespeare,
Richard III; ‘A lofty cedar tree . . . On whose top branches kingly eagles perch . . . the
highest bough of all’, Marlowe, Edward II; ‘Thus yields the cedar . . . Whose arms
gave shelter to the princely eagle, . . . Whose top-branch over-peer’d Jove’s spreading tree’, Shakespeare and others, 3 Henry VI; the Richard III passage has multiple
links with the Arden lines; the other two are the only ones in which the ‘top’
branch or bough of a tree is associated with nesting (or, in Marlowe, perching);
both, like the Arden lines, have a figurative meaning.
climbed the top bough of the tree (15): ‘catched at the highest bough’, Dekker, Old
Fortunatus (with a similar figurative sense); ‘tree tops’, Shakespeare, Romeo and Juliet.
climbed . . . the tree (15): ‘he climbed a tree’, Chettle and Munday, The Death of
Robert Earl of Huntingdon; ‘climbs up a tree’, The Thracian Wonder; ‘wouldst climb a
tree’, Shakespeare and others, 2 Henry VI.
climbed . . . to build my nest (15–16): ‘climb’d unto their nest’, Shakespeare and others, 3 Henry VI; ‘must climb a bird’s nest’, Shakespeare, Romeo and Juliet; the only
juxtapositions of the verb ‘climb’ and ‘nest’.
2. Compare also ‘troubled minds that wakes’ in The Rape of Lucrece, 126; ‘that wakes’ connects
with the ‘watchfulness’ of line 2 of Mosby’s soliloquy.
3. The phrase ‘golden time’ also occurs in Sonnets, 3.12, and Twelfth Night, 5.1.382.
222Appendix 1
Sought to build my nest among the clouds (16): ‘covet to build thy nest in the sun’,
Lyly, Sapho and Phao; ‘among the clouds’, Heywood, The Four Prentices of London.
Each . . . airy gale doth shake my bed (17): ‘by whirlwinds shaken’, Anon., An
Alarum for London; ‘a wind Not of sufficient power to shake a reed’, Heywood,
1 Edward IV; ‘wind-shaken’, Yarington, Two Lamentable Tragedies; ‘as mountains
are for winds, That shake not, though they blow perpetually’, Shakespeare, The
Taming of the Shrew.
gentlest airy gale (17): ‘gentle gale’, Greene, Selimus; Marlowe, Edward II; Peele,
The Battle of Alcazar.4
But whither doth contemplation carry me? (19): ‘O whither doth my passion carry
me’, Marston, Histriomastix.
The way I seek to find (20): ‘Seeking a way . . . Not knowing how to find’,
Shakespeare and others, 3 Henry VI; ‘If thou seekst that way, there thou shalt find
her’, Porter, The Two Angry Women of Abingdon; ‘How could he see his way to seek
out you?’, The Two Gentlemen of Verona.
I seek to find (20): Chapman, The Blind Beggar of Alexandria.
where pleasure dwells (20): Greene and Lodge, A Looking Glass for London.
hedged behind me (21): ‘Had hedged your person’, Greene, Selimus; ‘England,
hedg’d in with the main’, Shakespeare, King John; ‘To hedge me in’, Shakespeare,
Julius Caesar; ‘hedg’d me’, Shakespeare, The Merchant of Venice; these are the only
examples of ‘hedge’ used as a verb meaning ‘close, enclose’.
perish thou (23): Marlowe and Nashe, Dido Queen of Carthage.
Greene doth . . . weed thee up (24): ‘So one by one we’ll weed them all at last’,
Shakespeare and others, 2 Henry VI; ‘But say this weed her love from Valentine’,
Shakespeare, The Two Gentlemen of Verona; the only instances of ‘weed’ used figuratively as a verb.
ear the land (24): Shakespeare, Richard II.5
To make my harvest nothing but pure corn (25): ‘That thrust his sickle in my harvest
corn’, Kyd, Soliman and Perseda; ‘I to my harvest whose corn is now come out of the
blade’, Lyly, Love’s Metamorphosis.
And for his pains I’ll hive him up awhile | And, after, smother him to have his wax; |
Such bees as Greene must never live to sting (26–8): ‘When like the bee tolling from
every flower The virtuous sweets, Our thighs pack’d with wax, our mouths with
honey, We bring it to the hive, and like the bees, Are murd’red for our pains’,
Shakespeare, 2 Henry IV; a multiple parallel: ‘for his/our pains’, ‘hive’, ‘wax’,
‘bees’; three of the Arden elements (‘hive(s)’, ‘bees’, ‘sting(s)’) are present in ‘He is
4.‘Gentlest airy’ is an emendation of Q’s ‘gentle stary’, which Wine interprets as ‘gentle
starry’. Craik rightly regards ‘gentlest airy’ as much more likely to be what the playwright
intended. The inventory of parallels is not affected by acceptance of Craik’s reading rather
than Wine’s.
5. Compare ‘ear so barren a land’ in Shakespeare’s dedication to Venus and Adonis; ‘He that ears
my land’ in All’s Well That Ends Well (1.3.44); ‘O then we bring forth weeds . . . and our ills
told us | Is as our earing’, Antony and Cleopatra (1.2.109–11); and ‘ear’ again meaning ‘plough’
in Antony and Cleopatra, 1.4.49.
Appendix 1
223
not worthy of the honey-comb That shuns the hives because the bees have stings’,
Anon., Locrine.6
wax . . . bees . . . sting (27–8): ‘bees in swarms, and bring forth wax and honey’,
Lyly, Love’s Metamorphosis; ‘Some say the bee stings, but I say, ’tis the bee’s wax’,
Shakespeare and others, 2 Henry VI; the only places in which wax is mentioned in
connection with bees; the Shakespeare parallel is the closer, since it also mentions
the sting.
Such bees as Greene must never live to sting (28): ‘Let not this wasp outlive, us
both to sting’, Shakespeare and Peele, Titus Andronicus; the only example of ‘(out)
live . . . to sting’; ‘Injurious wasps, to feed on such sweet honey, And kill the bees
that yield it with your stings’, Shakespeare, The Two Gentlemen of Verona; though
there are more than five examples of ‘bees’ collocated with ‘stings’, this one is
closely related to the Arden passage because of the shared idea of killing bees that
yield something desirable.
Who, when they shall see (31): ‘who when they see’, Chettle and Munday, The
Death of Robert Earl of Huntingdon; Daniel, Cleopatra; Shakespeare, Richard II.
sit in Arden’s seat (31): ‘sit in friendship’s seat’, Kyd, Soliman and Perseda; ‘sit in the
seat’, Jonson, A Tale of a Tub.
insult upon (32): Dekker, Old Fortunatus; Heywood, The Four Prentices of London;
Jonson, Every Man in His Humour.
for my meed (32): ‘for his meed’, Shakespeare, Richard III; the only other example
of ‘for [possessive pronoun] meed’ used ironically.
fright me (33): Anon., The Tragedy of Caesar and Pompey; Marston, Antonio’s
Revenge; Shakespeare, A Midsummer Night’s Dream; Shakespeare, Julius Caesar.
I’ll none of that (34): Kyd, The Spanish Tragedy; Marlowe, The Jew of Malta;
Yarington, Two Lamentable Tragedies; ‘We’ll none of that’ (where ‘We’ means ‘I’),
Shakespeare, A Midsummer Night’s Dream.
I can cast a bone | To make these curs pluck out each other’s throat (34–35): ‘When two
dogs are at strife for a bone, it is commonly seen That the third comes and takes it
and wipes their mouths clean’, Munday, Fedele and Fortunio;7 ‘England now is left
To tug and scamble, and to part by th’ teeth The unowed interest of proud swelling state. Now for the bare-pick’d bone of majesty Doth dogged war bristle his
angry crest, And snarleth in the gentle eyes of peace’, Shakespeare, King John; the
dog-fight over a bone is vividly implicit in the imagery of the King John passage,
6. Wine retains Q’s ‘heave’ (‘heaue’), but Craik’s adoption of Delius’s ‘hive’ is strongly supported by the parallel with 2 Henry IV. This is so regardless of the authorship of the Quarrel
Scene. Failure to emend would eliminate only the Locrine link. In the 2 Henry IV lines,
metre suggests that ‘pack’d’ has been misplaced, and should follow ‘honey’.
7.Q Fedele and Fortunio accidentally transposes ‘dogs’ and ‘bone’. Mosby’s image of causing
dogs to fight over a bone recurs in Arden of Faversham, when Greene intervenes to stop Black
Will and Shakebag squabbling: ‘I pray you, sirs, list to Aesop’s talk: | Whilst two stout dogs
were striving for a bone, | There comes a cur and stole it from them both’ (9.30–2). The
sole allusion to Aesop within the recognized Shakespeare canon associates him with fables
about curs: ‘Let Aesop fable in a winter’s night, | His currish riddles sorts not with this
place’ (3 Henry VI, 5.5.25–6).
224Appendix 1
which shares with Arden a ferocity absent from Munday’s lines; although there is
no bone in the following lines, they are linked to Arden by the implicit dog image
and the attack on the ‘throat’: ‘What? Were you snarling all before I came, Ready
to catch each other by the throat’, Shakespeare, Richard III.
but she’s myself (37): ‘He is myself ’, Brandon, The Virtuous Octavia; Mary Sidney
Herbert, Antonius; ‘Silvia is myself ’, Shakespeare, The Two Gentlemen of Verona; the
first two examples use a pronoun, as does Arden, but the third is in another way
closer to Arden in connecting a woman to a male ‘myself ’.
And holy church rites makes us two but one (38): ‘Till Holy Church incorporate
two in one’, Shakespeare, Romeo and Juliet; ‘that great vow Which did incorporate
and make us one’, Shakespeare, Julius Caesar; ‘if we two be one’, Shakespeare, The
Comedy of Errors; ‘man and wife, being two, are one in love’, Shakespeare, Henry V;
in Errors, Adriana is also speaking of man and wife, ‘undividable incorporate’, so
that she equates Antipholus (‘thyself ’) with herself.
I may not trust you (39): ‘I may not trust thee’, Shakespeare, King John.
’tis fearful sleeping in a serpent’s bed (42): ‘a bed of crawling serpents’, Anon.,
Captain Thomas Stukeley; ‘sleeping in my orchard, A serpent stung me’,
Shakespeare, Hamlet; ‘do thy best To pluck this crawling serpent from my breast!
Ay me, for pity! What a dream was here! Lysander, look how I do quake with
fear’, Shakespeare, A Midsummer Night’s Dream; ‘Were there a serpent seen, with
forked tongue, That slily glided towards your Majesty, It were but necessary you
were wak’d, Lest being suffer’d in that harmful slumber, The mortal worm might
make the sleep eternal’, Shakespeare and others, 2 Henry VI; Stukeley affords the
only ‘bed’ of ‘serpents’; the Shakespearean examples all associate the danger of
the serpent with sleeping, Hermia having just awoken in Dream, which also has
‘fear’ linking with Arden’s ‘fearful’.8
I will cleanly rid my hands of her (43): ‘rid my hands of him’, Shakespeare, 2 Henry
IV; ‘rid his hands of her’, Shakespeare, The Taming of the Shrew; ‘rid my hands of
them’, Anon., Fair Em; ‘rid our hands of this fellow’, Anon., The Famous Victories
of Henry V.
sad and passionate (45): Shakespeare, King John.
Make me partaker of thy pensiveness (46): ‘Make us partakers of a little gain’,
Shakespeare and others, 1 Henry VI; ‘Wish me partaker in thy happiness’,
Shakespeare, The Two Gentlemen of Verona.
Fire divided burns with lesser force.— | But I will dam that fire in my breast | Till
by the force thereof my heart consume (47–9): ‘Fire that’s closest kept burns most
of all’, Shakespeare, The Two Gentlemen of Verona; ‘love’s hot fire . . . The more
thou dam’st it up, the more it burns’, Shakespeare, The Two Gentlemen of Verona;
‘Sorrow concealed, like an oven stopp’d, Doth burn the heart to cinders where it
is’, Shakespeare and Peele, Titus Andronicus; these are the only passages containing
8. Compare also ‘Who sees the lurking serpent steps aside; | But she sound sleeping, fearing no
such thing, | Lies at the mercy of his mortal sting’ (The Rape of Lucrece, 362–4).
Appendix 1
225
the central idea that the enclosed (or dammed up) fire burns most intensely; fires
within the breast are common.9
lesser force (47): Marlowe, Edward II.
like to a cannon’s burst | Discharged against a ruinated wall (51–2): ‘Hath planted a
double cannon in the door Ready to discharge it upon you’, Kyd, Soliman and
Perseda; ‘a cannon’s crack Discharged against the battlements of heaven’, Anon.,
The Troublesome Reign of King John; ‘ruinate the noble Theban wall’, Newton,
Thebais.10
relenting heart (53): Heywood, 2 Edward IV.
Thou know’st it well (55): Shakespeare, Richard III.
’tis thy policy | To (55–6): ‘it is your policy To’, Shakespeare and others, 1 Henry
VI; ‘’tis but his policy to’, Shakespeare and others, 3 Henry VI; ‘It is his policy
to’, Shakespeare and others, 3 Henry VI; ‘’tis some policy To’, Shakespeare, Love’s
Labour’s Lost; ‘it is King Edward’s policy To’, Peele, Edward I.
To forge distressful looks to wound a breast (56): ‘forge alluring looks, And feign
deep oaths to wound poor silly maids’, Kyd, Soliman and Perseda.11
to wound a breast | Where lies a heart that dies when thou art sad (56–7): ‘needs must
wound thy breast For it hath wellnigh slain my heart’, Peele, Edward I; the closest
parallel.12
a heart that dies (57): ‘the heart that dies’, Shakespeare and Peele, Titus Andronicus;
although the idea of a heart dying occurs more than five times, this is the only
example of the exact phrase, and ‘thy angry frown’ causes the dying in Titus, while
‘looks’ cause it in Arden.
when thou art sad (57): Marlowe and Nashe, Dido Queen of Carthage.
It is not love that loves to anger love.— | It is not love that loves to murder love (58–
9): ‘They do not love that do not show their love.—O, they love least that let men
know their love’, Shakespeare, The Two Gentlemen of Verona; ‘I cannot leave to
love, and yet I do; But here I leave to love where I should love’, Shakespeare, The
Two Gentlemen of Verona; ‘This hand, which for thy love did kill thy love, Shall
for thy love kill a far truer love’, Shakespeare, Richard III; ‘Love loving not itself,
none other can’, Shakespeare, Richard II; the first three examples share with Arden
9. The passage cited from Titus Andronicus seems to me to confirm Craik’s emendation of Q’s
‘part’ to ‘heart’ (which would have been spelt ‘hart’). The inventory of links is not affected
by the change. Compare the very close parallel to Mosby’s lines in Venus and Adonis: ‘An
oven that is stopp’d . . . | Burneth more hotly . . . | So of concealed sorrow may be said, | Free
vent of words love’s fire doth assuage’ (331–4).
10. Also ‘discharged cannon’ and ‘ruinate . . . buildings’, The Rape of Lucrece, 1043, 944.
11.The image, which has eyes feigning oaths, is confused, and whereas in Kyd ‘forge’ simply means ‘simulate’, in the Arden of Faversham passage it retains a hint of a blacksmith’s
weapon-making, and so interacts with the verb ‘wound’ to vivify the metaphor. The two
passages are more fully analysed in Chapter 4, Section II.
12. The closest parallel to Mosby’s ‘looks to wound a breast | Where lies a heart’ is ‘never wound
the heart with looks again’ in Venus and Adonis, 1042.
226Appendix 1
repetition of ‘love’ and play on the word, with the third also mentioning killing;
the final example, like the Arden lines, exposes a paradox.13
How mean you that? (60): Anon., King Leir; Greene, James IV; Jonson, Every Man
out of His Humour; Shakespeare, The Taming of the Shrew (twice).
conceal the rest (63): Greene, Orlando Furioso.
Lest that my words be carried with the wind (64): ‘good wind, blow not a word away’,
Shakespeare, The Two Gentlemen of Verona; ‘for fear the privy whispering of the
wind Convey our words amongst unfriendly ears’, Kyd, The Spanish Tragedy.
published in the world to both our shames (65): ‘publish to the world’, Greene and
Lodge, A Looking Glass for London; Munday and others, Sir Thomas More; ‘my published shame’, Peele, David and Bethsabe.
let our springtime wither (66): ‘when it begins to spring, I’ll let it wither while it
is in bud’, Chettle and Munday, The Downfall of Robert Earl of Huntingdon; ‘our
spring-time’, Shakespeare and others, 3 Henry VI.
loathsome weeds (67): Anon., A Knack to Know a Knave.
what hath passed betwixt us (68): ‘what hath pass’d between me and Ford’s wife’,
Shakespeare, The Merry Wives of Windsor; ‘what hath pass’d between you and
Claudio’, Shakespeare, Much Ado About Nothing.
I blush and tremble at the thoughts (69): ‘blush . . . and tremble at the’, Greene, John
of Bordeaux; ‘makes me tremble at the thought of it’, Anon., Locrine.
my former happy life (71): ‘his former life’, Anon., The Famous Victories of Henry
V; ‘your former life’, Anon., The Famous Victories of Henry V; ‘my former life’,
Yarington, Two Lamentable Tragedies.
honest Arden’s wife (73): ‘an honest plain carpenter’s wife’, Munday and others,
Sir Thomas More; ‘an honest man’s wife’, Shakespeare, 1 Henry IV.14
honest wife (73): Haughton, Grim the Collier; ‘honest loyal wife’, Heywood, 1
Edward IV.
sland’rous to (75): Shakespeare, King John; in both cases the meaning is, unusually, ‘disgraceful’.15
to all my kin (75): Shakespeare and others, 3 Henry VI; ‘all my kin’, Newton,
Thebais; Anon., A Knack to Know a Knave.
13. For similar repetition and play on ‘love’ see ‘My love to love is love but to disgrace it’, Venus
and Adonis, 412. With the second citation from The Two Gentlemen of Verona compare the
quibbling rhetoric of Arden of Faversham, 10.86–90, in which ‘love’ occurs five times, ‘leave’
twice, and ‘live’ or ‘life’ five times.
14. Alice’s declaration that she changed to again being ‘honest Arden’s wife, not Arden’s honest wife. | Ha, Mosby, ’tis thou hast rifled me of that’ is closely paralleled in The Rape of
Lucrece: ‘I was a loyal wife: | So am I now—O no, that cannot be, | Of that true type hath
Tarquin rifled me’ (1048–50); also ‘Pure Chastity is rifled of her store’ (692); no other author
of the period uses ‘rifled’ in connection with theft of a woman’s status as chaste wife.
15. The combination ‘sland’rous . . . to’ also occurs in The Rape of Lucrece, 1001. The adjective is
a favourite with Shakespeare, occurring a dozen times in his works; but to the instances in
King John and Lucrece may be added only one more (in Julius Caesar, 4.1.20) where the meaning, as in Arden, is ‘disgraceful’ (rather than ‘calumnious’).
