"Precision, consistency, and completeness in early-modern playbook manuscripts: The evidence from Thomas of Woodstock and John a Kent and John a Cumber" by Gabriel Egan*
In three influential essays of the 1980s, William B. Long challenged the standard New Bibliographical characterization of early-modern promptbooks and established a new othodoxy (Long 1985b; Long 1985a; Long 1989). Contrary to the assertions of W. W. Greg and others, Long argued, the play manuscripts used in the early-modern theatres, called promptbooks by New Bibliographers, could be sloppy. Long's essays have been widely received as successful attacks on the New Bibliography.1 According to Long, the vague, permissive, or self-contradictory stage directions that appeared in a dramatist's manuscript were not necessarily tidied for the convenience of the prompter, and no systematic effort was made to regularize inconsistent speech prefixes that gave multiple names for one character. Because such authorial peccadilloes were acceptable to performers, Long argued, the author's manuscript itself (rather than a fresh transcript of it) could normally serve as the document that regulated the performance as it happened.
Before Long's essays appeared, the New Bibliographical consensus established by Greg divided manuscripts into two main categories. The first was authorial papers in which inconsistency abounded, because an author in the heat of composition would be concerned with the grand scheme of a play's development and ignore minor details. The second category was promptbooks, which had to be orderly and complete if the performances were to run smoothly. Either kind of document could be used to print a play, and Greg's binary distinction gave editors a simple test for determining which kind of manuscript was used as copy for a particular early print edition. If the stage directions are vague, permissive, or self-contradictory, or the speech prefixes inconsistent, then the edition was set from authorial papers, and if not then it was set from a promptbook. Because Greg's taxonomy contained just two kinds of manuscript, authorial (sometimes called foul) papers and promptbook, the elimination of one candidate for copy instantly confirmed the other. For example, the presence in a print edition of inconsistent speech prefixes eliminated the possibility that the printer's copy was a promptbook, so it had to be authorial papers, and the presence of notes to have properties or actors ready (which an author would not write) meant that the printer's copy was a promptbook.
Long's argument, if correct, renders these simple tests invalid, as editors of Shakespeare have noted most readily.2 From the evidence of two manuscripts, the anonymous Thomas of Woodstock3 (British Library Egerton 1994) and Anthony Munday's John a Kent and John a Cumber (Huntington Library HM 500), Long offered a new characterization of the surviving playbook manuscripts that have a connection with the theatre, totalling sixteen documents in the essays considered here but later raised to eighteen (Long 1999, 414). The default assumption of the new orthodoxy developed by Long was that actors could perform from relatively untidy and inconsistent manuscripts. The present essay revaluates, in the light of subsequent discoveries, the logic of Long's argument and the manuscript evidence upon which it was based. Long's key inferences from theatre-historical data are rejected and the manuscripts are found to contain conflicting evidence that is no better explained by Long's new orthodoxy than by the old one it replaced.
The first of Long's essays was an article called "Stage-directions: A misinterpreted factor in determining textual provenance". Long began with Greg's personal failings, and in particular his strong belief "in spite of evidence published in his own researches, that a finished promptbook had to be carefully inscribed and meticulously written" (Long 1985b, 122). According to Long, Greg simply misapplied twentieth-century theatrical expectations to early-modern plays. From the manuscripts themselves Long discovered that
authorial stage directions are very seldom changed in the theater; speech heads are not regularized; copious markings do not appear to handle properties, entrances, and music. Regularization and completeness simply were not factors in theatrical marking of an author's papers. Theatrical personnel seem to have marked the book only in response to problems. (Long 1985b, 123)
Long assured the reader that his characterization of promptbooks--he preferred the term playbooks--was based on "the only detailed investigation of this material since Greg" (Long 1985b, 122) and he listed the sixteen manuscripts in question (Long 1985b, 135-6n3). However, Long offered no illustrative examples from the manuscripts themselves, quoting only the scholars he thought were misled by New Bibliographical assumptions originating with Greg. The commonest mistake Long found was the editorial assumption that the absence of scrupulous directions concerning stage business and properties in an early print edition proves that its printer's copy was not a promptbook. Rather, Long rightly pointed out, most of the surviving playbook manuscripts that have signs of use in the theatre have few added directions to supplement those provided by the dramatist.
One of Long's examples of an editor misreading an early print edition was J. R. Mulryne's assumption that because "Table and | Chess" appears as a marginal note in the first edition of Thomas Middleton's Women Beware Women, twenty-five lines before these items are called for by the dialogue (Middleton 1657, I5v), the note is a prompter's reminder to have the properties ready (Middleton 1975, xxiv-xxv). Long thought this possible but unlikely unless the printer had "copied [the words] exactly" and "positioned them as they were in the manuscript" (Long 1985b, 133). Just as likely, according to Long, the words appeared elsewhere in the manuscript (500 lines earlier near the start of the act) and were not for prompting purposes but rather were "a notation of unusual properties which had to be obtained" (Long 1985b, 134). Just why a printer would fail to copy the words exactly and would displace them by fifteen pages, onto a different sheet, Long did not explain. The idea that marginalia that have long been taken for readying notes are in fact notes for the procurement of properties was to become central to Long's analysis of the manuscript of Thomas of Woodstock.
