University of Virginia Library

PATTERNS OF PAPER USE IN THE WORKES OF BENIAMIN JONSON (WILLIAM STANSBY, 1616)
by
David L. Gants

While the numerous textual variants found in the Jonson Folio of 1616 have drawn a good deal of critical attention, the equally intriguing assortment of paper stocks used by William Stansby during the volume's printing has been largely ignored.[1] When the question of paper does arise, comment is usually restricted to the observation that the Folio exists on both "regular paper" and "large paper" stocks, although the basis for distinguishing between those forms is not always clear. The bookseller H. L. Ford is typical in differentiating them according to their current dimensions rather than original size or other characteristics. Ford suggested a more productive line of inquiry when he observed that "An examination of the sheets used in the copies under review discloses over forty differing water-marks, many of them appearing on identical leaves in each volume."[2] He went on, however, only to note that the paper is overwhelmingly watermarked with pots and that paper featuring a "bunch of grapes...is often found in the engraved title, but not in evidence in any other portion" (3).

The main challenge to booksellers like Ford and to scholars interested in paper has been a practical one: how to obtain precise images and measurements of watermarked paper easily, inexpensively, and in sufficient quantities so that they can be conveniently studied. The first of a series of important technical breakthroughs occurred in the late 1950s, when Soviet researchers began using beta radiography to obtain sharp images of the physical features of paper. A sheet of paper was placed between x-ray film and a beta source; the rays passed through in proportion to the thickness of the intervening paper, producing a negative image of the sheet when the film was developed. In the 1970s Robin Alston developed the "Ilkley Method," similar in basic procedures except that it substituted incandescent light and common photographic film for beta rays and x-ray film.[3] Also in the 1970s, Thomas Gravell

began experimenting with bibliographical applications of Dylux, a pH neutral, photosensitive paper developed by DuPont for proofing photographic negatives.[4] Like beta radiography and the Ilkley Method, a contact print is created by placing the Dylux beneath a sheet of paper; here, however, the energy source is fluorescent light, whose spectrum includes the visible but not the ultraviolet. Each method has its own peculiar advantages, although beta radiography and the Ilkley Method must be used under laboratory or darkroom conditions.

My own study of paper in Jonson's 1616 Folio has relied on beta radiography, because of the unparalleled clarity of images it produces, and Dylux, on account of its speed, convenience, and low cost. Through the offices of the Huntington Library, and from James Riddell's large private holdings, I acquired beta radiograph images of watermarks found in most of the paper groups used in the Folio.[5] For those watermarks not present in Huntington or Riddell copies, and for the numerous variant states that occurred in many of the paper groups, I used Dylux to capture examples from the holdings in the Folger Shakespeare Library, the Special Collections Department at the University of Virginia Library, and from my own copies. To exploit fully the information held in the Dylux images I needed to develop an image enhancement procedure that would highlight the watermarks, chain lines, and wire lines while minimizing the distraction caused by the ink of the text. This I accomplished using computer technology, converting the Dylux exposures to digital form and then manipulating them with inexpensive and readily available software.

Dylux has three technical features that make it extremely useful for such digital analysis: a sensitivity to small changes in the amount of light exposure, thereby giving it the ability to reveal not only watermarks but also chain and wire lines; a broad palette of color within the blue spectrum, allowing one to isolate a detail or group of details in part through color differentiation; and a chemical coating of fine grain which produces images that do not break down under repeated enlargement. These features make Dylux exposures particularly apt subjects for digital image enhancement. The software I employed can also reduce to manageable background haze the type clutter that frequently obscures a watermark. By applying various digital enhancement routines one can easily increase the clarity of a watermark image.

In the observations that follow, I suggest kinds of insights that can be gleaned from this scrutiny of paper. The study is based on a close examination

of 60 copies of the Jonson Folio and is further informed by a survey of paper stocks used in the 74 titles known to have been printed by Stansby from 1615 through 1617. The evidence provides insights into overall patterns in manufacturing the Folio, and it also confirms and expands earlier insights about the production of several problematic sections of the book, including about possible authorial intervention while the volume was at press. While serving as a methodological study, then, it also provides a keener understanding of the forces acting upon the Jonsonian text, and it offers new insights into the workings of one of the busiest printing houses in early modern London.

