The Synagogue’s Surroundings
The area inside the so-called portico, which is attached to the western synagogue wall (Sqs AC 28–30), was fully excavated. When the last remains of the ‘tesserae layer’ (Zangenberg et al. 2013) and the fill underneath it were removed, traces of trimming and leveling, which may indicate building activity that predates the synagogue, could be discerned on the bedrock. Excavations in 2013 further indicated that the ‘portico’ was built in two phases. The first phase consisted of three square pillar bases on each side of an entrance; in the second phase, the spaces between these pillar bases were filled with fieldstones to form a continuous wall (W7073). The distances between the pillar bases in each set equal the distances between the pillar bases inside the synagogue (c. 1.8 m), and the space between the pillar bases that directly flank the entrance corresponds to the breadth of the western synagogue entrance (L7094). This symmetry suggests that the pillars reflect a comprehensive building plan that belongs to the first construction phase of the synagogue.
Cleaning activities on both sides of the balk between AC 31 and AD 31, south of the southern synagogue wall, exposed the bedrock, which dramatically drops in this area. Below the topsoil and a compact layer containing much plaster was a hard-beaten earth and plaster floor, which sealed a thick layer of loose fill that contained a very large quantity of highly fragmented pottery, dating to the late first – late fourth centuries CE, i.e. earlier than the rest of the pottery which was found in and around the synagogue. This fill extended down to the bottom of the southern synagogue wall (W7131). Corresponding to the fill, were very short segments of walls that could either belong to the foundations of W7131 or to an earlier building of unknown character. Below this layer, another compact beaten-earth floor covered a small patch of soil containing very little pottery and a handle of a stone mug. Here, thepottery dated between the second and the late fourth centuries CE. The character of the pottery assemblages indicates that both pottery-bearing layers were brought from somewhere else on the site and dumped here to level the bedrock and prepare it for building the synagogue, shortly after 400 CE.
In order to further explore the area outside the southern synagogue wall, the excavation was expanded eastward into AE 31. Apart from W7163, which runs perpendicular to the southern synagogue wall and possibly belongs to an earlier building, remains of a narrow platform with three steps were found (W7338). This stepped platform may originally have provided access to a neighboring building or to an upper story of the synagogue. Apart from this feature, the entire excavated area south of the synagogue was open and covered with a beaten-earth floor.
Excavations in 2013 also targeted the area north of the synagogue, where at least two building phases, corresponding to destruction layers, could be distinguished just north of the newly discovered room attached to the northern synagogue wall (W7114). Since stratification inside the synagogue is meager and the few finds that were retrieved from probes underneath the synagogue floor were mostly non-indicative, the area north of W7114 will be crucial for reconstructing the building history of the synagogue.
The excavation was expanded into the eastern area of the synagogue, fully exposing the entire building down to the latest plaster phase of the synagogue floor (L7285). The fills below the topsoil were generally very homogeneous and consisted of firmly packed light brown soil, with many cobbles, small boulders, and chunks of plaster, along with numerous architectural fragments. The pottery was Late Roman and Byzantine in date, with a few sherds from the Crusader period—an assemblage which matches quite well other loci in the fill and from the stone tumble above the synagogue floor, which were excavated in previous seasons. The closer the loci were to the bemah in Sq AE 30, the more architectural fragments were found. Square AE 29 farther north, however, contained almost none.
Stratigraphic balk removal revealed large parts of the plaster floor and exposed the still-unexcavated section of the bench running along the southern wall (L7133), west of the southern entrance. Here, the excavation of a stone seat (B20587; width c. 0.55 m, depth c. 0.4 m, height c. 0.5 m; Figs. 4, 5) was completed. The seat was set on a bench (W7049), and two steps that were attached to the bench directly in front of it (L7240) served as a footstool or for access. When discovered, the seat was still plastered onto the bench. It was made of a single limestone block, and was probably used by the leader of the community during itsgatherings. This seat is the first ever found in situ in northern Israel (Zangenberg, forthcoming [b]).
Balk removal in the northern section of the synagogue revealed that a stone bench (W7140) ended approximately where the western aisle ends and the nave begins. In 2013, two small stretches of the bench were found farther east, the second forming a corner with what we initially called a ‘stylobate wall’ (L7290). The join between the bench and this ‘stylobate wall’ demonstrates that the latter was actually a low bench and not a wall.
Also exposed were the entire eastern synagogue wall and a floor (L7378) east of Bench 7290/7287, which runs north–south between the pillars. Almost no traces of the plaster floor were preserved in the eastern aisle. Instead, patches of roughly dressed flagstones form an uneven, floor-like surface, very different from the floor of the western aisle and the nave (Fig. 2). Further study is necessary to determine to what extent these remains may represent structures that predate the synagogue.