From his fresh survey of the evidence, Long offered new principles. Rather than prompting in the modern sense by poring over a promptbook, the company bookkeeper casually glanced over the playbook in front of him. Had the bookkeeper been paying close attention, Long reasoned, there would be no need for the fairly common marginalia that draw his attention to particular properties or entrances that require special consideration. Instead of systematically adding a layer of theatricalizing notes, regularizing the speech prefixes, and making the stage directions complete and unambiguous, the company did little to the dramatist's manuscript (Long 1985b, 125). If there was a specific problem--say, the need to cue an off-stage sound or to time precisely a crucial entrance--then marginalia might be added. But otherwise, the author's manuscript was considered perfectly usable to run the play and was not copied out first. Long's position was a revival of the theory of 'continuous copy', as E. K. Chambers called it (1924-25, 103), in which a single manuscript begins as authorial papers and ends up as a prompt-book. This New Bibliographical theory was invented by A. W. Pollard (1917, 58-62) and enthusiastically taken up by John Dover Wilson in the early volumes of his New Shakespeare series for Cambridge University Press (1921-66). Wilson later abandoned the theory as enthusiastically as he adopted it, claiming that he could not recall believing it (Shakespeare 1952, 118). Greg consistently opposed the theory, suggesting rather unfairly that Pollard (its founder) entertained it only when misled by Wilson (Greg 1955, 102-03). Long's own count that no more than half of the sixteen theatrical manuscripts in his survey are in their dramatists' handwriting (Long 1985a, 116n20) ought to have put him off the 'continuous copy' theory.
Long's second essay was an article focussing on one playbook manuscript, Thomas of Woodstock, and it provided detailed evidence to support his claims about the misconceptions of New Bibliography (Long 1985a). Marginalia that have been interpreted as notes reminding someone to have ready at the right moments the properties or people needed during the performance are more likely, Long argued, to be records of properties to be made, bought, or borrowed in advance of performance. For example, Greg understood the left-marginal note "Shrevs Ready"4 that appears thirteen lines before the centred entrance direction "Enter Trissilian wth the Shreeues of Kent and Northumberland" to be a prompter's warning to himself to check that the actors concerned were ready (Anonymous 1929, 179b; Greg 1931, 255). Long disagreed, since if the bookkeeper needed advance notice of this entrance, why not advance notice of other entrances? Long commented: "That such an inscription does not occur for the other roles demonstrates exactly the occasional marking and total lack of regularity found throughout the surviving playbooks" (Long 1985a, 95). Rather than giving a warning to be heeded during the performance, Long suggested, this annotation was made to record the need to cast actors and acquire properties for the parts of the shrieves. This sounds like special pleading, since something closer to "Shrevs needed" would seem the likelier phrasing for Long's purpose. Moreover, such a reinterpretation would deny the manuscript's role in regulating the moment-by-moment vicissitudes of performance. If notes formerly believed to comprise a layer of theatricalization (necessary to turn the author's manuscript into a promptbook) are only records of planned procurement, the manuscript loses its power to critique New Bibliography's characterization of promptbooks, since it is not that kind of manuscript.
To secure for the manuscript of Thomas of Woodstock a close connection with multiple performances over many years--lest it be thought irrelevant to his critique of Greg's characterization of promptbooks--Long described its involvement in three runs: the original and two revivals. The original performances Long dated to 1592-95 because twenty lines of dialogue are marked as "out" (Anonymous 1929, 176b), meaning that they are to be omitted (Long 1985a, 109). These lines refer to French fears about Edward of Woodstock, the Black Prince, and Long thought that the cut reflected theatrical prudence in the years when queen Elizabeth's relations with the French king Henry 4 were at a delicate stage. Because 1592-95 was a time when such a cut would be prudent, Long took this as his date for the first performances. Long found evidence for a revival of Thomas of Woodstock "ca. 1602-04" in the presence of annotations by a hand that appears also in the manuscript (British Library Egerton 1994) of the play Charlemagne "ca. 1600-04" (Long 1985a, 111). Long dated a second revival of Thomas of Woodstock by the presence of a hand also found in the manuscript of Walter Mountfort's The Launching of the Mary (another play in British Library Egerton 1994). Because the Master of the Revels Henry Herbert licensed The Launching of the Mary in 1633, Long assigned the second revival of Thomas of Woodstock to around that year (Long 1985a, 113).
All Long's dates are vulnerable. That a cut would be prudent in 1592-95 does not mean it was made then; the first performance cannot be so dated. A scribe's handwriting need not change appreciably over the years, so the supposed revivals cannot be dated by the presence in Thomas of Woodstock of hands appearing elsewhere at known times. MacDonald P. Jackson's exhaustive collation of evidence to attribute and date Thomas of Woodstock pointed to its having been written by Samuel Rowley after 1605 (Jackson 2001). The presence of two actors' names in the manuscript is evidence of "a revival about 1632-35, either by the second Prince Charles's Men or by the King's Revels Company or touring troupe of which they formed the nucleus" (Jackson 2001, 22). There are marks in the manuscript that look like those made elsewhere by the censor George Buc (Anonymous 1929, xxi), which can be dated 1603-22 when Buc was active in this capacity (Jackson 2001, 23). If the marks in Thomas of Woodstock are Buc's, the simplest explanation is that he was licensing the first performances and hence the manuscript was the 'allowed book' for performance, or as New Bibliographers called it, the promptbook.
