Sardis, 2015

The 2015 season at Sardis included an international staff of 69 scholars and students from 26 different universities and institutions, engaged in excavation, documentation, conservation and site enhancement, research, and publication. We are grateful to the Ministry of Culture and Tourism for permission to carry out archaeological research in Turkey, to Director General Abdullah Kocapınar, Assistant Director Zülküf Yılmaz, and Excavations Division Director Melik Ayaz and Assistant Division Directors Umut Görgülü and Güzen Köksal. This year’s representative from the Izmir Archaeological Museum was Mustafa Kiremitçi, whose expertise, humor, and tremendous good will made the season a great pleasure.

Excavation was carried out in three locations in the city (figs. 1, 2). The first sector is Field 49, a natural hill that was artificially terraced in the Lydian period (figs. 3, 4, 5). Two trenches exposed more of the complex Roman, Hellenistic, and Lydian remains on this hill in central Sardis. Our goal here is to understand the long history of this region, and particularly the Lydian phases when this terrace may have belonged to the palace of the Lydian kings.

The Roman phase here was probably begun in the early Roman period, probably after the earthquake of 17 AD, when the walls of destroyed buildings were robbed out and the area rebuilt. The buildings remained in use, with substantial modifications, through the sixth century AD.

One of the most remarkable objects found in recent years came from a Late Roman level: a bronze triangle with incised figures of Hekate in the three corners, inscriptions and symbols (M15.005:14186; figs. 6, 7). Hekate is identifiable by the objects she holds: torches, whip, snake, sistrum, and a sword, and epithets inscribed above each figure. Below each is written: am(e)ibousa (“the changing one”). This is only the third such triangle known, similar to examples from Pergamon and Apamea, and may have been used in theurgic rituals.¹

Hellenistic remains were complex, with multiple phases in both the northern and central trenches of the hill (figs. 8, 9). Within the terrace, parts of one or two rooms were excavated. One floor was found with traces of burning, fallen mudbricks and rooftiles, and other evidence of destruction (fig. 10). The pottery looked relatively early in the Hellenistic period, but two coins from this deposit remain to be cleaned and read.

As in previous years, we uncovered no architectural remains, and very few artifacts, dating to the Persian era.

Lydian levels were only reached with difficulty this summer, although a significant number of Lydian artifacts was recovered from later contexts, such as a fragment of a fine architectural terracotta depicting Pegasus (fig. 11). Excavation of the Roman robber’s trench revealed the top of the original Lydian limestone terrace wall, aligned with the section uncovered 40 m to its south (figs. 3, 12, 13). There are no obvious signs that this was built from reused blocks, and this probably belongs to the original Lydian phase of the terrace.²

Further excavation within the Hellenistic “basement” revealed more of the stone packing of this terrace, and the bottom of the east-west limestone wall uncovered last year (fig. 14). The wall was set above the packing, showing that it belongs to a later phase of building — a fact already suspected from the use of reused faceted wall base moldings in its foundations.³

The limestone terrace wall replaced an earlier monumental terrace wall made of massive boulders, which has been uncovered on both the north and west slopes of the hill. Excavation on the north side of the hill exposed a further 25 m of the wall, for a total length now of 45 m, revealing for the first time the extent and scale of the Lydian terracing program at Sardis (fig. 15).

Field 55 is an artificial terrace below Field 49 and ByzFort, near the center of the ancient city (figs. 1, 2). In the early Roman period the terrace belonged to a sanctuary of the Imperial cult, whose temple is called the “Wadi B temple” (fig. 16). In the later Roman period, the sanctuary was destroyed and the area reused for domestic and other purposes. A massive wall was built of spolia, mostly statue bases and architectural fragments from the Wadi B temple, and the terrace itself was repaired or strengthened with spolia. A room was later created in the corner between the spolia wall and the terrace. Another structure, probably a house, was built on the terrace itself (fig. 17). These later Roman buildings were apparently destroyed by an earthquake leaving the spaces filled with fallen marble and masonry.

Excavation reached floor levels both outside and on top of the terrace, and revealed rich assemblages of metal, glass, ceramic, stone, and other artifacts broken in situ on the floors and buried by the earthquake debris above, a sort of Late Roman Pompeii (figs. 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28). In the lower room, artifacts included two sigma tables, two marble platters, a polykandelon of bronze and glass, with suspension chains, a lamp stand, two bronze jugs, and other bronze vessels, iron agricultural implements, ceramic basins, jugs, and other artifacts (figs. 29, 30).

On top of the terrace, we completely exposed one room, with a tile floor and painted walls (figs. 31, 32). Among the artifacts here were three iron folding chairs; a fourth chair was found just in front of the door in the northern room (figs. 33, 34). A scales and weights, glass artifacts, pottery, and other artifacts were found on the floors (figs. 35, 36, 37).

The date of the earthquake and collapse was further refined with the discovery of many more coins from the destruction level. In previous years the latest coins dated to ca. 408-410 AD. Coins discovered this year in the destruction debris, however, include coins of the fifth and sixth centuries AD, the latest of which is a follis of Maurice or Heraclius, dating to 584-612 AD. The earthquake could not have taken place before 584 AD; but the majority of the coins circulating at that time date to hundreds of years earlier.

Excavation of the marble-paved road in front of the Synagogue continued as part of the Touristic Enhancement Project (below) and to reveal more of the monumental three-bayed arch discovered last year (Trench RT 15.1; figs. 1, 2, 38, 39, 40, 41). Expansion as far as the modern highway uncovered most of the southern set of piers that supported this arch, and the remains of its fallen superstructure. We have now exposed as much as possible of the footprint of this structure without cutting the modern road.

