fashionably late to the dating game
One tries to stay abreast of all the cool science stories wafting about the World Wide Web, but occasionally, to our great chagrin, we miss something that is so, like, utterly crucial. This is one of those days. Jen-Luc Piquant belatedly stumbled across a fascinating news item from June about a Penn State biologist who has drawn on his scientific expertise to develop a new method (based on a similar technique used to study genetic mutations) for dating manuscripts and artwork from several centuries ago. We aren't sure how this tasty tidbit escaped our notice, but it did, despite the Penn State press release and subsequent coverage in such regularly-perused venues as SEED and the Washington Post. Still, it's just too cool not to post something about it, so we prefer to think of this as coming fashionably late to an already hoppin' party.
The hero of today's bibliophilic tale is Blair Hedges, a professor of biology at Penn State University who specializes in evolutionary dynamics, but has a passionate side interest in old prints and maps, particularly from the Renaissance period. In fact, he's spent years scouring the Caribbean in search of ancient maps of the sea's many islands. Many of the maps he acquired were undated, their provenance a subject of much contention among experts. Dating manuscripts is traditionally a tricky business; antiquarians rely on analyzing fonts, type of paper, or printer's watermarks, among other features.
The problem with such an approach is that it's a bit too subjective and inexact, by scientific standards; sometimes the best even the top experts can do is make educated "guesstimates." Furthermore, truly gifted forgers, like that infamous murderous Mormon, Mark Hoffman (featured in the terrific book, The Poet and the Murderer), can produce convincing forgeries able to fool most experts. Hoffman produced his forged Oath of a Freeman -- a masterpiece in its own right, as forgeries go -- by acquiring old parchment paper from the desired time period, and making his own engraved plates. Even documented provenance can sometimes be cleverly faked.
Hedges decided there should be a better method, and set about developing one -- a research project he funded himself. He compares his new "print-clock" technique to the molecular clock techniques used to time genetic mutations over several generations, a key tool for dating genetic material in his biological research. In the case of book dating, he measured time-related changes (deterioration) in over 2500 Renaissance works, figuring that the changes were simply random errors in the printing devices, akin to the random errors in DNA that give rise to genetic mutation.
First, he took digital photographs of the texts or prints to be studied -- anything from an engraved book jacket, a biblical text, or an illustrated work like a 15th century map. Then he used image-analysis software and standard statistical methodology to analyze those images. The software specifically detects and counts breaks in the lines of woodblock prints, and can measure fading of the etched and engraved lines of copperplate prints. He published his findings in the June 21 issue of the esteemed Proceedings of the Royal Society.
For the method to work, Hedges had to figure out how to calibrate his "print clock." He discovered that the engraved wood blocks and copperplates used for printmaking deteriorate at a clock-like, constant rate over time, irrespective of print runs. Historians have long known that there are noticeable changes in print quality found in later editions of printed texts, which they have attributed to the wear and tear of the printing process itself, or the number of times an impression was made with a particular block or plate.
In contrast, Hedges says his analysis demonstrates that changes in print quality are caused by aging of the wood and copper materials that make up the plates during storage between print runs. Why is this helpful? Because the deterioration process is a measurable physical characteristic that he likens to the "random radioactive decay of geologic clocks and the random genetic mutations of molecular evolutionary clocks." Specifically, he found that the number of breaks in the lines of images printed from woodblock carvings increased over time, while the image intensity became more pale in copperplate prints. (A line break is a faded line in a drawing that may have been bolder in an earlier edition of the same text.)
To take advantage of this property, Hedges studied images with known print dates made with the same woodblock or copperplate; this gave him a reference point for determining how the line breaks and other print features deteriorated over time. In the case of woodcuts, it was Bordone's Isolario, an atlas with maps of the Caribbean islands. Three editions have known print dates: 1528, 1534, and 1547. Using the print-clock technique, Hedges was able to determine that his undated copy was printed in mid-February of 1565, give or take a year, so it would have been the fourth edition. For the copperplates, Hedges focused on two books of early maps of Jamaica, Cuba, and Hispaniola; one was printed in six editions over five decades, with the first appearing in 1572, while the other had three editions, the first having been run in 1596.
Now Hedges is chomping at the bit to apply his new technique to undated printings of Shakespeare's Hamlet and Romeo and Juliet, as well as undated prints by Rembrandt. And according to David Gants, an expert on early books at the University of New Brunswick in Canada, if Hedges' print-clock method turns out to be equally accurate for dating woodcuts from various times in the period where books were hand-pressed, "it will -- in one stroke -- render antiquated the existing strategies for dating books."
Pretty cool, huh? Nor is this the first time science has come to the aid of antique books and art, or even archaeological finds. All kinds of cutting-edge instrumentation has been brought to bear on such objects, often with fascinating results, from ultraviolet imaging to x-rays and synchrotron radiation. For instance, in 1999, a team of archaeologists from the University of Chicago's Oriental Institute submitted priceless artifacts recovered from a dig in the Amuq Valley in Turkey, to scientists at the Advanced Photon Source (APS) at Argonne National Laboratory for analysis. They wanted to know how certain artifacts had been made, but the only way they knew to do that was to take the relics apart -- something they were understandably loathe to do. But synchrotron radiation can analyze the internal structures of materials down to the atomic level without destroying or damaging them.
The APS is essentially a gigantic x-ray machine, featuring a ring about the size of four football fields. Inside that ring, electrons accelerate to the speed of light along the curved path, emitting electromagnetic radiation (a.k.a. "light") in various wavelengths ranging from x-rays to infrared. Different materials absorb and emit different kinds of light, which can be detected and analyzed to determine which are present in a given object.
Last year, SLAC made international headlines when it announced the results of an analysis of a priceless Archimedes manuscript (see image, above). The original 174-page manuscript described, among other insights, Archimedes' explanations of the pulley, the fulcrum and the lever, as well as his famed "Eureka!" insight into the physics of flotation that turned him into one of the world's earliest streakers. But the original Greek parchment scrolls were erased and re-used twice, once in the 10th century and again in the 12th, and re-inscribed with the text of a prayer book. It was common practice in that time period, but it did irrevocably alter the history of mathematics, since Archimedes' work was largely unknown during the Italian Renaissance as a result. The parchment was so badly damaged -- moldy, with charred edges -- that traditional imaging methods weren't effective for certain pages.
SLAC's Stanford Synchrotron Radiation Laboratory (SSRL) came to the rescue, revealing the long-obscured original text beneath the later inscriptions. It turns out that the inks used in all the relevant time periods contained iron pigment, and the SSRL can measure extremely small concentrations of iron in proteins. The facility's x-ray beam is tuned to the specific energy for iron, which causes the remaining traces of iron ink to fluoresce. This can be detected and rendered into a visible image, in a manner similar to old dot-matrix printers.
That's science for you: nailing down elusive dates for artifacts, making long-invisible inky inscriptionss visible again, and solving a few historical mysteries in the process. I'm sure we can look forward to even more "Eureka" moments as scientific progress marches on.