Kyd and Shakespeare: Authorship versus Influence

This article explores Shakespeare’s verbal indebtedness to works that have been attributed to Thomas Kyd, encompassing plays such as Soliman and Perseda, King Leir, and Arden of Faversham. Significantly, Martin Mueller has created an electronic corpus called Shakespeare His Contemporaries, which consists of over 500 plays dated between 1552 and 1662. Shakespeare His Contemporaries lists play pairs that share large numbers of dislegomena consisting of four words or more, and therefore provides empirical data that can help researchers to explore the intertextual relationships between early modern texts. This article investigates the nature of these parallels, drawing upon the idea of Shakespeare’s aural, or ‘actor’s memory’, and concludes that in order to distinguish between authorship and influence in contested texts like Arden of Faversham, more work needs to be done to ascertain the patterns of influence in Shakespeare’s plays. Contributor biography: Darren Freebury-Jones completed his PhD at Cardiff University. His doctoral thesis, ‘Kyd and Shakespeare: Authorship, Influence, and Collaboration’, investigated Brian Vickers’s arguments for an expanded Thomas Kyd canon. Freebury-Jones will serve as an editor and authorship consultant for the first edition of Kyd’s works since 1901. He recently investigated the boundaries of John Marston’s dramatic corpus for Oxford University Press. His work on the canons of dramatists like Shakespeare, Marlowe, and Kyd can be found in such journals as Journal of Early Modern Studies, Archiv für das Studium der neueren Sprachen und Literaturen, Notes and Queries, and A Quarterly Journal of Short Articles, Notes and Reviews. In this article I provide a brief overview of the scholarship concerning Shakespeare’s drama in relation to Thomas Kyd’s dramatic corpus, thus suggesting the ways in which the academy has been slowly drifting toward a collaborative model of authorship and influence.1 In Section Two I examine verbal affinities between Shakespeare’s early drama and Kyd’s accepted plays, The Spanish Tragedy (1587)2 and Soliman and Perseda (1588),3 whilst drawing upon the idea of Shakespeare’s ‘actor’s memory’. Finally, in 1 I would like to thank the two anonymous reviewers, whose detailed comments helped to improve and clarify this article. 2 I have used Martin Wiggins and Catherine Richardson’s British Drama 1533-1642: A Catalogue: Volume II: 1567-1589, 10 vols (Oxford: Oxford University Press, 2012) and British Drama 1533-1642: A Catalogue: Volume III: 1590-1597 (Oxford: Oxford University Press, 2013) for the dates of first performances. 3 Soliman and Perseda, like The Spanish Tragedy, was printed anonymously. The play has only been accepted as Kyd’s relatively recently, despite the fact that, as Terence P. Logan and Denzell S. Smith noted in 1973, ‘The only serious candidate’ for the play’s authorship is Kyd and ‘Almost every scholar who discusses either the play or the playwright acknowledges the connection’. Terence P. Logan and Denzell S. Smith, The Predecessors of Shakespeare: A Survey and Bibliography of Recent Studies in English Renaissance Drama (Lincoln, NE: University of Nebraska Press, 1973), p. 233. For further discussion on Kyd’s authorship of the Turkish tragedy, see Thomas Hawkins, The Origin of the English Drama, 3 vols (Oxford: Clarendon Press, 1773), I.x-xi; F. G. Fleay, A Biographical Chronicle of the English Drama, 2 vols (London: Reeves and Turner, 1891), II.26; Gregor Sarrazin, Thomas Kyd und sein Kreis: eine litterarhistorische Untersuchung (Berlin: E. Felber, 1892), p. 42; Adolphus William Ward, A History of English Dramatic Literature (London: Macmillan, 1899), p. 311; James E. Routh Jr., ‘Thomas Kyd’s Rime Schemes and the Authorship of Soliman and Perseda and of The First

Section Three, I compare results for the contested Kyd texts, The True Chronicle History of King Leir (1589) and Arden of Faversham (1590).This study is intended to broaden our understanding of Shakespeare's patterns of influence, in relation to plays written by his contemporaries.

I. The Argument for Shakespeare's Aural Memory
Many scholars have been willing to accept the notion that Shakespeare followed the standard practice of borrowing from his fellow dramatists.For example, Hardin Craig suggested in 1951 that Shakespeare had acted in King Leir and was thus able to recall the play. 4In 1958, Thomas H. McNeal listed numerous verbal links between Shakespeare's plays and King Leir.He concluded that Shakespeare borrowed 'in both phrase and paraphrase' from the old play throughout his career. 5Bart Van Es notes that 'Shakespeare's early drama is often spectacularly imitative and as a result his personal voice is much less distinct', 6 while Charles R. Forker has suggested that '[m]uch of this assimilation was undoubtedly unconscious, at least in the case of verbal echoes, since Shakespeare seems to have known many of the plays from practical experience in the theatre'. 7I argue that Shakespeare's ability to weave verbal details from other plays into his own passages is in part attributable to his career as an actor.
