University of Virginia Library

Today's Shakespeare Texts, and Tomorrow's
by
Fredson Bowers [*]

The standard Shakespeare texts of today are almost exactly one hundred years old. In 1863-66 William George Clark and William Aldis Wright published what is now known as the Old Cambridge Shakespeare, the first full editorial recasting of the text since the Malone Variorum of the turn of the century. The series of volumes was then revised by Wright alone in 1891-93. With its sometimes incomplete and inaccurate collations of the basic quartos and folios, this edition became the modern foundation, or source, text of Shakespeare. In November, 1864, the two editors subscribed from Trinity College, Cambridge, the Preface to the Globe Edition, a modification that in some few cases accepted emendations of doubtful words unaltered in the original Old Cambridge. In substance, this critical edition of 1864 has remained up to the present the only complete text worth mentioning formed from a systematic reexamination of the textual situation, save in a limited sense for the New Cambridge edition, started in 1921 by Sir Arthur Quiller-Couch and John Dover Wilson, the final volumes of which are only now appearing with the assistance of other editors. However, the New Cambridge text, despite its auspices, has not replaced the Old Cambridge, or the Globe, as the standard authority.

The Old Cambridge was redone from the ground up. The editors made a fresh examination of the various early editions, with a view to establishing anew the respective authority and relationship of the original documents; on the basis of this new look, they proposed to construct a conservative eclectic, or critical, text from the best original

editions. Given the elementary state of textual scholarship at the time, the results were much better than might have been anticipated from the vantage-point of our superior knowledge.

In only two respects affecting a handful of plays were the editors led astray and their text founded on imperfect authority. First, in the assignment of primary authority to wrong editions, Clark and Wright were deceived by the falsely dated Pavier Quartos of 1619 and mistook certain of these second editions as the first, and hence of superior authority in their readings. Next, they were misled into assigning quite unwarranted value to the readings of what we now know as bad quartos, that is, the earliest editions of certain plays that seem to have been printed not from a manuscript in an authoritative line but instead from papers that represent a memorial reconstruction, perhaps made by actors to secure a pirated text to play in the provinces. Following the general opinion of their day, the editors took these highly corrupt memorial texts of such plays as Hamlet, King Lear, The Merry Wives of Windsor, and Richard III, to represent early Shakespeare drafts. Satisfied thus of an authoritative origin, although the texts were manifestly imperfect, they felt free in the conventional manner to emend the Folio with an excessive number of readings brought over from versions that were, in fact, thoroughly untrustworthy. Indeed, in Richard III their critical estimate of the bad-quarto and of the corrected Folio versions was precisely wrong, and they based their edition on the corrupt form, not on the relatively pure Folio.

By present-day standards the greatest weakness of the Old Cambridge editors was their inability to cope properly with complex problems of textual transmission. Such problems fall roughly into three categories: (1) The pre-printing history of the manuscript that was utilized by the printer of a Shakespeare play, that is, whether it were authorial working-papers, authorial or scribal fair-copy, or theatrical promptbook. (2) The transmission of this manuscript through the typesetting in the printer's shop, including the whole effect of the printing process on the faithfulness of this transmission. (3) The physical relationship of certain quartos, or early editions, to each other and to the later counterparts of the texts in the First Folio.

Under the first heading — the pre-printing history of the manuscript — may be added those plays with memorially cobbled-up bad-quarto texts given a quite exaggerated authority as authentic early drafts. Their evil effect on the traditional texts of Henry VI, of Romeo and Juliet, Henry V, Richard III, to say nothing of King Lear, has been constant and serious. But the matter of manuscript transmission

goes deeper. If, as Dr. Alice Walker argues, the Quarto Othello were printed from a corrupt manuscript copy of a late promptbook version of the text, and the purer Folio derived instead from a fair-copy (or earlier promptbook) at no more than one remove from Shakespeare's own working-papers, then the conjectural derivation of the texts gives an editor a generally firm backing for preferring one version — because of its transmissional closeness to the author — as likely to be more authoritative in its readings than the other. Clark and Wright, like other critics of their day, were unable to solve such puzzles. Thus when faced with a divergent textual tradition, each part of some authority, as in Othello or in Hamlet, they could rely on no principle for their choice between the hundreds of variant readings except their literary judgment applied to the traditional preferences of earlier editors.

Although not infallible, the critical sense of these editors was broad, humane, and shrewd; and their excellent taste served in some part to repair the blunders of their occasionally faulty textual theories. Despite their wrong preference for the bad quarto of Richard III over the corrected Folio version, for instance, they substituted on purely literary grounds a surprising number of variants from the Folio that later and more scientific editors were to demonstrate must be the right ones. Nevertheless, enough faulty readings from the bad quarto remain in the Old Cambridge and Globe texts of Richard III to make them quite unsatisfactory representations of what Shakespeare actually wrote.

Since textual bibliography was unknown in the mid-nineteenth century, the proofreading of early texts while in press, and the consequent variation among readings between different copies of the same edition, was not envisaged. Nor were the means even suspected that enable modern critics to study the compositors of Shakespeare editions for their variable fidelity to copy. In short, no technical examination was possible of the physical documents on which the redone Old Cambridge text was based.

Linked to the two preceding categories was the absence of any method for determining the physical relationship between editions in cases of difficulty. Eighteenth-century editors had discovered that certain Folio versions, like Romeo and Juliet for example, were simple reprints of some specific earlier quarto edition. But that the complexly variant Folio Hamlet or Othello or King Lear or Richard III might have been printed not from a different manuscript but instead by marking up a copy of a quarto to bring it into general conformity with some manuscript — a point of absolutely crucial editorial importance

— this could never have occurred to Clark and Wright. Hence except in the simplest cases they confessed their helplessness to determine the relationship between two editions of some independent authority, either in the textual history before printing, or in the subsequent transmission of the text and the development of two separate textual traditions. Wanting the necessary determination of the facts, the Old Cambridge editors were once again thrust back on their literary judgment to solve the considerable problems of textual authority and variation.

In the main, general satisfaction with the overall features of the Globe text has acted as a serious brake on any proposals of editors once again to scrap authority and to resurvey the Shakespeare textual problem quite free from traditional assumptions. Between the publication of the Old Cambridge text in 1863 and the start of publication of the New Cambridge edition in 1921, the various complete editions of Shakespeare were little more than refinements of the basic Globe text. Indeed, when in recent years the publishers of the G. B. Harrison complete Shakespeare polled a wide selection of teachers, the vote was heavily in favor of a reprint of the Globe, with new introductions and notes, instead of a newly contrived text.

