44Dating and sources: competing narratives

External evidence for dating King John

King John was not entered at any time in the Stationers' Register. The only unambiguous contemporary references to it is by the invaluable Francis Meres, who mentions King John in his list of Shakespeare's tragedies in Palladis Tamia (1598). No other contemporary record of publication or performance has been discovered.

Internal evidence

45Recent work on the chronology of the plays has maintained and strengthened the traditional dating of King John in the mid 1590s; three recent studies confirm the position of King John as one of a group of plays that includes A Midsummer Night's Dream, Richard II, and Love's Labour's Lost (Wells and Taylor 89-109, 119, Vickers 104-11, Jackson passim). These scholars apply multiple criteria in developing their chronologies, including a range of sophisticated computer-generated stylistic tests. Tests of this kind are, of course, approximate, and vary in their results; they tend to work best when applied to works by a single author, making the assumption that the author worked on one work at a time. In the period generally, however, the norm was for collaboration rather than single authorship, and we have no clear assurance that Shakespeare or anyone else worked on one play at a time. No wonder, then, that the most useful and reliable information to be gained from this kind of internal evidence is the way it can relate groups of plays together, providing a more "fuzzy" chronology than critics, fascinated by the concept of an author's consistently developing art, might like. Since Shakespeare's art demonstrably did change and develop, the broad outlines of the chronology remain secure.

46Statistical approaches of this kind favor a position for King John between Richard II and 1 Henry IV, despite the attractiveness of seeing the play as a "transition" between the modern artifacts of the two tetralogies. Since the general acceptance of the concept that Shakespeare created two epic historical tetralogies, critical discussion of the histories has been dominated by this paradigm. Not surprisingly, the two histories that fall outside the chosen eight have fallen on relatively hard times in terms of critical focus and performance--King John and Henry VIII. It is certainly reasonable to think of the three plays of Henry VI together (though we may decide that they were not wholly written in a linear fashion), and it is equally reasonable on the basis of both external evidence and stylistic grounds to assume that Richard III followed soon after. It is no less true that the two parts of Henry IV and the following Henry V are closely interlinked by style and share many characters. Richard II, however, can perhaps be seen to be almost as much an anomaly as King John. Both Richard II and King John focus on a single plot line rather than multiple threads, and, perhaps because of their single plots, both are written throughout in verse; in addition both feature a chiastic, or "hour-glass," structure where one figure fades as another grows in stature. Richard II was certainly written at a time when Shakespeare was experimenting widely with form and with heightened language; like Romeo and Juliet, Love's Labour's Lost, and A Midsummer Night's Dream it delights in moments of high lyricism. King John is no less experimental, but its exploration is of intensely political interaction, a language suitable for political obfuscation, and the attendant ironies of a plot that is finally driven by non-human agency (see the Critical Introduction).

47Since internal evidence points so strongly to a date in the mid-1590s, it comes as something of a surprise that there has been a debate, at times intense, over the date of King John, with some scholars arguing forcibly for a much earlier date than that suggested above.

The debate

48The debate centers on the relationship between The Troublesome Reign of King John (TRKJ) and Shakespeare's King John. Did Shakespeare use TRKJ as his main source for his King John, or did someone else use his King John as the main source for TRKJ? TRKJ was published in 1591, whereas King John did not appear until the First Folio of 1623. Early editors and critics reached the obvious conclusion that Shakespeare had used TRKJ as his principal source in writing King John; the first critic to challenge this chronology was Peter Alexander, who commented in 1929 (Shakespeare's 'Henry VI' and 'Richard III') that

49it is difficult to understand how this work, so well digested in the scenes as to permit Shakespeare to follow it nearly scene by scene, should yet show so little corresponding modesty or cunning in its writing as to appear like a tissue of borrowed and only half-assimilated phrases from Henry VI, Richard III, as well as King John itself. (85)

50This belief--that a writer whose language is as banal as that of TRKJ could not at the same time create an effective plot--has been the mantra that those who would place King John as the earlier play return to again and again. One of the most recent defenders of King John as the prior play, Brian Boyd, writes:

51So boldly and brilliantly has history been reshaped that even those who argue for the priority of The Troublesome Raigne sometimes feel that nobody but Shakespeare could have moulded the material so well, that somehow or other he must at least have had a hand in The Troublesome Raigne before rewriting it as King John.

