Introduction
The accepted date for Macbeth is 1606. This date, first proposed by the 18th C scholar and editor Edmund Malone, has never been seriously challenged. Yet the earliest surviving text of Macbeth is that of the 1623 First Folio, published seven years after Shakespeare's death. There is good reason to believe that the text did not remain stable in the seventeen years between first performance and first publication, but underwent some revision. Most scholars agree that the revisions were undertaken not by Shakespeare himself, but by Thomas Middleton. Both external and internal evidence support this. External evidence consists in the conditions of authorship and performance in the Jacobean era; internal evidence can be found within the play's text itself.
External evidence that Macbeth is not entirely Shakespeare's work
We tend to assume that a writer is a solitary genius, relying on imagination to create works that are largely self-contained. However, this belief is at odds with the realities of theatrical production and performance in Shakespeare's day. Playwriting was largely a collaborative enterprise. When a company wanted to put on a new play or revive an old one, a writer or a team of writers was assigned to come up with a script. Scripts for new plays typically retold a story already in circulation. Scripts for revivals augmented elements that had proven popular in previous stagings, removed less successful elements, and updated any topical references. Gary Taylor points out that Shakespeare would not be excluded from this process of having his scripts adapted:
Theater historians have demonstrated that adaptation was, like collaboration, a routine practice in the early modern theater industry. Adaptation was particularly likely to affect plays that remained in the theatrical repertoire. Because Shakespeare produced more staples of that repertoire than any other playwright, we would certainly predict that at least some of his plays were adapted. (p. 241)
We can see this process at work by comparing two texts of Hamlet: the Second Quarto of 1604 (Q2), and the First Folio of 1623 (F). Act II scene ii in the Folio text has some 15 prose lines where Rosencrantz and Guildenstern explain to Hamlet why the Players are down on their luck:
HAMLET How comes it? Do they grow rusty?
ROSENCRANTZ Nay, their endeavor keeps in the wonted pace. But there is, sir, an aerie of children, little eyases, that cry out on the top of question and are
most tyrannically clapped for ’t. These are now the fashion and so berattle the common stages (so they call them) that many wearing rapiers are afraid of goose quills and dare scarce come thither.
HAMLET What, are they children? Who maintains ’em? How are they escoted? Will they pursue the quality no longer than they can sing? Will they not say afterwards, if they should grow themselves to common players (as it is most like, if their means are no better), their writers do them wrong to make them exclaim against their own succession?
ROSENCRANTZ Faith, there has been much to-do on both sides, and the nation holds it no sin to tar them to controversy. There was for a while no money bid for argument unless the poet and the player went to cuffs in the question.
HAMLET Is ’t possible?
GUILDENSTERN O, there has been much throwing about of brains.
HAMLET Do the boys carry it away?
ROSENCRANTZ Ay, that they do, my lord—Hercules and his load too. (II.ii.360–385)
As this answer to a different question describes, this passage, alluding to the new fashion for child actors that arose in 1600, is important in dating Hamlet to 1601. Yet the passage is cut from the Q2 text of 1604, and is found only in F some nineteen years later. Harold Jenkins has convincingly argued that this passage is not a late addition to the F text, but an excision from Q2 (pp. 2–3, 470–473).
The presence of this passage in F and its absence from Q2 reveals much about the textual transmission of Shakespeare's plays generally. It is unlikely that Shakespeare prepared any of his plays for publication, and his plays have not come down to us in stable, final forms. Plays were printed not from a final manuscript put together by the dramatist, but from whichever acting script was at hand. It seems that with Hamlet, a commentary on a 1600 innovation was out of date by 1604, and therefore excised; the Q2 version of the play therefore omits the passage. But when Shakespeare's friends collected his plays and for posthumous publication, the script they used was one that included the passage, so it found its way into F.
