Quadrat DemonstrandumVickers, Sir Brian. The One “King Lear.” Cambridge, MA: Harvard University Press, 2016. xxiii, 387 pp. Cloth, $45.00 (ISBN 978-0-674-50484-4).
2017; Bibliographical Society of America; Volume: 111; Issue: 1 Linguagem: Inglês
10.1086/690603
ISSN2377-6528
Autores Tópico(s)Shakespeare, Adaptation, and Literary Criticism
ResumoPrevious articleNext article FreeReview EssayQuadrat Demonstrandum Vickers, Sir Brian. The One “King Lear.” Cambridge, MA: Harvard University Press, 2016. xxiii, 387 pp. Cloth, $45.00 (ISBN 978-0-674-50484-4).Peter W. M. BlayneyPeter W. M. Blayney Search for more articles by this author PDFPDF PLUSFull Text Add to favoritesDownload CitationTrack CitationsPermissionsReprints Share onFacebookTwitterLinked InRedditEmailQR Code SectionsMoreIf we limit the definition of bibliography to the study of printing and the book trade in early modern England (as many do), it must be acknowledged that the goal of the major pioneers of that discipline, from Edward Arber to Charlton Hinman, was to advance the study of Shakespeare. In the 1960s, indeed, it was a simple fact that university-level bibliography was taught only in Departments of English, not History. Many Shakespeareans therefore think of bibliography as merely a subordinate subject whose rudiments they were taught as novices—and for many their knowledge remains rudimentary. But those who speak with authority in their own field sometimes begin to believe themselves equally expert in all things early modern, as I learned well during sixteen years as a daily reader at the Folger Shakespeare Library. Those who devote most of their time to studying and teaching Shakespeare can seldom spare the 10,000 hours it reputedly takes to achieve mastery in a second discipline.I first met King Lear in 1969 while collating the text and accidentals of the 1608 Quarto and and its reprint of 1619. Only afterwards did I discover (first) what the Folio had done to the text in 1623 and (second) what editors had subsequently created by conflating the two versions. In 1972 I began a closer study of Q1 by seeking what else Nicholas Okes and his predecessors had produced in 1605–9. Available sources listed nearly sixty items—of which several proved to contain sections printed by others. I had already planned to search all anonymously printed books. But if others contributed to books bearing Okes’s name it seemed likely that he sometimes contributed to their books, so I began examining as many books of 1605–9 as I could access.1 By the time I submitted the typescript for Origins to Cambridge University Press in 1978 my list had grown to eighty-eight books, twenty-seven of which were the work of more than one press, and I had examined every page of every accessible copy or fragment of those books (more than 500 items).2 I had also paged through every copy in Cambridge of Okes’s later books and several hundred other Jacobean books both there and elsewhere. After a hiatus I returned to my real interests in 1984, since when my full-time occupation has been scrutinizing English books of 1501–1616 and studying the history and records of the Stationers’ Company.While I do not question the eminence of Brian Vickers in his own field, analytical bibliography is not his forte. He has read quite widely among the bibliographers considered “canonical” by literary scholars. But some of those he cites and follows are less reliable than he believes, and there is anyway a fundamental difference between reading modern accounts of how early modern books were supposedly printed and actually examining them. Even role-models such as W. W. Greg, Charlton Hinman, and D. F. McKenzie could and did make mistakes, while a 2007 essay by Adrian Weiss that Vickers justly describes as “admirably concise and comprehensive” is by no means consistently accurate and has misled him on several occasions.3It is symptomatic that Vickers cannot coherently describe the Lear problem in his preface. There are important differences between the Quarto of 1608 (Q) and the text included in the Folio in 1623 (F). According to Vickers, “the Quarto lacks 102 lines … not found in the Folio, whereas the Folio lacks 285 lines … not found in the Quarto” (ix). But if each of those 387 lines (and “also many smaller phrases and single words”) is simultaneously lacked by one text and not found in the other, where on earth was it found? It is an undeniable historical fact that in 1623 the Folio both omitted 285 lines that had already been printed and added 102 that had not. Convinced that those 102 lines must have been part of the only Shakespearean text there ever was, and that they are missing from Q only because the printer chose to omit them, Vickers seems unwilling to call them “added” even when trying to explain the problem.Nicholas Rowe knew nothing of Q when he edited the plays from the 1685 Fourth Folio, first in 1709, and with revisions five years later. In 1725 Alexander Pope became the first editor to incorporate many of the Q-only lines from a copy