Appendix 1
227
Even in my forehead is thy name engraven (76): ‘doth engrave Upon thy brows the
drift of thy disgrace, Thy new-vowed love in sight of God and men’, Greene,
James IV.16
A mean artificer, that low-born name (77): ‘a mean and low-born maid’, Anon.,
The Maid’s Metamorphosis; ‘low-born’, Daniel, Cleopatra; the only examples of the
compound; ‘Another lean unwash’d artificer’, Shakespeare, King John; the only
contemptuous, snobbish use of ‘artificer’.
Woe worth the hapless hour (78): ‘Woe worth this hapless heavy hap’, Newton,
Thebais; ‘hapless hour’, Greene, Selimus.
let me breathe curses forth (80): ‘What curses breathe these men!’, Jonson,
Every Man out of His Humour; ‘the curses which the Furies breathe’, Marlowe, 1
Tamburlaine; ‘let the Church, our mother, breathe her curse’, Shakespeare, King
John; ‘Thus have you breath’d your curse’, Shakespeare, Richard III; ‘breathe out
curses’, Anon., Locrine.
stand so nicely at your fame (81): ‘stand upon such nice excuses’, Anon., Edmond
Ironside; ‘stand . . . on nice points’, Shakespeare and others, 3 Henry VI.
so nicely (81): Greville, Alaham; Shakespeare, Richard II.
the credit I have lost (82): ‘hath lost his credit’, Greene, George a Greene; ‘Now is
my credit lost’, Greene, James IV.
I have neglected (83): Anon., Fair Em.
matters of import (83): Anon., Look about You; Marlowe, The Massacre at Paris.
an honest maid (88): Shakespeare, The Merry Wives of Windsor; ‘honest maid’,
Greene, James IV.
Whose dowry would have weighed down all thy wealth (89): ‘Her dowry shall weigh
equal with a queen’, Shakespeare, King John; ‘no less weight Than Aquitaine, a
dowry for a queen’, Shakespeare, Love’s Labour’s Lost; ‘Her dowry wealthy’,
Shakespeare, The Taming of the Shrew; the first two are the only passages juxtaposing ‘dowry’ with ‘weigh’ or ‘weight’, and the last the only one juxtaposing
‘dowry’ with ‘wealth’ or ‘wealthy’.
weighed down (89): ‘weigheth down’, Anon., The Toublesome Reign of King John;
Lyly, Sapho and Phao; ‘weigh down’, Anon., Lust’s Dominion; Greville, Alaham.
Whose beauty and demeanour far exceeded thee (90): ‘beauty that exceeds’, Greene,
James IV; ‘exceeds her as much in beauty as’, Shakespeare, Much Ado About Nothing;
‘That . . . I might in . . . beauties . . . Exceed account’, Shakespeare, The Merchant of
Venice.
16. Omitted from this record is ‘that forehead . . . | Where should be graven . . . The slaughter of
the prince’ in Richard III, 4.4.140–2, because ‘graven’ is the Q reading, where F, followed
by Evans and most other modern editors, has ‘branded’. However, John Jowett accepts Q
as his control text and so reads ‘graven’ in The Tragedy of King Richard III, ed. John Jowett
(Oxford: OUP, 2000). Compare also ‘my digression is so vile, so base | That it will live
engraven in my face’, The Rape of Lucrece, 202–3 (of sexual sin, as in Arden). Passages in
which faults are ‘written’ or ‘branded’ in the forehead or brow are fairly common in plays of
1580–1600, but use of the verb ‘(en)grave’ in such contexts is not.
228Appendix 1
That showed my heart a raven for a dove (97): ‘Who will not change a raven for a
dove?’, Shakespeare, A Midsummer Night’s Dream; ‘Seems he a dove? his feathers
are but borrow’d, For he’s disposed as the hateful raven’, Shakespeare and others, 2
Henry VI; ‘Dove-feather’d raven’, Shakespeare, Romeo and Juliet.17
I knew thee not (99): Anon., The Rare Triumphs of Love and Fortune; ‘I knew not
thee’ (‘I knew thee not’ in octavo 1595), Shakespeare and others, 3 Henry VI.
And now the rain hath beaten off thy gilt | Thy worthless copper shows thee counterfeit
(100–1): ‘Our gayness and our gilt are all besmirch’d With rainy marching in the
painful field’, Shakespeare, Henry V; ‘if you do not all show like gilt twopences to
me’, Shakespeare, 2 Henry IV; ‘Iron of Naples hid with English gilt’, Shakespeare
and others, 3 Henry VI; in the first example, as in Arden, rain affects gilt; the other
two excerpts apply the image figuratively to persons, as does Arden; three instances
of the literal uttering of copper for gold in Wilson’s The Three Lords and Three
Ladies of London have not been recorded.18
It . . . mads me (102–3): ‘this mads me’, Dekker, Old Fortunatus; Porter, The Two
Angry Women of Abingdon; ‘this . . . mads me’, Shakespeare, Richard II; ‘he mads me’,
Jonson, Every Man in His Humour.
how foul thou art (102): ‘how foul she is’, Shakespeare and others, Edward III; ‘How
foul it is’, Shakespeare, 2 Henry IV; ‘how foul . . . is thine image’, Shakespeare, The
Taming of the Shrew.
Go, get thee gone (104): Haughton, Grim the Collier; Jonson, A Tale of a Tub;
Shakespeare, The Comedy of Errors; Shakespeare, The Taming of the Shrew (twice);
‘Get thee gone’, without the preceding ‘Go’, is common.
I am too good to be thy favourite (105): ‘I am . . . too good to be your concubine’,
Shakespeare and others, 3 Henry VI; ‘Thou art . . . Too good to be so’, Shakespeare,
Richard II.19
find it true (106): Greene, James IV; Haughton, Grim the Collier; Yarington, Two
Lamentable Tragedies.
often hath been told (107): ‘Often . . . told’, Shakespeare, The Merchant of Venice.
told me by my friends (107): ‘advised . . . by my friends’, Jonson, Every Man out of
His Humour.
Nay, hear me speak, Mosby, a word or two; | I’ll bite my tongue if it speak bitterly
(110–11): ‘I prithee hear me speak. —You speak too bitterly. —Hear me a word’,
Shakespeare, Richard III.
Nay, hear me speak (110): ‘Nay, hear them speak’, Anon., A Larum for London; the
less exact parallels, ‘Nay, hear me’ and ‘Hear me speak’, are common.
17. I have excluded ‘These ravens will seize upon thy dove’ (Chettle and Munday, The Death
of Robert Earl of Huntingdon), from which the idea of a wolf in sheep’s clothing, to shift the
figure from birds to animals, is absent. Just outside the 1580–1600 limits is ‘a raven’s heart
within a dove’, Twelfth Night, 5.1.131.
18. Compare ‘some with cunning gild their copper crowns’, Troilus and Cressida, 4.4.105, where
again the image is applied to a woman’s faithfulness.
19.I have excluded ‘It is too good to be true’ (Lyly, Mother Bombie), where the meaning is
different.
Appendix 1
229
I’ll bite my tongue (111): ‘bite their tongues’, Lyly, The Woman in the Moon; ‘bite
his tongue’, Shakespeare and others, 2 Henry VI; ‘bite thy tongue’, Shakespeare
and others, 3 Henry VI; ‘bite our tongues’, Shakespeare and Peele, Titus Andronicus.
speak bitterly (111): ‘bitterly to speak’, Shakespeare, Richard III.
stormy look (113): ‘you have such a February face, | So full of . . . storm and cloudiness’, Shakespeare, Much Ado About Nothing; ‘stormy forehead’, Greville, Mustapha;
‘stormy brow’, Marston, Histriomastix; ‘Suffolk’s cloudy brow [expresses] his
stormy hate’, Shakespeare and others, 2 Henry VI; these are the closest parallels,
since ‘stormy look’ does not occur.
do penance (115): Jonson, Every Man in His Humour; ‘done penance’, Marlowe,
Doctor Faustus.
prayerbook (116): Heywood, 2 Edward IV; Shakespeare and others, 3 Henry VI;
Shakespeare, Richard III; Shakespeare, The Merchant of Venice; included as a compound because often printed as two words.
holy word (117): Shakespeare, King John; Marlowe, The Massacre at Paris.
converted me (117): ‘convert me’, Marlowe, The Jew of Malta.
I will tear away the leaves (118): ‘that I’ll tear away’ (where ‘that’ is writing on
paper), Shakespeare, The Two Gentlemen of Verona.20
thereon will I chiefly meditate (121): ‘Whereon dost thou chiefly meditate?’, Kyd,
The Spanish Tragedy.
Wilt thou not hear? (124): ‘Wilt thou not hear thy father?’, Anon., Captain Thomas
Stukeley.
Why speaks thou not? (125): ‘Why speakst thou not?’, Anon., Look about You;
‘Why speakest thou not’, Chettle and Munday, The Downfall of Robert Earl of
Huntingdon; Greene, Alphonsus, King of Aragon (twice); Kyd, The Spanish Tragedy.
Thou hast been sighted as the eagle is (126): ‘eagle-sighted’, Shakespeare, Love’s
Labour’s Lost; ‘A lover’s eyes will gaze an eagle blind’, Shakespeare, Love’s Labour’s
Lost; like Alice Arden’s words, both passages relate to love, and, although the keen
vision of the eagle is proverbial, LION yields no further parallels among the plays
searched.
The fearful hare (127): Anon., The Maid’s Metamorphosis; Yarington, Two
Lamentable Tragedies; ‘the fearful flying hare’, Shakespeare and others, 3 Henry VI;
‘the fearfulness of the hare’, Lyly, Midas.21
spoke as smoothly as an orator (128): ‘smooth-tongued orators’, Nashe, Summer’s
Last Will and Testament; ‘speak smooth’, Peele, Edward I; ‘smooth and speak him
fair’, Shakespeare and Peele, Titus Andronicus.
hear or see or speak (129): ‘I hear, I see, I speak’, Shakespeare, The Taming of the Shrew.
20. The only other instance in plays of 1580–1600 of ‘tear away’ is in ‘Ah, do not tear away thyself from me’, The Comedy of Errors, 2.2.124; but this has been omitted as not involving literal
tearing.
21. By far the closest parallel to ‘heard as quickly as the fearful hare’ is found within an extended
description of the hunted hare in Venus and Adonis: ‘the timorous flying hare . . . poor Wat,
far off upon a hill, | Stands on his hinder-legs with list’ning ear, | To hearken if his foes pursue him still’ (674–99), where the hare’s fearfulness sharpens its hearing.
230Appendix 1
art thou sensible in (130): ‘Thou art sensible in’, Shakespeare, The Comedy of Errors;
‘sensible in’, Shakespeare, Love’s Labour’s Lost; ‘I am . . . sensible in’, Shakespeare,
Hamlet.
this little fault (131): ‘a little fault’, Chettle and Munday, The Downfall of Robert
Earl of Huntingdon; Greene, James IV; Shakespeare, Love’s Labour’s Lost; ‘little
faults’, Shakespeare, Henry V.
I deserve not (132): Greene, Selimus.
I deserve not Mosby’s muddy looks. | A fount once troubled is not thickened still; | Be
clear again, I’ll ne’er more trouble thee (132–4): ‘A woman mov’d is like a fountain
troubled, Muddy, ill-seeming, thick, bereft of beauty’, Shakespeare, The Taming
of the Shrew; ‘The purest spring is not so free from mud As I am clear from treason’, Shakespeare and others, 2 Henry VI; ‘whose filth and dirt Troubles the silver
spring where England drinks’, Shakespeare and others, 2 Henry VI; ‘Thou sheer,
immaculate, and silver fountain, From whence this stream through muddy passages Hath held his current and defil’d himself ’, Shakespeare, Richard II; ‘Here
stands the spring whom you have stain’d with mud’, Shakespeare and Peele, Titus
Andronicus; ‘I trouble now the fountain of thy youth, And make it moody’, Anon.,
The Troublesome Reign of King John; within the period, these are the fullest dramatic
parallels to Arden’s imagery of muddied waters used figuratively of emotional or
moral states; all are counted in the calculations because their verbal links with
Arden differ (and they occur in only five plays).22
I am a base artificer; | My wings are feathered for a lowly flight (135–6): since Mosby
is sarcastically referring to limits on his aspirations, one is reminded of the wings
that the artificer Daedalus created for Icarus, and of the image in the prologue
to Marlowe’s Doctor Faustus: ‘His waxen wings did mount above his reach, And,
melting, heavens conspired his overthrow’; but Shakespeare specifically names
Daedalus and Icarus in 3 Henry VI and follows with the line ‘The sun that sear’d
the wings of my sweet boy’; otherwise the nearest parallels are ‘mount aloft . . . And
outstrip the feathered fowls in flight’, Anon., King Leir; ‘the wings of my welltempered verse . . . thrice haughty flight . . . Their mounting feathers scorch not
with the fire’, Peele, David and Bethsabe; though rather doubtful, all four links are
counted in calculations.
a base artificer (135): ‘Another lean unwash’d artificer’, Shakespeare, King John.
Make love to (138): Jonson, The Case Is Altered; Marlowe, The Jew of Malta;
Shakespeare, The Taming of the Shrew; Shakespeare, The Merry Wives of Windsor;
Shakespeare, Hamlet.
Why, ’tis unpardonable (138): ‘O, ’tis . . . unpardonable’, Shakespeare and others,
3 Henry VI.
as gentle as a (140): Dekker, The Shoemakers’ Holiday; Shakespeare, Romeo and
Juliet; ‘gentle as a’: Greene, Selimus.
22.The muddied fountain image-complex is discussed in Chapter 1, Section V. Here Wine
adopts the emendation ‘fount once’ for Q’s ‘fence of ’. Further Shakespeare parallels include
three in The Rape of Lucrece: ‘Mud not the fountain that gave drink to thee’ (577), ‘toads infect
fair founts with venom mud’ (850), and ‘The poisoned fountain clears itself again’ (1707).
Appendix 1
231
too blind (141): Brandon, The Virtuous Octavia.
weeds [spring] in gardens (143): ‘weeds have sprung To stain the beauty of our
garden plot’, Anon., The Troublesome Reign of King John; ‘weeds . . . [will] o’ergrow
the garden’, Shakespeare and others, 2 Henry VI; ‘our . . . garden . . . Is full of weeds’,
Shakespeare, Richard II; surprisingly, these are the only juxtapositions (within the
range ‘NEAR.50’) of ‘weeds’ and ‘garden’.23
roses grow on thorns (143): ‘The rose although in thorny shrubs she spread Is
still the rose’, Greene, James IV; ‘Thorns lie in garrison about the roses’, Jonson,
Cynthia’s Revels; ‘a red rose from off this thorn’, Shakespeare and others, 1 Henry
VI; ‘Hath not thy rose a thorn’, Shakespeare and others, 1 Henry VI; although ‘No
rose without a thorn’ is proverbial, these are the only links discovered to plays of
1580–1600.24
sweet-set tongue (147): ‘sweet tongue’, Shakespeare and Peele, Titus Andronicus;
Shakespeare, Love’s Labour’s Lost; Lyly, Love’s Metamorphosis; ‘sweet tongue’s’,
Shakespeare and others, Edward III; ‘tongue so sweet’, Shakespeare, Romeo and
Juliet; there are other linkings of sweetness and tongues but these are the closest in
wording to Arden.
forget this quarrel (148): Shakespeare and others, 1 Henry VI.
Then with thy lips seal up this new-made match (150): ‘The duty that I owe unto
your Majesty I seal upon the lips of this sweet babe’, Shakespeare and others, 3
Henry VI; ‘and, lips, O you The doors of breath, seal with a righteous kiss A dateless bargain to engrossing death’, Shakespeare, Romeo and Juliet; ‘The match is
made, she seals it with a cur’sy’, Shakespeare and others, 3 Henry VI; the sealing
of lips in silence, the sealing of lips with a kiss, and the sealing of contracts with a
kiss are common, but the first two parallels cited here are the only ones in which
a compact is sealed specifically with ‘lips’, and the third is unique in having in
common with the Arden line ‘match’, ‘made’, and ‘seal(s) with’.25
seal up this new-made match (150): ‘To seal love’s bonds new made’, Shakespeare,
The Merchant of Venice; ‘his new-made bride’, Shakespeare and others, 3 Henry VI;
‘the new-made bridegroom’, Shakespeare, Romeo and Juliet; the only instances of
23.Compare Hamlet’s ‘unweeded garden’ (1.2.135). Alice’s ‘Flowers do sometimes spring in
fallow lands, | Weeds in gardens’ (142–3) implies that fallow lands normally grow weeds,
as in Henry V: France’s ‘fallow leas | The darnel, hemlock, and rank femetary | Doth root
upon’ (5.2.44–6). The Rape of Lucrece offers: ‘Unwholesome weeds take root with precious
flow’rs’ (870).
24.See also ‘Roses have thorns, and silver fountains mud’, Sonnets, 35.2, where the speaker,
like Alice Arden, is trying to excuse a lover, and the second half of the line, about muddied
fountains, links with Arden, 8.132. Venus and Adonis has ‘though the rose have prickles’ (574),
and The Rape of Lucrece has ‘I know what thorns the growing rose defends’ (492).
25.A stanza in Venus and Adonis is also close to the Arden line: ‘Pure lips, sweet seals in my
soft lips imprinted, | What bargains may I make, still to be sealing?’ Venus asks of Adonis,
continuing with an analogy of buying and selling that leads into ‘Which purchase if thou
make . . . | Set thy seal manual on my wax-red lips’ (511–16). Here again a contract is made
with the lips; compare ‘And seal the bargain with a holy kiss’, The Two Gentlemen of Verona,
2.2.7, and ‘And seal the title with a lovely kiss’, The Taming of the Shrew, 3.2.123.
232Appendix 1
‘new-made’ in connection with marriage or betrothal, with ‘seal’ providing a further link in the Merchant lines.
How now, Bradshaw, what’s the news with you? (152): ‘How now, sirrah, what’s the
news with you?’, Anon., Guy Earl of Warwick (twice); Haughton, Englishmen for My
Money; ‘How now, sir boy, what is the news with you?’, Heywood, 1 Edward IV;
the only instances of the formula with ‘How now’, as distinct from ‘Now’ (in The
Shoemaker’s Holiday or ‘And now’ (in Hamlet).
importuned me to give you (154): ‘importune you To let him’, Shakespeare, The Two
Gentlemen of Verona; ‘importune you To keep’, Shakespeare and others, Edward III;
the only examples of the verb ‘importune’ followed by pronoun, ‘to’, and verb.
a cup of beer (155): Anon., Thomas Lord Cromwell; Anon., Club Law; Chettle, Day,
and Haughton, The Blind Beggar of Bednal Green; Nashe, Summer’s Last Will and
Testament; ‘a cup o’ thy small beer’, Jonson, Every Man in His Humour.