Like his dates, Long's reinterpretation of readying notes as procurement notes is open to objection. The feature of Thomas of Woodstock that gave Long's article its title is the annotation "A bed | for woodstock" (Anonymous 1929, 180b), which he decided could not be a readying instruction because it occurs just eight lines before the bed is needed, which is not enough advance warning (Long 1985a, 108-09). Yet, as Long admitted, those eight lines contain an act interval marker ("Actus quintus") and the bed is needed for the start of the new act. Because he thought the play was first performed in the early 1590s, before act intervals were routinely observed (Taylor 1993b), Long assumed that the interval marker reflected performance only in the revivals. With composition now dated after 1605 it is just as likely that the intervals were observed in the first performances. "A bed | for woodstock" makes perfect sense as a readying note and the eight lines of anticipation plus an interval provide enough time to prepare the bed and get Woodstock in it. Long offered a peculiar objection to this interpretation: ". . . there is no provision for checking to see if the player is in it, although he must be there by the time the curtains [of the bed] are drawn" (Long 1985a, 108). It is not clear why Long thought the words "for woodstock" are insufficient provision.
Long offered two more arguments against "A bed | for woodstock" being a readying signal. The first was that the words "A bed" would imply that the company owned more than one, since otherwise the annotation ought to read "the bed". Long affected to know that the actors' economic state ruled out their possession of multiple beds:
. . . it is certainly unlikely that an acting company in this financially troubled period would possess more than one [bed]. Could the actors even have afforded one? Such a bed, of course, could have been borrowed, and more than likely was. (Long 1985a, 108)
For "A bed | for woodstock" to be a readying note, Long reasoned, the company would have to possess several beds, one of which this note calls for. Since the company cannot have owned several beds, "A bed | for woodstock" cannot be a readying note but must indicate the intention to procure one. However, even if the original company and performance date were securely known, we have nothing like enough detailed knowledge of any company's finances to be sure that they would have to borrow a bed. As with his support for the 'continuous copy' theory, Long's interpretation of a distinction in articles (a/the) comes directly from Pollard, New Bibliography's founder. Pollard argued that the direction "Enter Piramus with the Asse head" in Shakespeare's A Midsummer Night's Dream (Shakespeare 1623, N4v) could have been written only by someone who knew that there was just one such head in the company's stock (since otherwise the wording would be "an Asse head") and hence was a note made by the prompter (Pollard 1917, 68-69). All subsequent commentators have considered Pollard's faith in the significance of articles to be unfounded. Long's final reason for insisting that "A bed | for woodstock" cannot be a readying note is that the bed is too important to overlook, being "not something that anyone would need to be reminded about; without it the whole first half of the act would be impossible" (Long 1985a, 108). That is to say, a readying note would be as pointless for something of great significance in the show as it would be for something of little significance. According to Long, only for matters that fall between these extremes would a readying note be justified. Woodstock's bed, like Goldilocks's, would have to be just right.
Although Woodstock's bed is not a convincing illustration of Long's idea that Thomas of Woodstock's marginal notes record plans for procurement, other examples are compelling. One is the left-marginal note "Booke" occurring eighteen lines before Richard asks Bushy "what readst thou?" (Anonymous 1929, 166b). This cannot be a backstage reminder to have a book ready, for Bushy has been on the stage since the start of the scene fifty-three lines earlier and no-one has entered or exitted in the meantime. Either Bushy brought the book on with him, or it was placed on the stage at the start of the scene, and either way the supposedly anticipatory annotation is in fact thirty-five lines too late. It might be argued that "Booke" occurs at a moment when the character speaking, Tresilian the Lord Chief Justice, places his hand on a book that he has brought on with him, and hence that the note is an additional stage direction. Tresilian is at this moment giving a legal judgement (answering king Richard's "resolue vs therefore" on the lawful treatment of rebels), so a book would be an appropriate object to handle at this moment; Bushy could perhaps be browsing the same book eighteen lines later. However, this interpretation is rather more forced than Long's simple explanation that "Booke" was a note reminding someone to procure one.
A similar balance of probabilities in Long's favour governs the right-marginal note "Paper", which precedes by five lines king Richard's "reach me that paper Busshey" (Anonymous 1929, 168a). Just before the note, king Richard asks "wher are those Articles", referring to quite separate petitions from parliament, and York offers to fetch in the officials bringing them. It is just possible that "Paper" is an anticipatory note to have these off-stage parliamentary petitions ready, but since they are not used--the officials are never brought in and the scene develops in a different direction--it is more likely that the note refers to Bushy's paper. Since no-one goes off to fetch it, Bushy must have brought the paper with him thirty-three lines earlier at the start of the scene. Thus, rather than anticipating the need for Bushy's paper, the note "Paper" is too late. Rather less securely in Long's favour is the left-marginal note "Blankes" that appears five lines into the third act, six lines before Tresilian shows to king Richard his blank charters, a device for taxation (Anonymous 1929, 170b). Appearing almost at the start of the scene, "Blankes" could be a reminder to have them ready for that scene rather than, as Long argued, a note to procure them. To apply Long's own logic, blank parchment was so ordinary a commodity that it is hard to see why a special note would be needed for its procurement.
