LEARNING ABOUT GREEK EPIGRAPHY
Broadly defined, epigraphy is about the study of ancient texts. Epigraphists might study vellum scrolls, painted hieroglyphs, inscribed steles, and any number of other materials which have survived to the present. They have been curating, studying, and publishing ancient texts, trying their best to understand what was originally written on them. It takes a long time to learn an ancient language, so epigraphists usually focus on a specific area and culture.
DEVA is contributing to Greek epigraphy, where most texts happen to be carved into stone. The Athenian Agora was an ideal place to start the project due to its historic importance and rich collection of inscriptions. DEVA’s methods should work well with any Greek texts carved into marble – not to mention the similar kinds of texts in other languages found worldwide, as well as any kind of inscribed surface.
ACCESSING ANCIENT GREEK TEXTS
Curating and studying ancient texts is a complex process. There many different media involved, and collections are scattered over the world. Texts could be carved into a marble stele just as well as an outcrop of bedrock in a mountainous countryside. Keeping track of their locations – in museums, at archaeological sites, or in the outside world – is an important part of an epigraphist’s job.
Anyone studying an ancient text would like to be able to look up its contents and search for comparable documents quickly. Epigraphy has developed various kinds of specialized libraries where students and scholars can access transcriptions of ancient texts. Greek texts have been compiled and published in reference series like the Inscriptiones Graecae (IG), one of the most important and oldest series established in 1815 (under a previous name), which catalogues texts by region.
Many of these series are not very user-friendly. The IG is published as a series of big, expensive tomes that are not widely available outside of academic libraries. Even if you get your hands on a volume, the commentaries were until recently written in Latin, requiring an education in classical philology in order to understand them. The IG does not include every Greek inscription, either. Many other series index and publish about ancient texts. The Supplementum Epigraphicum Graecum (SEG) compiles the many different publications about Greek epigraphy as they come out each year.
From IG I2 (1924), p. 148
DOCUMENTATION METHODS IN GREEK EPIGRAPHY
In practice, these textual publications are simplifying a real, three-dimensional object into a string of letters. At the Agora, we are working with stones whose surfaces have incised or painted markings that spell out the underlying text. The versions printed in an IG volume or the PHI start with the standard alphabetic characters for ancient Greek, but the ancient stone has more to it than just that text.
For example, there could be misspellings or signs of erasures and later changes that should be pointed out in the transcription. It is important to show where the text appears on the stone, especially since ancient texts from thousands of years ago are often damaged and incomplete. Modern transcriptions use special symbols called an editorial sigla to indicate these additional features. The Leiden conventions are commonly used for Greek epigraphy to show these kinds of attributes through dotted letters and various styles of bracketing.
From: Roehl 1894, Imagines inscriptionum Graecarum antiquissimarum / Guarducci 1950, Inscriptiones Creticae IV.141
See Roehl, H. 1894. Imagines inscriptionum Graecarum antiquissimarum, Berlin, and Guarducci, M. 1950. Inscriptiones Creticae, vol. 4, Tituli Gortynii, no. 141
If we want to review or improve a transcription, we need more information. With the time and resources, we could theoretically travel to the site or museum and study the stone in person. The best way to get a better understanding is to bring back some kind of copy, both as evidence to support our interpretation of the stone and to help others to look for new features that we might have missed. There are several ways to record inscriptions, including photography, drawings, and the “squeeze.”
Photography can reproduce the surface of a stone, though it can be hard to see a feature until the light is oriented in a particular direction. Drawings are another common method for showing the lettering in its original context, though they are still interpretations whose accuracy and detail depend on the illustrator. The “squeeze” has a long history for Greek epigraphy and involves a piece of paper or latex moulded to the inscribed surface. This replica can then be taken to an epigraphical library, though a new squeeze needs to be made for every collection.
The Agora stones have seen many a squeeze. After working on many Agora stones, the DEVA team has gotten used to finding traces of this technique – like hairs left behind from the brush used to mould the paper onto the stone.
DIGITAL EPIGRAPHY
The digital age has had a major impact on epigraphy. Nowadays, collections of ancient texts are easy to find online. Of the many epigraphy resources, it is worth mentioning the Searchable Greek Inscriptions from the Packard Humanities Institute (PHI), which has digitally transcribed inscriptions from a huge number of print volumes. The Attic Inscriptions Online is another noteworthy resource that includes commentary and links to learn more about many ancient texts from the Agora and beyond.
