Sardis, 2013

Nicholas Cahill

Introduction

Archaeological fieldwork at Sardis was carried out from June 4 until Aug. 15, with a team of 49 archaeologists, staff, and specialists. We are particularly grateful to Ümit Güngör of Ege University, Izmir, who served at short notice as assistant director, and to Ünal Çınar of the Antalya Museum, who represented the Ministry of Culture. We again thank the Ministry of Culture and Tourism for continued permission to carry out archaeological research, and especially General Director Abdullah Kocapınar, Excavations Division Director Melik Ayaz; and Sevgi Soyaker, director of the Manisa Museum.

Field 49

Since 2009 a primary focus of excavation has been a pair of natural spurs of the acropolis, known as ByzFort and Field 49 (figs. 1, 2). These two spurs were enclosed in the Lydian period by monumental terrace walls (fig. 3), and in past years we have proposed to identify this area as the palace of the Lydian kings, based on the architecture and artifacts recovered there. Our goals in 2013 were to uncover more of the Lydian buildings on top of this terrace. As in previous years, however, the excavation and recording of later features of the Hellenistic, Roman, and Byzantine periods occupied most of the season.

Three trenches were opened, at the north, center, and southern parts of the hill (fig. 4). In the southern trench one major feature is an early Roman building, built after the earthquake that destroyed Sardis in 17 AD. As reported in previous years, one room contained a furnace; the walls of the other room were painted with a fine design. This year’s excavations focused on the earlier levels below the Roman building. Two interesting groups of artifacts, probably ritual deposits, were discovered under the earth floor of the first room (figs. 5, 6). Each consisted of two small vases, one used as a lid for the other. Each one contained a bronze nail, a bronze needle, other pointed metal artifacts, a coin, and an egg, pierced deliberately at one end, perhaps by the nail or needle (fig. 7). The coins both date to the reign of Nero; the reverse of one coin had a lion cut into it. They are remarkably similar to offerings described in the excavations of the sanctuary of Artemis in 1913.1 The skeleton of a young pig was found buried beneath the floor of the other room, and might be part of this or a similar ritual offering (fig. 8).

The Hellenistic phases on this hill are very complex. In the southern trench there are at least three or four major architectural phases, represented by subterranean foundations 1.7 m thick (fig. 5). The massive construction and occasional artifacts attest to very high-status buildings here in the Hellenistic period; their poor preservation may be due to the earthquake of 17 AD and extensive stone-robbing.

In the central trench, preservation of Hellenistic structures was better, and we uncovered part of a Hellenistic building perhaps dating to the second century BC. These walls, however, were much less massive than those in the south (fig. 9).

As in earlier seasons, remains of the Persian period were almost absent in most trenches. The exception is at the northern end of the hill. Here, a pit some 4.5 m deep crosses the excavation area. Pottery from the pit dates it to the second half of the sixth or fifth century BC. This might be a robber’s trench for some earlier, monumental Lydian feature such as a terrace wall, which was entirely removed in the Achaemenid period. There was no architecture or other signs of occupation, only of stone-robbing.

Buildings of the Lydian period on this hill were constructed on a massive terrace which established the general outline of the hill for the next thousand years. The first terrace was built in the seventh century BC of massive boulders, uncovered on both the northern and southwestern slopes of the hill. This was replaced in the first half of the sixth century by a wall built of limestone ashlars. This limestone terrace wall has been uncovered in the southern trench, but is so far missing in the central and northern trenches.

A major goal of the past years was to uncover the Lydian buildings on top of this monumental terrace. In the southern trench, preservation was not good. The only architectural feature was a relatively narrow mudbrick wall, preserved between the massive Hellenistic foundations (fig. 10). Nearby, however, floors and walls were not preserved, but only a layer of burned debris containing fine pottery and luxury artifacts. In previous years these included a chalcedony seal, part of a burned ivory inlay, fragments of jasper, and other luxury objects. In 2013, this stratum produced a number of human bones, a lump made up of at least 20 bronze arrowheads, apparently melted together and perhaps originally belonging to an armory (M13.015:13746), and other Lydian artifacts (figs. 11, 12, 13). The assemblage with its human bones and weapons is reminiscent of the destruction level found in houses around the city, and may date to Cyrus the Great’s capture of Sardis in about 547 BC.

