In his latest Info World column “Filling in the Margins,” Jon Udell writes:
Xerox’s Palo Alto Research Center (PARC) alumnus Austin Henderson says that “one of the most brilliant inventions of the paper bureaucracy was the idea of the margin.” There was always space for unofficial data, which traveled with the official data, and everybody knew about the relationship between the two.
As Udell makes clear, he’s paraphrasing the forthcoming research of Austin Henderson, and it’s an interesting comment. It’s not, however, quite accurate.
Yes, marginal glosses are used in medieval (and earlier—think Egyptian papyri) manuscripts just that way. But the “everybody knew about the relationship,” well, no, they didn’t, and no, we don’t. Those marginal comments, or glosses, were used by readers to make reflective annotations, to add reference material by other authorities, to make corrections, or even to doodle. Scribes also used margins to make corrections, or sometimes, just comments. And of course marginal comments weren’t restricted to the right and left margins, or even the top and bottom—readers made interlinear comments too, like these in the Book of Lindisfarne which provide Old English translations of the Latin Gospel. And then for certain texts, and classes of texts, the glosses were soon seen as a sort of textual appendix, one that was right on the page with the main text (what, you didn’t know hypertext was a manuscript tradition? Think of the Talmud.) Remember that manuscripts were copied by hand, often by professional scribes, in or out of the monastary, and often by private individuals. Professional or not, scribes get tired, and hungry, and have trouble with the light, and often, are copying texts in languages they can’t read. And sometimes, a scribe didn’t realize a marginal gloss was a gloss, and so the gloss was incorporated into the main text. OK, a lot of times— it happens so regularly that it’s a field of paleographic specialization. Once a gloss is incorporated into the body of a text, it’s frequently transmitted, so the error perpetuates, and even propagates.
My point, which I realize is somewhat divorced from John Udell’s context, is that as we work out semantic data and metadata and document standards, we need a way to do “digital marginalia” so that meta data identifying marginalia travels with it, because it’s a real pain comparing versions of an ancient text in an effort to determine whether text that appears corrupt is in fact part of the text, or a scribal error of addition. I don’t even want to think about doing that with a digital record.