| ||
The Rationale of Old-Spelling Editions of the Plays
of Shakespeare and his Contemporaries
by
John Russell Brown
There was a time, in the days of Pope and Dodsley, when editors 'modernized' the spelling of Elizabethan and Jacobean printed plays without thinking twice about the matter, but in more recent years, especially after the first volume of McKerrow's critical edition of Nashe in 1904 and the first issues of the Malone Society's diplomatic reprints in 1907, editors have seemed to prefer the 'old spelling' of original first editions with just as little thought.[1] While the Malone Society has issued its reprints for over fifty years, its editorial policies have remained unchanged in all essentials; and its list of subscribers continues to grow. At the present moment scholars are engaged on large-scale, critical editions of the plays of Peele, Greene, Dekker, Heywood and Massinger, and all have chosen to prepare old-spelling editions. So far the only old-spelling editions of Shakespeare's complete works have been facsimiles or diplomatic reprints, but work is now in progress on an old-spelling, critical edition. The decision made by McKerrow and others in the early years of this century has become an orthodox decision, and few editors seem to question its validity; apart from the more or less popular editions of Shakespeare and some text-book anthologies, very few modern-spelling texts of Elizabethan or Jacobean plays are now being published.
When a technical decision of this kind remains orthodox for more than fifty years it might be supposed that those years were a time of modest, conservative effort in the particular branch of studies involved. But this is far from the truth. Since McKerrow's first editorial work there have been great advances in textual studies; editors now know far more about Elizabethan printing practices and Elizabethan dramatic documents, and, on some aspects of their task, they now know that
The choice cannot be completely isolated from other choices, as whether or not to retain the original punctuation, capitals, type-variation and lay-out. Nor can the choice be made without considering the readers for whom a projected edition is intended. The problem may best be examined by reference to two distinct kinds of old-spelling editions, each exemplified by a particular example, and by confining the discussion to Elizabethan and Jacobean printed plays, omitting the special problems of manuscript originals.
The Facsimile Reprint
The simpler kind is the facsimile reprint, as represented by the Malone Reprints. The aim of the editors of these texts was summed up by A. W. Pollard at the inception of the society, and this aim has recently been endorsed by Professor F. P. Wilson, the present General Editor; it is to do
This kind of old-spelling edition must be judged today by the same yard-stick. Yet the answer will not be the same, for this particular measure has changed: the textual student now asks questions which, in 1907, concerned only the bibliographer, and he has won for himself the new and barbarous name of 'biblio-textual student'; increasingly he needs textual and bibliographical authorities. When he considers a crux he must consider not only the sense, spelling, punctuation and other details that can be reproduced in type-facsimile; he must also be on the look-out for the slightest irregularity in the original printing. So he may be able to argue that the type has been tampered with during some stage of proof-correction, and thereby discern the cause of an error and, perhaps, its most plausible correction. An example of this is found in The Honest Whore, Part 2 (1630) where the first edition reads:
old Iacomo, sonne to the Florentine Iacomo, a dog, that to
meet profit, would to the very eyelids wade in blood of his
owne children.
(A4)
Iacomo, sonne to the Florentine
Old Iacomo, a dog, that to meet profit,
Would to the very eyelids wade in blood
Of his owne children.