’Tis almost supper time (156): Yarington, Two Lamentable Tragedies.
We have missed of our purpose (157): ‘I missed my purpose’, Jonson, Every Man out
of His Humour; ‘miss of ’ (meaning ‘miss’) Greene, Selimus; Haughton, Englishmen
for My Money; Lyly, Love’s Metamorphosis.
mixed with bitter gall (165): Yarington, Two Lamentable Tragedies; ‘bitter gall’,
Brandon, The Virtuous Octavia; Shakespeare, Romeo and Juliet; Yarington, Two
Lamentable Tragedies; Yarington uses ‘bitter gall’ twice, once prefaced by ‘mixed
with’, as in Arden.
to shun suspicion (166): Marlowe, The Jew of Malta; Porter, Two Angry Women of
Abingdon; Chettle and Munday, The Downfall of Robert Earl of Huntingdon.
to the gates of death to follow thee (167): ‘follow to the gates of death’, Peele, The
Battle of Alcazar; ‘followed . . . to the gates of death’, Anon., The True Tragedy of
Richard III; these are the only two of six references to ‘gates of death’ that also
include the verb ‘to follow’.
part b
Below are recorded phrases and collocations in Doctor Faustus, 18.99–118 and
19.132–90, that are found, through Literature Online searches, not more than five
times in drama of the period 1580–1600.
Was this the face that launch’d a thousand ships | And burnt the topless towers of Ilium?
(18.99–100): ‘Helen, whose beauty summoned Greece to arms, And drew a thousand ships to Tenedos’, Marlowe, 2 Tamburlaine; ‘I never vowed . . . The desolation of his native Troy, Nor sent a thousand ships unto the walls’, Marlowe and
Nashe, Dido Queen of Carthage; these are the only instances of ‘a thousand ships’,
and each, like Faustus’s, is about Helen and the siege of Troy; ‘Was this the face
That’, Shakespeare, Richard II.
launch’d . . . ships (99): ‘all our ships were launched’, Marlowe and Nashe, Dido
Queen of Carthage; ‘Why are thy ships . . . launched?’, Marlowe and Nashe, Dido
Appendix 1
233
Queen of Carthage; there are two non-Marlovian instances of the verb ‘launch’, but
neither mentions ‘ships’, and one is not even nautical.
topless towers (100): ‘Towers that topless touch the clouds’, Greene and Lodge, A
Looking Glass for London.
make me immortal with a kiss (101): Marlowe and Nashe, Dido Queen of Carthage.
Her lips suck forth my soul (102): ‘she . . . sucks my soul forth with a melting kiss’,
Marston, Jack Drum’s Entertainment; ‘suck away their souls’, Shakespeare, Henry V;
Marston’s is much the closer parallel, but these are the only two examples of souls
being sucked.
I will be Paris, and for love of thee | Instead of Troy shall Wittenberg be sack’d (106–
7): ‘So thou wouldst prove as true as Paris did, Would, as fair Troy was, Carthage
might be sacked, And I be called a second Helena!’, Marlowe and Nashe, Dido
Queen of Carthage; ‘poor Troy must now be sacked’, Marlowe and Nashe, Dido
Queen of Carthage; there are no other instances of ‘be sacked’, and these refer to
Troy, the first naming Paris, as does Faustus.
wear thy colours (109): ‘wear those colors’, Shakespeare, 1 Henry IV; ‘wear his
colors’, Shakespeare, Love’s Labour’s Lost; where ‘colours’ means ‘ensign’.
plumed crest (109): Greene, Selimus; Shakespeare and others, 1 Henry VI; ‘highplumed crests’, Heywood, The Four Prentices of London.
Clad in the beauty of (113): Anon., The Tragedy of Caesar and Pompey.
More lovely than (116): Anon., The Taming of a Shrew.
azur’d arms (117): ‘azure arms’, Lyly, The Woman in the Moon.
none but thou shalt be my paramour (118): ‘None but thou Shall be his son-in-law’,
Jonson, The Case is Altered; ‘none but thou’, Kyd, Soliman and Perseda; ‘none but
thou and I’ in Heywood’s Edward IV is discarded as different in meaning.
be my paramour (118): ‘be his paramour’, Shakespeare, Romeo and Juliet; ‘be thy
paramour’, Shakespeare and others, 1 Henry VI; ‘be Rasnes’ paramour’, Greene
and Lodge, A Looking Glass for London.
but one bare hour (19.134): ‘but one hour’, Anon., Lust’s Dominion; Lyly, Love’s
Metamorphosis; ‘one bare hour’, Haughton, Englishmen for My Money.
thou must be damn’d (135): ‘I must be damned’, Anon., The Troublesome Reign
of King John; ‘Bacon must be damned’, Greene, Friar Bacon and Friar Bungay;
‘Thou . . . must be damned’, Marlowe, The Jew of Marlowe.
Stand still, you ever-moving spheres of heaven, | That time may cease, and midnight
never come | . . . That Faustus may repent (136–41): ‘Let never silent night possess this
clime. Stand still, you watches of the element; All times and seasons rest you at
a stay, That Edward may be still fair England’s king’, Marlowe, Edward II; there
are no other examples of ‘Stand still you’, and the two passages have several other
words and ideas in common.
spheres of heaven (136): ‘sphere of heaven’, Marlowe, Edward II; though heavenly
spheres are often mentioned, this is the only instance of the exact phrase ‘sphere(s)
of heaven’.
234Appendix 1
Fair nature’s (138): Haughton, Englishmen for My Money.
Perpetual day (139): Heywood, The Four Prentices of London.
A year, a month, a week, a natural day (140): ‘but king for a year, nay but half a year,
nay a month, a week, three days, one day, or half a day, nay an hour, ’swounds half
an hour’, Anon., The True Tragedy of Richard III; this reads like burlesque.
The stars move still (143): ‘as the star moves not but in his sphere’, Shakespeare,
Hamlet; ‘you stars, that move in your right spheres’, Shakespeare, King John; the
only cases of stars said to move.
Faustus must be damn’d (144): ‘I must be damned’, The Troublesome Reign of King
John; ‘Bacon must be damned’, Greene, Friar Bacon and Friar Bungay; ‘Thou . . . must
be damned’, Marlowe, The Jew of Malta.
leap up (145): ‘leaps up’, Dekker, Old Fortunatus; Jonson, Every Man in His
Humour; the Dekker example is counted, though the leaping is figurative, but
I have excluded Porter’s ‘leap up to the chin in a barrel of beer’ (Two Angry Women
of Abingdon), where ‘up’ goes with ‘to the chin’ not with ‘leap’.
See, see where (146): Warner, Menaechmi.
Christ’s blood streams in the firmament (146): ‘set black streamers in the firmament’, Marlowe, 2 Tamburlaine; the only example of anything streaming in ‘the
firmament’.
Christ’s blood (146): ‘the blood of Jesus Christ’, Yarington, Two Lamentable
Tragedies; ‘Christ, Whose blood must save me’, Anon., Arden of Faversham.
Rend not my heart for naming of my Christ (148): ‘Rend not my heart’, Marlowe,
Edward II; ‘I rent his name that rends my heart’, Marlowe, Edward II; there are several rendings of hearts, but even the second of these, with ‘rend(s) my heart’ and
‘name’, more closely matches Faustus’s utterance than any not listed here.
my Christ; | Yet will I call on him: (148–9): ‘call on Christ’, Anon., The Troublesome
Reign of King John; Marlowe, The Massacre at Paris.
O, spare me (149): Anon., Mucedorus; Marlowe, Edward II; ‘O, spare me not’ in
Shakespeare’s Richard II is omitted from calculations.
God | Stretcheth out his arm (150–1): ‘stretch out our arms’, Anon., Captain
Thomas Stukeley; ‘stretching out my arms’, Anon., Lust’s Dominion; ‘stretcheth
out his . . . arms’, Marston, Antonio and Mellida; ‘If God should stretch his hand’,
Yarington, Two Lamentable Tragedies.
God | . . . bends his ireful brows (150–1): ‘ireful brows’, Anon., A Knack to Know an
Honest Man; ‘God . . . whose . . . brow’, Munday and others, Sir Thomas More; the
latter is the only mention of God’s brow(s).
come, come, and (152): Shakespeare and Peele, Titus Andronicus; Shakespeare, 1
Henry IV; Shakespeare, Hamlet.
the heavy wrath of God (153): Anon., A Warning for Fair Women; ‘I have provoked
God to heavy wrath’, Yarington, Two Lamentable Tragedies; ‘heavy wrath’, Anon.,
Thomas Lord Cromwell.
wrath of God (153): Anon., A Knack to Know a Knave; Greene and Lodge, A
Looking Glass for London; Marlowe, 1 Tamburlaine.
Appendix 1
235
headlong run (155): Greville, Alaham; ‘run headlong’, Greene, James IV; Marston,
Antonio’s Revenge; Yarington, Two Lamentable Tragedies; ‘running headlong’, 1
Tamburlaine.
no, it will not (156): Anon., The Rare Triumphs of Love and Fortune.
You stars that reign’d at my nativity (157): ‘Smile, stars that reigned at my nativity’,
Marlowe, 1 Tamburlaine; ‘every star that reigned when I was born’, Anon., The
Wars of Cyrus; ‘I hope more happy stars will reign today’, Haughton, Englishmen
for My Money; ‘happy stars reigned at the disposition of her beauty’, Chapman,
An Humorous Day’s Mirth; there are no other instances of stars reigning, but the
Marlowe parallel is easily the best.
stars . . . | Whose influence (157–8): ‘the star whose influence’, Marlowe, The
Massacre at Paris; the ‘influence’ of the stars is mentioned several times, but this is
the only ‘whose influence’.
stars . . . hath allotted death (157–8): ‘allot me death’, Lyly, Love’s Metamorphosis;
‘this my stars to me allot’, Peele, The Arraignment of Paris; ‘the end that fate allotteth
me’, Lodge, The Wounds of Civil War.
foggy mist (159): Anon., Arden of Faversham.
Into the entrails of yon labouring cloud (160): ‘in the bowels of a freezing cloud’,
Marlowe, 1 Tamburlaine; ‘Within the entrails of a jetty cloud’, Peele, David and
Bethsabe.
vomit forth (161): ‘vomits forth’, Shakespeare, Richard III.
forth into the air (161): Shakespeare, The Merchant of Venice.
So that my soul may but (163): ‘So that Aeneas may but’, Marlowe and Nashe, Dido
Queen of Carthage; the only ‘So that . . . may but’.
ascend to heaven (163): Haughton, Grim the Collier; Shakespeare, King John;
‘ascend to fame’s immortal house’, Marlowe and Nashe, Dido Queen of Carthage;
the idea of souls going to heaven is common, but these are the only instances of
‘ascend to’.
Yet for Christ’s sake, whose blood hath ransom’d me (167): ‘[God] did send his own
dear son to pay his ransom with his precious blood’, Guy Earl of Warwick; ‘That this
my blood mought thy life’s ransom be’, Brandon, The Virtuous Octavia; ‘No drop
of blood falls from a Christian heart But thy heart’s blood shall ransom’, Heywood,
The Four Prentices of London; ‘Christ, Whose blood must save me’, Anon., Arden of
Faversham.
for Christ’s sake (167): Anon., Arden of Faversham; Heywood, Edward IV;
Yarington, Two Lamentable Tragedies.
Impose some end (168): ‘impose a final end’, Brandon, The Virtuous Octavia.
incessant pain (168): ‘incessant torments’, Anon., Locrine.
live in hell a thousand years (169): ‘in darkness hurled A thousand years, as Satan
was’, Jonson, Every Man in His Humour; ‘a lease of my life for a thousand years’,
Shakespeare and others, 2 Henry VI; ‘live a thousand years’, Shakespeare, Romeo
and Juliet; ‘a thousand years’, Kyd, Soliman and Perseda; ‘thousand years’, Marlowe
and Nashe, Dido Queen of Carthage.
236Appendix 1
and at last (170): Anon., Arden of Faversham; Greene and Lodge, A Looking Glass
for London; Marlowe, The Massacre at Paris.
Why wert thou (172): Kyd, Soliman and Perseda.
Ah, Pythagoras’ metempsychosis, were that true, | This soul should fly from me and
I be chang’d | Unto some brutish beast (174–6): ‘O you departed souls, If Pythagorian
axioms be true Of spirits’ transmigration, fleet no more To human bodies, rather
live in swine, Inhabit wolves’ flesh, scorpions, dogs, and toads’, Marston, Antonio’s
Revenge.
brutish beast (176): ‘brutish beasts’, Anon., Mucedorus; Shakespeare, Julius Caesar;
‘brutish savage beasts’, Greene, Alphonsus, King of Aragon; ‘brutish animal’, Anon.,
Every Woman in Her Humour.
when they die | Their souls are soon dissolv’d in elements (177–8): ‘Until our bodies
turn to elements’, Marlowe, 1 Tamburlaine; the only mention of anything changing
into elements.
Be plagu’d (179): Greene, Orlando Furioso; Jonson, Every Man out of His Humour;
Lyly, Sapho and Phao.
Curs’d be the parents that engender’d me! (180): ‘cursed be the time Of thy nativity’,
Shakespeare, 1 Henry VI; ‘Cursed be my birthday’, Anon., The Troublesome Reign of
King John; ‘Cursed be the day wherein I was born, and accursed be the hour when
I was begotten’, Anon., The Famous Victories of Henry V.
the parents that engender’d me (180): ‘the mother that engend’red thee’,
Shakespeare, Julius Caesar; ‘engendered me’, Marlowe, The Massacre at Paris.
little water drops . . . into the ocean (185–6): ‘in the sea . . . little water drops’, Marlowe
and Nashe, Dido Queen of Carthage; ‘water-drops’, Shakespeare, Richard II; Peele,
The Arraignment of Paris.
Look not so fierce on me! (187): ‘looks so fierce’, Marlowe, 1 Tamburlaine; ‘fierce
looks’, Shakespeare, King John.
Adders and serpents (188): Peele, The Battle of Alcazar.
Ugly hell, gape not! (189): ‘Hell gapes for me’, Greene and Lodge, A Looking Glass
for London; ‘hell gapes’, Kyd, Soliman and Perseda; ‘gape hollow hell’, Anon., The
Rare Triumphs of Love and Fortune; ‘though hell itself should gape’, Shakespeare,
Hamlet.
Appendix 2: Literature Online Data for
Chapter 2
Results of searches of Arden of Faversham, 6.5–31, for phrases and collocations
occurring five or fewer times in plays of 1580–1600:
Come, Master Franklin, you shall go with me (5): ‘Come, Master Banister, you shall
with me’, Anon., Thomas Lord Cromwell; ‘come, Egeus, you shall go with me’,
Shakespeare, A Midsummer Night’s Dream; ‘You must go with me, Master Doctor’,
Shakespeare, The Merry Wives of Windsor; the only parallels juxtaposing ‘come’
and/or the appellation ‘Master’ with ‘you shall/must go with me’.
This night I dreamed (6): ‘This night He dreamt’, Shakespeare, Richard III; ‘I
dreamed upon it all this night’, Anon., The Taming of a Shrew;1 the only instances of
‘dreamed’ or ‘dreamt’ coupled with ‘this’ rather than ‘last’ (or ‘tonight’); the first
example is the closer, since it is followed by an implicit ‘that’, actual in Arden.
being in a (6): Marlowe, Edward II.
in a park, | deer . . . herd’s approach (6–9): ‘hunt a deer in a park’, Lyly, Midas; ‘How
are we park’d and bounded in a pale, A little herd of England’s timorous deer’,
Shakespeare and others, 1 Henry VI; ‘in the park, Seeking to hide herself, as doth the
deer’, Shakespeare and Peele, Titus Andronicus; ‘deer . . . park-corner’, Shakespeare
and others, 3 Henry VI; ‘I chased the deer . . . But where’s your park?’, Dekker, The
Shoemaker’s Holiday; the only instances of ‘deer’ collocated with ‘park’, the latter in
verbal form in 1 Henry VI, which also shares ‘herd’ with Arden.
A toil was pitched to overthrow the deer (7): ‘pitch . . . your toils . . . And rouse
the . . . deer’, Marlowe and Nashe Dido, Queen of Carthage; ‘hunting the
deer . . . pitch’d a toil’, Shakespeare, Love’s Labour’s Lost; ‘pitched . . . his . . . toil’,
Marlowe, The Massacre at Paris; ‘Now have we got the . . . deer | Within the compass of a deadly toil’, Marlowe, The Massacre at Paris; the first two examples have
all three elements, ‘toil’, ‘pitched’, and ‘deer’, whereas the others have only two,
though ‘compass’ parallels Arden’s later ‘rounded’ (14).
1.Annals dates The Taming of a Shrew 1589, but this date rests on the assumption, rejected by
recent editors, that it was a source of Shakespeare’s The Taming of the Shrew (1590–1), rather
that derivative from it, as the Oxford Textual Companion holds (109–11).
238Appendix 2
I upon a little rising hill | Stood whistly watching for the herd’s approach (8–9): ‘stand
upon this molehill’, Shakespeare and others, 3 Henry VI; ‘a little hill’, Peele, The
Old Wives Tale; ‘this little hill’, Shakespeare and others, Edward III; ‘rising of the
hill’, Shakespeare, Love’s Labour’s Lost; ‘wishtly look’d’, Shakespeare, Richard II;
‘wistly follow whiles the game’s on foot’, Shakespeare and others, Edward III; the
first item is included because, although the hill is a mere molehill, the reference
is to standing ‘upon’ it, and the last item because it associates the rare ‘wistly’ (or
‘whistly’) with hunting ‘game’, a word also used later in Arden’s speech (19); six
items are included because they link to the Arden passage through different combinations of pertinent words. Riverside Shakespeare, unlike other editions, retains the
anomalous spelling ‘wishtly’ in Richard II.
Stood . . . watching for (9): ‘stand watching for’, Wilson, The Three Ladies of London;
‘watching for’, Shakespeare and others, 2 Henry VI; Anon., Club Law.
Even there, methoughts (10): ‘even there, methinks’, Shakespeare, King John.
gentle slumber (10): Lodge, The Wounds of Civil War.
sweet repose . . . rest (11–12): ‘sweet repose and rest’, Shakespeare, Romeo and Juliet
(1595); ‘sweet repose’, Daniel, Cleopatra.
rounded me with (where ‘rounded’ means ‘surrounded, encircled’, 14): ‘Rounded
with’, Greene and Lodge, A Looking Glass for London and England; Shakespeare, A
Midsummer Night’s Dream.
methought, was (15): Shakespeare, King John.
blew an evil-sounding horn (16): ‘blew your horns’, Chettle and Munday, The
Downfall of Robert Earl of Huntingdon; ‘blew my horn’, Greene, George a Greene
(1590); present tense blowing of horns is common.
evil-sounding (16): ‘ill-sounding’, Chapman, An Humorous Day’s Mirth; this
compound is doubtfully included: three of the other four compounds of adjective
plus present participle are by Shakespeare, including ‘harsh-sounding’ in Richard
II, but these are excluded from calculations.
at the noise (17): Greene and Lodge, A Looking Glass for London and England;
Jonson, A Tale of a Tub.