The universal adoption of the Globe act, scene, line numbering as the standard for reference may have aided in maintaining its authority. But in some considerable part the simple psychological result of having been brought up on Globe readings prepossesses both editor and reader in their favor, whether consciously or unconsciously. Sir Walter Greg testified to the tug of familiarity when he wrote that the most difficult obstacle he had to overcome in taking a fresh view of Marlowe's Doctor Faustus before re-editing it was to reconcile himself to the less familiar 1616 corrected readings instead of those from the earlier bad quarto that had been traditional in every edition he had read from his youth. Similarly, in his 1936 introduction to Richard III, Kittredge notes, "In some cases the Quarto reading has been adopted in many modern editions and has become so consecrated by usage that one finds it hard to be conscientious and accept the Folio text. Thus in Clarence's dream, 'that grim ferriman' may seem better than 'that sowre Ferry-man' (1, 4, 46), but 'sowre' is probably what Shakespeare wrote." Peter Alexander in his Complete Works of 1951 chose sour; but as indicated by such examples as his selection of Quarto spy for Folio see at I.i.26, he did not fully emancipate himself from traditional unauthoritative readings by applying with rigor a consistent textual

theory based on the modern recognition of the correct relationship of the two texts.

Indeed, how powerful is the pull of the familiar, of what one has read for years (though it be wrong) may be illustrated by the quite indefensible though invariable editorial preference, whether old or modern, for the Romeo and Juliet bad-quarto reading a rose by any other name instead of the good second quarto's correct a rose by any other word would smell as sweet.

The general influence of the Globe text has not been fostered entirely by tradition, however. More practical considerations may intervene. For instance, an editor of a new complete Shakespeare for general reading almost necessarily must use pages from some preceding edition to mark up and send to the printer. He may re-collate a number of the early authorities, as Kittredge did, in order to escape reliance on the unsatisfactory Old Cambridge listing of variants, or the incomplete Variorum volumes, some of them with almost unusable textual collations; but he cannot retranscribe the texts from the originals without an impossible expenditure of time and money. As I can testify, in cases of serious doubt it takes a strong-minded editor to go counter a sufficient number of times to the weight of authority in the pasted-up pages of the edition he has selected as his printer's copy. The disputed reading has usually become established by tradition. It has been the choice of a number of intelligent and respected critics and is therefore "safe." In this manner, through all the spread-out history of its derivatives, the old Globe text has continued to exert a strong influence on its successors.

This is not to say that no improvement has occurred. On the contrary, a steady move has been observable towards the purification of the text, less in the direction of independent emendation, or the introduction of brand-new readings, than towards the restoration of original readings wrongly emended by eighteenth-century editors and subsequently established as traditional. Or, as important, the restoration of readings from the more authoritative, or substantive, early editions when competing authority is present, as with Richard III, and when the Globe's false view of the origin and value of the Quarto version led it to adopt some quantity of variants from an inferior text.

In the process of restoring the text of Shakespeare from unnecessary emendation, the publication of the great Oxford English Dictionary has played a supremely important part. It is quite clear that Shakespeare's language had changed sufficiently by the eighteenth century to lead the early critics to assume that a number of words were corrupt

and in need of emendation. The Oxford English Dictionary is not infallible, and excessive reliance on its authority can have an inhibiting effect on one's survey of the purity of Shakespeare's language in the early editions; nevertheless, it is a vast storehouse of definitions and quotations that demonstrate the general correctness of the original texts as against non-linguistic emenders. In this connection, the major efforts to analyze Shakespeare's grammar and syntax, as by Schmidt, came after the Globe text had been formulated; even the invaluable, though inaccurate, Bartlett Concordance dates from 1894. Hence most of the scholarly studies of Shakespeare's language and the tools to implement such enquiries postdate the Old Cambridge editors and have assisted only their followers.

In the course of the successive editing of complete texts of Shakespeare a few landmarks emerge. The Oxford text of 1892 by W. J. Craig, or the text by W. A. Neilson in 1906, are examples of old-fashioned but shrewd literary taste applied to the problems without the necessary technical information to direct the editing to certain overlooked basic criteria. The 1936 text of George Lyman Kittredge, though often traditional, benefitted materially from his wide acquaintance with Elizabethan idiom and his profound understanding of the processes of Shakespeare's thought and expression. The 1951 text of Peter Alexander is in some respects the best of the collected editions, but it is marred by the fact that Alexander was brought in to rescue another editor and therefore did not have control of the operation from the beginning. His penetrating mind gives us, often, our best readings, but these may be mixed with faulty ones not weeded out with sufficient rigor from the old Collins text that served as basis. The revised Heritage text of 1958-59 is an improvement, but it has not had wide circulation. In some recent years Charles Sisson has championed his palaeographical theories for improving the complete text; but his critical taste is far from impeccable, and his reliance on alleged palaeographical evidence too often merely masks an erratic series of subjective speculations. In 1957 John J. Munro attempted a fresh survey, complete with collations. His scholarship proved unequal to the task.

Generalization about these texts is difficult and can scarcely be accurate in detail. However, it may be said that a few of them (and a number of unmentioned commercial ventures) have inevitably gained by the constant though usually speculative modern discussion of textual cruxes that in a few rare cases has served to illuminate some dark corner of Shakespeare's meaning. The few discoveries of new documents, like the unique first edition of Titus Andronicus in 1905, have

been singularly important but seem to have dried up. More useful has been the slow effect on editors' understanding of the analysis of the relationship of the basic textual documents to each other, as in the demonstration that the Folio Lear was printed from a copy of the Quarto brought into general conformity with some superior manuscript. Additional light has been cast on the kinds of manuscripts likely to have served as printers' copy and their possible relation to the original autograph. The identification of the scribe Ralph Crane who copied out several Folio plays like The Merry Wives and The Winter's Tale has explained some peculiarities in these texts and helped an editor to estimate the value of the readings preserved in the Folio. Thus a superior although far from complete understanding of the nature and origin of the variants in the early authoritative texts has assisted modern editors of the complete works.

As against the permanent value of some details of their texts must stand the undeniable fact that most twentieth-century complete editions have been undertaken as commercial enterprises and not as pure works of scholarship. The demands on a putative editor of the works are so severe, and the amount of information that he must sort out, digest, and attempt to apply even to the simplest problems is so enormous, as to make it a physical impossibility for most editors to produce what can honestly be called a rigorous and comprehensive work of scholarship in the formation of a new complete text. Inevitably, most have contented themselves with revising the efforts of their predecessors, largely on the basis of the same evidence, and using the same editorial criteria.

