My last post focused on my frustration with the assumption that digitization is primarily about access to text:
But access is not all that digitization can do for us. Why should we limit ourselves to thinking about digital facsimiles as being akin to photographs? Why should we think about these artifacts in terms only of the texts they transmit? Let’s instead think about digitization as a new tool that can do things for us that we wouldn’t be able to see without it. Let’s use digitization not only to access text but to explore the physical artifact.
I spent the remainder of that post brainstorming some suggestions about what digitization might enable other than access to text, and there were some great comments about the ramifications of textualizing the digital that I’m still mulling over. In this post I want to offer some examples of why we might want to look at books rather than digital surrogates as a way of approaching the relationship between digital and physical from another angle.
So why might we want to look at physical books rather than digital surrogates, other than a fetish of smell and a sense of the magical presence of the original? Here are a few examples that start to get at what physical books offer that digital surrogates miss.
The making of the book
One of the things that we learn from examining books is how they are made. There are all sorts of things you can see in physical books that reveal their making: watermarks and chain lines in the paper (which can help date a book, as Carter Hailey has shown ); sewing structures in bindings (which can reveal if pages have been cut out or added in later); and pasted-in cancels or errata slips (which show changes made due to correct errors or in response to censorship).
It’s true that some of these features could be incorporated into digital surrogates; as my photo above shows, it’s absolutely possible for digital images to reveal paste-ins such as this one. But it’s also true that most users of digital surrogates of early modern books rely on EEBO, in which the equivalent page appears thusly:
The text might be the same, but there is no indication in the EEBO reproduction that the text appears on a slip of paper that has been pasted onto the page (although a keen eye might notice that the lines are not quite square with the rest of the page). If all you care about is text, that’s fine, but you’re missing a key part of the book’s history. And what happens if you were looking at a copy that didn’t have the errata slip (not all copies did)? You might never know it was an option–see the later section, “a copy is not an edition” for more on that.
My larger point here is that while digitization could convey some aspects of this category of information, they generally do not.
The history of the book’s use
The best thing about old books, I think, is their longevity and the traces of the history that they carry with them. Inscriptions, marginalia, doodles, vandalism, erasures, cutting out images and leaves–none of those are captured if your focus is solely on the text, and all of them have something to tell us about how a book was used.
(This is an image of one of the Folger’s three copies of Sacro Bosco’s Sphaera Mundi; this one happens to be heavily marked up, with additional diagrams penned in and plenty of annotations, but the other two are significantly less annotated. It looks yellow, by the way, because I took the photo myself, sans flash, per Folger requirements. There’s not a whole lot of light in the Old Reading Room, and I did a sort of half-hearted job of color-correcting my picture.)
Here’s a detail from a blank leaf from the middle of a 1550 Chaucer Works that’s covered in marginalia. I count at least four hands in this picture, three sixteenth-century and one twentieth-century. There’s other marginalia in this book, too:
So add two more hands (this time a seventeenth-century and a twentieth-century one that is mistaken about the date) to the collection of readers who have left their traces in this book. And then there’s this, from the same volume:
So that’s one more inscriber (although he doesn’t appear to have been an owner of this book). There’s also the cover of the book, with two more names incorporated into the binding, eighteenth-century descendants of Frances Wolfreston. I’ve written about this book and this collector before, so I won’t go on further here. But this kind of passage through history tells us not only about one particular family and one particular book, but gives a window into different responses to and uses of Chaucer’s poetry.
I suppose it’s possible that digitizing could capture this; if I can take pictures of these inscriptions, there’s no reason a digitization project couldn’t. Oh, except time and money. This is a big book of more than 300 pages, and only one of two copies we have of this imprint, and one of five copies of this edition, and one of I-don’t-know-how-many mid-sixteenth-century copies of Chaucer’s Works. Are we talking about a project that will digitize all pages, cover to cover, of all these books? Who’s going to fund that? Is it worth funding that as opposed to, say, funding digitizing all the books that were once owned by Ben Jonson?
