Standing Type in Dicey/Marshall Issues of The Berkshire Tragedy

Abstract

Before the nineteenth century and the introduction of stereotyping, the tacit assumption has been that once street-literature publications such as broadside and chapbook ballads had been printed, the type would be distributed, and if there were subsequently a demand for more copies the text would be set once more from scratch. Any woodblock images would be retained, although the same block might or might not be reused if the ballad was printed again. Standing type was not unknown in the hand-press period, but there was clearly a need to keep the cases of type replenished, and the overall quantity of type in any particular printing house must have been a limiting factor.¹ Title-pages and pages set in type that was not in such constant use (perhaps including italic) were more likely to have been retained than body text that contained sorts that were in regular use. Nevertheless, there are known instances of short books (up to three or four sheets) kept in standing type, from quite an early date, and it does not seem impossible that single-sheet ballads might have fallen into the same category. McKerrow observes that the amount of type required to keep pages of verse standing need not be excessive.²

On the other hand, there are many instances where the same ballad has been printed more than once by the same printer—including the Dicey/Marshall firm, the dominant London printers of street literature for much of the eighteenth century,³ and James Catnach and John Pitts in the early decades of the nineteenth century—and where even quite casual observation will reveal that they are not from the same setting of type. Perhaps for that reason as much as any other, the possibility of standing type in earlier ballad printing has rarely been considered. Moreover, the idea of rapid printing and distribution feeds into a narrative of street literature as ephemeral, quickly and shoddily produced, and just as quickly discarded, by printers who occupied the lower echelons of the trade.⁴ Even Charles Hindley's account of the Catnach press, though he clearly does not find the idea of street literature kept in standing type surprising, still manages to perpetuate that same narrative, recording that Catnach was a great buyer of worn-out and damaged printer's stock at auction sales, even while he ‘had a great horror in laying out his money in new and improved manufactured type, because, as he observed, he kept so many standing formes’.⁵ Among the corpus of extant printed ballads, Bergel, Howe, and Windram have noted that the body text of The Wandering Jew's Chronicle was reprinted from standing type on two separate occasions: firstly in the reign of George I, when the booksellers/printers in question were C. Brown, T. Norris, and J. Walter; and then again in the reign of George IV, when the ballad was twice printed by Pitts.⁶

The present study seeks to examine this phenomenon in rather greater detail in respect of The Berkshire Tragedy, of which there are six different issues that can be ascribed to the Dicey/Marshall firm on the basis, variously, of imprint, provenance, typography, iconography, and ornaments. The Berkshire Tragedy itself is a fairly typical ‘murdered sweetheart’ ballad. It is first known in two chapbooks with the imprint of John Keed, in Edinburgh and York, dated 1744.⁷ These both include a prose appendix purporting to give the last dying words and confession of John Mauge, a miller, executed at Reading for the murder of his pregnant sweetheart, Anne Knite. There is no known record of the crime or of the execution of the perpetrator, and a number of eighteenth-century murder ballads are believed to be fictional, in the manner of what in the nineteenth century were known in the trade as ‘cocks’. The prose section does not appear with subsequent printings. The ballad itself is listed among the ‘old ballads’ in the William and Cluer Dicey catalogue of 1754,⁸ and again in the Cluer Dicey and Richard Marshall catalogue of 1764.⁹ Other copies were printed by Sympson (Stonecutter Street), Evans (Long Lane), Pitts, and Catnach in London, and by Cheney (Banbury), Wrighton (Birmingham), and Turner (Coventry). A broadside otherwise without imprint is dated 1796. The ballad was recast in the early nineteenth century into a shorter form, The Cruel Miller, and later collected as a folk song.

