Papyrus: Expert Q&A
On November 28, 2006, papyrologist Roger Macfarlane answered questions about the Oxyrhynchus papyri, the NASA-developed multispectral imaging being used to decipher them, and more.
Q: I recently attended a lecture in California given by a professor at Stanford. He is using high-energy X-rays to read faint or erased medieval texts written on parchment. The method depends on the ink having iron content. I was wondering if that method might have any application to the work that you are doing? Ken Posse, Fort Collins, Colorado
Roger Macfarlane: There is a fundamental difference between the ink used on most papyri we deal with and the ink found on medieval manuscripts. The earlier inks are nearly always made of carbon, which is charred vegetable matter, suspended in liquid such as gum arabic. The stickiness of the gum allows the carbon-based pigment to stay in place until the liquid evaporates or is absorbed into the medium upon which the ink is applied. In late antiquity, iron-based inks replaced carbon-based inks. These later inks are chemical compounds based upon iron gall or iron oxide. They are applied in the same way, but the traces that are left behind are actually caused by oxidation.
Carbon-based inks respond differently than the iron-based inks do. In our multispectral imaging work, we rely on the principle that the carbon always remains black no matter what wavelength of the light spectrum one views the traces in. This is true because the carbon absorbs light very effectively. Certain matter that may obscure the ink to the naked eye, such as an overlying pigment or thin layers of gesso plaster, does so because it reflects light. If such an overlayment has a spectral window through which light can pass, we can sometimes find the right wavelength in which the light passes to the background ink and reflects back through the overlayment. The result is still the same to the naked eye, but the filtered reflection can cast an image that the camera can discern. The best results usually come in the infrared sector of the light spectrum when we deal with carbon-based inks.
Since medieval texts tend to be written in iron-based ink, one standard method of reading erased texts, or "palimpsests," is to use ultraviolet light, because the traces of the iron-based ink tend to fluoresce in those conditions. The work of the Brigham Young University team has concentrated on papyrological texts, although we have the relatively simple equipment necessary to read later texts as well. In fact, we are working with one palimpsest papyrus text—a rarity, but not unheard of—and have found success using our infrared technology
Q: How does the ancient process of papyrus creation differ from today's modern paper-making?
Scott Marlowe, Dallas, Texas
Macfarlane: Papyrus-manufacture in ancient Egypt was a sufficiently hearty industry that in many periods of antiquity papyrus was relatively inexpensive. It was cheap enough that used papyrus was the basis for the sort of papier-mache called "cartonnage" that wraps mummies. And papyrus was cheap enough that papyrus was used as filling inside mummified corpses of crocodiles at Tebtunis. Those who put papyrus to these funerary uses paid no attention whatever to the content of the texts on the papyrus they chose. I act the same when I cover my kitchen table with newspaper before carving my kids' jack-o'-lantern pumpkins. Conversely, papyrus was sufficiently dear that even well-to-do users would sometimes turn over a roll of previously written papyrus and write on the back. Many documentary texts, such as police reports or imperial decrees, are written on the so-called "back side" of papyrus pieces that had been used for literary texts much earlier. Also, there are known instances in which users would wash away the text on the surface of the papyrus they wanted to reuse. Such palimpsests are much more uncommon than the medieval practice of scrubbing a text off a vellum sheet for reuse, because papyrus was relatively less expensive.
Another factor to consider in answer the question of papyrus' relative cost is the common discovery of other writing media, including texts preserved on pottery shards (ostraca), on textile swatches, on metal (i.e., lead or silver plate), or on wooden tablets. Papyrus texts tend to be more common than any of these other media, and it may be fair to say that papyri tend to have, collectively, a wider range of uses than any other portable writing medium.
If you have a chance to see the NOVA scienceNOW segment about our work again, you should note Joshua Sosin's really clear description of papyrus-manufacture. Pliny the Elder discusses the process in his 1st century treatise, the Natural History. Stripped fibers of the papyrus reed are laid in parallel on one layer, and then on top of that is placed a layer at perpendicular angle. The cell structure of the two fibrous layers is broken by pressure, and the gluey material in the cells causes the two layers of fibers to cohere into a unified sheet. After drying, the resulting sheet of papyrus can then be trimmed, polished, and written upon. Papyrus is remarkable for its sturdiness and flexibility. If treated right, it holds up well to the passage of time. Without care, it deteriorates and crumbles. For what it's worth, Eric G. Turner, the late British dean of papyrology, noted that "it is salutary for modern technologists to note that the oldest papyrus is also the best made."
Q: I am concerned about how we choose to interpret some of the more fragmentary documents. For instance, one note is described as a list of items for a sacrifice, but, without context, how do we know this was not simply a grocery list? Brian C., Pittsburgh, Pennsylvania
Macfarlane: Your skepticism is well placed. There are grocery lists, to be sure, among our papyrus hordes. But there are more clues in a papyrus fragment than a neophyte will sometimes appreciate. Watching an experienced papyrologist at work can be as fascinating as watching a forensic detective work through a crime scene. Indeed, the shape of the letters on a text, the frequency of abbreviations, the context in which the fragment was found—all these clues must be taken into account by the papyrologist who encounters a text.
