Recycling is great and all, but how would you feel if you found out that, in the name of thrift, someone had recycled all of Albert Einstein’s notes on the theory of relativity, causing them to be lost forever?

Here’s a similar situation. Spoiler alert: It has a happy ending!

After our piece on the scanning effort underway in the Vatican Library, several people told us we should check out the Archimedes Palimpsest.

If you’re not up on your ancient Greek scientists, you may think you’ve never heard of Archimedes, but chances are, you have. First of all, he’s the guy who supposedly shouted “Eureka!” (Greek for “I have found it!”) when he discovered the principle of displacement of fluid. The best part is that he supposedly discovered this in the bathtub (which is where we get all our best ideas), and was so excited about it that he ran down the street naked. Second of all, he’s the guy who said, “Give me a lever and a place to stand and I will move the earth.”

A palimpsest is what you get when you take parchment (a prepared goatskin) that someone’s written on, scrape off most of the stuff that’s been written on it, and use the mostly blank parchment to write something new. Ancient recycling. But because only most of the stuff can be scraped off, you can still see traces of what had been written on it before.

The original Archimedes text was made during the 10th century A.D.(Archimedes himself lived during the 3rd century B.C.)  Exactly who made it, when, where, and why, we don’t know, but it was probably written in the second half of the tenth century and almost certainly at Constantinople, because it was the one place where ancient mathematics was studied and copied, researchers say.

But a couple of centuries later, someone decided that Archimedes was too geeky for anyone to want to read about (well, really, researchers seem to think that the piece was too technical to be appreciated by 12th century audiences), and it was subjected to the scraping procedure. As one of the researchers joked, “It wasn’t worth the parchment it was written on.” Someone went on to use its 87 parchments—43 ½ goats’ worth—for a prayer book.

In the early 20th century, researchers looking at a 12th-century prayer book saw there was a document underneath. Upon reading it, they thought it was about Archimedes, so they contacted researcher Johan Ludvig Heiberg, who specialized in Archimedes and Greek thought. Using only a magnifying glass, in 1906 he translated the previously written Archimedes material beneath the prayer book’s text. He also took photographs of the pages. But he couldn’t read about 10 to 20 percent of the text. Plus, he didn’t understand all of it.

Nonetheless, everybody was pretty excited—the news of a long-lost Archimedes document made the front page of the New York Times—but in the meantime, the palimpsest disappeared, perhaps during the breakup of the Ottoman Empire after World War I. It finally turned up in 1998, at a Christie’s auction, where it was sold for $2 million on behalf of a French family. The problem was, they hadn’t taken very good care of the manuscript; in addition to the candle wax dripped on it during its stint as a prayer book, since 1908 it had also gotten burned, wet, moldy, and repaired with Elmer’s wood glue. To add insult to injury, sometime after 1938, four of its pages had been painted over by a forger trying to make the book look as though it had Byzantine illustrations.

After its purchase at auction by an anonymous investor, things started to get really interesting. The palimpsest had become a much more challenging restoration project than when Heiberg tackled it in 1906. But modern researchers also have many more tools in their arsenal than a magnifying glass and Elmer’s wood glue. Earlier in the 19th and 20th centuries, chemical techniques that researchers used to read palimpsests often ended up damaging the manuscript in the long run. But now, researchers had a new technique: Multispectral imaging.

It took four years just to take the book apart carefully enough not to destroy what was left. After that, researchers took pictures of the pages with all sorts of lighting—ranging from X-rays to ultraviolet, 14 wavelengths in all—to make it easier to read the Archimedes writing. The ink for the Archimedes writing was made out of a different material from the prayer book text—it had a lot more iron in it—so it responded differently to different kinds of lighting. For example, in looking at one page with UV lighting, researchers actually found the word “Archimedes.”

Imaging the document was a challenge.  To read the text clearly, researchers needed a higher resolution than the equipment’s 300 dpi, and the various filters they were using to produce the different wavelengths of light made the letters blurry. To achieve their target of 800 dpi, researchers took multiple pictures of each part of the parchment, 5,400 in all, and then electronically “stitched” them together.  Filters on the light itself rather than through the cameras addressed the blurry letters problem. 

Getting through the 20th-century forged “illuminations” was another issue. Eventually, researchers used the Stanford Linear Accelerator to generate the X-ray wavelengths needed to read the text through the illuminations. Even with such a powerful light source, it took 17 minutes to scan a page. The upshot is that, over the 12-year process, a number of things were discovered about the Archimedes Palimpsest that hadn’t been discovered in 1906, including places where Heiberg got it wrong: diagrams that Heiberg didn’t include in his transcript of the piece, and passages that he’d left text out of his transcript.

The other useful part of this process is that restorers now know a lot more about how to make ancient writing more legible, even on a manuscript as abused as this one, and about how to define metadata for academic documents. For that reason, the Archimedes Palimpsest project received the "2003 Imaging Solution of the Year" from Advanced Imaging Magazine.

It also turned out that, in addition to the Archimedes manuscripts in this palimpsest, there was also one of the writings of Hyperides, an Athenian orator from the 4 th century B.C., and a 3rd century A.D. commentary on Aristotle. “Normally when you're looking at medieval manuscripts that have been scraped off, you don't find unique texts,” project director William Noel says in a 2012 TED talk. “And to find two in one manuscript is really something. To find three is completely weird. And we found three.”

He didn’t say whether they shouted “Eureka!” and ran naked down the street. 

Related Posts