The six different Dicey/Marshall printings are most obviously distinguished one from another by text in either four or five columns, by the woodcut printed between the title and body text, by the presence or absence of an imprint, and by the correct or incorrect spelling of the word ‘Tragedy’ in the title:

• A [title across two text columns] The BERKSHIRE Tragedy: | OR, | The WITTAM MILLER. | With an Account of his Murdering his Sweetheart, &c. | [image A] | [text in four columns separated by ornamental borders, roman, no stanza divisions]

  • Cambridge, MA, Houghton Library, EB75 P4128C no. 19

• B [title across two text columns] The Berkfhire Tragedy, | OR, | The WITTAM MILLER, | With an Account of his Murdering his Sweetheart, &c. | [image B] | [text in four columns separated by ornamental borders, roman and italic, no stanza divisions]

  • ESTC T205905

  • Oxford, Bodleian Library, Douce Ballads 3(ib)

• C [title across two text columns] The Berkfhire Tragedy, | OR, | The WITTAM MILLER, | With an Account of his Murdering his Sweetheart, &c. | [image A] | [text in four columns separated by ornamental borders, roman and italic, no stanza divisions]

  • Brighton, Jubilee Library, box of 41 Broadsheets

  • Cambridge University Library, Madden Ballads 1.44 (currently listed along with Oxford, Bodleian Library, Douce Ballads 3(ib) as ESTC T205905)

• D [title across two text columns] The Berkfhire Tragedy, | OR, | The WITTAM MILLER, | With an Account of his Murdering his Sweetheart, &c. | [image A trimmed] | [text in five columns without borders, roman and italic, 22 eight-line stanzas]

  • ESTC T2154 6

  • London, British Library, 1876.e.i.(ii.)

  • San Marino, CA, Huntington Library, 289727

• E [title across two text columns] The Berkfhire Trgedy, | OR, | The WITTAM MILLER, | With an Account of his Murdering his Sweetheart, &c. | [image A trimmed] | [text in five columns without borders, roman and italic, 22 eight-line stanzas]

  • ESTC T204012

  • Oxford, Bodleian Library, Antiq. c. E.9^25)

  • Oxford, Bodleian Library, Harding B 6(96)

• F [title across two text columns] The Berkfhire Trgedy, | OR, | The WITTAM MILLER, | With an Account of his Murdering his Sweetheart, &c. | [image A trimmed] | [text in five columns without borders, roman and italic, 22 eight-line stanzas] | [horizontal rule across one column] | Printed and Sold at No. 4, Aldermary Church Ya[rd.]

  • ESTC T21547

  • Cambridge University Library, Madden Ballads 1.43 (horizontal rule stops short of left-hand edge of text column; top of ‘P’ of ‘Printed’ on level with rule; imprint torn after ‘No.’)

  • London, British Library, L.R.3i.b.19.(289.) (horizontal rule extends beyond left-hand edge of text column; ‘P’ of ‘Printed’ tilted left; imprint illegible at end ‘Churc[h Yard.]’)

  • San Marino, CA, Huntington Library, 289726 (horizontal rule stops short of left-hand edge of text column; top of ‘P’ of ‘Printed’ on level with rule; space between ‘r’ and ‘m’ of ‘Aldermary’; imprint illegible at end ‘Church Ya[rd.]’)

  • image A: rectangular woodcut across two text columns: (background left) house, figure looking over fence, (foreground left) woman's body lying on back with dress spread out, (centre) woman on knees in posture of supplication facing man holding stick, (right) body suspended from gallows.

  • image B: two woodcuts side by side across two text columns: (left) rectangular image, (foreground) man and woman holding hands, (background left) house on hill, (background right) row of houses on hill; (right) body suspended from gallows.

To anticipate a little, there is compelling reason to believe that all but one of these issues are typographically related: that is to say, they are printed from the same type, though not always from the same layout of that type.

It will be most convenient to commence with the one printing that does not fall into that category, represented by a broadside in Thomas Percy's collection now in the Houghton Library and probably one of the ballads that Percy acquired from Cluer Dicey c. 1761 (A).¹⁰ It is printed in four columns, entirely in roman type, with ornamental borders separating the columns but without stanza divisions. The woodcut (image A) can reasonably be interpreted as an iconographic representation of the story told in the ballad: the scene of the murder in the centre, with the man holding the murder weapon (which actually looks more like a club than a stick pulled from the hedge as described in the text) and the woman in a posture of supplication (‘But she fell on her bended knee / And did for mercy cry’); to the left, her body lying on its back floating down the stream and the figure of her father looking over a fence (‘The body it was found / Floating before her father's door’); to the right, the murderer's body suspended from the gallows. This would seem to justify, at least in this instance, the claim of the Dicey catalogues that their old ballads were ‘Printed in a neater Manner, and with Cuts more truly adapted to each Story, than elsewhere’. There are a number of minor textual variants between this printing and the rest of the sequence discussed below.