Who was tasked with making such changes to the acting version of a play? Then as now, the task of preparing a script for publication tended to be ad hoc. Scenes were cut, rewritten, or even added based on strictly expedient concerns: the topicality or popularity of a certain theme, the availability of certain actors, the amount of time a performance would take, the construction of the theatre where the play was to be staged, etc. None of this would be surprising to any working actor even today. The idea that a playscript is somehow sacrosanct—that altering lines or deleting entire scenes is a violation of the integrity of a work of art—would not have occurred to performers in Shakespeare's day any more than it does to theatre hands today. So when changes were made to any of Shakespeare's plays as they were being readied for performance, there is no reason to assume that Shakespeare alone made them.
Even though we have just one text of Macbeth, then, we cannot assume that the First Folio faithfully reproduces just what Shakespeare wrote. It merely represents one of the many possible acting scripts of the play that could have existed at the time. Macbeth is also among the shortest of Shakespeare's plays, suggesting that it might represent a trimmed playscript. Nicholas Brooke says:
The text is a good one, overall, which needs very little verbal emendation, and since it is well supplied with stage directions it is usually thought to derive from the theatre prompt-book, probably via a copy omitting any passages that had been cut in performance. (p. 49)
The conditions under which plays circulated in Shakespeare's day, both in print and onstage, thus make it unlikely that the Folio text of Macbeth reproduces exactly and only what Shakespeare wrote.
Internal evidence for Macbeth's being a collaborative play
The text shows internal evidence of non-Shakespearean revisions as well. Brooke points to various problems with the Folio text of Macbeth, particularly between Act III scene v and Act IV scene i:
What distinguishes 3.6 and 4.1 is that they cannot be performed as they stand, and so with 3.5 form a unit that differs from any other part of the play.... It seems as though the material available for those three scenes consisted of a revision not finally tidied up for performance. (p. 52)
The most noticeable sign that Macbeth was altered in some way is a glaring inconsistency between III.vi and IV.i. In the earlier scene, Lennox reports to an unnamed lord that Macduff is in Macbeth's bad books for speaking too plainly and refusing to attend the latter's feast. Lennox then asks the lord if he knows where Macduff is hiding. The lord replies that Macduff has gone to England, where Duncan's son has taken refuge with the English king. The following dialogue ensues:
LORD And this report
Hath so exasperate the King that he
Prepares for some attempt of war.
LENNOX Sent he to Macduff?
LORD He did, and with an absolute “Sir, not I,”
The cloudy messenger turns me his back
And hums, as who should say “You’ll rue the time
That clogs me with this answer.” (III.vi.41–49)
These lines make clear that Macbeth has sent for Macduff; that Macduff has refused the summons, and has fled for England to join forces with Malcolm; and that knowing this, Macbeth is preparing for war. But confusingly, in IV.i, Macbeth is shocked when Lennox tells him that Macduff is in England:
LENNOX Macduff is fled to England.
MACBETH Fled to England?
LENNOX Ay, my good lord.
MACBETH (aside) Time, thou anticipat’st my dread exploits.
The flighty purpose never is o’ertook
Unless the deed go with it. (IV.i.161–165)
This contradiction suggests that at some point, some scenes of the play were rearranged, and that the Folio text is based upon some manuscript where that rearrangement was underway, but not yet finalized.
Another clue to the not yet finalized nature of the text is the presence of two songs indicated only by their first lines. In III.v, the scene where Hecate first appears onstage, she exits to a song:
(Music and a song)
Hark! I am called. My little spirit, see,
Sits in a foggy cloud and stays for me. (Hecate exits.
Sing within “Come away, come away,” etc.) (III.v.sd, 34–35)
Likewise, in IV.i, a stage direction indicates another song marking Hecate's exit from the stage:
(Music and a song: “Black Spirits,” etc. Hecate exits) (IV.i.sd after line 43)
To be entirely accurate, the Folio text does not actually say Hecate exits here. However, the editorial emendation in the text at the Folger website makes sense; Hecate has no more lines in a scene that continues for considerably longer.
