Excavations at Zeugma

One of the richest and most important archaeological sites of Graeco-Roman date excavated in recent decades in the eastern Mediterranean is ancient Zeugma, located on the Euphrates River in southeastern Turkey. Also known as Seleucia and founded by Seleucus I Nicator ca. 300 B.C.E., Zeugma gained worldwide attention from the discovery of outstanding mosaics of Roman imperial date in many of its houses. The site was explored initially in the 1970s and then more fully in the 1990s and early 2000s. Soon afterward, approximately 30% of the site was inundated by construction of a hydroelectric dam downstream near Birecik, but not before many of the mosaics and adjacent wall paintings were lifted and taken to the archaeological museum at Gaziantep, the modern regional capital. Several of the mosaics rival the best pavements found at Antioch-on-the-Orontes, a larger town in the same general region. One of the teams participating in the rescue excavations was Oxford Archaeology (OA), funded by the Packard Humanities Institute (PHI) and led by William Aylward of the University of Wisconsin–Madison. This three-volume work is the team’s final report on excavations conducted in one sector of Zeugma in its only season on the site (2000). There is no comprehensive publication of all the town’s sectors investigated so far, but several reports by Australian, French, Swiss, and Turkish scholars have already appeared.

This publication is impressive for the scope and precision of its scholarship and for the thoroughness of its documentation, primarily covering an area in the city’s central residential district where parts of 13 houses were found. All the material is superbly illustrated. This is a collaborative work, drawing on the expertise of specialists in many disciplines. The three volumes contain chapters on specific aspects of OA’s excavations and a few related topics, all of which follow a detailed overview of the history of Zeugma written by Aylward. He traces Zeugma’s development from its foundation in Hellenistic times through its transformation into an imperial Roman garrison town in the later first century C.E. and its subsequent flowering in the second half of the second century and first half of the third century, when the houses received their finest decoration.

A major contribution of the excavations at Zeugma is confirmation of the date of the Sassanian sack in 252/3 C.E., providing a terminus ante quem for material sealed by the destruction. Coins found in the debris excavated by OA verify this date, with the latest issues attributable to Trajan Decius and Trebonianus Gallus. Curiously, no human skeletons were found in the collapsed homes at Zeugma, nor is there much evidence of military combat, suggesting that the inhabitants fled the city before the attack and the Roman army mounted no defense. These facts seem to contradict Aylward’s contention that soldiers may have been billeted in the houses of Zeugma prior to the Sassanian assault. There also is evidence that the citizens intended to return to their homes, as evidenced by the hoarding of various items, as seen in the House of the Tesserae and the House of the Hoards explored by OA, and in the storage for safekeeping of a bronze statue of Mars and a bronze candelabrum in the House of Poseidon excavated by the French team.

In the Late Imperial and Early Byzantine periods, there was a resettlement of Zeugma, which lasted until the mid seventh century, when this city and others in the same region apparently became victims of Arab incursions. OA found a substantial peristyle house of late fourth-century date with a staircase leading to a second story. The French excavation also yielded evidence of many other dwellings of a more rural type. There is solid evidence of a major military installation (this could be the legionary headquarters) on the site, but perhaps not a true fortress, sought by the Swiss team and discussed in volume 3. In the Umayyad period, Zeugma became an insignificant outpost, last attested in literary sources of the year 1048.

Following Aylward’s historical overview is a chapter on conservation of the site of Zeugma during the rescue excavations, written by Nardi and Schneider, conservators from the Centro di Conservazione Archeologica, Roma (CCA). The CCA worked with the PHI team to preserve as much of the excavated area as possible, as the water level in the reservoir behind the Birecik dam rose steadily. All the small finds from the excavations were taken to a field laboratory for documentation and treatment. Mosaics and wall paintings in ancient houses were meticulously recorded and consolidated with hydraulic lime and a lime wash; where they could not be removed, they were protected with a 5 cm thick coating of hydraulic mortar. After backfilling, the excavated structures were reburied before inundation by the dam’s flood waters, with the intention of their being reclaimed by future archaeologists. Overall, an area of 8,700 m² was sealed.

Of chief interest in volume 1 of this publication are the 13 houses of Middle Imperial date excavated by OA, which are analyzed for their architectural design and phasing as well as their decoration with mosaics and wall painting, in addition to numerous furnishings. As Tobin explains, several of these structures have a Seleucid foundation over which Roman dwellings were built, with alleys separating the houses. Regrettably, the excavations failed to yield a complete plan of any of the houses, preventing us from understanding the overall circulation within. Several common features recur in the domiciles. They tend to be two stories high, with pier-and-panel construction (employing limestone ashlar blocks for the piers) in the lower-level walls and mudbrick walls in the upper story. The homes were built on terraces carved from the sloping terrain, and they have rock-cut cisterns as well as storage pits. Plumbing pipes reached the upper floor, and one house has a fountain. At the center of each building is a court, usually with columns on at least one side, screening off a reception space, and other connected rooms line the other sides. A gallery may lead to a triclinium, as in the House of the Bulls. The houses excavated by OA are more modest in scale and less elaborately decorated than the domus of Zeugma’s wealthier citizens, such as the House of Poseidon and the House of the Synaristosai, built on a promontory overlooking the river.

