 | |  |
Notes
See, for example, Charlton Hinman, The Printing and Proof-Reading of the First Folio of Shakespeare (1963), I, 180-225, 369-424; Philip Williams, "New Approaches to Textual Problems in Shakespeare," SB, 8 (1956), 3-14; Fredson Bowers, Bibliography and Textual Criticism (1964), pp. 180-201.
See, for instance, Alice Walker's "introductory" remarks in Textual Problems of the First Folio (1953), esp. pp. 8-11, and Hinman, II, 511-512.
Discussion of the relatively minor differences in F1's reprinting of these plays may be found in Reid, "Spellings of Jaggard's Compositor B in Certain Plays in the First Folio of Shakespeare," Diss. Virginia 1972, pp. 2, 366-379.
The printing of the pages studied here, including the question of copy, is discussed in Hinman, II, passim, and Reid, "Spellings," pp. 366-379. One dissenting voice recently raised is that of Andrew S. Cairncross, in "Compositors E and F of the Shakespeare First Folio," PBSA, 66 (1972), 369-406. Cairncross believes that many of the pages attributed to B were set by E. Apart from its inability to explain the intercalary formes associated with E, Cairncross' argument uses spellings which for various reasons often seem to have little evidential value. The discussion immediately below touches on some of the problems—the influence of rhyme and justification and the matter of compositorial uniformity. Another is the question of what kinds of spellings constitute good evidence for identifying a workman. The assumption adopted here is that not tolerances of copy spellings, but preferences—primarily defined by actual changes of various copy spellings (when known) to the preferred form, and only secondarily by habitual reproduction of that form—are the basic spelling evidence for identification of compositors.
More on these matters may be found in Reid, "Justification and Spelling in Jaggard's Compositor B," SB, 27 (1974), 91-111; "Compositorial Spelling and Literal Rhyme: The Example of Jaggard's B," The Library, 5th ser., 30 (1975), 126-133; "Spellings," pp. 172-184.
T. H. Howard-Hill mentions these two issues specifically in "Spelling and the Bibliographer," The Library, 5th Ser., 18 (1963), 10-11. For the sake of convenience and clarity, I use the term "constancy" to refer to the stability of a compositor's spelling practice over a period of time, and reserve the term "consistency" for the uniformity of his treatment of analogous forms. "Consistency" has been used, not illogically, to cover both meanings by Howard-Hill and others, but a convenient distinction between the temporal and analogical senses seems useful.
The Pavier Quartos and the First Folio of Shakespeare, Shakespeare Studies Monograph Series, No. 2 (1970), ch. 2.
See, e.g., D. F. McKenzie, "Compositor B's Role in The Merchant of Venice Q2 (1619)," SB, 12 (1959), 75-90, and Kable, esp. pp. 8-14; also Williams, "New Approaches," p. 12.
Peter W. M. Blayney, "'Compositor B' and the Pavier Quartos: Problems of Identification and Their Implications," The Library, 5th Ser., 27 (1972), 179-206; John F. Andrews, "The Pavier Quartos of 1619: Evidence for Two Compositors," Diss. Vanderbilt 1971.
More detailed examination may be found in Reid, "Spellings," ch. 2. The matter of -ie/-y touched on below is discussed at length on pp. 79-95.
Thus Andrews' use of mechanical evidence seems, at least at first sight, to be more successful than his use of spellings; very often the differences in spelling habits identified in Q2 Lr. fail to materialize consistently in the other Pavier Quartos. Some spellings may be useful as evidence supporting a well-established case for two compositors, but it seems unlikely that many will provide positive evidence that cannot be neutralized by conflicting spelling evidence. Were two compositors finally proven, then spellings would probably have to be relied on to identify these with the compositors of F1, since the different formats of the books of 1619 and 1623 would make some mechanical evidence inapplicable to both.
Thus it seems we must reject Andrews' conclusion that "Kable's generalizations seem more likely to be applicable to Compositor G [his current name for the second compositor, changed from his original "F" to avoid confusion with the F hitherto identified with A in F1] than to Compositor B" (p. 342). The spelling practices of B in F1 are like those of Andrews' "G" as much as they are like those of his "B," and B's spellings are more often like the spellings of both, combined, than they are like those of either. Hence, unless we are to assume that between 1619 and 1621 B altered his practices in such a way that they came to incorporate those habits of "G's" that previously distinguished his work from B's, we must conclude that generally B's and "G's" spelling practices are essentially identical and that Kable's statistics therefore give us a fair picture of most of B's habits, though they are sometimes inaccurate, may include some words not actually set by B, and may even misrepresent some of B's spellings of a few words.
Throughout I use single quotes to distinguish words as substantives (given in modern American spelling) from their actual spelled forms in the texts, which are italicized here. This and several other related matters are discussed in my "Spellings," pp. 27-35.
This problem receives more detailed treatment in Reid, "Spellings," pp. 184-191. Allowing for a certain proportion of insignificant spellings in a compositor's pages also helps circumvent the problems raised by the recent debate over whether lost rounds of proof-correction occurred in printing Renaissance books. When we study only the earlier forms of surviving press-variants and yet allow for unexplained anomalies, we automatically treat as insignificant any presumably uncorrected spellings in a compositor's pages that may be atypical because they in fact represent a proof-reader's unidentified corrections, and thus we virtually eliminate the possibility that such spellings can adversely affect conclusions. The question of B's consistency which is touched upon below receives more thorough discussion in Reid, "Spellings," pp. 130-133.