I

Allan Stevenson has observed that, when planning the various components of a proposed book, the printer or publisher "generally arranged for paper sufficient for that book only and paper homogeneous in size and quality."[6] Economic circumstances fostered such practices, for "Paper was too expensive a commodity, too space consuming, to make any other system really practicable" (21-22). As a consequence of this practice, books from the common press period are often partially or completely printed on paper bearing a common watermark or watermark pair. One also frequently finds the other extreme, that is, a book with "a considerable diversity of papers, mixtures arising from certain practices of gathering and distribution within the paper trade" (21). These groups of papers usually consist of leftover sheets from another project that a printer "would save for later use, or else return to his publisher or patron or paper merchant"[7]—who would then, presumably, pass them on to another customer.

These phenomena are commonly referred to as "runs" and "remnants," and evidence of both occurs with regularity in Jonson's Workes. Six groups of paper stocks make up the main supply used in printing 225 of the 257 sheets in the Folio, each group contributing most or all of the paper for a significant portion of production. The book is a folio-in-sixes and collates ¶6 A-4P6 4Q4; the runs of paper occur as follows:

             

Paper Stocks [8]	Gatherings
1	G-P
6 & 7 (combined)	3X-4I
6 & 15 (combined)	4K-4Q
10	3G-3N
12	A-E, R-2A, 2Z-3D, 3N-3T
10, 12, & 13 (combined)	2G-2X

The distribution of like groups of paper into these recognizable and dominant patterns indicates that Stansby used large agglomerations of mostly homogenous paper stocks in his printing work.

In addition, three other smaller groups dominate the production for short spans of time:

       

Paper Stocks	Gatherings
3	Q, 2C-2F
5	2B-2D, 3E-3F
11	F, 3P-3R

Stansby regularly used remnants he had stored or had acquired from others; they often fill small, temporary gaps or pad the dominant stocks to make them last longer. Occasionally these remnants will provide the major paper source for a short period, as in P3.4, Q1.6, R3.4, 2L, 2O-2P, 2V2.5, 2Y (reset), 3D3.4, 3E, 3F1.6, 3S, 3V, 3X3.4, and 4B.

When one of the groupings gives way to the next, the transitional sections show a blending of the outgoing and incoming stocks over one or more formes. The manner in which runs and remnants appear, disappear, and then reappear can reveal clues about the day-to-day activity in a printing house. For example, at certain points in the Folio's printing, Stansby's use of two distinct paper groups alternates back and forth with strict regularity, as in gatherings L-O, where Group 10 appears briefly in six of twelve sheets, then fades out again. The same phenomenon surfaces in gatherings 3P-3S between Groups 11 and 12, and in a more complicated dance among Groups 10, 12, and 13 in gatherings 2H-2M. These patterns may indicate that, at least for a short time, Stansby used two presses to print the Workes ; as one press crew used up their supply of paper he allocated more paper from a new supply while the second crew continued working with the original allotment.[9] Stansby's business was quite active during this period, averaging at one point 840 edition sheets per year, and the printing house's busy production schedule must have at times required the shifting of different jobs between presses.

An examination of paper use over time can also shed light on the sometimes muddy relationship among the owners of various of Jonson's texts. Stansby had to negotiate with a number of booksellers in order to secure the rights to print Jonson's Workes . Walter Burre owned complete or partial rights to seven of the nine plays, John Smethwicke owned Every Man out of His Humour, Matthew Lownes owned Poetaster, and various parts of the poetry, entertainments and masques were owned by Stansby, Edward Blount, Richard Bonion, Thomas Thorpe and Henry Walley. In addition, Stansby sold part of the edition to Richard Meighen, a Stationer for whom he had done some work the previous year.[10] Given that the bargaining for rights may