The one exception, perhaps, has been the New Cambridge edition, of which John Dover Wilson has been the chief entrepreneur. Wilson, who was an early and enthusiastic convert to the "New Bibliography," made a unique effort to investigate the conjectural nature of the manuscripts, and also the bibliographical relationships of the texts, as an informed basis for a new view of the textual problems in each play. It is perhaps ungrateful to remark that his highly speculative mind produced more random insights than it did comprehensive working hypotheses that have stood the test of informed scrutiny. It follows that despite some most illuminating revelations, the textual theory that for each play guided his principles of editing has too often proved faulty in its major hypotheses. True, the stimulation provided by an editor who seemed actually to be a pioneer in textual theory instead of a laggard camp-follower served to focus more attention on the problems of Shakespeare's text than ever before. This attention was

valuable in establishing a climate of interested opinion that led, on occasion, to some fruitful discussion and re-assessment. But the virtues of the uneven New Cambridge text are due more often to Wilson's literary acumen than to the so-called bibliographical theory that was supposed to inform his texts. Indeed, in such plays as Romeo and Juliet his literary taste runs quite counter to his textual theory, and his edition exhibits an enlightened but conventional text that in its choice of readings is often squarely opposite to what would have been produced if he had actually applied his textual hypotheses. On the other hand, the virtues of critical taste and bibliographical theory operate to make his Hamlet text a superior one in many respects.

The single editor of a complete Shakespeare ought to put a uniform impress on his work. His standards theoretically should be much the same for different plays, and his judgment in large part similar. Actually, practice does not confirm this theory quite as one would hope. When about forty years separate the first volume of the New Cambridge Shakespeare from the last that Dover Wilson issued unaided, it is clear that changes in knowledge, in standards, and in the aims of Shakespearean scholarship have occurred that have outdated some of his early texts. Hence any scholar who concentrates on one play, and releases it, before embarking on the formulation of the next cannot possibly provide a series of editions that represent a uniform point of view producing equally valid results.

On the other hand, the alternative may seem to present an effective impossibility. If, say, Wilson had edited his early plays but held them from publication until the last of the canon had been prepared, as he gained in experience and as the advances of textual and linguistic scholarship over the period of forty years affected his views of the text, we should have expected him constantly to have gone back to the completed texts to revise them so that at one time — when all could be published as a unit — the whole would have been uniform and up-to-date. Such a procedure is theoretically possible, but publication under these circumstances of continual revision of past work while pushing into new territory would have taken over the forty years of Wilson's normal rate of progress. One may well query, first, what scholar can know enough to begin a complete edition of Shakespeare at, say, the age of twenty-five, and then work at it constantly with the object of triumphantly publishing his results at the age of seventy or seventy-five? Perhaps Lloyds of London would issue a policy to his apprehensive publisher, but would the publisher pay the premiums and continue to guarantee publication while all around him his rivals

were blithely cashing in on new editions that were saturating the market? Moreover, the task itself would be almost impossible for an academic with a regular teaching schedule to occupy his major energies.

Are we then to have complete Shakespeares constructed only by amateurs with private incomes and proved hereditary longevity? What publisher would engage himself to such a project? Very few indeed, perhaps, outside of the two wealthiest university presses with their Bible subsidies, Clarendon and Cambridge in England. Actually, Clarendon is at present engaged to the production of a new and complete Shakespeare freshly edited from the ground up, though with serial publication in view. But Dr. Alice Walker, its sole editor, did not assume charge of the project in its new form until she had reached the stage of ripe experience, and whether her productivity will equal the inevitable race against time is in some question. The initial volume has been considerably delayed.

The problem then arises, with the weight of research that is now incumbent on a Shakespeare editor, whether any one person can ever again do the whole works in a text that is not merely a reprinted modification of some existing edition, instead of a completely new, rethought, and independent effort constructed from a return to the original editions (not pasted-up pages of some other editor) and taking full account of all modern textual scholarship? I shall return to this problem in a moment. Just now I ask the question in order to introduce what has come to be the modern attack on the problem, that is, the assignment of each play to a different editor under the general supervision of a chief, or advisory, editor.

This is the standard way of producing the new paperback editions such as the Pelican Shakespeare, in the charge of Alfred Harbage. Obviously much depends upon the choice of sub-editors and on the rigor with which the general editor supervises and puts his own uniform impress on their work. Certainly, it is quite possible for a general editor to succeed in enforcing uniformity of styling, as in the typography, the treatment of stage-directions, speech-prefixes, and so on. But his individual editors will inevitably not all be men of comparable judgment or experience. One simply cannot find thirty-odd scholars each to do a text of approximately equal standards, the more especially when many of them will regard the assignment not as a work of independent scholarship but as a hack, or money-making venture that will not deserve a major effort on their part. In these circumstances a particularly heavy burden is placed on the general editor. Having just emerged from under the sheltering wing of Professor

Harbage, I can testify to the amazing care he has devoted to reviewing the work of his editors of the Pelican Shakespeare, not only in technical details but especially in the criticism of the text. Not all general editors have his comprehensive knowledge, or his conscience, however.

The most ambitious Shakespearean editing venture of recent years, the New Arden Shakespeare, lacked such a guiding intelligence in its early stages. The general editor was a brilliant scholar but no textual critic, and as a result the editors of the individual volumes had an unduly free hand without sufficient check from superior authority and experience. The chief virtue of the New, as it was of the Old Arden, is the commentary notes. As a result, some editors seem to have been chosen less for their textual experience than for their interest in the criticism of Shakespeare, although a few of the more recent volumes are redressing the imbalance. Hence the texts run the gamut. From the early volumes we have an eccentric King Lear by a critic who denied the likelihood of press-variants in the Folio text, although Dr. Hinman has since shown them to exist in some quantity. Or an infuriating-to-read Antony and Cleopatra in which a ridiculous attempt is made to keep as much of the Folio punctuation as possible despite the modernization of every other feature of the text. A reader becomes almost schizophrenic trying to adjust to an obsolete system of punctuation when everything else is in modern dress. The idleness of the attempt may be suggested by the fact that of all features of an author's manuscript the Elizabethan compositor seems to have followed the punctuation the least faithfully, even when it existed — as mostly it did not. Thus the Antony and Cleopatra retains as Shakespeare's — or at least as playhouse — punctuation what is, in fact, Folio Compositor B's own favorite system. The effect can be devastating on a reader unaccustomed to, or not expecting, the Elizabethan system of rhetorical pointing instead of the syntactical system that should accompany modern spelling.

On the other hand, such volumes as Much Ado About Nothing have been edited with scrupulous care for the bibliographical history of the text and what can be conjectured about the nature of the underlying manuscript.

