During 2004 and 2005 three seasons of archaeological excavations were conducted in the Ohel Yizhaq Synagogue on Ha-Gāy Street, in the Old City of Jerusalem (Permit Nos. A-4128, A-4436; map ref. NIG 2220/6316; OIG 1720/1316), as part of rebuilding the synagogue. The excavations, on behalf of the Antiquity Authority and financed by Everest foundation, were directed by H. Barbé and T. De‘adle, with the assistance of V. Essman and V. Pirsky (surveying), T. Sagiv (photography), D.T. Ariel (numismatics) and E. Boaretto (14Carbon dating).
The Early Roman Period (first century BCE–first century CE; Fig. 1)
Two portions of a masonry core (W125 and W409; width over 3 m), oriented north–south, were uncovered in the southeast corner of the excavated area, c. 7 m below the elevation of Ha-Gäy Street. A large staircase abutted the masonry on the east side (W416; Fig. 2). Three steps were exposed and a fourth was reached in a limited sounding. The second and third steps were covered with slabs, identical in the type of limestone, the dressing and the wear to those found on top of the masonry core. Three dry-stone walls, built of large masonry (W126, W127, W415), bordered the masonry core on the west side. Two walls of similar construction (W412, W413) were oriented east–west and linked up with W415. Several voussoirs, still in situ on W413, indicate that the inner face of these walls, built of limestone stretchers and dressed with a claw chisel, were the remains of a vaulted hall.
The excavation did not yet reach the floor of the vaulted hall. However, the lowest excavated layer coincided with a destruction layer (L4036) that comprised the collapsed stones of the vaulted hall (elevation of 723.7 m above sea level). Maps based on a topographical study of Jerusalem (L.H. Vincent and A.M. Steve. 1954. Jérusalem de l’Ancien Testament, Recherches d’archéologie et d’histoire, Pls. I, II) show contour lines at 720 m above sea level in the immediate surroundings of Ohel Yizhaq. Hence, the lowest point reached in the excavation is close to the floor/bedrock.
A section of a wall (W215) was uncovered in the northwest corner of the excavation, in a probe beneath the flagstones of the Byzantine paved street (the secondary cardo). Its slightly damaged eastern face was apparently coated with a lime layer. Its western face abutted a paved floor (L2054) whose exposure was very limited. However, earlier work in this area had revealed a Herodian street, c. 1.7–2.1 m beneath the Byzantine street. The 1.8 m difference observed at our excavation between the level of the Byzantine flagstone street and that of the paved floor on the west side of W215, may suggest that the latter was a Herodian street.
The pottery in the destruction layer of the vaulted hall dated to the end of the Second Temple period (first century BCE–first century CE). The associated coins included issues dating to the prefect Ambibulus (11/12 CE; IAA 101904), Agrippa I (41/42 CE; IAA 76762, 76776, 76777) and the procurator Festus (59 CE; IAA 101906). None was later than the second year of the Jewish War (67/68 CE; IAA 76774, 101903, 101905). The deposits sealing these remains and the thickness of the fills (up to 2 m) attest to the scale of the destruction, as well as to the leveling activities of subsequent rebuilding at the site.
The Roman Period (second–third centuries CE; Fig. 3)
The fills covering the Early Roman remains were cut by the foundation trench of a wall (W402; exposed length 4.9 m, thickness 1.2 m, average height 0.6 m) whose base rested on the masonry core of the staircase (W409). It indicates that W402, oriented north–south, was not erected before the end of the first century CE. Some of the stones in W402 were more than 1 m long, rendering it a monumental aspect. To the east of W402, three additional walls (W407, W408 and W414) were built at a later stage, forming a closed rectangular space. Wall 414 abutted the east face of W402. It was oriented east–west and its north face consisted of large, dry-laid blocks, dressed with flat bosses, made with a punch or point, which were surrounded by claw-chiseled margins. The technique is similar to Herodian dressing and we presumed that the stones were re-used, although the general unity of construction disagrees with this hypothesis. In any case, the north face of W414 turned out to be the outer face of a building that extended southward. The west face of the north–south oriented W408 abutted the north face of W414 and its foundations rested on the steps of the Early Roman staircase. A few slabs that adjoined W408 belonged to a pavement whose west part corresponded to the top of the masonry core of the Herodian staircase. Two slit windows (height 1 m, width 0.6–0.7 m, depth 1 m) were in the inner face of W408, which was entirely coated with a lime layer, indicating the thickness of the wall. Wall 408 seemed to continue northward but no additional segment was found. Only the southern face of W407 was exposed in the present excavation; it was oriented the east–west and built of small stones.