With falchion drawn, and bent it at my breast (18): ‘with purple falchion’,
Shakespeare and others, 3 Henry VI; ‘Thy murd’rous falchion, . . . The which thou
once did bend against her breast’, Shakespeare, Richard III; ‘Bend not your dangerous weapon at my breast’, Anon., Lust’s Dominion; ‘though thy sword were at my
breast’, Anon., The Wars of Cyrus.
With this I waked (20): ‘With this I depart’, Shakespeare, Julius Caesar; ‘With
this we charg’d’, Shakespeare and others, 3 Henry VI; ‘With this she fell distract’,
Shakespeare, Julius Caesar; ‘With this we both fell out’, Jonson, Every Man out of
His Humour; there are no other cases of ‘With this I/we/he/she/they’ with the
sense ‘whereupon’.
I . . . trembled every joint (20): ‘I tremble every joint’, Jonson, A Tale of a Tub;
‘I . . . tremble every joint’, Anon., A Warning for Fair Women; ‘my trembling joints’,
Marston, Antonio’s Revenge; Shakespeare and Peele, Titus Andronicus; the first two
Appendix 2
239
items more closely match Arden than the second two, but ‘My trembling joints’
occurs in Arden at 4.95.
obscurèd in a little bush (21): ‘obscured in’, Marlowe, The Massacre at Paris;
Shakespeare, Romeo and Juliet; Shakespeare, As You Like It; ‘hid in a bush’, Anon.,
Guy Earl of Warwick; ‘in a bush’, Shakespeare and others, 3 Henry VI.
a lion foraging (22): ‘behold his lion’s whelp | Forage’, Shakespeare, Henry V; ‘the
Nemean lion . . . from forage will incline to play’, Shakespeare, Love’s Labour’s Lost;
‘The lion doth . . . grace his foragement’, Shakespeare and others, Edward III; the
only passages in which a lion is associated with foraging.
dreadful forest king (23): ‘forest kings’ (specified as lions): Yarington, Two
Lamentable Tragedies; ‘the kingly lion . . . forest’, Shakespeare and others, 3 Henry
VI, ‘a lion . . . is a most dreadful thing’, Shakespeare, A Midsummer Night’s Dream;
lions are associated with kings on five other occasions, two in Shakespeare plays,
not counting a reference to King Richard Coeur-de-lion in King John; the 3 Henry
VI instance distinguishes itself from these by placing the lion in a ‘forest’.
timorous suspect (24): Jonson, The Case Is Altered.
I stood in doubt (29): Anon., Edmond Ironside.
Such great impression took this fond surprise (his dream, 30): ‘Such terrible impression made my dream’, Shakespeare, Richard III.
God grant this vision bedeem me any good! (31): ‘God grant these dreams to good
effect be brought’, Kyd, Cornelia; ‘this vision’, Lyly, Endymion; Anon., The Maid’s
Metamorphosis, twice; Shakespeare, Hamlet.2
2. I suspect that Arden’s unknown ‘bedeem’ is an error for ‘beteem’, meaning ‘grant, allow’,
as in Hamlet’s ‘he might not beteem the winds of heaven | Visit her face too roughly’
(1.2.141–2).
Appendix 3: Literature Online Data for
Chapter 3
Results of searches of Arden of Faversham, 14.1–76 for phrases and collocations
occurring five or fewer times in plays of 1580–1600:
so long in killing (1): ‘so long in getting’, Greville, Alaham; ‘so long in passing’,
Nashe, Summer’s Last Will and Testament; the only instances of ‘so long in’ followed
by verbs.
killing a man (1): Heywood, The Four Prentices of London.
I think we shall never (2): Shakespeare, Henry V; ‘I think we shall’, Marston,
Antonio’s Revenge; Marston, Jack Drum’s Entertainment.
never do it (2): Anon., A Knack to Know a Knave; Anon., A Warning for Fair Women;
Anon., The Rare Triumphs of Love and Fortune.
give it over (2): Kyd, The Spanish Tragedy; Shakespeare, 1 Henry IV.
though we be hanged (3): ‘both hanged’, Greene, James IV.
at his door (4): Anon., A Knack to Know an Honest Man (in a stage direction);
Anon., Warning for Fair Women; Munday and Chettle, Sir Thomas More.
I have lived in (5): Porter, The Two Angry Women of Abingdon.
twelve years (6): Wilson, The Three Ladies of London.
for taking the wall (7): Jonson, Every Man in His Humour.
cracked . . . blades (8–9): ‘crack a blade’, Greene, Orlando Furioso; Peele, Edward I.
monstrous lie (10): Anon., Guy Earl of Warwick; Lyly, Endymion; Peele, Edward I.
paid me tribute (11–12): ‘pay me tribute’, Marlowe, 2 Tamburlaine; Shakespeare
and others, 2 Henry VI; ‘paid tribute’, Wilson, The Three Ladies of London; Hughes
and others, The Misfortunes of Arthur.
op’ning her shop windows (13): ‘open my shop windows’, Dekker, The Shoemakers’
Holiday; ‘shop window’, Dekker, The Shoemakers’ Holiday; the only mentions of
shop windows.
cross word (14): ‘cross words’, Shakespeare, Hamlet.
cross word of a tapster (14): ‘word of a tapster’, Shakespeare, As You Like It; ‘of a
tapster’, Anon., Every Woman in Her Humour; Shakespeare, Love’s Labour’s Lost.
with my dagger (15): Shakespeare and Peele, Titus Andronicus.
held him by the ears (15): ‘hold . . . by the ears’, Marlowe, Edward II.
in Thames Street (16): Jonson, Every Man in His Humour.
Appendix 3
241
run over me (17): Haughton, Englishmen for My Money.
I made no more ado but (17): ‘I made me no more ado but’, Shakespeare, 1 Henry IV.
beat them about his head (19): ‘He breaks it about his head’ (stage direction),
Greene, Orlando Furioso; the only instance of ‘about his head’ associated with
aggression.
the constable . . . his watch (20): Anon., Look about You.
broken a sergeant’s head (21): ‘a broken head’, Lyly, Midas.
with my sword and buckler (22–3): ‘my sword and buckler’, Porter, Two Angry
Women of Abingdon; ‘with his sword and buckler’, Anon., The Famous Victories of
Henry V; ‘his sword and buckler’ (in a Q 1600 stage direction), Shakespeare, 2
Henry IV.
tenpenny (23): Greene, James IV; included because a quasi-compound.
a quart pot (24): Nashe, Summer’s Last Will and Testament; Shakespeare and others,
2 Henry VI; Anon., Jack Straw; Greene, John of Bordeaux; ‘the quart pot’, Marston,
Jack Drum’s Entertainment.
in their hand (24): Anon., King Leir; Greene, Alphonsus, King of Aragon; Marston,
Histriomastix; Shakespeare, Henry V.
Will it please your worship (24–5): ‘Will it please your honour’, Anon., The
Troublesome Reign of King John; ‘Will it please your ladyship’, Jonson, Cynthia’s
Revels; ‘Will it please your grace’, Hamlet.
pulled down (26): Greene, Friar Bacon and Friar Bungay; Marlowe, Edward II.
the next night (27): Haughton, Grim the Collier; Shakespeare, Romeo and Juliet;
Shakespeare, The Merchant of Venice.
have I not done (27–8): Greville, Alaham; Anon., A Warning for Fair Women.
Hence . . . Here comes (30): Shakespeare, As You Like It; in each case dismissing
somebody as a new arrival approaches.
they both shook hands (32): ‘they shook hands’, Shakespeare, As You Like It; in
each case two men formerly at enmity are making peace.
railed on (34): Shakespeare, As You Like It.
was cause of all (34): Daniel, Cleopatra.
No sooner came (35): Greville, Mustapha.
in at doors (35): Marlowe, The Jew of Malta.
gave him money (36): ‘give him money’, Chettle and Munday, The Downfall of
Robert Earl of Huntingdon; ‘gives him money’ (stage direction), Anon., Captain
Thomas Stukeley.
bring you word (37): Anon., The True Tragedy of Richard III; Greene, Alphonsus,
King of Aragon; Jonson, A Tale of a Tub; Marston, Jack Drum’s Entertainment.
divers of his (39): Anon., Jack Straw.
his neighbours and his friends (39): ‘neighbours and friends’, Anon., Jack Straw;
Heywood, 1 Edward IV; Shakespeare, The Taming of the Shrew; ‘my fellow friends,
neighbours’, Greene, George a Greene; the only collocations of the plurals.
sup with you at our house this night (40): ‘you see my house, And sup with me this
night’, Heywood, 1 Edward IV; the only instances of ‘sup with’, ‘house’, and ‘this
night’.
242Appendix 3
Ah, gentle Michael, run thou back again (41): ‘Good Margaret, run thee to the parlor’, Shakespeare, Much Ado About Nothing; instructions in each case.
when my husband (42): Shakespeare, The Merry Wives of Windsor; ‘when my husband’s’, Anon., A Knack to Know an Honest Man.
Bid Mosby steal from him and come to me (43): ‘bid thy master rise and come to me’,
Shakespeare, Richard III; ‘and come to me’, Anon., The Troublesome Reign of King
John; although the Richard III parallel is the closer, there is no further example of
‘and come to me’.
And this night (44): Shakespeare, As You Like It; Jonson, Every Man in His Humour;
Anon., A Knack to Know an Honest Man.
this night shall (44): Chapman, The Blind Beggar of Alexandria; Heywood, The
Four Prentices of London; Kyd, The Spanish Tragedy; Shakespeare, The Merry Wives
of Windsor.
made sure (44): Marlowe, The Jew of Malta; Heywood, The Four Prentices of
London.
I’ll go tell him (45): Chettle, Day, and Haughton, The Blind Beggar of Bednal Green;
‘Go tell him’, Anon., Lust’s Dominion; ‘I will tell him’, Shakespeare, A Midsummer
Night’s Dream; ‘go tell him’, Shakespeare, Romeo and Juliet.
as thou goest (46): Heywood, 1 Edward IV; Shakespeare, The Merchant of Venice.
our guests (46): Chettle and Munday, The Downfall of Robert Early of Huntingdon;
Heywood, 1 Edward IV; Chettle and Munday, Sir Thomas More (twice).
spare for no cost (47): Greene, Orlando Furioso; Jonson, A Tale of a Tub; Marlowe,
Edward II; Shakespeare, Hamlet.
such cheer (48): Anon., The Maid’s Metamorphosis; Anon., The Taming of a Shrew;
Chettle and Munday, Sir Thomas More; Peele, The Arraignment of Paris.
I do mean to (48): Greene, Selimus; Porter, The Two Angry Women of Abingdon;
Shakespeare, The Merry Wives of Windsor; Shakespeare, Romeo and Juliet.
And welcome shall you be (50): Haughton, Grim the Collier; ‘and . . . welcome shall
you be’, Shakespeare, As You Like It; ‘welcome shall you be’, Anon., A Knack to
Know a Knave.
Ah, gentlemen (50): Haughton, Englishmen for My Money; Marlowe, Doctor
Faustus (3 times).
How missed you of your purpose (51): ‘I missed my purpose’, Jonson, Every Man out
of His Humour; ‘miss of ’ (meaning ‘miss’), Greene, Selimus; Haughton, Englishmen
for My Money; Lyly, Love’s Metamorphosis.
’Twas long of (meaning ‘on account of ’) (52): Chettle and Munday, The Downfall
of Robert Earl of Huntingdon.
that unlucky (52): Heywood, 2 Edward IV; followed by a noun.
Thou dost me wrong (53): Heywood, 2 Edward IV; Peele, The Arraignment of Paris;
Shakespeare, Julius Caesar (twice).
as much as any (53): Anon., Edmond Ironside.
I’ll tell you how it was (54): Haughton, Englishmen for My Money; ‘I’ll tell you how’,
Chapman, The Blind Beggar of Alexandria; Marlowe, Doctor Faustus; Shakespeare,
The Merry Wives of Windsor; Shakespeare and others, 3 Henry VI.
Appendix 3
243
When he should have (55): Anon., Caesar and Pompey.
over his head (56): Anon., The Entertainment at Mitcham; Peele, David and Bethsabe.
comes . . . at him (56): ‘come at him’, Chapman, The Blind Beggar of Alexandria.
slinks away (57): ‘slink away’, Shakespeare, The Merchant of Venice.
hand and feet (58): Jonson, A Tale of a Tub; Marlowe, 2 Tamburlaine; Marlowe and
Nashe, Dido Queen of Carthage; the list in Jonson’s Every Man in His Humour, including ‘eye’ and with the singular ‘foot’ (‘eye, hand, and foot’), has been ignored.
one and two (58–9): Chettle, Day, and Haughton, The Blind Beggar of Bednal
Green; Jonson, Every Man in His Humour.
his costard (59): Haughton, Englishmen for My Money.
his sword-point (59–60): Anon., Edmond Ironside.
half a yard (60): Haughton, Englishmen for My Money; Lyly, Midas.
out of danger (60): Greville, Mustapha; Jonson, A Tale of a Tub.
If the devil come (61): Shakespeare, 1 Henry IV.
have no more strength than (61–2): Shakespeare, A Midsummer Night’s Dream.
I’ll stand to it (63): Anon., A Knack to Know an Honest Man; Greene, Friar Bacon
and Friar Bungay; Greene and Lodge, A Looking Glass for London; Porter, The Two
Angry Women of Abingdon; Shakespeare, As You Like It.
as good as a castle . . . better than a sconce (63–4): ‘as good as a sconce’, Wilson, The
Cobbler’s Prophecy; a sconce is a small fort.
began to faint (65): Kyd, Soliman and Perseda.
arming-sword (66): Hughes and others, The Misfortunes of Arthur.
I wonder why (68): Chettle and Munday, The Downfall of Robert Earl of Huntingdon;
Lodge, The Wounds of Civil War; Marlowe, Edward II; Marston, Jack Drum’s
Entertainment; Peele, Edward I.
stood still (68): Daniel, Cleopatra; Greville, Mustapha; Haughton, Englishmen for
My Money; Chettle and Munday, Sir Thomas More; Porter, Two Angry Women of
Abingdon.
Faith, I was (69): Anon., Love and Fortune; Porter, Two Angry Women of Abingdon;
‘In good faith, I was’, Porter, Two Angry Women of Abingdon; Jonson, Every Man in
His Humour.
I was so amazed (69): Chettle, Day, and Haughton, The Blind Beggar of Bednal
Green; ‘so amazed’, Greene, Friar Bacon and Friar Bungay; Newton, Thebais.
Ah, sirs (70): Lyly, Campaspe; Shakespeare, Richard III.
had he . . . been slain (70): ‘had he been slain’, Shakespeare and others, 3 Henry VI.
For every drop of . . . blood (71): ‘for every drop of blood’, Heywood, 1 Fair Maid
of the West; Lodge, The Wounds of Civil War; Shakespeare and others, 1 Henry VI.
crammed in angels (72): ‘cram wealth in’, Dekker, The Shoemakers’ Holiday;
‘cramm’d in’, Shakespeare, The Merry Wives of Windsor; ‘cramm’d up in’,
Shakespeare, Love’s Labour’s Lost; ‘cram . . . in’, Newton, Thebais; Arden Q reads
‘cramme’; ‘angels’ are here coins.
hugged thee (73): ‘hug thee’, Anon., Look about You; Chettle, Day, and Haughton,
The Blind Beggar of Bednal Green (twice); Haughton, Englishmen for My Money.
244Appendix 3
hugged . . . in my arms (73): ‘hug . . . in my arms’, Anon., A Knack to Know an
Honest Man.
Patient yourself (74): Shakespeare and Peele, Titus Andronicus.
cannot help it (74): Anon., Captain Thomas Stukeley; Heywood, 1 Edward IV
(twice); Jonson, Every Man in His Humour.
dog him (75): Anon., A Warning for Fair Women; Warner, Menaechmi; Wilson, The
Three Lords and Three Ladies of London.
Appendix 4: Literature Online Data for
Chapter 6
part a
Results of control tests in which Sonnets 109 and 110 are searched for phrases and
collocations occurring five or fewer times in drama first performed 1590–1610:
SONN ET 109
O, never say (1): Shakespeare, Cymbeline.
false of heart (1): Shakespeare, King Lear; Shakespeare, The Winter’s Tale.
my soul which in thy breast doth lie (4): ‘my heart is in thy breast’, Shakespeare,
Love’s Labour’s Lost; ‘my heart . . . in thy breast’, Kyd, Soliman and Perseda; the only
examples of ‘my’ followed by a noun and ‘in thy breast’.
Like him that travels I return again (5): ‘my return from travel’, Armin, The
Two Maids of More-Clacke; ‘returned from travel’, Beaumont and Fletcher, The
Coxcomb; ‘travels upon your return’, Mason, The Turk; ‘travel . . . or return again’,
Shakespeare, All’s Well That Ends Well; the Shakespeare link is the closest, because
‘travel’ is a verb and it has ‘return again’ in common with the sonnet; but all four
citations qualify as rare in their juxtaposition of returning and travelling.
Never believe (9): ‘but never believe it’, Beaumont and Fletcher, The Woman
Hater; ‘never believe their protestations’, Daniel, The Queen’s Arcadia; ‘Never
believe it’, Shakespeare, Hamlet; ‘never believe me’, Shakespeare, Richard II; the
only instances where ‘believe’ is imperative.
in my nature (9): Shakespeare, Twelfth Night.
in my nature reign’d | All frailties that besiege all kinds of blood (9–10): ‘like a fever
she Reigns in my blood’, Shakespeare, Love’s Labour’s Lost; ‘a fury reigns Over my
boiling blood’, Chettle, Hoffman; ‘the cold that besiegeth our wam blood’, Nashe,
Summer’s Last Will and Testament.
nature . . . be stain’d (9–11): ‘hath stained nature’, Greville, Mustapha; Barnes’s
‘Stained with the guilt of nature’ in The Devil’s Charter is excluded because ‘the
guilt of nature’ is the agent of the staining.
246Appendix 4
so preposterously (11): Shakespeare, Henry V; Shakespeare, Othello; the contexts
are all of shameful and unnatural behaviour, Scroop’s treason in Henry V and
Desdemona’s alleged ‘erring’ against ‘nature’ (as in line 9 of the sonnet) in Othello.
wide universe (13): ‘the wide vessel of the universe’, Shakespeare, Henry V.
my rose (14): Anon., Timon (University Play); Beaumont and Fletcher, Philaster;
in Dekker’s The Shoemaker’s Holiday, Rose is a proper name, so is discounted.
