Humanities Computing

Archaeological Computing at Aphrodisias, Turkey

Christopher Ratté

[Ed: Links to web pages and/or e-mail addresses which have become inactive since the publication of this article have been enclosed in curly brackets { }. Replacement links have been provided where possible.]

Prospection has its limits, however, and in the end there is no substitute for digging, in time-honored fashion, by hand with the simplest of tools -- pick, shovel, trowel, spoon and needle. By the same token, no computer is as efficient as the eye and hand of an experienced draftsman, but computers allow archaeologists to store and manipulate images in ways that open up exciting new perspectives on our evidence. Computing technology plays an increasingly important role in archaeological excavations around the world, including New York University's long-term research project at Aphrodisias in Turkey.

The Site

Aphrodisias lies in southwestern Turkey, in a fertile valley 100 miles southeast of the port of Izmir. Famous for its sanctuary of Aphrodite, the city's patron goddess, Aphrodisias enjoyed a long and prosperous existence from the first century B.C. through the sixth century A.D. Today, many of the city's ancient monuments remain standing, and excavations have unearthed numerous fine marble statues and other artifacts. The great beauty and extraordinary preservation of this site combine to bring the civic culture of the Graeco-Roman world vividly to life.

The Santuary of Aphrodite, viewed looking east.

The first systematic excavations at the site were begun under the direction of the late Professor Kenan Erim of the Department of Classics at New York University in 1961. These excavations concentrated on the city's central monuments and on the recovery of sculpture, with spectacular results. The religious heart of the city was the Sanctuary of Aphrodite, situated on the northern edge of the central part of town. To the south lay two large public squares, which together made up the Agora, or civic and administrative center (see below). On the south side of the Agora, pendant to the Sanctuary of Aphrodite, is the Theater, a large open-air structure with an estimated seating capacity of 5,000.

All these buildings date to the late first century B.C. and first two centuries A.D., the most prosperous period in the history of the site, and they were richly adorned with well-preserved statues and relief sculptures. Later architectural projects include the fourth-century city walls, and the conversion of the Temple of Aphrodite into a Christian church. Like other provincial towns throughout western Asia Minor, Aphrodisias fell on hard times in the late sixth and the early seventh centuries. The town survived until the fourteenth century, when the site was finally abandoned.

The Council House on the north side of the Agora, viewed looking southeast

After Professor Erim's untimely death in 1990, a new program of research at Aphrodisias was begun under the sponsorship of the Institute of Fine Arts in cooperation with the Faculty of Arts and Science. The main goals of this program are to study, conserve, and publish the results of the earlier excavations, and to investigate, through extensive surveying and limited new excavation, the relationship between the excavated monuments and their larger urban context. The use of computers and computing technology -- unknown at Aphrodisias before the 1990s -- is a crucial element of this new research program.

Before discussing this program, it may be useful to give a brief description of the staffing and physical set-up of the dig. The field staff includes twenty-five to thirty students and specialists of various kinds, including conservators, architects and photographers, as well as archaeologists. About half of the staff are NYU faculty and graduate students, and about half come from other universities in the United States, England, France, Germany, Austria and Turkey. With such an international team, good methods of communication during the off-season are extremely important.

The colonnade of the Agora.

During the two-month summer field season, the dig occupies two large permanent building complexes on the site. In addition to sleeping quarters, a kitchen and other practical facilities, the camp includes a small library, two drafting rooms, two conservation laboratories, a darkroom, several study rooms and secure outdoor work areas, and a number of depots for permanent storage of archaeological finds. There is also a museum on the site, operated and staffed by the Turkish Department of Monuments and Museums. The site and museum are open to tourists all year long, but the dig camp is closed, and no excavation takes place during the off-season.

Computing at Aphrodisias

Back in New York, the dig maintains a permanent archive and research office at the Institute of Fine Arts. The research office houses archival documents such as the excavators' field notebooks, photographic negatives, color slides and original drawings, as well as an Apple PowerMac 7100 desktop computer.

Databases

Both among specialists and among the interested public, Aphrodisias is best known for its extraordinary sculpture (see right). As such, the study and publication of the approximately 10,000 pieces of sculpture found in the earlier excavations is a high priority. At the time of excavation, each piece of sculpture was assigned an inventory number, written in ink or black paint on the object. The number was also recorded on a five-by-seven-inch notecard, together with basic information such as where the object was found, its dimensions and the state of its preservation.