of Q2 (1619). But he overlooked several passages (including the servants’ dialogue from the close of Act 3) and printed one only as a footnote. In 1733 Lewis Theobald went a step or two further, but one passage in 1.2 remained excluded until 1770, when Charles Jennens finally produced a Lear that can be called completely conflated. Vickers asserts that before the publication of what he calls “the manifesto” of those who distrust conflation, “For nearly four hundred years this version gave us ‘the bitter-sweet of this Shakespearian fruit,’ as Keats described it” (x).4 But even counting from Pope rather than Jennens, 1983 minus 1725 is nowhere near 400.The early conflaters had done no meticulous textual research. Pope and his successors took the word of Heminge and Condell that the plays included in the earliest Folios were “perfect of their limbes; and … absolute in their numbers,” and assumed (unlike Rowe) that the seven plays added in 1664 were not covered by that warranty. But if earlier editions contained additional lines that they considered credibly Shakespearean (a flexible definition before a canon had been established) they included them anyway.During the two centuries after the Jennens Lear, literally millions of copies of conflated editions were printed. They differed in minor respects according to the theories or whims of their actual or purported “editors,” but all depended on the belief that Shakespeare must have intended all the extant lines to coexist in a single play. That was the version in which almost everyone thereafter first encountered King Lear: the version on which the entire critical tradition was based from the 1770s to the 1970s. That was the version that Vickers had known, loved, and studied for over twenty years before anyone had the temerity to point out that the allegedly bibliographical arguments devised to explain and defend the editorial tradition had two things in common. They were not really bibliographical and did not fit the evidence.For anyone as intimately familiar with the composite play as is Vickers, it cannot be easy to read either of its sources as anything but a defective version of it. Wherever customary words are absent, memory will silently supply both the line and the accepted reasons why it “belongs” there. With the Folio that might perhaps be defensible, because whatever it omits is known to have existed in 1608. But there is absolutely no historical, bibliographical, or textual evidence that most of the 1623 additions had yet been written in 1608, let alone that they were present in the printer’s copy for the Quarto.In 1976 Michael Warren gave a conference paper on the ways in which F’s alterations affect the characters of Albany and Edgar, and soon learned that at least three graduate students were also independently doubting the validity of conflation: Georgia Peters Burton, Steven Urkowitz, and myself. Our conclusions influenced several others (including Gary Taylor, who had been shown some draft material I lent to someone else), and when four of the growing group participated in a seminar at Boston in 1980, the suggestion that the two texts might be incompatible began to spread with unexpected speed. I doubt that either Taylor (the loudest opponent of conflation) or Vickers (whose fellows are “some of the leading scholars and editors of our time”) really knows which side currently has the most support, although when the latter twice calls the two-text view “the new orthodoxy” (x, 310) he may provide a clue.5In Vickers’s opinion “The revisionists’ case rests entirely on their claim that Shakespeare himself made the Folio cuts,” and he argues that even if Warren’s description of the differences between the respective Albanys is accurate,It is … true that cuts will affect some roles more than others, but the Two Versions theorists illicitly infer intention from effect. No evidence exists that Shakespeare intended to change the roles of Albany and Edgar, and several experienced scholars have argued that had he meant to do so he would not have been satisfied with cuts alone, but would have added new material(xiv).The contributors to The Division of the Kingdoms do seem inordinately insistent that revision by Shakespeare himself is the only possible (or acceptable) solution, and that “revised” must mean “superior.” As one of the founding “Two-Versions theorists” I can assure Vickers that I have never believed that the person responsible for the Folio adaptation (the word I prefer) was the man who wrote the original play. But the questions of who made the changes and whether they are improvements are quite distinct from the basic problem, which is simply whether or not the two primary texts are compatible. Moreover, those who assert that the version on which two centuries of criticism depend must necessarily be the original and best are equally guilty of inferring intention from effect. What else have they from which to infer it?The Folio’s emasculation of Albany is not simply an inevitable consequence of cutting. The