Lydian levels were not reached in the central trench. In the northern trench, no Lydian features of the sixth century BC were found. Among the finds from later strata here are fragments of iron lamellar armor, coming from the robber’s trench and therefore either of Lydian or Persian date (M13.011:13636; fig. 14), and a column krater, also of either Lydian or Persian date (P13.065:13408; fig. 15). However, at a depth of almost 6.5 m below surface was a mudbrick wall with postholes to support a substantial wooden framework (fig. 16). This was buried not by the burned and artifact-rich stratum found in the southern trench, but by an almost sterile layer of unburned green mudbrick. The relatively few artifacts from this stratum seem to date to the seventh century or earlier, at least a century earlier than the burned layer in the south. This building may be contemporary with or earlier than the boulder terrace wall on the north slope of the hill, and may be similar to a “basement” excavated on the adjoining terrace, ByzFort, in the 1980s and 1990s.2

  • Fig. 1

    Plan of Sardis. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 2

    Plan of central Sardis. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 3

    Terraces on the lower slopes of the Acropolis in central Sardis; reconstructed view looking south. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 4

    Plan of Field 49, showing trenches and geophysical survey. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 5

    View of F49 13.1, showing Roman oven (center); late Roman tile-floored basin (lower right), Hellenistic walls (below). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 6

    Ritual deposit in situ under floor of the early Roman room in F49 13.1. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 7

    Ritual deposits found below the floor of the early Roman room in F49 13.1. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 8

    Skeleton of young pig found below the floor of early Roman room in F49 13.1. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 9

    View of trench F49 13.2 showing Hellenistic room with door. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 10

    F49 13.1, Lydian mudbrick wall and post-Lydian limestone and rubble walls. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 11

    Lydian bronze leaf-shaped arrowheads, corroded or melted into a solid mass, from trench F49 13.1 (M13.015). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 12

    Bronze repoussé plaque from trench F49 13.1 (M13.013). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 13

    Lydian bronze bowl from trench F49 13.1 (M13.021). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 14

    Iron lamellar armor from trench F49 13.3, in a Persian context but either Lydian or Persian in date (M13.011). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 15

    Trench F49 13.3, Late Lydian column krater from Persian-period pit (P13.065). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 16

    View of trench F49 13.3 showing early Lydian mudbrick wall at right; Hellenistic foundation at upper left. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

Wadi B Temple and Field 55

A second excavation sector was on a lower terrace near the center of the ancient city, known as Field 55 (figs. 1, 2). Previous excavation showed that this belongs to a sanctuary with an octastyle temple, about the same size as the temple of Zeus at Aizanoi or the temple of Augustus and Roma in Ankara (the so-called “Wadi B Temple”), facing north onto a 100-meter-square plaza (“Field 55”), reached by an axial staircase (fig. 17).3 The temple and terrace date to the first century AD, and are thought to belong to the imperial cult.

The objective in 2013 was to investigate a group of architectural blocks and inscriptions on the east side of the terrace, which had been partly exposed in 2005. Part of the eastern terrace wall was exposed, and the area outside was found buried by a thick layer of fallen marble blocks (figs. 18, 19). Nearly all of these are reused elements, mostly from the Wadi B temple, but also including blocks from other buildings, honorific and dedicatory inscriptions, and sculpture. Removing the fallen blocks showed that they had been built into at least two features: a late phase of the terrace wall, faced with reused marble blocks, and crosswall 2.6 m thick, built entirely of spolia.

The architectural blocks and sculptures provide a great deal of new information about the temple and other buildings of the sanctuary. Among the important finds are the bottom half of a figural Corinthian capital whose upper half had been discovered in 2005 (fig. 20). Fragments of two similar capitals were also found. Fragments of architectural sculpture include legs of the right scale and workmanship to belong to the pediment of the temple, and a number of fragments of a figural frieze (figs. 21, 22, 23). The scale and style of these fragments suggest that they all belong to the same structure, most likely the temple itself, and if so, this must have been an extraordinarily richly decorated building. Other blocks include architraves, a floral frieze block, cornices with lions’ head spouts, column drums and bases, moldings, and a large number of other fragments. One architrave has cuttings for bronze letters, but a convincing reading has not yet been established. The style of the sculptural reliefs and architectural ornament suggest a date in the Julio-Claudian period (S13.041:13479; fig. 24).

Inscriptions from this deposit include perhaps 83 separate texts, although many are represented only by a few letters. A number of texts honor priests and priestesses of the province of Asia and of the Thirteen Cities, reinforcing the identification of this as a neocorate temple dedicated not by the city of Sardis, but by the Roman province of Asia. One text mentions building or repairing the roof of the sanctuary of Menogenes ([Meno]geneion), a person perhaps referred to in other texts from this assemblage. A small but particularly interesting text mentions the Herakleidai (the dynasty of Lydian kings); Sadyattes, an early king of Lydia; as well as word for “he ruled.” Georg Petzl tentatively suggests that it may be a chronicle of early Lydian history.

