Facebook Twitter Gplus RSS
magnify
Home Miscellanea Preliminary Lessons Learned from Typesetting a Broadside Ballad
formats

Preliminary Lessons Learned from Typesetting a Broadside Ballad

Published on January 8, 2015 by in Miscellanea

Over the past several months the English Broadside Ballad Team (EBBA), supplemented by an additional group of UC Santa Barbara graduate students, has been engaged in the act of re-creating, from scratch, a 17th century Broadside Ballad.  Beginning with old shirts and rags, we worked in the UCSB Art Studio department’s papermaking lab to make our own paper using methods as close to historically accurate as possible.  We wrote our own ballad text; and, with the help of our musicologist Eric Bell, learned to sing it to a period appropriate tune. And, finally, we typeset and printed the ballad on a replica of an historical press at the UC Riverside Library’s printing lab run by Sara Stilley.

The intent of this activity was to see how a physical engagement with all stages of the printing process might inform our understanding of the ballads themselves.  A complete, peer-reviewed, multi-media volume designed both to demonstrate the making process and to critically engage it through a collection of essays written by project participants will be released under the Early Modern Center (EMC) imprint later in 2015. (Watch the comments to this post for a link to the publication once it goes live.)   This post offers a preview of some of the ways in which this experience has already affected my understanding of Broadside Ballads.

It is now obvious to me that, by and large, the printers of broadside ballads took far more care in producing their works than critics have been want to acknowledge. Scholars have tended to point to the myriad typos and or run-on/folded lines that appear in the ballads as signs of the careless and rushed manner in which any printer would, naturally it is assumed, approach the production of the “lowly ballad.”

This assumption has hardly been questioned in the literature; however, in comparison to the ballad that we typeset and printed, most 17th century broadside ballads stand out as exemplars of editorial care and effort.  Our first run ballad had no less than 12 typographical errors; and, despite our extreme efforts to mock-up the layout of the ballad prior to typesetting, we were forced to fold three different lines of print.  Most printing operations in London in the seventeenth century employed, on average, 3 to 5 workers for each print run.  Our production employed three Professors, one Post-Doctorate, and seven graduate students—all of whom spent a significant amount of time writing our ballad text so that it would fit nicely into two even columns with no folded lines and reading and proofreading the typeset before its initial run.  The first run product was then proofread again before going to final printing and, despite this effort, three errors still appeared in our final print.

The moral of the above story is that typesetting and printing is hard—really hard; and the simple fact is that the average 17th century print shop produced fewer errors per page than a dedicated and motivated room of present day, PhD level scholars.  As such, it seems that, contrary to the accepted opinion, 17th century printers of broadside ballads actually devoted significant care to their work and did not hurriedly throw together their product.  This suggests that they, at least, as the producers of the work, did not see the ballads as lowly at all, but as works worthy of care and attention.

Another lesson learned relates to the actual printing method of the ballads.  We know that, as a general rule, when printing an item that contained both type and blocks, it was standard for historical printers to run the paper through the press twice.  The forme for the first run would include all of the typeset to be printed on the sheet with a printed space left blank (using furniture to hold the type in place) where the block impressions were meant to appear.  Then, after the ink from the first run had dried, the typeset would be removed from the form and the blocks put in place.  The sheet would then be run through the press a second time, resulting in a completed sheet with both type and block print.

Double printing was a standard printing practice with the hand press because it was then and remains now nearly impossible to insure that sorts of type and woodblocks will stand at the same height.  In printer’s parlance, regardless of the effort spent producing a block, blocks hand-carved from wood blanks that are hand milled are seldom “type high” when compared to sorts of type which are produced through a regularized, mold-based manufacturing process.    This variation in height results in uneven pressure being applied to the sheet as it is run through the press and, as result, in uneven inking in the finished product when both woodblocks and type sorts are used in the same forme.

Because of their presumed lowly nature, it is a commonplace in ballad scholarship to claim that ballads were printed in a single printing for brevities sake.  As such, when printing our ballad, we naturally filled our form with both type and block so that we could produce a single-run printing.  What followed was nearly four hours of intense and dedicated work devoted to trying to create an even, type-high setting.  It proved, for all intents and purposes, impossible to produce anything that resembled the relative even inking that appears on an overwhelming majority of extant 17th century ballads.  While variations in inking do obviously appear in historical printings, compared to our prints, which were completely unreadable in this regard, extant 17th century broadside ballads stand as icons of even and readable inking.

Several additional facts are worth noting regarding our printing process.  First and foremost, the process was not simply supervised but largely executed by the trained hand of an expert in historical printing processes.  Also. the several hour long process that ultimately led to our finally producing one (yes, only one) usable print, involved sacrificing over 30 pieces of paper and two sheets of clear Mylar film to the process.  And finally, in the end, we were only able to produce this print by cheating the system and using clear tape to secure pieces of paper to the platen of the press (the portion that folds down on top of the paper and type-set forme and acts as the pressure plate during the printing.)  We had tried unsuccessfully to place these thin wedges under the blocks themselves in the forme, but each time we did so we would solve one problem but create another through the overall disruption of the tightly-packed forme.

There are several aspects of our story that should inform our understanding broadside ballads.  First, we can assert with absolute certainty (something you rarely get a chance to do as an early modernist) that 16th and 17th century printers had no access to clear tape or Mylar film.  Second, our method of taping pieces of paper to the platen to create even inking would fail during a multiple print run as the paper wedges moved significantly even after one print.  And finally, given the value of paper itself during the period, it is unlikely that any printer would be willing or able to sacrifice 30 sheets of paper to the setup of each print run.

All of the above realities suggest that it would, in fact, have been both more economical and faster to use a tradition double pass to print broadside ballads.  As one of the most prolific forms of printed output in the period, this fact, by extension, suggests that we must rethink not only our understanding of the ballads themselves but of the printing process and economy in general.  Additionally, when placed in the context of the care and attention paid to typesetting and proofreading as discussed above, it seems obvious that 17th century printers, at a minimum, had a much higher opinion the of the value of the broadside than have a majority of ballad scholars.

 
 Share on Facebook Share on Twitter Share on Reddit Share on LinkedIn
No Comments  comments 
© 2015 by Carl G Stahmer
All Rights Reserved