 | |  |
Notes
I dedicate this paper to the memory of Fredson Bowers whose death on 11 April 1991 deprived the bibliographical world of its guiding force. It is a great personal loss as well. He gave validity to my research when I had no idea that it was anything more than an accumulation of amusing bibliographical details. The idea of formulating my methods of analysis was entirely his. Without that guidance, it would all have amounted to nothing. His criticisms of my thinking and writing were blunt but respectful of my efforts which, at times, fell far short of the mark. For this I am grateful. Sit tibi terra levis. I thank The Huntington Library for an Andrew W. Mellon Fellowship which partially supported the research for this paper, and the staff of Reader Services and the Special Reading Room for their cheerful and efficient assistance. I also thank Mac Pigman (California Institute of Technology) for posing and discussing many of the issues considered here.
This paper assumes familiarity with concepts, principles, methods, and examples presented in earlier papers: "Reproductions of Early Dramatic Texts as a Source of Bibliographical Evidence," TEXT, 4 (1989), 237-268; "Font Analysis as a Bibliographical Method: The Elizabethan Play-Quarto Printers and Compositors" (hereafter "Font Analysis"), Studies in Bibliography, 43 (1990), 95-164; "Bibliographical Methods for Identifying Unknown Printers in Elizabethan/Jacobean Books" (hereafter "Printer Identification"), Studies in Bibliography, 44 (1991), 183-228; and my review of W. Craig Ferguson's Pica Roman Type in Elizabethan England (1989) in Papers of the Bibliographical Society of America, 83 (1989), 539-546 (hereafter "Review").
See "Introduction," "Textual Notes," and "Critical Notes" in C. T. Prouty ed., George Gascoigne's A Hundreth Sundrie Flowres', The University of Missouri Studies, 17 (1942). Prouty's numbering of poems and lineation throughout the edition provides a convenient method of reference to the texts and follows the format "page:line." References to a second edition are to John W. Cunliffe ed., The Complete Works of George Gascoigne, vol. 1 (1907). In general, Prouty's analysis of the bibliographical and textual evidence was skewed by the assumption of a rapid, concurrent printing in Bynneman's shop with Gascoigne doing daily proof until about 19 March 1573 when he supposedly left for Holland. The single fact of Gascoigne's whereabouts during the period of about September 1572 to 25 May 1573 is his attachment to an English regiment in Holland on the latter date. Gascoigne's attendance at the Montague marriage sometime in the fall of 1572 seems plausible, but the rest of Prouty's account is speculative.
The most exhaustive and authoritative discussion of shared printing is found in Chapter 2, "Printing-house Methods", of Peter W. M. Blayney's The Texts of 'King Lear' and their Origins, vol. 1 (1982) (hereafter simply "Texts"). It is impossible to note adequately my debt to Blayney's work in every instance. With respect to the sharing climate, see pp. 51-52, 57.
Some appearances of Bynneman's ornamented initials as follows (single measurement for square pieces; others are height x width; no date for 1573 appearances). Doubleruled 'Vine-F' [16mm]: STC11635, K3; STC22243 (1576), T4. Double-ruled, black arabesque 'I' [16mm x 15mm]: STC11635, A4v; STC6901 (1572), *3v; STC20309, F3v; STC3548 (1574), A4. 'Vine-I' [21mm]: STC11635, A2; STC22241, a2; STC15003 (1574), A4; Bynneman's section (3A-K2) of STC22242 (1574, Middleton), 3A1. 'Vine-O' [20.5mm x 19.5mm]: STC11635, K4; STC3737, A1. Double-ruled 'Sunflower-H' [16mm]: STC11635, B3; Bynneman's section (3A-T7v, 4E-K4) of STC4395 (1571, East and Middleton), 3H5v. 'Cropped Vine-T' [22mm]: STC 11635, Q1; STC22241, Lv4; STC15541, *4; STC25429, E2. Double-ruled black arabesque 'O' [21mm x 22mm]: STC11635, R4; STC13063 (1574), H5. Middleton's 'W' [9mm]: STC11635, 2M3v; STC4395 (1571), A3; STC4055 (1572), *7v, D1. No noted appearance of the ruled 'R' in 2A1v in books by Middleton or East; used later by R. Robinson and R. Braddock.