Papyrology is among the most precise of the disciplines included within Classical Studies. Papyrologists tend to be very meticulous, because they have to consider so many options and because they must avoid jumping to conclusions. And papyrologists tend to be among the most collegial of all classicists, benefactors of that famous amicitia papyrologorum, "friendship among papyrologists," because they must frequently rely on the forbearance of their peers in forming and publishing their findings.
Q: Could this infrared technique be used for advanced medical imaging? Diana Foster, Potomac, Maryland
Macfarlane: The particular type of multispectral imaging that we use for reading damaged papyri may not yield results in medical applications. But in the world of papyrology, we are pleased by a nascent application of CAT scanning to papyrus rolls. Dr. Brent Seales at the University of Kentucky is developing the technology of radio-tomography for reading papyrus scrolls before they are unrolled. Early application of this technology enables Dr. Seales to read some texts on underlying layers of text without removing the upper layers.
The papyrological imaging team at Brigham Young University can read through some overlying materials in some instances. And we are developing other technologies for hopefully achieving even better results more consistently.
Q: What environmental conditions do papyrologists need to prevent the papyrus from decaying? Kirk Daniel, Rockville, Maryland
Macfarlane: Dampness is papyrus' worst enemy. The hot, dry sands of Egypt are actually the best environment for preventing papyrus from decaying. The rubbish mounds of Oxyrhynchus offered many layers of textual preservation for archeologists to excavate. The lowest layers, which would represent the Ptolemaic Era, were not at all well preserved when Grenfell and Hunt began excavating in the late 19th Century. The oldest layers were subjected to water rising from below, and the newest, uppermost levels suffered the ravages of weathering.
Even though papyrus was much used in ancient Greece and Rome, only the papyri that have been preserved in cataclysmic circumstances—such as the carbonized library at Herculaneum that was entombed in the pyroclastic surges of Vesuvius' eruption in A.D. 79, or the Derveni Papyrus that was accidentally spared from the funeral pyre by a muddy skin that formed around it and became baked like terracotta—survive the vicissitudes of time. Indeed, had the Herculaneum library been kept safe on a shelf in the humidity of the southern Italian climate, the papyrus fabric of the scrolls would have crumbled many centuries before their carbonized remains were discovered in 1752.
Q: Could multispectral imaging be dangerous to papyrus, just as flash photography is dangerous to old documents like the Constitution? Kirk Daniel, Rockville, Maryland
Macfarlane: The danger of flash photography is, of course, the sudden pumping of high-energy light onto the surface of the document. To obviate the effects of lighting, in our work with the multispectral imaging apparatus, we use special lights that have a high color temperature but a low thermal temperature. The lights, in other words, replicate the color spectrum of the sun—with the full range of color that you get outside on a clear day; however, they are very cool and do not raise the surface temperature of the papyrus we are imaging.
We are sensitive to the risks associated with pouring light onto the surface of a document. Therefore, we are always careful to demonstrate to curators and librarians that our equipment does not raise the surface temperature of their precious documents at all. We use remote-sensory thermometers to monitor the temperature on the surface of the document.
Some of our most important work has been done with the Herculaneum Papyri, which are carbonized papyri—black background with black ink. Even when it was necessary to keep one papyrus under the lights for as much as 30 or 45 minutes, we could never monitor more than a 1-degree Fahrenheit rise in temperature vs. ambient temperature.
Q: If papyrus is more durable than paper, and we can clean some old paper documents, then why can't we clean some of the papyrus? Cristin Coleman, Jackson, Mississippi
Macfarlane: It would not be right to say that we cannot clean papyrus. Careless attempts, however, in the past have wreaked many disastrous effects on unique texts. If a non-invasive process can be developed whereby reliable readings of damaged texts can be obtained, then curators will be greatly relieved.
Among the Tebtunis Papyri there is a series of fragments that preserve portions of Sophocles' play called Inachos, which is otherwise not extant. Among these portions of Sophocles' text, there is one fragment that is still covered by a layer of the gesso plaster that made the piece cohere in the mummy cartonnage but that obscures the underlying text. The curator of the Tebtunis collection could pay the high price of conservation and have the plaster removed from the papyrus surface, thereby running the risk of damaging the text underneath the plaster. Our multispectral-imaging apparatus, however, proved capable of capturing an image of the Demotic Greek documentary text that lies under that plaster. Some papyrologists who favor the study of documentary papyri may favor the removal of the plaster from this one fragment, so as ascertain with surety what is written beneath. Those who want more Sophocles will not want to touch this papyrus again.
Attempts over the years to remove surface film and dirt from above papyrological texts have, of course, not always been disastrous. However, the perfection of non-invasive technologies will always win the hearts of those who want to assure that any given text may be passed on to the next generation.