also have included some solicitation of funding from the original copyright holders, and recognizing that paper constituted a significant expense in a book's production, it would not be surprising if Stansby arranged part or all of his paper supply individually with the different interested parties.[11] In fact, this is what we see in the printing of the plays. Every Man Out, owned by Smethwicke, is printed almost exclusively on Group 1 paper, with some remnants in the final quires. This selection of paper changes abruptly to Group 12 with the commencement of printing on Cynthia's Revels, owned by Burre. One gathering into Poetaster, owned by Lownes, the type of paper used switches again, this time to a mixture of Groups 3 and 5. Finally, with the printing of Sejanus and the remaining plays, all owned by Burre, the sheets are again from Group 12, a type that continues with only a few interruptions through Volpone, Epicoene, and parts of the Alchemist and Catiline. Interestingly, the remainder of the Folio (the poems, entertainments, and masques) shows a similar consistency of paper use, with the early sections of Epigrams printed on the leftover Burre stock, the section from the middle of Epigrams through Hymenaei printed primarily on Groups 6 and 7, and the Haddington Masque through the end on Groups 6 and 15. The homogeneity of paper use in the final 20 quires indicates that the various owners of the smaller works likely collaborated to purchase the necessary paper, or that by the time of printing the remaining gatherings Stansby had reached agreement with the rightsholders to purchase the titles outright.

II

Throughout most of the Folio, one group or set of groups will give way gradually to the next with evidence of mixing at the transition points. Occasionally the change among types of paper used will be abrupt, as is the case when production shifted from material owned by one investor to another. However, such a sudden change may also indicate a possible disruption or delay in printing. When these abrupt shifts correlate with similar evidence of interruption such as changes in headlines or typography, then we can begin to develop confidence in the probability of these suppositions. In particular, this type of evidence can shed further light on three problematic sections in the Folio: the order of the settings of gathering 2Y, the timing of the resetting and printing of the initial gatherings of Every Man out of His Humour, and the delay in printing the first play of the collection, Every Man in His Humour.

When Percy Simpson edited Epicoene for the Oxford Ben Jonson, he as-

sumed that Jonson would have insisted his large presentation copies contain as few flaws as possible; consequently, Simpson adopted the dictum that the large paper copies always reflected the latest, most-corrected state of a forme (5:148-149). Because gathering 2Y exists in two settings, one on regular paper and one on large paper, Simpson identified the large-paper setting as the later one. In his review of the Oxford edition in 1957, Johan Gerritsen used arguments based on evidence from recurring headlines to reverse the order of settings (121). The paper evidence supports Gerritsen's version, for the sheets in the large paper copies all show a nearly unbroken string of marks from the Groups D1 and D2 running from gatherings 2S through 3F. On the other hand, the regular paper copies show a disruption in paper usage between gathering 2X and 2Z, with most of 2Y printed on stock not found in the current sequence. Gerritsen also posited that the printing of the second setting took place between gatherings 3E and 3F ("Stansby," 54). Again, the paper evidence bears out and expands somewhat Gerritsen's conclusions. Reset sheet 2Y3.4 bears mostly marks from Group 12, with a few from Group 8, placing it sometime during abrupt changes in paper stocks used for gathering 3C and sheet 3D1.6—the prior contains mostly Group 12, while the latter bears almost exclusively Group 8 watermarks. Reset 2Y1.6 shows mostly Group 8 with a few sheets from Group 13, a mixture that appears elsewhere only at 3D3.4. Finally, reset 2Y2.5 exhibits paper from Groups 5 and 10, with a few rare appearances from groups 12 and 19. This closely matches the mixture of paper groups used in sheets 3E3.4 and 3F1.6. It appears, then, that the printing of reset 2Y was concurrent with the printing of gatherings 3C-3F, and that the reset gathering was printed inner sheet first, followed by the outer and then the middle:

               

Sheet:	3C3.4		3D1.6	3D2.5	3D3.4		3E1.6	3E2.5	3E3.4		3F1.6
		2Y3.4				2Y1.6				2Y2.5
Paper roup
5	x			x	x			x	x	x	x
8		x	x		x	x	x	x
10										x	x
12	x	x		x	x					x	x
13					x	x