This is the font that W. Craig Ferguson mis-identified as a Tavernier. The printer omitted a major revision to my "Review," p. 542, beginning at line 3 "The mis-labeled. . . ." The revision is found in "Printer Identification," note 27, p. 208, "Furthermore, the misidentified. . . ."
Details of S-face lower-case variants are found in "Font Analysis," pp. 110-113, with comment on the 'w4' in p. 112.
Ferguson overlooked this 79mm S-font. Additionally, Middleton used it in A Caveat STC12788 (1573), A1-4; An Answer STC540 (1573), A3v-4v; and East in Examen STC11844 (1578), A2-4. These 79mm S-fonts present a paradox. The bare height of the S-face is 82mm: it is physically impossible to reduce this height during lock-up. A 79mm font could conceivably be expanded to 82mm by leading (inserting a thin strip of lead between each line of type) but this was a practical impossibility since the leading would have to be a mere 0.1mm thick. In other instances of type cast on an undersized body, new punches were cut with shortened ascenders and descenders (see "Font Analysis," note 11, p. 152). No such modification occurred for these 79mm fonts. The only possible explanation for the 79mm body is an intended use in headings along with a smaller font as is done in some editorial links in 2M-S.
See Texts, p. 51, for discussion of logical expectations based upon textual divisions and the sharing pattern in Bulkeley where Eld printed twenty-four sheets, Okes the next six, and Eld the last five; and "Printer Identification," p. 224, An Apology STC19295.
See "Printer Identification," p. 223, for examples of symmetrical patterns and note 44 for the apparently symmetrical pattern in All Fools STC4963 which suggests concurrent shared printing except for the identification of Eld-Y1 in AB, GH, and K.
See Texts, pp. 58-59, regarding problems arising from SR entries. The special problem here is that the Register containing book-entries for July 1571 to July 1576 has not survived. The practice of dating individual entries began with Register B in July 1576. See W. W. Greg, Some Aspects and Problems of London Publishing (1956), pp. 23-31.
Two classes of job-lots of papers are encountered in Elizabethan books corresponding to differences in the production and delivery system. First, bales of paper arrived in England for sale to printers and publishers containing homogeneous watermarks consisting of either a single pair of twin watermarks or groups of twins from moulds with very similar or nearly identical watermarks fashioned for a single, large paper mill or for a consortium of smaller mills. Such a job-lot may appear to yield a significant parallel distribution pattern in two sections of a book or in two or more books, but the pattern is illusory, reflecting only the order in which papers from different vats in a production center were gathered and baled for shipment. Second, valuable printing evidence is often provided by papers exhibiting heterogeneous, unrelated watermarks produced in the numerous "single-vat-and-pair-of-moulds" family operations located along networks of streams in various continental locations. Paper factors collected these lots of paper and baled them for shipment to England. The random sequence of individual lots in a bale can provide a rather accurate indication of printing sequence. A parallel distribution of watermarks by sheet will emerge if a daily allotment of papers was supplied to two sharing printers from the same job-lot of heterogeneous papers. An offset in the distribution pattern is clear evidence of when the sharing printer began work relative to the primary printer. If the distribution reveals an exclusive watermark in each concurrent pair of sheets (e.g., no overlap of the watermarks with prior and subsequent sheets) and this relation is constant throughout the book, there seems little reason to doubt concurrent shared printing. An extensive survey of watermarks in books throughout a production year can locate a book in the production schedule given a sequence of heterogeneous job-lots and an overlapping at the beginning or end of the book. For background on watermark evidence, see Allan Stevenson, "New Uses of Watermarks as Bibliographical Evidence," Studies in Bibliography, 1 (1948), 151-182; "Watermarks are Twins," Studies in Bibliography, 4 (1951-52), 57-91; "Chain Indentations in Paper as Bibliographical Evidence," Studies in Bibliography, 6 (1955), 181-195; and David L. Vander Meulen, "The Identification of Paper without Watermarks: The Example of Pope's Dunciad," Studies in Bibliography, 37 (1984), 58-81. The literature on watermarks is extensive, but the outdated A Short Guide to Books on Watermarks (Paper Publications Society, Hilverson, Holland, 1955) provides an early overview. A valuable resource is Edward Heawood's "Sources of Early English Paper Supply," The Library: 2nd ser., 10 (1929-30), 282-307, 429-454; 11 (1930-31), 263-302, 466-498; 3rd ser., 2 (1947), 119-141.
