In 2015–2020, a research excavation was conducted in the Protestant Cemetery on the slopes of Mount Zion (License Nos. G-33/2015, G-57/2016, G-92/2017, G-56/2018, G-75/2019, G-54/2020; map ref. 221724–820/630809–43; Fig. 1). The excavation, undertaken on behalf of the German Protestant Institute of Archaeology (GPIA; DEI), was directed by D. Vieweger with the assistance of F. Schöpf (2015–2020), K. Soennecken (2015–2017; 2020), K. Palmberger (2015–2017) and J. Zimni (2018–2020) (area supervision), P. Leiverkus (measurements, photogrammetry, IT-support), Y.I. Sfez (Numismatics), O. Harush and T. Yashuv (drawing of findings), H. Maaß (plans; 2017), J. Reinhard (measurements, photogrammetry; 2017), R. Hunsdorf (measurements; 2016), L. Whittaker (archaeobotany) and M. Lavi (metallurgical laboratory). The project is financed by Gerda Henkel Stiftung und Düsseldorfer Wissenschaftsförderungsgesellschaft.
Archaeological research in the Anglican-Prussian Protestant Cemetery began with the explorations of Charles Warren and Charles Wilson, who discovered rock-cut steps along the cemetery’s northern border (Wilson 1871). Henry Maudsley investigated a larger area of the rock scarp along the southwestern hill of the compound during the years 1873–1874 (Conder 1875). This area was again explored by Frederick Bliss and Archibald Dickie in 1894–1897 (Bliss 1898). Among other finds, they uncovered a gate with three different thresholds (Fig. 2) dating from the Early Roman to the Byzantine periods, which they identified with Josephus Flavius’ ‘Gate of the Essenes’ (War 5:142–148). Almost a century later, between 1977 and 1988, Bargil Pixner, a monk of the nearby Dormition Abbey, re-excavated the area around the gate, with the aim of uncovering what he envisioned as a residential quarter of the Jewish Essenes sect associated with the ‘Gate of the Essenes’ (Pixner, Chen and Margalit 1989)—an idea which is widely rejected by scholars today. Nevertheless, the excavations by Bliss and Dickie and by Pixner uncovered a number of phases of the gate and remains of the Byzantine-period city wall to which it belonged.
The present excavation, encompassing part of the area excavated by Pixner and extending northeast of this area (designated Area I; Fig. 1), revealed remains from the eighth century BCE up until the Islamic periods (Fig. 3; Palmberger and Vieweger 2015).
No architectural remains dated to this period were uncovered in the excavation. Nevertheless, the excavation did yield a rather substantial amount of pottery sherds dating from the Iron Age II, suggesting that this area was in use at that time; however, the nature of this use remains unknown.
A wall uncovered in Pixner’s excavation to the west of the Byzantine-period city wall and identified as an earlier city wall was suspected as belonging to the Iron Age (Chen, Margalit and Pixner 1994). However, as the entire fill around it had been removed and the finds have never been fully published, its date remained unclear. Therefore, the excavation campaign in 2020 aimed to explore this wall more closely, by excavating a previously unexposed section of the wall. In contrast to Pixner’s results, the excavation suggests a Hellenistic/Hasmonean date for this wall.
Another wall, substantial and constructed of massive ashlars, was built in a general west–east direction to the east of the city wall uncovered by Pixner. As some of its ashlars were used as floor foundations in the Early Roman period, it is very likely that it too was built during the Hellenistic period, a situation similar to that of the fortifications at Samaria-Sebaste (Crowfoot, Kenyon and Sukenik 1942:28–31, Pl. XLIII:2).
The bedrock in the center of the excavation area is stepped (step riser 0.5–0.7 m). The steps descend from north to south, parallel to a north–south wall, which belongs to a building that is assumed to be from the Hellenistic period; this suggests that they were hewn in the Hellenistic/Hasmonean period. The northern edge of the stepped bedrock appears to have been altered to serve a different function, probably at a later period, and by the Byzantine period this part of the bedrock served as a bedding for a mosaic floor in a domestic room (see Room F, below).
Early Roman Period
A street which lead from Bliss’ and Pixner’s city gate to the northeast, into the inner city, and parts of which were exposed in previous excavations (Bliss 1898; Pixner, Chen and Margalit 1989) could be correlated with the Early Roman phase of the city gate. The street was paved with large stone slabs, which covered a drainage channel that ran below the street. Some of these stone slabs had been removed or destroyed, possibly by Jewish rebels fleeing from the invading Romans in 70 CE (Josephus War 6:370–371).
The layout of the architectural remains along the street seems to indicate that it was lined with houses, although robbing and reuse of the building materials has left few remains of these structures. The remains include several channels, some for supplying the houses with fresh water and others that served for draining wastewater into the main sewage channel below the street. One of these channels led into a rock-cut pool with steps descending eastward (steps riser c. 0.4 m)—a ritual bath (miqveh; Fig. 4)—most likely to supply it with fresh water.
Late Roman Period
The only architectural remain from this period is the gate’s middle threshold; coins found in the surrounding contexts attest to its date. The rest of the remains comprise mainly soil fills and a massive refuse dump (Fig. 5), containing organic waste, such as bones, charcoal and other burnt material. The pottery indicates that the dump was in use during the Late Roman period and ceased to function when it was covered with layers of chalk lime and sealed below later structures around the time that the Byzantine-period city wall was erected in the fifth century CE.
A fragment of a unique inscribed stone (height 20 cm, width 14 cm, thickness 9 cm; Fig. 6)—to date, the only one of its kind found in Jerusalem—bearing the cursus honorum of a person of senatorial rank (Eck, Vieweger and Zimni 2020) was reused in the Byzantine period as a building stone.
Nine rooms (A–I; see Fig. 3) dating from the fifth and sixth centuries CE were uncovered at about 0.8 m above the Street leading from Bliss’ and Pixner’s city gate to the northeast. Much of the architecture in the southeastern part of the excavation area, such as rooms G, H and I, was destroyed in the Islamic period when terrace walls were built in their stead. This applies as well to scant remains of additional rooms that could be observed to the east of the rooms and were also likely part of the Byzantine residential quarter. A mold for producing cross-shaped items (Fig. 7), found in a drainage channel north of Room B, and oil lamps bearing crosses and Greek inscriptions, found along the wall of this room, attest to the Christian affinity of this residential quarter. It is well known that Mount Zion became a flourishing center for Christian pilgrims during the mid-fifth century CE, when the church of Hagia Sion was founded nearby.
The Byzantine structures generally reused architectural elements robbed from earlier Roman structures, and in some cases the buildings’ foundations were set on bedrock. Rooms C and D were built directly above the Roman street, using it as a foundation. While the Roman street was no longer in use, the large drainage channel below it was still removing wastewater out of the city. A system of water channels belonging to this period was exposed between the city gate and the adjacent residential quarter (Fig. 8), apparently used for both water supply and drainage.
Room A. A floor made of beaten earth and pebbles was uncovered in this room. A rectangular plastered cistern was found below this floor, its well-head situated at the floor level. An accumulation of ash with charcoal and many iron nails covered the floor, likely deriving from a burnt wooden roof.
Room B. This room, exposed to the southeast of and adjacent to Room A, was vaulted and had a simple white mosaic as a floor. The entrance to this room was created by modifying the Early Roman rock-hewn miqveh. Also found in Room B was a clay water pipe that seems to have channeled rainwater into a small channel underneath the floor level through a small settling basin and into a cistern below the room. The fill within the cistern indicates that it was probably used as a trash dump subsequent to its period of primary use. A chalk-plastered installation lined with worked stones, possibly used to store food, was also found in Room B.
Room E. This room, adjacent to Room B, to its southeast, had a floor of flat paving stones. A metal hanger for an oil lamp of a type found in a monastery at Bet She’an and dated to the sixth century CE (Fitzgerald 1939: Pl. III) was found underneath the floor, suggesting that the building complex was constructed in the sixth century CE.
Room C. This room, built atop the Roman street, revealed the typical Byzantine plaster with triangular incisions and remains of a white mosaic floor. As the southeastern part of this room was destroyed by later structures, its complete layout could not be reconstructed. A plastered channel found outside of the room, abutting its northwestern wall, was oriented east–west, in the direction of the city gate; it therefore is likely to have been a drainage channel.
Room D. This room, adjacent to Room C, to its east, was nearly rectangular. It was probed in several places during the Bliss and Dickie excavation (Bliss 1894:253). The northwestern wall of Room D features a plastered water channel running into the drainage channel below the Roman street. The southeastern part of this room was founded on the hewn bedrock, which seems to have been worked into the shape of the room. Large in situ worked stones that were laid on top of the rock-cut foundation formed one of the walls of the room.
Room F. The possibility that another room, Room F, existed on the southeastern side of Room D is indicated by a continuation of the rock-cut foundation, which bore scant remains of a mosaic bedding.
Room G. This room borders the northeastern side of the rock-cut foundation of Room D. Its main preserved feature is a rock-cut channel leading into the drainage channel below the Roman street. A soil accumulation including tiny potsherds mixed with reddish brown earth formed the bedding for a floor south of the rock-cut channel. This floor was exposed only in the southwestern part of Room G, as the room extends beyond the excavation limits.
Room H. This room, east of Room G, had a plastered rock-cut wall delineating it on the northwest. The plaster was of the incised type, like the plaster in Room C. The room was paved with a mosaic floor made of tesserae of various colors (Fig. 9).
Room I. This room was separated from Room H by a thin wall. The small remnant of a mosaic floor was part of a simple white mosaic with a repeating floral pattern.
The Early Islamic remains consist mostly of fills. This area saw mainly the reuse of Byzantine structures and very limited building activity. Byzantine Rooms A and B were filled in and built over with plaster floors, into which small channels were installed; no other Early Islamic structural remains were uncovered in this area. The Byzantine cistern in Room A was reused in the Islamic period, as attested by small walls that were built above it to allow access from the above-lying living surface. The enclosed space above the cistern was entirely filled in (fill height c. 0.7 m) during the Late Islamic period, putting the cistern out of use.
An Early Islamic limekiln (diam. 2.2 m, depth 1.8 m) was built adjacent to Byzantine Rooms A and B, most probably destroying several other Byzantine rooms. The pottery from the limekiln dates from the Iron Age II through the Late Islamic period, the latter providing a terminus post quem for its filling and suggesting that it continued in use into the Late Islamic period. The surrounding area appears to have been levelled at that time employing massive chalk fills.
A noteworthy Late Islamic architectural feature comprises two northeast–southwest terrace walls built of unworked fieldstones, running across the site and further disrupting the Byzantine-period architectural remains.
The plaster floors and water channels of the Early Islamic period and the limekiln which was installed in that period and probably continued to be used in the Late Islamic period indicate that this area was used for industrial purposes throughout much of the Islamic periods.
The six seasons of excavation shed light on the chronology of the fortification walls of the southwestern hill of Jerusalem. They unearthed new finds pertinent to reconstructing the changing urban landscape of Mount Zion, mainly between the Early Roman and the Islamic periods. It seems that the nature of activity on the southwestern slope of Mount Zion underwent a radical change around the transition from the Byzantine period to the Islamic period. At this time, the affluent Byzantine residential quarter transformed into an industrial quarter.
Bliss F.J. 1894. Second report on the Excavations at Jerusalem. PEFQ 26:243–257.
Bliss F.J. 1898. Excavations at Jerusalem, 1894–1897. London.
Chen D., Margalit S. and Pixner B. 1994. Mount Zion: Discovery of the Iron Age Fortification below the Gate of the Essenes. In. H. Geva. Ancient Jerusalem Revealed. Jerusalem. Pp. 76–81.
Conder C.R. 1875. The Rock Scarp of Zion. PEFQ 7:81–89.
Crowfoot J.W., Kenyon K.M., and Sukenik E.L. 1942. The Buildings at Samaria. London.
Eck W., Vieweger D., Zimni J. 2020. Die Basis einer Ehrenstatue mit dem cursus honorum eines senatorischen Amtsträgers in Jerusalem. ZPE 216:273–278.
Fitzgerald G.M. 1939. A Sixth Century Monastery at Beth-Shan (Scythopolis). Philadelphia.
Palmberger K. and Vieweger D. 2015. Die Grabung im Anglikanisch-Preußischen Friedhof auf dem Zionsberg in Jerusalem. ZDPV 131:201–207.
Pixner B., Chen D. and Margalit S. 1989. Mount Zion: The “Gate of the Essenes” Re-excavated. ZDPV 105:85–95.
Wilson C.W. and Warren C. 1871. The recovery of Jerusalem: A narrative of Exploration and Discovery in the City and the Holy Land. London.