That "Come away, come away" and "Black Spirits" are indicated only by their opening words is extremely unusual. Every other play in the First Folio includes songs in full. Brooke says that such contradictions and omissions in the text
could quite easily be the result of a late decision to alter this part of the play, chiefly no doubt to introduce Hecate, but not exclusively for that. (p. 52)
Brooke's speculation here reflects the scholarly consensus, which accepts that the scenes from III.v to IV.i inclusive represent a revision of a prior version that did not include Hecate. And the likeliest candidate for the dramatist who worked to revise the play script is Thomas Middleton.
Thomas Middleton as the collaborator
Like Shakespeare, Middleton was a consummate theatre professional. He is known to have written portions of at least two other plays that are largely Shakespeare's, Measure for Measure (1604) and Timon of Athens (1606). His role in adapting some scenes of Macbeth has been known since the 18th C, largely due to the overlap between one of his plays, The Witch, and the Hecate scenes in Macbeth.
The Witch, probably written around 1613–1616, was first printed only in 1778. The same year, Edmund Malone and George Steevens pointed out that the witch scenes in Macbeth were derivative of Middleton's play. Malone's essay "An Attempt to Ascertain the Order in which the Plays Attributed to Shakspeare were Written" was part of Steevens and Samuel Johnson's magisterial 1778 edition of Shakespeare's complete works. In a footnote to that essay, Steevens noted many verbal similarities between Shakespeare's play and Middleton's. In the body of the essay, Malone himself argued that Middleton's play must have preceded Shakespeare's:
Mr. Steevens has lately diſcovered a Mſ. play, entitled THE WITCH, written by Thomas Middleton, which renders it queſtionable, whether Shakſpeare was not indebted to that author for the firſt hint of the magic introduced in this tragedy. The reader will find an account of this ſingular curiosity in the note [superscript numeral indicating Steevens' footnote].—To the observations of Mr. Steevens I have only to add, that the ſongs, beginning, Come away, &c. and Black ſpirits, &c. being found at full length in The Witch, while only two firſt words of them are printed in Macbeth, favour the ſupposition that Middleton's piece preceded that of Shakſpeare; the latter, it ſhould ſeem, thinking it unneceſſary to ſet down verſes which were probably well known, and perhaps then in the poſſeſſion of the managers of the Globe theatre. The high reputation of Shakſpeare's performances (to mention a circumſtance which in the course of theſe observations will be more than once inſiſted upon) likewise ſtrengthens this conjecture; for it is very improbable, that Middleton, or any other poet of that time, ſhould have ventured into thoſe regions of fiction, in which our author had already expatiated. (pp. 324–330; most of each page is taken up by Steevens' lengthy footnote)
From the time that The Witch was rediscovered, then, scholars have accepted that Macbeth includes contributions from Middleton. The full versions of "Come away, come away" and "Black spirits" being found in The Witch is a compelling piece of evidence in this regard. But other parts of the play have been attributed to Middleton as well. Tsundoku's answer has gone over the various attempts to attribute specific passages and scenes in Macbeth to Middleton, and the shifting arguments over the years about the extent of the contribution. Until fairly recently, those arguments were based on scholarly judgments regarding verbal correspondences. R. V. Holdsworth, for example, notes that stage directions of the form Enter X meeting Y are not unusual in Middleton, but rare in Shakespeare's plays (Brooke p. 58). In Macbeth, that formulation occurs in I.ii and III.v. Both of those scenes have been attributed to Middleton by other scholars as well.
Instead of relying on scholarly judgments alone, more recent stylometric approaches use computer-aided analysis to confirm or rebut attributions of the sort at issue here. In 2014, Gary Taylor compared parts of IV.i and all of III.v across a wide corpus of Elizabethan and Jacobean drama found in the databases Literature Online (LION), Oxford Scholarly Editions Online (OSEO), and Early English Books Online-Text Creation Partnership (EEBO-TCP). His paper is in part a response to the paper by Markus Dahl, Marina Tarlinskaya, and Brian Vickers that Tsundoku discusses in his answer Those scholars contend that Macbeth is entirely Shakespeare's work. Taylor, however, argues on the basis of his computer-aided analysis that III.v and two passages from IV.i are far more likely Middleton's than Shakespeare's. The passages from IV.i are these:
HECATE O, well done! I commend your pains,
And everyone shall share i’ th’ gains.
