The Integrity of King Lear
Sidney Thomas, Syracuse University
In a brief article published a decade ago, I ventured to challenge the then relatively new theory of a two-text King Lear, hoping that by calling attention to some of the theory's weaknesses and exaggerations, I could help to prevent it, as I said then, from hardening into 'a new orthodoxy'.1 What I anticipated has now happened: not only, as Jay Halio has recently observed, has 'strong support for a revision hypothesis [ … ] grown among scholars',2 but the new orthodoxy I feared has proved to be even more rigid and uncompromising in its assertion of the absolute truth of its position than any one could have expected, and what I half-jokingly predicted, the metastasis of the two-text theory of King Lear to other plays in the canon, has taken place with a vengeance.
Thus, in the latest and most extensive treatment of the revisionist theory, Grace Ioppolo recommends that 'the Quarto and Folio texts of Hamlet, Troilus and Cressida, Othello, and other plays should also [in addition to King Lear] be printed in separate versions if scholars are ever to come to terms with all that they offer'.3
The adherents of the revisionist hypothesis have now proclaimed victory. In the approving words of Ioppolo, 'the new revisionists have achieved a coup d'état which offers a new constitution for how scholars read, study, and teach Shakespeare's canon and also redefines the canon itself (p. 3). I am therefore impelled to take up once again the argument for a King Lear conceived and written by Shakespeare as an integral work, and left untouched by him except for minor changes made currente calamo, or in the course of rehearsal or initial staging.
To begin with, there is not a shred of external evidence to support the notion that Shakespeare revised any of his plays after its first performance except, perhaps, to correct an obvious gaffe. Significantly, the very Oxford editors who argue for the hypothesis of Shakespearean revision in one passage of their Textual Companion4 take a very different tack in another passage. Referring to the contention that Shakespeare's plays might have been posthumously revised, they declare:
Most obviously, Shakespeare could not veto or influence any changes imposed upon his plays in the theatre after his death. Fortunately the economics and mechanics of the pre-Restoration repertory system made it impractical to reshape a play every time it was revived; when later adaptation did occur, it usually involved the addition of discrete chunks of material. Therefore the number of changes affected by such intervention should be small. (p. 15)
But if the economics and mechanics of the repertory system made posthumous revision unlikely, did they not also make revision by Shakespeare himself equally unlikely? The very same article cited by the Oxford editors as a basis for their dismissal of the probability of posthumous revision provides strong evidence against the theory of authorial revision. Rosalind Knutson, after a careful and thorough analysis of Henslowe's theatrical records, concludes:
For the last decade of the Elizabethan period, for the one company with a playhouse document that shows patterns of revival and the commercial value of plays in the repertory, the assumption that revision accompanied revival cannot be supported. Furthermore, the assumption that revision was necessary to make old plays profitable cannot be supported. [ …] Only in a few isolated cases is there evidence that the plays being revived were also revised. [ … ] On the basis of evidence in Henslowe's Diary, therefore, revision for the occasion of revival was neither commonplace nor economically necessary.5
But, it may be objected, the evidence of Henslowe 's Diary need not apply to Shakespeare's company or to Shakespeare himself. There is, however, a striking piece of evidence to suggest that for the management of the King's Men, as well as for Henslowe, revival did not necessarily entail revision. In an often-cited 1604 letter from Sir Walter Cope to Robert Cecil, Lord Cranborne, Cope reports his efforts to arrange a theatrical performance for the Queen, and after detailing his frustrations, records his success: 'Burbage ys come, & Sayes ther ys no new playe that the queene hath not seene, but they have Revyved an olde one, Cawled Loves Labore lost, which for wytt and mirthe he sayes will please her excedingly.'6 Surely, if the 'old play' being revived (the most topical of Shakespeare's comedies) had been revised to bring it up to date, Burbage would have mentioned it to Cope, and Cope would have reported it to Cecil. We can be virtually certain, therefore, that the play revived in 1604, was basically identical with the play written and first acted about 1594, and published in 1598. Moreover, if there had been a new version produced in 1604, we would expect the 1623 Folio text to embody that version, rather than the 1598 text, as it essentially does.
Of course, the example of Love's Labour's Lost proves nothing about the text of King Lear. Shakespeare may have chosen not to revise the one, but to revise the other. It does, however, induce us to question Gary Taylor's 'obvious conclusion, that Shakespeare occasionally—perhaps if we could only see it, habitually—revised his work'.7 So far from being obvious, the theory of habitual, or even occasional, Shakespearean revision has nothing to support it. No play in the canon, it is safe to say, more desperately needed revision to make its wealth of topical allusions (many still a mystery to us) meaningful to an audience after the passage of ten years than Love's Labour's Lost. That it seems not to have received such revision should make us cautious about any theory of habitual revision.