SONN ET 110
’tis true, I have gone (1): ‘’Tis true, I have heard’, William Rowley, A Shoemaker,
a Gentleman; ‘true, I have lost’, Shakespeare, As You Like It; ‘true I have married’,
Shakespeare, Othello (where the context is the same as that in which ‘so preposterously’ occurs).
to the view (2): Middleton, The Puritan; Shakespeare, Antony and Cleopatra;
Shakespeare, Hamlet.
Gor’d mine own thoughts (3): ‘gor’d the gentle bosom of the peace’, Shakespeare,
Henry V; ‘My fame is shrowdly gor’d’, Shakespeare, Troilus and Cressida; ‘the gor’d
state’, Shakespeare, King Lear; these parallels are, somewhat doubtfully, included,
because the single word ‘gor’d’ is used in each case in a context in which the sense
is figurative: in the Troilus and Cressida instance, as in the sonnet, shame is involved.
mine own thoughts (3): Daniel, Philotas; Shakespeare, The Winter’s Tale; Wilkins,
The Miseries of Enforced Marriage.
is most dear (3): Jonson, Every Man In His Humour; Dekker and Midleton, 1
Honest Whore.
old offenses (4): Tomkis, Lingua; ‘old offence’, Lyly, The Woman in the Moon.
affections new (4): Fletcher, The Faithful Shepherdess; ‘new affection’, Anon., A
Warning for Fair Women.
Most true it is (5): ‘it is most true’, Chapman, Sir Giles Goosecap; Shakespeare,
Othello (still in the same context as the Othello links to 109.11 and 110.1).
look’d on truth (5): ‘look on truth’, Jonson, Cynthia’s Revels.
look’d . . . askance (5–6): ‘look . . . askance’, Chapman, The Gentleman Usher;
Marston, Histriomastix; Jonson, Every Man in His Humour; ‘look askance’,
Shakespeare, The Taming of the Shrew.
look’d on truth . . . strangely (5–6): ‘look strangely on’, Shakespeare, 2 Henry IV;
Heywood, The Wise Woman of Hogsdon; ‘look not so strangely upon’, Tomkis,
Lingua; ‘look strangely upon’, Anon., The Maid’s Metamorphosis (in a stage direction, so doubtfully included).
gave my heart another youth (7): ‘gives my heart matter of joy’, Barry, The Family of
Love; ‘given my heart a winking’, Shakespeare, Hamlet; ‘so strong a passion As love
doth give my heart’, Shakespeare, Twelfth Night; in each case the heart is the recipient, in contrast to cases of giving one’s heart.
my best of love (8): ‘the best of love’, Middleton, Michaelmas Term; ‘your best of
love’, Barry, Ram Alley; ‘her best of love’, Armin, The Two Maids of More-Clacke;
‘my best love’, Beaumont and Fletcher, Philaster.
Appendix 4
247
have no end (9): Chapman, Byron’s Conspiracy.
Mine appetite (10): Chapman, A Humorous Day’s Mirth; Heywood, The Rape of
Lucrece; Marston, What You Will.
try an older friend (11): ‘try a friend’, Heywood and William Rowley, Fortune by
Land and Sea.
an older friend (11): ‘an old friend’, Anon., Captain Thomas Stukeley; Drayton,
Munday, Hathway, and Wilson, 1 Sir John Oldcastle; Dekker and Webster,
Northward Ho.
I am confin’d (12): Middleton, Your Five Gallants; Anon., Lust’s Dominion.
Even to thy (14): Heywood, The Royal King and the Loyal Subject; Greene, George
a Greene; Dekker and Middleton, 1 Honest Whore.
thy pure . . . breast (14): ‘thy pure bosom’, Shakespeare, The Two Gentlemen of
Verona.
most most loving (14): ‘most most dear’, Dekker, The Whore of Babylon; Dekker
and Middleton, 1 Honest Whore; ‘most most monstrous’, Marston, The Fawn; ‘most
most private’, Marston, Sophonisba; in these cases ‘most most’ is a double intensifier
of a following adjective; ‘most most unlooked for’ occurs in Shakespeare’s Hamlet
only in the ‘bad quarto’ of 1603 and so is ignored.
loving breast (14): Shakespeare and Peele, Titus Andronicus.
part b
Results when the first seven stanzas of A Lover’s Complaint (lines 1–49) are searched
for phrases and collocations used by Shakespeare but not John Davies of Hereford
and vice versa:
From off a hill whose concave womb (1): ‘a molehill from whose hollow womb’,
Davies, Commendatory Poems.
A hill whose concave womb reworded . . . this double voice (1–3): ‘earth . . . whose hollow womb resounds’, Shakespeare, Venus and Adonis; ‘Tiber trembled underneath
her banks To hear the replication of your sounds Made in her concave shores’,
Julius Caesar.
double voice (3): ‘double, like the voice and echo’, Shakespeare, 2 Henry IV; ‘a
voice . . . As double’, Othello.
to list, meaning ‘to listen to’ (4): Shakespeare, The Comedy of Errors.
list the sad-tun’d tale (4): ‘list a brief tale’, Shakespeare, King Lear.
the sad-tun’d tale (4): ‘sad tale’, Shakespeare, Richard II; ‘sad tales’, The Rape of
Lucrece’; ‘sad tale’s’, The Winter’s Tale; ‘saddest tale’, A Midsummer Night’s Dream;
Davies does not mention sad tales; ‘ill-tuned repetitions’, Shakespeare, King John;
‘care-tun’d tongue’, Richard II; ‘new-tun’d oaths’, Henry V; the tunings are of
speech, not music.
Ere long (5): Shakespeare, Sonnet 73, Coriolanus, 2 Henry IV, King John, King Lear,
Measure for Measure, A Midsummer Night’s Dream, The Tempest, Hamlet (twice).
Tearing of (6): Davies, The Holy Rood.
248Appendix 4
Tearing of papers (6): ‘I do tear his paper’, Shakespeare, The Two Gentlemen of
Verona; this is much closer to the Complaint than the previous item, since Silvia,
like the maid in the poem, is tearing the letter of a false suitor.
breaking rings (6): ‘cut the wedding-ring, And break it’, Shakespeare, The Comedy
of Errors; an instance of breaking a ring in Davies’s The Muse’s Sacrifice relates to
breaking out of a circle of men.
breaking rings a-twain (6): ‘bite the holy cords a-twain’, King Lear; the rings are
symbols of love in the poem, while the reference is to marriage bonds in the play.
Storming her world with sorrow’s wind and rain (7): ‘Sorrow . . . But like a stormy
day, now wind, now rain, Sighs dry her cheeks, tears make them wet again’,
Shakespeare, Venus and Adonis; ‘Strives in his little world of man to out-storm
The . . . wind and rain’, King Lear (adopting, like most editors except Evans in The
Riverside Shakespeare, the emendation ‘out-storm’ for Q’s ‘outscorne’); ‘puffing
with wind and rain’, As You Like It; ‘roaring wind and rain’, King Lear.
Upon her head (8): Davies, Commendatory Poems, Humour’s Heaven, Microcosmos.
might think (10): Shakespeare, Sonnet 138.
beauty spent and done (11): ‘decay’d and done’, Shakespeare, The Rape of Lucrece
(in a context in which the word ‘beauty’ is repeated and ‘beauteous’ also occurs);
‘beauty . . . wasted, thaw’d, and done’, Venus and Adonis; ‘My inch of taper will be
burnt and done’, Richard II (Richard’s ‘inch of taper’ is his life); Davies has no similar ‘and done’ collocations.
fell rage (13): Davies, The Holy Rood.
Some beauty peep’d through lettice of sear’d age (14): ‘So thou through windows of
thine age shalt see Despite of wrinkles this thy golden time’, Shakespeare, Sonnet
3: ‘My way of life Is fall’n into the sear, the yellow leaf . . . old age’, Macbeth; the
Riverside edition retains the original spelling ‘lettice’ for ‘lattice’.
napkin . . . to her eyne . . . tears (15–18): ‘Thy napkin cannot drink a tear’, ‘His napkin, with his true tears all bewet’, Shakespeare and Peele, Titus Andronicus; ‘to dry
his cheeks, A napkin’, Shakespeare and others, 3 Henry VI; instances of a napkin
wiping a brow in Hamlet and Othello have been omitted as less closely parallel,
though Davies does not use the word ‘napkin’ at all.
reading what contents it bears (19): ‘they read . . . those contents’, Shakespeare,
King Lear.
clamors of all size, both high and low (21): ‘songs of all sizes’, Shakespeare, The
Winter’s Tale; ‘sing both high and low’, Twelfth Night: the only instances of ‘size’
and ‘both high and low’ connected with sound; ‘both high and low’, The Merry
Wives of Windsor; in this play the phrase is not used of sound, but the four-word
sequence is not used by Davies, though he uses ‘high and low’.
Sometime . . . anon (24–6): Shakespeare, Venus and Adonis (twice); Romeo and
Juliet.
orbed earth (25): ‘orbed ground’, Shakespeare, Hamlet.
hair . . . | Hanging . . . | braided (29–35): ‘braided hanging mane’, Shakespeare,
Venus and Adonis.
Appendix 4
249
her pale and pined cheek (32): ‘her pale cheek’, ‘her cheek was pale’, ‘his pale
cheeks’, Shakespeare, Venus and Adonis; ‘cheeks neither red not pale’, The Rape of
Lucrece; ‘Why is your cheek so pale?’, A Midsummer Night’s Dream; ‘his cheek look’d
pale’, his cheek looks pale, 2 Henry IV; ‘Calphurnia’s cheek is pale’, Julius Caesar;
‘Make pale our cheeks’, Richard II; Shakespeare and others, ‘your cheeks . . . pale
they look’, 1 Henry VI.
still did bide (33): ‘bides still’, Shakespeare, The Comedy of Errors.
break from thence (34): ‘break from hence’, Shakespeare and others, 3 Henry VI;
‘break from’, King Lear, The Winter’s Tale; Shakespeare and Fletcher, The Two
Noble Kinsmen.
crystal, and . . . jet (37): ‘crystal knots in mould of jet’, Davies, Microcosmos.
one by one (38): Shakespeare, Venus and Adonis, 2 Henry IV, Julius Caesar, Troilus
and Cressida, The Winter’s Tale.
a river . . . | whose weeping margent: (38–9): ‘the beached margent of the sea’,
Shakespeare, A Midsummer Night’s Dream; ‘o’er the sea margent’, Shakespeare and
Wilkins, Pericles (adopting the almost universally accepted emendation of Q’s
nonsensical phrase, retained in the Riverside edition, ‘ere the sea marre it’); Davies
never applies the word ‘margent’ (margin) to the verge of a body of water.
monarch’s hands that lets not bounty fall | Where want cries some, but where excess
begs all (41–2): ‘So distribution should undo excess, And each man have enough’,
Shakespeare, King Lear.
many a one (43): Shakespeare, Coriolanus.
Crack’d many a ring (45): ‘Cut the wedding-ring, And break it’, Shakespeare, The
Comedy of Errors (as in item in line 6).
a ring of posied gold and bone (45): Shakespeare, ‘the posy of a ring’, Hamlet; ‘a
hoop of gold, a paltry ring . . . whose posy’, The Merchant of Venice; ‘posied’ means
inscribed with a posy or motto, as are the rings in the two plays.
sleided silk (48): Shakespeare and Wilkins, Pericles.
Appendix 5: Control Test for Chapter 7
In case Shakespeare’s texts were, for some mysterious reason, such exceptional
sources of rare spellings as to furnish a disproportionate number of links to early
modern poems of anybody’s authorship, I tested the first 329 lines of Davies’s
Humour’s Heaven on Earth (1609) in exactly the same way as I had tested A Lover’s
Complaint. The following spellings proved rare, but present at least once, in LION’s
drama and poetry databases (with those found in only one or two writers preceded
by an asterisk): 4 *reedeme (redeem); 30, 41, 49 *casd (cased); 37 elles (ells); 39 bumme
(bum); 40 sloppes (slops); 40 *trusst (trussed); 43 clogges sb. (clogs); 45 soales (soles);
48 cloakt (cloaked); 66 taffataes (taffetas); 70 *shuttes (shuts); 103 pinckt (pinked); 120
*poiz’d (poised); 125 *perbrake (perbreak); 134 *turkies (turkeys); 156 giu’s (gives);
171 *hie-way (highway); 176 distasts (distastes); 208 mesure (measure); 226 imbrodred
(embroidered); 240, 248, 256 *formositie (formosity); 242 vnderpropt (underpropped);
249 stubberne (stubborn); 296 inspite (in spite); 323 *artezan (artisan).
A Lover’s Complaint had twenty-eight links to Shakespeare’s works, three to
Davies. Humour’s Heaven, 1–329, has ten links to Davies works, six to Shakespeare’s: Shakespeare links
A Lover’s Complaint:28
Humour’s Heaven, 1–329:
6
Davies links
3
10
The only plays with two or more links to Humour’s Heaven, 1–329, are Marston’s
What You Will (1607) with three, and Dekker’s The Shoemaker’s Holiday (1620) and
Campion’s The Mask at Lord Hay’s Wedding (1607), each with two. I have regarded
links to other components of the volume Humour’s Heaven on Earth (‘The Second
Tale’ and ‘The Triumph of Death’) as counting as links to works by Davies, just
as Complaint links to the sonnets were counted. The Davies links are: bumme: The
Scourge of Folly (1611) twice, Commendatory Poems (ed. 1878); cloakt: Humour’s Heaven
(1609) (‘Triumph of Death’), Microcosmos (1603); perbrake: The Scourge of Folly (1611);
hie-way: Humour’s Heaven (1609) (‘Triumph of Death’); distasts: The Scourge of Folly
(1611); mesure: Humour’s Heaven (1609) (‘Second Tale’); formositie: The Holy Rood
(1609), The Muse’s Sacrifice (1612) twice. Because it has the apostrophe, I discounted
as modern the metrical spelling lab’ring, though it occurs in only three texts,
including Davies’s The Holy Rood; also beheau’n (272), which is a modern metrical
Appendix 5
251
spelling once changes in u/v conventions are taken into account: but it occurs only
in Davies’s Microcosmos (1602) and Mirum in Modum (1602), while the full form,
beheauen, is found only in Davies’s Bien Venu (1606).
Of the spellings occurring in no more than two writers, Davies has perbrake,
hie-way, and formositie (in two works). Gordon has shuttes and casd, Marston has casd
and trusst, and Shakespeare has poiz’d (Troilus and Cressida, 1623) and turkies (1 Henry
IV, 1598). The two that appear in Shakespearean plays turn out to be very common
indeed in the LION database as a whole, most examples tending to appear later in
the seventeenth century and in the eighteenth. In contrast Davies’s hyphenated
hie-way occurs in only two other works, a poem by Colvil and a play by Drue;
formositie is used by only five other writers; and perbrake by five, from 1553 to 1592.
This spelling evidence for Davies’s authorship of Humour’s Heaven on Earth,
1–329, is less overwhelming than the spelling evidence for Shakespeare’s authorship of A Lover’s Complaint, but it certainly makes a striking contrast. Shakespeare’s
six links out of seventy-five afforded by drama and poetry combined match precisely his 8 per cent of the amount of text searched.
Davies was evidently a more orthodox and consistent speller than Shakespeare,
and the lexical element seems even more influential on the Humour’s Heaven results
than on the Complaint results. Perbrake and formositie, though qualifying only by the
barest of margins as non-modern spellings, are typical Davies exoticisms. They
are supported by items excluded from the rare spellings, such as beheau’n: LION
detects no instances of this or of beheauen (even with v replacing u) beyond the four
by Davies. The Humour’s Heaven form brast (burst), though not qualifying as rare,
occurs in two further works by Davies, but never in Shakespeare. Similarly, the
variant ware for ‘wore’ (Humour’s Heaven, 19, 25, 58) appears a further fifteen times
within Davies’s poetry, but is never used by Shakespeare, the one instance in his
canon falling within Peele’s share of Titus Andronicus (1594, 1.1.6). Straightforward
spellings (not odd verbal forms) that are not rare would also point us to Davies
rather than Shakespeare if we were in any doubt over which of the two wrote
Humour’s Heaven, 1–329: sutable (113) and anoy (181), for example, are common in
Davies, but are absent from Shakespearean texts. And, were we still in doubt, the
four instances of sith, three of pow’r, two of heau’n, and one of it’s would be decisive.
But the most significant result of this control test is the negative one: a poetic
text by Davies does not produce a disproportionate number of rare spellings and
variant forms shared with Shakespearean texts. A Lover’s Complaint does, and to a
remarkable degree.
Bibliography
Allen, Percy, Shakespeare, Jonson, and Wilkins as Borrowers (London: C. Palmer,
1928).
Anon., The Lamentable and True Tragedy of M. Arden of Feversham (Menston: Scolar
Press, 1971).
Austin, Alfred, Savonarola, 2nd edn. (London: Macmillan, 1891).
Barker, Simon and Hilary Hinds, eds., The Routledge Anthology of Renaissance
Drama (London: Routledge, 2003).
Bartlett, John, A Complete Concordance or Verbal Index to Words, Phrases . . . with a
Supplementary Concordance to the Poems (London: Macmillan, 1894).
Bate, Jonathan, The Genius of Shakespeare (London: Picador-Macmillan, 1997).
Bayne, Ronald, ed., Arden of Feversham (London: Dent, 1897; repr. 1955).
Bell, Ilona, ‘ “That which thou hast done”: Shakespeare’s Sonnets and A Lover’s
Complaint’, in Shakespeare’s Sonnets: Critical Essays, ed. James Schiffer (New York
and London: Garland, 1999), 455–74.
Belsey, Catherine, ‘Alice Arden’s Crime’, Renaissance Drama, n.s. 13 (1982), 83–102.
——The Subject of Tragedy (London: Methuen, 1985).
Blayney, Peter, The Texts of ‘King Lear’ and their Origins, vol. 1, Nicholas Okes and
the First Quarto (Cambridge: CUP, 1982).
Boyd, Brian, Why Lyrics Last: Evolution, Cognition, and Shakespeare’s Sonnets
(Cambridge, MA: Harvard UP, 2012).
Burrows, John, ‘Questions of Authorship: Attribution and Beyond’, Computers
and the Humanities, 37 (2003), 5–32.
——‘A Second Opinion on “Shakespeare and Authorship Studies in the TwentyFirst Century” ’, Shakespeare Quarterly, 63 (2012), 355–92.
Cairncross, A. S., ‘Pembroke’s Men and Some Shakespearian Piracies’, Shakespeare
Quarterly, 11 (1960), 335–49.
Carroll, Jayne M., ‘The Mystery of Arden of Faversham: Investigating the Play’s
Authorship through Elements of Language and Imagery’, MA thesis, University
of Auckland, 2002.