And now about the cauldron sing
Like elves and fairies in a ring,
Enchanting all that you put in. (ll. 39–43)
FIRST WITCH Ay, sir, all this is so. But why
Stands Macbeth thus amazedly?
Come, sisters, cheer we up his sprites
And show the best of our delights.
I’ll charm the air to give a sound
While you perform your antic round,
That this great king may kindly say
Our duties did his welcome pay. (ll. 141–148)
Taylor's paper does not consider other passages in Macbeth that have been ascribed to Middleton at various times. Even I.ii, a point of contention discussed at length in the Dahl, Tarlinskaya, and Vickers paper, is ignored. In 2020, Vickers published a rebuttal of Taylor's paper in the Times Literary Supplement. I am not able to read that paywalled essay; however, Vickers and his co-authors are outliers in insisting upon the integrity of Macbeth as entirely Shakespeare's. Most scholars and editors accept that the play shows Middleton's hand. Disagreements remain over exactly which passages are to be attributed to Middleton, but that he was involved in revising the play is rarely disputed.
The exact date of this revision is also disputed. Brooke claims, somewhat unusually, that Middleton revised Macbeth some years before he wrote The Witch:
I suggest that Macbeth was revised in 1609–10, and The Witch written in 1615. (p. 66)
Two factors work against this claim. First, Middleton is not known to have written regularly for Shakespeare's company, the King's Men, until 1615 or so. Second, since The Witch too is dated 1615 or 1616, it seems unlikely that Middleton would incorporate songs from a successful play such as Macbeth into it. By contrast, The Witch itself was not a success. Middleton, in a prefatory note, called it "this ignorantly ill-fated labour of mine":
Witches are, ipso facto, by the law condemn'd, and it only, I think, hath made her lie so long in an imprison'd obscurity.
By the law condemn'd suggests that The Witch was censored rather than disliked by audiences. The play is topical, dealing allusively with the divorce of the Earl of Essex, in which King James took an interest. James's own aversion to witchcraft as well as the touchy subject matter probably got the play banned. The majority opinion, that Middleton used material from his "ill-fated" play in his revision of Macbeth, seems more likely.
What led to Macbeth's being adapted by Middleton?
The question remains: what were the specific circumstances under which Macbeth was adapted, and why was Middleton a good choice for the adaptation? One reason was probably to highlight the witchcraft elements. A. R. Braunmuller writes:
In early modern England, no less than in many societies today, sensational events—accusations of witchcraft, gruesome murders, and the like—were fodder for every available medium (at that time, drama, pamphlet, and ballad), or, as one later contemporary wrote, such events were "Staged, Book'd, and Balleted." (p. 258; the internal quotation is from Thomas Taylor, p. 94)
Interest in witchcraft had increased since James VI of Scotland succeeded to the English throne as James I of England after Queen Elizabeth's death in 1601. James himself believed that storms his ship had encountered when he was returning to Scotland after marrying Princess Anne of Denmark were the result of witchcraft. Several Danish women were tried and convicted of casting spells on her ship. James instituted similar trials in Scotland as well, several alleged witches being burned at the stake as a result. Witchcraft was thus a topical subject for Macbeth. The North Berwick witch trials, and James's subsequent book Daemonologie (1597), are important sources for the play.