We know frustratingly little about Shakespeare's acting career.The first allusion to Shakespeare as an actor and dramatist features in Robert Greene's Groatsworth of Wit Bought with a Million of Repentance (1592).Greene warns his fellow dramatists and University Wits, Nashe, Peele, and Marlowe, about actors, 'those Puppets (I meane) that spake from our mouths, those Anticks garnisht in our colours', and one actor in particular.Shakespeare, or 'Shake-scene', has had the audacity to turn his hand to writing plays: Yes trust them not, for there is an upstart Crow, beautified with our feathers that, with his Tygers heart wrapt in a Players hyde, supposes he is as well able to bombast out a blanke verse as the best of you: and being an absolute Johannes fac totum, is in his own conceit the only Shake-scene in a countrey. 8rt of Jeronimo ', Modern Language Notes, 20.2 (1905), 49-51; Philip Edwards, Thomas Kyd and Early  Elizabethan Tragedy (London: Longmans, Green and Company, 1966), p. 23; Arthur Freeman, Thomas Kyd: A  Study of Facts and Problems (Oxford: Clarendon Press, 1967), pp.140-166; Thomas Merriam, 'New Light on a  Possible Kyd Canon', Notes and Queries, 240 (1995), 340-341; Lukas Erne, Beyond The Spanish Tragedy: A Study of the Works of Thomas Kyd (Manchester: Manchester University Press, 2001), pp.157-167. 4See Hardin Craig, 'Motivation in Shakespeare's Source of Materials', Shakespeare Survey, 4 (1951), 26-34. 5 Thomas H. McNeal, 'Margaret of Anjou: Romantic Princess and Troubled Queen', Shakespeare Quarterly, 9 (1958), 1-10 (p.5). 6Bart Van Es, Shakespeare in Company (Oxford: Oxford University Press, 2013), p. 36. 7Charles R. Forker, 'The Troublesome Reign, Richard II, and the Date of King John: A Study in Intertextuality', Shakespeare Survey, 63 (2010), 127-148 (p.127). 8Robert Greene, Greene's Groatsworth of Wit Bought with a Million of Repentance, ed.D. Allen Carroll  (Binghampton, NY: Medieval and Renaissance Texts and Studies, 1994), pp.83-84.
There is a 'list printed in the 1616 Jonson Folio', which tells us that Shakespeare was 'one of the "principall Comedians" in Ben Jonson's Every Man in his Humour' in 1598. 9 He is also 'listed among "The principall Tragedians" in Ben Jonson's Sejanus'. 10Finally, Shakespeare is listed as one of the principal actors in his own plays, in the First Folio (1623).John Davies of Hereford tells us that he often played 'Kingly parts in sport'. 11 consider it most likely that Shakespeare began his career as an actor-dramatist for Pembroke's Men, as proposed by J. O. Halliwell-Phillipps during the nineteenth century. 12Shakespeare seems to have written his earliest plays, such as Henry VI Part Two (1591), Henry VI Part Three (1591), and The Taming of the Shrew (1592), with that company before it disbanded in 1593.Following the company's collapse, some of Pembroke's players were able to produce memorial reconstructions of Shakespeare's texts, as was convincingly argued by Madeleine Doran in 1928, 13 and Peter Alexander in 1929. 14  (1595).I agree with Hart that these texts are 'garbled abridgements of the acting versions made by order of the company from Shakespeare's manuscripts'. 15. S. Cairncross argued that Pembroke's Men 'existed before 1592, probably as early as 1589', and it was 'Shakespeare's company, as it was, for a time at least, Kyd's'. 16imilarly, Terence Schoone-Jongen points out that 'Pembroke's 1592-93 court performances indicate it probably had existed long enough to attract the court's attention, and presumably had actors and/or writers talented enough to attract such attention'. 17He notes that 'Surviving evidence' linking Shakespeare's early acting career with 'Pembroke's Men is more plentiful than surviving evidence for some of its fellow playing companies'. 18It seems likely that, as an actor-dramatist for Pembroke's Men, Shakespeare would have developed an ear for the 'useful phrases from a probably unsorted store of theatrical utterances that had become commonplace in the minds of their users'.19Indeed, T. W. Baldwin suggested in 1959 that Shakespeare 'would learn, from acting in the old plays' of authors 'such as Kyd'. 20I propose that proper acknowledgement of Shakespeare's beginnings as an actor can tell us much about the hybrid nature of his plays, or what we might call, as Gloucester puts it in Henry VI Part Two, his 'books of memory'. 21eoffrey Bullough observed that Shakespeare 'seems to have forgotten nothing that he read or heard, or rather, his powers of associative memory were such that if he required a parallel or contrast for plot and incident or a poetic image, something relevant and vivid floated up from his unconscious'. 22Significantly, John Tobin notes that [b]ecause plays were very seldom performed in an uninterrupted run, actors needed powerful memories.It was a time when the aural rather than the visual understanding was much greater than in our own time, but even so, the capacity of actors to hold in their heads a large number of roles from many different plays was extraordinary, and new plays were constantly being added to the repertory. 23e 'capacious, book-like memory' Shakespeare required in order to succeed as an Elizabethan player meant that he could draw from a variety of plays for the verbal details of his own works. 24However, we cannot rule out the possibility that Shakespeare simply remembered other dramatists' lines from having seen their plays during performance.Let us consider Hamlet's recitation of the Player's speech, which 'engages directly with the mechanics of recall'. 25The eponymous character offers 'the Player a cue line to stimulate his memory of the speech; he misremembers, and he corrects his memory'. 