This family of related watermarks appears exclusively in The Huntington's copies of the following books by Bynneman. 1572: Of Ghosts STC15320, An Answer Q1 STC25427. 1573: A Hundreth STC11635, An Answer Q1,3 STC25427,29, Historicae Brytannicae STC 20309, Arte of Reason STC15541, T. First Parte STC22241, T. Garden STC12464 [?mixed with an anomalous "hand-star" paper]; the following books from 1573 were not checked for papers: STC3737, 7623, 11985, 13603, 13846, 19060, 23003, 24171, 24788, 25010. 1574: Sermons STC4449, A Viewe STC15003, A Catholike STC17408, Historia STC25004, T. Defense STC25430, and all of T. Three Partes STC22241 including Bynneman's section 2A-K2. 1575: T. Posies STC11636. 1576: T. Steele Glas STC11645 (A-I2) and The Complaint of Phylomene (K-O). The longevity of moulds was quite remarkable, some producing papers for nearly a half-century. The "hand-star" watermarks appear in a 1604 book printed by Eld. The commonness of these papers in Bynneman's books confirms Richard Tottle's allegations regarding the efforts of French papermakers to sabotage his efforts to found an English paper mill: "the Ffrenchemen did by all meanes possible labor to distroye [theire] worke begonne . . . as by procuringe all our ragges (beinge the chief substance that paper is made of) to be brought over to them, by bringinge in greate aboundaunce of paper at that tyme and sellinge it (although to losse) better chepe then they were hable to doe" (see Edward Arber, A Transcript of the Registers, I, 242, for the complete letter). The time referred to is approximately 1564 onward, with Tottle's effort in approximately 1573.
The fouling process and the principle of random recurrence according to the probability ratio are discussed in Section VI, "Font Analysis," pp. 126-141, especially pp. 129-131, 137-139.
No example of S-face 'g1' or other fouling letters occurs in STC5952, 11759, and 22991. The introduction of 'w4' occurred late in the printing of STC5952. The 'w4' fouling in STC5952 seems to be limited to the final sheet Hh and the preliminaries (A1-8, bc) given the minimal emphasis use of the font in the remainder of the book although some long quotation passages (twenty percent of U5-8 and T1-5, and nearly complete pages at C8-D1 and Q5-7) should yield 'w4' if it was in the font. The clustering of alternate settings suggests that the compositor set several 'vv' from the Y1 case, then several 'w4' from the S1 case on affected pages. The ratios reveal a high-density alternating pattern (STC5952 A2-4v: 'w4'/'vv': 1/4, 7/14, 14/9, 14/4, 10/11, 6/12); STC11759, A2-5v: 3/8, 0/22, 8/4, 5/10, 3/14, 0/11, 0/7; STC25427, E1-2v: 0/8, 9/20, 3/26, 0/8; I1-3: 5/0, 14/0, 5/17, 0/13, 8/3). It is not clear whether the erratic variation is attributable to the preferences of two compositors working on several books simultaneously in irregular stints or the fouling of only one Y1-case. The non-appearance of 'w4' in Y1 settings in 2E, A of the Latin STC19139 and I-2A, * 2-3v of STC19137 in 1572, and the books of Groups I and II in 1573, all of which were printed after the entry of S-face 'g1', probably reflects a complete purging of the 'w4' from the 'v' sort box after its final appearance in 2L2v of STC25427.
The census of May 1583 credited three presses to both Bynneman and Middleton, and one press to East (see Arber, Transcript, I, 248). However, this number probably included the spare proofing press kept by many printers. Judging from Middleton's known output in 1573, he in all probability did not use even two presses simultaneously for any significant length of time unless half of his output is hidden in shared sections. Bynneman's output is credible for two presses and an intermittent use of a third. For extensive discussion of this problem and annual production rates, see Texts, pp. 41, 42-43, 57-58, 61-63.