52The editor of King John for the Signet Shakespeare, William Matchett, makes similar claims, emphasizing what he sees as weaknesses in TRKJ caused by its author's inept imitation of Shakespeare: "What is taken as 'better plotting' in TRKJ can be consistently explained as an expansion and cheapening of Shakespeare's implications" (136). Like many advocates of King John as the prior play, Matchett prefers not to accept the proposition that the reverse process, in which Shakespeare improved upon the dramatic effectiveness of his original, is more likely than a hack writer muddling something already demonstrably superior. On the other hand, Matchett finds the plot superior in TRKJ in the scene where Arthur is to be blinded; here he makes the opposite argument, that Shakespeare would not have made something more muddled that was clearer in his original. Matchett focuses on the apparent confusion in King John between the order King John gives to kill Arthur, and Hubert's intention of blinding him; he comments, "we have the curious Shakespeare of certain textual scholars, the man who grew toward mastery of his craft through carelessness in handling a perfectly clear source. That Shakespeare was sometimes careless--that he was careless here--there is no doubt; but his carelessness is of a differing kind if he is not following a source which has already solved the problem he then creates" (140). I have no wish to try to have it both ways like those I am citing, but I cannot resist pointing out that Shakespeare frequently used a kind of shorthand in his plotting as compared to his sources, making unclear what was perfectly clear; I think of the several places in Antony and Cleopatra where Plutarch is far more precise than Shakespeare in providing information and motivation for the protagonists. Perhaps because it is his intention to complicate the moral simplicity of Plutarch, Shakespeare never makes clear the reason for the final disaster where Antony believes himself betrayed. Plutarch explains it in detail.

53The most recent editor of the play, L.A. Beaurline, restates the basic argument several times, that the author of TRKJ could not possibly have created a plot so well-constructed:

54. . . the author of TRKJ seems insensitive to the very principles on which his play is constructed. A greater dramatic artist conceived the design by which the scattered historical events were synthesized from the Chronicles. (197)55. . . the ironies and the dilemmas that are built into the scenes of the two plays are too sophisticated to have been invented by the author of the dialogue and the monastic episodes in TRKJ. (198)

56A related argument is derived from the unquestioned fact that if TRKJ is the prior play Shakespeare used his source more closely than was his usual habit. Even critics who do assume that TRKJ is Shakespeare's source sometimes feel a need to justify the way Shakespeare made use of it, following it more closely than he does with other source texts: "King John's plot may be from The Troublesome Raigne, but Shakespeare condensed his material to emphasize repeated reversals of expectations" (Vaughn, 415). Stimulated perhaps by the rather poor press King John received from critics in the early part of the twentieth century, the chief proponent of King John as the prior play, E.A.J Honigmann, begins his critical introduction to the play by remarking: "To praise the contrivance of a play which deviates very little from its 'source-play' would be dangerous. The critics, consequently, have been content to admire the few virtues in John not plundered directly from the T.R." (lix).

The argument

57The basic argument is this: TRKJ has generally weak dialogue and language, and such a writer could not have produced a plot as well-constructed as that shared by both TRKJ and King John. But is it demonstrably true that no playwright could have produced a well-constructed plot other than Shakespeare? Beaurline certainly thinks so:

58. . . no other playwright of the time--not Marlowe, Peele, Greene, Kyd, or Lyly--has the combinative and structural powers of the early Shakespeare; yet the preternaturally gifted author of Troublesome Reign has supposedly done his basic work for him. (197)

59The suggestion that only Shakespeare was "preternaturally gifted" in plotting is surely a major overstatement. Shakespeare certainly tended to combine more than one version of a plot, and often to interweave additional materials, but all the evidence suggests that plot was the least of his concerns; much critical energy has been expended puzzling over loose ends in many of his plots. Beaurline might have heeded A.R. Braunmuller's perceptive suggestion that the claim for the precedence of King John over TRKJ

60conceals a germ of Bardolotry: Shakespeare was a master of both plotting and dramatic language; he did not need and would not use, even as a guide, another writer's arrangement of an action; any author who seems his peer or superior must some way owe that quality to Shakespeare. (12)

61Braunmuller goes on to anticipate and counter Beaurline's dismissive comment on Shakespeare's contemporaries by suggesting that "there were several playwrights in the late 1580s and 1590s that could work up a plot from Holinshed and the other chroniclers with at least as much skill as Shakespeare shows in the first tetralogy" (12), adducing both Peele and Marlowe as examples. Recent scholarship from critics less Shakespeare-centered has provided a further corrective to the untested assumption that only Shakespeare could have compiled the plot for TRKJ. Most recently, Brian Vickers has argued on stylistic grounds that the play was actually written by George Peele; drawing evidence from internal use of language: vocabulary, parallel passages, a tendency to "self-repetition," the use of the vocative, self-address in the third person, alliteration, and the use of mixed Latin and English phrases. Of course, those who see King John as the prior play can argue that Peele was the author who rewrote Shakespeare's play. Charles Forker's edition of The Troublesome Reign for the Revels series confidently assigns the play to Peele; his Introduction fully sums up the arguments in favour of Peele's authorship.