severe abridgement of his quarrel with Gonorill in 4.2.29–68 takes the equivalent of twenty-five verse lines from him but only eight from her. It therefore allows Albany almost no opportunity to display the dogmatic moral rectitude that leaves him the undisputed “winner” of the argument in Q. It leaves only an exchange of two brief insults each, and Gonorill gets the last word. In 5.1 Albany’s first speech is cut from eight to three and a half lines. He is then interrupted by a half-line from Regan that is followed by two and a half lines in which Gonorill attempts to take control. Had the primary motive for all those cuts been to make F more “theatrical” by accelerating the action and pruning unnecessary verbiage, one could certainly assume that the diminution of Albany was just an inevitable side-effect. But in the first act the Folio adds to his line-count, and I cannot agree with Warren’s suggestion that “Albany, who is bewildered and ineffectual in either text, is more patently so in Q, where he is given no opportunity to urge patience [or] to warn Goneril of the unwisdom of her acts.”6Albany’s first addition is spoken in unison with Cornwall at 1.1.162, when they try to interrupt Lear’s furious treatment of Kent with “Deare Sir, forbeare.” His second is the interjection of “Pray Sir be patient” at 1.4.261, between the closing half-line of one of Lear’s rants and the complementary half-line that begins Gonorill’s response. Neither interpolation fits the metrical context, and both go unheeded by Lear (unsurprisingly, if they are new since 1608). And I cannot see how having his first two utterances completely ignored makes F Albany less ineffectual than his Q counterpart—although not knowing how the quarrel began he is certainly “bewildered” in both versions. After Lear and the Fool have left, a longer addition in F allows Gonorill to try justifying her actions to Albany. He fails to react as desired and responds only “Well, you may feare too farre” (1.4.328). This indecision prompts a six-line lecture from his wife, and the overall result (as in 4.2) is to strengthen her at his expense.As the two prepare to leave the scene the Folio makes one more change, traditionally dismissed as merely the correction of an auditory error in Q (the alleged similarity of pronunciation having first been deduced from the variant). But while Q’s doctrinaire moralist tells Gonorill that “Striuing to better ought [anything], we marre whats well,” his weakened F counterpart offers only the less decisive “Striuing to better, oft we marre what’s well.” (If it ain’t broke, often don’t fix it.) At the other end of the play we find changes of yet another kind, when two speeches spoken by Albany in Q are reassigned to Edgar. The first is the urgent order to whomever is sent to rescue Cordelia, “Hast thee for thy life.” The second is the final speech of the play: four lines that implicitly identify their speaker as the person now in charge of the kingdom’s future.If the omissions, additions, alterations, and speech reassignments involving Albany all appear to point the same way, it may not be completely naive to suspect the presence of an intention.Vickers suggests that “If you were to complete either version by adding the passages preserved by the other, you would have, in terms of characters and events, two identical plays” (ix), but that is untrue. If one believes (as both Vickers and I do) that Q was printed from an autograph and authoritative manuscript, then at the end of the play “Breake hart, I prethe breake” is Lear’s dying line rather than Kent’s reactive one, and the final speech implicitly leaves the reunited kingdom in the hands of the only legal successor (Albany) rather than Edgar. Those two attributions alone necessarily influence one’s view of what the preceding drama “has been about,” and they are by no means the only significant differences that are neither omissions nor additions. But one can always weasel one’s way out of Quarto readings one dislikes by dismissing them as simply evidence of “the execrable printing of the Quarto” (ix)—thus offering the sincerest form of flattery to the acumen of Pope, Hanmer, and Warburton.For Pope and his immediate successors, conflation was simply the way to include as many lines as possible that had been published under Shakespeare’s name. All men of letters knew that texts of all kinds and of all dates could be preserved in imperfect forms. There seemed nothing exceptional about each version of Lear having “lost” different groups of lines. Not until Victorian times was it felt important to wonder how each version had acquired its distinct contents. The Folio’s omissions were easy enough to explain away, because cutting scripts for performance was a common theatrical practice. To explain the lines found only in the Folio was less easy until the twentieth century invented “memorial contamination”—an infinitely flexible means of corruption that could be “deduced” from almost any kind of departure from a reading preferred by