The acme of domestic architecture at Zeugma/Seleucia is the period from the mid second through the first half of the third century C.E., more specifically from the Antonine through Severan periods, when the houses received their richest embellishment of mosaic and wall decoration, discussed in this publication respectively by Dunbabin and Bergmann. This parallels a contemporary development of mosaic art in the regional hub at Antioch, located ca. 100 km to the southwest, and from which workshops almost certainly came to fulfill commissions at Zeugma. Another individual, named Zosimos, traveled to Zeugma from Samosata to the northeast (the erstwhile capital of Commagene) and signed pavements in a few of the richer homes at Zeugma. Two figural mosaics were found in the houses excavated by OA, one of them a multipaneled floor with Dionysiac masks including a head of Silenus, individual birds, and Eros (M17), ornamenting the House of the Bull. It has a counterpart in a pavement from a different domicile at Zeugma, bought on the art market by Bowling Green State University in the 1960s and published recently (S. Langin-Hooper, R. Martin, and M. Önal, with R. Molholt, “Zeugma as the Provenance of 12 Mosaic Fragments at Bowling Green State University,” JRA [2013] 439–55). The other figural mosaic uncovered by OA has an emblema of Nereids riding on sea creatures (M23) and was found in the House of the Fountain. Concerning geometric mosaics in the PHI corpus, we note two pleasing examples from the House of the Tunnel, one of which displays a centralized design of an interlooped guilloche band and a wavy ribbon (M26), while the other has a panel of intersecting circles colored red, yellow, and green and placed on a trellis with yellow and white poised squares (appx., Room 13H). The Sassanian sack of Zeugma establishes a valuable terminus ante quem for dating the mosaics of Zeugma and related pavements from Antioch, although making the chronology of individual floors more precise within the mid second- to mid third-century span is sometimes difficult. Unfortunately, little technical data about the mosaics were recorded by the field archaeologists.

Balancing the mosaics in domestic interiors is an extensive series of frescoes, several rising to a significant height, which are typically divided into large panels outlined by red bands in the walls’ main zone. In some cases, the panels exhibit painted imitation of yellow marble with wavy striations, as in Room 13A of the House of the Tunnel. In Room 2K of the House of the Bull, one sees white panels displaying suspended garlands tied by ribbons; thin plant candelabra separate the panels, and the dado is dotted with spiral plant forms. In general, these frescoes were found less well preserved than those in the wealthier houses at Zeugma, which include monumental figures, some inscribed with mythical names. As Bergmann notes (168), the wall paintings at Zeugma demonstrate close affinities to frescoes of similar date in the terrace houses at Ephesos, as well as other sites. The reader, however, wishes there was more discussion of possible decorative ensembles in the dwellings at Zeugma (this is done in only one instance), looking for evidence of an aesthetic rapport between floor mosaics and wall paintings in individual spaces. This type of analysis also is needed at other archaeological sites in Asia Minor.

Subsequent chapters of volume 1 deal with graffiti found on the walls of homes, stone inscriptions related to the cult of Antiochus I of Commagene (in addition to private funerary inscriptions), a new sculptural relief or stele commemorating the Commagenian king, and the geophysics of archaeological research at Zeugma. The graffiti, documented by Benefiel and Coleman, date from an apparent decline of the Roman houses prior to the Sassanian attack, when some dwellings were subdivided. The graffiti depict boats or gladiators, the latter perhaps reflecting games held in the city. Particularly significant for the religious life of Zeugma are stone inscriptions, discussed by Crowther, and a newly discovered relief depicting Antiochus I, analyzed by Rose. We learn, among other facts, that attendants making offerings to the king were required to wear Persian costume. The new dexiosis relief resembles another, fragmentary stele commemorating the same ruler found on Belkis Tepe, the hill dominating the city, in the 1970s. The present example represents the king in Persian dress and with a feathered crown shaking hands with nude Apollo, who has a radiate crown. Because this object was discovered in a trench in the residential part of the city, it is suggested that Zeugma may have had two temene honoring Antiochus, one on the sacred hill and the other in the city below. Altogether, 10 reliefs of this type honoring Antiochus I have been found at various sites within his kingdom, and all date to the latest phase of the ruler cult. Chapters 11–13 of volume 1 present magnetometry, electrical resistivity, and ground-penetrating radar analysis of Zeugma. An area of about 20,000 m² was surveyed, helping to reconstruct the ancient city plan. Street alignments of Hellenistic date on the promontory overlooking the river at Zeugma match those found in Apamea on the opposite bank, whereas later Roman residences had different street alignments.

