University of Virginia Library

Press-Variants and Proofreading in The First Quarto of Othello (1622)
by
Millard T. Jones

The aim of this article is to present a list of the press-variants revealed by collation of the nineteen copies of Q1 Othello (1622) included in the Bartlett and Pollard Census,[1] so that a scholar using one copy of the book may know how its text differs from the texts of the other eighteen copies. Since, however, the press-variants seem to provide evidence about the degree to which proofreading in Nicholas Okes's printing-house did or did not safeguard the transmission of the manuscript copy underlying the quarto, this article offers an interpretation of that evidence secondarily. Indeed, the strong probability—discovered by Dr. Alice Walker and further demonstrated by Professor Fredson Bowers[2] —that an annotated quarto served as copy for the Folio has made the degree of Okes's accuracy more crucial than ever to those who try to account for the correspondences and variants between the Q1 and Folio texts. Moreover, the evidence about proofreading indicates that the printshop's effect on the transmission of the text needs to be accounted for in more detail before it is safe to say, as Professors Nevill Coghill and E. A. J. Honigmann[3] have suggested in recent years, that some variants between the two substantive texts of Othello represent different stages of Shakespeare's thoughts, rather than the inaccuracies of scribes and compositors coupled with playhouse cuts in the quarto copy and some editorial interference in the Folio.

Bibliographical descriptions of Q1 are to be found, of course, in Bartlett and Pollard's Census and in W. W. Greg's A Bibliography of English Printed Drama to the Restoration. To conserve space, here I merely list the nineteen copies, giving the Census number, the abbreviation used to refer to the copy in the table of variants, and the copy's location:

• (796) Bodl, Bodleian Library
• (797) MB, Boston Public Library

---

178




• (798) BM1, British Museum
• (799) BM2, British Museum
• (800) CLU-C, Clark Library, U.C.L.A.
• (801) Cohen, Estate of the late W. W. Cohen, N.Y.C.
• (802) Dyce, Victoria and Albert Museum
• (803) Edin, Edinburgh University
• (804) CtY-EC, Elizabethan Club, Yale University
• (805) DFo1, Folger Library
• (806) DFo2, Folger Library
• (807) MH, Harvard University
• (808) CSmH, Huntington Library
• (809) NNPM, Pierpont Morgan Library
• (810) NN, New York Public Library
• (811) Petw, Petworth House, Sussex
• (812) Pforz, Pforzheimer Library
• (813) PPRF, Rosenbach Library, Philadelphia
• (814) Trin, Trinity College, Cambridge

The quarto includes forty-eight leaves collating A2 B-M[4] N2. Since several of the nineteen copies collated are imperfect, they yield only the following numbers of sheets printed by Okes: Sheet A (15 ⅔), B (18), C (17 ¾), D (17), E-H (19), I (18 ½), K-M (18), N (15). Of the book's twenty-four formes—that is, the four-page groups that printed one side of a quarto sheet—nine contain press-variants.

Press-Variants in Q1 Othello (1622)

Sheet B (outer forme)

• Corrected: Bodl, CLU-C, CSmH, CtY-EC, DFo1-2, NN, Trin
• Uncorrected: BM1-2, Cohen, Dyce, (Edin in facsim.), MB, MH, NNPM, Petw, Pforz, PPRF
• [Gummed label]
• B3:15 Now euer ] How euer

Sheet C (inner forme)

• Corrected: Bodl, BM1, Cohen, CLU-C, CSmH, CtY-EC, Dyce, (Edin in facsim.), DFo1-2, MB, MH, NN, Petw, Pforz, Trin
• Uncorrected: BM2, NNPM, PPRF
• C2:15 Saunce ] Since

Sheet G (outer forme)

• Corrected: All copies except Bodl
• Uncorrected: Bodl
• G3:4 purſe ] purfe [with "r" inverted]
• G3:25 to me ] tome

Sheet H (outer forme)

• 1st Stage Corrected: Bodl, Edin, Petw, PPRF [see footnote 4]

---

179




• Uncorrected: Cohen, MB, NNPM
• H1:13 Exit ] [S.D. omitted]
• H1:28 keepe ] leepe
• H2v:4 and ] aud
• H2v:8 Occupation's ] Oceupation's
• H2v:27 God buy you ] God buy, you
• H2v:29 note,O ] note O
• H4v:5 Let ] Tis
• H4v:14 caſtigation, ] caſtigation
• H4v:36 T'would ] Twould
• 2nd Stage Corrected: BM1-2, CLU-C, CSmH, CtY-EC, DFo1-2, Dyce, MH, NN, Pforz, Trin
• H1:1 [indention] Iag. To ſcan ] [no indention] To ſcan
• H1:21 Chamberers ] Chamlerers
• H2v:10 thou ] you
• H2v:27 mine ] thine
• H2v:28 thine ] mine