the investigator. And so the Lear of 1608 was often dismissed as a “Bad Quarto,” despite being longer than the allegedly superior (but “abridged”) Folio text.In 1931 Madeleine Doran drew attention to evidence suggesting that Q was probably printed from an autograph manuscript.7 Unlike her less convincing notions about the printer’s copy for F, that part of her work has been widely respected, even by those most insistent that F represents a more finished and improved version of the same play. It also allows ardent conflaters to argue that Q’s 285 lines (supposedly cut by irresponsible actors) are themselves textually credible. But Vickers is reluctant to believe that anyone could possibly have decided to add anything to the One True Text between 1608 and 1623. He has therefore devised a theory that allows the lines found only in F to have been cut from Q by a different but equally irresponsible agent: the printer Nicholas Okes.“King Lear at the Printer”Before explaining his hypothesis, Vickers devotes a chapter to a discussion, “based on the standard scholarship, which I synthesize as clearly as I can,” to ensure “that readers understand the terms and technical processes that feature in subsequent discussions” (3). But he is himself no bibliographer, as he has already demonstrated in the preceding “Note on References,” where he claims that “G3v27 refers to the third page, verso, of sheet G, line 27” (xxi). Anyone really familiar with the standard scholarship could have told him that G3v (more properly G3v) is the sixth page of quire G, namely the verso of the third leaf. The difference between a leaf and a page is as elementary as that between an act and a scene.The word stationer did not combine “in varying proportions, the roles of printer, bookseller, and publisher” (4). As a common noun it meant simply bookseller; when used to mean a freeman of the Stationers’ Company it did not imply anything about his particular trade (printer, bookseller, typefounder, binder, claspmaker, or something entirely unconnected with books).Shakespeare wrote King Lear for the King’s Men: an acting company which (we are told) “according to the conventions of early modern publishing, thus owned the right in the copy” (4). But although the players certainly owned the manuscript(s) they bought, and while the legally unenforceable customs of their profession gave them the exclusive right to perform the play (at least in London), by the rules of the book trade the Company of Stationers could grant the exclusive right to publish it to any free Stationer who legally acquired a manuscript text. The players had no more right to publish a play than did the Stationers to perform it.“When a stationer had acquired the right to publish a book he could either print it himself, if he had a large enough business, or hire a printer” (8). If he was a printer he could print it no matter how small his printing house; if he wasn’t, he had to hire a printer no matter how large his business.The definition of format that Vickers quotes from Weiss is both clumsily expressed and imprecise: “the number of type-pages printed on one side of a sheet, which is folded one or more times” (8). But the folded sheet does not have to have text printed on every page (or indeed, on any): sheet I of Pericles Q1 has only three type-pages on each side (leaf I4 is blank), but is nevertheless quarto, not tertio. The format is simply and definitively the number of leaves into which a sheet is folded.Weiss has also misled Vickers about paper, causing him to suggest that “the printer was ‘invariably paid by the publisher’ [my italics] either for the paper alone or ‘for both paper and printing’.” (8). But in the first of those supposed alternatives the printing is apparently not paid for at all. In reality there was another option, namely for the publisher to supply the paper himself and pay the printer only for the printing, and on pp. 98–100 of Origins I have explained why that seems likely to have happened with Lear.The composing stick described and illustrated by Joseph Moxon in 1683 would hold only six lines of pica,8 not the eight suggested by Vickers on p. 10, where endnote 27 leads us to the mistaken suggestion that “earlier in the century the stick would not have had an adjustable slide” (351). This probably derives from a misunderstanding of Philip Gaskell’s comment (citing Christopher Plantin) that “Wooden composing sticks without slides were used for bookwork in the sixteenth century.”9 Gaskell, however, neither said nor meant that all European compositors (or even most of them) used only fixed wooden sticks before 1683. English printers were certainly using Moxon-style sticks (which had two adjustable slides, not just one) at least in and after the 1560s. If Vickers were to inspect Okes’s other books of 1607–8 he would find items whose measure (not counting the shoulder notes found in more than a third of them) ranges from 40 mm (a duodecimo with framed pages) to 223 mm (the full-width preface to a two-column