62Both Vickers and Forker build on the admirable work done by Scott McMillin and Sally-Beth MacLean on the dramaturgy of the plays written for the Queen's Men in The Queen's Men and their Plays (1998). In their analysis of the nine plays known to have been performed by the Queen's Men, McMillin and MacLean establish that there was a kind of "house style" for the plays, both in plot and language. Queen's Men plays adopted the theatrical literalism of morality plays and interludes, and developed a "medley" style in language, switching from prose to fourteeners to rhymed verse to blank verse as the tone of the moment required. TRKJ fits comfortably into this pattern, and shares many qualities both in plotting and language with the other plays in the group. Since one of the plays (Old Wives Tale) was known to be written by Peele, the claim by Vickers and Forker that he alone is responsible for TRKJ is certainly possible, even likely, though McMillin and MacLean make a strong case for more collaborative composition as the norm for the plays written for the Queen's Men. For our purposes, the vital point is that TRKJ is in no way exceptional in its plot, and not even especially inept in its language. Peele and his peers were typically limited in their style but ingenious in their plotting--precisely what we find in TRKJ.

63Once it is clear that TRKJ is in no way especially remarkable as a play of its period, the basic reason for arguing for the priority of King John effectively disappears. We may be pardoned if we feel some relief that this should be so, since the argument putting King John first requires its proponents to substantially rewrite the generally accepted chronology of the early plays, pushing the early histories back several years, and requiring that even Romeo and Juliet be written as early as 1592. It is true certainly that the chronology of the early plays is based on more conjecture than we might like, but the ingenuity required to justify early dates for so many plays is considerable.

64The usefulness of the approach by McMillin and MacLean is that they view TRKJ horizontally, as one of a coherent group of plays written for a specific company with specific actors. They explore the network of interconnected influences; the Queen's Men were struggling to maintain their audience as new fashions, inspired especially by Marlowe, were taking over the stage. Critics who have attempted to place King John before TRKJ have argued vertically, looking at two specific plays in a disputed chronology, but failing to see the network as a whole. There has been a great deal of influence-hunting as part of the attempt to prove that one or other play came first, but many influences (often called "sources") are readily capable of being interpreted in opposite ways. Editors and textual critics have accordingly developed a series of narratives to explain their preferred train of influence.

The "bad quarto" or "memorial reconstruction" narrative

65The most dogged champion of King John as the prior play has been the editor of the Arden 2 King John (1954), E.A.J. Honigmann. In his discussion of the text of King John, he argued that TRKJ has many characteristics of a "bad quarto," "reported text," or "derivative" play (lvi, 174); it is, he says, a "hotch-potch of a numerous collection of old plays" (lv); according to this narrative, the anonymous author of TRKJ used King John simply for its plot, possibly as a memorial construction after hearing it in the playhouse (lvi), remembering the scenes, then adding his own vacuous language and some additional comic, anti-Catholic scenes. Honigmann finds many other possible sources for King John, though these can of course be seen as working in either direction. To bolster his case further, he sees a number of topical references, none of which stand up to close inspection, largely because, as Braunmuller points out, in terms of potential topical allusions, the entire Elizabethan period is an "embarrassment of riches" (3). Alice Walker, in her review of Honigmann's edition, pointed out that several stage directions in the two plays are so similar that is is difficult to imagine them being created without one being the model for the other; she describes these as "documentary evidence" for the priority of TRKJ. Her initial example is the very first entrance in each play:

66Enter K. Iohn, Queene Elinor his mother, William Marshal Earle of Pembrooke, the Earles of Essex, and of Salisbury. (TRKJ, TLN 1-2)67 Enter King Iohn, Queene Eleanor, Pembroke, Essex, and Salisbury, with the Chattylion of France. (King John, TLN 2-3)

68This stage direction is especially interesting, since Essex appears only in this one scene of King John and is never mentioned by name in the play; for this reason, Walker points out, the theory of some kind of memorial construction is difficult to justify.