Carroll, Jayne M. and MacDonald P. Jackson, ‘Shakespeare, Arden of Faversham,
and “Literature Online” ’, Shakespeare Newsletter, 54 (2004), 3–4, 6.
Chambers, E. K. The Elizabethan Stage, 4 vols. (Oxford: Clarendon Press, 1923).
Chatterley, Albert, ‘Thomas Watson’, Oxford Dictionary of National Biography,
online edition.
Bibliography
253
Clemen, Wolfgang H., The Development of Shakespeare’s Imagery (London: Methuen,
1951).
Cockerell, Sydney, Letter to The Times, 2 October 1952, 7.
Collins, Stephen, Letter to The Times Literary Supplement, 31 August 2012, 6.
Craig, Hugh, ‘George Chapman, John Davies of Hereford, William Shakespeare,
and A Lover’s Complaint’, Shakespeare Quarterly, 63 (2012), 147–74.
Craig, Hugh and John Burrows, ‘A Collaboration about a Collaboration: The
Authorship of King Henry VI, Part Three’, in Collaborative Research in the Digital
Humanities, ed. Marilyn Deegan and Willard McCarty (Farnham: Ashgate,
2012), 27–65.
Craig, Hugh and Arthur F. Kinney, eds., Shakespeare, Computers, and the Mystery of
Authorship (Cambridge: CUP, 2009).
Craik, Katherine A., ‘Shakespeare’s A Lover’s Complaint and Early Modern
Criminal Confession’, Shakespeare Quarterly, 53 (2002), 437–59.
Craik, T. W., ed., Minor Elizabethan Tragedies (London: Dent, 1974).
Crane, Mary Thomas, Shakespeare’s Brain: Reading with Cognitive Theory (Princeton:
Princeton UP, 2001).
Crawford, Charles, ed., England’s Parnassus (Oxford: Clarendon Press, 1913).
Dessen, Alan C. and Leslie Thomson, A Dictionary of Stage Directions in English
Drama, 1580–1642 (Cambridge: CUP, 1999; repr. 2000).
Doran, Madeleine, Endeavors of Art: A Study of Form in Elizabethan Drama
(Madison: University of Wisconsin Press, 1954; repr. 1964).
Early English Books Online (EEBO) [website], http://eebo.chadwyck.com (accessed
26 June 2013).
Egan, Gabriel, ‘Impalpable Hits: Indeterminacy in the Searching of Tagged
Shakespearian Texts’. Paper delivered at the 33rd Annual Meeting of the
Shakespeare Association of America, Bermuda, 17–19 March 2005. Available
online at http://gabrielegan.com/publications/Egan2005a.htm (accessed 26
June 2013).
Elliott, Ward E. Y., ‘Notes from the Claremont Shakespeare Clinic’, Shakespeare
Newsletter, 61 (2011/12), 105, 108–12.
Elliott, Ward, E. Y. and Robert J. Valenza, ‘And Then There Were None:
Winnowing the Shakespeare Claimants’, Computers and the Humanities, 30
(1996), 191–245.
——‘Glass Slippers and Seven-Leagued Boots: C-prompted Doubt about
Ascribing A Funeral Elegy and A Lover’s Complaint to Shakespeare’, Shakespeare
Quarterly, 48 (1997), 177–207.
——‘The Professor Doth Protest Too Much, Methinks: Problems with the Foster
“Response” ’, Computers and the Humanities, 32 (1998), 425–88.
——‘Did Shakespeare Write A Lover’s Complaint?: The Jackson Ascription
Revisited’, in Words That Count: Early Modern Authorship: Essays in Honor of
MacDonald P. Jackson, ed. Brian Boyd (Newark: University of Delaware Press,
2004), 117–40.
254Bibliography
——‘Oxford by the Numbers: What Are the Odds That the Earl of Oxford Could
Have Written Shakespeare’s Poems and Plays?’, Tennessee Law Review, 72 (2004),
323–453.
——‘Two Tough Nuts to Crack: Did Shakespeare Write the “Shakespeare”
Portions of Sir Thomas More and Edward III? Part I and Part II: Conclusion’,
Literary and Linguistic Computing, 25 (2010), 67–83 and 165–77.
Erne, Lukas, Beyond ‘The Spanish Tragedy’: A Study of the Works of Thomas Kyd
(Manchester and New York: Manchester UP).
Everitt, E. B., The Young Shakespeare: Studies in Documentary Evidence: Anglistica 2
(Copenhagen: Rosenkilde and Bagger, 1954).
Farmer, Alan B. and Zachary Lesser, ‘Vile Arts: The Marketing of English Printed
Drama, 1512–1660’, Research Opportunities in Renaissance Drama, 39 (2000),
77–165.
‘Fisher’s Exact Test Calculator for 2x2 Contingency Tables’ [computer software],
http://research.microsoft.com/en-us/um/redmond/projects/mscompbio/
fisherexacttest/ (accessed 26 June 2013).
Fleissner, Robert F., ‘ “The Secret’st Man of Blood”: Foreshadowings of Macbeth
in Arden of Faversham’, University of Dayton Review, 14 (1979–80), 7–13.
Fleming, Juliet, Letters to The Times Literary Supplement, 10 and 24 August 2012, 6.
Foster, Donald W., Elegy by W. S.: A Study in Attribution (Newark: University of
Delaware Press, 1989).
Fraser, Russell A. and Norman Rabkin, eds., Drama of the English Renaissance
I: The Tudor Period (New York: Macmillan, 1976).
Freeman, Arthur, Thomas Kyd: Facts and Problems (Oxford: Clarendon Press, 1967).
——‘The Writings of Thomas Achelley’, The Library, 25 (1970), 40–2.
Garner, Bryan, ‘Shakespeare’s Latinate Neologisms’, Shakespeare Studies, 15 (1982),
149–70; repr. in A Reader in the Language of Shakespearean Drama, ed. Vivian
Salmon and Edwina Burness (Amsterdam and Philadelphia: John Benjamins,
1987), 207–28.
Go, Kenji, ‘Samuel Daniel’s The Complaint of Rosamond and an Emblematic
Reconsideration of A Lover’s Complaint’, Studies in Philology, 104 (2007), 82–122.
Greenblatt, Stephen, Will in the World: How Shakespeare Became Shakespeare (New
York: Norton, 2004).
Greene, Robert, Friar Bacon and Friar Bungay, ed. Daniel Seltzer. Regents
Renaissance Drama (Lincoln, NE: University of Nebraska Press, 1963).
——The Scottish History of James the Fourth, ed. Norman Sanders (London:
Methuen, 1970).
Greg, W. W., ‘Authorship Attributions in the Early Play-Lists, 1656–1671’,
Edinburgh Bibliographical Society Transactions, 2 (1938–45), 305–29.
——‘Shakespeare and Arden of Feversham’, Review of English Studies, 21 (1945),
134–6.
——The Shakespeare First Folio: Its Bibliographical and Textual History (Oxford:
Clarendon Press, 1955).
Bibliography
255
Gurr, Andrew, The Shakespearian Playing Companies (Oxford: Clarendon Press,
1996; repr. 2001).
Hamilton, Charles, Cardenio, or, The Second Maiden’s Tragedy (Lakewood, CO:
Glenbridge, 1994).
Harbage, Alfred, rev. Samuel Schoenbaum and S. Waggonheim, Annals of English
Drama 975–1700 (London and New York: Methuen, 1989).
Hart, Alfred, ‘The Number of Lines in Shakespeare’s Plays’, Review of English
Studies, 8 (1932), 19–28.
——Shakespeare and the Homilies (Melbourne and London: Melbourne UP and
OUP, 1934).
——‘Shakespeare and the Vocabulary of The Two Noble Kinsmen’, Review of English
Studies, 10 (1934), 274–87.
——Stolne and Surreptitious Copies: A Comparative Study of Shakespeare’s Bad Quartos
(Melbourne: Melbourne UP, 1942).
——‘Vocabularies of Shakespeare’s Plays’ and ‘The Growth of Shakespeare’s
Vocabulary’, Review of English Studies, 19 (1943), 128–40 and 242–54.
Headlam, W., ‘Arden of Feversham: “Pathaires” ’, Athenaeum, 26 December 1903.
Healy, Margaret, Shakespeare, Alchemy and the Creative Imagination: The Sonnets and
‘A Lover’s Complaint’ (Cambridge: CUP, 2010).
Here’s History Kent [website] http://www.hereshistorykent.org.uk (accessed 26
June 2013).
Heywood, Thomas, The Captives, ed. Arthur Brown, Malone Society Reprint
(Oxford: OUP, 1953).
——The Escapes of Jupiter, ed. Henry D. Janzen, Malone Society Reprint
(Oxford: OUP, 1978).
Hieatt, A. K., T. G. Bishop, and E. A. Nicholson, ‘Shakespeare’s Rare Words:
“Lover’s Complaint”, Cymbeline, and Sonnets’, Notes and Queries, 232 (1987),
219–24.
Holdsworth, R. V., ‘Middleton and Shakespeare: The Case for Middleton’s Hand
in Timon of Athens’, PhD thesis, University of Manchester, 1982.
Hopkinson, A. F., ed., Arden of Faversham (London: privately printed, 1890); rev. in
Shakespeare’s Doubtful Plays, 1898, 1907.
Hoover, David, ‘The Rarer They Are, the More There Are, the Less They Matter’.
Paper delivered at the Digital Humanities conference, Hamburg, 16–20 July 2012.
Availableonlineathttp://www.dh2012.uni-hamburg.de/conference/programme/
abstracts/the-rarer-they-are-the-more-there-are-the-less-they-matter/
(accessed 26 June 2013).
Hyde, Patricia, Thomas Arden in Faversham: The Man Behind the Myth (Faversham:
The Faversham Society, 1996).
Jackson, MacDonald P., ‘Material for an Edition of Arden of Feversham’, BLitt thesis, Oxford University, 1963.
——Shakespeare’s ‘A Lover’s Complaint’: Its Date and Authenticity (Auckland:
University of Auckland, Bulletin 72 English Series 13, 1965).
256Bibliography
——‘Punctuation and the Compositors of Shakespeare’s Sonnets 1609’, The
Library, 5th series, 30 (1975), 1–24.
——‘Compositorial Practices in The Revenger’s Tragedy, 1607–08’, Papers of the
Bibliographical Society of America, 75 (1981), 157–70.
——‘John Webster and Thomas Heywood in Appius and Virginia: A Bibliographical
Approach to the Problem of Authorship’, Studies in Bibliography, 38 (1985),
217–35.
——‘The Additions to The Second Maiden’s Tragedy: Shakespeare or Middleton?’,
Shakespeare Quarterly, 41 (1990), 402–5.
——‘Shakespearean Features of the Poetic Style of Arden of Faversham’, Archiv für
das Studium der neueren Sprachen und Literaturen, 230 (1993), 279–304.
——‘Editing, Attribution Studies, and “Literature Online”: A New Resource in
Renaissance Drama’, Research Opportunities in Renaissance Drama, 37 (1998), 1–16.
——‘New Work on the Compositors of Shakespeare’s Sonnets (1609)’, Shakespeare
Newsletter, 48 (1998), 31–4.
——‘Aspects of Organisation in Shakespeare’s Sonnets’, Parergon, 17 (1999), 109–34.
——‘Rhymes in Shakespeare’s Sonnets: Evidence of Date of Composition’, Notes
and Queries, 244 (1999), 213–19.
——‘Shakespeare’s Sonnets: Rhyme and Reason in the Dark Lady Series’, Notes and
Queries, 244 (1999), 219–22.
——‘Late Webster and his Collaborators: How Many Playwrights Wrote A Cure
for a Cuckold?’, Papers of the Bibliographical Society of America, 95 (2001), 295–313.
——‘Vocabulary and Chronology: The Case of Shakespeare’s Sonnets’, Review of
English Studies, 52 (2001), 59–75.
——‘Dating Shakespeare’s Sonnets: Some Old Evidence Revisited’, Notes and
Queries, 247 (2002), 237–41.
——‘Determining Authorship: A New Technique’, Research Opportunities in
Renaissance Drama, 41 (2002), 1–14.
——‘The Distribution of Pronouns in Shakespeare’s Sonnets’, AUMLA: Journal of
the Australasian Universities Language and Literature Association, 97 (2002), 22–38.
——‘Shakespeare’s Richard II and the Anonymous Thomas of Woodstock’, Medieval
and Renaissance Drama in England, 14 (2002), 17–65.
——Defining Shakespeare: ‘Pericles’ as Test Case (Oxford: OUP, 2003).
——‘A Lover’s Complaint Revisited’, Shakespeare Studies, 32 (2004), 267–94.
——‘Anything for a Quiet Life, IV.ii.1–44’, Notes and Queries, 251 (2006), 86–9.
——‘Compound Adjectives in Arden of Faversham’, Notes and Queries, 250
(2006), 51–5.
——‘The Date and Authorship of Hand D’s Contribution to Sir Thomas
More: Evidence from “Literature Online”’, Shakespeare Survey, 59 (2006), 69–78.
——‘John Webster, James Shirley, and the Melbourne Manuscript’, Medieval and
Renaissance Drama in England, 19 (2006), 21–44.
——‘Shakespeare and the Quarrel Scene in Arden of Faversham’, Shakespeare
Quarterly, 57 (2006), 249–93.
Bibliography
257
——‘The Date and Authorship of Thomas of Woodstock: Evidence and Its
Interpretation’, Research Opportunities in Medieval and Renaissance Drama, 46
(2007), 67–100.
——‘Is “Hand D” of Sir Thomas More Shakespeare’s? Thomas Bayes and the
Elliott–Valenza Authorship Tests’, Early Modern Literary Studies, 12.3 (2007),
1.1–36 [online journal] http://purl.oclc.org/emls/12-3/jackbaye.htm (accessed
26 June 2013).
——Review of Brian Vickers, Shakespeare, ‘A Lover’s Complaint’, and John Davies of
Hereford, Review of English Studies, 58 (2007), 723–5.
——‘Shakespeare’s Sonnet cx1 and John Davies of Hereford’s Microcosmos (1603)’,
Modern Language Review, 102 (2007), 1–10.
——‘New Research on the Dramatic Canon of Thomas Kyd’, Research
Opportunities in Medieval and Renaissance Drama, 47 (2008), 107–27.
——‘Arcite’s Horsemanship: A Reading in The Two Noble Kinsmen, II.v.13’, Notes
and Queries, 254 (2009), 605–7.
——‘Shakespeare or Davies? A Clue to the Authorship of “A Lover’s Complaint”’,
Notes and Queries, 254 (2009), 62–3.
——‘Parallels and Poetry: Shakespeare, Kyd, and Arden of Faversham’, Medieval and
Renaissance Drama in England, 23 (2010), 17–33.
——‘Deciphering a Date and Determining a Date: Anthony Munday’s John a Kent
and John a Cumber and the Original Version of Sir Thomas More’, Early Modern
Literary Studies, 15 (2011), 2: 1–24 [online journal] http://extra.shu.ac.uk/
emls/15-3/jackdate.htm (accessed 26 June 2013).
——‘Collaboration’, in The Oxford Handbook of Shakespeare, ed. Arthur F. Kinney
(Oxford: OUP, 2012), 31–52.
——‘Looking for Shakespeare in Double Falsehood: Stylistic Evidence’, in The
Quest for ‘Cardenio’: Shakespeare, Fletcher, Cervantes, and the Lost Play, ed. David
Carnegie and Gary Taylor (Oxford: OUP, 2012), 133–61.
——‘Reviewing Authorship Studies of Shakespeare and his Contemporaries, and
the Case of Arden of Faversham’, Memoria di Shakespeare, 8: On Authorship, ed.
Rosy Colombo and Pamela Guardamagno (Rome: Bulzoni, 2012), 149–167.
——‘A Lover’s Complaint and the Claremont Shakespeare Clinic’, Early Modern
Literary Studies, 16.3 (2013), 1–12.
Jacob, Edward, The History of the Port and Town of Faversham (London, 1774).
Jowett, John, ‘Henry Chettle and the First Quarto of Romeo and Juliet’, Papers of the
Bibliographical Society of America, 92 (1998), 53–74.
——Review of Brian Vickers, Shakespeare, ‘A Lover’s Complaint’, and John Davies of
Hereford, Shakespeare Quarterly, 60 (2009), 493–7.
——‘Disintegration, 1924’, Shakespeare (forthcoming).
Juola, Patrick, ‘Authorship Attribution’, Foundations and Trends in Information
Retrieval, 1 (2008), 233–334.
Kerrigan, John, Motives of Woe: Shakespeare and the ‘Female Complaint’ (Oxford:
OUP, 1991).
258Bibliography
King, Ros, ‘Arden of Faversham: The Moral of History and the Thrill of Perfor
mance’, in The Oxford Handbook of Tudor Drama, ed. Thomas Betteridge and
Greg Walker (Oxford: OUP, 2012), 635–52.
Kinney, Arthur F., ed., An Anthology of Plays and Entertainments (Oxford: Blackwell,
1999).
Kökeritz, Helge, Shakespeare’s Pronunciation (New Haven: Yale UP, 1953).
Kyd, Thomas, The Works of Thomas Kyd, ed. Frederick S. Boas (Oxford: Clarendon
Press, 1901).
——The Spanish Tragedy, ed. Philip Edwards (London: Methuen, 1959).
Lake, David J., The Canon of Thomas Middleton’s Plays (Cambridge: CUP, 1975).
——‘Three Seventeenth-Century Revisions: Thomas of Woodstock, The Jew of
Malta, and Faustus B’, Notes and Queries, 228 (1983), 133–43.
Lambarde, William, A Perambulation of Kent (London, 1570; 2nd edn. 1596).
Langley, Russell, Practical Statistics Simply Explained (New York: Dover Publications, 1971).
Laws, Jennifer, ‘The Generic Complexities of A Lover’s Complaint and its
Relationship to the Sonnets in Shakespeare’s 1609 Volume’, AUMLA: Journal
of the Australasian Universities Language and Literature Association, 89 (1998), 79–97.
Levenson, Jill, ‘Anonymous Plays: Arden of Faversham’, in The Predecessors of
Shakespeare: A Survey and Bibliography of Recent Studies in English Renaissance
Drama, ed. Terence P. Logan and Denzell S. Smith (Lincoln, NE: University of
Nebraska Press, 1978), 240–52.
Lillo, George, The London Merchant, ed. William H. McBurney (Lincoln,
NE: University of Nebraska Press, 1965).
Literature Online (LION) [website], http://lion.chawyck.co.uk (accessed 26 June
2013).
Lodge, Thomas and Robert Greene, A Looking-Glass for London and England, ed.
W. W. Greg, Malone Society Reprint (Oxford: OUP, 1932).
Love, Harold, Attributing Authorship: An Introduction (Cambridge: CUP, 2002).