The task of sifting through this mass of evidence would have been daunting under any circumstances, but it was made especially complicated by the fact that the five-by-seven-inch cards were the only extant records of many of the objects. Even worse, there were no records of the present whereabouts of the majority of the pieces. Before this material could really be studied, therefore, it was necessary to go through the several storage depots on the site and in the nearby museum, and to identify and organize their contents. At the same time, the inventory cards were checked for accuracy, and new photographs were taken of every object that was reexamined.

In order to facilitate this process, we decided to convert all of the original notecards to a computer database. The database program we are using is FileMaker Pro, chosen because of its speed, cross-platform capability and ease of use. We initially expected that we would scan in the original cards, convert them to text, and then divide them up into the various fields.

In the end, it proved faster and more accurate to re-type the contents of the cards directly into the database. We set up the database so that there would be separate fields for each category of information -- such as the height, width, and thickness of the object -- but in the interests of time we began by entering information into only six fields: the inventory number written on the object, the inventory number recorded on the inventory card (which was not always the same), the present location of the object, the numbers of the new photographs, the entire contents of the original notecard, and a final field for additional information, such as corrections to the data recorded on the original card. We have now completed the conversion of the sculpture inventory cards, and are beginning work on the cards for other classes of objects, such as pottery and coins. These databases will be used to generate the catalogues that will accompany the published final reports on each category of object.

In addition to computerizing old records, we are also trying to take advantage of the opportunity provided by computer technology to make the records of the current excavations more useful. Detailed information about all new inventoried objects is entered immediately into a database separate from (but linkable to) the database of finds from the previous excavations. In addition, our photographic records and drawing lists are also kept in database format.

Perhaps the most useful new type of record now being kept is a database of all excavated contents. This document includes a record of every layer or part of a layer dug at the site since 1993 (when new excavations were begun), with information in approximately 30 fields about the location and contents of the layer and its relationship to surrounding stratigraphic features. The records also include cross-references to the object, photograph and drawing records, as well as to the excavator's field notebooks.

Spatial Analysis

Resistivity surveying with the Pentax PTS-2 Total Station.

The first step was simply to make a new series of plans of the site using modern surveying instruments: a Pentax PTS-2 Total Station with a Pentax SC-5 field computer, or data recorder (see above). The topographical survey of Aphrodisias is based on a number of permanently marked points whose location in the local Turkish survey grid is known. When properly positioned over any of these points, the total station can be used to determine the precise three-dimensional location of any other point within range.

Measurements are taken by means of an infrared beam emitted from the survey instrument and then reflected back to it by a prism set up on the point whose location is being surveyed. The survey data thus recorded is temporarily stored in the attached field computer, and later dumped in DXF file format from the recorder into the laptop in the drafting room. It is then imported into the CAD program used at the site, Microstation 95 for Macintosh.

To date, we have used this survey data mainly in order to generate two-dimensional plotting sheets of survey points. These plotting sheets are then printed out on our laser printer at a variety of scales, and used as the basis for detailed drawings of individual areas, done by hand in pencil on graph paper. These initial drawings are usually made at a scale of 1:25 or 1:50, and a variety of less detailed drawings are then patched together from them (see right). We do not yet have an efficient way to automate the production of the primary large scale drawings, since photogrammetric and other methods tried at other sites have been unsuccessful. Now all the drawings based on them can be done, at least in part, on the computer, more quickly and easily than before. We are acutely conscious, however, of the fact that we have barely begun to exploit the potential of our survey data -- for example, for 3-D modeling.

The other main focus of research into the spatial organization of the site has been the investigation of unexcavated areas around the city center. Here we are fortunate in that Aphrodisias is relatively flat, uncultivated (it is an archaeological park), and evenly buried; it is, in other words, an ideal candidate for geophysical prospection. A geophysical survey of the areas north and south of the city center was begun in 1995; this survey has been carried out by Lew Somers of Geoscan Research (USA).

The surveying instrument that has proved most successful is an electrical resistivity meter, which measures the resistivity of subterranean objects, such as stone walls, to electrical current. The resistivity meter consists of two electrodes, metal spikes set a specific distance apart on a rigid frame. Current flowing into the ground through one electrode is measured by the other, and the measurement naturally differs according to the resistivity of the earth and buried materials between the electrodes. The measurements are recorded on a small field computer connected to the instrument.

