St. Catherine’s Monastery occupies a spot at the base of Mount Sinai where Moses is believed to have encountered the burning bush. We know monks have been there since at least the late 4th century, since the peripatetic Egeria noted its existence in the early 380s, but it’s possible monastic life there stretches back even further. So: really old.
St. Catherine’s has a massive store of ancient manuscripts, second in size and scope only to the Vatican’s. For the last decade, St. Catherine’s head librarian Fr. Justin (a Texan) has been digitizing the collection in order to preserve it and make it available to scholars.
Photographing manuscripts is time consuming but not particularly difficult. More tantalizing, however, are the palimpsests. Since parchment was durable and hard to come by, it was common practice to scrape the text from a manuscript and reuse the now-blank surface for new texts. These scraped texts are called palimpsests, and every paleographer worth the title has wondered what was lost in the scraping and over-writing process. Maybe that second book of the Poetics?
It turns out that some treasures are indeed locked away beneath medieval texts, and now we have the technology capable of retrieving some of them. The process proved its worth when Michael Toth, formerly policy director at the National Reconnaissance Office, saw the potential to use modern imaging techniques to uncover hidden text. He assembled a team and offered to analyze a medieval prayer book that had been written over the earliest known manuscript of a work by Archimedes. Using digital imaging, the team was able to read the Archimedes Codex for the first time in 800 years, revealing texts of Balancing Planes, On Floating Bodies, The Method of Mechanical Theorems, and the Stomachion.
Toth’s latest project is much larger. St. Catherine’s holds about 150 medieval palimpsests, and Toth’s team is going through the slow process of imaging all of them. Here’s how it works:
The palimpsest imaging system exploits three basic strategies. LEDs on tripods in each corner of the camera room can bathe a manuscript page in light of a specific color or wavelength range, from ultraviolet to infrared. Each color interacts with ink and paper in different ways, allowing the camera to capture a series of slightly different images.
Next there’s back and side lighting. Sometimes the undertext ink is gone, but infinitesimal grooves remain where the ink ate into the animal hide. These etchings can be illuminated because the grooved parchment is thinner, allowing more backlight to shine through. This technique had never been applied to palimsests. The side lighting works similarly, creating tiny shadows in grooves and irregularities on the pages.
Finally, there’s fluoresence — a favorite of both real and TV crime scene investigators. Whether dealing with blood at a crime scene or parchment, fluorescence works the same way. When light of some wavelengths hits certain organic materials, their molecules absorb the light, then reemit it at a different wavelength. Filters bring out the resulting glow. With manuscripts, the organic material is the animal-skin parchment. Ink blocks some of its fluorescence, making it appear visibly darker in photographs.
“Probably nothing we do is unique,” says Bill Christens-Barry, the team’s electronics guru, who works as an independent contractor based outside Baltimore. “But we’ve found ways of optimizing each of these techniques.”
Under some of these imaging conditions, undertext shows up more prominently. Computer programs essentially subtract the difference between images where both types of text are prominent and those where mainly overtext is visible. The differences are converted into color: In some processes, undertext becomes remarkably legible in an artificial red, and overtext is suppressed in a muted gray. If that technique doesn’t work, the team can perform more complicated analyses and digital manipulations to bring out the text.
We don’t yet know what the team has found: they just create the images, while scholars around the world sort out the words. Even if there are no long-lost text, the patterns of palimpsest usage–what was scraped and what was saved–will reveal attitudes towards texts and preservation of knowledge.
In other paleography news, I have a follow-up to my story on The Oxyrhynchus Project. A newly deciphered manuscript from the Oxyrhynchus dump reveals a previously unknown poem to Nero’s wife, Poppaea Sabina. The manuscript is rather bizarre because it depicts a divinized Poppaea and was written in Egypt, in Greek, a couple centuries after her death, when the reputation of both of them as bloodthirsty lunatics was quite thoroughly cemented. It’s quite possible that the original text dates to Nero’s reign and was written by a court poet to dispel rumors that Nero kicked the pregnant Poppaea to death. It was subsequently copied and circulated not out of any love of Nero, but because people just liked the poem. In any case, it’s another remarkable find from Oxyrhynchus.