26Hamlet is able to recall a thirteen-line speech, with 'good accent and good discretion' (Ham., II.ii.469-470), despite his having only heard the 'speech once, but it was never acted, or, if it was, not above once ' (II.ii.437-438).It is conceivable that Shakespeare's 'prodigious skills of memorization required for the theatre' would similarly enable him to recall a number of speeches from plays he had engaged with, either as a spectator or actor. 27In authorship studies, there is much emphasis placed on authorial self-borrowing but, in my view, a dramatist's patterns of influence can also serve as useful authorship markers.Notably, Lukas Erne observes that 'Shakespeare, perhaps more than anyone else, seems to have specifically profited from Kyd's works'. 28en Jonson coupled Kyd's The Spanish Tragedy with Shakespeare and George Peele's Titus Andronicus (1592) in his Induction to Bartholomew Fair (1614): 'He that will swear Jeronimo or Andronicus are the best plays yet, shall pass unexcepted at here, as a man whose judgement shows it is constant'. 29Shakespeare's first tragedy, like Kyd's most famous play, is written in the Senecan mode.Erne summarizes Kyd's influence over Shakespeare's subsequent tragedies thus: His second tragedy, Romeo and Juliet, did what only Kyd's Soliman and Perseda among extant plays had done before on the public stage, namely to place a conflict of love at the centre of a tragedy.His third tragedy, Julius Caesar, covers the same period of Roman history as Kyd's Cornelia, and Shakespeare's Brutus may well owe something to Kyd's.Finally, the chief source of Shakespeare's fourth tragedy, Hamlet, is undoubtedly Kyd's work of the same name. 30hich lists play pairs that share large numbers of unique tetragrams plus (four-word sequences or more). 34Mueller notes that 'it is quite rare for two plays-texts that are typically between 15,000 and 25,000 words long-to share more than one or two of the dislegomena' [n-grams that occur in only two plays in Mueller's corpus] analyzed here'. 35Mueller's database lends weight to the hypothesis that Shakespeare recycled verbal details from earlier plays. 36In the following table I present Mueller's data for unique n-grams of four or more words, shared between Kyd's accepted tragedies and Shakespeare's plays: Although John Southworth argues that 'Shakespeare's familiarity with Kyd's Spanish Tragedy' was 'more likely to derive from having acted' in it, the very popularity of the play presents a difficulty, for many of its phrases seem to have been embedded in the minds of Kyd's contemporaries. 37The play was parodied by dramatists such as Nashe, Heywood, Marston, Dekker, Jonson, Field, Beaumont, and Shirley, while Mueller's database records large numbers of repetitions between the tragedy and sixteen plays by different authors (although it is perhaps worth noting that six of these plays are Shakespeare's). 38onversely, as Freeman pointed out, '[t]hat Soliman never attained the popularity of The Spanish Tragedy is evident, both from its scant printing history and the paucity of allusions to it in its own time'. 39Given that Kyd's Turkish tragedy is 'generally assumed to have been printed not long after it was entered in the Stationers' Register in November 1592', 40 it seems possible that the verbal affinities with Henry VI Part Three (which was almost certainly on stage by September 1592, given that an explicit allusion to the play occurs in Greene's prose tract) are due to Shakespeare's 'fabulous "aural memory"'. 41Significantly, Hart highlighted inter-play borrowings from Soliman and Perseda, Edward II (1592), and Arden of Faversham in what he considered to be memorial reconstructions of Shakespeare's Henry VI plays.Hart's findings suggest that Kyd's play had been in the repertory of Pembroke's Men. 42Erne argues that Kyd's Turkish tragedy belonged to Pembroke's Men 'until at least 1597', although 'we do not know for which company Kyd wrote his play'. 43Shakespeare's ability to recall the verbal details of these texts could therefore be the result of his having acted in them.
Southworth tells us that '[p]erforming in a play brings to the actor a general familiarity with the text as a whole', for 'he needs to give half an ear to what is being spoken on stage if he is not to miss his entrance cues'. 44He claims that '[i]t is apparent' Shakespeare played Erastus in Kyd's Soliman and Perseda. 45Indeed, we find the unique pentagram (five-word sequence) in Mueller's spreadsheet, 'And thanks unto you all', shared between Erastus's speech, 'And thankes unto you all, brave worthy sirs./ Impose me taske, how I may do you good; / Erastus will be dutifull in all', 46 and King Edward's lines, 'Thanks, brave Montgomery, and thanks unto you all./ If fortune serve me I'll requite this kindness' (3H6, IV.viii.76-77). 47We might note the similar contexts in which this formation is employed: both characters are thankful and offer requital.However, this is the only unique word sequence between Kyd's play and Henry VI Part Three that occurs during Erastus's dialogue. 48Other word sequences in Mueller's database occur in scenes during which Erastus is on stage.Ferdinando's interrogative, 'Dasell mine eyes, or ist Lucinas chaine?' (S&P, II.i.244), provides a cue for Erastus to speak and matches Edward's line, 'Dazzle mine eyes, or do I see three suns?' (3H6, provides a cue for Erastus to speak, while matching (in language, but not in thought) Henry's prophecy that Richmond will become King: 'Make much of him, my lords, for this is he ' (3H6, IV.vii.75).