The picture of today's texts sums up to something like this. The complete editions now current, edited under the hand of one man, have revised and improved the basic Globe text, but unsystematically according to the taste and knowledge of the editor as he has preferred a variant reading here, or another there, or a traditional emendation. Several, such as the texts of Kittredge or Alexander, have been the

work of outstanding Shakespearean scholars who have contributed their special judgment to the problem and have been at least conversant with the major textual scholarship of their day. But with the exception of the Dover Wilson New Cambridge, in none of these complete editions has a real attempt been made to cast off the shadow of the Globe and to reconstruct a new text of Shakespeare as an independent act of scholarship. The attempt of Wilson, admirable as has been its purpose, has failed in the last analysis through the substitution of highly speculative methods of analyzing textual history for scientific bibliography, and the failure to do very much towards applying the general textual theory for each play to the peculiar problems of its text.

The collected editions of Shakespeare by various hands under a general editor have suffered the inevitable unevenness of execution in different volumes according as the sub-editors have had suitable experience with the editing of texts or else were chiefly literary men engaging themselves to a technical task for which they were not equipped. The Pelican is perhaps the most evenly edited as a general proposition; the New Arden, though some texts are so bad as to demand re-doing, also has a few examples of the very best in modern editing.

At the present time something of a point of rest seems to have been reached. Conventional texts for teaching purposes continue to be issued as commercial ventures, but none is likely to make its mark as an original or lasting work of scholarship. Independently of the editors, however, a great deal is being learned about the nature of the documents on which Shakespeare's texts rest, the most notable work of textual or bibliographical scholarship of the century having been issued just this year in 1964. This is Charlton Hinman's two-volume study of the printing of the First Folio, which recovers in the most extraordinary detail the day-by-day operations of Jaggard's presses and the compositors who set the type.

Much general information now available has not been digested as yet, and perhaps is not ready for digestion, in the sense that adequate interpretation and exploitation (let alone application) have not been made of the factual findings that have rapidly multiplied since the end of the Second World War in the hands of such younger men as John Russell Brown, Philip Edwards, D. F. McKenzie, Philip Williams, Charlton Hinman, Robert K. Turner, and George W. Williams. The year 1945 may properly be taken as marking the coming of age of textual bibliography, yet no comprehensive new survey of the text by a competent scholar has been made in the last two decades despite

some admirable dealing with the problems of individual plays.

When one turns to speculations about the texts of tomorrow, a most significant fact instantly emerges. No complete edited text of Shakespeare is available in the only authoritative form that can exist, that is, in the old spelling and punctuation of the original documents in which it was printed. This fact is shocking, but it is true. All available complete texts are modernized, many of them with quite illogical anomalies remaining. In 1930 Sir Edmund Chambers (William Shakespeare, I, 276) wrote: "A text in Elizabethan spelling is still a desideratum. The Old Spelling edition of F. J. Furnivall, W. G. Boswell-Stone, and F. W. Clarke (1907-12, 17 plays issued) remains incomplete." Since 1912, therefore, or for over fifty years, no one has succeeded in bringing out an edited complete text of Shakespeare in its most authentic and authoritative form. Facsimile and diplomatic reprints are available for some, though not for all plays, but these are not the answer to the problem of a reading edition based on and in large part reproducing the early authorities. R. B. McKerrow, a great scholar, published proposals in 1939 for such an edition, but his early death and the effects of the World War joined with various difficulties to prevent its continuation in other hands. In the past few years the Clarendon Press has revived the long-quiescent project and, at this moment, is awaiting copy for the first volume of the projected series. The fate that overtook the 1907-12 edition, or the McKerrow proposals, will not again operate here, one hopes.

Wherein is an old-spelling edition more authoritative in its form than a modernized text? For one thing, an old-spelling edition must refer directly back to the originals, whereas every modernized text is formed by annotation of the pasted-up pages of some preceding editor's version, and is likely, as a consequence, to repeat, inadvertently, some of its errors that were not detected. A long history of the careless transmission of error by this means can be written. More important, the old-spelling text comes as close to Shakespeare's original intentions as can be managed, since the earliest printed texts were set up, some directly, from a Shakespearean manuscript but most from a manuscript that at least had a reasonably close transcriptional link with the Shakespearean holograph. Hence, despite the imposition of the compositor's style, and his occasional error, on the manuscript copy from which he set type, in the absence of the original papers no authority exists superior to the printed editions manufactured directly from these manuscripts.

Still, the question remains wherein a good modern text is of lesser

authority than an old-spelling edition constructed from the original documents. Are not the words — what Shakespeare wrote — the same? For instance, if one were to listen to actors, would one be able to distinguish whether one were hearing a modernized or an old-spelling text? The answer is, they are not the same and by their nature can seldom be the same in precise terms. To be practical, one can say that an old-spelling edition is likely to be a work of scholarship. The editor will take the trouble to compare letter for letter, word for word, and punctuation mark for mark, the readings in a number of copies in order to ascertain which are the result of press-alterations made during the course of printing and what is the authority of such variants. The editor of a typical modern text is more likely to rely for convenience on the faulty Old Cambridge collations to give him the readings of the various early editions and a conspectus of editors; at the most, and then only in exceptional cases of conscience, may he check his paste-up sheets against some single copy of the appropriate early edition.

The old-spelling editor, acutely conscious of the forms that words take in their spellings and in their connective punctuation, studies the printing of the early editions for what evidence can be brought to bear on the text. He is concerned with variations in these forms, whether due to compositorial preference or to the possibility that they may sometimes reflect the spelling of the manuscript underlying the print. From all the available evidence he endeavors to come to some conclusions about the nature of this underlying manuscript and therefore about its degree of trustworthiness. If he thinks it to be a Shakespearean autograph, perhaps as Antony and Cleopatra is, he will be relatively conservative in his emendation. If a scribal copy at some remove from the original, like the Crane manuscript behind the Folio Merry Wives of Windsor, he will be bolder in his treatment of deficiencies in the text. Nothing that might affect the authority and trustworthiness of every possible detail escapes his attention.

No law bars a modernizing editor from an equally scrupulous scholarly examination before making up his text. But the plain fact is that though John Russell Brown's modernized edition of John Webster's several plays was based on just such a searching enquiry, to date no comparable edition of Shakespeare has appeared. Hence, on the record, the modernizing editor works basically at secondhand, without the minute and scrupulous examination of the source materials that marks the activities of the old-spelling editor.

