The 2018 Season of Excavations at Tell es-Safi/Gath: A First Summary
Aren M. Maeir
This year’s season of excavations at Tell es-Safi/Gath was conducted from June 26 to July 20, 2018. It was the 22nd season of the project and the 21st season of excavation. The team was comprised of archaeologists, students and volunteers from all over the world, including groups from Bar-Ilan University, Brigham Young University, Colorado Christian University, University of Melbourne, University of Northern Colorado and Yeshiva University. The excavation staff included: Aren Maeir (director), Jeff Chadwick (field director, Area D East), Jill Katz (field director, Area Y), Maria Enukhina (field director, Area M; photographer), Vanessa Workman (field director, Area D West), Bracha Soibelman (registrar and team coordinator), Rotem Liran (logistics), Ron Lev and Liora Horwitz (archaeozoology), Sue Frumin (archaeobotany), Andrew Creekmore (Remote Sensing), Adi Eliyahu (materials science) and (in alphabetical order) Michael Aeschlimann, Brent Davis, David Kotter, Gemma Lee, Alice McInnes, Eitan Meir, Seth Rodriquez, Shemtov Sasson, Nohar Shahar, Nati Orbach (area supervisors and assistant supervisors). The team stayed at the Zafit Regional School at Kibbutz Kfar Menahem.
In contrast to previous seasons of excavation in which we had always excavated in various areas in the “upper city” (and areas in the “lower city”), this season was the first season in which we only excavated in the lower city. As the project is now nearing its 25th anniversary, and is close to completion of field work, for the next few years, we will concentrate most of our efforts on the study of the lower city. While clearly, there is still much potential for the study of the archaeological remains in the upper city, we believe that what we have revealed over the last 2 decades provides a nice cross section of the cultural history of the upper city throughout its major phases of occupation. In depth, further study of the upper city will have to await new projects on the site, sometime in the future.
The reason that we have chosen to focus on the lower city is the fact that over the last decade, in the extensive excavations in Area D, and more recently in Areas K and K2, well-preserved remains of the Iron Age were discovered, immediately below surface, in various parts of the lower city. In addition to this, based on aerial and terrestrial photos from the first half of the 20th cent CE, as well as from the archaeological evidence from the survey and excavations, it appears that the lower city was hardly settled in modern times, and apparently, for the most part, was never substantially settled after the Iron Age IIA destruction of the city by Hazael, King of Aram Damascus, ca. 830 BCE.
This picture was further strengthened through remote sensing conducted in the lower city during the 2017 season. Following an extensive magnetometry survey conducted by Andrew Creekmore and his team, it was quite clear that there were many potential areas for further excavation (most likely of Iron Age dating based on their orientation) in various parts of the lower city.
All this indicated the very high potential, that a combination of excavations and remote sensing of various kinds, might enable us to understand the urban plan of and extensive portion of the lower city of Gath during the Iron Age, something that is rarely available in other large Iron Age cities in the Southern Levant.
With this backdrop, the objectives of the 2018 season were set as follows:
- Define two new areas of excavation in the eastern portions of the lower city, based on several promising “anomalies” that could be seen in the magnetometry survey, to “ground truth” a sampling of the remote sensing survey.
- Continue excavation in Area D East, to further define and understand the fortifications and apparent gate found in previous seasons in this area. Particular focus would be placed on confirming that in fact there was a gate, understand how it functioned, and more clearly define the architectural and stratigraphic sequence in this area.
- Conduct small scale stratigraphic probes in Area D West, in the portion of this area where a metallurgical production zone was located in previous seasons. The aim was to define the borders of the metallurgical activity, and better define the stratigraphic and contextual sequence below it.
Results of the 2018 Season:
Area D West (supervisor V. Workman): Excavations in this area were limited to several small stratigraphic probes that were conducted in and around the metallurgical production zone that had been discovered in previous seasons. In addition to defining the phasing of several of the architectural features in this area, the stratigraphic phasing in this location was rechecked. A very interesting result relates to the dating of the metallurgical activities. Up until now, all evidence of these activities came from Stratum D3, the stratum destroyed in the “Hazael destruction” of ca. 830 BCE. And in fact, for the most part, this was the stratigraphic picture this season as well. But in one of the probes, portions of a crucible was found in stratigraphic layer below Stratum D3, most probably Stratum D4. This may very well indicate that the metallurgical activities commenced earlier. This in fact makes sense as it seems likely that the Stratum D3 metallurgical activities are connected to the cultic activities and temple situation to the west. As these cultic activities are seen in the earlier Strata D4 and D5, earlier metallurgical activities would fit in very well with this.