(I.i.123-127)
As more press-variants are discovered between different copies of the first and authoritative editions of Elizabethan and Jacobean plays, the textual student is tied more firmly to the necessity of consulting those originals. Even the examination of the variants which other students have discovered requires at least some form of photographic reproduction; sometimes the order of two states can only be determined by observing displaced or damaged type, or some irregularity in lay-out. An example of this is found in the outer forme of sheet H of John Webster's The White Devil (1612): here the differences between two states are very considerable, involving the addition or deletion of a whole line, an entry and an exit, and the substitution of one perfectly acceptable word for another; yet the argument for the order of printing, and hence for the authority of the two sets of readings, rests on the exact arrangement of type for H2v and 3—the precise disposition of the lines of type on both pages, and an irregularity in the indentation of the first line of one of the speeches.[4] Both spacing and indentation are normalized in a Malone Society Reprint,[5] and so such an edition would be an insufficient authority for a student working on the variants of this text. For work on plays with press-variants, the most convenient, and generally available, textual authority would be a
Another important advance in textual studies has been the development of techniques for distinguishing the type set by individual compositors in Elizabethan printing-shops. By undertaking spelling-tests, and by tracing the recurrences of identifiable pieces of type, Dr. Alice Walker,[7] Dr. Hinman[8] and other scholars have been able to evaluate more precisely than before the authority of a text set by any known compositor, and the varying authority of one set by two or more of them. This means that a student will now wish to apply any and every test for distinguishing the identity and number of compositors employed on the text he is studying. For applying spelling-tests, the Malone Society Reprints are barely satisfactory—their shortcoming in this respect is that, in prose-lines or full lines of verse, their normalized spacing makes it more difficult to judge whether words have been shortened as an aid to justification. But there are other tests for which such old-spelling editions are no use at all; such are the analysis of variations in indentation or in centering of stage-directions, details which are always normalized in the Malone Reprints.[9] A student will be very unwilling to be deprived of these tests, for they are among the easiest to apply. He will be unwilling, too, to be prevented from judging the 'look' of a whole page as set in the original, for sometimes one of two compositors will give a looser, sharper or more regular appearance to a page of print than his fellow, and such a distinction can be discerned by a trained eye in a moment and, sometimes, greatly facilitate the
Progress has also been made in the bibliographical problem of reconstructing the press-work of early books, and, again, the textual student has learned these techniques. From a study of the press-work of a book, he can sometimes find help towards identifying the work of two or more compositors, discovering the method of proof-correction used, or adding to his understanding of the authority of a complex text; occasionally it can help him to argue that a text was set hurriedly or leisuredly, and this in turn can influence his assessment of the accuracy of its compositors and hence of the authority of the whole work.[10] These investigations depend almost wholly on evidence which is not, and could not be, reproduced in any facsimile reprint: identification of skeleton-formes, for instance, involves the precise size and position of headlines, or minute variations in the length or width of rules, or imperfections in types and ornaments; proof that a printer used two cases of type, as James Roberts did for Hamlet (1604/5),[11] may depend on minute variations in the forms of type such as only a photograph could reproduce.
More recently, new knowledge has been gained about composing methods and it is now known that, both for folios and quartos, an early printer might cast-off his copy and set by formes, not by consecutive pages.[12] This is of considerable importance to students who wish to analyse compositors' habits and to make editorial judgements: inaccurate casting-off might cause a compositor to change his habitual spellings in order to lengthen or shorten a passage of prose at the foot of a page, or it might cause him to modify the lining of the verse as found in his copy. All textual students will wish to know precisely when and where they should expect such interference and here again a facsimile reprint is of no use to them. Such investigations involve a knowledge of press-work, the identification of odd pieces of type and the analysis of occasional intrusions of italic type, and for all this a
The influence of inaccurate casting-off on the lining of verse plays is one aspect of a problem which confronts most students working on Elizabethan and Jacobean plays. It is widely known that the availability of text-space influenced the line-arrangement and the exact position of stage-directions in early plays, but the extent of this kind of textual interference may be properly appreciated only by working continuously with original editions or photographic reproductions of them. The Malone Society has high standards in this respect, asking its editors to