Some repeated phrases, however, cannot be explained by the theory that Shakespeare played Erastus and recalled his own lines or cue-lines.Erastus is not present on stage when Amurath, accompanied by Soliman, Haleb, and the Janissaries, says: 'I would not hence till I had let thee know' (S&P, I.v.53).Amurath attacks Haleb for thwarting 'a Monarchs holy oath' (I.v.54).Shakespeare recalls this line when Margaret calls Warwick a 'Proud setter-up and puller-down of kings!/ I will not hence, till, with my talk and tears' (3H6, III.iii.157-158).Significantly, the next unique word sequence occurs during Soliman's speech, and Erastus has been murdered two scenes previously: My last request, for I commaund no more, Is that my body with Persedas be Interd, where my Erastus lyes intombd.(S&P, V.iv.140-142)In Shakespeare's play, the King says, 'Let me entreat-for I command no more-' (3H6, IV.vii.59).Erastus is also absent from Act Two Scene Two, when Basilisco says, 'Why so?I am in honor bound to combat him' (S&P, II.ii.52), which matches Henry's contextually dissimilar line, 'Why, so I am, in mind' (3H6, III.i.60).Shakespeare thus seems to have been familiar with the language of the play as a whole.Hart noted that '[t]here seems no good reason why any actor who had no part in a scene should be excluded from the stage during its rehearsal'. 49Similarly, Brian Vickers observes that All actors were expected to attend rehearsals on the day of the performance, and they could hardly avoid attending to the play during performance, since, with the exception of the principal roles, others were doubled, and actors would be watching out for their entry cue.Further, in the confined space of the Elizabethan theatre, and even more so in the venues encountered on tour, it would have been impossible not to know what was happening on stage. 50 is therefore not unreasonable to suggest that Shakespeare familiarized himself with scenes from Soliman and Perseda during which he was not required on stage.
We find a similar pattern to those shared by Henry VI Part Three and Soliman and Perseda in the matches between Kyd's play and The Two Gentlemen of Verona (1594).Just one of these unique n-grams, the pentagram, 'not and therefore she is', is spoken by Erastus, in the line, 'I kept it not, and therefore she is lost' (S&P, I.iv.123), which matches Thurio's line, 'I claim her not, and therefore she is thine' (TGV, V.iv.133). 49 However, other lines provide cues for Erastus, such as Perseda's declarative, 'And all my former love is turnd to hate' (S&P, II.i.152), during the lovers' quarrel, which provides a unique tetragram (four-word sequence) match with Proteus's speech, when he renounces his love for Julia: 'So the remembrance of my former love / Is by a newer object quite forgotten' (TGV, II.iv.192-193).It is conceivable that Shakespeare, if he played Erastus, noted the formulation, 'The sweet glances of', while giving 'half an ear to what' was 'being spoken on stage', 51 for Basilisco's line, 'The sweet glances of many amorous girles' (S&P, I.iii.129), is spoken shortly before Erastus enters.Valentine repeats this four-word unit at the beginning of Shakespeare's play: 'To the sweet glances of thy honoured love' (TGV, I.i.4).All of the remaining unique word sequences occur shortly before Erastus enters, which would seem to support Southworth's claim.However, Erastus is the male protagonist of Kyd's tragedy and is therefore on stage for much of the play.The evidence suggests that Shakespeare could have played Erastus, but it is hardly surprising that many of these repeated phrases can be found in scenes during which Erastus is present.

III. Shakespeare's Verbal Indebtedness to Kyd's Contested Plays
In a general essay published in the Times Literary Supplement in 2008, Brian Vickers argued for a new Kyd canon, ascribing to him-along with the traditionally accepted Kyd plays, The Spanish Tragedy, Soliman and Perseda, and Cornelia (1594)-King Leir, Arden of Faversham, Fair Em, the Miller's Daughter of Manchester (1590), and parts of Henry VI Part One (1592) and Edward III (1593). 52In this essay, Vickers examined authorial self-repetition seen in Kyd's use of n-grams, using evidence produced by modern anti-plagiarism software.
