University of Virginia Library

Of the fourteen plays printed in the Comedies section of the First Folio, only four were reprints of previously published quarto editions. For the remaining ten texts which were set up from manuscript copy, it is important therefore to discover how they were affected by the four compositors known to have set them in type. Little is understood of the habits, capacities and accuracy of these compositors: most work has been done on compositor B, virtually nothing on C and D, and compositor A of the comedies to Winter's Tale is not, as I shall show, the compositor A of the histories and later, familiar to editors mainly from the writings of Dr. Alice Walker. No attempt to study the influence of compositors on substantive readings can succeed unless the parts of the text which each compositor set are correctly assigned. The best kind of evidence useful to distinguish one compositor from another is supplied when the compositor is least influenced by the orthography of his copy, and it follows from this that evidence resting to any extent on what a compositor may or may not have done with the inferred forms of copy which does not exist, under conditions which can only be guessed at and never reproduced, is less likely to be good evidence. It is unfortunate, therefore, that compositor identification in the Folio has depended perforce mainly on this latter kind of evidence, particularly the testimony of spellings.

In the following paper fresh attributions of pages of the Folio are made from combinations of evidence. This includes the compositors' various habits of spacing after commas in short lines, and at the ends of lines, their typographical arrangement of turned-over verse lines, their preferences in dealing with 'll and th' elisions, and some fairly common spellings supplementing the familiar do, go, and here. The evidence of dashes and catchwords is shown to be of small value in the

comedies, but the new criteria are so generally helpful that there are few pages in the comedies for which the compositors can now be held to be uncertain.

There has been some disagreement about the identification of the compositors of later texts in the Folio, but Hinman's conclusions on the compositors of the comedies have recently been discussed only in an article by Professor A. S. Cairncross which appeared after my own investigations were completed.[1] I propose here to decide the attribution of pages about which Hinman was in doubt and to question some other of his identifications with the aid of fresh evidence which was not available to him before 1963. I should therefore briefly describe the kinds of evidence that compositor study of the Folio has drawn upon and give some account of the evidence I have used here.

It is not necessary to give details of the early work of Satchell and Willoughby, who mainly used spellings to distinguish between compositors, and the work of Hinman before 1963, because this is already incorporated in Printing and Proof-Reading and in Alice Walker's highly-commended Textual Problems of the First Folio.[2] She listed spellings which could be used to identify pages set by compositors A and B, and other spellings they tended to use.[3] Many of these spellings have been tested and approved by later investigators and A and B can usually be readily identified in the Folio by the criteria she supplied. There remains some doubt, however, about particular pages in which the evidence is scanty, and it is conceivable, since she worked from WT onwards, that compositor A of the early comedies is not the A of the histories and tragedies.[4] It is not always easy to use spellings, which of course provide linguistic rather than bibliographical evidence, for, without precise knowledge of the spellings of the copy from which the compositors set their Folio pages, it is hard to identify their shares of texts printed from manuscript copy of unknown character.

However, Miss Walker, in common with later investigators, drew upon evidence of a more strictly bibliographical kind. She observed:

The compositors were further distinguished by their use of parentheses and italics. Similar kinds of evidence were used in other studies of single texts. Although these confirmed her criteria, they did not significantly add to them. Greater advance was made by Hinman who, having already identified an apprentice compositor (E) who usually set from quarto copy in the Folio,[6] distinguished two more Folio compositors, C and D. One of his main aims in Printing and Proof-Reading was to show precisely how work on the Folio was shared amongst Jaggard's compositors. From evidence of type recurrence, changes in box-rule arrangement, running-title irregularities, page-numbering errors and similar bibliographical peculiarities, he determined the order in which the formes of type were composed, printed and distributed into the type-cases. Study of distribution evidence enabled him to show how many sets of type-cases were in use at any one time, and therefore, how many compositors were at work. Individual compositors normally worked at the same case and so knowledge of the case used in the composition of a passage of text usually suggested which compositor set it. Obviously, if two type-cases were in use together, the text set from them was most likely to have been set by two different compositors since it was impractical for them to set simultaneously from a single case. Occasionally, two compositors shared a page, and when, as in I5, distinctive types from different type-cases were found in a single page (usually in separate columns), spelling evidence revealed who the compositors were. Generally, however, knowledge of which compositors had worked from the respective cases in other pages of the quire was sufficient to identify the compositors of shared pages. However, at other times, a compositor who usually worked at one case would move to another to share the composition of a page with a compositor setting from his habitual case. Page bb3v was shared by compositors A and B, setting from A's case x; page mm4 was shared by B and A who set from B's case y. When this occurred, and there is reason to believe that it happened more often than Hinman was able to detect, case evidence alone could not distinguish the compositors. Identifications were made principally (with such other minor criteria as dashes and catchwords) from the evidence of spellings. Spelling evidence again was generally all that showed Hinman that another compositor had set a single page in a quire quite regularly shared by two other compositors.