Lyly, John, The Plays of John Lyly, ed. Carter A. Daniel (Lewisberg, PA: Bucknell
UP; London and Toronto: Associated University Presses, 1988).
Macdonald, Hugh with D. Nichol Smith, eds., Arden of Feversham 1592, Malone
Society Reprint (Oxford: OUP, 1947 for 1940).
Mackail, J. W., ‘A Lover’s Complaint’, Essays and Studies, 3 (1912), 51–70.
McLuskie, Kathleen E. and David Bevington, eds., Plays on Women (Manchester:
Manchester UP, 1999).
Maguire, Laurie E., Shakespearean Suspect Texts: The ‘Bad’ Quartos and their Contexts
(Cambridge: CUP, 1996).
Mahood, M. M., Shakespeare’s Wordplay (London: Methuen, 1957).
Marlowe, Christopher. Doctor Faustus, ed. John D. Jump (London: Methuen, 1962).
——The Complete Plays, ed. J. B. Steane (Harmondsworth: Penguin, 1969; repr.
1975).
——Doctor Faustus A- and B-texts (1604, 1616), ed. David Bevington and Eric
Rasmussen (Manchester: Manchester UP, 1993).
Bibliography
259
——The Jew of Malta, ed. David Bevington (Manchester: Manchester UP, 1997).
Merriam, Thomas, ‘Possible Light on a Kyd Canon’, Notes and Queries, 240 (1995),
340–1.
——‘Marlowe and Nashe in Dido Queen of Carthage’, Notes and Queries, 245
(2000), 425–8.
——‘Untangling the Derivatives: Points for Clarification in the Findings of the
Shakespeare Clinic’, Literary and Linguistic Computing, 24 (2009), 403–16.
Middleton, Thomas, Thomas Middleton: The Collected Works, gen. eds. Gary Taylor
and John Lavagnino (Oxford: Clarendon Press, 2007).
Mincoff, Marco, Shakespeare: The First Steps (Sofia: Bulgarian Academy of
Sciences, 1976).
Mosteller, Frederick and David L. Wallace, Inference and Disputed Authorship: The
Federalist (Reading, MA: Addison-Wesley, 1964).
Muir, Kenneth, ‘ “A Lover’s Complaint”: A Reconsideration’, in Shakespeare
1564–1964: A Collection of Modern Essays by Various Hands, ed. Edward A. Bloom
(Providence, RI: Brown UP, 1964), 154–66.
——Shakespeare the Professional and Related Studies (London: Heinemann, 1973).
Murray, John J., ed., The Tragedye of Solyman and Perseda (New York: Garland, 1991).
Neill, Michael, ‘“This Gentle Gentleman”: Social Change and the Language
of Status in Arden of Faversham’, Medieval and Renaissance Drama in England, 10
(1998), 73–97.
——Putting History to the Question: Power, Politics, and Society in English Renaissance
Drama (New York: Columbia UP, 2000).
Nicoll, Charles, The Lodger: Shakespeare on Silver Street (London: Allen Lane, 2007).
Nigri, Lucia, ‘“Speaking the truth”: On the Power of Words in Arden of Faversham’,
in Memoria di Shakespeare, 8: On Authorship, ed. Rosy Colombo and Daniela
Guardamagna (Rome: Bulzoni, 2012), 373–84.
Nosworthy, J. M., ‘Notes on Henry Porter’, Modern Language Review, 35 (1940),
517–21.
——‘Henry Porter’, English, 6 (1942), 65–9.
O’Connor, Frank, Shakespeare’s Progress (New York: Collier Books, 1961).
Oliphant, E. H. C., ‘Problems of Authorship in Elizabethan Dramatic Literature’,
Modern Philology, 8 (1911), 411–59.
——‘Marlowe’s Hand in Arden of Feversham: A Problem for Critics’, New Criterion,
4 (1926), 76–93.
——ed., Elizabethan Dramatists Other than Shakespeare (New York: Prentice-Hall,
1931).
The Oxford English Dictionary, Compact Edition (Oxford: Clarendon Press, 1971).
Palfrey, Simon and Tiffany Stern, Shakespeare in Parts (Oxford: OUP, 2007).
Percy, William, William Percy’s ‘Mahomet and His Heaven’: A Critical Edition, ed.
Matthew Dimmock (Aldershot: Ashgate, 2006).
Porter, Henry, Henry Porter’s The Two Angry Women of Abingdon: A Critical Edition,
ed. Marianne Brish Evett (New York and London: Garland, 1980).
260Bibliography
Preacher, K. J., ‘Calculation for the chi-square test: An interactive calculation tool
for chi-square tests of goodness of fit and independence’ [computer software]
http://www.quantpsy.org/chisq/chisq.htm (accessed 26 June 2013).
Price, Hereward T., ‘The First Quarto of Titus Andronicus’, in English Institute
Essays 1947 (New York: Columbia UP, 1948; repr. New York: AMS Press, 1965).
Rasmussen, Eric, ‘Shakespeare’s Hand in The Second Maiden’s Tragedy’, Shakespeare
Quarterly, 40 (1989), 1–26.
Robertson, J. M., Shakespeare and Chapman (London: Unwin, 1917).
——The Problems of the Shakespeare Sonnets (London: Routledge, 1926).
Rollins, Hyder Edward, ed., A New Variorum of Shakespeare: The Poems (Philadelphia
and London: Lippincott, 1938).
Sarrazin, G., ‘Wortechos bei Shakespeare’, Shakespeare Jahrbuch, 33 (1897), 121–65;
and 34 (1898), 119–69.
Schäfer, Jürgen, Shakespeares Stil: Germanisches und romanisches Vokabular (Frankfurt:
Athenäum, 1973).
——Documentation in the OED: Shakespeare and Nashe as Test Cases (Oxford:
Clarendon Press, 1980).
——Early Modern English Lexicography, 2 vols. (Oxford: Clarendon Press, 1989).
Schmidt, Alexander, Shakespeare Lexicon, 2 vols., rev. Gregor Sarrazin (Berlin:
Georg Reimer, 1902).
Schoenbaum, S., Internal Evidence and Elizabethan Dramatic Authorship: An Essay in
Literary History and Method (Evanston, IL: Northwestern UP, 1966).
——William Shakespeare: A Compact Documentary Life (Oxford: Clarendon Press,
1977).
Schoenfeldt, Michael, ed., A Companion to Shakespeare’s Sonnets (Oxford: Blackwell,
2007).
‘Script Example: How Many Words Are Unique to Each Shakespeare Work?’
[webpage], http://wordhoard.northwestern.edu/userman/scripting-example.
html (accessed 26 June 2013).
Shakespeare, William, The Works of William Shakespeare, 9 vols., ed. William
George Clark, William Aldis Wright, and John Glover (Cambridge: Macmillan,
1863–6).
——The Second Part of King Henry VI, ed. John Dover Wilson, New Shakespeare
(Cambridge: CUP, 1952).
——The Second Part of King Henry the Sixth, ed. Andrew S. Cairncross, Arden
Shakespeare (London: Methuen, 1957).
——The Winter’s Tale, ed. J. H. P. Pafford, Arden Shakespeare (London: Methuen,
1963).
——The Third Part of King Henry the Sixth, ed. Andrew S. Cairncross, Arden
Shakespeare (London: Methuen, 1964).
——William Shakespeare: Sonnets 1609 (Menston: Scolar Press, 1968).
——A Midsummer Night’s Dream, ed. Harold F. Brooks, Arden Shakespeare
(London: Methuen, 1979).
Bibliography
261
——The Sonnets and A Lover’s Complaint, ed. John Kerrigan (Harmondsworth:
Penguin, 1986).
——William Shakespeare: The Complete Works: Original-Spelling Edition, gen. eds.
Stanley Wells and Gary Taylor (Oxford: Clarendon Press, 1986).
——The Merchant of Venice, ed. M. M. Mahood, New Cambridge Shakespeare
(Cambridge: CUP, 1987).
——The Poems, ed. John Rowe, New Cambridge Shakespeare (Cambridge:
CUP, 1992).
——The Riverside Shakespeare, 2nd edn., ed. G. Blakemore Evans et al. (Boston
and New York: Houghton Mifflin, 1997).
——Shakespeare’s Sonnets, ed. Katherine Duncan-Jones, Arden Shakespeare
(London: Nelson, 1997; rev. edn. London: Methuen Drama, 2010).
——King Edward III, ed. Giorgio Melchiori, New Cambridge Shakespeare
(Cambridge: CUP, 1998).
——The Tragedy of King Richard III, ed. John Jowett, Oxford Shakespeare (Oxford:
OUP, 2000).
——The Complete Sonnets and Poems, ed. Colin Burrow, Oxford Shakespeare
(Oxford: OUP, 2002).
——Timon of Athens, ed. John Jowett, Oxford Shakespeare (Oxford: OUP, 2004).
——Cymbeline, ed. Martin Butler, New Cambridge Shakespeare (Cambridge:
CUP, 2005).
——William Shakespeare: The Complete Works, ed. Stanley Wells, Gary Taylor, John
Jowett, and William Montgomery, 2nd edn. (Oxford: Clarendon Press, 2005).
——William Shakespeare: Complete Works, ed. Jonathan Bate and Eric Rasmussen.
RSC Shakespeare (Basingstoke: Palgrave Macmillan, 2007).
——The Norton Shakespeare: Second Edition, gen. ed. Stephen Greenblatt
(New York: Norton, 2008).
——Timon of Athens, ed. Anthony B. Dawson and Gretchen E. Minton, Arden
Shakespeare (London: Cengage Learning, 2008).
Shakespeare, William et al., Sir Thomas More, ed. John Jowett, Arden Shakespeare
(London: Methuen, 2011).
Slater, Eliot, ‘Shakespeare: Word Links between Poems and Plays’, Notes and
Queries, 220 (1975), 157–63.
——The Problem of ‘The Reign of King Edward III’: A Statistical Approach (Cambridge:
CUP, 1988).
Smith, M. W. A. and Hugh Calvert, ‘Word-Links as a General Indicator of Chronology of Composition’, Notes and Queries, 234 (1989), 338–41.
Smith, Warren D., ‘The Henry V Choruses in the First Folio’, Journal of English and
Germanic Philology, 53 (1954), 38–57.
Southworth, John, Shakespeare the Player: A Life in the Theatre (Stroud: Sutton
Publishing, 2000).
Spevack, Marvin, A Complete and Systematic Concordance to the Works of Shakespeare,
9 vols. (Hildesheim: Georg Olms, 1968–80).
262Bibliography
——The Harvard Concordance to Shakespeare (Cambridge, MA: Belknap Press of
Harvard UP, 1973).
Spurgeon, Caroline F. E., Shakespeare’s Imagery and What It Tells Us (Cambridge:
CUP, 1935; repr. Boston: Beacon Press, 1958).
Stevenson, Warren, ‘Shakespeare’s Hand in The Spanish Tragedy 1602’, Studies in
English Literature: 1500–1900, 8 (1968), 307–21.
——Shakespeare’s Additions to Thomas Kyd’s ‘The Spanish Tragedy’: A Fresh Look at
the Evidence Regarding the 1602 Additions (Lewiston, NY; Queenstown, Ontario;
Lampeter, Wales: Edwin Mellen Press, 2008).
Sturgess, Keith, ed., Three Elizabethan Domestic Tragedies (Harmondsworth: Penguin, 1969).
——Review of The Tragedy of Master Arden of Faversham, ed. M. L. Wine, Durham
University Journal, n.s. 36 (1974), 245.
Swinburne, A. C., A Study of Shakespeare (London: Chatto & Windus, 1879; repr.
Heinemann, 1918).
Sykes, H. Dugdale, Sidelights on Shakespeare (Stratford-upon-Avon: Shakespeare
Head Press, 1919).
Symonds, John Addington, Shakespeare’s Predecessors in the English Drama (London:
Smith Elder, 1884, repr. 1900).
Tarlinskaja, Marina, Shakespeare’s Verse: Iambic Pentameter and the Poet’s Idiosyncrasies
(New York: Peter Lang, 1987).
——‘The Verse of A Lover’s Complaint: Not Shakespeare’, in Words That Count:
Early Modern Authorship: Essays in Honor of MacDonald P. Jackson, ed. Brian Boyd
(Newark: University of Delaware Press, 2004), 141–58.
——‘Who Did NOT Write A Lover’s Complaint?’, Shakespeare Yearbook, 15 (2005),
434–82.
Taylor, Gary, ‘Shakespeare and Others: The Authorship of Henry the Sixth, Part
One’, Medieval and Renaissance Drama in England, 7 (1995), 145–205.
——‘Middleton and Rowley—and Heywood: The Old Law and New Attribution
Technologies’, Papers of the Bibliographical Society of America, 96 (2002), 165–217.
——Review of Warren Stevenson, Shakespeare’s Additions to Thomas Kyd’s ‘The
Spanish Tragedy’: A Fresh Look at the Evidence Regarding the 1602 Additions and
Hugh Craig and Arthur F. Kinney, eds., Shakespeare, Computers, and the Mystery
of Authorship, Medieval and Renaissance Drama in English, 34 (2010), 198–201.
——‘A History of The History of Cardenio’, in The Quest for Cardenio: Shakespeare,
Cervantes, Fletcher, and the Lost Play, ed. David Carnegie and Gary Taylor
(Oxford: OUP, 2012), 11–61.
——with John V. Nance, ‘Four Characters in Search of a Subplot’, in The Quest for
Cardenio: Shakespeare, Cervantes, Fletcher, and the Lost Play, ed. David Carnegie
and Gary Taylor (Oxford: OUP, 2012), 192–213.
Taylor, Gary and John Lavagnino, gen. eds., Thomas Middleton and Early Modern
Textual Culture: A Companion to The Collected Works (Oxford: OUP, 2007).
Taylor, Michael, ‘The Year’s Contributions to Shakespeare Studies: Critical
Studies’, Shakespeare Survey, 61 (2008), 359–83.
Bibliography
263
Tynan, Kenneth, He That Plays the King (London: Longmans, Green, and Co., 1950).
Valenza, Robert J., ‘Are the Thisted–Efron Authorship Tests Valid?’, Computers
and the Humanities, 25 (1991), 27–46.
Vickers, Brian, Shakespeare, Co-Author: A Historical Study of Five Collaborative Plays
(Oxford: OUP, 2002).
——‘Incomplete Shakespeare: Or, Denying Coauthorship in 1 Henry VI’,
Shakespeare Quarterly, 58 (2007), 311–52.
——Shakespeare, ‘A Lover’s Complaint’, and John Davies of Hereford (Cambridge:
CUP, 2007).
——‘Thomas Kyd, Secret Sharer’, The Times Literary Supplement, 18 April
2008, 13–15.
——‘Shakespeare and Authorship Studies in the Twenty-First Century’, Shakespeare
Quarterly, 62 (2011), 106–42.
——‘Identifying Shakespeare’s Additions to The Spanish Tragedy (1602): A New(er)
Approach’, Shakespeare, 8 (2012), 13–43.
——Letters to The Times Literary Supplement, 3 and 17 August 2012, 6.
Vincent, Paul, ‘Unsolved Mysteries in Henry the Sixth, Part Two’, Notes and Queries,
246 (2001), 270–4.
——When ‘harey’ Met Shakespeare: The Genesis of ‘The First Part of Henry the Sixth’
(Saarbrücken: Verlag Dr. Müller, 2008).
Walker, Alice, Textual Problems of the First Folio (Cambridge: CUP, 1953).
Wallenberg, J. K., The Place-Names of Kent (Uppsala: Lundequistska Bokhandeln,
1934).
Warren, Roger, ‘ “A Lover’s Complaint”, “All’s Well”, and The Sonnets’, Notes
and Queries, 215 (1970), 130–2.
Wells, Stanley, Letter to The Times Literary Supplement, 7 September 2012, 6.
Wells, Stanley and Gary Taylor with John Jowett and William Montgomery,
William Shakespeare: A Textual Companion (Oxford: Clarendon Press, 1987).
White, Martin, ed., The Tragedy of Master Arden of Faversham, New Mermaids
(London: Ernest Benn, 1982; and rev. edn. introduced by Tom Lockwood,
London: A and C Black, 2007).
Wiggins, Martin, ‘Lord Mayor’s Showman’, review of David Gunby, David
Carnegie and MacDonald P. Jackson, eds., The Works of John Webster, Volume
Three, The Times Literary Supplement, 4 April 2008, 26–7.
——ed., A Woman Killed with Kindness and Other Domestic Plays (Oxford: OUP,
2008).
Wilson, J. Dover, ‘Bibliographical Links between the Three Pages and the Good
Quartos’, in Shakespeare’s Hand in ‘The Play of Sir Thomas More’, ed. Alfred W.
Pollard (Cambridge: CUP, 1923), 113–41.
——The Manuscript of Shakespeare’s ‘Hamlet’ and the Problems of its Transmission, 2
vols. (Cambridge: CUP, 1934).
Wine, M. L., ed., The Tragedy of Master Arden of Faversham (London: Methuen, 1973).
Zisser, Shirley Sharon, Critical Essays on Shakespeare’s ‘A Lover’s Complaint’: Suffering
Ecstasy (Aldershot: Ashgate, 2006).
Index
Entries in bold denote references to tables. Entries of the form ‘XnY’ denote f­ ootnote
Y on page X. This index provides guidance to the book’s content but does not include the names of every scholar mentioned in the text, footnotes, or comprehensive
Bibliography of works cited.