Conversely, in his 2014 monograph Determining the Shakespeare Canon: Arden of Faversham & A Lover's Complaint, MacDonald P. Jackson argued for Shakespeare's authorship of scenes Four to Nine (the middle section of the play, or Act Three in older editions) of Arden of Faversham. 53Jackson placed much emphasis on Arthur F. Kinney's conclusion, based on function-word and lexical-word tests, that the play shows 'no sustained affinities with Kyd' but that 'Shakespeare was one of the authors; and his part is concentrated in the middle section of the play'. 54Jackson therefore ascribed the 51  remainder of the play to an unknown co-author who was probably not Kyd.He claimed that there was a 'disparity' between the large number of verbal matches with Shakespeare in the middle portion of the play, as opposed to the remainder of Arden of Faversham. 55However, Jackson's 'underlying assumptions regarding the general distribution of linguistic items' do not appear to be supported by the data he has provided elsewhere. 56n his 2008 paper, 'New Research on the Dramatic Canon of Thomas Kyd', Jackson accused Vickers of 'entering Kyd into a one horse race, which Kyd cannot fail to win'. 57e aimed 'not to argue for Shakespeare's authorship, in whole or in part', but 'to demonstrate the inadequacy of Vickers's case for expanding the dramatic canon of Thomas Kyd'. 58 However, given that his replication of Vickers's method (using objective plagiarism software) was far more reliable than the laborious method of collecting parallels utilized in his monograph, I reproduce some of his results.Jackson was able to detect more unique three-word sequences between the domestic tragedy and Shakespeare's Henry VI Part Two and The Taming of the Shrew than Vickers had (at that time) discovered with Kyd's plays. 59Jackson lists (by my count) forty unique trigrams (three-word sequences) between Henry VI Part Two and scenes that he does not attribute to Shakespeare in Arden of Faversham.He lists only ten verbal matches between Shakespeare's play and the middle portion of the domestic tragedy. 60He also lists thirty-eight matches between The Taming of the Shrew and scenes outside of the middle portion of Arden of Faversham, with just six matches between Shakespeare's comedy and the portions he ascribes to Shakespeare. 61If we take Jackson's figures and adjust them to composite word counts, we are given an average of 0.03 matches between scenes Four to Nine of Arden of Faversham and Henry VI Part Two (combining the overall word count for these scenes in Arden of Faversham with the total word count for Shakespeare's play gives us a total of 30972) and 0.09 matches with the 'non-Shakespeare' scenes (which give us a composite count of 42480 words).Similarly, The 55 MacDonald P. Jackson, Determining the Shakespeare Canon: Arden of Faversham and A Lover's Complaint (Oxford: Oxford University Press, 2014), p. 65.For further discussion on Jackson's methodology for recording parallels in his monograph (in which he abandons the use of anti-plagiarism software), and his arguments concerning the distribution of verbal links with authorship candidates in Arden of Faversham, see Darren Freebury-Jones, '"A raven for a dove": Kyd, Shakespeare, and   Drama, 47 (2008), 107-127 (p.108). 58See Jackson,'New Research',p. 119. 59 In his 2008 essay, Vickers provided crude counts for triples.For example, he recorded thirty-one matches between the three traditionally accepted Kyd plays and King Leir.In a paper given at Somerville College, Oxford, on 16 May 2015, he listed a far more impressive ninety-six matches between King Leir and The Spanish Tragedy alone. 60 See Jackson,'New Research',  See Jackson, 'New Research', pp.115-116.  Ting of the Shrew averages 0.02 matches with the middle portion of the play (with a combined total of 26720 words) and 0.10 with the 'non-Shakespeare' scenes (38228 words).The overall pattern of unique matches does not support Jackson and Kinney's argument that Shakespeare's contribution is concentrated in the middle portion of Arden of Faversham.John P. A. Ioannidis notes that researchers sometimes fail 'to notice statistically significant relationships', or perhaps, in this case, 'there may be conflicts of interest that tend to "bury" significant findings'.62 The distribution of parallels would seem to suggest that either Shakespeare's hand can be found throughout Arden of Faversham, or that Shakespeare was very familiar with the domestic tragedy, and thus borrowed verbal details from the play as a whole.
Arden of Faversham was first entered in the Stationers' Register on 3 April 1592.It was published in Quarto that same year by Edward White.Jackson acknowledges that 'no Shakespeare play' seems to have been 'written before Arden of Faversham'. 63onetheless, Jackson considers it unlikely that Shakespeare 'imbibed' another 'playwright's words through hearing them' during performance. 64He queries: 'Even if' Shakespeare '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?' 65 We might ask ourselves: did Arden of Faversham really exert a greater influence over Shakespeare's dramatic language than other plays assigned to Kyd by Vickers?While pursuing this line of inquiry, we should remember that Soliman and Perseda is 'generally assumed to have been printed not long after it was entered in the Stationers' Register in November 1592', 66 and acknowledge that although King Leir was entered in the Stationers' Register in 1594, 'no copy of it seems to have been actually printed until more than a decade later'. 67Here I examine the relationship between Shakespeare's drama and the latter play. 68ueller contends that King Leir 'belongs to a very small set of stories to which Shakespeare returned again and again throughout his career', and that '[w]ithout The True Chronicle Historie we would not have King Lear or As You Like It, while Richard III, The Merchant of Venice, and Hamlet would be quite different plays.From such a perspective The True Chronicle Historie emerges as a play with a remarkably consequential career'. 69Philip Henslowe's records show that the play was performed in April 1594 at the Rose Theatre, by the 'Quenes men & my lord of Susexe to geather'. 70ichard Knowles argues that 'there being no clear evidence that Shakespeare ever was a Queen's Man, and some reason to think otherwise, there is accordingly no reason to think that he ever acted in Leir'. 