For example, numbers 82 and 83 were transposed in L3v-4 along with the furniture as both appear in the gutter margin. Page numbers 513 and 516 in STC3737, correctly imposed in the outer forme at 3X1 and 3X2v, remained on the stone (probably along with the correct new numbers) and were transposed during incorrect imposition in the inner forme at 3X2 and 3X1v respectively. In other instances, the numbers were simply transposed or new numbers omitted, as in STC25010 (237, 238, 239, 238, 241; 301, 302, 330, 340), STC13063 (42, 43, 42, 45, 46; 157, 158, 157, 160; 325, 326, 372, 328). A more complicated instance occurs in T. Fourth Parte STC22243 (1576) where T3 is numbered 149 and T3v is numbered 1410. The compositor intended to scavenge a "14" from the outer forme (as he did for T1v-2) and did so. New numbers "5 0" (for page 150) were probably placed on the stone but during imposition a left-over "1" was set instead of the "5," producing the "1410."
See W. W. Greg's discussion of the printer's notation for signatures and page numbers in "An Elizabethan Printer and his Copy," The Library, 4th ser., 4 (1924), 105-107.
STC25427,28,29 answered An Admonition to the Parliament STC10847,48, which was printed surreptitiously by the secret Puritan press in summer 1572. Queen Elizabeth apparently put Lord Burghley to the task of suppression through the agency of the Bishop of London. The authors were imprisoned by 7 July but the queen's furor continued unabated and broke forth in the proclamation of 11 June 1573, which attacked the book and commanded "al and euery Printer, Stationer, Booke bynder, Marchaunt" (and everyone else) to turn in copies to the Bishop of the affected diocese (see Arber, Transcript, I, 464). The Bishop of London reported to Burghley that none had been turned in by 2 July 1573 (Arber, I, 466). Elizabeth issued several proclamations against seditious books in general, but this seems the only book that merited its own proclamation. The book was still "hot" in 1578 when Thomas Woodcock was imprisoned for selling it (Arber, I, 484). An Answer was still in the stock of the bookseller Thomas Chard in 1583-84, which confirms the printing of large editions of Q1-3 as is suggested by the nature of the controversy; see Robert Jahn, "Letters and Booklists of Thomas Chard (or Chare) of London, 1583-84," The Library, 4th ser., 4 (1924), 229, 235. The significant fact is that Q3 STC25429, despite the addition of five sheets of argument, makes no reference to the Puritan's response to Q1 STC25427 entitled A Replie to An Answer STC4711. Moreover, the specific page references in A Replie show that the author Thomas Cartwright knew only An Answer Q1 STC25427. The dating of the series of Puritan treatises and Whitgift's replies is controversial and comment must await a future note. It is clear, however, that STC4711 and STC25429 were being printed during the same period.
The folio format permitted the use of larger titling fonts. A dramatic text in quarto used 96mm roman for the Act/scene heads and pica roman for the stage directions and speech prefixes. The compositor shifted to double-pica black letter in the cast (K2v) and the orders of "the dumme showes" in Iocasta, setting "The Argument" in double-pica italic followed by the 96mm roman verse.
In this context, the compositor's setting of the letters of H. W. and G. T. at the beginning of the "F.J." manuscript illustrates an interpretation that reflects continuous copy. The letters are treated as integral components of the main text and not as preliminaries. Hence they are headed by pica rather than double-pica sub-titles which places them on the same level as the titles of the poems and the narrative links. Furthermore, "H. VV. to the Reader." is headed by a large 56mm "Cornucopia-I" and the first line is set in double-pica, indicating the very beginning of the "F.J." subtext. G. T.'s letter is headed by a smaller 21mm 'R'. Both letters are set full measure throughout and no tail-piece or other ornament separates them. In contrast, "The Printer to the Reader." clearly was a separate manuscript and exhibits the end-of-text "V" format followed by a 17mm x 65mm setting of flowers in A3 (unsigned). The conclusions of the two verse texts of the plays naturally eliminated the "V" format but both set a "Finis" followed by an ornament (see K1v, X4v); the absence of catchwords is normal at the end of independent texts.