While the completion of reset gathering 2Y apparently took place fairly quickly, the resetting and printing of the initial pages of Every Man Out seems to have been a much more complicated affair. Kevin Donovan's work on headlines indicates that Stansby printed reset sheets G1.6 and G2.5 first after finishing Every Man Out gathering P, and printed reset G3.4, H, and I3.4 at a later time.[12] Riddell subsequently examined the paper use patterns based on a smaller sample of Folios and found that his evidence supported Donovan's conclusions regarding sheets G1.6 and G2.5 ("Printing," 156). Because the stocks upon which the reset G3.4, H, and I3.4 were printed were either too common (Group 11) or unique in the Folio (Group 37), he could not identify

when those remaining five sheets were printed. Nor does headline evidence reveal anything about these sheets, for their resetting and printing did not occur while the Folio was at press but rather were put off until later. In this case an examination of the paper stocks used in volumes printed immediately after the completion of the Folio is most helpful.

The clearest evidence for dating is found in gathering H; as Riddell noted, the watermarks on these sheets appear nowhere else in the Folio. This paper Group 37 surfaces in a number of other works printed by Stansby with imprint dates of 1617, constituting the first two quires of Joseph Hall's A Recollection of such Treatises (STC 12707), the first half of Richard Hooker's Lawes of Ecclesiastical Politie (STC 13716), and the middle section of Samuel Purchas's immensely popular Purchas his Pilgrimage (third edition; STC 20507). In each of these works, as in the Folio's gathering H, the paper group appears as a distinct, intact run rather than as remnants spread out over a long period. If this paper group came into Stansby's shop as a well-defined bale, as seems to have been the usual practice during production of the Folio, then we can surmise that Stansby paused in the middle of the massive Purchas folio to print the reset sheet from Every Man Out, then began the Hooker and Hall volumes.

Sheets G3.4 and I3.4 are more problematic as they are both printed on paper from Group 11, a fairly common type of paper in the Folio. Looking only at paper used in the Folio, Riddell surmised that G3.4 "was set at about the same time that the last two plays, The Alchemist and Catiline, were going through the press." He also supposed that the other four reset sheets "were run off more or less together, probably at the same time as G3.4" ("Printing," 156). However, an examination of the paper used in other titles Stansby printed during and immediately after the Folio was at press helps us align the reset sheets within the larger production schedule. Headline evidence shows that these Folio sheets were probably printed together but not imposed in a skeleton forme used to print the rest of Jonson's Workes ; the paper use sequence shows this same Group 11 was the dominant stock used to print the first part of the Purchas volume. In addition, in 1617 Stansby printed the first two-thirds of Davids Learning (STC 23827), a commentary by Thomas Taylor on Psalm 32, on Group 11 stock, while the remaining third plus preliminaries are on Group 37. Judging from the sequence of paper use in these works, Stansby printed sheets G and I3.4 first, followed by gathering H, and did so while he also was printing the Taylor and Purchas volumes, and just before he began those of Hooker and Hall.

In the two cases just discussed, evidence drawn from the use of distinct stocks of paper further illuminates earlier insights into how the resetting of pages was handled in Stansby's printing house. However, the case of Every Man In involved not a resetting but rather a delay in printing. Headline evidence indicates that printing of the play took place in two phases, with its first gathering A printed between the middle of Epicoene (2X5-3D6), and the remaining gatherings B-F between Catiline (3L4-3S4) and Epigrams (3S5-3Z1). The outer and middle sheets of gathering A consist of paper from Group 12, while the inner sheet is a mix of Groups 5 and 12, distribution that matches

paper used during gatherings 2Z through 3C of Epicoene, as expected. Riddell's examination of paper use supports the headline evidence here as well.

The remainder of Every Man In was printed on a mixture of paper groups, in particular a number of rare types, allowing us to expand upon the outline derived from the headlines. When setting and imposing this play, Stansby's men employed the rules from the skeleton formes used in the printing of Catiline to print the final five gatherings B-F, allowing us to place the production of this section after the completion of gathering 3R. However, a closer analysis of the complicated distribution of paper in these gatherings reveals that printing may not have advanced in a straightforward fashion, with B-E printed concurrently with the Epigrams and F not printed until early in the Masques (4F3 ff.).