What we know, from contemporary references, about Shakespeare's creative processes and relationship to his own work also militates against the theory of authorial revision. Heminge and Condell, who were Shakespeare's close colleagues for many years, provide in their address 'To the great Variety of Readers' (whether written for them by Ben Jonson does not matter) in the 1623 Folio (A3) a picture of a man from whom they never received a blot in his papers, who unfortunately never lived to oversee the publication of his plays, and whose work they were now presenting to the reader in its perfect form, free of the corruptions of previous unauthorized publications. All these statements emphasize one thing: as editors, Heminge and Condell were giving the public what Shakespeare had written once and for all, restored to its initial perfection as it had originally existed. There is no mention, not even a hint, that they were presenting what Shakespeare himself had revised and presumably further 'perfected', certainly an effective selling-point if they could have made it. To claim that what they did say is meaningless hyperbole, not to be taken seriously (though I do not believe this to be so) ignores the fact that a more useful hyperbole was available to them, if the revisionists are right.
If the external evidence for revision is nonexistent or dubiously conjectural, on what, then, do the proponents of the two-text Lear mainly rely? It is the so-called internal evidence of the texts themselves, the revisionists assert, that overwhelmingly supports their theory that the Q and F Lears are two separate plays that must be printed, staged, and studied as independent entities. One of the basic underpinnings of this theory (the argument that the characterization of Edgar and Albany, as well as their relationship to each other, is radically different in F from what it is in Q) was first elaborated by Michael Warren in a pioneering essay some years ago and has remained an article of faith for the revisionists ever since.8 If, therefore, Warren's thesis can be shown to rest on unfounded assertions and assumptions, as I believe it does, the two-text theory of Lear can be seriously questioned. Here is Warren's principal argument in detail:
The elevation of Edgar at the close and relative reduction of Albany that distinguish F from Q can be documented from three other places. At 5.3.229 [V.3.204; TLN 3180] in Q, Edgar says to Albany 'Here comes Kent sir', but 'Here comes Kent' in F. The transfer of the command 'Hast thee for thy life' (5.3.251 [V.3.225; TLN 3209] from Albany in Q to Edgar in F gives Edgar a more active role in the urgent events; indeed, Q may indicate that it is Edgar who is to run. [… ] In Q [… ] Edgar concludes the play stunned to silence by the reality of Lear's death, a very young man who does not even answer Albany's appeal 'Friends of my soule, you twaine / Rule in this Realme' (5.3.319-20) [V.3.293-94; TLN 3294/95]. [… ] This characterization of Edgar is a far cry from the Edgar of F who comes forward as a future ruler when he enables Albany to achieve his objective of not ruling. (pp. 104-05).
I begin with Warren's first piece of documentation, the omission of 'sir' in F. Leaving aside the distinct possibility that the omission of 'sir' in F is a compositorial error, and accepting for the moment the authenticity and intentionality of the F reading, it is still possible to reject it as 'documentation' of the argument that Edgar in F is deliberately made more assertive and less deferential to Albany than he is in Q. For, some fifty lines earlier in the scene (v.3.169; TLN 3141) Edgar has addressed Albany in F (as in Q) as 'Worthy Prince'; and a few lines later in F as 'my Lord' (v.3.172; TLN 3144), and again in F (v.3.223; TLN 3206) as 'my Lord'. Above all, there is Edgar's reference in F (as in Q) to Albany as 'this high illustrious Prince' (v.3.125; TLN 3090). To claim, as against these examples of Edgar's proper deference to Albany, that the absence of 'sir' at the end of a casual observation (incidentally, not directly addressed to Albany in F) is crucial evidence of a change in the relationship between Edgar and Albany in F is making far too much of very little.