See W. W. Greg, "An Elizabethan Printer and his Copy," pp. 107-108. The uncanny correspondences between the two setting problems should not go unmentioned: both involve a transition to a translation of a sub-text from Orlando Furioso, the one an allegory, the other "allegorized," one set from continuous copy with an ornament inserted but no page break, the other set from a junction of two manuscripts with flowers inserted and an inappropriate page break.
Middleton's compositors set poem titles in either roman or italic pica thus: (1) one or more lines of italic; (2) the first line in italic followed by one or more lines of roman; (3) the first line in italic followed by one or more lines of 67mm roman.
The very long lines encountered in the text of 2S1v-2 expanded the measure from 86mm to 103mm and required either modification of the skeleton or construction of a new one. The furniture of 2S1v-2 was replaced since the catchwords as well as page numbers move to the new letterpress margins. 2S3v is normal in this respect, but the page number (343) of 2S4 is 100.5mm from the left margin although the catchword is spaced the normal 86mm.
A third date found in No. 74, "Gascoignes voyage into Holland. An.1572," is given as Gascoigne's date of departure to Holland: "In March it was, that cannot I forget, | In this last March upon the nintenth day, | . . . the very twentith day we set | Our sayles abrode ---" (187:31-36). 19 March 1572/73 served as the cornerstone of Prouty's hypothesis of rapid printing and publication in early April following the belated delivery of copy for the misplaced link. He mistakenly attributed the authorship of the correction notice and the concluding editorial link to Bynneman despite the internal evidence of Gascoigne's authorship. Prouty's conjecture that Gascoigne returned to England during the fall of 1572 seems certain, but that he remained until 19 March 1572/73 is contradicted by the printing and textual evidence. This date is clearly false in that context. Beyond that, the specificity of the date is quite extraordinary in a poem, and even more so because of Gascoigne's emphatic avowal "that cannot I forget." If he indeed was under threat of an investigation by the Privy Council (see Prouty, p. 16) and fled, it would have been of legal importance to publish the claim that he did not sail for Holland until 19 March although Dutch ports were reopened to English shipping in January 1572/73. Flight to avoid investigation and possible prosecution is a prima facie admission of guilt. He had been jailed before for debt (see Prouty, p. 17) and could probably have expected to be again. The published date would serve as testimony that he had been in England all the while even though the Privy Council's agents could not find him. If we trust his prefatory letter to the second edition, A Hundreth was sensational both in court and abroad and the 19 March date probably became "common knowledge." In the final analysis, this date is a fiction like everything else in A Hundreth—the "printer," H. W., G. T., the "sundrie Gentlemen," The Reporter, the "written regyster," and Master F. J. and everyone else in the castle in northern England (see Robert P. Adams, "Gascoigne's 'Master F. J.' as Original Fiction," PMLA, 63 [1958], 315-326; Frank B. Fieler, "Gascoigne's Use of Courtly Love Conventions in 'The Adventures Passed by Master F. J.,'" Studies in Short Fiction, 1 [1963], 26-32; and Leicester Bradner, "Point of View in George Gascoigne's Fiction," Studies in Short Fiction, 3 [1965], 16-22). In short, the printing and textual evidence can be trusted to show that Gascoigne was gone before printing began in late January.