Long's last example is a clear case: the left-marginal note "3: B" occurs just as three blank charters are sealed after being reluctantly signed by three men of Dunstable (Anonymous 1929, 174b). The obvious interpretation is that it is an abbreviation denoting the three blanks needed for the scene. There is no opportunity to bring them on during the scene, so they must be readied at its start 104 lines before the marginal note, which therefore cannot be anticipatory. Even if the note refers to some other properties (three beeswax seals?), it is too late and Long must be right that its purpose was to record materials needed for the play rather than to regulate its performance.5 Thus three notes, "Booke", "Paper" and "3: B", turn out to be just as Long described them--procurement rather than readying notes6--but two others ("Shrevs Ready" and "A bed | for woodstock") are more likely to be readying than procurement notes and one ("Blankes") could be either. The evidence from the manuscript is not as clear-cut as Long suggested and substantial theatre-historical elements in his narrative about it are mistaken. How these matters ought to affect our understanding of the New Bibliographical characterization of promptbooks will be taken up in the conclusion.
Long's third essay on this theme was concerned with the manuscript playbook John a Kent and John a Cumber and argued that like the others it shows how little was done in the theatre to alter the dramatist's text, and that none of the things proposed by New Bibliography as the preparation of a promptbook was done with significant regularity (Long 1989). Errors in the manuscript of Thomas of Woodstock are strong evidence that it is a scribal copy of authorial papers (Anonymous 1929, vi-vii), and as Jackson pointed out this is certainly the case if Rowley is the author since the writing does not match his samples in Philip Henslowe's Diary (Jackson 2001, 50-51). Apart from its annotations, John a Kent and John a Cumber is entirely in the hand of Anthony Munday, who signed it at the end as an author would; after his signature someone else has written the date "Decembris 159x" (Munday 1923, 13b). The last digit of the date (represented here by x) is unclear and to the present author's eyes any of the various suggestions is possible. Greg read the date as "1595", Chambers saw "1596", I. A. Shapiro and Roslyn L. Knutson saw "1590" and Grace Ioppolo recently supported "1595" (Greg 1908, 172; Chambers 1923, 446; Shapiro 1955; Knutson 1984, 1; Ioppolo 2006, 101). Combining Shapiro's date with connections such as shared hands and actors' names that link John a Kent and John a Cumber to the manuscript playbook Sir Thomas More (British Library Harley 7368) and the playhouse plot of 2 Seven Deadly Sins, Long decided that all three "can be assigned to the apparently sporadic confederation of the Lord Admiral's and the Lord Strange's companies" in the early 1590s (Long 1989, 129).
David Kathman has since convincingly reassigned 2 Seven Deadly Sins to the Chamberlain's men in 1597-87 and Andrew Gurr has shown that there was no confederation of the Admiral's men and Strange's men in the early 1590s (Gurr 1993). However, connections between John a Kent and John a Cumber and Sir Thomas More remain: Munday composed and wrote out all the former and most of the latter, the two are bound with pieces of one leaf from a medieval vellum manuscript used as wrapping, and Hand C of Sir Thomas More (seen also in the plot of 2 Seven Deadly Sins) wrote the labels on both wrappers (Thompson 1919, 328; Munday 1923, vi; Greg 1931, 224-25). Long concluded that Sir Thomas More and John a Kent and John a Cumber were bound in their wrappers by Hand C around the same time. However, as John Jowett pointed out, wrapping plays in vellum was not usual theatre practice and the vellum manuscript in question (a compilation of canon law by bishop Bernardus Papiensis) was just the sort of thing Munday could have purloined from a Catholic household when hunting recusants in the 1580s and 1590s; this may well be how he acquired the manuscript Life of Thomas More by Nicholas Harpsfield, the major source for his play about More (Bergeron 2007; Jowett 2010).
Long designated Hand C of Sir Thomas More as Bookkeeper-1 of John a Kent and John a Cumber, since it made sense that a person who troubled himself to preserve books using wrappers and who wrote out playhouse plots was a theatre man concerned with production, although he might also have been a player. He need not have provided the wrappers, but at some point he wrote the labels on them. There is no reason to suppose that being the bookkeeper was any one person's sole occupation, since there seems to have been too little work to do. Bookkeeper-1 made all seven of the annotations in John a Kent and John a Cumber (Long 1989, 131). Just as he had argued happened with Thomas of Woodstock, Long saw Bookkeeper-1 solving specific problems in John a Kent and John a Cumber rather than undertaking a thoroughgoing theatricalization, and he went no further than a few marginal notes to make entrances and music happen earlier than the dramatist allowed (Long 1989, 133-35).
Repeatedly, however, Bookkeeper-1 did not move stage directions up one line when entering characters are spoken of as visible before their entrance direction. On three occasions identical in form to the anticipations highlighted by Long the bookkeeper did nothing:
. . . least you spill the conceit, for heere they come.
Enter Pembrook, Moorteon, Oswen, Amery, to them this crew
[JOHN A KENT]
And see where Iohn a Kent is first come in. .
Enter Iohn a Cumber lyke Iohn a Kent
[JOHN A CUMBER]
yonder is Iohn a Kent [I] heere Iohn a Cumber. .
Iohn a Kent in his owne habit, denvyle, Griffin, Powesse, Euan,
If Long's interpretation is right, we simply have to accept that Bookkeeper-1 was utterly inconsistent. Rather than regularize the script, as New Bibliography would predict, he solved a staging problem in respect of two entrances (for John a Kent on 2b and John a Cumber on 11b) and left the same problem unsolved for these three others. Indeed, the last of the above three examples occurs only ten lines after the second of the two occasions that Bookkeeper-1 fixed the problem, so we would have to accept that his inconsistency could manifest itself within a few lines.