Yet even supposing that the results might be the same in the actual

words — although this is a tall order unlikely to be fulfilled — the fact that the forms of the words and punctuation will differ is often of considerable consequence. For old-spelling texts the original evidence is revealed to the reader, with no impenetrable screen of silent editorial decision concealing from him the essential facts. On the contrary, a modernizing editor must constantly be making up his mind about the import of the evidence and presenting the results all predigested to the reader — and in most texts without the information that he has altered the originals. When in A Midsummer Night's Dream Titania urges her fairies to steal the honey bags from the bees, she observes, Since this was written and printed before the introduction of the genitive apostrophe, no external evidence indicates whether the fiery glow-worms eyes refers to the eyes of one glow worm or of several. Professor Arthur Brown, from whom I draw these two illustrations, also notices the lines in the Dream in which Lysander pleading his claim to Hermia, argues, My fortunes every way as fairly ranked: does this mean that my fortunes (in the plural) are (elided) as fairly ranked. or is fortunes the contraction for singular fortune is? The apostrophe for the contraction is often omitted in early texts. A modernizing editor must make a decision that will, inevitably, lead the reader to suppose that whatever form he chooses is what Shakespeare wrote. Well, maybe!

The old-spelling editor does not mislead the reader thus but reports the evidence, always emended if necessary; but if he is wise, he adds a footnote directing the reader's attention to the difficulty, together with a conspectus of what reading critical opinion has previously favored and the reasons for his own conclusion.

Even if a modernized text has notes that would refer to such ambiguities as the glow-worms eyes or fortunes every way, the editor's modernizing may still ignorantly destroy a meaning intended by the author, and without a note. Sometimes quite subtle puns or word-plays are contained in variant spellings. Unless an editor is acutely aware of these possibilities, he will conceal the depth of these meanings by his unwitting destruction of the evidence in the process of modernization.

A typical case comes to hand in the Elizabethan doublet travel-travail. The first means the process of going from one place to another; the second means toil, labor, or else anguish, pain, distress, effort. Sometimes Shakespeare keeps the precise meanings of the two absolutely distinct. Sometimes, like other Elizabethans, he seems to intend both meanings simultaneously, a device aided by the fact that travel was often spelled travail anyway. Faced with such situations, a modernizing editor may keep an appropriate travail spelling to indicate the probability of word-play, if he recognizes it, or he may obscure the possible evidence by modernizing all travail forms to travel whenever the sense permits. If he does so, he certainly changes Shakespeare's probable intentions in an unwarranted manner and detracts from the vitality of the language and the subtle meaning that Shakespeare may have intended. Here are a few examples in which, in the original editions, the spelling is travail, with what seems to me to be the strong possibility of a pun, but the Globe text silently changes to travel and is conscientiously followed by Kittredge and Alexander.

In The Tempest after Alonso and his party have exhausted themselves in search for Ferdinand, the conspirators discuss the planned murder. Sebastian remarks that they should take the next advantage, to which Antonio answers, in the Folio,

From what crystal-ball did the Globe editors decide that travail here meant travel instead of the oppression of the party's physical powers and spirits by the labor and anguish of the search? If a primary meaning were to be sought, indeed, travail fits the context better than travel; and at the least Shakespeare may well have been playing on both meanings.

A lesser example may be found in As You Like It when Rosalind, Celia, and Touchstone seeking the court in the forest come upon Corin and Silvius, and Rosalind begs rest and food for Celia, with

Here the primary meaning of travel is clear enough, but the secondary meaning of travail is also likely, although modernized texts follow the Globe in concealing the fact.

In Love's Labour's Lost in the preliminaries to the masque (V.ii. 191-197) the Boy accosts the disguised Berowne with,

This is the reading of the authoritative quarto. But as an illustration of the ambiguity of the word, the compositor of the unauthoritative Folio reprint altered the spelling travaile to travel, and so do the Globe, Kittredge, and Alexander. Yet with the emphasis upon weerie steps and weerie miles, although the primary meaning of travel or journey is clear the simultaneous word-play on the pain and effort of stepping off the weary miles should also be a distinct possibility, which is lost in modernized texts.

In some circumstances punctuation by altering the modification can alter the meaning. The repunctuation of the old texts that must be undertaken by a modernizing editor may, without the reader's knowledge, support one meaning over another, or support one idiomatic usage against its contrary. Many examples are small in themselves, but cumulatively they end by having an important effect on our critical understanding of Shakespeare's language, and even of his plain intentions.

For instance, at the beginning of Hamlet Marcellus tells Bernardo that Horatio thinks they have imagined the Ghost, and

That is, the good Second Quarto punctuates so, with a comma pause after along. But the Folio reads, Here no comma appears after along, but the pause is placed after us. The point may be trivial, but a Shakespearean idiom is involved (whether one intreats or intreats along), and a reader ought to want to read this line as Shakespeare intended when he wrote it. The Globe edition, followed by Dover Wilson and by Peter Alexander, omits all commas so that one reads, This alteration of the punctuation of either authority dodges the whole question, which is — does with us modify the elided come before along, or does it modify to watch. That is, should one paraphrase, "entreated him to come along with us in order to watch," or "entreated him to come along in order to watch with us"? The Folio reads the first, but the Quarto the second. Kittredge retains the Quarto comma after along, and in my opinion correctly, since the us must refer to both Marcellus and Bernardo, but Horatio has come only with Marcellus for the purpose of joining Bernardo to watch the minutes of the night. The Globe modernization towards light punctuation, perhaps unthinkingly followed by Wilson and Alexander, removes the Quarto's signpost punctuation and leaves the matter ambiguously open to the reader's conjecture — to the loss of Shakespeare's probable intention. A linguist and lexicographer would certainly be interested in the difference, and so should we if we are concerned with fine shades of meaning.

If this example seems like nitpicking (even though it be multiplied many hundreds of times to the loss of Shakespeare's precision of meaning), then we may contemplate the effect of the dispute between Quarto and Folio about the modification of some key phrases in Hamlet's "What a piece of work is a man" speech to Rosencrantz and Guildenstern (II.ii.12-16). The Second Quarto reads,

According to this punctuation, man is infinite not only in faculties but also in form and moving; his action is express and admirable; he is like an angel in apprehension, indeed (as an exclamation) he is like a god.

On the other hand, the Folio, which is followed by all editors, has this:

One must admit that the sense is perceptibly affected by the reassignment of express and admirable to man's form and moving instead of to his faculties, and by the distinction between the action of an angel and the apprehension of a god. Both readings can and have been argued; and sometimes, in a moment of weakness, one may wonder idly whether our general preference for the Folio reading has not been shaped by the generations of early editors who relied on the Folio over the Quarto in most cases of variation — though I take it the Folio is really correct. These days we are inclined to defend the general textual authority of the Hamlet Second Quarto, especially if it were printed from Shakespeare's own working papers, whereas the Folio may represent a copy of a copy of a copy of these papers. That there may be two sides to this moving speech is concealed from any reader of a modernized text, however.