According to the title in "The contents," "Lastly the dolorous discourse of Dan Bartholmew," the first direction could mean anywhere after the beginning of "Dan" in Ee2v (412), which makes no reference to "dolorous discourse," or within the series of poems introduced by the sub-title "Dan Bartholmew, Dolorous discourses" in Ff2v (420). Although "Supplication to Care" does not correspond to any sub-title, it seems clear from the conclusion of No. 5, "To Care I make this supplication," and the sub-title of No. 6, "His libell of request exhibited to Care," that the link belongs before this point in Hh1 (433). It is clear that whoever supplied page 430 for the correction notice picked the first textual break within the "Dolorous discourses" which occurs at the end of the first discourse occupying Ff2v-Gg3v (420-430). The link makes no textual sense in Gg3v (430): its opening reference to the preceding "triumph," "The vaunting verses with many mo," corresponds to page 420. Several textual inconsistencies in "Dan" clearly point to Gascoigne's failure to integrate completely the poems into the editorial frame through a much needed stage of revision. The first discourse in Ff2v (420) begins "I have entreated care to cut the thread," which can only refer to No. 6 in Gg3v which begins "O Curteous Care . . . O knife that canst cut off the thred of thrall." Next, "The contents" lists "his triumphes," The Reporter's first link makes the transition "His triumpes here I thinke will shewe no lesse," and the sub-title also is in the plural: "Dan Bartholmew his Triumphes." But only one "triumph" poem is included. The Posies inserts two additional "triumphs": the "third Triumphe" is No. 34 (2Q4, 327) of "The devises" moved to the new position. The "second Triumphe" either had not been composed or was omitted during the preparation of printer's copy. Gascoigne clearly attempted to employ the "chain" device to link the sequence of poems through a narrative flow from one to the next. The core group may have originally been composed as a chained posie, but the integration into the editorial frame was never accomplished. Perhaps Gascoigne's abandonment of "Dan" and its subsequent incorporation into G. T.'s editorial frame is a symptom of this failure. See also Prouty's comments, pp. 277-294 (it is worth noting two important typographical errors in Prouty's notes to "Epitaph uppon capitaine Boucher" [No. 68] and the Montague masque, p. 282. Both dates should read "1572," not "1573," and are correctly given in the "Introduction"). The reconstruction is unclear as to the amount of delay between the delivery of the manuscripts of "Dan" and the misplaced link but raises the distinct possibility that Gascoigne became confused about the order of materials in the "Dan" manuscript by the time he wrote the correction notice.
This is absolutely clear from Joseph Moxon's contemporary description of the process. Mechanick Exercises on the whole Art of Printing, ed. Herbert Davis & Harry Carter (1958), pp. 239-244. Invaluable insight is provided in W. Speed Hill's comparison of the casting-off marks in the Pullen manuscript with the folio produced from this manuscript by John Windet. See "Casting Off Copy and the Composition of Hooker's Book V," Studies in Bibliography, 33 (1980), 144-161, with a reproduction at p. 147. See also W. H. Bond, "Casting Off Copy by Elizabethan Printers: A Theory," Papers of the Bibliographical Society of America, 42 (1948), 281-291.
The transfer of No. 34 from its intended position in the "triumphs" section (see note 32 above) in "Dan" (as found in The Posies) may reflect this compositional sequence. The Reporter's first link probably was the first to be written and introduces the "triumphs" with No. 34 probably in mind. The three "triumphs" in The Posies obviously were juxtaposed without any compelling internal textual rationale other than general subject matter. It is plausible that, while later assembling materials for "The devises," Gascoigne realised that No. 34 was needed in the courtship sequence of No. 32-35. He inserted it with the title "Another shorter discourse to the same effecte," a reference to No. 33, "The Lover declareth his affection." No. 35 then refers to the momentary "blisse" of No. 34 occasioned by "And how thou seemdst to like me well" (line 15). No. 34 is essential here, since No. 35, "The lover disdaynefully rejected contrary to former promise," requires this stage of recognition by the lady.
Greg and Prouty report a total of four cancellation stubs. Further, the appearance of half of the watermark in B3 of The Huntington's copy indicates cancellation: half-sheet imposition by the "work and turn" method produces a conjugate fold with the entire watermark. Hence, the whole sheet B was printed for at least five extant copies.
Bynneman was quite adept at resolving organizational problems. For example, Supposes was set line-for-line from A Hundreth up to and including the pagination gap at 36-45 for The Posies. Because of Gascoigne's reorganization of the texts in the second edition, the text of Supposes is found in the middle of the book with the original B3 and page 1. Bynneman simply set roman numeral pagination ("clix") to that point to solve the problem.
A second possibility is that the letter contained explicit references to the "missing text." However, it seems that the economics of cancellation would lead to toleration of the inconsistency, especially in view of the fact that Gascoigne was not present to demand the correction. One kind of compositorial error is conceivable as the cause of the cancellation but can be rejected: a mis-imposition of either B1-2v or B1v-2. Such an error would have been caught almost immediately and corrected. For example, only one exemplar of sheet B of Monsieur d'Olive (Clark copy) with mis-imposed pages in the outer form survived, probably in a copy scavenged from discarded sheets by compositors for sale on the black market.
| | |