Sheet I (outer forme)

• Corrected: BM1, Dyce, (Edin in facsim.), NNPM, Petw
• Uncorrected: Bodl, BM2, Cohen (with variant sig. hand-lettered 'I3'), CLU-C, CSmH, CtY-EC, DFo1-2, MB, MH, NN, Pforz, PPRF, Trin
• Sig. I3 signed I4 ] I3 signed I5

Sheet I (inner forme)

• Corrected: Bodl, BM1-2, CLU-C, CSmH, CtY-EC, Dyce, Edin (with I2 and I3v in facsim.), DFo1-2, MH, NN, NNPM, Petw, Pforz
• Uncorrected: Cohen, MB, PPRF, Trin
• I1v:10 the loſſe ] this loſſe
• I2:16 Cypres ] Cipres
• I2:26 (vnhandfome, warrior ] (vnhandfome warrior
• I2:27 vnkindenſſe ] vnkindneſſe
• I3v:5 say (as . . . abroad) ] say, as . . . abroad,
• I3v:8 Conuinced ] Coniured
• I3v:16 Oth. ] Oth,
• I3v:19 that's ] thar's
• I3v:20 He fals downe. ] [S.D. omitted]
• I3v:31 he foames ] he he foames
• I3v:37 thou ] thon
• I4:7 Good sir ] God sir
• 14:20 vnfuting ] vnfitting
• I4:25 the Ieeres,the Iibes ] the geeres,the gibes
• I4:34 cunning ] cunuing

---

180





Sheet K (outer forme)

• Corrected: Bodl, BM1-2, CLU-C, CSmH, CtY-EC, Dyce, (Edin in facsim.), DFo1-2, MH, NN, NNPM, Petw, Pforz
• Uncorrected: Cohen, MB, PPRF, Trin
• K2v:20 Exit. ] [S.D. omitted]

Sheet L (inner forme)

• Corrected: Bodl, BM1-2, CSmH, DFo1-2, MB, NN, NNPM, Petw, PPRF, Trin
• Uncorrected: Cohen, CLU-C, CtY-EC, Dyce, (Edin in facsim.), MH, Pforz
• Pag. L1v numbered 78 ] L1v numbered 84
• L2 numbered 77 ] L2 numbered 85

Sheet N (outer forme)

• Corrected: All copies except Bodl (but Sheet N lacking in BM2 and DFo2; Dyce and Edin in facsim.)
• Uncorrected: Bodl
• N1:8 liquid ] liquit
• N1:19 liue ] loue

Without considering kinds of bibliographical evidence beyond the scope of this study, one cannot determine with any certainty the method of printing employed in the quarto.[5] The consistency with which press-variants appear in only one forme of a sheet suggests, however, that Okes's shop may have followed a systematic order of work such as Professor Bowers has proposed was standard for two-skeleton books during the Elizabethan period. (Sheet I, the one sheet with press-variants in both the inner and outer formes, has in I(o) only a signature miscorrected and a battered 'e' replaced, sorts of correction that a pressman might initiate, according to Moxon.[6]) The quarto, Hinman has established, used two skeletons (pp. 54-59), and the systematic method of printing described by Bowers may well account for the sheets showing uncorrected states in only one forme: the pressman would pull a proof and begin actual printing from the uncorrected type of the forme brought to press; when the pressman was forced to pause in the printing of forme one so that the compositor could correct the type in accordance with the corrector's marked proof, the pressman would pull a proof of uncorrected forme two but take up the edition-run again with corrected forme one; by the time the press-run of that forme was finished, corrected forme two would have been available to perfect all the sheets, forme two thus appearing in only one state while forme one might survive in variant states.[7]