If the facsimile reprint were the only alternative to a copy of an original edition, it would be good sense to try to improve the standards of reproduction of such reprints. But there is an alternative in the photostat or some other mode of photographic reproduction. Professor R. C. Bald suggested some years ago that
Of course even photographs have their limitations. But it would be unwise to be deterred from issuing them by the ill-informed practices of J. S. Farmer in preparing his Tudor Facsimile Texts, or by a knowledge of the kind of errors which can be introduced by the retouching of some kinds of photographic plates. The needs of the textual student would best be served by issuing some reliable form of photographic reproduction, which had been overseen by a responsible scholar, as Sir Walter Greg oversaw the excellent series of Shakespeare Quarto Facsimiles which are now being published by the Oxford University Press. Such an overseer could select the cleanest available copy of the first edition and, where press-variants had been discovered, prepare a composite copy which would give the preferred states of all variant formes, whether 'corrected' or 'uncorrected'. In addition, it would be good to adapt the Malone Society's current procedure and print the overseer's report on press-variants (if possible, with reproductions of the rejected variant formes) together with his lists of misprints and doubtful readings (which would be informed by a scrutiny of all available copies). Finally there could be a brief introduction, again on the model of the Malone Society Reprints.[16]
The various photographic processes have differing advantages and disadvantages: prints made from microfilms are, for instance, very inexpensive, but books printed from photographic plates are easier to handle and might prove more susceptible to editorial supervision. Considerable experiment would be required before the best way of providing a large number of high-grade photographic facsimiles is discovered, and this may not prove to be the cheapest method of reproduction. If its cost cannot be kept low enough, some form of subvention or special subscription would be required: those who gave their support to such a venture would know that they were providing the
It might be contended that many Elizabethan and Jacobean plays do not present problems complex enough to justify the provision of photographic facsimiles. But as textual students learn more about the working of compositors in the early printing-houses, so 'unimportant' plays come into prominence in their studies; thus, for example, The Two Maids of Moreclack has become necessary for a study of the Pied Bull quarto of King Lear because it was partly set by the compositor who, a year earlier, had set King Lear.[18] When one is working on a complex dramatic text, it can be a great help to observe one of the compositors of that text working on another play, no matter how trivial the textual problems of that second play may be. The ideal for a textual student is the complete corpus of Elizabethan and Jacobean drama in photographic reproductions, overseen and commented upon by responsible scholars. And with advances in cheap photographic processes there is little reason why concerted effort should not provide this in the course of fifty years or so. The sooner the student's needs are recognised and the mechanical possibility acknowledged, the sooner the ideal can be realized.
The Old-Spelling Critical Edition
A far more complicated problem is the rationale of old-spelling critical editions of Elizabethan and Jacobean plays. And here also it must be expected that the advances in textual studies have modified both purposes and achievements. At the beginning of this century it might have been supposed that these editions were especially prepared for the textual student; now, in the introductory remarks to his old-spelling edition of Dekker's Dramatic Works, published in 1953, Professor Bowers has stated clearly:
The simplest example will help to show how some modern readers are served. Let us neglect, for the time being, all occasions when there may be some doubt about what old-spelling an editor should print or when there is an ambiguity in the meaning of the text, and let us suppose that a 'modern reader familiar with seventeenth-century usage' is confronted with an old-spelling, critical edition of a play of that period: what will it mean to him? Let us take, for example, the word owl. When Mr. F. L. Lucas, the learned editor of Webster, found 'Oowle' in the original edition of The Duchess of Malfi, he wished to keep that spelling in his edition of the play, commenting: "Oowle can only mean "Owl" and is far too expressive to be given up."[21] Another reader, of a more precise turn of mind, might seize on the same spelling as an example of ME 'Q' becoming late ME 'ū', under the influence of the 'w',[22] and might presume that Webster required such a pronunciation. But someone familiar with both seventeenth-century usage and recent textual studies will know that these readers are making unwarrantable assumptions; at once several questions arise:
- 1). Was this spelling in the printer's manuscript copy, or did a
compositor introduce it? Or, to put this another way, is this a characteristic
spelling of Nicholas Okes' 'Compositor A' who set 'Oowle' on El (II.iii.9)
but 'Owle' on K4 (IV.ii.360),[23] and was the position in which the word occurred on the printed page such that his spelling might have been influenced by the need to justify the line of type or a desire to emphasise the beginning of a line or speech?59
- 2). If 'Oowle' was in the copy, did the scrivener, who prepared this manuscript,[24] introduce this spelling, or was it in the author's manuscript?