The debris sealing these walls contained Roman pottery, dating to the second and third centuries CE, together with numerous fragments of roof tiles (tegulae and imbrices). Some of the tiles bore the GXF stamp of the Tenth Roman Legion (Fretensis), which was based in Jerusalem at that time. Among the coins dating to the same period, some were minted in Aelia Capitolina, the latest of which by the emperor Elagabalus (218–220 CE; IAA 76767, 76778).
The Byzantine Period (fourth–seventh centuries CE; Fig. 4)
This period is mainly represented by a relatively large section of the secondary cardo (L1057, L2038) that was oriented north–south (exposed length c. 14 m, width 3.0–3.5 m; Fig. 5). The flagstones of this pavement, some of impressive size (length 1.6 m, width 1 m), were laid diagonally in relation to the axis of the street. Grooves on some of the flagstones were meant to hold the chariot wheels. Their various orientations were apparently the result of street repairs.
One flagstone of the cardo was temporarily lifted to probe the fills beneath, aiming to obtain material concerning the construction of the street. Most of the pottery dated to the time of the city’s destruction in 70 CE and a few potsherds, found immediately beneath the pavement, were clearly dated to the Byzantine period. The coins discovered in this probe should aid in refining the chronology.
The façade of a building was built at the east edge of the flagstone street (W211 and W212). The base of W212 rested on the cardo’s flagstones; one of the base stones was the shaft of a small column, which clearly indicates repair. The foundations of W211 penetrated into the underlying fills, attesting to the absence of flagstones at the time of its construction. The building’s doorway opened onto the street and was intentionally blocked during the Early Islamic period. It is assumed that the building was erected between the end of the Byzantine period and the beginning of the Early Islamic period.
The Byzantine period in the south part of the excavation is represented by a drain (L4008). Its west side was built of limestone ashlars and its east side had cut a large Early Roman wall (W402). A new wall (W404) that abutted the inner face of an earlier wall (W408) and whose foundation trench had cut the fills of the Roman period was built. It was constructed in two, south and north parts, leaving an empty space for the drain, which suggests that W404 and the drain, were contemporary. The drain flowed from east to west, toward the contemporaneous Ha-Gāy (El-Wad) Street, in the opposite direction of the site’s natural topography.
The Early Islamic Period (eighth–eleventh centuries CE; Fig. 6)
A plaster floor (L1070) is attributed to this period, as well as a small wall (W209) that was built directly on the cardo flagstones and abutted the blocked doorway of the building (W214). Pottery fragments found in the fills between W209 and W211 (L2010) show that the building erected on the street was no longer in use during this period. However, the Byzantine street continued, albeit damaged and reduced in size.
The Crusader Period (twelfth–thirteenth centuries CE; Fig. 7)
During this period, a new building was erected over the east edge of the cardo, its north wall abutting the south corner of the Byzantine-period building. The façade of this building consisted of two walls (W134, W135), flanking a doorway that opened onto the street. The foundation trench of W135 had cut the plaster floor of the Early Islamic period. An east–west partition wall (W131) formed the south limit of a room. Its north face was coated with a lime layer and abutted, on the west side, a short wall (W132), which enclosed a small plastered structure (L1065), most likely a pool, whose walls were coated with two layers of plaster. The bottom plaster layer contained potsherds and the upper layer consisted of lime mixed with crushed potsherds, which rendered it a pink-orange hue. Since this technique clearly indicates that the coating was meant to be water-proof, the structure, undoubtedly, contained liquids. This building served, most likely, as a shop or a workshop. Documents dating to the Crusader period show that this street was called Tanners’/Curriers’ Street and Furriers’ Street, attesting to the presence of craftsmen who produced or sold leather and fur goods.