What if Long is wrong, and the manuscript is witness to something other than a completed process of minor improvements that readied the play for performance? A year before Long's essay on John a Kent and John a Cumber appeared, T. H. Howard-Hill published his account of John Fletcher and Philip Massinger's Sir John Van Olden Barnavelt (British Library Additional Manuscript 18653), a scribal transcript by Ralph Crane containing a bookkeeper's annotations (Howard-Hill 1988). Howard-Hill concluded that the manuscript is a theatrical document used for making the actors' parts but not for running the play in performance. Although the manuscript has clearly been marked up with theatrical concerns, "further decisions about casting, parts, and stage movement would have been essential before the play could have been staged" (Howard-Hill 1988, 154). Crucially, chairs are brought in so the French ambassadors may sit while pleading with the Prince of Orange for Barnavelt's life, but nothing is provided for the Prince and his train to sit on; that they remain standing is impossibly indecorous and cannot reflect the final staging (Howard-Hill 1988, 166). If Howard-Hill is right, this manuscript has no place in Greg's binary taxonomy, but could find a home in Fredson Bowers's broader range of possibilities, which included such incomplete documents as "foul papers or fair copies partially marked by the prompter as a preliminary for transcription into prompt".8
Peter W. M. Blayney thought that the vellum binding of Sir Thomas More--made from the same manuscript used to bind John a Kent and John a Cumber and presumably created at the same time--was added before it was sent to Edmund Tilney, Master of the Revels, for licensing (Blayney 1972, 171), and we know that much needed to be done before Sir Thomas More could be acted (Greg 1911, 3a). Like that of Sir Thomas More, the manuscript of John a Kent and John a Cumber has an unusually precise and complete set of authorial stage directions, yet it too might also have been unready for performance, although its subject matter was inherently less alarming to a censor. Long believed that the presence of the wrapper indicated the bookkeeper's completion of his work on the manuscript: "Hand C-Bookkeeper 1 is . . . someone concerned with production of the play; his writing of the plots is the logical extension of his efforts to ensure a smooth production, as is his care of the books when not in use, as evidenced by his preparation of the wrappers" (Long 1989, 131). The link with Sir Thomas More undermines Long's assumption that the manuscript of John a Kent and John a Cumber is witness to a completed process of readying the script for performance.
The editorial consequences of Long's essays are important because they provided part of an empirical base for the widespread rejection of New Bibliography in the 1980s and 1990s and its replacement with what has become known as the New Textualism. This essay's revaluation of Long's work undermines his part of the empirical basis for that rejection. Long's attempts to discover the provenance of the manuscripts under discussion relied upon unsafe speculations about dates of performance and revival, upon which he built hypotheses regarding unknowables such as the state of company finances. His deduction of company stock from the wording of a stage direction was unreliable and his revival of the 'continuous copy' theory from early New Bibliography ran counter to the primary evidence: the manuscripts are predominantly scribal. Since Long was writing it has become clear that there often existed more manuscripts of each play than had been supposed by Greg and his followers, who assumed a norm of two: foul-papers and promptbook. For example, we now know that scribal transcripts provided copy for most of the 1623 Folio of Shakespeare (Taylor 1993a). There does not appear to have been much commercial production of play transcripts as a means of public dissemination (Howard-Hill 1999), so we should not discount the likelihood that playing companies made transcripts for their own purposes. Revival of the discredited 'continuous copy' theory served the same function within Long's logic as the binary taxonomy served within Greg's, enabling the inference of certain behaviours from the absence of countervailing evidence. That is to say, Long assumed that the manuscripts he examined represented all that was normally done to ready a play for performance because, in his view, there was usually only one manuscript for each play. That assumption is no longer viable.
Long's conclusions were illustrated with detail from just two manuscripts. He showed that certain annotations in Thomas of Woodstock that used to be understood as readying notes are in fact concerned with the procuring of properties for a production. But other annotations in that manuscript--"Shrevs Ready", "A bed | for woodstock", and perhaps "Blankes"--cannot safely be reassigned in this way. New Bibliographers were right to point to the existence of such readying notes in this manuscript, and in other manuscripts not considered by Long. For example, Thomas Dekker's The Welsh Ambassador (Cardiff Central Library Manuscript 4.12) has a great many notes in the form "bee redy . . ." followed by a character's name, and other manuscripts have similar annotations, as does a print copy of The Merry Milkmaids marked for performance.9 Long's analysis of John a Kent and John a Cumber showed that a bookkeeper altered the timing of a few of the dramatist's stage directions, but the bookkeeper was inconsistent in his behaviour if, as Long believed, he had completed the task of fixing problems he found in the manuscript. However, if there were multiple transcripts of each play we should be cautious in supposing that these two manuscripts were used to run performances. It would follow from the presence of Buc's marks in Thomas of Woodstock that this manuscript was the licensed book, which New Bibliographers assumed was the promptbook. But Gurr may be right that the licensed book was too valuable to be used directly to run performances and was kept as a reference document from which a promptbook would be transcribed as needed (Gurr 1999). Without more evidence, just what the manuscripts of Thomas of Woodstock and John a Kent and John a Cumber were used for cannot be determined.