Presumably the Weïrd Sisters had proved popular in Macbeth, and expanding their role would have been appealing to the spectators. Since Hecate is the goddess of witchcraft, introducing her was an obvious way of doing so. Macbeth himself has mentioned Hecate in his soliloquy as he approaches the sleeping Duncan:
Now o’er the one-half world
Nature seems dead, and wicked dreams abuse
The curtained sleep. Witchcraft celebrates
Pale Hecate’s off’rings, and withered murder,
Alarumed by his sentinel, the wolf,
Whose howl’s his watch, thus with his stealthy pace,
With Tarquin’s ravishing strides, towards his design
Moves like a ghost. (II.i.61–68)
A second reason was that Macbeth, originally written for the Globe, was being performed in a new space. The King's Men had acquired a controlling interest in the Blackfriars in 1608, and began performing there after 1609. An indoor theatre, the Blackfriars had innovations not available at the outdoor Globe. Artificial lighting was one such; another was having onstage actors raised and lowered using equipment concealed in the flies. Having witches arrive and depart onstage by flying would not have been possible at the Globe; the Blackfriars opened up that possibility. Hecate's exits probably made use of this (Brooke 35). It has been suggested that the songs in III.v and IV.i were added to cover the noise of the mechanism used as Hecate flies off (Braunmuller 257).
Hecate's opening speech, where she complains about having been left out, is in fact a metatheatrical commentary both on her newly added role in Macbeth and on the possibilities of the new theatre. She upbraids the Weïrd Sisters:
Saucy and overbold, how did you dare
To trade and traffic with Macbeth
In riddles and affairs of death,
And I, the mistress of your charms,
The close contriver of all harms,
Was never called to bear my part
Or show the glory of our art? (III.v.3–9)
The previous version of Macbeth did not call upon Hecate to bear [her] part, which was newly written for the adaptation; nor did that version show the glory of the theatrical art, because that glory depended on the resources of the Blackfriars. When Macbeth was adapted for performance in that theatre, the new version added the role of Hecate in order to showcase the spectacle possible in the new space.
Middleton was ideally placed to undertake the adaptation. A freelancer who (unlike Shakespeare) was not tied to any one company, he had written The Witch for the King's Men. The suppression or failure of that play meant that he had material at hand to incorporate into Macbeth. Perhaps more important, he was adept at spectacle. Middleton had written several masques for the court of James I. These allegorical entertainments involved music, dancing, and acting, as well as elaborate stage design and costumes. As such, they were important in the development of opera, which brings together various arts in the same way.
Caroline Baird has shown that Middleton's familiarity with masque is key to understanding his plays:
Thomas Middleton's intimate knowledge of the Jacobean court masque enabled him to exploit its conventions, iconography, and structural functions in order to experiment innovatively with how masques, or masquing components, might be deployed in the playhouse. Although Middleton was not a prolific masquewright ... he wrote more masques into his plays than any of his contemporaries, and up to 1611 was the only dramatist to take the device outside the repertoire of the children's troupes. (p. 58)
Middleton's chief contribution to Macbeth appears to have been the insertion of masque-like elements. Gary Taylor comments:
Middleton's Hecate brings into Shakespeare's Scottish tragedy the verbal and visual semiotics of civic humanist London pageantry. (p. 263)
Middleton's contribution was tremendously successful, and Macbeth became associated with visual and musical spectacle. These spectacular elements were front and center in the revision by William Davenant, which ruled the stage during the Restoration era. Davenant had been with the King's Men in the 1630s. When London theaters reopened after the Puritan interregnum of 1642–1660, he staged his rewritten Macbeth, which Brooke describes as "a spoken verse drama with substantial musical interludes involving some at least of the main characters." Brooke quotes from the memoirs of John Downes, who was a prompter in Davenant's company:
The tragedy of Macbeth, altered by Sir William Davenant; being dressed to all its finery, as new clothes, new scenes, machines, as flyings for the witches; with all the singing and dancing in it; the first composed by Mr. Lock, the other by Mr. Channell and Mr. Joseph Preist; it being all excellently performed, being in the nature of an opera, it recompensed double the expense; it proves still a lasting play. (pp. 37–38)
Davenant's operatic Macbeth highlighted and augmented the masque elements that Middleton introduced into the play. It was not until 1744 that David Garrick put on a performance of Macbeth that went back to the Folio text (Braunmuller pp. 61–62; Stone, passim). While Davenant's redaction of Macbeth is lamentable, it does indicate the theatrical appeal of Middleton's contributions to Shakespeare's play.