The dominance of Group 12 paper in most of gatherings C, D, and sheet E3.4 correlates with paper distribution in 3R, and supports the supposition that they were printed at the end of Catiline. Sheet C1.6, however, contains sheets from Groups 16, 25, and 33 in a mixture that does not come into the Folio until 3V3.4, well into the printing of the Epigrams. Likewise, sheet B1.6 shows the presence of Group 25, placing it at the same time as 3V3.4. Next, sheets B3.4, B2.5, and E2.5 have a mixture of Groups 6, 7, 11, and 12 that place them at the same spot in the production sequence as sheet 3X3.4, also in the Epigrams. Closing out this middle section of Every Man In, sheet E1.6 has a number of marks from Group 15, a lot of paper that does not come into use until 3Z, late in the Epigrams. The last gathering in this play, F, comprises paper from Groups 6, 7, and 11 in a ratio that matches the paper use only in 4I, the end of Hymenaei (4G6-4I5).

For most of the Folio, paper and headline evidence indicates that printing proceeded in a relatively orderly fashion, with gatherings usually printed in the order they were to be bound. With the printing of Every Man In, however, paper use patterns point to a rather more jumbled sequence. Stansby began by printing gathering A concurrently with Epicoene. He then seems to have printed concurrently with the Epigrams all of D, the inner and middle sheets of C, and the inner sheet of E. He then printed the outer sheets of B and C, followed by the remainder of B along with E2.5, and completed this section with E1.6. He then put off the final gathering until he was well into printing the Masques. That Stansby chose to print these sheets out of sequence indicates that something may have occurred to force him into this more complicated procedure.

Critics have for more than 100 years discussed the dating of the revisions Jonson made to Every Man In, citing internal and external evidence to support proposed dates ranging from 1604 to 1613.[13] James Riddell has argued that Stansby was forced to put off the printing of Every Man In until late in the Folio because Jonson had not yet completed his final alterations.[14] While not conclusive, the sequence of printing that I have constructed from the

paper evidence supports Riddell's dating of the revision concurrently at least in part with the Folio's production. The breaking up of the play's printing into three separate pieces, one in the middle of Epicoene, one during the early gathering of the poems, and one in the middle of the masques, as well as the general postponement of the play's printing despite its initial place in the Folio, indicates that Stansby may not have had the completed manuscript when work began. Furthermore, the jumbled order in which gatherings B-E went through the press signifies that Jonson may have been sending pieces of the manuscript to Stansby when he completed the revision, irrespective of overall order. With completion of the Folio looming, one can easily imagine Jonson scrambling to finish the promised revisions before Stansby sent the manuscript to the compositors. There is also contemporary evidence that Stansby sent pages to his authors for correction, so the idea that Jonson both corrected and revised during the printing of his Workes is not unreasonable.[15]

Overall, the picture that emerges from the preceding observations shows a printing house bustling with activity, and Stansby himself appears as a master at organizing work. The methods and procedures behind these observations also demonstrate the potential of blending traditional bibliographical scholarship with the tools of digital technology and point to new ways that scholars may build upon the work of their predecessors.

III

The following chart illustrates the distribution of paper stocks throughout the Jonson Folio based on an examination of sixty copies.[16] I have listed individually the fourteen most prevalent stocks of regular paper (1, 3-13, 15-16, and 33) along with the six most prevalent stocks of large paper (M1-4 and D1-2). When a sheet contains reset text, and when a gathering contains two distinct settings, the number of sheets bearing the first setting will be followed (in parentheses) by the number of resettings. Those stocks that occur rarely or sporadically I have combined under the headings "Misc. Regular" and "Misc. Large." The total occurrences are displayed in columns labeled by gathering and sheet (outer, middle, inner). Finally, I have indicated at the top of the chart where a new title commences (including the page in parentheses), and hence where ownership may change.

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

[Description: Data table]

Notes