Yet the case for the early punctuation goes deeper even than such important questions of linguistic analysis, or of basic meaning. Elizabethan syntax and the rhetorical punctuation system that accompanied it are wedded and cannot be divorced without really serious loss. Any modernizing editor finds acute difficulty in applying present-day logical syntactical punctuation to Elizabethan structures that are perfectly suited to the graduated pauses of rhetorical pointing but not to strict and limited modification. All sorts of expedients must be adopted in modernizing, such as series of dots or a far too liberal sprinkling of dashes as clumsy substitutes for the original. But the fact remains that when Shakespeare's plays are read in hobbling and anachronistic modern punctuation, a real injustice is done to the free flow of the verse, or prose, and to the loose rhetorical structure.

The answer, quite definitely, is not to use the old punctuation with modern spelling, as in the New Arden Antony and Cleopatra, an even more unworkable anachronism. Instead, it is to read Shakespeare in an edited form that reproduces the original, corrected and modified as necessary to form a usable and accurate reading text. Formerly, the old pedantic use of the long 'f' that looks like an 'f' to the uninitiate, and then the confusing retention of the 'u' for medial 'v' and of 'v' for initial 'u', and the invariable appearance of 'i' for 'j', militated against the ease of reading old spelling for all but the expert. However, these typographical conventions have no real purpose in a reading text, even one suitable for scholars, and they are now being discarded as an obstacle to the reader. Consequently, any reader who is concerned with the most accurate and intimate approach to Shakespeare will require a critically edited old-spelling text, not a modernized version that constantly draws a veil between him and the subtleties of the original. This is not mere antiquarianism, but solid common sense. Students constantly show us that one does not need to be a trained

scholar to read such editions, properly prepared, with ease and with an added pleasure.

Finally, the presence of an authoritative old-spelling edited text of Shakespeare, with full collations, would vastly improve the quality of the modernized versions that will always be necessary for school use and for the most popular reading. A new and definitive norm would be established on which these popular texts could be based, and by this means one could end the long chain of accumulated error that has resulted from the carelessness of selection of textual authority, the paste-up system of preparing printer's copy, and the thoughtlessness that follows from depending too closely on the work of other editors.

Such enlightened old-spelling complete editions are sure to be the basic texts of tomorrow. Not one can be bought today, and the New Oxford edition, which will be in this form, will very likely not be completed in the lifetime of some of us here. Not in mine, surely. Tomorrow's texts, thus, or perhaps the day after tomorrow's will make available reading editions either in usable old-spelling versions for the informed and cultivated general reader, or for popular use more accurate modernizations than are now generally available. I hope you will receive this as a prophecy.

If these critical old-spelling texts are as desirable as I assert they are, why do we not have them now? One answer, though it is incomplete, is that no Shakespeare text can be very profitable if it does not see widespread use in the schools. Hence publishers are conservative and do not wish to engage themselves, in what is a highly competitive business, to an expensive undertaking that will limit the use of a reading edition to only one segment of the public. Moreover, in their favor let it be said that the means for making old spelling palatable to non-scholars by dropping its distinctive typographical difficulties, which are of no true significance, are of comparatively recent acceptance.

Another important reason, possibly, is psychological. Modernized texts can be semi-respectable hack work, but an old-spelling edition, just now at least when we are in the pioneering stage, requires basic scholarship. In the lack of publishers' enthusiasm, academic modesty has persuaded workers in the Shakespeare industry (as it is familiarly known) to tried and true formulas that are simpler than the unknown terrors of old-spelling. The commercial academics well know that any attempt on their part to treat old-spelling editions as casually as they have dealt with modernized versions would call upon their heads the wrath of all informed critics, because they would have been dealing

with a serious matter of scholarship, not just with innocuous texts garnished with pretty pictures for the passive general reader, and school children. The day will come when versions of some authoritative basic old-spelling edition will be contrived for general use, but the camp-followers must wait for that day when a source and model will be available. They do not have the scholarship or the skill to construct an acceptable old-spelling text for themselves.

We await, then, the authoritative critical edition of Shakespeare that will serve as the basic model for more popular modifications in the original spelling as well as in modern form. Why are the scholars not delivering it? The answer is that anyone who sets himself to edit the ultimately authoritative old-spelling Shakespeare that will preempt the field, as the modernized Old Cambridge, and its Globe derivative, did in its day, knows that the time has not come for the start, since too much vital information is still uninvestigated and too many crucial problems unsolved. Since this situation will last for a number of years, the need is sufficiently urgent for some transitional old-spelling edition, such as the New Oxford may well prove, to be issued in the interim. It will have the field to itself for time enough to justify its existence, and more.

I turn now, though very briefly, to some examples of the problems that need solving before this ultimately authoritative edition can be produced to serve as the progenitor of tomorrow's superior texts. First, let it be said that present-day editors have long since exhausted traditional materials. Just about every emendation has been proposed that is likely to be adopted, and editing has largely resolved itself to the exercise of personal choice among the known alternatives. The purpose of the new scholarship that will produce tomorrow's texts is threefold: first, to evolve a more scientific and logically rigorous method to govern the critical choice of alternatives in respect to the words of the text; second, to indicate the degree of confidence that a reader can feel in the various critical choices that have been made; and, third, to bring some order and authority into the presentation of what are known as the accidentals of a text, that is, the system of spelling, punctuation, capitalization, and word-division that clothes the words and sometimes shares directly in the communication of their meaning.

First, to evolve a more scientific and logically rigorous method to govern the critical choice of alternatives in respect to the words of the text. This investigation has two prongs, the linguistic and the bibliographical. Modern linguists — spawned by cold-war necessities — have

scarcely tackled the problems of Renaissance English, and when they have done so the results have not always been valid because some scholars were too bibliographically naive to choose the proper texts for analysis. Time and again an editor of Shakespeare comes up short against a wall of ignorance about the possibility or impossibility of certain grammatical or syntactical constructions, and finds no specialist in the Elizabethan language to help him. Is a seemingly nonce word a Shakespearean invention, a rare form, or a possible corruption? To what extent were various dialect forms accounted acceptable in cultivated speech? (Some suggested emendations by linguists, or defenses of possibly corrupt readings, reach into every nook and cranny of country dialect although spoken from the mouths of kings on the stage.) What spellings of certain words are possible and what impossible? What limitations can be placed upon violation of grammatical concord? The truth is that despite three centuries of study we still know far too little about Shakespeare's language to answer a large number of questions that inevitably arise. The Early Modern English Dictionary, which would serve as a new foundation, seems to be bogged down and in need of transfer to hands that will make it a going concern.