‘a-twain’ 136, 201
Achelley, Thomas 120
adjectives, compound 76–7, 77
adultery 9, 12, 125
‘affectedly’ 203
‘airy’ 51
The Alchemist 156
Alison, Richard 204
Allen, Percy 98n21
Allott, Robert 120
Antony and Cleopatra
reflexive conceit in 30
vocabulary of 91, 195, 198
‘appertainings’ 195, 201
‘aptness’ 190–1
Archer, Edward 13
Arden of Faversham
Alice in 12, 104–5
Arden’s dream in 54–5, 55
Arden’s speech 55–6, 58, 66,
102–3
attribution of authorship 1, 14, 53,
117–20, 126
co-author of 74–5, 84
compound adjectives in 76–7
Craig-Kinney tests on 38–9, 41,
48–52, 56
first publication of 13
geography in 110–13
hunting imagery in 65, 93
imagery and wordplay in 28, 87–90,
96–103
and A Lover’s Complaint 150, 156
metrical features of 115–16
non-Shakespearean features of 78–9,
81
‘qualm’ in 94–6
quarto text of 14–15, 81–3
reflexive conceit in 29–30
scene 3 65–6
scene 8 67–8
scene 14 72–3
Shakespeare’s contribution to 1–2,
5–6, 38, 75
social hierarchy in 61–3
sources of 15
stage directions in 105–10
staging of 121–2
textual parallels with 16, 21, 57,
65–6, 80
As You Like It, and Quarrel Scene 73
The Atheist’s Tragedy 203
attribution studies 5, 56, 116
authorship, positive and negative
evidence of 131, 167,
217–18
‘ay, but’ 79–80, 82
bad quartos 108, 112, 122
‘Beauty’s Periphrasis’ 188
The Betraying of Christ 189
Browne, William 165
Burrow, Colin 202
Burrows, John 39, 42, 68
Butler, Martin 188, 261
Cairncross, Andrew S. 122, 126
‘cannon’ 34
266Index
Cardenio 151n22
Carroll, Jayne 66
Chapman, George
links to A Lover’s Complaint 132,
134, 156–7, 171, 217
spellings in 148, 152
chi-square test 69n29, 115
children’s companies 104–5
Claremont Shakespeare Clinic 38,
84n65, 130–1, 209–12
‘closet’ 88–9, 98
Cockerell, Sydney 1
Collins, Stephen 116
The Comedy of Errors
Adriana’s entreaty in 31–2
dating of 77
Pinch in 76
‘company’ 99
compositors, and spelling variations 51,
141–2, 159, 161, 164, 167–8
computational stylistics 45, 48, 77, 124,
126, 217
‘copesmate’ 34, 36
Coriolanus, Craig-Kinney tests of 43
Cornelia 16, 52, 86, 113–15
Craig, Hugh 40, 63, 217
Craig-Kinney tests
and Arden authorship 39, 49, 63–4,
81, 123–4
and Henry VI trilogy 46
methodology of 44–5, 51
Vickers’s critique of 41–4
‘credent’ 195, 200–1
Cymbeline
links to Arden 92n15
publication of 195
rare vocabulary in 185–92, 194, 201,
205
vocabulary of 183, 191, 194, 201–5
De Vere, Edward 119
Dekker, Thomas 80, 119
Dido Queen of Carthage 105, 118
Doctor Faustus
co-authorship of 46
LION analysis of 24–6, 25, 71, 118
domestic tragedy
Arden as 6, 75, 84
in Yarington’s work 22
dove and raven, imagery of 34–5, 37,
51, 99
dramatic poetry 86
Drayton, Michael 119–20
Dudley, Robert 15
Dahl, Marcus 53
darkness, sheeting 89, 91
Davies, John
as author of A Lover’s Complaint 3–4,
135, 141, 187
extent of canon 156n26
poetry of 136, 138–40, 152, 156,
165–7, 212–13
spellings used by 160, 162–4, 163
Tarlinskaja on 209
Fair Em the Miller’s Daughter 53, 106,
113
‘fancy’ 203–4
‘feat’ 187, 190–1, 194
Ferrers, George 119
Fitten, Jack 76–7
Fletcher, John 5, 132, 170
‘fold’ 91
‘forbod’ 147, 153, 201–2
fountains, imagery of 28–9, 34
‘eager mood’ 92–3
eagle, imagery of 31, 33, 36
The Earl of Westmorland 188
Edmond Ironside 44
Edward II
‘closet’ in 88–9
similar phrasings to Arden 16, 80
Edward III
authorship of 13–14, 19n31, 38
Craig-Kinney tests of 44
links with Arden 67, 71–2, 124–5
lions in 58
and A Lover’s Complaint 139, 150
rare words in 186
Shakespeare’s contribution to 2, 5
Edwards, Richard 119
Eld, George 129, 147, 167
Elliott, W. E. Y. 22n38, 38–9, 84n65
enclitic phrases 208
Endymion 15, 28
‘everlasting night’ 89, 91–2
Exact Probability Test 49
Index
franklin, as social class 12n5
Freeman, Arthur 47
French language 182
function-word testing 42, 44, 46,
50–2, 84n65
A Game at Chess 142
Garner, Bryan 171, 173, 175, 182–3
Gascoigne, George 119
‘gentle’ 11, 51, 61–4, 83–4
The Golden Age 132
good quartos 141, 159
Greenblatt, Stephen 60
‘gyves’ 149, 160, 186–7, 190–1
Hamilton, Charles 151n22
Hamlet
links to Arden 156
neologisms in 173
reflexive conceit in 30
setting of 151
vocabulary of 174–5
‘handkerchief ’ 196n26
Hart, Alfred 76, 81, 169–70, 173–4,
176–8, 180–2
Harvey, Richard 74
Hathway, Richard 118–20
Haughton, William 74
Hengist 43
Henry IV
Part 1, vocabulary of 175
Part 2
imagery and wordplay in 27, 88
links to A Lover’s Complaint 156
setting of 151
vocabulary of 175
Henry V
choruses in 169
lions in 58
and A Lover’s Complaint 138
social hierarchy in 60–1
Henry VI trilogy
Craig-Kinney tests on 46–7
links with Arden 16, 21–2, 67, 72–3,
84, 121
Part 1
authorship of 5n16, 21, 39, 46
LION searches of 20
Marlowe as co-author 113
267
Nashe’s contribution to 64
Part 2
Act 3 of 66
compound adjectives in 77
hunting imagery in 93
imagery and wordplay in 87
language of 82–3, 95–8
Margaret’s entreaty in 32–3
Shakespeare’s contribution to 64–
5, 68, 70, 125
shared imagery with Arden 66
textual parallels with Quarrel
Scene 33–4, 36–7
Part 3
authorship of 5, 39, 68–9
‘bad octavo’ of 165n47
hunting imagery in 55–6, 58
links with Arden 72, 124
use of language in 79, 98
and A Lover’s Complaint 137, 150
Henry VIII, vocabulary of 174
‘hew’ 159–60
Heywood, Thomas
spelling of 142
on Watson 119
and A Lover’s Complaint 132, 134,
156–7
‘high and low’ 137, 139–40, 199
‘hive’ 51
Holinshed, Raphael
on geography of Kent 112
story of Arden in 15, 62, 111, 121
The Holy Rood 166n53, 197n28
Humour’s Heaven on Earth 139, 191n19,
203, 214, 216
hyphened words 147
indefinite article 166
Isle of Sheppey 111
Jacob, Edward 14
James IV
‘qualm’ in 96
and Shakespeare’s drama 186
textual parallels to Arden 22
The Jew of Malta
co-authorship of 46
links to Doctor Faustus 25
links with Arden 80–1
268Index
Jonson, Ben
canon of 53
epigram to First Folio 60
spelling of 142
and A Lover’s Complaint 132
Jowett, John
on ‘spongy’ 202
and A Lover’s Complaint 136
and stage directions 108
Julius Caesar
‘closet’ in 89
and A Lover’s Complaint 137
‘purchase’ in 99n22
Juola, Patrick 40
Kent, geography of 111–12
King John
imagery and wordplay in 86–7
reflexive conceit in 30
King Lear
and A Lover’s Complaint 137, 156
quarto of 200
setting of 151
spellings in 158–9
in Thisted-Efron tests 178
vocabulary of 198
Kyd, Thomas
expansion of canon 53, 114–15
as poet 85–6
use of stage directions 106
and authorship of Arden 2, 15–16,
47–8, 79, 113–14
and Craig-Kinney tests 52
land, struggle for 62
Latin 106, 119, 181–2, 195
Latinisms 102, 174–5, 174, 182–3
‘laund’ring’ 196, 198
lexical-word tests 43–6, 48, 50–1, 65
‘linger’ 91–2
LION (Literature Online)
and Arden of Faversham 50, 55, 66–7,
72–3, 73
and A Lover’s Complaint 131, 136,
143
Shakespeare’s works included
in 18–19, 144
textual parallels to Doctor Faustus 25
textual parallels to Quarrel Scene 21
textual searches in 17–19
and variant spellings 51
Wiggins’s criticisms of 6
lions, imagery of 58
lips, glued 98
A Looking Glass for London and England,
textual parallels of 26, 95
Love, Harold 85
A Lover’s Complaint
Claremont Clinic tests on 210–12
date of composition 171n11, 175–6,
186
debate on authorship of 3–4,
215–16
hyphenation in 147–8
imagery of 200
metrical features of 207–9
neologisms in 171–5, 183
peculiar and rare words in 177, 182,
188–92, 195–6, 201–5
in quarto text of Sonnets 129–30
rhyme in 213–14
and Shakespeare’s drama 131–4,
136–40, 185–6
spellings in 142–7, 149–55, 157–61,
164
textual errors in 160
Thisted-Efron tests on 179–81
wordplay in 197–9
‘A Lover’s Complaint Revisited’ 130–
1, 133, 135
Love’s Labour’s Lost
Berowne’s speech in 33
imagery in 27, 56, 58
links with Arden 121
rare spellings in 150
setting of 151
vocabulary of 95, 175
Love’s Labour’s Won 118
‘lungs’ 202–3
Lyly, John 15, 28
Lyly, Thomas 204
Macbeth
Arden’s foreshadowing of 11n3, 91, 98
water-dogs in 93
spelling in 160
vocabulary of 100, 175, 191
Mackail, J. W. 169, 171
Index
Marlowe, Christopher
and authorship of Arden 2, 15–16,
63, 83, 110–11
and Henry VI Part 2 46–7
influence on Shakespeare 24
peculiar words in 181
textual parallels of plays 24–5, 26
Marston, John 132
The Massacre at Paris 25, 47, 55–6
Measure for Measure, vocabulary of 189,
191
Medea, Studley’s translation of 98
memorial error 15, 81, 105,
108–10
The Merchant of Venice
and A Lover’s Complaint 133
vocabulary of 99, 197
Meres, Francis 118–20
Merriam, Thomas 79
The Merry Devil of Edmonton 56
The Merry Wives of Windsor
hunting imagery in 94
vocabulary of 175, 199
metaphors, moribund 87, 99
Microcosmos 165n45, 193n21, 202–3
Middleton, Thomas
authorship studies on 53
collaborations with Shakespeare 5,
45
spelling of 142, 167
and Timon of Athens 116–17
A Midsummer Night’s Dream
hunting imagery in 94
social hierarchy in 61
vocabulary of 175
Mincoff, Marco 69
Mucedorus 91
‘napkin’ 196
Nashe, Thomas
and Henry VI Part 1 46, 64, 69
verse style of 5
Neill, Michael 62, 83
neologisms 172, 174–6, 174, 176, 183,
216
Nobody and Somebody 201
non-Shakespearean words
in A Lover’s Complaint 176
use of term 169
269
The Old Wives Tale 132–3
Oliphant, E. H. C. 71, 74,
108–9, 111
Othello
links to A Lover’s Complaint 133
spellings in 148
in Thisted-Efron tests 178
vocabulary of 175, 196n26
‘outward’ 188–94, 200, 205
Oxford Textual Companion, chronology
of 18
peculiar words 170, 176–8, 180–2
Peele, George
and Titus Andronicus 18n31, 39, 45,
67
writing style of 5
‘peeping through’ 192
‘pelleted’ 195, 197–201
Pembroke’s Men 122
Pericles
co-authorship of 38
and A Lover’s Complaint 133, 150
‘sister’ in 137n15
vocabulary of 174, 194
‘perticular’ 156n25
The Phoenix and Turtle 160, 168, 172–6,
210n7
‘physic’ 186–7, 189–91, 200–1, 205
‘platted’ 201, 203
poetry, spelling in 152
Porter, Henry 74–5
principal component analysis
(PCA) 41–3, 115, 210n8
‘qualm’ 94–6, 98
Quarrel Scene
Alice’s entreaty in 31–3
authorship of 14–15
computerised testing of 39n62
links to early collaborations 68–71,
70, 73, 121–5
quarto text spellings 51
rare phrasing in 26–9
summary of 9–13
textual parallels with 17–24, 21,
26, 33–4, 35–7, 55, 57–8,
118
quirks, accumulation of 130–1
270Index
The Rape of Lucrece
and Claremont tests 209, 211
and A Lover’s Complaint 3, 138, 156,
197
neologisms in 172
rare words in 202
shared imagery with Arden 66, 90,
93
textual parallels with Quarrel
Scene 23, 27, 30, 33–4, 35–6,
121
in Thisted-Efron tests 178
verse form of 129, 209, 214
reflexive conceit 29–30, 72
regression effect 44, 52
The Revenger’s Tragedy 132, 147, 167
rhyme royal 129, 185, 207, 213
Richard II, reflexive conceit in 30
Richard III
darkness in 91
and A Lover’s Complaint 138
reflexive conceit in 30
simultaneity in 110
‘rifled’ 34
Romeo and Juliet
bad quarto of 108, 151, 155n24
in Thisted-Efron tests 178
vocabulary of 175, 197
‘rudeness’ 190–1
Sapho and Phao 95
Schäfer, Jürgen 171, 173, 175
‘seared’ 188, 190, 200, 205
The Second Maiden’s Tragedy 145–6,
150–1, 151n22, 155–6
seeing 78–9, 81
Select Second Husband for Thomas
Overbury’s Wife 203, 205
Selimus 22
sexual possession 62
Shakespeare, William
acting career of 123n43
collaborations of 68
differences from Kyd 86
Droeshout engraving of 60
early collaborations of 70, 71, 79,
124–6
female roles in 104
and hunting imagery 58, 65, 94
Latinate neologisms in 174
and A Lover’s Complaint 3–4
materiality and canon 5–6
neologisms of 170–1, 175, 176
non-dramatic verse of 134, 152, 173,
177, 207–8
spellings used by 157–60, 162, 163
and Timon of Athens 117
types and tokens in 172–3
use of language 27, 102–3, 171
use of rhyme 213–14
vocabulary of 170–1
Shakespeare First Folio
Arden’s omission from 13, 38
copyright of 38n61
‘disintegration’ of 85
Droeshout engraving in 60
sharp-sightedness 33
‘sheets’ 91
The Shepherd’s Pipe 165
simultaneity 110
Sir Thomas More
Hand D in 44, 141–2, 150, 159–60,
167
links with Arden 72
‘sister’, as a verb 137n15
‘sith’ 163–6
‘slackly’ 189–90
Slater, Eliot 185–6
Smith, Henry D. 169
social hierarchy
characters’ awareness of 12
language of 84
mobility within 60
Soliman and Perseda
and Craig-Kinney tests 52
authorship of 86, 97, 106
‘ay, but’ in 79
links to Arden 16, 77–8, 89, 92–4,
96–7, 103
stage directions in 106–8
‘somnet’ 158
Sonnets
‘audit’ in 90
and Claremont tests 211
‘forge’ in 96
images of cold and new growth
in 27
links to Shakespeare’s drama 134–5
Index
A Lover’s Complaint in 184
metrical segmentation of 207
neologisms in 172
publication of 129
quarto of 4, 141, 161, 167
in Thisted-Efron tests 178
The Spanish Tragedy
Additions to 40, 45, 59
authorship of 16
‘ay, but’ in 79
dating of 86
Hieronimo’s soliloquy in 102
hunting imagery in 93
images of cold and new growth
in 27–8
peculiar words in 181
and Shakespeare’s drama 186
spelling preferences 141, 161
spellings
of grammatical endings 148–9
rare 143–4, 149–53, 157, 161–2, 167,
216
Spenser, Edmund 165, 198, 207–8, 212
Spevack, Marvin 5n17, 173, 177, 180,
182
‘spongy’ 187, 191, 201–3
Spurgeon, Caroline 27
stage directions 105–8
Stationers’ Register
Arden in 13, 57
Soliman and Perseda in 86
Sonnets in 129, 194
status, language of 62, 83
‘steel’ 88, 94, 97
Steevens, George 197, 200
Stevenson, Warren 40–1, 45, 53, 59
‘storm’ 200
Sturgess, Keith 9, 38, 105
style, measurement of 217
Summa Totalis 139, 163n43, 192
Swinburne, Algernon 1, 3, 14
syntactic markers 42
Tamburlaine
links to Arden 118
links to Doctor Faustus 25
vocabulary of 181
The Taming of the Shrew
female roles in 105
271
links to Arden 47n19, 98, 103, 113,
115
textual parallels of 16, 28
Tarlinskaja, Marina 114–16, 130,
207–9, 212
Taylor, Gary 20
‘The Hour-glass’ 91–2
Thisted-Efron tests 178–9, 179, 181–2
Thorpe, Thomas 4, 129, 141, 168, 215
time, in Arden and Lucrece 90–1
Timon of Athens, as collaboration 116–17
Titus Andronicus
authorship of 39, 18n31
Craig-Kinney tests of 45
imagery in 38
links to Arden 73, 115, 124
and A Lover’s Complaint 133
and Marlowe 71
Peele’s work in 69
reflexive conceit in 30
tokens and types 172–3, 173, 177, 180,
182
trigrams
unique 47n19
in Vickers’s method 53–4, 89, 113
triple rhymes 184, 213–14
Troilus and Cressida
neologisms in 173
spellings in 160
textual parallels to Arden 29
utterance junctures in 208
vocabulary of 174–6, 189, 191
The True Chronicle History of King Leir
30n50, 113
‘tush’ 78–9, 81
Twelfth Night, vocabulary in 197, 199
The Two Angry Women of Abingdon
74–5
The Two Gentlemen of Verona
hunting imagery in 93
reflexive conceit in 30
use of language in 79, 99
Two Lamentable Tragedies, textual
parallels of 22–3, 25–6
The Two Noble Kinsmen
co-authorship of 38
Fletcher’s contribution to 170
and A Lover’s Complaint 150
vocabulary of 174, 190n17
272Index
The Valiant Welshman 189
Venus and Adonis
and Claremont tests 209–11
lamenting in 198
links with Arden 56–9, 57, 66, 98,
121
and A Lover’s Complaint 137–9, 193
reflexive conceit in 30
in Thisted-Efron tests 178
vocabulary of 91, 172, 175
Vickers, Brian
on Arden of Faversham 47–8, 50, 52,
86, 113–14
on authorship studies 40–2, 53–4,
56, 59
on Craig-Kinney tests 45–6
and Cymbeline 187, 190
on imagery 197–8
and LION searches 20
on A Lover’s Complaint 3–4, 130,
135–6, 141, 162–6, 171–2,
183–5, 212–15
other attributions of authorship
43–4
on rare words 169, 201–2, 205–6
use of plagiarism software 53–4,
103, 113
and Shakespeare as co-author 5
and Soliman and Perseda 89
Vincent, Paul 20, 66
Warner, William 156–7
Warwickshire 58
water-dogs 93–4
Watkins, Rowland 91
Watson, Thomas 119–20
Webster, John, characteristic phrases of 20
Whetstone, George 203
Wiggins, Martin 6, 104–5, 108–11
Wilkins, George 5
Wilson, J. Dover 141, 159
Wine, M. L. 65
Winter, Thomas 203
The Winter’s Tale, sound in 199
Wit’s Bedlam 156, 166n53, 190–1, 201
Wit’s Pilgrimage 139, 202, 204
word, definitions of 170, 177, 180
wordplay
in Arden 10, 66
Shakespeare’s use of 86, 88