A counter-narrative

69The next editor of King John, R.L. Smallwood, dismissed the narrative of memorial construction leading to a bad quarto:

70We might, however, still imagine a man writing the play after seeing King John if we supposed him to be a vigorous Protestant burning to explore issues that he felt Shakespeare had inexcusably left out of his play, going back over the chronicles to do so and picking up a good deal of extra, irrelevant information in the process and including it in his play as a sign of independent research, and then, having deliberately cleared his mind of all memories of Shakespeare's language except for a few details of history which, though mainly available in the chronicles, he had most precisely learned by heart from Shakespeare, writing a play that eschewed all of the theatrical high-points that King John had achieved, and substituting a longer exploration of historical and religious material. (371)

71As Smallwood comments, "[s]uch a man is indeed difficult to imagine."

72A decade later the issue became the subject of a debate in the pages of Shakespeare Quarterly. In 1986, Sidney Thomas pointed out that another stage direction in King John seems to show that Shakespeare was working from a copy of TRKJ as he radically rewrote the play. The curious direction is Enter a Sheriffe(TLN 50); no Sheriff speaks and no Sheriff is given an exit. Thomas points to the equivalent scene in TRKJ, where a Sheriff enters and duly passes on information to the assembly concerning the background of the quarrel between Robert and Philip Faulconbridge. The following year SQ published what they called an "Exchange," providing a forum for varying narratives. Honigmann returned to the fray, insisting again that his narrative could withstand the forces of skepticism, and reiterating his belief that "those who prepared the copy for Shakespeare's Folio sometimes printed from, or consulted, quarto texts. Even 'bad' quartos were used" (Honigmann, 1982 62). Thus, he proposed, those who prepared the Folio King John consulted TRKJ to assist with stage directions when they were "illegible, or simply missing" (Honigmann, 1987 124). He made an interesting comparison with the relationship between TRKJ and the anonymous The Taming of a Shrew, going so far as to suggest that "the authors of AS and TRKJ, perhaps one and the same man, could have had access to unpublished texts of TS and KJ [King John]" (125). In the same exchange, Thomas is moved to something close to strong language:

73The relationship between The Taming of a Shrew (AS) and The Taming of the Shrew (TS) that Honigmann instances to buttress his argument is not only problematic but irrelevant to the question of the relationship between TRKJ and KJ, despite Honigmann's amazing and completely unsubstantiated suggestion that the same author may have written both AS and TRKJ. (Thomas, 1987 130)

An alternative narrative

74In the same exchange, Paul Werstine weighed in with an alternative narrative to explain the puzzling Sheriff. Using the manuscript of Edmund Ironside as an example, Werstine suggested that the Sheriff's ghost may have been created by an annotator preparing the manuscript for use in the theater: "a dramatist lists a character's name in a stage direction and then provides the character with a speech in the scene that follows. The hand that annotates this manuscript for the theatre denies the character his speech but fails to eliminate his name from the entrance direction and thereby makes the character a ghost" (128-9). Possible though this narrative may be, research published by Gary Taylor since Werstine was writing suggests that the manuscript the compositors were working on was not closely connected with the theater.

The demise of the "bad quarto" narrative

75In the same eventful year, 1987, A.R. Braunmuller's insightful and scholarly Oxford edition appeared, as did the companion volume to the Oxford Works, William Shakespeare, a Textual Companion. Both made important contributions to study of King John. Braunmuller politely demolished the many arguments proposed by Honigmann and Matchett, and introduced into the discussion the discovery by Gary Taylor and John Jowett that the Folio copy of King John was the result of a transcription by two scribes (Textual Companion 317).

76It is no surprise then that Beaurline's attempt to resuscitate the precedence of King John requires that he shift his argument from the discredited theory of a bad quarto to the proposition that the author of TRKJ had access to an "author's plot" (207) of some kind, conveniently providing some stage directions he could imitate without all the baggage of Shakespeare's language. That there is no evidence of any kind for such author's plots as part of the process of the transmission of Shakespeare's texts is not mentioned. The appearance the following year of Laurie Maguire's Shakespearean Suspect Texts made even more convincing the earlier arguments that TRKJ is in no way likely to be a memorial construction (314-15). One cannot help but feel that those arguing for the priority of King John were increasingly forced to grasp at straws as their narratives became less and less likely to convince. No wonder then that Honigmann's last gasp reverted to vaguer and more general arguments than textual to press his still insistent claim that King John was written first.