Q: Are papyri from locations other than Oxyrhynchus being imaged? What are the most surprising documents we've found so far from any location? Dawn Campbell, Youngsville, North Carolina
Macfarlane: The NOVA scienceNOW footage that was run in the November 2006 report includes material shot while the Brigham Young University team was working with the Tebtunis Papyri that are archived at the Center for the Tebtunis Papyri in the Bancroft Library at the University of California, Berkeley. Work with the Tebtunis Papyri resulted in some of our most exciting discoveries to date.
We have also worked with papyri and ostraca (inscribed ceramic shards) in the Petrie Museum of Egyptian Archaeology at University College London, and with many other papyrological and archeological texts. The team's work with the Herculaneum Papyri, carbonized papyri rescued from Vesuvius' volcanic matrix, is perhaps our most far-reaching and most satisfying. Herculaneum texts are in the British Library, Oxford's Bodleian Library, and the Institut de France in Paris. The Derveni Papyrus, a national treasure archived at Thessaloniki, Greece, is an artifact imaged by our team. Even the small papyrological collection in the Brigham Young University Library has papyri that have responded well to our technological application.
Surely the most surprising documents imaged so far have been the Herculaneum Papyri, because their obscured texts—black, carbon-based ink on blackened, carbonized papyrus background—have responded very well to the application of our technology. We anticipated that we would do well with this technology, but nobody really knew how well. That surprising result has been very gratifying.
Q: Would it be possible to illuminate the documents with infrared light in a dark box and then use an ordinary digital camera to photograph them? This could be very inexpensive and, if it worked, many more documents could be processed. Tom Roderick, Lithonia, Georgia
Macfarlane: In our hardware setup, ambient light makes no real difference. We add lighting (that includes the wavelengths of the infrared in its synthetic spectrum) and capture the reflection of that light in various wavelengths. The lights themselves cost very little.
We could use a lower-grade camera to save money. The one we use is manufactured to scientific-grade specifications, allowing less than one flawed pixel per million.
But the expense of our system comes in the logarithms that run the camera/computer interface and allow the computer to stretch the grey-scale in such a way that the projected image is able to read more readily than the black-to-grey original.
Q: What about the multispectral imaging process makes it so time-consuming? Why can't we go through all of the remaining fragments in the span of a few months? J. Johnson, Oxford, Mississippi
Macfarlane: One time-consuming factor in our work with the Oxyrhynchus papyri is associated with the fact that we are in some sense still developing our technology. It interests us that a given fragment may respond differently from other fragments. We are still trying to figure out why. Further, we are still collecting data in a scientific way, i.e., still working with the possibility that we might not necessarily discern the value of one particular reading, one that appears better in other circumstances.
The largest project we have done to date is the multispectral digitization of the entire corpus of Herculaneum Papyri, a project that involved the recording of some 50,000 unique images over several months. In that work, we determined that one filter would usually give us the best results. And we stuck with that one filter, unless it appeared that a different wavelength would give us better results. Then we changed to another filter temporarily.
The other end of the spectrum is represented by our most technically demanding project to date, the carbonized Derveni Papyrus. In that work we have shot two complete series of 15 filters, spanning the wavelengths from 400 to 1,000 nanometers, altering shutter speeds, apertures, lighting angles, and so forth.
With the Oxyrhynchus collection, our most daunting challenge will be the vastness of the collection, with its roughly 400,000 individual papyrus pieces. The size of the collection is incomprehensible. To accelerate our approach to it, we are working with the development of new equipment. In the meantime, though, we continue to crawl through in a scientific—and not unsatisfying—rate of return.
Q: How can interested persons help experts get through all of that papyrus? Are there opportunities for students to get involved? Joyce Greco, Rollins College, Ocoee, Florida
Macfarlane: It is an overstatement to say that papyrologists are a dying breed. It is, however, no exaggeration to say that the specializations that combine to make a good papyrologist are failing with time. As more resources go into the development of transitory or emerging disciplines, there seems to be less for the classical disciplines.
Study of the papyri from Herculaneum requires a thorough knowledge of classical Greek. If today's universities train no students in the rigors of ancient languages, then tomorrow's discoveries cannot occur.
In the past, access to papyrological texts has also presented an obstacle keeping "interested persons" from developing a widespread interest in the fascinating contents of texts from Egypt or Europe. The National Endowment for the Humanities has funded quite generously for the past decade two separate papyrological initiatives, the Philodemus Project and the Advanced Papyrological Information System (APIS). While the Philodemus Project is inherently esoteric, several fundamental studies of the ancient philosopher Philodemus have come to light. APIS is different in opening wide previously impassable restrictions that kept us from peering in at the content of the papyri. Via APIS, libraries and museums across the world are now uploading copious metadata and digitized images of literary and documentary papyri, and these data are becoming accessible to the general public. As time passes, more institutions are joining APIS, even as more engaging computer interfaces are being designed for the use of scholars and laypeople as well. Many of the APIS texts have Greek transcriptions and many have English translations as well.
Fostering an atmosphere in which classical languages and classical culture are valued will continue to effect progress with the vast stores of information still lurking unread in the texts from ancient civilizations.