Another broadside comes with an altogether different setting of type, with body text again in four columns but in a mixture of roman (12frac12; stanzas) and italic (9frac12; stanzas), without stanza divisions, with the same column breaks but with the borders between them made up of a different set of ornaments, and with a different setting of type in the title (B). The accompanying image is entirely different, comprised of two woodcuts side by side (image B): to the left, a scene dominated by a man and woman holding hands in the foreground; and to the right, a body suspended from a gallows. These woodcuts are noticeably both cruder and more generic than the image A woodcut, and although they were probably brought together for the purpose they are no more than loosely relevant to the ballad story. Then, exactly the same setting of body text in four columns in roman and italic, confirmed by the further coincidence of distinctive pieces of type (see below), and with the same ornamental borders, is found in another printing (C).¹¹ Here, though, the image is once again the woodcut specific to The Berkshire Tragedy (image A). This pair of four-column printings (B and C) is sufficient to indicate that the body text was held in standing type. The title may also have been retained, although the wording is quite sparse so it is difficult to be certain.

Turning to the remaining broadsides (D, E, F), here the body text is in five columns, without borders between them, but with stanza divisions (divided into twenty-two eight-line stanzas), and all of them have the image A woodcut, but trimmed at the right-hand side. The setting of body text is exactly the same throughout all of these five-column printings, although they can still be distinguished by the spelling of ‘Tragedy’ or ‘Trgedy’ in the title, and by the presence or absence of an imprint. Aside from ‘Trgedy’, the title appears to remain the same, and is the same as in B and C. And when examined closely, it is apparent that the type sorts used for the body text in these five-column printings (D, E, F) are exactly the same as those used in the four-column printings (B and C). The four- and five-column printings are clearly not the same setting as such, but the coincidence of distinctive pieces of type is so great as to mean that the set lines of type themselves must have been rearranged and reused (re-imposed).

The distinctive features are many and various. They include the 12frac12; stanzas in predominantly roman type and 9frac12; stanzas predominantly in italics, although those would not in themselves necessarily confirm reuse of the same sorts. Similarly, things like small caps in ‘Wittam’ (2.3), the singular ‘cloath’ (17.3), the ungrammatical ‘Has You’ (20.3), or the unwarranted repetition ‘now now’ (21.6), could conceivably (if improbably) have been copied from one setting to another. But there are many other things that cannot be explained in that way, of which the best examples are: the letter ‘r’ dropped below the line in ‘Oxford’ (1.5, 2.5), turned small cap ‘t’ (5.3), ‘sopely’ (9.6), ‘miscbief’ (9.7), turned italic ‘u’ in ‘nnto’ (11.4), ‘wlthing’ (11.8) , ‘befor,’ (15.1). Additionally, there are specific occurrences of italics in stanzas of roman type: ‘not fit to’ (7.4), ‘that’ (10.5, 21.6), ‘to’ (10.7), ‘the’ (10.8) , ‘sent’ (21.8), ‘take’ (22.1); and of roman and small caps in italic stanzas: ‘But’ (11.3, 14.5, 18.3), ‘that’ (17.3), ‘The’ (17.5), ‘The’ (18.5), ‘to’ (19.7) . Some further recurrent quirks are lower-case ‘i’ for the first-person pronoun, small cap ‘t’ for upper-case roman ‘T’, and very many instances of italic ‘r’ for roman ‘r’. Some other markers, such as worn or damaged letters, are less conclusive but are still consistent with the scenario being proposed: the worn or damaged descender of ‘J’ in ‘Judge’ (17.5), for example. Presumably the inappropriate use of roman, italic, lower case, and small caps are all the result of shortages of the correct sorts. Indeed, the setting of a substantial part of the ballad in italics quite likely suggests a shortage of roman type.