- 3). If 'Oowle' was in the author's original manuscript, was the spelling just a flourish, or an accident, or the result of a significant, though probably unconscious, act?
- 4). If 'Oowle' was a significant authorial spelling, how should it be pronounced and how should this pronunciation contrast with the author's pronunciation as a whole?
- 5). Is the capital 'O' authoritative and significant?
Those scholars who have prepared a modern edition reproducing the spelling of a first edition would claim that its spelling is nearer to the author's original spelling than that of a modernized text.[25] This, of course, is true, for some of the author's spellings will survive the modifications of scribes, compositors and proof-correctors, and the number of survivals may be high, especially in books printed before
If a 'modern reader familiar with sixteenth- and seventeenth-century usage' cannot (or, rather, should not) take the spelling of an old-spelling critical edition seriously with respect to the author's intentions, there must be some other reasons for presenting him with such a text. One reason that has been advanced is that it is appropriate to read an author in the spelling of his first edition, even if that was not the author's spelling—an old play in an old spelling. Sir Walter Greg has said:
It is hardly relevant to bring up the question of pronunciation in this connection, for if it would be hard to make a consistent attempt to speak the speeches from an autograph manuscript as the author would have pronounced them, it would be impossible to pronounce them in any meaningful fashion from the doubly or trebly confused orthography of a printed book. And failing a consistent Elizabethan pronunciation, there seems little point in restoring a partial 'Elizabethan' pronunciation to those few words whose old spellings more clearly suggest a sound different from the customary modern ones.
Perhaps one of the chief reasons why scholars prepare modern, old-spelling editions is the difficulty of preparing modern-spelling ones. The ambiguity of the original spelling presents the most intractable problems. Not that a reader familiar with seventeenth-century literature will be perplexed by 'lose' for 'loose' and 'lose' in modern- or in old-spelling texts; nor 'curtsy' for 'curtsy' and 'courtesy'; he needs to be watchful for such ambiguities in both kinds of texts. But sometimes the ambiguity of an original edition embraces two modern words which are not clearly related in form or sense: so in 'How now brother what trauailing to bed to your kind wife'[29], 'trauailing' is an old spelling for both the modern 'travelling' and the modern 'travailing'; or 'Machiuillian',[30] besides being the equivalent of the modern 'Machiavellian',
So far only ambiguous spellings actually found in first editions have been considered, but the argument must be taken further. Several facts need to be remembered: firstly, that a compositor could remove ambiguous spellings as well as introduce them; secondly, that some spellings were ambiguous to Elizabethans which are now understood in one sense only; thirdly, that ambiguity of spelling does not necessarily imply ambiguity of meaning, even if it could be proved that the author himself was responsible for it—he could have used the ambiguous form unintentionally. It follows that an experienced reader of an old-spelling text knows that many of the author's ambiguous spellings may have been lost and that any ambiguous spelling in the text may be fortuitous; he would in fact be on the look-out for double meanings at all points, regardless of the spelling of any particular word. Old spelling is therefore no guarantee that a reader will appreciate all the author's meanings: its ambiguities will often mislead the inexperienced reader and must always be questioned by the experienced in the light
Early dramatic texts have some ambiguous spellings which could not possibly imply ambiguity of meaning, but rather two mutually exclusive meanings. Such are
Old Spelling | Modern Counterparts |
heare, etc. | hear, here |
I | I, Ay |
of | of, off |
the | the, thee |
then | then, than |
to | to, too |
whose | whose, who's |
It has been said that an old-spelling, critical editor 'could console himself' in difficulties such as these 'with the knowledge that his scholarly readers were in possession of sufficient evidence to make up their own minds'.[32] But this is no longer true: a choice between two exclusive meanings, as between modern to and too, may often depend on a knowledge of a compositor's predilections, of text-space and type-shortage, and so to attempt a decision in any particular instance, a 'scholarly reader' must be furnished with a copy of the original edition or a photographic reproduction of it. Without provision for all this, no reader should dare to make up his mind. Clearly this is an editor's responsibility, and the most convenient way of assuming it is to give
The time has come for editors to consider whether they should not adopt the policy of 'all and nothing' with regard to the spelling of Elizabethan and Jacobean plays: for those who can attempt to interpret the old-spelling, there can nowadays be the photographic reproduction of a first edition; for those who cannot (or in the time at their disposal do not wish to exert themselves to that extent), there should be a critical edition, as fully modernized as possible—one that can be relied on not to give meaningless or erroneous detail, and one that interprets the ambiguities of the original text in the light of the fullest possible knowledge. Anything between these two extremes would be incomplete for the specialist, and misleading for other readers.