Never say die: now the author of TRKJ acted in King John

77Honigmann's final salvo in the debate appeared in 2000. This time he found evidence for the priority of King John by locating a series of Shakespearean "self-repetitions" in TRKJ and King John, arguing that the author of TRKJ must have used a Shakespearean original. I have neither the space nor (I will admit) the patience to work through his examples to show how commonplace they are, and how they are clearly derived initially from TRKJ in order to show its supposed dependence; it is precisely the kind of horizontal study exemplified by McMillan and MacLean that shows the hollowness of this kind of argument. Honigmann returns to his main theme, that the author of TRKJ could not possibly have created the plot in its complexity. But even here he wants to have it both ways: the plot is good, but it is also inept:

78. . . a beautifully plotted play in its analysis of complex political manoeuvres and its dramatic control, at times appears to confuse or forget its own logic. . . . [the author of TRKJ] wrote a play that resembles Shakespeare's in its plotting yet goes off the rails whenever it differs from Shakespeare in its plotting. (178)

79Finally, in order once again to provide an acceptable narrative that will explain the mechanism for the author's borrowing of the plot from King John, Honigmann proposes that the author "had acted in King John--and, it would follow, in other plays by Shakespeare, Marlowe, Peele and Kyd" (180). There is, of course, absolutely no evidence to support this claim.

Applying Occam: returning to the original narrative

80The original argument postulated on purely esthetic grounds by Peter Alexander has led to seventy years of intensive scholarship, spearheaded by his pupil, Ernst Honigmann. Despite the scholarly good sense of Braunmuller, the debate has continued until very recently, with no "smoking gun" discovered by those who wish to challenge the generally accepted date. The continuing confirmation of a later date for the composition of King John through newer techniques for evaluating internal and stylistic development, the difficulty of the "knock-on" effect requiring that many other plays be re-dated to earlier years, and the clear inaccuracy of the claim that an author of the talent of the writer of TRKJ could not have compiled its plot--all lead to the simple solution that Shakespeare used TRKJ as his source, relying on it more closely than was his habit at other times. The horizontal approach of McMillin and MacLean, particularly in the analysis of the rivalry between the Queen's Men and the increasing popularity of Marlowe's plays, provides the clearest sense yet of the place of TRKJ in the period as a whole. Their further speculation, that the apprentice Shakespeare may actually have contributed in a minor way to TRKJ, elegantly explains the intersecting questions of attribution on the title pages TRKJ Q2 and Q3, the lack of entry in the Stationers' Register for King John when the Folio was being planned, and Shakespeare's unquestionably atypically detailed use of TRKJ as his main source. There is no way to establish the truth or falseness of this immensely attractive narrative, but it does provide a synthesis of the available evidence.

Conclusion

81The challenge to the Queen's Men provided by the popularity of Marlowe (and Kyd) is a reminder that at least some in their audience changed in their tastes in a way that must have influenced authors writing for all companies. We may be reminded of the tetchy comments of Ben Jonson on members of his audience who "will swear, Jeronimo or Andronicus, are the best plays yet"; but his censure of those whose judgment has been "constant , and hath stood still these five and twenty or thirty years" (Bartholomew Fair, Induction) is at the same time praise of those who have indeed changed. It is reasonable to assume that that authors were interested in reaching the most responsive and critically aware members of their audience. Shakespeare's own plays-within-plays illustrate the development in the audience from naivety to sophistication as they move from requiring an argument or dumb show to explain the action ahead of time--and requiring the player, when he came in, ever to begin with telling where he was, or else the tale would not be conceived--to the kind of opening scene in King John, where the exposition emerges seamlessly from action and dialogue. This audience clearly filled in gaps in Shakespeare's dramatic shorthand without straining after detail. A more extended vertical analysis of the audiences of the period along these lines may also provide further information in the dating of plays.

82The debate sustained by Honigmann and others has not shaken the essential validity of the network of evidence both external and internal that establishes the sequence of Shakespeare's plays, or of groups of his plays, putting the composition of King John securely in the mid 1590s, well after the publication of TRKJ. A less bard-centered approach both to Shakespeare's own plays and those of his lesser contemporaries would encourage textual scholars to spend less of their considerable energies and intellects seeking to prove that Shakespeare wrote all that was worthwhile, when others manifestly had good, if minor, talents. There are more important things to explore in King John than the quality of its plot.