But whatever the method of printing employed by Okes, the quality of the proofreading and correction left much to be desired. Seven of the nine variant formes, all except H(o) and I(i), benefited little from stop-press correction. The change at B3:15 probably resulted from an accident at press, not from proof-correction; and the efforts to correct I(o) and L(i) may or may not have resulted from proofreading. Admittedly, the order of states of B(o) is not certain: the state showing 'How euer' at B3:15—that is, the state represented above as the earlier—has the reading required by the context and therefore might be regarded as a revised state; but as Hinman points out, the large number of copies in both states argues against deliberate revision, and the gummed catchword on B3 in the state reading 'Now euer' suggests its lateness, the variant probably the effect of careless replacement of a type accidentally pulled at press (p. 113). The state of I(o) with the signature 'I4' on I3 is definitely later than that with 'I5', for the corrector also replaced a damaged 'e' in 'newer' at I2v:22, in the process displacing the 'a' in the same line. Confronted by four wrong page numbers on L(i), the corrector revised only two, neither of them correctly. The correct pagination on L1v, L2, L3v, and L4 would have been, respectively, 74, 75, 78, and 79: the page numbers in the unrevised state are 84, 85, 80, and 81 and in the revised state, 78, 77, 80, and 81. In formes B(o), I(o), and L(i), Okes's workmen bungled the corrections attempted, all relatively nonsubstantive errors whose detection would not have required proofreading.

The remaining four of the lightly revised formes indicate proofreading very cursory in nature, the corrections not implying reference to copy. The first example of deliberate correction in the quarto, the change of 'Since' to 'Saunce' on C(i), probably restores the meaning of the copy, inasmuch as 'Saunce' was an acceptable spelling of the French sans required by the context. But neither that change nor the addition of 'Exit.' at K2v:20 is beyond the ability of an average reader. On G(o) the corrector merely righted an inverted 'r' and inserted a space between 'to' and 'me'. His corrections on N(o) were equally obvious, spelling 'liquid' rather than 'liquit' and changing 'loue' to 'liue' (where the sense of the latter word was clearly needed). If the proofreader did consult the manuscript in order to check any of the readings, his failure to correct more than one page per forme and more than one or two errors per page suggests that he was not systematically reading proof against copy, nor even reading proofs completely, since the corrected formes contain relatively numerous errors. C(i), for example, has the non-word 'reſterine' (C1v:20), 'rhem' for 'them' (C3v:12), and very illogical punctuation (allowing for Jacobean standards) in Brabantio's speech at the top of C4. The proofreader's very superficial and selective examination of C(i) and K(o) indicates that he sampled formes rather casually, making sure that there was no very large number of unsightly errors, except when he encountered deficiencies as obvious as those in uncorrected H(o), which caused him to take special pains.

The large number of typographical errors in uncorrected H(o) probably accounts for its receiving more attention than any earlier forme, though its corrections, especially those in the first corrected state, do not imply reference to copy. This state revises only easily detected errors: substituting a 'k' for an 'l' to produce 'keepe', the word apparently required; inverting the upside-down letter in 'aud'; correcting a misspelling; eliminating one comma, and adding two others and an apostrophe. Indeed, as Hinman observes, the proofreader's emending 'Tis' to 'Let' provides strong evidence that he acted without consulting copy during the first stage of correction, for the erroneous 'Tis' probably originated from the compositor's misreading of 'Til', which appears spelled with two 'l's in the Folio (pp. 165-166). But, perhaps because the need to eliminate many obvious errors had made the proofreader more attentive, as well as determined to see that the errors he had marked had in fact been corrected, he read the forme again. The need for his first change, an added speech-prefix at H1:1, could only have been perceived if he had read the previous line, printed on G4v, where the speech begins. Too late to put the speech-prefix in the proper place, he had it set before the speech's second line. Then, besides correcting the obvious error in 'Chamlerers', the reader changed three pronouns on H2v which may or may not have required reference to copy.

Again in I(i) the proofs received considerable attention, corrections being made not merely on one page, but on every page in the forme. Many of the corrections here set right obvious errors very similar in kind to those in the other corrected states: changing 'this' to 'the' (I1v.10), substituting a period for a comma (I3v:16), righting one wrong-font error (I3v:19) and two inverted types (I3v:37 and I4:34), providing a stage-direction implied by the text (I3v:20), and eliminating repetition of a word (I3v:31).

But some of the proof-corrections in I(i) were different in kind from those in the other corrected formes. The proofreader rejected acceptable spellings in favor of optional spellings at I2:16 and I4:25. And, possibly, he marked changes in punctuation involving parentheses in ways misunderstood by the typesetter, whose changes in such punctuation produced nonsense at I2:26 and I3v:5. Most importantly, the proofreader recovered the manuscript readings for two words in separate places whose contexts already made some sort of sense. The two alterations—'Conuinced' from 'Coniured' (I3v:8) and 'vnfuting' from 'vnfitting' (I4:20) — provide the surest instances in the quarto of proofreading from copy, since both errors, especially the latter, would have been difficult to detect through the reading of only the printed proof. Incidentally, the two readings corrected by reference to copy illustrate sorts of errors compositors were likely to make, 'Coniured' representing a misreading and 'vnfitting' a memorial substitution.