Resistivity survey in progress, with fortification wall in background

When measurements are taken at regular intervals over a large area, the information thus recorded can be used to generate maps of the buildings and other features that lie beneath the surface. The specific instrument used is a Geoscan Research (UK) RM-15 Advanced Resistance Meter, operated at 100 v output, 0.1 ma current and 137 Hz. The resistivity surveys have been performed at one data sample per square meter density, with the instrument electrode spacing set to 1 meter.

At the end of every survey session, the survey data is downloaded into Lew Somers's laptop computer, and processed by Somers using Geoscan's proprietary software, Geoplot. Somers then prints out preliminary maps at a scale of 1:1000 on a portable inkjet printer. These maps are used to assess the day-to-day results of the survey in the field. At the end of the season, Somers gives us his data files in a number of different formats, including TIFF files of the processed data, which we can easily manipulate in Photoshop, and paste into our existing Microstation drawing files.

Among the most interesting results of the geophysical survey is the clear appearance in a number of places of a grid pattern of light grey stripes, which are low resistivity areas, or areas with low concentrations of stone. Test excavations carried out at three locations have shown that, as we suspected, these low resistivity stripes are streets. When the results of the survey are plotted on the larger city plan, moreover, it is apparent that the streets detected in the presumably residential neighborhoods north and south of the center of the city have the same orientations as the main axes of the Agora. The geophysical survey has thus dramatically altered our understanding of ancient Aphrodisias, showing as never before realized that this was a planned city, laid out on a rectangular grid. This survey has thus more than met its original objectives of clarifying the organization of the public buildings of the city center, and the relationship between the city center and the rest of town.

A third project which is still in its infancy is the investigation of ancient Aphrodisias in relation to the surrounding countryside and other nearby ancient settlements. Here we are combining old-fashioned archaeological surveying -- driving around the countryside and asking local residents if there are any ruins on their property -- with the study of satellite images and the use of an inexpensive hand-held Global Positioning System to locate specific sites. Still unexplored is the possibility of using a GIS package to integrate the several different methods of spatial analysis now being carried out at Aphrodisias.

Image Processing

A panorama of the site's theater {http://www.arch.at/sites/aphro/sights/index.html}. It can be rotated to the left or right when the cursor becomes a target.

The panoramas may be viewed on the ARCH website (http://www.arch.at/) as well as on our own site (www.nyu.edu/projects/aphrodisias). They have been very effective when projected at a large scale during public lectures and demonstrations. At the moment, the panoramas that have been made are too few to be of much use for research, but this technology is potentially extremely useful for remote study of the site during the off-season.

Internet

At present, we publish annual or biennial preliminary reports on each field season in a conventional print journal, the American Journal of Archaeology. For us, the great limitation of conventional publishing is in the number of illustrations that can accompany the text; it is often said that the heart of an archaeological report lies in the illustrations, with the text serving in large part as commentary on them. Our reports generally tend to run from 25 and 30 pages in length, and we are usually allowed between 20 and 25 illustrations, but we could certainly provide 100 or more.

On the Internet, of course, that is easily done, and in the future we intend to publish in both media. The images that accompany the electronic publication would also become part of a much larger digital library of images on the site. The limitations of conventional publication are especially acute when it comes to information originally recorded electronically, such as the geophysical survey data. This data can be represented in a virtually infinite number of ways, and it is extremely frustrating to have to limit ourselves to one or two images for publication. Through electronic publication, we could easily provide a much larger number of images to interested scholars. We will be investigating the possibility of making the original data accessible on the Internet, as well.

On the research front, we are now making all the text files, databases, CAD files and digital images stored on our computers fully accessible to staff at any time, wherever they may be -- and during the off-season, our staff are very widely dispersed indeed. Of course, access to these files is closely protected, so we are developing strict rules to regulate the modification of the files. It is ironic that due to the current instability of the telephone connection at Aphrodisias, one place where it is difficult to gain access to our data files in this way is at the site itself, but that is arguably the place where they are needed least.

Conclusions

Archaeological research has experienced a number of technological revolutions, such as the introduction of radiocarbon dating in the 1950s, and this is also a time of exciting and dramatic change. At Aphrodisias, new technology has enabled us not only to improve the way we do such mundane things as keep track of the thousands of artifacts recovered in our excavations, but also to study our site much more effectively -- not as a collection of monuments, but as a whole city, constantly changing and inhabited not just by statues, but by living people over a period of roughly seven hundred years.

Posted May 18,1998. Reviewed January 26, 2004.