The Collation

Research and Exploration at the Folger

Two disciplines separated by a common language

I should have seen it coming when the Art History professor and the English professor started talking with each other about “print culture” (names omitted to protect reputations). It soon became clear that one had been talking about the circulation of printed pictures, the other had been talking about the circulation of printed words, and neither wanted to let on that they hadn’t been talking about both all along. Full disclosure: when I first came to the library world from the art world, I had no idea that familiar picture-printing terms have different and sometimes contradictory meanings in word-printing. This post is for anyone else who didn’t know that they didn’t know this.

My intention is to use bold when a shared word appears in its book sense, and bold italics when a shared word appears in its picture sense. Let’s see if I can avoid getting them mixed up. If I do get mixed up, or if you can think of other examples, please speak up in the Comments. 1

In the picture world, an edition is the number of copies in a print run. The term usually only comes up when the number is artificially capped—for instance, “printed in an edition of ten” or “number fourteen of an edition of twenty-five” (which would be written simply “14/25” in pencil on the print). [Updated to include the following image] See, for instance, the Folger’s copy of J.W. Winkelman’s 1994 etching “Early Stage,” which is number eighteen of an edition of one hundred:

J.W. Winkleman's "Early Theatre" (1994) showing its edition statement
J.W. Winkelman’s “Early Stage” (1994) showing its edition statement

For book people, an edition isn’t a number, it’s the collective noun for all of the copies resulting from (roughly) a single setting of type. 2 Deliberate changes to the settings of type constitute different editions (second edition; revised edition; annotated edition; etc.) and the publisher usually proclaims the difference on the title page.

So how do picture people describe deliberate changes to the same work? Simple. Each visually identifiable stage in the life of a picture’s printing surface is a different state (second state; state 1 of 3; early state; etc.). Prints from the early modern period frequently exist in more than one state, since the printing surfaces were often touched-up as they wore out, or altered to replace one printseller’s name with another’s when the plate (metal printing surface) or block (wooden printing surface) changed hands. The earliest state of this print by Martin Droeshout says “Sold by Roger Daniell at the Angell in lumbard Streete” in the oval near the bottom:

Close-up of engraved imprint
Detail of an engraved title page in the Folger collection (click to go to full image)

In a later state, the text in the oval has been changed to “Are to be sold by Thomas Johnson in Brittaynes Burse.”

Detail of imprint
Detail of an engraved title page in the British Museum collection (click to go to full image and catalog record)

On the other hand, a state in bibliography is a copy or group of copies within the same impression (i.e., print run) that differs from the others in some  way that the publisher does not want to proclaim as different. 3

However, an impression in picture printing refers to a single copy of a print.

Detail of artist’s note in the margin of a printed image (click to go to full image)

And a copy in picture printing refers to a reproduction of a pre-existing picture. Sometimes a copy reproduces a picture in a different medium. The etching below is Wenceslaus Hollar’s copy of a painting by Hans Holbein:

Monochrome image of a baby in rich clothes
Etching of a painting of Edward VI as Prince of Wales (click to go to zoomable image)

The original picture is a much larger, and very colorful, oil painting.

Colorful oil painting of a baby in rich clothes
Painted portrait of Edward VI as Prince of Wales now in the collection of the National Gallery of Art, Washington, DC (click to go to catalog description)

Other times, a copy in the picture world is a replica. This plate 4 of Boscobel House was removed from a copy of the second edition of Sir Thomas Blount’s Boscobel, or, The history of His Sacred Majesties most miraculous preservation after the battle of Worcester, which was published 1769. It is a close copy of a Wenceslaus Hollar print from the first edition of the book, published in 1660.

Engraving of a forested scene.
Illustration from the second edition of Thomas Blount’s Boscobel: or, the compleat history of His Sacred Majesty’s most miraculous preservation after the battle of Worcester. Worcester : printed for S. Gamidge, 1769 (Click to go to zoomable image)

Why was a copy used rather than additional impressions from the original plate? Presumably, the plate etched by Hollar was no longer available, so the plate in the book was made from a new plate.

Speaking of metal printing plates, sometimes picture terms are simply misunderstood in the book world. You know how an engraving is printed from a sheet of metal with lines cut into it, and how a woodcut is printed from a piece of wood with spaces between the lines cut out of it?  (See Woodcut, engraving, or what?.) Long ago, a policymaker at ESTC assumed that if early modern prints from wood are woodcuts, then early modern prints from metal must be metal cuts:

"Physical descr. ... maps (metal cuts)...
Screen shot from the English Short Title Catalogue (click to go to full record)

But in fact, prints made from metal or wood by cutting out the spaces between the lines are metal cuts and woodcuts, respectively (with metal cuts being quite rare, especially after the 16th century). And prints made by cutting lines into metal or wood are engravings  – copper engravings, steel engravings, wood engravings, etc. (with wood engravings being quite rare until the 19th century).