Long's essays were a necessary corrective to a prevailing editorial over-confidence in determining (with the aid of simplified rules from Greg) the manuscript provenance of early editions of plays produced in the late-sixteenth and early-seventeenth centuries. Greg inferred the characteristics of playbooks by combining knowledge of the manuscripts he had seen--not nearly all those he could have seen--with a set of hypotheses about performance practices, and, crucially, a survey of anomalies in print editions. This last area of evidence is largely absent from critiques of Greg's work on manuscripts. The famously absurd repetitions in Q2 Romeo and Juliet--Romeo and Friar Laurence giving essentially the same "grey eyde morne" speech and Romeo apparently poisoning himself twice (Shakespeare 1599, D4v, L3r)--are virtually unperformable and best understood as authorial false starts that would have disappeared before creation of a script that the actors could use. There is nothing like these repetitions in the surviving playbook manuscripts, and it is reasonable to call them signs of the author in Q2's underlying manuscript. Equally, there exist features in early print editions that are difficult to attribute to anyone but a scribe, such as the use of Latinate markers (of the kind "Actus Quintus") and stage directions that open a scene by listing everyone in it regardless of when they enter. Long's conclusions have found favour within a general movement to embrace indeterminacy in early-modern studies, on the assumption that he had proved the impossibility of making any determination of the copy underlying an early printing. In fact, Long illustrated only that over-confident determinations had been made on too-slight evidence, and that where there are no manifest impossibilities in a script (such as those in Q2 Romeo and Juliet) authorial papers cannot easily be distinguished from papers used to run a performance.
Long's essays on the surviving manuscript playbooks was widely received as a successful attack upon New Bibliography, but Greg and Bowers were both New Bibliographers and their mutually incompatible ideas could not all be undermined at once. As a critique of Greg's account of the making of a promptbook, Long's interpretation assumed that the surviving documents represent the full extent of what was done in readying plays for the stage. Even Greg would admit that the making of a promptbook might not be completed (as it almost certainly was not in the case of Sir Thomas More), so, without further evidence that Long was looking at documents ready to be used to run a performance, his claim that promptbooks could be irregular and untidy is unproven. Equally unproven, of course, is the assumption that they were regular and tidy. Long's critique relied on the binary notion that a document is either untouched by theatrical annotation or ready to be performed. If Bowers was right and there existed many kinds of play manuscripts in various states of precision, consistency, and completeness, rather than just one or two, Long may have mistaken for promptbooks documents of another kind. Howard-Hill's reinterpretation of the manuscript of Fletcher and Massinger's Sir John Van Olden Barnavelt as an intermediate theatrical document, used to make the parts but not to run the performances, gives a concrete illustration of the vulnerability of Long's logic.
* I am grateful to the Huntington Library in San Marino, California, for a fellowship in 2008 that made possible the research presented in this article.
1Among the many important studies that have approvingly cited Long's essays on this topic are Werstine 1988; Werstine 1990; Trousdale 1990; De Grazia & Stallybrass 1993; Tipton 1998; Gurr 1999; Kidnie 2000; Werstine 2001; Edelman 2002; Calore 2003; Tribble 2005; and Keefer 2006.
2The latest Arden Shakespeare Hamlet credits Long with having "undermined confidence that a text containing indeterminate stage directions necessarily pre-dates theatrical performance" (Shakespeare 2006, 501), as does the latest Arden Pericles (Shakespeare 2004, 84). Long's account of the manuscript of Thomas of Woodstock, challenged in the present essay, was used by the editor of the latest Arden editor of Richard II as a brake upon the editorial search for authority in variant readings (Shakespeare 2002, 524-25). The Oxford Shakespeare 3 Henry VI credits Long with overturning R. B. McKerrow's idea that manuscripts used in the theatres had to be tidy (Shakespeare 2001, 100) as does the Oxford Shakespeare Richard III (Shakespeare 2000, 123); for both editors Long's essays are central to diagnoses of the copy for early editions. These examples could be multiplied from other recent editions of Shakespearian and non-Shakespearian plays. The importance that play editors attach to Long's essays justifies the critique they receive here.
3The manuscript does not name the play, and the Malone Society's editor settled on a "compromise between conflicting claims" for its title, arising from the subject matter (Anonymous 1929, v). Michael Egan recently renamed it to express an opinion that has not gained significant support (Egan 2006).
4The convention of representing by roman, italic, and boldface type the different hands and styles of script found in a manuscript is not followed in this essay. Where such distinctions are relevant to the argument they are addressed explicitly in the main text. Manuscripts are here quoted from Malone Society Reprint editions wherever possible, using leaf number and side (a or b).
5Tiffany Stern read this note and the preceding one for "Blankes" as instructions to a "scroll-scribe"--whose job it was to make the property documents needed for each play--telling him to set aside the necessary materials (Stern 2009, 188-89). This play requires no such documents to be read aloud, so it is unlikely a scribe was asked to read the whole manuscript just to find two notes that do not call for his specialist skills, and we should assume that they were treated like the other properties called for in the annotations.
6It remains a mystery why a bookkeeper would consider it convenient to bury procurement notes in the middle of a manuscript rather than placing them at the beginning or end, where they might easily be recovered later.