71However, plays frequently passed into the repertoires of different companies, and Shakespeare and Peele's Titus Andronicus (as stated on the title page of the First Quarto) is known to have been performed by Pembroke's Men, Derby's Men, and Sussex's Men.I propose that Sussex's Men acquired the play from Pembroke's, 72 and were therefore able to perform King Leir in conjunction with Queen's Men, 73 just as they were 'able to play "Titus & ondronicus"' when Pembroke's Men collapsed and the play 'became temporarily derelict'. 74Notably, David George suggested in 1981 that 'Sussex's Men were willing' to 'help Pembroke's all they could' in 1594, and 'probably Pembroke's Men were trying to raise capital for one more try at independent acting' by selling some of their plays, 75 while Lene B. Petersen observes that 'Queen's Men plays' often recurred 'quasi-simultaneously in other companies' repertoires'. 76hether Shakespeare had acted in King Leir or not, Knowles's argument that 'the evidence for Leir's influence on Shakespeare's early plays is small at best and illusory at worst' is more than countered by the data contained in Mueller's Excel document. 77ueller's automated results reveal that Shakespeare shares a large number of unique word sequences with the old play, which suggests that King Leir exerted a considerable influence over Shakespeare's dramatic language.As Mueller put it in 1994: probability arguments are often subject to dispute, and source or allusion hunters are frequently criticized for deriving strong claims from weak resemblances, but sceptics are apt to underestimate the consequences of the fact that probabilities for random occurrence drop precipitously as soon as even a few independently very common features recur in combination. 78ere are ten unique n-grams shared between King Leir and Henry VI Part Three; eight with Richard III (1593); eight with King John; eight with Henry IV Part One (1597); and seven with Much Ado About Nothing (1598). 79I focus here on unique n-grams shared between the old play and Shakespeare's Richard III.Some of these parallels can be found in lines delivered by the (would-be) murderous Messenger.He tells Leir and Perillus: Feare nothing, man, thou art but in a dreame, And thou shalt never wake untill doomes day. 80 Richard III, the Second Murderer prepares to kill Clarence.He reassures his companion: 'Why, he shall never wake until the great judgement day' (R3, I.iv.100).There appears to be some substance in Meredith Skura's argument that Scene Nineteen of King Leir served as a 'source for the murder of Clarence'. 81The characters and plot situations could hardly more alike.Other recurring n-grams, however, are contextually dissimilar.The Messenger tells Ragan: I weigh no more the murdring of a man, Then I respect the cracking of a Flea.(KL, xv.1214-1215)Richard tells Prince Edward: Nor more can you distinguish of a man Than of his outward show.(R3, III.i.9-10) The only similarity here, apart from the placing of this four-word unit in the verse lines, is that both characters are villains (though we might also note that this matching phrase embraces the two-word sequence 'no/r more').Other word sequences are also used in different contexts.The remorseful Leir tells Perillus: It may be, if I should to her repayre, She would be kinder, and intreat me fayre.(KL, x.919-920) 78 Mueller, 'From Leir to Lear ', p. 197. 79 https://scalablereading.northwestern.edu/?p=312 80 King Leir, xix.1616-1617, in The History of King Leir, ed.W. W. Greg (Oxford: Malone Society Reprints,  1907).All further references are to this edition and will be given parenthetically. 81Meredith Skura, Shakespeare the Actor and the Purposes of Playing (Chicago: The University of Chicago Press, 1993), p. 285.
Shakespeare draws on (by my argument) Kyd's phraseology for the moment when Richard threatens Queen Elizabeth: Either be patient and entreat me fair, Or with the clamorous report of war Thus will I drown your exclamations.(R3, IV.iv.152-154)Some of these n-grams give us a possible insight into Shakespeare's associative memory, while other unique sequences suggest that Shakespeare was so familiar with the verbal fabric of the play that he could retrieve phrases irrespective of context.As is the case with Kyd's Soliman and Perseda, the parallels are not limited to a single character's lines or cue-lines.They suggest an intimate familiarity with the text as a whole.
John Jones argues that a study of 'so-called reported texts confirms one's common sense expectation that having been on-stage fortifies the memory of the reporter.(An actor's recall of his own lines is obviously better again)'. 82However, it would be erroneous to group Shakespeare with actor-reporters when investigating memorial repetitions.Shakespeare was not an actor-reporter attempting to reconstruct whole scenes or speeches.If he had indeed performed in Soliman and Perseda and/or King Leir, he is likely to have had a 'general memory for the whole performance' as an actorturned-dramatist, which enabled him to repeat phrases both consciously and unconsciously. 83Moreover, as Baldwin suggested, 'As the play was being planned, constructed, and fitted, he would at least hear, and would doubtless participate in, the discussions which arose between author and actors'. 84Unfortunately, there are simply not enough known actor-dramatists during this period for one to conduct a systematic study of their patterns of verbal borrowing.Vickers notes that 'Apart from Robert Wilson, with his rather crude morality plays, and Richard Tarlton, with his extemporal jests, Shakespeare was the only dramatist in the early 1590s who was also an actor'. 85e would have to turn to Ben Jonson for a comparable example of a successful actorturned-dramatist's verbal recall.Judging by the data contained in Mueller's Excel document, Jonson also seems to have had a remarkable retentive memory of serviceable phrases within the theatrical vernacular of his time. 86If Shakespeare had been able to borrow phrases from The Spanish Tragedy, Soliman and Perseda, and King Leir through having read them, he must have somehow acquired copies of these plays prior to publication.In my view, it seems more likely that such repetitions are the products of Shakespeare's aural memory, and that he had either seen the plays during performance  84 Baldwin, Literary Genetics, p. 55. 85 Vickers, 'The Quarrel Scene', p. 6. 86 We might also add Thomas Heywood to the equation.Heywood seems to have begun his acting and writing career with the Admiral's Men; he (like Jonson) shares a large number of matches with a great many plays written by other dramatists according to Mueller's document.
or had played in them.Readers might, however, be surprised to discover that King Leir and Shakespeare's King Lear (1605) do not make Mueller's list of plays sharing large numbers of unique matches.
