University of Minnesota
University of Minnesota researchers are employing technology and the discerning eyes of tens of thousands of volunteers around the world to decipher texts salvaged from that ancient trash pile.
Translating ancient documents
By Kirsten Weir
Two thousand years ago (give or take a few), a resident of Oxyrhynchus tossed a piece of papyrus onto the town's trash heap. There it lay, parched by the Egyptian climate, preserved for posterity.
Now, University of Minnesota researchers are employing technology and the discerning eyes of tens of thousands of volunteers around the world to decipher texts salvaged from that ancient trash pile.
The modern chapter of this exceedingly long story began in 1896 when British archaeologists discovered the Oxyrhynchus rubbish mounds. The find was at first unimpressive—then dazzling. It included some of the earliest copies of the New Testament, fragments of the Gospel of Thomas and other non-canonical Christian and Jewish theological writings, poems of Pindar and fragments from Sappho, parts of lost plays of Sophocles, the oldest diagrams of Euclid's Elements, a life of Euripides...as well as private letters, business contracts, tax documents, census returns, even grocery receipts for dates and olives.
"It's every kind of writing you can imagine," says Nita Krevans, a professor in the CLA's Department of Classical and Near-Eastern Studies and the project’s co-principal investigator along with Philip Sellew, whose expertise includes early Greek and Coptic Christian texts that have been preserved on papyrus.
The documents may be mostly small fragments, but they are keys to vast untapped knowledge about Egyptian life. Some date from 300 BCE, but most were penned during the first and second centuries CE; they were written primarily in Ancient Greek, Egypt's official language after Alexander's conquest in 332 BCE.
So this is a story of how a city dump turned out to be an unequalled archive of ancient life and times. Of how it yielded comprehensive records of a large and prosperous city that today lies buried under the modern town of el-Bahnasa, and of writings by some of the ancient world's greatest artists, scholars, and religious writers. And of how modern-day CLA scholars are part of this historic exploration.
A staggering task
After a fair bit of digging it became apparent that the very richness of the find presented a major problem. The fragments number around a half million; many are faded and torn, the antique ink abraded. In more than a century since they were discovered, only about 1 percent have been transcribed and published. While modern scholars are certainly able to read the Greek texts, even sifting through the mounds is a challenge of staggering proportion.
But a new project, Ancient Lives, is speeding up that glacial pace. It's an international, interdisciplinary collaboration involving the Egypt Exploration Society, which owns the Oxyrhynchus papyri collection; Oxford University Department of Physics, which stores it; and two U of M colleges—CLA via the Department of Classical and Near-Eastern Studies, and the College of Science and Engineering, which are developing technology to help translate it.
On the Ancient Lives website you can find images of hundreds of thousands of the fragments and an invitation to transcribe them by matching handwritten letters to the Greek characters that appear in a key at the bottom of the screen.
"We're basically asking volunteers to speed up the transcription process," says Marco Perale, a CLA papyrologist (papyrus expert) and postdoctoral researcher.
Ancient Lives grew out of Galaxy Zoo, a project launched in 2007 to recruit amateur science enthusiasts to help identify galaxies from images posted on the website. Lucy Fortson, associate professor of physics and astronomy in the College of Science and Engineering, was involved with that project from its early days. Volunteers—there are already 120,000 of them, pore over the online papyrus images, matching individual letters to the provided set of Ancient Greek characters. "The large majority are amateurs," Fortson says. "Many don't even read Greek. It's a pattern-matching exercise—you just match the shapes."
Behind the scenes, Fortson and Anne-Francoise Lamblin from the U of M's Minnesota Supercomputing Institute are developing software to analyze the volunteers' findings and create a master transcription based on the most common responses from each volunteer transcriber.
Volunteer transcribers are doing an impressive job, producing transcriptions that agree with experts about 80 percent of the time. Smart as the software may be, however, it by no means replaces classics scholars, so Marco Perale, a CLA papyrus expert and postdoctoral researcher, and his counterparts in Oxford review the consensus transcriptions, translate the text, interpret it, and determine which scraps are worthy of publication.
The project is fast gaining fans. When Theresa Chresand, a sophomore Greek major, learned about it, she got hooked, now spends a lot of her free time on Ancient Lives, and has even recruited friends to join her. "I had no idea what papyrology was until I got involved in the project," she says. Now she's considering it as a career option.
Get in on the fun—go to the Ancient Lives website.
This story was adapted from Reach magazine, a publication of CLA.
Kirsten Weir is a Minneapolis-based science writer.