These kinds of sites allow you to browse and search transcriptions of texts, not recordings of the stones. Epigraphical fields that work with glyphs have been pushing the development of more visual resources. For example, Egyptologists or researchers of ancient Maya hieroglyphics, have many reasons to capture more than just the underlying texts and have built collections of imagery online. Similar kinds of image libraries have been designed for Greek and Latin texts, especially through scanning collections of paper squeezes in epigraphical libraries. Some examples include the McGregor Squeeze Collection shared by The University of British Columbia libraries, or Krateros at the Princeton Institute of Advanced Study.
Computers offer more possibilities than simply sharing scanned photographs and drawings of ancient inscriptions. Many epigraphists have explored special photographic and computational techniques to create a kind of “digital squeeze.” A popular method called Reflectance Transformation Imaging, or RTI, starts with dozens of photos taken from the same position with the lighting changed to different angles. Using software like Relight, we can compute a model of the surface that can simulate lighting from any direction. The results are interactive, something like a virtual flashlight that can help find subtle incisions on an inscription. The technique has been widely used since it was introduced in the early 2000s for museum objects and rock art, though it has not been been as popular yet for Greek epigraphy. Multi-spectral photography can bring out hidden features, especially paper documents like those examined by the Livingstone project.
3D MODELLING AND PHOTOGRAMMETRY
Another new technology with great potential for epigraphical study is 3D scanning. A 3D model can be dynamically illuminated, much like an RTI, but it can also be rotated to other points of view (an RTI is fixed to one viewpoint). The mesh can also be reshaded using nonphotorealistic methods that can clarify inscribed reliefs. These approaches been very powerful for studying cuneiform tablets, whose wedge-shaped impressions on a rounded clay slab can be tricky to show on a flat page. The Stanford Cuneiform Tablet Visualization project published exciting early results in 2002. Since then, these techniques have become even more powerful with software like GigaMesh.
But there are plenty of challenges to creating the 3D scans. Laser scanners are designed to operate at a specific size and resolution, and they usually work best with non-reflective surfaces. Translucent materials can disrupt the scanning process by scattering the light in unpredictable ways. This has been a major obstacle to digitizing Greek inscriptions, especially in Athens where inscriptions are often carved into marble. The transparent crystals in marble can degrade the quality of the scan.
A relatively new arrival is photogrammetry, a technique to measure 3D points by comparing photographs taken from different vantage points. Thanks to new computational techniques, the latest programs can now automatically extract millions of 3D points from photos, which is enough to create a detailed 3D model with realistic textures. Photogrammetry is more flexible than laser scanning as well. It can work with any camera and model 3D features from tiny details to whole landscapes.
DEVA has been using photogrammetry to great effect on Agora stones, including pieces that would frustrate most laser-scanning equipment. We have been able to uncover new features on many inscriptions using techniques similar to those used in GigaMesh, RTI, and similar approaches, but we have adapted them to the unique challenges posed by Agora stones.
FURTHER READING
On epigraphic methods:
- McLean, B.H.. 2002. An Introduction to Greek Epigraphy of the Hellenistic and Roman Periods from Alexander the Great down to the Reign of Constantine (323 B.C. – A.D. 337) (University of Michigan Press).
On some of the technologies in GigaMesh:
- Mara, H., and S. Krömker. 2017. “Visual Computing for Archaeological Artifacts with Integral Invariant Filters in 3D,” in Eurographics Workshop on Graphics and Cultural Heritage, GCH 2017 (The Eurographics Association), 37–47.
About photogrammetry in archaeology and epigraphy:
- Magnani, M., M. Douglass, W. Schroder, J. Reeves, and D.R. Braun. 2020. “The Digital Revolution to Come: Photogrammetry in Archaeological Practice” American Antiquity 85.4: 737–60.
- Ravanelli, R., L. Lastilla, and S. Ferrara. 2022. “A high-resolution photogrammetric workflow based on focus stacking for the 3D modeling of small Aegean inscriptions” Journal of Cultural Heritage 54: 130–45.
- Sapirstein, P. 2018. “A high-precision photogrammetric recording system for small artifacts” Journal of Cultural Heritage 31: 33–45.