broadside).10 The most common sizes range through more than a dozen steps from 53 to 104 mm.Weiss leads Vickers into an even more important error when he claims, “The reference standard for type-body width was based upon the letter ‘m’, the widest letter. Blanks in this width were termed em-quadrats, em-quads, or simply ems.”11 At this point Weiss inserts a footnote (26):This width was unique to each individually designed typeface, so no universal value for a given size category of typefaces was possible. For example, different pica roman typefaces differed in the width of ‘m’, so it was not possible to state a width such as ‘22mm’ that would apply to all the different designs in the pica size.No pica “m” ever even approached 22 mm wide (perhaps a misprint for 2.2 mm), but the note is anyway wholly untrue. It probably stems from a misinterpretation of one of Moxon’s remarks:It is generally observed by Work-men as a Rule, That when they Cast Quadrats they Cast them exactly to the Thickness of a set Number of m’s or Body, viz. two m’s thick, three m’s thick, four m’s thick, &c.(171)Moxon’s meaning would have been clearer if he had glossed plural “m’s” by plural “Bodies” rather than singular “Body”—but what he here spells with a single lower-case roman “m” is the word now usually spelt “em,” and which meant then what it still means today: the body-height of the fount, not the actual width of either an upper- or lower-case “m.” As he explains less ambiguously on p. 103 when discussing punches:some are m thick; by m thick is meant m Quadrat thick, which is just so thick as the Body is high: Some are n thick; that is to say, n Quadrat thick, viz. half so thick as the body is high.The name presumably originated in the early decades of printing, because in many blackletter faces the widest letters (especially textura capitals M and W) are roughly a body-height wide, which is not true in roman faces. The en, however, was probably never the actual width of either kind of N: just a nickname because a lower-case “m” has two arches and an “n” only one.To return to Vickers’s own mistakes, on p. 13 we are told that “Once printed, the damp sheets would be hung along the center crease for drying, until the reverse side could be printed.” This too is completely wrong. As sheets came off the press they were stacked in a heap. Before perfecting it was important to leave that heap for long enough to reduce the ability of the ink to set off on the tympan-cloth of the press—but just as important to prevent the wet paper from drying enough to begin shrinking, which would make it difficult to ensure proper register of the pages on either side of each leaf.12 The sheets were not hung up to dry until both formes had been printed.It is an exaggeration to claim that Charlton Hinman “identified” the Folio proofreader as Isaac Jaggard (16). He did spend several pages trying to justify the claims that “we may consider it very probable that most if not all of the proof-reading … was done by one man; and probable also that this man was Isaac Jaggard.”13 But on the previous page he had more realistically admitted that “we do not and cannot positively know who the proof-reader … was, or even that there were not two or more readers”, and after much speculation the case for Isaac is acknowledged in a footnote to boil down to the fact that “no other member of the firm who is likely to have acted in a supervisory capacity is known”. The only other members of the firm who are known at all are William Jaggard (blind) and John Leason (William’s teenage apprentice), so Isaac’s actual role can be no more than an unsupported guess.Those pages, however, may have inspired another of Vickers’s quirks. Bibliographers customarily attribute the printing of a book to whoever occupied the printing house, owned the printing materials, and both hired and paid the workmen, but Vickers has taken literally the imprint of the Folio, which reads, “Printed by Isaac Iaggard, and Ed. Blount.” Like many hundreds of early modern imprints,14 this means what we would now express as printed for those men: Blount was never a printer, and Isaac did not become the master of the Jaggard printing house until his father died at about the time those words were printed. Vickers usually equivocates, referring only to an unspecified “Jaggard” or “the Jaggards” (and indexing them only as “Jaggard, William and Isaac”), but on pp. ix and 309 he explicitly attributes the printing of Folio Lear to Isaac rather than William (whose forename he never uses outside the index). Isaac was certainly the Jaggard who published the Folio, but William printed it.Vickers suggests on p. 20 that when uncorrected Q reads “three snyted” at E1r13 where “three suyted” is required, the reading is “probably a simple case of a turned letter, ‘n’ for ‘u’, letters that were in contiguous sort boxes and often confused.” But if the letter is an inverted “u” it is merely upside-down rather than “confused,” and the lay of the case is irrelevant. Given that “u” and “n” are usually