And THEN there’s the too-common assumption that “book” is synonymous with “codex of printed words,” despite the fact that bound manuscripts, bound prints, and bound hybrids of all three are also books… [Erin wanders off down the hall, muttering to herself. Again.]

  1. On the assumption that most Collation readers are already familiar with the the book-world meanings of the terms, and because this post is overdue, I’m only illustrating the picture-world meanings.
  2. Philip Gaskell describes an edition as “all the copies of a book printed at any time (or times) from substantially the same setting of type, and includes all the various impressions, issues, and states which may have derived from that setting.” He goes on to describe “substantially the same setting” as meaning, roughly, that less than half the type has been reset. (Gaskell, A New Introduction to Bibliography, Second Impression, Oak Knoll Press, 2006, p.313.
  3. Again, here’s Gaskell for more nuanced definitions of impression—“all the copies of an edition printed at any one time” (p.314) which, in the hand-press period, essentially is the same as an edition—and state—“all other variants from the basic form of the ideal copy” (p.315) including stop-press corrections, inserting or removing preliminaries, adding errata leaves, inserting or removing text during the process of printing.
  4. The standard manual for cataloging rare materials (commonly known as “DCRM(B)”) provides the following definition of plate: “A leaf that is chiefly or entirely non-letterpress, or a folded leaf of any kind, inserted with letterpress gatherings of text. A plate usually contains illustrative matter, with or without accompanying text, but may contain only text, e.g., an engraved title page or a folded letterpress table.” (Association of College and Research Libraries, Descriptive Cataloging of Rare Materials: Books Washington (DC: Cataloging Distribution Services, 2007), p. 203.


  • With regard to the bibliographical term “state,” it’s critical to also include the last phrase of Gaskell’s paragraph on the subject (p. 316):

    “… differences of state are generally the attributes of individual formes, or sometimes of individual sheets.”

    Although it’s all too common to find copies described as being of a certain state, it’s misleading to do so. This is especially the case when an impression has multiple stop-press corrections during the print run, whether in the same forme or in different formes, so any given copy will exhibit a combination of states.

    I’d go a little further than Gaskell, and omit reference to individual sheets – a sheet is almost always printed from at least two formes, and if there are variants in both formes, it makes more sense to specify the states of each forme separately, and then describe the combinations of states that are found in various copies of the sheet. This can perhaps be useful in establishing the order of printing, though such evidence can be tricky to evaluate.

  • Thanks, Erin, for another post that helps expose some of the misconceptions and fuzzy thinking about book history. You provide an excellent reminder of the need to be precise about our use of terms that at least aim for some technical precision. I fear, though, that there’s some inevitable tension between the goal of technical precision and the messiness of practice in the book trade. I remember Peter Blayney cautioning in his Folger Institute course years ago that we really didn’t have a handle on distinctions among edition, issue, and state. Didn’t the fluidity and ad hoc practices of the printing house work against the terminological precision we would all like to impose? Nevertheless, and again, you remind us of the need to query for ourselves our own understandings of these issues–and to articulate our own usages and the complicating factors we find in our research.

  • In response to both John Lancaster and Kathleen Lynch, I would agree that it’s true that bibliographers continue to quibble over the precise usage of their terms (and I would agree that Gaskell is sometimes more fixed than the fluidities of early printing practice). But I think Erin’s main point is larger than that: there is a real difference between how book historians and art historians talk about their objects. The quibbling that bibliographers do over printing practices isn’t going to affect the distances between the two groups if we don’t understand the differences between our training and our vocabulary.

  • Great article! In regards to the section (and earlier linked post) on woodcuts and engravings, and to make things a bit more confusing, it seems worthwhile to point out that wood engravings are, like woodcuts, a type of relief print. While metal engravings (and etchings) are intaglio prints. The terms “engraving” and “cut” have more to do with (rather fuzzy) differences in how the marks in the plate/block are made, as opposed to how they are printed.

    • Good point. I deliberately restricted the earlier post on woodcuts and [metal] engravings to the 16th and 17th century for precisely that reason! Wood engravings are printed in relief. From the late 18th century until the 1830s, they usually read quite clearly as white lines cut out of a black background. See, for example, (it’s most obvious if you zoom in on a dark area). Later 19th-century wood engravings often convincingly imitate etchings, confusing things even more.

      The main technical difference between a woodcut and a wood engraving is that woodcuts are made on the softer “plank” side of a chunk of slow-growing hardwood, while wood engravings are made on the much harder (and smaller) *end* of a chunk of slow-growing hardwood (the ends are bolted or glued together if a surface bigger than three inches or so is needed).

Leave a Reply

  • (will not be published)