In 1933, Joseph Quincy Adams argued that the publication of the 1605 Quarto of the old play was a deliberate attempt to take advantage of Shakespeare's tragedy.He noted that the title under which the old play was entered in the Stationers' Register, in 1605, was The Tragicall historie of Kinge Leir, whereas the play was really a comedy with a very happy ending, and no writer handling the Lear story had ever given it a tragic conclusion until Shakespeare put on the boards his entirely altered version. 87e entry of Shakespeare's tragedy in the Stationers' Register on 26 November 1607 appears to make an especial effort to avoid confusion with the old play: 'Master William Shakespeare his historye of Kinge Lear as yt was played before the kinges majestie at Whitehall'. 88W. W. Greg asserted in 1940 that 'I do not think there can be any doubt that the prominence given to the author's name on the title-page' was 'due to a desire to distinguish the piece as clearly as possible from its predecessor'. 89Greg also suggested that the King Leir 'manuscript which Stafford acquired and printed in 1605' had 'presumably remained for the eleven intervening years in the hands of stationers'. 90evertheless, Shakespeare recalls a sufficient number of the play's details to suggest that he either saw or acted in it, and 'it would seem that as' Shakespeare 'wrote, ideas, phrases, cadences from the old play still floated in his memory below the level of conscious thought'. 91(Lr., 87 Joseph Quincy Adams, 'The Quarto of King Lear and Shorthand', Modern Philology, 31.2 (1933), 135-163  (p.136  20 (1940), 377-400 (p.381).90 Greg, 'The Date of King Lear', p. 385. 91Greg, 'The Date of King Lear', p. 397.92 I have extracted these matches from Marcus Dahl's unpublished document, 'Lear vs Leir parallels'.I wish to thank Dahl for giving me permission to reproduce his findings.
Here we see a somewhat tenuous contextual correlation triggering the same combination of words.In the old play, Perillus tells the King, 'I had ynough, my Lord, and having that, / What should you need to give me any more?' (KL, x.890-891), while Regan spites Shakespeare's Lear: 'I dare avouch it, sir.What, fifty followers?/ Is it not well?What should you need of more?' (Lr., II.ii.410-411).The character relationships in these examples are markedly different, for Perillus loves Leir and is content with his station in life, while Regan is a schemer who denies Lear his (now) fifty followers.Such verbal echoes are perhaps, as Skura puts it, 'accidental, like a tune that you hear and find yourself helplessly singing over and over'. 93Other passages from King Leir seem to have persisted in Shakespeare's long-term memory, for he repeats the sentiment of Leir's line, 'And think me but the shaddow of my selfe' (KL, xiv.1111), in the exchange between Lear and the Fool in Act One Scene Four.Lear asks, 'Who is it that can tell me who I am?' (Lr., I.iv.212), to which the Fool responds: 'Lear's shadow' (I.iv.213).As we can see here, Shakespeare's verbal indebtedness to the old play in King Lear is hardly as prevalent as in his early works, such as Henry VI Part Three and Richard III.I now return to Arden of Faversham to show that the verbal evidence for Shakespeare's part authorship of the domestic tragedy seems weak in light of Shakespeare's relationship with plays that scholarly consensus assigns to Kyd, as well as the contested text, King Leir.
Cairncross argued that Arden of Faversham belonged to the repertory of Pembroke's Men, for whom Shakespeare perhaps began his career as an actordramatist. 94Nonetheless, Jackson argues that Shakespeare parallels contained in the middle portion of Arden of Faversham must be authorial, for Shakespeare 'could not have played a role in both' Scene Six and Scene Eight of the domestic tragedy. 95owever, as we have seen in the cases of Soliman and Perseda and King Leir, Shakespeare's verbal borrowings from plays attributable to Kyd are not confined to scenes in which he may have acted.In fact, Mueller's data strongly suggests that Shakespeare was no more influenced by Arden of Faversham than he was by other plays in Vickers's 'extended' Kyd canon.I reproduce Mueller's results for unique tetragrams in Table 2 below: 93 Meredith Skura, 'What Shakespeare Did with the Queen's Men's King Leir and When', Shakespeare Survey,  63 (2010), 316-325 (p.316). 94See Cairncross, 'Some Shakespearean Piracies', 335-349. 95Jackson, Determining, p. 122.The plays with the most pervasive influence on Shakespeare's dramatic language in this table are The Spanish Tragedy and King Leir, not Arden of Faversham.It seems fair to say that if we were not confident in Kyd's authorship of The Spanish Tragedy and Soliman and Perseda, or if we were to collect parallels with King Leir, throw in some impressionistic evaluations of certain passages, along with some misleading data drawn from assumptions about the distribution of linguistic items, it would not be too difficult to provide a superficially impressive case for Shakespeare's part authorship of these texts.