All these blocks and sculptures were reused in a later Roman building, which is also of great interest (figs. 25, 26, 27). Its massive scale and use of marble spolia suggest that it was an important public building, but its nature is uncertain. Its date is also uncertain, as we were not able to reach floor level. Earlier excavations of the temple itself, the source of many of these blocks, suggest that it suffered two destructions: one in the second century AD, and another in the late fourth or early fifth century AD. The building in which these blocks were reused may be a product of this later destruction of the temple, although the earlier date is not currently ruled out.

Finally, this later Roman building seems to have been destroyed in an earthquake. Blocks were found fallen and shattered in place, sometimes fallen like dominoes as if they fell all together. This might be the same earthquake attested in earlier excavations by deep fissures that run through the terrace from east to west, one of which was traced to a depth of about 9 meters (fig. 28). A team of archaeoseismologists is studying these earthquake faults, and we hope that further excavation of the building will provide a closer date.

  • Fig. 1

    Plan of Sardis. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 2

    Plan of central Sardis. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 17

    Plan of Wadi B temple and Field 55. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 18

    Marble collapse in Field 55 excavations. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 19

    Plan of marble collapse in Field 55 excavations. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 20

    Figural capital from Field 55 excavations: top half discovered 2005, bottom half discovered 2013 (S05.016). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 21

    Sculptural fragments from the collapse in Field 55 excavations, perhaps belonging to a pedimental figure (S13.060). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 22

    Relief fragment from the collapse in Field 55 excavations (S13.059). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 23

    Relief fragment from the collapse in Field 55 excavations (S13.059). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 24

    Fragment of Julio-Claudian head, probably from a relief, from Field 55 excavations (S13.041). (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 25

    Plan of late Roman features in Field 55 excavations. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 26

    View of Field 55 excavations at end of season. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 27

    View of Field 55 excavations at end of season. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 28

    Excavations in the center of Field 55 in 2005, showing earthquake fault. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

Site Conservation, and the Temple of Artemis

The temple of Artemis at Sardis was excavated over a century ago; in the intervening years, the building has become discolored by black cyanobacteria and green lichens (fig. 29). Not only do these damage the stone, but also they provide a canvas for people to scratch their names into the walls, doing further, permanent damage. In order to preserve this building we have developed new methods to remove the biological film, leaving the marble undamaged, and last year received funding for a five-year program to clean the entire temple and restore buildings in the sanctuary. In 2013 our efforts were aimed at refining the methods of cleaning, fine-tuning the system to account for the very different degrees of biological growth on different parts of the building in order to bring the stone to a uniform color (figs. 30, 31).

A program to stabilize the small chapel attached to the temple was continued as well. The north wall of the building was pointed with lime putty and capped. We are developing a program to stabilize the other walls of the building. Other buildings throughout the site including Building A and the Roman Agora were stabilized and restored as necessary to prevent further collapse due to stone robbing.

A major effort in site conservation and restoration is the “Touristic Enhancement Project,” a project to conserve and open to visitors the region including the Lydian Fortification, the monumental Lydian gate, Late Roman houses, the Synagogue, a Roman colonnaded avenue, and other buildings in the vicinity. Work this year included designing a new roof over the 20-meter-thick Lydian mudbrick wall and adjacent buildings. This will be built of translucent glass and allow visitors to see the wall from different viewpoints. The existing shelter over the mudbrick fortification was replaced until a permanent roof is built.

Research and publication projects included the Lydian levels of the “House of Bronzes” sector; architecture and finds of the Synagogue; architecture of the Temple of Artemis; Hellenistic and Roman pottery, including the development of a new chronology for Hellenistic coarsewares from the site; coins discovered since 1971; inscriptions on stone; figural terracottas; and Byzantine seals. Analysis of marble from the temple of Artemis showed that it was built from marble from the Mağara Deresi quarries just south of the site.

Traces of pigment on a late Archaic molding from a cemetery were analyzed with a portable XRF machine, and proved to be Egyptian blue, copper-based green (malachite or verdigris), and red ochre. The machine was also used to perform elemental analysis on ceramics from the site; however, careful experimentation showed that burial conditions and cleaning treatments significantly affected the trace elements of the sherds, rendering the analysis unreliable.

(Adapted from Kazı Sonuçları Toplantısı 36.2: 413-430)

  • Fig. 29

    Northeast anta of the temple of Artemis, before cleaning. The overhanging architrave block has shielded the anta below from rain, so lichen and cyanobacteria have not grown, and the original color of the marble is preserved. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 30

    Marble cleaning test: Building Q before cleaning. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

  • Fig. 31

    Marble cleaning test: Building Q after cleaning. (©Archaeological Exploration of Sardis/President and Fellows of Harvard College)

Notes