The Collation

Research and Exploration at the Folger

Myth-busting early modern book illustration, part one

There’s a common core of misconceptions that many readers of this blog will be accustomed to dispelling thanks to their interest in Shakespeare and Early Modern Europe. “O Romeo, Romeo, wherefore art thou Romeo?” doesn’t mean “Where’d you go, Romeo?!” Historic window glass didn’t “flow” to become thicker at the bottom over time. The printing press didn’t destroy manuscript culture. But what about myths we propagate without knowing it? When this came up in discussion at the Teaching Book History workshop (described last month by Sarah Werner) our self-deprecating laughter quickly turned to earnest requests for examples. 

The example that immediately came to my mind was the notion that copper plates wear out because the intense pressure from the rolling press compresses the soft copper over time (though I believe the term I used in the workshop was “squooshes” rather than “compresses” since it’s rare that a good example and proper vocabulary come to my mind at the same time). ((For a quick review of how engravings are made and printed, see the 7 February 2012 Collation post “Woodcut, engraving, or what?“))

A rolling press, for printing from intaglio plates
A rolling press, for printing from intaglio plates

This compression, according to the hypothesis, is the reason why prints in later editions of a book are paler than those in earlier editions. Compare a detail of a ship from the 1552 (first) edition of L’isole piu famose del mondo and the 1556 (second) edition. The area of cross-hatching on the sail makes a particularly good example of engraved lines becoming thinner (paler) and shorter over time:

Engraving details showing fainter, shorter lines in the later edition
Detail of a ship’s sail from the upper right of the “Spagnola” map in Thomaso Porcacchi, L’isole piu famose del mondo… 1572 edition (on left) and 1576 edition (on right). Click image to open a pan-and-zoom window.

While it is true that copper is soft (or else you couldn’t cut lines into it with hand tools) recent studies by S. Blair Hedges, a scientist at Penn State, demonstrate that plates did not compress with use, and that even if they had, lines would become longer and wider, not shorter and thinner (imagine rolling out pie dough, marking an “X” on it, then using the rolling pin some more. The “X’ will stretch and grow). Instead, Hedges’ work supports the idea that plates wore out through erosion when they were polished for re-use after storage. ((Hedges is a biology professor and print collector; this research is the result of his realization that the “molecular clock” he uses to date divergences among species based on the random mutations in the genome over time could be used as a model to create a “print clock” of accumulated changes in a printing surface over time.))

Stair-step graphic showing shallower triangle grooves with each of 4 editions
Generalized diagram showing change in width of an engraved triangular-shaped groove in a metal (copper) plate through time (four editions), and the printed lines produced by that groove. The cross-section of the ink-filled groove is black. Before each printing, the plate surface was polished (eroded) to remove nicks and corrosion, narrowing the width of the triangular groove. (Copyright S. Blair Hedges; made available for educational purposes)

Lines become thinner not in relation to the number of impressions taken from the plate, but in relation to the time between printing sessions. ((Hedges’ work also takes into account variations in inking (an under-inked plate will also make paler, thinner lines), re-touching of plates, and other variables too numerous to detail here.)) Using digital images and scientific software designed for image analysis, it is possible to gather data from multiple copies of different dated editions of a print, graph them to show where the different editions cluster in terms of line density, then see where data collected from an undated edition of a print falls within the time scale. Hedges was able to show similar correlations between time and image deterioration in woodcut illustrations by counting line breaks (the result of increasing cracks and losses in the raised surface of the block).

Why isn’t this information better-known among people who teach and study the history of the book? Perhaps the most obvious reason is that it’s hard enough to keep up with academic journals such as Renaissance Quarterly and Printing History, so when Hedges’ work appeared in Proceedings of the Royal Society A : Mathematical, Physical, and Engineering Sciences in 2006 ((Hedges, S. B, 2006. A method for dating early books and prints using image analysis. Proc. R. Soc. A : Mathematical, Physical, and Engineering Sciences 462:3555-3573. Download PDF here. Also see Electronic appendix PDF.)) and Proceedings of  SPIE (the international society for optics and photonics) in 2008 ((Hedges, S. B. 2008a. Image analysis of Renaissance copperplate prints. Proc. SPIE 6810: 681009. 20 pages. E-print. “SPIE” gave up trying to make its name match its acronym in 1981, having previously been the Society of Photographic Instrumentation Engineers, then the Society of Photo-Optical Instrumentation Engineers.)) it didn’t exactly find its audience.

For a general overview of the research and practical how-to instructions, see The Print Clock: A method for dating early books and prints.

What else might we learn about early modern prints from a massive database of crunchable numbers? Blair Hedges reports that “discriminant function analysis” of three different sets of early modern prints and three different Renaissance maps “was able to group art prints from different sets and maps based on different mapmakers” based on the density of the line work. ((Hedges, S. B. 2008a. Image analysis of Renaissance copperplate prints. Proc. SPIE 6810: 681009-18.)) This sounds very much like the kind of statistical analysis of word patterns and types already familiar to many in literary studies.

Can the study of book illustrations give anything back to the scientists? You know those worm holes that show up as white spots in woodcuts?

Woodcut illustration showing prominent white circles indicating wormholes

As a print collector, Blair Hedges knows worm holes, and as a biologist, he knows worms (or rather, the wood-boring beetles that make what we refer to as worm holes). It turns out that different species of wood-boring beetles make different-sized holes, so he started counting and measuring white spots on book illustrations known to have been printed in particular places at particular times. It turns out that two species whose current European habitats overlap almost completely were clearly separated until modern times, one in northern Europe, the other in southern Europe. ((See Hedges, S. B. 2012. Wormholes record species history in space and time. (main article). Biology Letters (E-print) and Hedges, S. B. 2012. Wormholes record species history in space and time. (supplement). Biology Letters (E-print).)) There must be plenty of other opportunities for studies like this, it’s just that people in the humanities and the sciences need to pool our knowledge. Let’s get busy!


  • Erin – greetings. I think the major factor in plate-erosion must surely have been simply the heavy wiping to remove surface ink – both before and after taking each impression. Watching someone clean and re-ink a plate is highly instructive in this regard. That said, I don’t doubt periodic refurbishment also played a part. All best wishes, Laurence.

    • Good to hear from you, Laurence! I’m glad you mentioned wiping the plate, since Blair Hedges specifically addresses that in his 2006 article, but I didn’t mention it here for lack of space. (For people unfamiliar with wiping a plate, see for example this clip).

      Wiping the plate before and after each impression is a vigorous activity, but if that were the cause of the measurable thinning of the lines over time, there wouldn’t be a jump in deterioration between editions of an early-modern illustrated book. Instead, the deterioration would be gradual, with prints made late in the run for the first edition looking quite similar to prints made early in the run for the second edition, and that isn’t what Hedges’ evidence shows for the 16th and 17th centuries (once protective coatings started to be put onto copper plates before they went into storage there was no longer a need to polish off scratches and corrosion).

  • Thanks for this informative piece. I’ve been wondering about corrosion and copper plates for a while. Somewhere in the 17th century(*), there was a re-edition of a work by Bruges engraver Hubert Goltzius on coins. The re-edition apparently used heavily corroded plates which weren’t cleaned, possibly because they wanted to preserve the very fine detail used by Goltzius.

    (*) I tend to forget these details but I can look them up if you want to. Perhaps Goran will be able to provide a copy in the Folger.

  • Fascinating! This piece will fit well with my discussion of EM printing and bookmaking technologies and Hamlet tomorrow. I’m looking forward to part two (and beyond?)!

Leave a Reply

  • (will not be published)