The Shakespeare play with the most matches with Arden of Faversham, according to Mueller's document, is Richard III, which William Wells regarded as 'a study in Kydian methods'. 96These plays share eight unique n-grams in total. 97Six of these n-grams occur in scenes that Jackson does not ascribe to Shakespeare.The evidence therefore supports the theory that Shakespeare appropriated phrases from a play that, like Soliman and Perseda and King Leir, seems to have antedated his whole corpus.The next Shakespeare play with the most unique links with Arden of Faversham, according to Mueller's corpus, is The Merchant of Venice (1597), with seven matches in total.Only one of these n-grams features in a scene Jackson assigns to Shakespeare.The third and final Shakespeare play with a high number of unique matches in Mueller's corpus is Troilus and Cressida (1602), which also shares seven n-grams of four or more words, four of which occur in scenes that Jackson does not attribute to Shakespeare.
Mueller's data also conflicts with Jackson's argument that the accepted Kyd plays are like each other and 'the putatively Kydian plays' are not. 98When Arden of Faversham is tested against over 500 plays for unique n-grams of four or more words (tetragrams are statistically rarer than trigrams in early modern drama), the play with the most unique matches is Kyd's Soliman and Perseda, with a total of eighteen.Mueller has noted elsewhere that 'the odds of getting between 10 and 15 shared tetra-or pentagrams in a random draw are on the order of 1:10,000'. 99Kyd's The Spanish Tragedy and Soliman and Perseda share eight unique n-grams of four or more words, which corroborates with Mueller's observation that 'on average plays by the same author share five dislegomena, and the median is four.Roughly speaking, plays by the same author are likely to share twice as many dislegomena as plays by different authors'. 100Similarly, Mueller's document reveals that there are eight unique n-grams of four or more words shared between King Leir and Kyd's Soliman and Perseda, which is the same total we find for Kyd's accepted tragedies.Arden of Faversham and King Leir share eleven unique n-grams, which also provides compelling evidence for common authorship. 101Contrary to Jackson's criticisms of Vickers's parallel-checking methodology, we cannot suppose that Kyd has been entered into 'a one horse race' here. 102Furthermore, as I have shown elsewhere, Mueller's database demonstrates that the plays Vickers assigns to Kyd (as well as the traditionally accepted Kyd plays) lead the race in terms of the frequency of short strings of words, longer collocations, common phrases, and rare/unique phrases shared with Arden of Faversham. 103In short, Jackson's argument in The New Oxford Shakespeare: Authorship Companion that '[i]n the two-horse race, Shakespeare beats Kyd' appears to be erroneous. 104 Over a decade later, Alfred Hart provided what remains the most comprehensive examination of unauthorized texts such as The first part of the Contention of the two famous houses of Yorke and Lancaster, with the death of the good duke Humphrey (1594) and The true Tragedie of Richard Duke of Yorke and the death of good King Henry the Sixt

82
John Jones, Shakespeare at Work (New York: Oxford University Press, 1995), p. 43. 83Margrethe Jolly, The First Two Quartos of Hamlet: A New View of the Origins and Relationship of the Texts (Jefferson, NC: McFarland and Company, 2014), p. 106.

Table 1 :
Southworth, Player, p. 41.  52Brian Vickers, 'Thomas Kyd, Secret Sharer', Times Literary Supplement, 13 April 2008, 13-15.  5ckson first argued for Shakespeare's hand in this play in 1963.See MacDonald P. Jackson, 'Material for an edition of Arden of Faversham' (B.Litt.thesis: Oxford University, 1963). 54Arthur F. Kinney, 'Authoring Arden of Faversham', in Shakespeare, Computers, and the Mystery of Authorship, eds.Hugh Craig and Arthur F. Kinney (Cambridge: Cambridge University Press, 2009), pp.78-99 (p.99).Jackson does not acknowledge the conflicting findings of other teams, such as those of Lene Buhl Petersen and Marcus Dahl, who have used similar quantitative analysis of function and lexical words.For example, Petersen has applied 'discriminant analysis' to 'principal data components' with 'cross-validation'.According to Petersen's use of Principal Component Analysis, 'Arden of Faversham cross-validates as Kyd'.Unlike Kinney, Petersen concludes, sensibly, that 'these classifications are by no means to be taken as truths'.See Lene Buhl Petersen, Shakespeare's Errant Texts: Textual Form and Linguistic Style in Shakespearean 'Bad' Quartos and Co-authored Plays (Cambridge: Cambridge University Press, 2010), pp.213-214.
the Authorship of Arden of Faversham's Quarrel Scene', Archiv fuer das Studium der Neueren Sprachen und Literaturen, 253.1 (2016), 39-64; '"Fearful Dreams" in Thomas Kyd's Restored Canon' (forthcoming); 'In Defence of Kyd: Evaluating the Claim for Shakespeare's Part Authorship of Arden of Faversham' (forthcoming).For a careful rebuttal of Jackson and his New Oxford Shakespeare colleagues' more recent arguments, based on computational stylistic tests, see Freebury-Jones, 'Augean Stables' (forthcoming). 56Sabine Bartsch, Structural and Functional Properties of Collocations in English (Tübingen: Narr, 2004), p. 103. 57MacDonald P. Jackson, 'New Research on the Dramatic Canon of Thomas Kyd', Research Opportunities in Medieval and Renaissance ). 88 Geoffrey Bullough, Narrative and Dramatic Sources of Shakespeare: Volume IV, 8 vols (London: Routledge, 1973), IV.269. 89W. W. Greg, 'The Date of King Lear and Shakespeare's Use of Earlier Versions of the Story', The Library,