Many blocks of the superstructure were discovered fallen among these piers and onto the pavement. No further voussoir blocks from the large, 13-meter arch were discovered, but only voussoir blocks from the smaller vaults. This reinforces the hypothesis proposed last year that the superstructure fell more or less in place in an earthquake, and the blocks were broken up and burned for lime but not moved a great distance.

Further study of the arch and contemporary structures throughout the Roman empire suggests that it may have been as tall as 24 m high, taller than the Marble Court nearby and completely dominating the surrounding landscape (figs. 42, 43, 44). No monumental arches this large or larger are known to us; this does seem to be the largest such arch in the Roman world. After excavation this year we now estimate that only about 2.58% of the volume of the arch survives. The date of the arch remains uncertain, but it is later than the mid-first century AD and probably earlier than the third century.

This was the second full-scale year of the five-year project to remove biofilm from the temple of Artemis, using ammonium quaternate biocide in a new technique developed by Sardis conservators Michael Morris and Hiroko Kariya (fig. 45, 46). This year a larger team of 12 women cleaned the South Stele Bases, more of the south wall of the cella, twelve of the peristyle column foundations, seven of the eight foundations of the cella colonnade, and one of the column foundations of the west porch (figs. 47, 48, 49). We also cleaned three of the capitals now set up around the temple, and displaced blocks set to the north of the temple. Much of the work, however, focused on the standing columns of the east end of the temple, three of which were cleaned (figs. 50, 51, 52).

Other work included creating 3d models of Church M using two different techniques, laser scan and photogrammetry from a small drone. This will allow us to document the year-by-year movement of the walls and develop methods of reinforcing the foundations.

At the request of the Ministry of Culture we continued to work on a stoa to house four Hellenistic column capitals now placed around the temple. We applied to the II. İzmir Koruma Kurulu for permission to build the stoa last summer, and permission was granted in spring 2015. During the summer we leveled the area chosen for the stoa, and constructed bases for the four capitals.

As in 2014, significant effort was required to consolidate the wall paintings in Field 55 and to clean them so they could be drawn. Last year’s treatment and roofing was a success: despite torrential rains, the areas treated last year remained firmly attached to the walls. We therefore followed the same procedure, using injection grout and lime putty to adhere the plaster to the walls and then carefully cleaning the accretion with ethanol and deionized water.

Work continued on the Touristic Enhancement Project, designing protective roofs over the Synagogue and mudbrick fortification. SmithGroupJJR used its laser scanner to make an accurate 3d model of the Lydian Fortification and its surroundings, and of the Synagogue, to facilitate and check the design. We are now completing the design work for the Synagogue roof.

A commission from the Ministry of Culture visited Sardis with the goal of studying the site boundaries and development. Sardis is unusual in its wide extent of occupation and tombs (fig. 53). Occupation in the Lydian period extends as far south as the temple of Artemis and fields opposite the sanctuary, and as far north as modern Sart Mahmut, a distance of over 2.5 km. Tombs belonging to the city extend even further, and satellite communities, extramural sanctuaries, and other settlements range beyond this. The protection of this wide area is a matter of ongoing concern.

The latest volume of Sardis Reports, Churches EA and E at Sardis by Hans Buchwald, was published in spring 2015. Ongoing research and publication projects include excavation sectors, architecture of the Temple of Artemis and the Wadi B Temple, sculpture, wall painting, coins, human bones, pottery, inscriptions, and other artifacts.

A unique event last summer was an open-air performance of Kamran İnce’s Sardis Symphony, performed by the Bilkent Symphony Orchestra on June 17, 2015 in front of the temple of Artemis. The symphony was commissioned in 1999 by the former director of the excavation, Prof. Crawford H. Greenewalt, jr., but had never been performed before in Turkey. The movements of the symphony describe in music the dramatic topography of Sardis: the Tmolus mountains, the Hermus river, the Acropolis, Necropolis, and Bin Tepe. Kamran İnce himself came to conduct the orchestra, and at his suggestion, they added a number of arias from his opera The Judgement of Midas, which was also commissioned by Prof. Greenewalt. The concert was held at sunset, against the backdrop of the Acropolis bathed in the red light of the setting sun, and the two columns of the temple illuminated from below (figs. 54, 55). Between 900 and 1200 people attended, including the local workmen and their families.

Another event in honor of Prof. Greenewalt was the opening of the Greenewalt Library at Ege University, Department of Archaeology, on June 3, 2015. Prof. Greenewalt had donated his scholarly library of some 4,500 books to Ege University, and Prof. Gül Gürtekin-Demir, Dr. Ümit Güngör, and their students Gencay Öztürk and Sinem Çakır had catalogued and shelved them in a newly-renovated space. The opening was held in conjunction with the 2nd KERAMOS symposium on ancient ceramics, organized by Prof. Gürtekin-Demir in Prof. Greenewalt’s honor.

• 1The triangle is now published by Will Bruce and Kassandra Jackson Miller, “Towards a Typology of Triangular Bronze Hekate Bases: Contextualizing a New Find from Sardis.” Journal of Roman Archaeology 30: 509-516.
• 2In 2017 it was discovered that this wall is actually an earlier Hellenistic feature, and not Lydian.
• 3In 2016 it was discovered that this wall is actually an earlier Hellenistic feature, and not Lydian as originally believed.