There are a few instances where all the printings do not look quite identical, although there are probably innocuous reasons for some of these, such as light or heavy inking. ‘For’, which is printed clearly enough in the four- column printings, is damaged in the five-column printings (3.5). In the italic section, in the five-column printings the ascender of the letter ‘d’ is worn or damaged in ‘deny’ (14.7), ‘second’, ‘siez‘d’ (15.5), the second ‘d’ of ‘deed’ (15.8) , and ‘bound’ (19.1), and the letter ‘f in ‘befor,’ (15.1) is badly damaged, whereas the equivalent characters in the four-column printings are generally in a better state. The line that in the four-column printings reads ‘But Satan did me still persuade’ looks like ‘ ut Satan did me stili persuade’ in the five-column printings, where the ‘B’ must have dropped out or otherwise failed to print and the penultimate word is ‘still’ with damage to the ascender of the second ‘l’ (13.1). The five-column printings also have ‘Have mercy on me thee pray’, and ‘S I the second time was siez‘d’, where once more letters have dropped out or failed to print (22.7, 15.5). In the latter instance, the ‘o’ of ‘So’ is present in B and in the Cambridge exemplar of C, but is missing from the Brighton exemplar, presumably indicating that it dropped out while C was being printed, and therefore that C was printed after B.

The two lines ‘And there examined again / about the bloody deed.’ come at the head of a new column in the four-column printings, but conclude a stanza in mid-column in the five-column printings (15.7—8). In the latter, ‘there examined’ appears as ‘thereexamined’, with ‘ere’ seemingly pushed a little below the line and most of ‘examin’ pushed a little above it. Indeed, the whole of this stanza is noticeably uneven. It looks, then, as if the two lines ‘And there examined again / about the bloody deed.’ had to be pushed into position when making up the stanza for the five-column printings and this caused some disruption to the evenness of the type.

Something similar seems to have happened with respect to the lines ‘And perfect flames of Hell did flash / wlthing my guilty face.’, which fall in midcolumn in the four-column printings but at the foot of a column in the five- column printings, where they read ‘And perfect flames of Hell did flash / wlthingce.myguilty fa’ (11.7-8). The erroneous ‘wlthing’ is retained, but it looks as if some of the other characters became loosened or dropped out and were replaced incorrectly. Bowers observes that when a forme is unlocked, lines at the top or bottom of the page are most likely to be prone to displacement and disruption (‘pieing’).¹²

Elsewhere, when C, D, E, F are compared with B, ‘hair,’ is damaged, to varying degrees, and the end of line comma cannot be made out (8.5). The pronoun ‘I’ has dropped below the line at the beginning of a line in E and F (with ‘Trgedy’ in the title), but not in D (with ‘Tragedy’) (6.1). Also in Eand F, there are signs of incipient damage to the bottom right-hand serif of the ‘M’ of ‘WITTAM’ in the title. Near the end of the ballad, the final word of the line ‘Lord grant me grace while I do stay’ is quite badly damaged and not really legible in two of the three exemplars of F (21.5). There are further differences between exemplars of F involving the imprint, probably as a result of damage to, or looseness of, the type. ‘Yard.’ is wholly or partially illegible, and the extent of the horizontal rule, orientation of the ‘P’ of ‘Printed’, and spacing in the word ‘Aldermary’ are all affected.

As already noted, all but one of these Dicey/Marshall printings carry a woodcut that was most probably designed specifically for this particular ballad (image A). It is placed directly beneath the title and spans two columns of text, regardless of whether the text is in four or five columns. When matched with the four-column text, the full extent of the image is printed, including the rectangular frame on all four sides (A and C). When matched with the five-column text, however, the frame and a small portion of the scene, amounting to approximately one centimetre, is cut off at the right-hand side (D, E, F). Also, in the five-column printings the woodcut extends well to the left of the left-hand edge of the first column of text, whereas in the four-column printings it is more or less flush with the edge (C), or even slightly indented (A). (In the four-column B, with image B, the woodcut is more or less flush with the edge of the text.) Evidently, spatial constraints created by the re-imposition in five columns required a realignment of the image, and a small portion of the right-hand side of the woodblock was physically removed (sawn off) in order that it could be accommodated across the first two columns. The left-hand border of the woodcut, which appears solid in A, shows signs of damage elsewhere, most likely worm damage. Although far from an exact science (in some instances dependent on the examination of digital images, and potentially influenced by variation in the weight of the inking), the damage appears worse in the five-column printings.¹³

