University of Virginia Library

Summaries of Evidence

I do not propose to discuss here the pages for which Hinman's identifications are acceptable. The evidence used is set out summarily


Page 76
in tables for each text printed as appendices; these include concise figures for spacing. When 'do/go/here' evidence is ambiguous, and even when it is plentiful, it is best to use all the discriminatory spellings for a particular text that can be found. In practice, however, 'do/go/here/indeed/mistress' were the only discriminants which occurred frequently enough throughout the comedies to be useful. After the comedies had been examined for the presence of these spellings, and other potentially discriminating spellings such as Miss Walker's A and B spellings, I surveyed the distribution of a miscellaneous selection of other spellings. These were 'deed' (since 'indeed' was characteristic of C, 'indeede' of the other compositors), 'doe't/ doo't/do't', and the more common elisions, 'a'th'/by'th'/i'th'/o'th'/ to'th' and 'hee'l/shee'l/they'l/you'l' and their variant spellings. The results of this investigation, which revealed different compositorial practices confirming the attributions already made from the main evidence, is set out in a table in the appendices.[31] This evidence is incorporated in the summaries of compositorial practices in the individual comedies which are also printed as appendices.[32]

In Tmp. the spacing confirms Hinman's attributions except for B4 which is more likely to have been set by C than A: the presence of 'heere' gives strong support to this identification.[33] On B3v the spacing is not in accord with A's usual practice but the three 'here' spellings, and the spelling 'indeede' which C rarely uses, do not allow this page to be assigned to C. For TGV I have had the help of a table compiled by Miss Walker for the Oxford Old-Spelling Shakespeare. Her spelling evidence and the spacing confirm Hinman's division of this text between compositors A and C. He was doubtful about assigning B6 and B6v to C but the spacing shows him to have been correct.

In Wiv., where again I have benefited from an analysis of the composition by Miss Walker, the spellings of all the B pages save E1 and E6 are supported by the spacing evidence. These two pages contain many long lines which have probably disturbed B's customary preference for spaced commas inside lines.[34] Page E6, over which Hinman was in doubt, can certainly be assigned to B for it shows seven instances of 'Mr' without a point, B's almost invariable spelling. B's pages apart, spellings do not always distinguish the contributions of A and C, the other compositors of Wiv., as there is much prose and C's distinctive


Page 77
'heere' is liable to be affected by the need to justify. He was in any case apt to accept 'here' from his copy. The spellings suggest that A set D3, D3v, E3 and E5 although the only 'here' of E3 is in a long line. However, the spacing evidence is consistent with these identifications. The spacing evidence, particularly the spaced terminal commas, similarly confirms the attributions from spelling evidence of D2, D2v (which Hinman had assigned to A), and D4-6v to compositor C.

The compositors of MM and Err. are much more difficult to distinguish and I must devote more space to discussion of their problems. According to Hinman, compositor D is encountered for the first time in MM where supporting evidence from the order of printing is almost totally lacking. Hinman found that compositor C set F2v where alone there is a clear preference for the speech-prefix 'Isa.', with doe and goe, compositor B set five pages, F5v, F6, G2v, G3 and G3v, and the remaining pages of quires F and G were set by A and D, their stints being separated by D's tolerance of 'do' and 'go' when it occurred in copy. Hinman commented:

we cannot here adduce the combined evidence of types and spellings. In several quires in this section of the Folio (and in some, alas, that raise very tricky identification problems), such attributions to particular compositors as we make will for the most part have to be based upon spelling peculiarities alone.[35]
In these quires there are many 'do/go' spellings, and 'here' occurs so frequently throughout that it is very likely to have occurred frequently in the copy for MM (and, indeed, in the texts which precede it in the Folio). In order to assign pages to D, Hinman assumed that 'do' and 'go' were also present in the copy, which is probable enough, and he drew upon the evidence of D's relative tolerance of these spellings which he had observed when establishing the presence of a fourth compositor in the Folio.[36] However, in order to give F1v, F6v, G1v, G4 and G4v to compositor D, he had to allow him eleven 'heere' spellings as against only four 'here's. 'Here' is the spelling which distinguishes D from compositor C, so on his own showing, Hinman's compositor identifications are hardly satisfactory. They would have been more persuasive had D been setting from his case z, but that case, with the assumption that a new compositor had joined in work on the Folio, is not seen until quire K where also Hinman identified compositor D.

As the customary spellings are inadequate, it is necessary to draw upon a wider range of evidence. Other compositorial practices


Page 78
described earlier offer helpful support in some pages, but no evidence is more generally useful than that of the spacing of terminal and medial commas. The spacing evidence of quires F and G by columns runs as follows:

Spacing in Quires F and G, by Columns[*]

Col. a  Col. b 
Page  Hinman's comp.  Hill's comp.  Term. #,  Med. #,  Non sp.  Hill's comp.  Term. #,  Med. #,  Non sp. 
F1  A?  12 
1v   7*  5* 
2:1  21  25 
2v   11 
10*  7* 
3v   11  3*  13  1* 
14  25 
4v   12  17 
16  24 
5v   14  21 
22  3*  26 
6v   1j  3*  1j  2* 
G1  14  2*  5* 
1v   D?  10  1*  10  1* 
15*  6* 
2v   16  4*  13  0* 
1*  8* 
3v   19  16  1* 
D?  A(F)  20 
4v   D?  A(F)  27  19 
29  1:1  20 
5v   A(F)  A(F)  13  13 
A(F)  17  21 
6v   A? 
Patently, many of the pages which Hinman assigns to D have contrary spacing practices from column to column. Also, the spacing of medial commas on Hinman's D pages is not consistent with compositor D's practice in the case z pages of quire K where he does not prefer to use spaces after commas in short lines.[37] Further, the absence of spaced terminal commas from C's page F2v and the number of them elsewhere suggests that Hinman's attributions need closer examination. The table also shows that compositor B's strong inclination towards the use of spaces after commas is confirmed by the pages which Hinman


Page 79
assigned to him and that there are other columns which might also, if the spellings were agreeable, have been set by this compositor.

Compositor B's spellings are so distinctive and frequent that his contribution to MM may be dealt with summarily.

Hinman noted that the spellings of G1v resembled compositor B's but assigned the page to D against the evidence of 'heere' because 'there is no other evidence to suggest that B had any share in the setting of the last three formes of quire G. . .'.[38] The printing of quire F was so irregular that an appeal to normality carries little weight, but as the table shows, B is found in the last three formes of quire G, in column a of G1. There are no contrary spellings, the spacing agrees with what is known of B's practice, the spellings are incompatible with the habits of the other compositors, including D, and as can be seen from the earlier table of spacing in MM, the two columns of G1 contrast in spelling and spacing evidence. G1 was therefore shared by B and another compositor.

Other pages show different spacing habits between columns: this suggests but of course does not prove that two compositors shared in the composition. There are other pages and columns in which the incidence of spaced terminal commas points to compositor C. Hence F4 and F4v are more likely to have been set by C than any other compositor (certainly not A or D who did not favour internal spaces), and so too are F5a, G4a, G6vb where the distribution of spaced terminal and internal commas contrasts with the practice of the other column of the page. The spacing evidence is most useful to distinguish C, who, as Tmp., TGV and Wiv. show, will accept 'do', 'go' and 'here'


Page 80
occasionally from copy, but it is less useful to distinguish A and D, particularly when many of the columns have large numbers of long lines. Their stints have been assigned, as far as other spellings permit, on the principle of minimising the number of 'do/go' spellings allowed to A. The evidence I have used and my final allocation of columns to A, C and D is set out in a table facing this page. It may be that C's partner in the shared pages of quire F was sometimes A rather than D to whom, where the spellings are not conclusive, I have assigned columns so that quire F was shared by two compositors other than B.

The picture that this division of work presents of activities in Jaggard's printing-house is most unusual.[39] It is scarcely to be conceived that the Folio would ever have been published had this been the ordinary method of working. Yet the absence of corroboration from type recurrence and distribution which Hinman noticed is perhaps more readily explained if the compositors shared many pages of these quires and distributed whole pages, rather than the portions each had set, into the type-cases which were free at the moment. In any event, Hinman's account of the printing shows that when the printing of Wilson's Christian Dictionary came to an end and compositor D was free to work more often on the Folio, normal routine was upset. The irregular distribution of these quires arose, I suggest, not because compositor D's practice was irregular (as Hinman thought) but because the division of setting amongst four compositors working from two cases led to unusual practices.

A's 'indeede' is found on G4va (2439) and G5v (2718j), 'she'll' on G5v (2654j), 'we'll' on G5v (2743, 2690j), and there is an indented verse turn-over on G5va. Characteristic C spellings are 'they'll' on F2 (236j), 'you'll' on F3v (685j) and G6v (2935), 'deed' on F4 (848) and G6 (2853) but 'deede' on G4a (2291), 'doe't' twice on F4 (799, 802j), 'shee'll' on G2b (1841), and 'wee'll' on G5 (2638j) and G6v (2937). An indented turn-over occurs on D's F1b, 'indeed' on F1va (131j), 'indeede' on F3 (558, 561, 550j, 581j) and F2b (331j), 'youle' on F3 (589), and 'doo't' on F5b (1071, 1074). A is characterised by his preference for italicising 'Duke/s', 'Prouost', and 'Frier' and his intolerance of 'do', 'go' and 'heere'. He seems to have preferred 'Duk.' to 'Duke.' as the speech-prefix. Compositor C on the other hand did not generally italicise these words although it is consistent with his habits elsewhere in the Folio that occasionally he should do so. He was more prone than A to accept 'do' from his copy, though not as often as D, and his


Page 81
readiness to take 'here' from copy has been noted in the preceding comedies. He inclined to the speech-prefix 'Duke.' Compositor D did not tolerate 'heere' and did not italicise 'Duke/s', but apart from his willingness to accept 'do' and 'go' from his copy, his habits are similar to A's.

These were the general grounds for the compositor identifications from spellings in quires F and G. There is not much useful evidence on F1 and F1v. Compositor C clearly set F1vb. Page F2 is divided on spacing evidence between C and D, on account of the 'do' spellings and 'Frier.' Page F2v contains the 17 'Isa.' speech-prefixes which led Hinman to assign this page to a compositor (C) who otherwise, since other pages show mixed forms of the speech-prefix, did not participate in the setting of MM. In the other sections of the text which I give to C he preferred 'Isab.' but the unitalicised 'Duke' and 'Prouost', together with the spacing, obliges me to conclude that in this first page in which he encountered 'Isa.' compositor C followed copy. The spacing of F3 is inconclusive, but the 'do/go' spellings suggest D rather than A. Spacing and Hinman's conclusion that the pages of the first forme were set by the same compositor makes it most likely that F3v was set by C, despite such A spellings as 'ile' and 'here'. In F4, F4v



Page 82
and F5a the large number of spaced terminal commas makes C the compositor, but D seems to have set F5b where there are two 'do' spellings and a great number of medial commas without spaces. The 'heere' spellings and 'Frier' (1501) show that F6vb is C's as also the other column may be, but the 'do' spellings there encourage me to maintain the division of quire F between C and D. Compositor B's partner in G1b is assuredly D as 'do', 'Duke/s' and 'Frier' show, and the spacing and 'do' of G2a show that that column was also set by D. His partner may well have been A since the spacing is inconclusive, but the unitalicised 'Friar' (1566) and the preponderance of 'Duke.' speech-prefixes indicate C, to whom, with the support of his characteristic 'shee'll' (1841), G2b may also be given. Spacing and spelling evidence agree to show C as the compositor of column a of G4 which was evidently shared by two compositors. Compositor A is more likely, on the showing of four 'Duke's and two 'Frier's, than D to have been C's partner, and similar evidence divides G4v between the same compositors. Spacing and spellings alike show C to have set the whole of G5 and A to have set G5v. Contrasting spellings divide G6 between A and C, but only the elisions show C to have set the short page G6v.

