University of Virginia Library

III

Three decades ago, Charlton Hinman "noticed by chance" that what seemed to be the same damaged letter "m" appeared frequently in the Shakespeare

First Folio. By examining the Folio under slight magnification, Hinman was able to identify 615 distinctively damaged pieces of type and to trace their 13,000 appearances throughout the Folio. Their patterns of recurrence allowed Hinman to identify the pairs of type-cases used by the Folio compositors. When identifiable types from B1r appeared only on Cr4 and types from B2v appeared only on C3v, for example, Hinman could demonstrate that two pairs of type-cases were being used: one into which B1r was distributed and from which C4r was set, and one into which B2v was distributed and from which C3v was set. This type-recurrence evidence supported the spelling evidence which indicated that Compositor A set C4r and Compositor C set C3v. "In such circumstances," Hinman concluded, "there can be no serious doubt that the two pages . . . were set simultaneously."[20]

My investigation of the A1 quarto of Faustus followed Hinman's methods as modified by Blayney for application to quarto texts (see Texts, pp. 93-94). I examined the unique Bodleian copy of A1, type by type, under 7x magnification with a calibrated lens. Each type that differed in any significant respect from the usual appearance of that character was noted on a photocopy facsimile. 2250 types were so marked. I then examined each sort individually and rejected as evidence most of the types originally marked, usually because, as Hinman discovered, certain sorts are often similarly damaged because they are especially vulnerable to a particular kind of injury; identical injuries in several types are useless for identifying individual types. The blackletter font used to print Faustus, which was hardly fresh in appearance when Simmes first used it in 1594, had, in the intervening decade, been used for at least twenty-seven books.[21] By the time Faustus was printed, the font was old and battered: letters in which a vertical joins a curve (b, d, h, m, n, p, q, r, u) are almost all damaged at the join; ascenders and descenders (b, d, h, l, k, p, q,) are frequently bent or broken. Nearly every letter has a nick, chip, bend, break or bite (though not necessarily a distinctive one).[22] Nevertheless

I found sufficiently distinctive damage to allow for accurate identification of 599 individual types and made a sketch of each, recording the position of its initial appearance along with any subsequent appearances. I then repeated the examination of the entire quarto, type by type, in order to ensure that no appearances of the 599 distinctive types had been overlooked. The result was 89 appearances made by 37 types (listed in Appendix I).

Although 37 recurring types out of a possible 599 is a surprisingly low number, it should be stressed that because A1 survives in only one copy a type-recurrence study is inherently limited in its ability to gather large amounts of reliable evidence.[23] Hinman noticed that some identifiable type defects are liable to be inked over occasionally, so that a distinctive type might be unrecognizable in some copies, and he concluded that a "thorough investigation will certainly require the use of more than one copy" (Printing and Proof-Reading, I.55). When Blayney compared the damaged types he had identified in the Bodleian copy of Q1 Lear with those in the Trinity College copy, he found that more than a thousand supposedly damaged types were, in fact, simply inking flaws (Texts, p. 95). And Blayney has been criticized because he did not examine more of the extant copies. In similar studies, Anthony Hammond found it necessary to consult ten copies of Q1 The White Devil and Paul Werstine examined all of the extant copies of Q1 Love's Labour's Lost.[24]

Unfortunately, such comparison and cross-checking is impossible with A1 Faustus because the Bodleian copy is unique. The difficulty is felt even more keenly because of the uneven inking of the extant copy. On several pages the inking is quite heavy (A4r, B1r, B3r), and consequently some identifiable type damage may have been inked over. Conversely, on other pages the inking is so faint that there are sections of text (B3v:21-35, C3r:33-37, D3r:1-10, and E3v:29-34 are prime examples) in which what may be actual type damage cannot possibly be distinguished from mere inking flaws. I have no doubt that many distinctive types have chosen these pages on which to make their only reappearance, but, short of discovering a better inked copy of A1, there is no way to trace them.

Blayney, who rightly argues for greater rigor in type-recurrence studies

of quarto texts, has shown that a large quantity of evidence—certainly more than 5 or 6 types per page—is necessary in order "to allow the page-order of a quarto to be determined with any great precision" (Texts, p. 91). Given that the "evidence-density" of recurring types in the pages of Faustus is relatively low, it would be foolish to try to prove the order in which the pages were set, and I have not attempted to do so. Despite their limited numbers, however, the recurring types do form clear patterns from which several general conclusions regarding distribution may be drawn.

The black-letter font was apparently large enough to enable two entire gatherings (four formes) to be in type at the same time. Types from sheet A do not reappear until sheet C; types from B not until D; from C not until E; and from D not until F. It is reasonable to deduce that A did not need to be distributed until C was set; B not until D was set; C not until E; and D not until F (see Distribution Tables in Appendix II). But because the only identifiable type from outer D appearing in sheet F is a capital "W" (W3), which may simply have been cannibalized from the margin of an undistributed page, we have no evidence that outer D was necessarily distributed. Thus, the relatively large type supply meant that only seven formes needed to be distributed during the printing of this relatively short quarto.[25]

This type-recurrence study was undertaken primarily to determine whether or not the two compositors were setting simultaneously. The type evidence, minimal though it is, seems to point to the use of two separate pairs of type-cases. Twenty-two identifiable types from sheet C reappear for the first time in sheet E. Types last seen on C1r and C4v appear only on the pages of E set by Compositor X. Types from C2v, C3r, C3v, C4r appear only on the pages of E set by Compositor Y (see table 4).

Apparently, C1r and C4v were distributed into one pair of type-cases from which X set E2v, E3v(lower), and E4r; while C2v, C3r, C3v, and C4r were distributed into another pair of cases from which Y set E1r, E1v, E3r(lower), E3v(upper), and E4v(lower). Hinman observed that, in the Shakespeare

Folio, "a compositor would ordinarily distribute pages which he had himself set before he distributed material set by someone else" (Printing and Proof-Reading, I.124). Indeed, Compositor X is here found distributing the Faustus pages that he had set (he was responsible for all of C1r and most of C4v), while Y is found distributing the pages that he had set (all of C2v, most of C3r, some of C3v, and all of C4r). Compositor Y distributed two pages from the inner forme of C and two pages from the outer forme into his own case. The sudden influx of types from inner C on E3r suggests that the inner forme was distributed sometime after Y set E1r and E1v and before he began on E3r. In addition to distributing C1r and C4v of the outer forme, Compositor X may have distributed C1v and C2r of the inner forme as well (both of which he had set) although no identifiable types from these pages recur.

The evidence that two pairs of type-cases were in use does not, in itself, prove simultaneous setting by two compositors. Theoretically, the same compositor could set one page of a forme from one case and then another page from another case. However, the combined evidence of distinctive compositorial patterns of spelling-punctuation and the recurrence of identifiable types may be taken as proof of simultaneous setting. For while there would have been little point in employing two compositors, each working at his own type case, alternately to set pages and part-pages for the same forme, simultaneous setting, as Hinman points out, "would have secured the very important advantage of speeding up composition in relation to presswork" (Printing and Proof-Reading, I.109). It would appear, then, that Compositor X and Compositor Y, working from two pairs of type-cases, set the text of Faustus simultaneously.