Along the entire eastern side of the synagogue, ancient efforts at structural repairs were visible. Several walls running parallel to each other, and remains of two rows of benches along the eastern synagogue wall (W7193), indicate intensive building activity. Outside the synagogue, to the east of W7193, lay a thick tumble of large basalt boulders. Some of the boulders were roughly worked and appear to have fallen from W7193. The appearance of others suggests that they may have belonged to walls abutting W7193. One wall (W7326) clearly undercuts it, and therefore seems to predate the eastern synagogue wall. In addition, W7193 appears to have been repaired. As mentioned above, no traces of the plaster floor that is found elsewhere were identified in Sqs AF 29 and AF 30. Instead, some patches of a much coarser, possibly earlier, plaster layer were discovered.
It also appears that there were actually two lines of benches in the eastern part of the building: one running north–south between the pillars separating the nave from the eastern aisle (L7287/L7290), likely to be associated with the second phase of the synagogue, and another running along W7312 and W7193inside of the eastern synagogue, likely belonging to the early phase. Further study is necessary to understand the extent to which the sequence of benches implies spatial and functional rearrangements of this part of the synagogue building.
Another bench was found along the southern synagogue wall, inside the synagogue (southern end of Sq AE 30). This bench had two phases: first, a line of low blocks was installed along the wall; at some later point, a fallen column was reused as an additional bench (L7280) and was plastered onto the floor. In 2013, a small patch of mosaic was found just to the east of Pillar 7287, partly continuing underneath Column 7280—a nice, stratigraphically relevant feature. The fact that far more than 50,000 tesserae were found in and around the synagogue throughout the excavation seasons certainly indicates that large areas inside the synagogue had been covered with mosaic.
A careful examination of the floor inside the synagogue revealed several layers of gray plaster in the western aisle and in the central nave. While the plaster in the western aisle was almost completely preserved, the floor in the central nave was heavily damaged by fallen debris from the roof, revealing the bedding below the floor and the fact that in many spots there were several superimposed thin layers of plaster. This layering indicates that the plaster floor was refurbished and repaired several times (Fig. 2).
In 2012, Sq AE 29 was opened to expose the northeastern end of the synagogue down to the plaster floor. Just like in Sq AE 30, the fill was brownish gray and contained loose soil with pebbles, cobbles, boulders and chunks of plaster. The pottery was mixed: Late Roman, Roman–Byzantine and later pottery. In this fill, three well-preserved Byzantine gold coins were discovered: two solidi minted by Justin II (565–578 CE) and one tremissis of Tiberius II (578–582 CE).
Bemah. A large, almost square structure attached to the center of the southern wall, on the inside, was found in 2011 and fully excavated in 2012 and 2013. In accordance with recent synagogue research, we call this the bemah, i.e. a raised platform that supports the Torah shrine (see, e.g., Hachlili 2013:163–184). Excavation inside the bemah brought a number of interesting features to light: a square room measuring (c. 2 × 2 m; Fig. 6), containing small benches or shelves; a support pillar; and the remains of at least two floors. A plastered area right in front (i.e., north) of the bemah shows signs of patching and reworking that may be associated with structural changes of the bemah itself, and/or the synagogue building. A deep hole was dug through the floors into the ground below the bemah. In it, a bronze oil lamp from the Roman period was discovered (Fig. 7) along with Roman and Byzantine pottery.
Stone Table. A carved basalt stone shaped like a low table (c. 0.7 × 0.7 m, c. 0.35 m high; Figs. 8, 9) with four feet (length 0.15 m) was found in 2012; the area around it was carefully examined at the end of the 2013 campaign. This object is certainly among the most spectacular finds unearthed at Horbat Kur so far. It is slightly damaged on the upper left corner of the front face and lacks the right foot on the rear. The stone was found in secondary use as part of Bench 7290 at the northeastern part of the synagogue, between Sqs AE 29 and AF 29. Its original context and function are unknown and can only be determined on the basis of a careful analysis of its form and decoration.