All these pieces of evidence (based on an admittedly small number of surviving exemplars) point towards a putative sequence for these Dicey/Marshall broadsides. The firm's catalogue indicates that the ballad was available in 1754, while the presence of A in Percy's collection most probably means that particular printing was available in 1761. The woodcut evidence suggests that A probably preceded C. Given that the woodblock was apparently designed especially for The Berkshire Tragedy, it is a puzzle why a different image should have been printed with B. Perhaps C was in some sense a corrected state of B which substituted the ‘correct’ image. In any case, it is possible to be rather more confident that the four-column printings preceded the five-column printings. The No. 4, Aldermary Churchyard imprint probably belongs to the 177os/8os (F).¹⁴ The reason for the switch to five-column printing is not at all obvious, but perhaps it was thought important to introduce stanza divisions.

So, it looks as if the ornamental borders were removed from the already standing type in four columns, and the lines of type separated with additional spacing into groups of eight, which required their being reimposed in a new five-column layout. It seems essential to posit a scenario that kept the page of type together, albeit unlocked from the forme, and presumably with a good deal of spacing in the form of quads removed. In the process, some of the type may have become loosened and even dropped out, and in the case of ‘wlthingce.myguilty fa’ was replaced incorrectly. At the same time, the woodblock had to be removed and cut down in order to fit the new layout.

At some point, too, perhaps at the end of the sequence, the ‘a’ fell out of ‘Tragedy’ in the title, and instead of replacing it the ‘r’ and ‘g’ were closed up (presumably it was perceived as a spacing error rather than one of orthography).¹⁵ The ‘I’ dropped below the line in E and F (6.1), and the damaged word ‘stay’ in exemplars of F (21.5), perhaps point towards those particular printings, with ‘Trgedy’ in the title, coming at the end of the sequence. Certainly, the ‘Tragedy’/‘Trgedy’ variant and the presence or absence of an imprint point towards a certain amount of deliberate intervention, perhaps even during a single press run. Howbeit, if the switch to the five-column format with stanza divisions had anything to do with the claim that the Dicey/Marshall ballads were ‘Printed in a neater Manner […] than elsewhere’, the actual result was rather ironic.

The scenario posited here is at first sight rather unexpected. Is it not quite far-fetched to think not only that a ballad should have been kept in standing type, but also that the type should have been reused even when the layout was altered? After all, the type used for A must have been distributed after printing. On the other hand, there are a few documented instances where standing type was re-imposed so that it could be issued in another format. A celebrated example is The Case of the Planters of Tobacco in Virginia of 1733, which saw the first setting as a folio sheet kept in standing type for something like a couple of months before it was completely re-imposed to be printed again in an octavo format.¹⁶ What is being posited for The Berkshire Tragedy is not, therefore, without precedent, and modern presuppositions about ballads and ballad printing should not rule it out. The English Short Title Catalogue does note a small number of different issues of other ballads from the same setting of type, distinguished by variant imprints and/or the presence/absence of an imprint. Researchers working with Dicey/Marshall ballads are now coming to the view that the practice was not uncommon. The Bloody Gardener's Cruelty, for example, exists in at least three different states from the same setting of type, as determined by recurrent typographical peculiarities. It is a painstaking task to identify and enumerate such instances, and the present study of The Berkshire Tragedy can stand as a detailed example of the phenomenon, while further research in this area will potentially shed new light on the ballad trade in the eighteenth century. At the very least, it points towards the anticipation of recurrent demand for copies of particular ballads, but perhaps it also suggests that initial print runs might have been quite limited.

Footnotes