Even less convincing is Warren's second piece of documentation. 'The transfer of the command "Hast thee for thy life" (5.3.251) [v.3.225; TLN 3209] from Albany in Q to Edgar in F gives Edgar a more active role in the urgent events; indeed, Q may indicate that it is Edgar who is to run', says Warren. But what Warren neglects to point out is that it is Albany who gives the first decisive command, in F (as in Q) 'Run, run, O run' (v.3.221; TLN 3205) and that Edgar's words merely echo Albany's. Moreover, throughout v.3 it is always Albany who is in command, and directs the necessary action. To claim that 'in F Edgar grows into a potential ruler, a well-intentioned, resolute man in a harsh world, while Albany, a weaker man, abdicates his responsibilities' (p. 105) is to fly in the face of the evidence. Against Warren's sweeping statement may be placed R. A. Foakes's acute and, in my opinion, demonstrably correct comment on 'the new toughness and independence shown by Albany in the later scenes of the play'.9 At the very least, Albany is no weaker a character in F than he is in Q. Again and again in his exchanges with Edmund, Albany shows his resoluteness, his determination to assert his authority and control. At his first entrance in v.3, he makes it clear, in his exchange of words with Edmund, that he will brook no challenge to his own primacy:
Sir, you have shew'd to day your valiant
And Fortune led you well: you have the
Who were the opposites of this dayes
I do require them of you so to use them,
As we shall find their merites, and our
May equally determine.
(v.3.394; TLN 2984)
After Edmund's attempt to justify his action in disposing of his prisoners come Albany's curt dismissive words:
Sir, by your patience
I hold you but a subject of this Warre,
Not as a brother.
(v.3.53; TLN 2997)
Further, to speak of Albany in F as 'a weaker man' who 'abdicates his responsibilities' is manifestly untenable in the light of the ringing accusatory and condemnatory speeches he addresses to Edmund in F (as in Q):
Stay yet, heare reason: Edmund, I arrest
On capitali Treason; and in thy arrest,
This guilded Serpent: for your claime faire
I bare it in the interest of my wife,
'Tis she is sub-contracted to this Lord,
And I her husband contradict your Banes.
If you will marry, make your loves to me,
My Lady is bespoke.
(v.3.76; TLN 3027)
Continuing after Goneril's contemptuous interjection, 'An enterlude':
Thou art armed Gloster,
Let the Trumpet sound:
If none appeare to prove upon thy person,
Thy heynous, manifest, and many Treasons,
There is my pledge: Ile make it on thy heart
Ere I taste bread, thou art in nothing lesse
Then I have heere proclaim'd thee.
(v.3.84; TLN 3035)
In these two successive speeches, it is not only their content that shows Albany as a strong, self-assured leader. It is their poetic force, the sardonic wit of the first speech, the hammer blows of the second speech, that project a mature and powerful personality, no less so than in Q. If he abdicates his responsibilities in F at the end of the play when he asks Kent and Edgar to 'Rule in this Realme, and the gor'd state sustaine' he does no less in Q in virtually identical words. Nor do I think any significance should be attached to the fact (Warren's third piece of 'documentation') that the final speech, given in Q to 'Duke' or Albany, is given to Edgar in F. I believe that Q is in error here and that the speech with its clear emphasis on 'we that are yong' (in both F and Q) belongs to Edgar, the one survivor of the new generation.
Not only does the attempt to make the Albany of the closing scene of the play in F a differently conceived and developed character from the Albany of Q fail: the similar attempt by Steven Urkowitz to present the Albany of the earlier scenes of the play in F as a morally ambiguous figure, in contrast to the morally assured Albany of Q, is also unsupported by the evidence, as I have argued before.10 But if the case for a major revision of Albany in F collapses, what then is left of the overall argument that the F text of Lear is a substantial authorial reworking of the play embodied in the Q text, and that the two texts represent two independent works that must be printed, studied, and produced as separate entities? For it is the supposed change in the characterization of Albany in F on which the two-text theory of Lear largely rests, since, as Gary Taylor, one of the principal revisionists, concedes, 'the Folio-only passages do not include new narrative material or major, structurally important incidents; they do not consist of new scenes, but of alterations here and there'.11
The argument for supposed changes in characterization between Q and F, while one of the main props of the two-text theory, is not, however, the only one on which the revisionists depend. Equally important to their thesis is the treatment of all substantive differences in wording between Q and F, except for the most obvious compositorial errors, as evidence of authorial revision, designed to change or modify the dramatic impact or significance of a particular speech. In what the General Editors describe as 'the first fully annotated, critical edition of King Lear to appear for forty years' (p. i), Jay Halio gives, in support of 'the revision theory on which this edition is based' (p. 81), the most detailed, well-argued, and lucid account of the presumed evidence for revision found in the linguistic variants between Q and F that has yet appeared.
None the less, I find the case for revision, as he presents it, no more convincing than that earlier made by Warren, Urkowitz, and others.
It is impossible, in this article, to deal with all of the presumed evidence for revision presented by Halio in his textual analysis. I deliberately select his discussion of one of the key speeches in the play, Lear's opening statement: