Sardis, 2014

Archaeological fieldwork at Sardis was carried out from June 7 until Aug. 15, with an extended season for site conservation from Aug. 16 until Oct. 31. The team included 59 archaeologists, staff, and specialists. We were fortunate in having Özcan Şimşek of the Istanbul Archaeological Museum with us as representative of the Ministry of Culture. We 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, for their help and support.

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, 3). These two spurs were enclosed in the Lydian period by monumental terrace walls, 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 2014 were to uncover more of the Lydian buildings on top of this terrace.

A trench in the center of the hill begun in 2012 was expanded to 15 x 24 m (figs. 4, 5). Previous excavation had uncovered just the top of a limestone wall, which is exactly aligned with the Lydian limestone terrace wall in the south. As in previous years, a number of early Byzantine burials were scattered through this area, bringing the total number of burials on this hill to 41 (figs. 6, 7).

Roman remains were badly preserved here, but seem to belong to a large, perhaps open space with pithoi set into its floor (fig. 8).

A complex series of Hellenistic walls includes one room built partly of reused limestone blocks, with a door in its south wall (figs. 9, 10). The room seems to have been a basement, reached by a stairway, now destroyed but originally built on the stub of an earlier Hellenistic wall. Ceramics from the fill under the floor suggest a date in the first century BC. A landing at the bottom of the stairs was paved with four fired bricks, an early use of this technology in Anatolia.

Beneath this basement room was exposed a wall built of limestone ashlar blocks, the first such wall uncovered on this hill other than the terrace (fig. 11). In the foundations of the wall were built two faceted blocks, probably originally wall base blocks but here reused as spolia. Its relationship to the limestone terrace wall just to its west is obscured by the Hellenistic walls above. Further east, a stone packing on the other side of the Hellenistic room might be packing for the Lydian terrace, which would then be about 7 m wide. Notable Lydian finds from this trench include a terracotta revetment showing a charioteer (fig. 12); a bronze fibula fragment; and a very fine fragment of an Ephesian Ware dish (fig. 13).

At the north end of the hill, expansion of a trench dug in 2012 and 2013 revealed at least three phases of Roman buildings (figs. 14, 15). These were built over substantial Hellenistic foundations of a room 5.5 m wide, its subterranean foundations 2.6 m high and 1.0 m thick (fig. 16).

Earlier excavation here had uncovered the Lydian terrace wall of the 7th c BC, and a mudbrick and timber wall, 6.5 m below modern ground surface (figs. 15, 17). Remains of the 6th century BC phase, however, were almost entirely missing; and we hypothesized that a cut in the southern half of the trench, containing mostly Lydian material but with a significant mixture of Persian-period pottery, might represent a robber’s trench for some 6th century structure, such as a terrace wall, that had been entirely robbed out.

The complex Roman and Hellenistic remains prevented us from reaching Lydian strata, but one important discovery was a straight cut in the bedrock oriented to later walls here, predating the Hellenistic wall and so presumably made in the Lydian period (figs. 15, 17). This could be the south face of a feature that was entirely robbed out, such as a terrace wall; if so it would be almost six meters wide.

A second excavation sector was a lower terrace, known as Field 55 (figs. 2, 18). Previous excavation had showed that in the early Roman period this was a sanctuary, probably of the Imperial Cult, with a temple known as the Wadi B temple. The sanctuary dates to the 1st century AD; it was destroyed sometime during the Roman period, and many of its blocks were reused in a later, massive building on the east side of the terrace, constructed almost entirely out of spolia. This later building had been destroyed, probably in an earthquake. On top of the terrace were two rooms, perhaps of a later Roman house.

We expanded the area to more than 300 square meters to uncover more of the spolia building and walls to its north built of brick and fieldstones, with a door on one side (figs. 19, 20). Excavation did not reach the floor, however, but stopped at a layer of collapsed masonry, the remains of these internal walls which apparently fell in before the marble walls.

Among the spolia were architectural fragments belonging to several different structures. Study of these architectural fragments allowed a preliminary reconstruction of the superstructure of the Wadi B temple, as well as other buildings of the sanctuary. Sculptural fragments discovered in 2014 included a fragmentary Julio-Claudian head, and a later female head wearing a turban (fig. 21).

Among the numerous small fragments of inscriptions from the spolia is one recording that “The settlement of the Hekate-worshippers (honored) [ - - ] Aquila Ma[rcel?]lus, son of Laenas, the Asiarch,” documenting the existence of a community dedicated to Hekate, and another of many examples of Asiarchs honored at this sanctuary. A round altar reads “The association (thiasos) of the ‘spearmen’ dedicated (the altar) to Caesar Imperator Maximus and to Ba(k)chos Dionysos, son of Zeus; [ -deleted male name- ] took care of its erection and of the building.” According to Georg Petzl, it probably honored Octavian after Actium but before he was granted the title Augustus, dating the stone to between 31 and 28 BC.

From the later phase of this area, when the sanctuary had gone out of use and its blocks rebuilt in the spolia building, came a number of surprising discoveries. Among these is a reused Roman pedimented stele, incised with a menorah, lulav (palm branch) and ethrog (citron; fig. 22). This was found in a later, Medieval stratum; but a second menorah was found incised on the terrace wall itself (fig. 23).¹

On top of the terrace are at least two rooms, which are contemporary with the spolia building (fig. 24). Both had painted walls, but the wall paintings were so fragile that we could only expose and treat one wall. This reproduces colored marble revetment with pilasters standing on tall plinths (figs. 25, 26, 27).

These rooms too were found filled with collapsed and burned debris, very probably resulting from the same earthquake that destroyed the spolia building. The latest coins and pottery from the room date to ca. 408-410 AD, but this only provides a terminus post quem for the earthquake in the fifth century AD.²

A third excavation sector was opened as part of the Touristic Enhancement Project, which has been a central effort of work in recent years. We continued the design of two roofs, over the synagogue and the Lydian fortification. While we finish the design, however, we wished also to open more area to link these two sectors, and to demonstrate to visitors the remarkable continuity here, to show how the late Roman road passes directly over the Lydian gate of almost 1,000 years earlier, and under the modern Izmir-Ankara highway of 2,000 years later.

A large trench was therefore opened in front of the Synagogue, in order to expose more of the marble-paved road (fig. 28). The road surface, however, proved to be completely covered with large marble blocks, which had fallen here from a great height and smashed into the pavement (figs. 29, 30, 31, 32). Many of the blocks were then deliberately broken up after they had fallen, presumably to burn them for lime.

Study of the blocks showed that many of the blocks were voussoirs, in two main sizes. The smaller ones of ca. 4 m diameter would fit four piers excavated just north of here in 1963, and reconstructed as a tetrapylon (fig. 33). The larger voussoirs, however, measured approximately 13 m in diameter, far too large for this tetrapylon. Excavation uncovered the edge of a corresponding set of piers on the other side of the road, 13 m away and so exactly fitting the voussoirs, and suggesting that the road was spanned by a monumental marble arch. We may therefore restore a three-bayed arch set on eight piers, like the arches at Anazarbos or Thessaloniki (fig. 34, 35).

The structure is 8.6 m thick and, if symmetrical, would be about 33-34 m wide. This makes it among the largest arches known in the Roman world, with a 13 m central span. By great fortune the keystone on the inner side of the arch is preserved, with two inscriptions: the upper broken away, the lower recording that the supervisor of the work dedicated a statue of Herakles [---]phylax from his own funds (fig. 36). The date of the arch has not yet been fixed with certainty. Prof. Petzl dates the inscription on the keystone to the first or second century AD. Architectural ornament suggests a date in the second century, but other features such as the lewises in the foundations might suggest an earlier date. The structure undoubtedly had a complex building history, and may have been re-erected on one or more occasions.

A very peculiar feature of the arch is that most of the large voussoirs were cut from reused column drums from the cella of the Temple of Artemis, about 1200 m away. It seems that this earlier building was cleared of columns in the Roman period, and many drums reused here. This is an interesting early use of Roman spolia, and is paralleled by the Temple of Apollo at Corinth in the first century AD.³

Site conservation projects included work on Church M, the North Stoa of the Sanctuary of Artemis, Sector PN, wall paintings at Field 55, the Bath-Gymnasium Complex, and the Synagogue. The largest project, however, was in the temple of Artemis. In the century since it was excavated, the temple has become covered with lichen and black cyanobacteria, which both damage the marble and also provide a “blackboard” for visitors to write on (fig. 37). For the past three years Michael Morris and Hiroko Kariya have developed a new method to kill the biological organisms that have colonized the marble (fig. 38). A dilute solution of Preventol biocide is sprayed onto the stones and covered with wet burlap, also saturated with Preventol, which is then covered with plastic. The area is left covered for about 5 days, depending on the degree of infestation and the situation of the stone. During this time the Preventol soaks into the stone and kills the cyanobacteria deep within the marble, and softens the lichens. At the end of five days, the area is further scrubbed and briefly washed. Although the marble looks brown and blotchy after cleaning (fig. 39), the biocide continues to kill the bacteria for months, and it takes a year or more to complete the process. The process is slow, but gentle and effective; it does not simply clean off the marble blocks but directly addresses to fundamental problem, the infestation by colonies of living organisms. It works not just on the surface but deep within the stones; and most important, it is impossible to overclean the stone, and the technique leaves the surface patina undamaged.

After three years of development, a five year project was begun this year. The cleaning went more quickly and effectively than we had expected, and we continued the season for another two months, allowing us to clean much of the cella walls and the interior bases, leaving a large part of the temple as it looked in antiquity (figs. 40, 41, 42, 43).

Research and publications were an essential part of the season. A new project is a Hellenistic Working Group of archaeologists and historians, to study Sardis as a Hellenistic capital both locally and regionally, and the changes the city underwent as it was transformed from a Persian satrapal capital to a Greek polis. Work proceeded on new publications of excavation sectors such as HoB, buildings including the Synagogue and the Temple of Artemis, and artifacts including coins, inscriptions, pottery, and terracottas. The long-awaited publication of Churches EA and E by Hans Buchwald was published this spring. And finally, this new web site (http://sardisexpedition.org) includes all our publications for free download, an interactive database of artifacts, and essays in Turkish and English.

(Adapted from Kazı Sonuçları Toplantısı 37.2: 147-164)

• 1See Rautman 2015.
• 2Excavation in 2015, after this report was written, recovered coins of the sixth century AD from this same destruction layer.
• 3Frey 2015.