The serendipity of finding the unexpected
This sounds a lot like the sort of paean to open stacks and browsing that you hear from some book fetishists. But I’m talking about something a bit more complicated, I think. When you look at a digital surrogate, someone has already made the decision for you about what you want to see and how you are going to use it. There’s no reason you can’t use the surrogate differently, but every choice they’ve made impacts your ability to circumvent it. For instance, some of the biggest digitization projects out there don’t include blank leaves or pages in their digitization. ECCO is the chief culprit that comes to mind: they prioritize text, and so their surrogates start with the frontispiece or title page, then move to the dedication/preface/letter to the reader/start of the text. But you know what’s missing? The blank verso of the title page. Does that matter? I don’t know. It might. It depends on what you’re looking for. But you’ll never know that you might be looking for an answer that depends on that blank presence if you don’t know that it’s not there.
Jeffrey Todd Knight has written about this serendipity in his most recent article, “Invisible Ink: A Note on Ghost Images in Early Modern Printed Books”. His focus is on ghost images left behind when some of the Pavier Quartos were bound together in early collections with non-Shakespearean works, so that an image of the title page of Heywood’s A Woman Killed with Kindness appears faintly on the verso of the last leaf of Henry V. Knight draws out some of the implications of what this means for the Pavier Quartos and our understanding of Shakespeare book history–go read the piece yourself–but he also makes a broader argument for the need to consider the invisible, reminding us that “it has been easy to forget that text reproduction technologies, at every level, carry biases”:
The onscreen interfaces that give us Shakespeare and Heywood’s plays today are not transparent windows onto the text themselves; they define and regulate a field of visibility, as do all forms of curation going back to the early copies, which also carried biases.
What I would emphasize is that we don’t know what we’re missing until it’s possible to see it. We can’t see the blank pages or the invisible ink if we’re experiencing a book through decisions that have already eliminated the possibility of seeing them.
A copy is not an edition
This is true for all books, but especially early modern books: no two copies are the same, whether through the process of how they were printed or their subsequent use by readers. The practice of stop-press changes being made at any stage of production, and at multiple stages of production, and the habit of mixing sheets so that “uncorrected” sheets can appear in the same copy as “corrected” sheets, means that any book that had changes made during the printing process will exist in different states. Nor are those states typically indicated on a book’s title page or (depending on how important a book is seen as being and how well cataloged it has been) in its bibliographic record. You can only find these variants by looking at multiple copies of a single edition–hence the brilliance of things like the Hinman collator, the Lindstrand Comparator, the McLeod Portable Collator, and Hailey’s Comet, all devices that allow their user to compare multiple copies of books without actually reading them. (Reading, as any copy editor will tell you, will only distract you from what you’re actually seeing on teh page; collators–or, as I like to think of them, the original textual/machine hacks–let you look instead of read.)
Why do we care about these textual variants? We might care because they tell us something about how the book was made, because it might say something about economic or societal pressures (if they were changes introduced in response to something other than correcting an error), or because the presence of and differences between states might help us understand the range of textual meaning available.
One example of the stop-press changes can be seen in the second quarto of Hamlet, in which Hamlet’s lines to Osric exist in three states, two of which can be compared in the Shakespeare Quartos Archive:
We are in no danger of missing the many textual variants of any of Shakespeare’s plays. But imagine if we were talking about, say, a Webster play, or a poem in praise of Queen Elizabeth. Those have not been as extensively, even fanatically, collated as Shakespeare’s works have. Nor have the funds been lavished on reproducing many digital surrogates of a single edition, as they have with Hamlet.
Remember above when I pointed out that some copies of The General History of Virginia have an errata slip, but others do not? If you encounter a digital surrogate of a copy that has the slip, you will treat that copy as if it represents the entire run of that edition, as if all copies of that edition of General History have errata slips. What is the characteristic of a single copy becomes a characteristic of the edition. But a copy is not an edition; what is unique to a copy is not common to the edition.
This list of possibilities has gone on long enough. Nearly everything I write about on this blog dwells on what we discover from looking at one copy of one book. If you’re curious to see more examples of what you might find from looking at physical books rather than digital surrogates, try some of the following posts in addition to the ones I link to above about blanks and Frances Wolfreston: “an armorial binding mystery“, focusing on overlapping book stamps; “essayes of a prentise“, about the significance of the binding on a volume of James I’s writings; “David and Goliath, redux“, about finding the same pattern of embroidered bindings; and “bibles for historical occasions“, in which I think about, um, bibles used on historical occasions.
I would love to hear more thoughts from you on this subject, particularly of lucid examples of what we can learn from looking at physical books beyond the sort of intangibility of emotions that books can evoke.