7Kathman 2004. Two scholars have independently attempted to dismiss Kathman's crucial discovery that the Thomas Belt mentioned in the plot of 2 Seven Deadly Sins was apprenticed to the Chamberlain's man John Heminges on 12 November 1595, which Kathman took as proof that the plot post-dates this event (Gurr 2007, 82-84; Manley 2007). Since Belt was sixteen at the time of his apprenticeship--rather old for a boy to begin acting--they conclude that he probably was already taking small roles, as demanded of him in this plot, and hence the plot could date from the early 1590s.
8Bowers 1955, 11. Leslie Thomson found evidence that a bookkeeper left incomplete his labour of annotating for performance a quarto of The Two Merry Milkmaids printed in 1620, and concluded that "we assume more than we should about the meaning of 'marked for performance'" (Thomson 1996, 183, 190, 198-199).
9Thomson 1996. Stern's account of such readying notes treated the promptbook and the plot as a matched pair, so that the more the bookkeeper chose to put in the latter the less he needed to record in the former, which would help explain the apparently inconsistent readying notes in some manuscripts (Stern 2009, 227-31). However, her suggestion that plots were used like eighteenth-century call-sheets, directing messenger-boys to fetch actors, is unlikely because pre-Commonwealth theatre tiring houses were too small for this to be necessary (219-27). Michela Calore and Evelyn Tribble's analyses more plausibly situated plots within an integrated knowledge system (including the playhouse fabric itself) that facilitated the feats of memory called for by a high-turnover repertory schedule (Calore 2003; Tribble 2005).
Anonymous. 1929. The First Part of the Reign of King Richard the Second, or Thomas of Woodstock. Ed. Wilhelmina P. Frijlinck. Malone Society Reprints. London. Malone Society.
Bergeron, David M. 2007. "Anthony Munday (Baptised 1560, Died 1633)." doi:10.1093/ref:odnb/19531. Oxford Dictionary of National Biography. Edited by H. C. G. Matthew, Brian Harrison and Lawrence Goldman. Oxford. Oxford University Press. n. pag..
Blayney, Peter W. M. 1972. "The Booke of Sir Thomas Moore Re-examined." Studies in Philology 69. 167-91.
Bowers, Fredson. 1955. On Editing Shakespeare and the Elizabethan Dramatists. Philadelphia. University of Pennsylvania Library.
Calore, Michela. 2003. "Elizabethan Plots: A Shared Code of Theatrical and Fictional Language." Theatre Survey 44. 249-61.
Chambers, E. K. 1923. The Elizabethan Stage. Vol. 3. 4 vols. Oxford. Clarendon.
Chambers, E. K. 1924-25. "'The Disintegration of Shakespeare': The British Academy Annual Shakespeare Lecture Read 12 May 1924." Proceedings of the British Academy 11. 89-108.
De Grazia, Margreta and Peter Stallybrass. 1993. "The Materiality of Shakespeare's Text." Shakespeare Quarterly 44. 255-83.
Edelman, Charles. 2002. "The Battle of Alcazar, Muly Molocco, and Shakespeare's 2 and 3 Henry VI." Notes and Queries 247. 215-18.
Egan, Michael, ed. 2006. The Tragedy of Richard II Part One: A Newly Authenticated Play By William Shakespeare. 3 vols. Ceredigion. Edwin Mellen.
Greg, W. W. 1955. The Shakespeare First Folio: Its Bibliographical and Textual History. Oxford. Clarendon Press.
Greg, W. W., ed. 1908. Henslowe's Diary. Vol. 2: Commentary. 2 vols. London. Bullen.
Greg, W. W., ed. 1911. The Book of Sir Thomas More. Malone Society Reprints. Oxford. Malone Society.
Greg, W. W., ed. 1931. Dramatic Documents from the Elizabethan Playhouses: Stage Plots, Actors' Parts, Prompt Books. Vol. 1: Commentary. 2 vols. Oxford. Clarendon Press.
Gurr, Andrew. 1993. "The Chimera of Amalgamation." Theatre Research International 18. 85-93.
Gurr, Andrew. 1999. "Maximal and Minimal Texts: Shakespeare V. the Globe." Shakespeare Survey 52. 68-87.
Gurr, Andrew. 2007. "The Work of Elizabethan Plotters, and 2 Seven Deadly Sins." Early Theatre 10.1. 67-87.
Howard-Hill, T. H. 1988. "Crane's 1619 'Promptbook' of Barnavelt and Theatrical Processes." Modern Philology 86. 146-70.
Howard-Hill, T. H. 1999. "'Nor Stage, nor Stationers Stall Can Showe': The Circulation of Plays in Manuscript in the Early Seventeenth Century." Book History 2. 28-41.
Ioppolo, Grace. 2006. Dramatists and Their Manuscripts in the Age of Shakespeare, Jonson, Middleton and Heywood: Authorship, Authority and the Playhouse. Routledge Studies in Renaissance Literature and Culture. London. Routledge.
Jackson, MacDonald P. 2001. "Shakespeare's Richard II and the Anonymous Thomas of Woodstock." Medieval and Renaissance Drama in England 14. 17-65.
Jowett, John. 2010. 'The Wrappers for John a Kent and John a Cumber and Sir Thomas More': Private Communication to the Author, 16 March.
Kathman, David. 2004. "Reconsidering The Seven Deadly Sins." Early Theatre 7.2. 13-44.
Keefer, Michael H. 2006. "The A and B Texts of Marlowe's Doctor Faustus Revisited." Papers of the Bibliographical Society of America 100. 227-57.
