44Half-Sheet Imposition

Another possible explanation for the existence of BL1 half-sheet H is half-sheet imposition, which would require that the two variant states of the text were set and printed simultaneously. Gaskell identifies two methods of printing half-sheets:

In one, called half-sheet imposition, all the pages for a half sheet were imposed in one forme ... this forme was first printed on one side of the whole sheet, then the heap of paper was turned (end over end in quarto and octavo...) and printed from the same forme on the other side. Each printed sheet was then slit in half to yield two copies of the same half sheet. In the other method, the pages for two successive half sheets were imposed in two formes and printed in the normal way; again the printed sheets were cut in half, but this time each one yielded copies of the two successive half sheets which were different from each other, but which were sometimes indistinguishable from similar half sheets printed by half-sheet imposition.^[177]

45If half-sheet imposition was employed to set half-sheet H, it cannot have been by the first method outlined by Gaskell, as ‘work and turnʼ produces two identical copies of the text for each sheet of paper printed. The two states of half-sheet H for An Humorous Dayʼs Mirth are anything but identical. K. Povey identifies a way of printing two copies of a half-sheet by half-sheet imposition from one sheet of printed paper.^[178] This involves setting each page twice and imposing it in two formes. The paper is printed with the first forme and perfected with the second forme. Each sheet is cut in two, yielding two copies of the half-sheet for each sheet of paper printed. Using this method two very different states of the text can be produced simultaneously, without the need for Gregʼs accidental destruction theory. However, this new suggestion requires examination.

46The wider setting of the text and running titles both suggest the possibility of half-sheet imposition. Also striking is the ornament on H2 before the colophon: in the BL1 copy, seven and a half repetitions of the pattern are used, while in the corrected state (e.g., BL2), only six and a half repetitions of the same pattern occur, with an additional half-repetition of a different pattern. It would seem unprofessional for a compositor to use an incomplete, mismatched ornament on one state when a complete set of ornamentation was obviously available in the printing house, as the BL1 state indicates, unless the two states were printed at the same time. In this case the compositor might be stretched for ornamental type, if setting two copies of the same text simultaneously. Furthermore, the ornament does not appear in A Warning for Fair Women, also printed in 1599 in Simmesʼs shop, and signifying that the ornament was not in use in the only other extant play printed by Simmes that year.

47Also puzzling is use of a wider measure and different font. This irregularity might indicate the work of another compositor, one who had not previously worked on the text of the play, and who might have his composing stick set to a different measure. There are no stage directions in sheet H, so no comparison of the treatment of stage directions in the two states could be made. Spelling preferences can sometimes identify the work of different compositors, so a comparison of BL1 and BL2 copies was made. Words spelt differently in the two states of half-sheet H were then traced through sheets A-G to accumulate the spelling preferences of the compositor who set the rest of the play. Of course, the quantity of text set on half-sheet H is very small and the tests conducted were inconclusive because only one example of differentiation existed in either state of the text. Also, when a compositor has a preference for one spelling, he invariably also uses the non-preferred spelling at some point in his setting. The following results were found:

48Very little can be established from this analysis due to the small sample of spellings in half-sheet H; however there are some signs that where the rest of the text established a strong preference, BL2 follows it, but the other copies (represented by BL1) do not. This suggests, if anything, that the same compositor set the rest of the play plus BL1ʼs ‘resetʼ half-sheet H and a different compositor set the half-sheet H found in the other copies. However, the test does not take into account the influence of justification on spelling preferences. Wishing to avoid the pitfalls of scant evidence, the investigation is deemed inconclusive.