Conclusion
The Folio text of Macbeth is carefully prepared, but appears to be set from a prompt-book that had not been fully revised. The text shows inconsistencies in various scenes, most noticeably in III.v–IV.i, and arguably in I.ii, that are the result of incomplete revision. Computer-assisted stylometric analysis of III.v as well as a couple of passages from IV.i suggests that the adaptation was probably by Middleton. The occasion for the adaptation was likely a revival of the play at the Blackfriars theatre, which as an indoor theatre offered the opportunity for greater spectacle.
The case for Middleton's hand in Macbeth is strengthened by the mention of songs from his play The Witch in a couple of stage directions, as well as by his experience as a masquewright. To my knowledge, a full stylometric analysis of all the disputed scenes and passages in the play has not yet been undertaken. This makes it hard to pinpoint exactly how much of, say, IV.i is Shakespeare's and how much Middleton's. However, with the notable exception of Brian Vickers and some allied with him, most scholars and editors agree that the text of Macbeth cannot be attributed entirely to Shakespeare; that III.v is entirely Middleton's; and that Middleton has contributed other passages, perhaps comprising entire scenes, to the play.
References (other than Wikipedia)
- Baird, Caroline. "From court to playhouse and back: Middleton's appropriation of the masque." Early Theatre, vol. 18 no. 2 (2015), pp. 57–85. centaur.reading.ac.uk. Accessed March 14, 2021.
- Braunmuller, A. R., ed. Macbeth by William Shakespeare. 1606. Cambridge, UK: Cambridge UP, 1997.
- Brooke, Nicholas, ed. The Tragedy of Macbeth by William Shakespeare. 1606. The Oxford Shakespeare. Oxford: Oxford UP, 1990.
- Dahl, Markus, Marina Tarlinskaya, and Brian Vickers. An Enquiry into Middleton’s supposed “adaptation” of Macbeth. 2010. thomasmiddleton.org. Accessed March 13, 2021.
- James VI of Scotland. Daemonologie (1597). Archive.org. Accessed March 14, 2021.
- Jenkins, Harold, ed. Hamlet by William Shakespeare. 1601. The Arden Shakespeare, second series. London: Methuen, 1982.
- Malone, Edmond. "An Attempt to Ascertain the Order in which the Plays Attributed to Shakspeare were Written”. In The Plays of William Shakespeare, vol. 1. Ed. Samuel Johnson and George Steevens. 2nd ed. London: Bathurst et al., 1778. pp. 261–346. Archive.org. Accessed March 11, 2021.
- Middleton, Thomas. The Witch. Chris Cleary's Middleton site at tech.org. Accessed March 14, 2021.
- Shakespeare, William. Hamlet. Folger Shakespeare Library. Accessed March 8, 2021.
- Shakespeare, William. Macbeth. Folger Shakespeare Library. Accessed March 10, 2021.
- Stone, Jr., George Winchester. “Garrick's Handling of Macbeth.” Studies in Philology, vol. 38, no. 4 (1941), pp. 609–628. JSTOR. Accessed March 14, 2021.
- Taylor, Gary. "Empirical Middleton: Macbeth, Adaptation, and Microauthorship." Shakespeare Quarterly, vol. 65, no. 3 (Fall 2014), pp. 239–272. JSTOR. Accessed March 9, 2021.
- Taylor, Thomas. The Second Part of The Theatre of Gods Judgments. London: Richard Herne, 1642. Hathi Trust. Accessed March 9, 2021.
- Vickers, Brian. "Infecting the Teller: The Failure of a Mathematical Approach to Shakespeare’s Authorship." Times Literary Supplement, April 17, 2020. Accessed March 31, 2021.