Area D East (supervisor J. Chadwick):
Excavations in Area D-East in 2016 and 2017 suggested that a gateway existed in the gully running from the tell to the stream bed. In 2018, the gate area was defined and identified. Additionally, the stone foundation of the Iron Age city wall was identified, as well as several rooms inside the wall line. Dramatic evidence of defensive actions taken by the inhabitants of Gath during the Aramean siege of the 9th century BCE was also discovered.
The segment of the Iron Age city wall excavated was some 14 meters long, and evidently continues eastward into unexcavated squares. Ceramics recovered from surviving portions of the brick superstructure of the wall suggest that it was built during the 10th century BCE, during early Iron Age IIA. At the west end of the excavated city wall length, a protruding gate pier foundation was identified. This northern gate pier was oriented on a slightly inward angle from the line of the city wall, and was built atop the foundation of an earlier north-south fortification wall which has not yet been securely dated. Access into the city through this gateway apparently involved walking up from the streambed, up the gully southward, along the earlier fortification line, and then turning sharply left (south-eastward) to enter the gateway opening. A small court or plaza seems to have existed inside the gateway, flanked by domestic structures which abutted the city wall’s inner face line. This gateway does not seem to have been the main gateway into the lower city, and probably served as the “water gate.”
A fortification tower with two interior rooms was built against the outside of the Iron Age city wall, just east of the north gate pier, sometime after the city wall itself had been initially erected. That this two-room tower was a later addition to the fortifications seems clear since its stone foundation extended higher than the outside line of the city wall foundation, and its stonework was not integrated into the stonework of the city wall foundation itself. The tower’s construction is dated to later Iron Age IIA, in the 9th century BCE. The tower seems to have been built to enhance the protection of the city wall and gateway at this vulnerable location where the gateway gully met the streambed. An earlier north-south stretch of fortifications, built of very large stones, over which the northern gate pier was erected, is provisionally dated to Iron Age I. The Iron Age I/II (or early Iron Age IIA) city wall system seems to have been built to enhance and strengthen the earlier fortification system in this area along the stream bed. It may be that the gateway in this area was previously in use as an entry in the earlier fortification scheme.
During later Iron Age IIA, presumably during the late 9th century BCE Aramean siege of Gath, the gateway was closed in from the inside. Rooms south and east of the gateway, along the inside of the city wall line, were also filled with grey, ashy soil containing great amounts of LB and Iron I pottery sherds. These deep fills, which may have utilized soil from old LB/Iron I garbage dumps in the lower city, appear to have been a desperate effort to buffer and fortify the inside of the city wall line against siege breach efforts of the Aramean attackers.
Area M (supervisor M. Enuikhina):
The excavations in this area, situated in the eastern side of the lower city, were commenced this season, based on a series of linear anomalies, oriented east-west, north-south, seemingly looking like structures and roads, which were seen in the magnetometry conducted in the 2017 season. The excavations were aimed to “ground truth” the remote sensing.
Four squares were opened in this area with fantastic results. Mere centimeters below surface, rich remains of the “Hazael Destruction layer” were discovered. In all four squares, architectural remains of rooms filled with more than a half meter thick deposit of this destruction level, with scores of ceramic vessels (mostly smashed, but quite a few still whole). These vessels included a broad range of types, including small and large storage vessels, cooking and serving vessels, cultic related vessels (chalices), as well as vessels relating to various other functions.
Of particular interest were the two rounded stone basins discovered in two of the squares, very similar to stone basins previously found in Areas A and K. Most likely, these installations are olive presses, and perhaps serve as evidence of the importance of olive oil production in Iron IIA Gath. This is of importance as previously, some have suggested that olive oil production only commenced in Philistia in the 8th and 7th century BCE, centered at Philistine Ekron, and connected to the Neo-Assyrian Empire. If we are correct in the identification of these installations, it appears that olive oil production in Philistia was already important, at least as early as the Iron IIA (and perhaps earlier as well). It may very well be that the production and trade in olive oil was one of the economic strengths of Iron IIA Gath. And just as it has been suggested that Gath had a major role in the copper trade during this period, perhaps olive oil should be figured in as well.
Among the special finds from this season in Area M, several can be noted: a large jar that most of its body was covered in drilled holes – similar to several other jars found in Areas D and K in previous seasons; a chalice bowl with an application of an animal on its rim; and a complete, unbroken jar that was found in the last hour of the last day of excavation.
Further work in Area M will aim, inter alia, to expand excavations to understand the architectural contexts of these rooms, and to understand the function of the stone basins. In addition, we hope to be able to see whether additional features can be detected in the remote sensing, to enable to define more of these stone basins and related features, whether through excavation or only through remote sensing.
Area Y (supervisor J. Katz):
The excavations in this area, situated in the eastern side of the lower city, were likewise commenced this season, based on a series of five square shaped anomalies seen in the magnetometry, oriented NE-SW, seemingly looking like rooms or installations that had been exposed to high temperatures. As in Area M, the excavations here were aimed to “ground truth” the remote sensing.
To a large extent, the results in Area Y were quite surprising. As opposed to all other areas that so far have been excavated in the lower city (Areas D West, D East, K, K2 and M) where the Iron IIA remains were immediately below surface, in Area Y, Iron Age I remains were discovered immediately below the mixed upper sediments containing finds from modern through the Iron Age.
In fact, a large structure, built of thick brick walls were discovered in all the five squares that were excavated. In addition to this, three of anomalies seen in the remote sensing were excavated as well, and they were concentrations of burnt bricks. One these concentrations, on the NE excavated square was of particular interest. Here, a built installation with a seemingly gabled roof made of burnt bricks was discovered. This installation’s function is not clear. While reminiscent of pottery kilns, the finds associated with this structure have so far not provided a clear indication of its function. Whatever its function was, large concentrations of burnt bricks were found in its vicinity as well, including in a pit dug on its SE side. Extensive samples were taken from this feature and its surroundings, and hopefully, the results of their analyses (by A. Behar) will help clarify this installation’s functions.
On the southern side of Area Y, another interesting feature was discovered. Here, an extensive area (major portions of two squares) with layers of chalky material was discovered, seemingly laid out in purposeful manner. These chalk layers, apparently deriving from the nearby chalk cliffs of the upper tell, might have been part of a production process of building materials. In other areas of the excavation, and in many periods, building materials made from this chalk material have been discovered, including used as a plaster like material, within roofing and mudbricks and other features. Perhaps, these or similar construction-related materials were produced at this location.
An important characteristic of Area Y was the very poor preservation of artifacts in general. While there was sufficient pottery to date the features, in general, there was very little pottery, bones and other materials.
Perhaps the most unique aspect of Area Y is the Iron I dating of the features that were discovered very close to surface. This is quite different from all other excavation areas in the lower city, including Areas K and M which are only a few score meters away (to the east and south). In addition, the orientation of the architecture in Area Y is completely different from the Iron II architecture in other parts of the lower city. The finds in Area Y indicate that already during the Iron I, the lower city of Gath was quite extensive. Previously found to the west in Areas D West and D East, we now know for sure that eastern parts of the lower city were settled as well during the early Iron Age.
An additional point is that of the relevance of the finds from Area Y, from a methodological point of view, as an example of the possible pit falls of insufficient sampling strategies in the study of a large scale site. If the study of the lower city of Gath would have been limited to only a small sample of excavation areas, very different results might have arisen. Based on a limited sampling from few excavation areas, one might have assumed that the lower city was only settled in the Iron I (as seen in Area Y), or, based on other areas, mainly in Iron IIA, with some earlier evidence below this. Clearly then, this stresses the need for large scale sampling, using survey, excavation and remote sensing, to be able to truly understand the character and history of a large site, such as Tell es-Safi/Gath.
Remote Sensing (A. Creekmore):
The original plan for this year was to continue and expand the remote sensing in the lower city, and to supplement the magnetometric analyses conducted last year with a Ground Penetrating Radar survey of the lower city. It was hoped that this would provide additional, supplementary information to the already very exciting results of last year’s remote sensing. Unfortunately, Andrew Creekmore’s GPR equipment was not released by the Israeli customs during his entire stay in Israel, so this equipment could not be used. Despite this, thanks for Prof. Amotz Agnon of the Hebrew University of Jerusalem, we did manage to survey a limited portion of the lower city, with a GPR instrument that was provided by Prof. Agnon. Hopefully, this will serve as the first stage for an extensive GPR survey at a later date.