The date of the building is provided by the pottery found below and above its floor (L1062), attributed to the Crusader period (twelfth–thirteenth centuries CE). This dating is corroborated by an Ayyubid coin (al-'Adil I, 1198–1218 CE; IAA 76714). At a later date, when the doorway was still in use, a stone wall (W130) was erected against the outer face of the building, on the cardo flagstones. This small wall looks like a podium and may be part of a staircase masonry core. Some time later, but not after the fourteenth century CE, the doorway of the building was intentionally blocked (W136).
Farther south, part of the north doorjamb of an entrance (W410) was uncovered on the same axis and at the same elevation as the store/workshop building. This doorway, which was found blocked, opened onto a room whose floor had been preserved. In the threshold area (L4025), the floor was made of small stones but in the rest of the room, it consisted of much worn flagstones (L4028). The debris found on this floor contained pottery dating to the Crusader period, indicating that, like the other building, the room ceased to be used at that time.
A new pavement (L4011), composed of large flagstones, was laid above Floor 4028, at a date ranging between the thirteenth century CE—when the other building was abandoned—and the beginning of the fourteenth century CE—the beginning of the public bath’s construction (see below). Slabs and shafts of columns in secondary use, which were found farther east, seem to belong to Pavement 4011 that must have covered, originally, a large area. As this pavement was not connected with any architectural features, it is difficult to interpret, yet it may have been part of a Crusader-period street.
The Mamluk Period (fourteenth–fifteenth centuries CE; Fig. 8)
This period is represented by a public bath (hammam) that extended across the entire site. Despite some destruction in the central area, due to the construction of a water reservoir in the Ottoman period, and some limitations because of existing buildings, the excavation revealed the nearly complete plan of this hammam, including its west façade (W100, W200). The changing room (Fig. 9) was preserved in its entirety. A cross-ribbed vault covered a monumental door, which was flanked by benches on either side and was blocked at the end of the Mamluk period (end of fifteenth century CE). A drainage system (L1028) for emptying the latrines was preserved in the warm- dry rooms that led to the furnaces. The octagonal room was flanked by small rooms, one on the east and the other on the west. A similar plan existed for the warm-dry room located in the central part of the bath. During the 2005 season, part of an adjacent eastern room in a good state of preservation was uncovered. It had the base of a dome, with a few brick courses above W139, which was its easternmost border, whose interior décor was of stucco.
The pools (Loci 2055, 3025, 3026) that had plastered floors were built above the furnaces and produced steam while a double system of channels carried hot air. The first, leading from the furnaces, consisted of a network of closed channels built under the floors of the octagonal room and the two small rooms (L2040, L3027). The second, leading from the floors of the pools, ensured the circulation of hot air through pottery pipes to the warm-dry rooms. All the channels converged at the south end of the façade wall (W100), where the air was drawn up into a chimney (L1079). The material associated with the floors of these baths and their levels of abandonment can be attributed to the early fourteenth–late fifteenth centuries CE. A Carbon 14 analysis of mortar samples from the construction phase of the changing room and of samples of plaster that coated its walls after its conversion into a water reservoir, confirmed this dating. Although slightly smaller, the plan of this bath is identical with that of Hammam al-Ayn, located immediately to the north of the excavated site. An archival text from the year 1531 allows us to identify the hammam on the site of Ohel Yizhaq with the Mamluk Hammam, known as Mustahamm Daraj al-Ayn.
The excavations at Ohel Yizhaq provide important results and most of all, new information regarding the different periods at the site. The discovery of a well-preserved public bath, dating to the fourteenth century CE, add to the knowledge of the town planning in an area of the city, which was drastically remodeled in the Mamluk period. Excavating a portion of the ‘Secondary Cardo’ contributes to our knowledge of this main street and allows a more precise dating of its construction and the duration of its use. The exposure of the Roman-period monumental remains, north of the Decumanus, including a wall with loopholes, evidences a fortification system and hints at the nature of military settlement at the site. This new data is relevant to the location of the Tenth Legion encampment in the city. The excavations also uncovered significant remains from the Second Temple period, which substantially contribute to the proposed reconstruction of the topography of this central part of Jerusalem in the last century of Jewish autonomy before the destruction of the temple.