Clearly, the press-variants in I(i) reflect tendencies in proofreading not to be found in any of the other variant formes. Such unevenness in Okes's proofreading need not, however, imply a change in proofreaders. A proof-sheet from John Tichborne's A Triple Antidote, printed by Okes in 1609,

shows the marks of a proofreader "rushed or uneven in fulfilling his task, so the workmen who followed his instructions did so imperfectly."[8] Of course, A Triple Antidote was printed thirteen years before Othello. Yet the man who desultorily examined the Othello proofs before forme H(o) may have become so alarmed at the errors discovered in that forme, first at the typographical errors and then, during the reading of the first corrected state, at some of the snarls in meaning, that he grudgingly devoted considerable attention to I(i), and perhaps to the substantially invariant formes H(i) and I(o), which may have been proof-corrected before their press-runs began. In any case, formes after I(i) show no extensive correction, and as Hinman's analysis of uncorrected errors in the quarto indicates, the composition in the late formes of the book allowed fewer obvious errors than in the early formes (p. 298).

Whatever the identity of the man or men who proofread Q1 Othello, they seem to have exercised little control over the text's accuracy in matters of substance. Twenty-two of the book's twenty-four formes show little benefit from stop-press correction, and it is difficult to believe that the Okes shop read proofs antecedent to the corrected formes discussed here. If one wishes to assume that Q1 Othello went through what Professor D. F. McKenzie regards as the traditional minimum of three stages of proof-correction, then the press-variants are to be viewed as merely the last, but only surviving, stage of correction: "once the unique copy [manuscript] has been set and checked, it can be disposed of, once the single galley proof has been read and checked, it can be disposed of, once the revise has been read and checked, it can be disposed of, but once printing has started, the multiple copies are preserved and of course they are available for consultation in those portions of the edition still extant."[9] But it has been pointed out above that several conspicuous errors went uncorrected in C(i), perhaps because they appeared in pages other than C2, where the single correction took place. And, to cite only one more example, outer H after two stages of press-correction still contains two conspicuous errors—'valt [rather than vale] of yeares' (H1:22) and 'Let vs be merry [rather than wary]' (H3:29). That such errors could have survived not only the stage or stages of proofreading visible in the extant copies, but also the prior traditional stages of proofreading, is perhaps improbable.

In Q1 Othello, as in most of the other Okes books examined thus far by bibliographers, the proofreading seems to have been intended mainly to prevent an excessive number of obvious errors, not really to transmit an accurate text. Authorities differ about the standard of proofreading implied by the elaborate press-corrections in the Pied Bull Lear (1608), Greg having

pictured Okes as a printer who did not care as much about the accuracy of the text he was printing as about "the economical working of his business."[10] Professor Bowers, on the other hand, has seen in Lear indications "that Okes was doing his conscientious best to produce the best text he could from his miserable manuscript."[11] On the basis of the proof-sheet from A Triple Antidote, whose superficial and uneven correction agrees closely with most of the corrections in Othello, Professor Brown believes that Lear received much care, but that careful proofreading was atypical of Okes's printing-house.[12] The evidence provided by the press-variants in Othello is consistent with Professor Brown's assessment of Okes's methods, and with the conclusions of Professor Robert K. Turner, Jr., who has written several recent studies of books printed by Okes. Of Thierry and Theodoret (1621), Turner writes: "It was proofed primarily for mechanical rather than literary errors";[13] and of The Maid's Tragedy (1619): "The total effect of the proofreading on the text of Q1 seems to have been very slight; the compositors were almost entirely responsible for the text as it now stands."[14]

Since proofreading in Okes's shop evidently aimed at the detection of conspicuous typographical errors, leaving the responsibility for the text of Othello largely to the compositor or compositors, they may well have introduced a number of readings like the two in I(i) detected by the proof-reader when referring to copy, or more serious errors which make sufficient sense in context to prevent detection by the sort of proofreading employed in the greater part of the book. That even the reading of H(o) and I(i) was done with consistent care and more than sporadic reference to copy is certainly open to question. Without more knowledge about Okes's compositors, then, it is unwise to accept the corrections in I(i) as proof "that most of Q's errors were derived from Walkley's manuscript"[15] and much less wise to accept substantive variants between the Q1 and Folio texts as Shakespeare's first and second thoughts.

Notes