The Photographic and the Modernized Edition
Problems of spelling must be considered with other problems, of which the choice of punctuation is the most important. At first, spelling and punctuation seem to offer very much the same difficulties: as early compositors altered the spelling of their copy so they altered its punctuation in accordance with their own predilections, the availability of type, and the exigencies of justification; and the author's punctuation, like his spelling, may have been modified by professional scriveners, and printing-house editors and correctors. Moreover there are some ambiguities in Elizabethan usage, notably the use of '?' where a modern writer would use '!' and the occasional omission of queries. So it may be said that the right place for the 'old punctuation' is the first edition or a photographic reproduction of it. There is no point in reproducing it in a modern, critical edition, because only a reader who knows how and where the compositor may have modified the author's original punctuation can begin to interpret it properly; if he does not know, for instance, that Nicholas Okes' 'Compositor A', around 1612, very frequently added unnecessary punctuation to the ends of verselines or that William Jaggard's 'Compositor B', around 1622, frequently introduced parentheses,[33] and if he does not know where these compositors have been working and under what conditions, the punctuation of a first edition from either of these workshops would seriously mislead him with regard to its author's intentions; a reader of a modern edition with 'old' punctuation can take no single punctuation mark on
But there is one important difference between spelling and punctuation, namely that there is no one, standard, modern punctuation, as there is a standard spelling. This means that an editor may punctuate to please himself, but not to conform to any generally accepted model; so he needs a further directive than simply the call to modernize. When patient analysis of an original edition discovers no trace of the author's punctuation the modernizing editor must be told to modernize in the light of his own taste and his own knowledge of his author's style and artistic intentions. But when circumstances are more favourable and the editor can distinguish some individual marks of punctuation that were probably in his author's manuscript, then he may be counselled to attempt to develop a kind of punctuation for that particular text which will retain—or, better, represent—those authorial elements. This will never be an easy task but anyone who has taken part in a theatrical performance will know how radically a change of pointing can alter the speed, tone or force of almost every line of dialogue; in comparison the choice between lanthorn and lantern is often trivial. For this reason special care should be taken with the punctuation of dramatic texts and an editor should not rely solely on his own ideas of standard modern punctuation until he has tried every means of discovering the nature of the pointing of the author's manuscript behind the first authoritative edition. Very occasionally it may be possible for him to attempt a 'restoration' of the original punctuation; often he may be able to evolve a mode of punctuation that reflects some traces of it. Spelling must be either 'old' (reproducing the copy-text exactly) or 'new' (in accordance with the modern standard), but punctuation, by its nature, can have many variations, and therefore admits a compromise solution.
The use of italic type and capitals, and the treatment of compound words and elisions, may occasionally allow similar compromises, for a reader can accept modifications of standard modern practices in these respects without much difficulty. Capitals were usually used, not where emphasis was required (as A. W. Pollard once suggested they were[34]), but wherever a word began with 'C', 'T', 'M', or 'A', or some other group of letters depending on meaningless tricks of the author's hand-writing or the compositor's setting, or on the availability of type; an
The principles suggested for the treatment of spelling would hold good for punctuation and all these other details of an editor's task—with some modifications in favour of the restoration of the author's usages wherever an editor judges that he can undertake this with a fair chance of success. These principles are different from those which underlie most of the editorial work in progress today. Yet they have been determined in the light of new and developing textual and bibliographical techniques and take advantage of new techniques for preparing cheap photographic reproductions of the original editions; these are factors which have become important after the present orthodoxy was established, and therefore the alternative editorial principles might well be worth serious consideration.