Volume 2 deals primarily with the many kinds of portable objects of various materials found by OA. These shed light on commercial and economic life at Zeugma and the city’s trade abroad as well as the everyday functioning of private households and the residents’ diet. The pottery analysis by Kenrick, Doherty, and Reynolds is meticulous, classifying many types of vessels, including transport amphoras, from stratified archaeological layers. The examples reach chronologically from Hellenistic times to the Islamic period (there are seven major groups), and the results are quantified. The illustrations include color photographs of selected vessels and profile drawings of the catalogued items. This documentation will be very valuable for constructing an inventory of all the pottery found in the diverse excavations at Zeugma and for interpreting this material within a regional context. Zeugma had commercial access to the Mediterranean world, as we know for example from Koan amphoras with double-barreled handles. Another noteworthy fact about the pottery at Zeugma is the absence of tableware of second- and third-century date, with glass vessels apparently replacing ceramic ones for this purpose. In the Late Imperial period, however, imported tableware reappeared, coming from North Africa, Cyprus, and western Asia Minor.

The rest of volume 2 contains fine essays on ceramic lamps, clay sealings, terracotta figurines, and glass found in OA’s excavations. Hawari identifies nine distinctive lamp types of primarily local production, either moldmade or wheelmade in various periods. Thirty-one examples are complete, and the best formal parallels occur at Dura-Europos, although these objects represent only a fraction of lamp types known for the entire Euphrates Valley. Once again, a regional synthesis is needed. In a related chapter, Gingras and Aylward discuss terracotta figurines excavated by OA. Another study by Herbert deals with clay sealings, or bullae, used to close and notarize documents, which average 15 mm in diameter and bear impressions of individuals’ seal rings. The trenches investigated by the OA team yielded limited samples, in contrast to the masses of sealings found elsewhere on the site and implying the existence of an urban archive. Overall, Zeugma is unique for the quantity of bullae it yielded. As for the glass finds, discussed by Grossmann, they had varied uses and stretch chronologically from the Commagenian period to Late Imperial times. The earlier types include fragments of mosaic glass vessels, while the Roman era furnished an array of glass objects, including tableware, jars or bottles, and glass lamps of Late Imperial date. A sizable amount of window glass was uncovered, appearing thin and greenish in color, and which was set within iron frames. Both upper and lower stories of houses had glass windows.

The essays in the third and final volume fall into a few general categories. They include coins and other metal objects, milling and weaving equipment and textiles, military gear and military camps at Zeugma, environmental studies, and plant and animal remains. As explained by Butcher in the chapter on coins and coin hoards (for which the bibliography unfortunately is missing), Zeugma issued its own coins intermittently, beginning in the reign of Antoninus Pius and continuing to the mid third century, while coins from several other cities in the same general region also circulated. Third-century bronze coinage, the largest chronological group, includes issues of Antiochene, Mesopotamian, Pontic, and Peloponnesian origin, with the Mesopotamian issues predominating and seeming to reflect the importance of trade with the East. A characteristic image on locally made coins is a temple on a hill with the accompanying legend “ZΕΥΓΜΑΤΕΩΝ” (“of the citizens of Zeugma”). Four coin hoards were discovered by OA, the largest containing a puzzling mixture of diverse issues ranging from Hellenistic times to the mid third century C.E. (462).

Three other groups of metal objects, discussed thoroughly by Khamis and Scott and illustrated beautifully, include items made of copper alloy, iron, and gold. It is assumed that the most elaborate examples in the first category were imported from Italy. They include combinations of a patera and jug, intended for the washing of hands in Roman dining custom, and one bronze patera handle is ornamented with a naturalistic ram’s head. There also exists a fragmentary candelabrum in three pieces (BR20), an elaborate item testifying to Zeugma’s prosperity in the Roman era, in addition to a few bronze statuettes of Aphrodite and fragments of large statues. Among iron objects, nail fragments are most common, followed by tools, objects for household security such as padlocks of varying size, a parade helmet, and a brazier for heating or lighting (IR103 and IR105). Some iron window grilles or frames were made of star-shaped pieces, with one fine example (IR269) displaying wavy points on the stars.

Additional domestic items of special interest in this volume, examined by Parton and Cole, are milling and weaving equipment and textiles, with the textiles totaling 170 fragments, an astonishing number for the limited area excavated. Among these are fragments of shoe soles (TX26 and TX27) apparently belonging to a sandal, as well as rope or cord. Luxurious items of dress may have been imported from Tyre or Sidon or traded along the Silk Road. Curiously, no traces of wool fabrics were found, which contrasts with the findings at other sites in the same region. Warp-weighted looms were still in use at Zeugma in the third century, as indicated by the abundant loomweights recovered.