easy to identify, why has he not looked closely enough to tell us which letter it actually is? (And because it really is an “n,” why should we rule out a simple misreading of minims?)A few pages later Vickers is misled by a statement of my own. By 1979 the contents of the intended second volume of Origins had been outlined, and several chapters and sections drafted. On p. 144 of the first volume I unwisely inserted a hint about one of those sections but deliberately declined to explain it. Vickers quotes that sentence on p. 28: “And what Lear used in the quantities most unprecedented in the books of 1605–07 was space-metal.” But what he offers as an explanation is quite unconnected with what I meant: “Not being able to set the requisite amount of space after punctuation marks or between words, the compositors left all too many instances of type packed together.”There are two entirely separate issues here, and were Vickers more familiar with early modern English books he would know that declining to use a space after a punctuation mark was quite unexceptional (although probably discouraged by the more fastidious master printers). As an experiment, I took twenty-two other play-quartos: the nineteen other Shakespeare “Firsts” (including The Two Noble Kinsmen), plus Romeo Q2, Hamlet Q2, and the fragmentary 1 Henry IV Q0. In each case I began at the top of the first page containing dialogue (ignoring prologues), and read down until I encountered the first punctuation mark without even a hair-space after it. In only four books out of twenty-two did I have to turn the page, and in the Bad Quartos of Romeo and Hamlet the first unequivocal example was high on the second page (lines 7 and 4 respectively). The two significant exceptions were the only two printed by James Roberts, in which I had to reach the third page of The Merchant of Venice and the sixth page of Hamlet Q2 before finding an unquestionable example. Roberts’s successor, William Jaggard, was less fussy. Not having actually counted the unspaced points in the Jacobean editions of Lear I cannot compare statistics, but they are certainly plentiful in both Jaggard’s Q2 (1619) and his Folio text. I therefore see no obvious reason why their numbers in Q1 should indicate a problematic shortage of the four sizes actually called spaces: thick, middle, thin, and hair.As regards unspaced words: Vickers is right that Greg included twenty-two examples in his list of over 500 readings he classified as misprints.15 But in footnote ‡ on p. 65 Greg admitted that “Between some of the words which I have recorded as run together there is, in fact, a very slight gap,” and wrongly claimed that “It is too narrow to allow of an actual type-space.” He was apparently unfamiliar with how thin an early hair space could be, and by my count at least nine of his twenty-two examples are probably hair-spaced.The remaining thirteen are apparently genuine misprints with no space at all: an average of one careless slip every six pages. It is hardly possible that any of them represents a moment when the compositor reached into the space box (shared by all four sizes: they did not have separate boxes) and found it completely empty. That would have been an emergency requiring a complete halt until more spaces were somehow found. What I meant in Origins by “space-metal” was not the spaces used between words, but the quadrats (or “quads”) used beside the text either for indenting lines or for filling the areas between the text and the right-hand margin.On pp. 29–30 Vickers quotes me again, this time to the effect that “the whole text must have been cast off, in that the number of sheets it was to fill had been predetermined.”16 But when he suggests that “Blayney is rather evasive on this important issue” (30) he is mistaken. My previously quoted comment about space-metal was evasive; this one was just plain wrong. It was an example of what Vickers rightly disparages as “illicitly infer[ring] intention from effect” (xiv), and this illicit inference has unfortunately become the core of Vickers’s theory. By the time the long-delayed Origins finally came off the press I had realized that there was a fundamental contradiction between two of my assumptions, but considered it a fairly minor matter whose correction could wait a year or two. At least from sheet C (and almost certainly from the start) Okes resorted to seriatim setting despite his comparatively small stock of pica type. But if he did so because the copy was unusually difficult to cast off accurately (which seems to be the only plausible explanation), we cannot simultaneously assume that he had confidently predicted how many sheets it would fill. That Lear ends on the penultimate page of sheet L is the result of what happened during the printing, but it is illicit to infer that L4r is where Okes always intended it to end. If he unwisely offered Nathaniel Butter an “exact” guess at how much paper he would need, he may have guessed eleven sheets instead of ten and a half—or ele
Referência(s)