All four sides and the top of the basalt stone are carefully decorated with figurative elements (front face) and geometric patterns (other sides and top); the bottom is undecorated. The top of the stone features three rectangles set one inside the other (0.30 × 0.43 m, 0.21 × 0.33 m, 0.16 × 0.27 m) and cut successively deeper into the stone; the surface of the central one is much smoother than the rest of the stone. On the front face the stone displays the following carved objects (from left to right; Fig. 9): a lampstand (candelabrum) extending down the entire side, including the foot; an unknown object; a jug with high handle (damaged in antiquity); a ladle (simpulum) and a vase without handles in the center; a jug with a handle; an unknown object; and again a lampstand. The objects between the flanking candelabra appear to stand on a table or board. The rear of the stone table displays a different decoration, consisting of three elements: three rectangles, one inside the other on each side, flank by a frame (c. 0.20 × 0.23 m) with four squares. The central element may represent a two-winged door. Both other sides are decorated with a broad set of three rectangles one inside the other extending across almost the entire side (c. 0.12 × 0.35 m, c. 8 × 31 cm, c. 3 × 27 cm) and triglyphs above each foot. At the bottom of each foot there is a circular decoration (disc? wreath?). All the geometrical elements are well known from Hellenistic–Roman art and architecture. The objects on the front do not have any particularly Jewish connotations, but rather resemble vessels known from the Greco-Roman world of dining (Zangenberg, forthcoming [a]
The stone table was removed for safety reasons, exposing at least two layers of plaster below it. The table rested on the lower plaster floor, and a later plaster layer sloped up to its front side and partly covered the feet that face west. A gold coin that was minted by Justin II (565–578 CE) was found right between the lower plaster layer and the basalt table, and establishes a terminus ante quem for the construction of the lower plaster layer and a terminus post quem for the construction of the bench, including the insertion of the basalt table.
Cistern 7187 (Fig. 10)
In 2012, a probe was dug through the accumulated sediments in Cistern 7187, which was found at the end of the 2011 season north of the synagogue. The cistern (floor measuring c. 7.1 m north–south, 8.5 m east–west, depth 6.5 m) appears to be a naturally formed cavity whose walls were plastered. Several niches or ‘shelves’ are visible along the bottom part of the natural walls. One of the most notable of these is located on the eastern wall, where a high ‘shelf’ or ‘bench’, running south and then following the curve of the cistern to the west was hewn.
The probe was excavated along a north–south axis down to the cistern’s floor (c. 0.8 m deep). It provided several important insights on the build-up of the sedimentation inside the cistern and its contents. Three layers of sediment were identified. The uppermost consisted of numerous small potsherds, pebbles and plaster fragments—material that fell into the cistern after it had gone out of use and before the opening was sealed, sometime at toward the end of the Byzantine period. The pottery that was retrieved from this layer dates no later than the Late Byzantine period, with the possible exception of very few pieces that may date from the Early Islamic period. The second layer consisted of very compact gray mud with very few stones, pebbles or plaster fragments. None of the material dates later than the Late Byzantine period. This layer probably represents the use of the former cistern as a local dump. The lowest of the three layers (thickness c. 0.6 m) was clayish and consisted of fine mud; it very likely represents silt accumulated during the cistern’s ongoing use. The silt layer contained many complete or near-complete jugs and cooking pots (e.g., Fig. 10:1, 2), as well as a considerable quantity of faunal remains. The vessels were randomly distributed in the soil across the entire locus, between approximately 5 and 40 cm above the cistern’s floor. The absence of soot on the cooking pots indicates that they were not used for the preparation of food, but for drawing water, a feature that corresponds well with the findings from a contemporaneous cistern in Meiron (Meyers, Strange and Meyers 1981:94–105). The vast majority of the vessels fits well into the local Galilean typology; some subtypes were found here in complete form for the first time. Only one decorated jug—unknown in the Galilean repertoire so far—is perhaps imported (Fig. 11). According to preliminary analysis, all vessels from the silt layer date from around 400 CE; future research will have to show if and to what extent these dates correlate to the synagogue phases. A large number of soil samples were taken from the horizontal section of the probe dug into the cistern, and are now being analyzed. The amount of plant and small-animal remains (parasites, mollusks, rodents) is remarkable and will allow us to draw further conclusions about the regional climate and the living conditions of the ancient inhabitants of Horbat Kur.
The 2012 and the 2013 campaigns confirmed the assumption that the Horbat Kur synagogue followed a basilical-broadhouse layout and very likely had two main building phases. In addition, evidence for a pre-synagogue activity may have been found in Sqs AC 31 and AF 29, while post-synagogue structures were discovered in Sq AF 30. Especially important are several stone features (furniture?), such as the stone seat, the massive bemah and the decorated stone table.
The probe in Cistern 7187 also proved fruitful. The examples of complete vessels of common ware will supplement the Late Roman Galilean repertoire. Soil samples from the section will make it possible to reconstruct the climate and vegetation during the period before the cistern was sealed near the end of the Byzantine period.