Hinman's work is all the more admirable when one appreciates that the spellings he used to distinguish the five compositors of the Folio were the spelling variants of only three words: 'do', 'go' and 'here'. Although from time to time he referred to Alice Walker's A and B spellings, this was most often merely to confirm an identification made primarily on the evidence of 'do/go/here'. These are the only words of variable spelling which occur often enough in most pages of the Folio for distinct compositorial habits to be seen. The habitual or preferred spellings of the four main Folio compositors were found to be:

• A: doe, goe, here
• B: do, go, here
• C: doe, goe, heere
• D: doe, goe, here, but also do and go when (by assumption) they were in his copy.[7]

The features of this table which contribute to the difficulty of Folio compositor identification leap to the eye. When 'here' does not occur often in the text (there are many pages from which it is absent), compositors A, C and D are indistinguishable. Similarly, when 'do' and 'go' are infrequent, B and C cannot readily be distinguished by these spellings alone.[8] Moreover, compositor D's preferences are identical with A's and, although Hinman characterises D as a compositor who would reproduce a few 'do' or 'go' spellings from his copy, unless he did this, and unless 'do' and 'go' were in his copy, these spellings cannot distinguish him from A. It is possible therefore that compositor D may be found to have set pages of the Folio, from copy in which 'doe' and 'goe' were the only spellings of these words, which have hitherto been assigned to compositor A.

It is also useful to know how tolerant of their non-habitual spellings the compositors were (or, alternatively, how strongly they preferred one spelling to the other) since A and D could not be separated if A was equally prone to accept 'do' and 'go' from copy. A's preferences in the comedies are strongly defined and contrary spellings are infrequent. Compositor C on the other hand was more inclined to mix his spellings: he will tolerate 'here', and 'do/go' but not generally to the extent that he can be mistaken for D. Compositor D is fairly intolerant of 'heere' and this serves to distinguish him from C when they were setting together from copy with prevalent 'do/go' spellings.[9]

From considerations such as these and his close analysis of the Folio composition and press-work, Hinman assigned 48 pages to compositor A, 121 to B, 58 to C, and 28 to compositor D, of the 302 printed pages of the comedies. There were six pages in which compositors shared. The available evidence did not permit him to be certain about the compositors of 41 other pages although often, when he indicated alternatives, his first suggestion was, by my analysis, correct.[10]

The general principle can be suggested that in circumstances such as prevailed in Jaggard's printing-house, where the compositors were not expected to conform to rigid practices or house rules, the best kind of evidence for distinguishing compositors would be typographical, such as the case evidence Hinman used, or 'psychomechanical', such as varying practices in centering or arrangement of stage-directions, or the setting of dashes.[11] Often, however, these features are as scanty as characteristic spellings and sometimes they are not found on the pages where their testimony would be of greatest use. Typographical evidence may be affected by 'outside' influences such as the printing of non-Folio matter which caused the migration of types and even of compositors under circumstances that can only be guessed at. Psychomechanical evidence is affected by characteristic human fallibility or inconsistency. Compositor B's practice of putting the speech-prefix alone in catchwords is exceptionally consistent. Even so, it points to another intrinsic limitation of the kinds of evidence on which compositor

identification must most often draw: when the evidence is binary, that is, when there are only two ways in which a compositor could deal with a matter (such as putting the speech-prefix alone in the catch-word, or the speech-prefix and the first word or part word of the following speech), the characteristic will still only distinguish two compositors, even if it is adhered to consistently. When as in the comedies, there are four compositors (at least) and all but compositor B are inconsistent, this characteristic on a page has little or no value as evidence, however satisfying it might be to confirm an identification made primarily from other evidence. A clear distinction, furthermore, must be made between evidence which enables the compositor of a page to be identified with reasonable confidence, and evidence which tends merely to confirm the identification or to describe the habits of that compositor. The pertinence of these comments will be seen clearly in the ensuing discussion of the kinds of evidence other than spellings which I have used for the identification of the compositors in the comedies.