The bibliographical investigations that remain form a staggering prospect in that they seem to branch out in every direction. I select only two as illustrations. The basic all-purpose investigation must be into the specific habits of the compositors who set Shakespeare texts, including their faithfulness to copy both for words and for accidentals, or their tendency to certain identifiable kinds of divergence and error. Charlton Hinman's monumental work now identifies the five Jaggard compositors who set the First Folio and assigns the shares of these workmen even down to a part column on a page. As yet, the equally important quarto texts set in a number of printing shops have seen little scrutiny, and most of their compositors remain unknown.

The importance of the information that will follow an analysis of each compositor's work, in a manner as yet scarcely envisaged, cannot be over-estimated. By stripping the veil of print from the texts, one may recover a number of the characteristics of the manuscript that was given to the printer. From such evidence one may eventually determine, not impressionistically as at present but scientifically, which were Shakespeare's own papers and which copies by perhaps identifiable scribes like Ralph Crane. For instance, it is now known from details in the Folio print that at about the halfway point of The First Part of King Henry VI another hand, with different spelling

characteristics, took over the inscription of the lost manuscript. No scholar has yet looked into the effect on the text by playing the two compositors that set the Folio pages against the two inscribers, although the fact has been known for about ten years. Either of the hands could perhaps be Shakespeare's. The possibility is certainly worth looking into, to say the least.

One of the first tasks of any editor is to try to identify the kind of manuscript that served as printer's copy for a Shakespeare first edition and for any later authoritative form of the text. Indeed, his editorial practice will be largely dictated by what he finds in this enquiry. If the manuscript, he thinks, is a Shakespeare autograph, the printer's errors will often be simple ones of misreading the handwriting, mixed with some amount of memorial error. The editor will be inclined to give the benefit of the doubt to difficult words or phrases less likely to be the invention of the compositor than of the author, and he will feel that they can represent Shakespeare's own, or else in some recoverable form a recognizably corrupt version of what Shakespeare wrote. But although the printed text will be superficially cleaner and more correct if the printer's manuscript were a scribal copy, the editor will know that it is sure to harbor concealed corruptions of a different sort from the first, and he will incline to be much bolder in his questioning of passable but not entirely characteristic readings.

Already we know in the First Folio that the quality of the work of the veteran Compositor A was superior indeed to that of the apprentice Compositor E, named John Leason, who had joined Jaggard's establishment just a few months before he started to typeset Titus Andronicus. We suspect that Compositor B was likely to crowd his memory with too much manuscript material before turning to his type-cases, and hence that his proportion of paraphrase and substitution is higher than is A's. We can sometimes see that Compositor B is prone to tinker with the metre and perhaps voluntarily to change words that did not seem to him to be quite right.

The analysis of compositors may lead not only to some clearer concepts of the nature of the manuscripts for the different plays, but also to evidence about the disputed relationship of different printed editions. No editor can tackle the serious problems of Hamlet or of Othello unless he knows whether the quite different Folio version was itself set directly from a manuscript in another tradition from that of the Quarto, or whether a copy of the Quarto was marked up for the printer by comparison with this manuscript. An editor of Richard III had better determine whether a copy of the third or of the sixth

quarto was annotated to form the printer's copy for the Folio. The hypotheses that link some texts like the Quarto and Folio of The Second Part of King Henry IV are still far from demonstrated, to the despair of editors. Even the Folio Hamlet, for which the printer's copy seems to be an annotated quarto, has some nagging problems involving the compositorial treatment of the supposed copy that continue to disturb bibliographical critics. On the other hand, the application of the bibliographical method of textual study to the Othello problem has proved successful, just this year, in demonstrating the dependence of the Folio on the Quarto, and it may be hoped that the other texts, when rigorously examined, will also become subject to factual demonstration in a manner not previously contemplated.

Second, to indicate the degree of confidence that a reader can feel in the various critical choices that have been made. Some readings are subject only to critical hypotheses. In Hamlet's "To be or not to be" soliloquy whether the Quarto's enterprises of great pitch and moment is correct, or the Folio's great pith and moment is a critical matter pure and simple, not susceptible of bibliographical decision except as a general view of the Folio's authority and its connection with the Quarto might bear on the solution of an individual reading. On the other hand, when the Folio Richard III, probably set in main part from an annotated copy of the Sixth Quarto, repeats a reading that is known to be a printer's corruption since it first appeared in the unauthoritative reprint Second Quarto, bibliographical logic demands a return to the original reading of the first edition.

In the first instance, a reader can weigh the analogies in Shakepearean and general Elizabethan idiom of pitch and of pith and in context come to some agreement or disagreement with editorial choice. Since a hair's-breadth separates the appropriateness of either word, only provisional confidence can perhaps be given to any critical choice. Even in the Richard III example there is less than absolute certainty despite the bibliographical demonstration of superior authority. An editor should know that he cannot guarantee that the memorially reported First Quarto original reading is actually Shakespeare's, because too many words in this edition are demonstrably not authentic. Yet he can inform the reader that even though the word itself may not be authentic, the odds are that it comes closer to Shakespeare than can a simple printer's error in the second edition transmitted through to the sixth and thence to the Folio.

Practical certainty can be achieved in Richard III in another

textual situation, however. When the First Quarto reads one word, and the Second Quarto, or a later, corrupts to another, but the Folio returns to the reading of the first edition, the only assumption is that the Folio scribe corrected the error in his printed copy by reference to his manuscript. The manuscript behind the Folio text and the actor's memory that produced the First Quarto were therefore in agreement. Hence, unless one is to suppose that the word is a common error emanating from the promptbook, a real vote of confidence can be given the first reading as authentically Shakespeare's, and the other can be discarded utterly.

Another example shows how textual theory can assist us to estimate the confidence we may place in the authenticity of a reading, here an editorial emendation. This is the passage in which Hamlet toys with Polonius and leads him towards the theory of love madness with "For if the sunne breede maggots in a dead dogge, being a good kissing carrion. Have you a daughter?" At least, a good kissing carrion is the Second Quarto reading, followed by the Folio, although most editors emend to being a god kissing carrion. If like the Quarto the Folio were set directly from an independent manuscript, the odds would be high against a common error of this sort in two such disparate authorities, and the emendation would be doubtful indeed. On the other hand, if a copy of the Quarto were marked up for the printer by consultation with some other manuscript, all one need do to explain the preservation of the assumed error in the Folio is to conjecture that the collator overlooked the difference in the words, even though his manuscript presumably read god. This is far from an impossible situation, for just such oversights can be demonstrated in the Folio Richard III, and there are other possible cases in Hamlet. We must recognize that suggesting that this oversight could occur is very far from demonstrating that it did occur. Nevertheless, an emending editor who believes in the annotated-quarto theory can indicate to his reader a far greater confidence in the emendation of good to god than can an editor who takes it that the Folio was set from manuscript printer's copy and that the error good must go back to some common original.