They have only been stated in simple form in this paper; elaborations would have to follow. Fortunately, experience of modern-spelling editions has been gained in preparing the multi-volume editions of Shakespeare, like the Yale, New Cambridge, and New Arden Shakespeares, and in the new complete editions by Professors Alexander and Sisson; here many different editors have experimented with ways of collating the spelling of the original edition where this affects the choice of an emendation, of restoring and interpreting what they judge to have been the author's punctuation, lining and so forth, of representing ambiguities and, generally, of being both fully responsible and fully interpretative. Their methods have not always been informed by up-to-date textual and linguistic understanding, but, when they have been, these editors have achieved useful work. The highly developed methods for editing type-facsimile reprints would likewise aid the preparation and overseeing of the new photographic reproductions. For complex textual authorities, involving more than one substantive
These difficulties need not stand for long in the way of preparing modern photographic reproductions and modern, fully interpretative, critical editions, if such texts would truly fulfil the present-day needs of students and readers, and what can be guessed of those of tomorrow. In considering these alternative principles for editing and reproducing Elizabethan and Jacobean plays, each reader should consider his own experiences and ask how frequently he has wished for a closer approximation to the original edition than a type-facsimile reprint or an old-spelling, critical edition could provide, and how frequently he has wished that obscurities had been resolved when he has been reading an old-spelling, critical edition, without a primary concern for textual matters. He should ask, too, whether a photograph and a fully responsible, modernized, critical text would not answer all his needs.
Notes
But see F. Bowers, Textual and Literary Criticism (1959), pp. 117-150. for an extensive survey of the problem; also, the comment of C. J. Sisson in New Readings in Shakespeare, (1956), I, 35-37.
I.e., the state of the text which has departed least from the substantive reading of the author's manuscript. Usually this will be the uncorrected state; but when it can be shown that the author was responsible for the corrections, the corrected state would be preferred; cf. F. Bowers, The Library, 5th ser., vii (1952), 262-272.
Cf. Textual Problems of the First Folio (1953), passim, and Studies in Bibliography, VII (1955), 51-67, and VIII (1956), 95-111.
For examples of an editor's deductions from a study of press-work, see the Textual Introductions in Professor Bowers' edition of the Dramatic Works of Thomas Dekker, I (1953), II (1955), and III (1958).
Rule 7 of the Malone Society's rules for editors (Collections Volume IV (1956), p. 67) states that 'Wrong-fount letters are to be corrected, and no further notice need as a rule be taken. Under wrong-fount letters are included italic letters in a roman word or passage, and so forth. . . .'.
These problems are discussed in the introduction and notes of the present writer's forthcoming edition of The White Devil.
Microfilms would be a much cheaper alternative to photostats, but, for textual work (involving constant comparison of several different works, or collation of several editions of one work or of numerous copies of one edition), they are by no means so convenient.
Cf. W. W. Greg, The Editorial Problem (1942), p. li: 'So long as there is any chance of an edition preserving some trace, however faint, of the author's individuality, the critic will want to follow it.'
Cf. Muriel St. C. Byrne, The Library, 4th ser., IV (1923), 9-23, where it is shown that Munday's peculiar '-oo-' spellings were usually found in books printed before 1590, often found in those printed in the nineties, but seldom found in those printed after 1600.
Cf. Alice Walker, Studies in Bibliography, VIII (1956), 111: 'we should beware of supposing that, when Jaggard's compositors A and B substituted 'murther' for 'murder', as they usually plainly did, they were expressing anything more than a preference for one spelling over another'.
E.g., when there is no modern equivalent in sense or metrical value, or when the original spelling was meant to be dialectal or antique, or when it is necessary to retain a rhyme.
Textual Introduction, The White Devil (forthcoming), ed. J.R. Brown; and Alice Walker, Textual Problems (1953), pp. 9-10.
| ||