Discriminating spellings are scanty in Err. where, fortunately, Hinman's attribution of H4v, H5, H6, H6v and I1-2v to compositor B is amply confirmed by the spacing and other characteristic practices, such as the setting of 'Mr, without a point, which have been noticed in his pages previously.[40] The compositors of the remaining pages are not as readily distinguished, for again on pages which Hinman assigned to D, B's partner in Err., there is conflict of spacing evidence between the columns of some pages. Spaced terminal commas indicate that C was responsible for H1 and H1va where his characteristic 'heere' is seen, but not column b where, although there is one spaced terminal comma, the other spacing and two 'indeede's tell against him. He also set H2b where 'heere' is found; five 'Mistris' spellings in column a and six 'Mistresse's in column b show that H4 was set by different compositors. The spacing of H2v and H3a (which also has 'Mistresse' spellings) identifies C as the compositor but he is unlikely, despite the spaced comma in column a, to have set H3v which has many indented turn-overs of verse lines: these are also found on H3b and H4b which C did not set. (Compositor C also set H4a as the spacing and 'heere' show.) His partner in the shared pages where 'do' and 'go' sometimes occur is therefore more likely to have been D than A. I


Page 83
have assigned the remainder of Err., where indented verse turn-overs are frequent, to D.

Happily, after MM and Err. the printing of the comedies became more regular and never again are the compositors found sharing the setting of single pages in such numbers. Undoubtedly the introduction of a fresh type-case, case z, from which hereafter D usually set, contributed greatly to normalisation of the compositorial work from Ado. onwards. In the rest of quire I, quire K and L1 which make up Ado., Hinman distinguished three compositors. Two are B, and D setting from case z, whereas the compositor who set from case x was doubtfully A, C, or both. The spellings do not give good evidence because much of Ado. is prose, but the spacing evidence leaves little doubt that the compositor of case x was C, even though, because of the many long lines, some of the spacing counts are inconclusive. The spellings, especially 'indeed' which only C sets against the copy spelling, supports C's composition of the pages I attribute to him.

The spacing also shows him to have set M1-3v of LLL where Hinman was in doubt: all the evidence is consistent with division amongst B, C and D (setting from case z). Spacing also confirms D as the compositor of N6v of MND.[41] In MV there are anomalies of spacing in C's pages but the spaced terminal commas give good evidence of his presence. Page P1v, which Hinman assigned to C from the spellings, I have given to D on the evidence of indented turn-overs and spacing.[42] The absence of 'do/go' spellings cannot tell against D in MV because there are none at all in quarto copy. Accordingly I have reassigned Hinman's A pages of quire O which were set from case z to compositor D: 'youle' on O5v and the spacing supports this, but great confidence is not possible when the only spellings which distinguish D from A cannot, by the nature of the copy, occur very frequently. I should perhaps note that Hinman shifted his ground here: compositor D, who was earlier identified by his tolerance of 'do/go' when they were in his copy is now identified because copy 'doe' and 'goe' were 'needlessly' changed to 'do' and 'go'.[43] In short, compositor D not only tolerated these spellings but also occasionally used them against copy: in this his habit is like C's. The spellings of MV are not so clear that it can be certain that every page has been correctly assigned to A and D, and a closer analysis may find that all pages set from case z were set by compositor D.


Page 84

None of Hinman's compositor identifications in AYL has been changed but in Shr. pages T5-6v which he assigned to A are more likely to have been set by D. The compositor set from a new type-case which had been freshly distributed into, so type evidence is not available to aid the identification of the compositor.[44] There are too many 'do/go' spellings for compositor A (it is suggestive that there is only one such spelling in C's pages: apparently 'do/go' were not frequent in copy) and the indented verse turn-overs, which occur on all four pages, and D's characteristic 'weele' on T5, support reattribution to him. To these D pages may be added V3 and V3v of AWW where, because there is only one 'do/go' spelling, Hinman identified A. However, the elisions are strong evidence of compositor D and there are indented verse turn-overs on both pages: compositor D was responsible for them.

Compositor B set the whole of TN and shared WT, the last text in the comedies section, with compositor A. The compositor identifications from a detailed investigation of the composition of WT by J. H. P. Pafford were confirmed by Hinman. Dr. Pafford noticed that B's page Bb3 contains a fairly large number of uncharacteristic spellings; his table shows, for example, four 'here's.[45] But as Hinman commented, all but one of the non-B spellings in this page occur in long lines; all the 'here' spellings are in long lines, and three of them are 'here's' spellings, B's frequent justified form. Spacing evidence confirms this page as his.