University of Virginia Library

Conclusion

The coincidence of unusual clustering of stop-press corrections on one forme with a break in the pattern of headline reuse on that forme suggests a mechanical

explanation for the extensive adjustment of N(outer). That the changes on this forme were relatively undermotivated and difficult to make (compared to those elsewhere in the book) gives further reason to suppose a mechanical explanation rather than to assume that the printers were unusually fussy at this point. The largest single alteration, involving resetting four lines of type on N3r, occurred where an oversized headline (likely to make the type underneath it loose) was borrowed, quite possibly for the first time in the job, from the other compositor. However, the alteration was extensive only because a whole word had to be inserted into a block of prose (be hangers > be might hangers), necessitating the movement of type in four successive lines. This change was undermotivated since might was not essential to the line's meaning: the optative mood is already established by the phrasing I would it be hangers. The adjustment was in any case bungled, since be hangers > might be hangers was clearly intended. This bungling might encourage scepticism about the accepted order of correction (Wilson's order), making us wonder whether in fact the obviously incorrect reading be might hangers was the initial setting and during the run it was corrected to be hangers. The displaced letters in dazzie and responsiue, however, establish unequivocally the order of correction for the entire forme: Wilson was right. Figure 4 represents the forme of type as the compositor would see it, showing how easy it would be to insert a word in the wrong place because the type is a mirror image of the inked impression it makes.

If, as I propose, the difference in widths in the two compositors' headlines caused this forme's peculiarities, the effects were not uniform: N3r and N4v were set beneath overlong headlines but only the former underwent stop-press correction, and N2v was set with a correctly-sized headline yet it also underwent correction. However, it would take only one significant accident in one part of the forme for the entire forme to receive attention, and if a large number of (not necessarily adjacent) sorts were progressively loosened during machining until they were plucked from the forme while it was being inked, then the compositor would probably be asked to check copy before reinserting them, since the potential for mistakes is great when multiple sorts are to be reinserted at once. As an explanation, such an accident would economically combine the known facts regarding the first round of stop-press correction: the compositors set to different measures, they swapped headlines when this forme was set, the forme contains an unusually large number of the book's variants—38% of the variants falling within just 4% of the book— and the alterations were relatively undermotivated (when judged purely on poetical grounds) yet difficult to carry out. Such an accident would preclude removing the forme to the imposing stone, and repair on the bed of the press (presumably under pressure of time) is consistent with the misalignment of the letters zz in dazzie in the first round of corrections. (The misalignment of the first s in responsiue was part of a subsequent round of stop-press correction—reponsiue > responsiue and sellingly > fellingly—that required so little disturbance of type that it could easily be achieved without removing the forme from the press.) If there was such an accident then the readings of the corrected state resulting from the first round of changes, witnessed uniquely in the British Library exemplar, have authority arising from the compositor's likely consultation of copy when recovering from it.