III
The second issue that must be settled is the printing sequence of the four
sections of Complaints. Muiopotmos, the fourth
section (T-Z), bears an imprint date of 1590, while the earlier three
sections are all dated 1591. Various theories have been offered to explain
this apparent anomaly. It is worth noting that, given the fact that Complaints was entered on 29 December 1590, it is not
unlikely that a compositor setting up a title-page in 1590 would date it
thus. But Johnson rejects the view that Muiopotmos
was printed first, concluding that “the probable explanation of
the date of Muiopotmos is that when the printer
reached that point, he realized that the volume would be completed and ready
for sale in 1590 (Old Style) instead of 1591, and changed the date on the
title-page accordingly.”[26] Since
Johnson's arguments have not been previously refuted, some discussion is
necessary.
First, Johnson correctly assumes that the continuous, unbroken sequence of
signatures from A to Z demonstrates that the four sections were considered
parts of a single volume from the outset. The beginning of a new alphabet
for Daphnaida, conversely, probably indicates that it
was not intended as part of the volume, although the fact that Complaints ends on Z creates some uncertainty:
frequently enough, compositors begin a second alphabet in a single volume
without doubling the signature (i.e., Aa, Bb, etc.). The fact that only
three copies of Daphnaida are extant, one bound
together with Complaints, is a clear enough
indication that it was not considered part of the volume. However, the
assumption that the continuous signing demonstrates that the Complaints was printed seriatim (i.e., sheet A, then
B, then C,
and so on to Z) overlooks the realities of
textual processing in the Elizabethan printing house, particularly in regard
to the different challenge presented in preparing for the press a prose
manuscript as opposed to a verse manuscript. As Joseph Moxon explains at
great length, the prose manuscript can present considerable difficulty in
the process of casting-off, especially if the hand is uneven. Nonetheless,
in describing three methods of casting-off copy, he assures the reader that
a compositor is able to perform this complicated process with sufficient
accuracy to estimate that 127 pages of manuscript copy containing 191,135
pieces of type (including both letters and spaces) would transfer into 123
quarto pages set in english (94mm) roman type, or fifteen sheets of paper
with three type-pages remaining for the sixteenth sheet.
[27] In contrast, a novice could cast-off a
verse manuscript provided that he kept in mind that blank lines between
stanzas had to be included in the count. The task of predicting the
signatures of the three internal title-pages was rendered easier because of
the editorial decision to place each on the recto of the first leaf of their
respective sheets (or “gatherings”), leaving the verso
blank. Similarly, each of the other texts begins on the recto side of a
leaf. An experienced publisher's savvy is apparent in the grouping of poems
which produced this attractive layout and yielded but one
“wasted” type-page (X2
v was left
blank so that
Visions of the worlds vanitie could
commence on X3
r). In short, assuming that Ponsonby had
not cast-off the manuscripts himself in arranging them into their final
order (as he probably did), a compositor could have accomplished the
necessary analysis in about an hour to determine that the title of
Muiopotmos would occur on signature T1. As noted
earlier, the placement of the internal title-pages on the recto of the first
leaf of their respective gatherings probably resulted from a marketing
strategy rather than purely aesthetic concerns; that is, the three internal
sections could be marketed separately.
Moreover, Johnson does not consider the implications of Orwin's ownership and
use of two presses. In a one-press shop, “seriatim” can
only mean two or more sheets in sequence, one after the other, usually
accompanied by a distribution of the type from one or both formes of the
previous sheet in order to replenish the cases for the setting of the first
forme of the next sheet. However, in a two-press shop using the concurrent
production method, work is shared between the two crews and presses. Hence,
“seriatim” does not necessarily apply in its strict
denotative sense. For example, Orwin could have assigned sheet A to press 1
to begin the job, but because the manuscript had been cast off, he was free
to assign any other sheet to press 2. There was no compelling necessity to
assign sheet B to press 2, nor did press 1 have to begin the job with sheet
A.[28]
Second, Johnson's assumption “on a priori grounds” that
“the book was printed in... logical order” undergirds
his interpretation of “the typographical make-up of the four
title-pages” as the “key to the final solution of the
problem.” He cites the movement of the “four
hands” ornament and “London” upward in three
steps from the crowded #1 title-page setting to the position in title-page
#4 in arguing that, had the sections been printed in reverse order (i.e.,
#4, #3, #2, #1), “it seems most unlikely that the printer, having
achieved a satisfactory spacing of the matter... in #4 and #3... would have
quite unnecessarily disturbed this arrangement in #2.” Obviously,
he overlooks the logical implication of his observation that the large
amount of matter in #1 forced the crowding in the lower part of the
compartment: this crowding was necessary regardless of whether #1 was
printed first or last. In any event, the main problem is Johnson's failure
both to visualize these spacing variations from a compositor's perspective
and to appreciate the process of re-imposition. Such adjustments were
unavoidable given the sequence of resettings and re-impositions. All of the
type, including the blanks, in the upper portion of the compartment had to
be removed and reset in the stick with the new texts, and then re-justified
both horizontally and vertically each time it was imposed in the compartment
formed by the vertical and horizontal rules abutting the four border
pieces.[29]
Consequently, the notion that a compositor would actually be concerned about
preserving a “satisfactory spacing” ignores the simple
reality that a given spacing could not be preserved in these circumstances
even if the text reflected only a quite minimal change. Indeed, the
compositor's primary concern was simply to pack the compartment space with
types and blanks which, when locked up in the furniture, would firmly remain
in place despite the downward thrust of the platen during the many hundreds
of pulls on the bar. In short, the perception of a sequentially evolving
layout is illusory.[30]
Moreover, taken as physical evidence of changes in a typographical setting,
spacing variations simply cannot prove sequence even when the changes occur
during a single imposition. The direction of change is always ambiguous in
itself: one state of the setting can be declared as the earlier setting (or,
vice versa, the later setting) only if other evidence that is capable of
demonstrating direction is present.[31] In
reality, sequence can be proven conclusively only by progressive damage to
an identifiable typographical entity such as an ornament, initial, border,
rule, or type which reappears in a sequence of states of an impression or in
a sequence of impositions. In this case, the imprint provides such
typographical evidence. The first “r” in the pica roman
setting of the line “Churchyard at the signe of”
exhibits a few characteristic dents which can be observed at high
magnification (30X) in all title-pages except #4 on T1 where it is nearly
perfect. However, the “r” exhibits new damage which is
observable even without magnification in title-page #3 on L1. Just below the
junction with the shoulder the vertical member has been severely dented by a
blow to the left side, producing a pronounced arc to the right of the
vertical member. Similarly, the “a” in
“at” is undamaged on T1, but the shoulder is battered
downward toward the body on E1 and L1. In short, the title-page #3 (L1) was
the last title-page to be printed. The order in which the title-pages #1
(A1) and #2 (E1) were printed cannot be determined conclusively, but the
condition of the types on T1 indicates that it was first. Admittedly, this
seems a quite strange method of printing a book. Nonetheless, when a group
of key types is examined across six copies and consistently noted as
“perfect” or “good” (i.e., no
discernible
damage), and these occur in the context of
other identifiable types which indicate “same setting of
types,” the evidence contradicts the logical expectation of what
is normal. In conclusion, since
Muiopotmos was not
printed in the logical order suggested by the sequence of signatures, there
seems little reason to reject the correspondence of the entry date and the
“1590” imprint date for
Muiopotmos as the external basis for confirming that it was
printed first, with its title-page being set in 1590.
[32]