Discussion of some other words may be found in ch. 4 of Reid, "Spellings"; the complete Folio statistics on each word dealt with here are in Appendix I. I include here only those spellings that examination has shown are legitimate, or "pertinent," evidence of B's normal habits. Terms such as "the Folio evidence," "the Folio spellings," and "the Folio statistics" refer, of course, to the spellings in the Folio pages under consideration here (see fn. 4 and its text).
"Normal," or unexceptional, conditions means circumstances in a text under study which (as preceding examination has shown) did not cause B to adopt certain spelling practices that were exceptions to his normal habits. For instance, at the end of a crowded long line the spelling hart would not be legitimate evidence that hart was in B's copy, since we know that B often justified lines by using such short spellings, regardless of the spelling of his copy; likewise, hart in a rhyme with part would not be pertinent evidence, since B sometimes set this spelling in such a rhyme even when heart was in his copy. The following discussions assume reference to normal conditions as a general rule and assume that allowances must be made for the fact that between 1% and 5% of all spellings are, theoretically, insignificant evidence due to chance (i.e., mere random hits).
Kable, pp. 49-55, 59-62; E. E. Willoughby, The Printing of the First Folio of Shakespeare (1932), pp. 55-59; Walker, esp. pp. 8-9; Hinman, I, 182-183, 369-424; Cauthen, "Compositor Determination in the First Folio King Lear," SB, 5 (1952-53), 73-80, esp. 78; Andrews, pp. 345-389; Blayney, passim. Throughout I assume Andrews' assignments of much of Pq to his compositor "G" and Blayney's to his "G" and "H" are unproven.
For a more thorough discussion of the basic premises behind this approach, see the Introduction to Reid, "Spellings," esp. pp. 7-10, 22.
That is, Kable, Andrews, Blayney, and Reid, "Spellings." In a few cases the evidence cited for Pq supplements that given by Kable, or in rare cases by Andrews or Blayney. This supplementary evidence is taken only from Pq plays now considered Shakespeare's and is, therefore, not complete; these relatively few spellings were found with the aid of a Shakespeare concordance rather than through new, independent collations of all the plays, and hence the spellings in the Pq reprints of A Yorkshire Tragedy and Sir John Old-castle are not represented here. The statistics for these few cases appear in Appendix II of my "Spellings."
Kable's statistics come mostly from his "Appendix," but those for 'do', 'go', and 'here' are on pp. 12-13. For convenience I follow his listing of the words and his habit of considering homographs (e.g., 'art' sb. and vb.) in one discussion, if they appear to be spelled identically by B. However, in listing the evidence from F1, I identify the words by part of speech in round brackets. See Reid, "Spellings," pp. 32-33, for more on this point.
The fact that the number of occurrences in Pq and F1 are about equal suggests that Kable's evidence may be incomplete, but it does not necessarily imply that conclusions drawn from the full statistics would have to be different. Indeed, the evidence from F1 for B retaining art can stand on its own.
Andrews' figures on this word and 'bloody' are combined, but comparison of his totals with Kable's shows that he finds fewer instances of -ou- retained than does Kable (31 versus 40). Thus the apparent discrepancy between Folio B's rate of retaining -ou- and that in Pq may be very small. (Andrews argues that "B" never retains -ou- whereas G frequently does, but the distinction could be merely hypothetical.)
Andrews, pp. 281-289; Blayney, p. 188. The situation in Pq is now quite confused. Blayney's new evidence leads him to conclude that Kable has reversed his spellings from deere > deare to deare > deere, but I think this fanciful reconstruction must be rejected, as Andrews' figures show. It seems more likely that Kable's statistics are incomplete, as they sometimes are for other words; besides, Blayney's theory that deare is preferred in Pq would not explain the changes of -ea- to -ee- in forms (e.g., 'dearer') other than the simple one. If Andrews' figures are correct, there are three possible explanations for the Pavier evidence. (1) Andrews' explanation, that Compositor G prefers the -ea- spellings while "B" prefers -ee-; this explanation has one drawback—"B" gratuitously introduces an -ea-. (2) That "B" in 1619 had a preference for -ee- but temporarily varied from it (cf. his handling of -ie/-y); the fact that practically all the -ea- &c.nt;hanges occur in Lr. and in MV, plays apparently set at approximately the same time, would support this view. (3) That in 1619 B had not settled on a preference for either deare or deere; this view finds support in the fact that by 1621 he still had not developed a generic preference for either -ea- or -ee- but preferred both neere and yeare and was indifferent to the spelling in some other analogous words. The chance that the second explanation is correct makes impossible final conclusions about the value of -ea- as a clue to B's Folio copy, given the slimness of the Folio evidence. On an additional dearer reproduced in F1 (1H4, 3073), see Reid, "Spellings," p. 30, fn. 2.
Both Andrews (p. 355) and Blayney (pp. 185-186) produce statistics for this word that differ from Kable's, but they do not agree with each other any more than they do with Kable. The figures cited here are Andrews'.
One direction (MND, 2065) has Lion, but we should discount this because it is not in Q2 and, having been supplied by an annotator or by B himself, was set under peculiar and unknown influences. Otherwise, B uses Lyon invariably in the prefixes and the direction that was in Q2 (MND, 2021).
Andrews' figures on this word (p. 360) show more instances of the same spelling practice, but whether he includes only the spellings of 'lose' and not those of 'loose' is not clear.
Andrews (p. 364) provides more complete information on the hyphenation of this word in Pq, but he ignores the other elements of form discussed here. See n. 22.
Andrews (p. 378) contests this view, arguing that "B" prefers -th- and G -d-. The Folio evidence is unfortunately so meagre that it alone cannot settle the question. However, if Andrews' view is right, it forces the conclusion that curiously enough both "B" and G are very tolerant of the spellings they each do not prefer.
| | |