Kidnie, Margaret Jane. 2000. "Text, Performance, and the Editors: Staging Shakespeare's Drama." Shakespeare Quarterly 51. 456-73.
Knutson, Roslyn L. 1984. "Play Identifications: The Wise Man of West Chester and John a Kent and John a Cumber; Longshanks and Edward I." Huntington Library Quarterly 47. 1-11.
Long, William B. 1985a. "'A Bed / for Woodstock': A Warning for the Unwary." Medieval and Renaissance Drama in England 2. 91-118.
Long, William B. 1985b. "Stage-directions: A Misinterpreted Factor in Determining Textual Provenance." TEXT: Transactions of the Society for Textual Scholarship 2. 121-37.
Long, William B. 1989. "John a Kent and John a Cumber: An Elizabethan Playbook and its Implications." Shakespeare and Dramatic Tradition: Essays in Honor of S. F. Johnson. Edited by W. R. Elton and William B. Long. Newark DE. University of Delaware Press. 125-43.
Long, William B. 1999. "'Precious Few': English Manuscript Playbooks." A Companion to Shakespeare. Edited by David Scott Kastan. Blackwell Companions to Literature and Culture. Oxford. Blackwell. 414-33.
Manley, Lawrence. 2007. "Thomas Belte, Elizabethan Boy Actor." Notes and Queries 252. 310-13.
Middleton, Thomas. 1657. Two New Playes, Viz. More Dissemblers Besides Women [And] Women Beware Women. Wing M1989. London. For Humphrey Moseley.
Middleton, Thomas. 1975. Women Beware Women. Ed. J. R. Mulryne. The Revels Plays. London. Methuen.
Munday, Anthony. 1923. John a Kent and John a Cumber. Ed. Muriel St Clare Byrne. Malone Society Reprints. London. Malone Society.
Pollard, A. W. 1917. Shakespeare's Fight with the Pirates and the Problems of the Transmission of His Text. London. Moring.
Shakespeare, William. 1599. [Romeo and Juliet] The Most Excellent and Lamentable Tragedie, of Romeo and Juliet. STC 22323 BEPD 143b (Q2). London. Thomas Creede for Cuthbert Burby.
Shakespeare, William. 1623. Comedies, Histories and Tragedies. STC 22273 (F1). London. Isaac and William Jaggard for Edward Blount, John Smethwick, Isaac Jaggard and William Aspley.
Shakespeare, William. 1952. The Third Part of King Henry VI. Ed. John Dover Wilson. The New Shakespeare. Cambridge. Cambridge University Press.
Shakespeare, William. 2000. Richard III. Ed. John Jowett. The Oxford Shakespeare. Oxford. Oxford University Press.
Shakespeare, William. 2001. Henry VI, Part Three. Ed. Randall Martin. The Oxford Shakespeare. Oxford. Oxford University Press.
Shakespeare, William. 2002. King Richard II. Ed. Charles R. Forker. The Arden Shakespeare. London. Thomson Learning.
Shakespeare, William. 2004. Pericles. Ed. Suzanne Gossett. The Arden Shakespeare. London. Thomson Learning.
Shakespeare, William. 2006. Hamlet. Ed. Ann Thompson and Neil Taylor. The Arden Shakespeare. London. Thomson Learning.
Shapiro, I. A. 1955. "The Significance of a Date." Shakespeare Survey 8. 100-05.
Stern, Tiffany. 2009. Documents of Performance in Early Modern England. Cambridge. Cambridge University Press.
Taylor, Gary. 1993a. "Post-script." Shakespeare Reshaped, 1606-1623. Edited by Gary Taylor and John Jowett. Oxford Shakespeare Studies. Oxford. Clarendon Press. 237-43.
Taylor, Gary. 1993b. "The Structure of Performance: Act-intervals in the London Theatres, 1576-1642." Shakespeare Reshaped 1606-1623. Edited by Gary Taylor and John Jowett. Oxford Shakespeare Studies. Oxford. Clarendon. 3-50.
Thompson, Edward Maunde. 1919. "The Autograph Manuscripts of Anthony Mundy." Transactions of the Bibliographical Society. 1st Series 14. 325-53.
Thomson, Leslie. 1996. "A Quarto 'Marked for Performance': Evidence of What?" Medieval and Renaissance Drama in England 8. 176-210.
Tipton, Alzada J. 1998. "'The Meanest Man . . . Shall be Permitted Freely to Accuse': The Commoners in Woodstock." Comparative Drama 32. 117-45.
Tribble, Evelyn. 2005. "Distributing Cognition in the Globe." Shakespeare Quarterly 56. 135-55.
Trousdale, Marion. 1990. "A Second Look at Critical Bibliography and the Acting of Plays." Shakespeare Quarterly 41. 87-96.
Werstine, Paul. 1988. "'Foul Papers' and 'Prompt Books': Printer's Copy for Shakespeare's Comedy of Errors." Studies in Bibliography 41. 232-46.
Werstine, Paul. 1990. "Narratives About Printed Shakespeare Texts: 'Foul Papers' and 'Bad' Quartos." Shakespeare Quarterly 41. 65-86.
Werstine, Paul. 2001. "All's Well That Ends Well and Editorial Constructions of 'Foul Papers'." Archiv für das Studium der neueren Sprachen und Literaturen 238. 19-39.