Subsequent chapters (8–10) deal with the military presence at Zeugma and are written respectively by Scott, Elton, and Hartmann and Speidel. The topics range from arms and armor and other military objects to Zeugma’s military history in light of recent field research and finally military installations at Zeugma. The Roman army occupied the former Kingdom of Commagene, including the city of Zeugma, from the later first century C.E. onward (in the decades of the 60s or 70s), and among the various legions garrisoned there, only legio IIII Scythica seems to have had a permanent base at the site and to have built most of the military structures. Many kinds of military fittings were found in the destruction debris of houses explored by OA, but none seems to show signs of recent use. They include helmet fragments, spearheads, arrowheads, a few swords and daggers, scales for armor, belt fittings, harness fittings, and studs. Much of this material resembles military gear adopted by soldiers on the Danube frontier, particularly in Pannonia, suggesting that Zeugma had a direct connection to the army in that region. But how do we explain the objects’ presence in houses at Zeugma? Perhaps the equipment was left over from an earlier period in Zeugma’s history and was not in active use at the time of the Sassanian attack. Normally, troops would be stationed in a camp outside the town they were protecting rather than in private homes. The final word on this subject has yet to be written.

Complementing discussion of the domiciles at Zeugma is a chapter on the Swiss team’s investigations of military installations at this site from 2001 to 2003, following earlier research in 1997 and 1999. Remains of a military camp were found on the plateau of At Meydanı and in surrounding fields in the western part of the site. The camp was connected to the city by a road and was equipped with workshops, latrines, and baths. In the archaeologists’ opinion, this construction is not large enough to have been the permanent quarters of the Fourth legion, which occupied the site continuously until the year 195 C.E., when it was redeployed farther eastward. Numerous military gravestones, along with stamped tiles and dedications to the Roman emperor, all inscribed in Latin, were found distributed over the area. More recent research is clarifying and further exploring the army’s defensive installations and should illuminate more fully the relations between soldiers and townspeople at this site.

The remaining chapters (11–14) of volume 3 reveal a great deal about the natural environment around Zeugma in antiquity and the diet of the city’s residents and other domestic features, based on faunal and plant remains and the analysis of charcoal. Using techniques of geomorphology, isotopic studies, and palynology, Meiggs determined that the land around Zeugma originally had a moister climate than in the Graeco-Roman era and that its vegetation was once denser, all changed by extensive grazing around the site going back to Neolithic times. From analysis of faunal and plant remains, studied respectively by Charles, and Challinor and de Moulins, we learn that the diet of Zeugma’s residents included much pork, lamb, and mutton in addition to beef, camel, deer, hare, domesticated fowl, and fish; milk came from sheep and goats. The overall assemblage resembles that of inhabitants of contemporary Sagalassos in the province of Pisidia, extremely well documented by the Belgian team working there. The diet at Zeugma also is consistent with the Roman diet in other parts of the empire. An important supplement to the food of Zeugma’s population is pistachio nuts, perhaps influenced by Persian cuisine. (Pistachios are an important cash crop in the region around Zeugma today.) One house excavated by OA also yielded stored walnuts and pomegranates, and the very presence of the endocarp of walnuts, harvested in the autumn season, allows us to date the Sassanian attack to that time of the year, specifically within a period of a few weeks. This may indicate that the assault took Zeugma’s residents by surprise, belying the notion that the city had already been abandoned. Perhaps instead the inhabitants fled only days before the Sassanians arrived. Finally, from Gale’s analysis of charcoal samples (58 in all), consisting primarily of wood burned in the Sassanian sack, it could be determined that the local populace had access to a wide variety of timber, including pine, cedar, and cypress, some of it acquired by long-distance trade.

From this publication as a whole, the reader gains a very full portrait of life in ancient Zeugma, both from the careful and detailed analysis of excavations by OA, which were focused on one sector of the city, and from investigations of nearby military installations and the surrounding environment. This research sets a very high standard of archaeological interpretation, complementing the results of the Franco-Turkish investigations in other more prosperous and richly developed sectors of Zeugma. It is hoped that future collaboration among these teams will produce a still more profound understanding of this fascinating town, viewed within a regional context. In the meantime, current excavations have revealed the full extent of Zeugma in the Roman period, including its defensive perimeter and the surrounding necropoleis. New houses of Roman date also have been unearthed in the residential district, and a superbly designed shelter protects some of them (see the Zeugma website at http://zeugmaarchproject.com). Finally, in 2008, under Kutalmış Görkay, there began a fresh investigation of the sanctuary on Belkis Tepe, a focal point of the city’s religious life. In view of these developments and others to come, the future of archaeological research at Zeugma looks very bright and promising.

David Parrish
School of Visual and Performing Arts
Purdue University
dparrish@purdue.edu