Third, to bring some order and authority into the preparation of the accidentals of a text, especially those that share directly in the communication of meaning. The popularization of modernized texts, not always with direct reference back to the authoritative early editions, has led to much carelessness in viewing the spelling, punctuation, capitalization, and division of the originals. Since these features would

be modernized anyway, many editors have chosen to ignore the significance of the early forms. But even in such very minor examples as that offered by entreat along with us to watch the minutes of the night, or the travaile-travel word-play, the exact forms of the basic authorities are singularly important to construe. One of the chief functions of a critically edited old-spelling edition will be to settle in a definitive manner — with due regard for the different compositorial characteristics — the precise forms of the originals. This process will involve a minute collation of all first-quarto editions in search of press-variants, something that (shockingly) has not yet been done for every Shakespeare text. (Dr. Hinman has taken care of the Folio collation, of course.) It will also involve a careful scrutiny of ambiguities and an attempt to resolve them. For instance, when Falstaff in The Merry Wives in outrage at Pistol's refusal to carry the letter to the wives shouts, "I, I, I my selfe sometimes, leaving the feare of heaven on the left hand, and hiding mine honor in my necessity, am faine to shuffle: to hedge, and to lurch" (II.ii.23-25), the Folio, followed by all editors, prints I, I, as if it were the first person pronoun, and an actor would be justified in beating his breast to emphasize the repetition of the meaning, I, even I. But since the adverb ay (meaning yea) was also frequently spelled capital 'I' like the pronoun, an editor must consider carefully the possibility that Falstaff is, instead, saying, "Yes, yes, I myself." Finally, the re-examination will concern itself with the detection and emendation of error on a conservative basis and always with record of any alteration made in the original.

Doubtless not much that is very startling or novel will appear in this new definitive edition of tomorrow that is to replace the Old Cambridge text of today as a source authority. Laymen are far too optimistic when they think that scholarly re-examination of the original documents by textual critics must inevitably produce new lines, phrases, or words by Shakespeare not hitherto discovered. The prospect of any second coming is most unlikely, although it is to be expected that some few new emendations may be suggested, perhaps of a minor sort. It must be emphasized that the basic variants between early editions have been known for many years, that there is much more to a definitive edition than the editorial invention of new readings to replace those in the original thought to be in need of emendation, and that very few previously unsuggested editorial emendations of any marked importance, likely to secure general consent, will be proposed and adopted in the future.

Indeed, perhaps the most valuable result might be the confirmation of a certain number of existing traditional emendations, like some of the positively required additions from the bad quarto to the Folio Merry Wives text. As the obverse, we may look forward to the discarding once and for all of excessive emendation that has become traditional, as in other of these Merry Wives additions that have been accepted from old days by a series of editors who knew little of the suspect authority of any bad-quarto reading and had no idea of the correct relationship of the two forms that have been preserved for this text.

The restoration of the purity of Shakespeare's text from traditional emendation based on what are now known to be false premises will alter many readings, some of them very familiar. The greatest changes will unquestionably come in those plays where two texts, each of some independent worth, have been subject to random conflation or substitution of readings without regard for their textual history and relative authority. The re-evaluation of the evidence for the nature of the underlying manuscripts and their respective authority (or purity) is badly needed for such important plays as Hamlet and Othello but chiefly for that perennial problem, King Lear. The choice of readings not on a subjective basis but logically guided by the pre-printing history of the manuscripts and the post-printing physical relationship of the texts will certainly result in changes from tradition. Hamlet's flesh, for instance, will become too, too sullied, not solid. Cordelia's love will become more ponderous and not more richer, and so on. As an easier problem, the last vestiges of indefensible bad-quarto readings in plays like Richard III and Romeo and Juliet will be weeded out. In the future we shall no longer read spy for see in Richard or name for word in Romeo and Juliet on the basis of assumed literary superiority when the physical, bibliographical facts make the odds about 1,000 to 1 against their correctness. Despite all the editing that has been done in this century, Shakespeare's text of today is pockmarked with indefensible wrong words substituted one to two hundred years ago from early texts of inferior or of no authority.

No blinding new revelations about plot or character are to be anticipated, then: Hamlet will not be revealed as a woman in disguise, nor will Lear save Cordelia from hanging. But that unique instrument of Shakespeare's language in all its Elizabethan vigor and subtlety will have its eighteenth- and nineteenth-century tarnish rubbed off and will gain some new glints that cumulatively will strike all but the dullest eyes, and ears. As a bonus, the informed reader will be freed

from the strait-jacket of inappropriate and arbitrary modern punctuation contrived for a different kind of syntax, and will be able to read the texts with the indication of pauses, and their weight and duration, that seemed appropriate at the time. No one is so foolish any more as to argue that the punctuation — now demonstrated to be mainly compositorial — is Shakespeare's own or that it represents in any way the authority of the playhouse. But such as it is, it is the most authentic that we have, and it is usually well suited to clarify the casual Elizabethan syntax. Finally, the ability to read Shakespeare in his early spelling will reveal fresh nuances, associations, and plays on sounds and meanings, in depth, that have too often been destroyed by the heavy hand of modernizing editors.

These add up to a not inconsiderable benefit. They are well worth the extraordinary amount of time and treasure that will need to be devoted to the project. The prime desideratum is a computer-made old-spelling concordance that will assemble all the basic facts for editorial analysis. The problems of programming such a concordance and of getting it into usable form with a sufficiently comprehensive scope are staggering indeed, but not insoluble. Without it the necessary compositor studies, linguistic analyses, and then — that crown of all editing — the critical evaluation and coordination of the final results, can never be achieved. We may have transitional texts tomorrow without such a comprehensively planned concordance, but the final and authoritative form of Shakespeare's text that need never be changed except in minor detail awaits an electronic birth.

However, I do not put the machine above the critical mind, because — in the end — the sole function of linguistic analysis and of textual bibliography, with all its mechanical aids, is to guide an editor's critical intelligence to the truth. This process involves setting technical limits to conjecture by means of the scientific establishment of different degrees of factual probability ranging from the quicksands of mere possibility up to the starry level of irreversible demonstration.

The concordance will doubtless prove easier to come by than the ultimate editor. Perhaps, in the end, one man may be able to digest the widely assorted technical and critical problems and unify them into a single great work of scholarship. It may be, on the other hand, that the ultimate editor will, instead, be the director and the final arbiter of a uniformly trained small group of scholars, learning and then working together within a single room in a research institute. If the miracle of the King James Bible could emerge from some such form of one mind in bodies multiple, perhaps there is some hope for Shakespeare — in the twenty-first century.

Notes