The excavation took place inside a structure that belongs to the Franciscan Custodia de Terra Sancta. It is situated on the eastern side of Ha-Degel Street (‘Aqabat el-Beiriq), a narrow winding street that connects the main streets of Ma’alot Ha-Midrasha (‘Aqabat et-Takiyya) and the Via Dolorosa (Ṭariq el-‘Alam). These two main streets descend the western hill of the Old City from the direction of Bet Ha-Bad Street (Khan el-Zeit) toward the Tyropoeon Valley (Ha-Gai Street, Ṭariq el-Wad). Ma’alot Ha-Midrasha Street is known mainly for the beautiful buildings from the Mamluk and the early Ottoman periods along its eastern side. A few years ago, remains of the Mamluk-period Suq Khan el-Zeit were excavated at the western end of the street (Permit No. A-6532). Recently, an excavation was conducted in a structure called Dar el-Consul, which uncovered, among other things, remains of the Roman Cardo (Permit No. A-7976). The Via Dolorosa is known mainly for the sacred Christian sites along it. Earlier remains are known only at the Fourth Station (Veronica’s House), where rock-cut remains of an ancient structure and a mosaic floor were discovered in the crypt of a church. These, however, were not methodically excavated, and underwent only rudimentary documentation. Based on a fragment of an inscription incised in stone, the site is identified with a church dedicated to Saints Cosmas and Damien from the Early Islamic period (Wilkinson 1977:162).

The building in which the excavation took place is typical of the Ottoman period and consists of three vaulted spaces. The entrance to the building is through a small courtyard, which lies c. 1 m lower than present-day street level. The excavation took place in a rectangular space (c. 3 × 4 m; Figs. 2–4) roofed with a pointed vault—the westernmost of the three spaces. The architectural remains comprised three main phases, which apparently date from the Early Islamic period (Phase C), the Mamluk period (Phase B) and the early Ottoman period (Phase A).

Phase C (Fig. 2). Two rectangular paving slabs (L119; width 0.8–1.0 m, thickness c. 0.25 m; Figs. 5, 6) made of hard mizi ahmar limestone were partially uncovered at a depth of c. 2.5 m below the floor of the building, in the northwestern corner of the space. The southern of the two slabs was c. 5 cm higher than the northern slab, suggesting that it was part of a sidewalk. A round drain was cut into the northern slab (L127; Fig. 7), framed by a rectangular groove suitable for a cover slab or grate. The drain led to a plastered channel (L128) that continued eastward (see Fig. 6). Both the slabs are completely smoothed, apparently due to prolonged use. A trench excavated east of the slabs revealed that they were set on a bedding of soil and small stones (L126), which did not contain any finds that could date Paving 119.
Phase B. In this phase, a vault was built under and beside Paving 119. Only the top of the vault was uncovered; a surface (L120, L125), which was built of stones of various sizes in secondary use interspersed with fills of plaster and lime, cut the continuation of Paving 119 on the east and south. The top of the surface descends from east to west, where it cuts the bedding of Paving 119 (L126; Fig. 8). At the top of the vault were three broken, dressed stones bearing plaster remains, which apparently served as the vault’s keystones (Fig. 5). A paving slab was inserted between the stones of the arch (Fig. 9), as well as a fragment of a small column from the same hard mizi ahmar limestone (Fig. 10), apparently remains from the destroyed building of Phase C. The top of the vault was missing at the eastern end of the excavation; this may have been an opening. Above the vault and Paving 119 was a thick layer of fill (L112, L121, L123, L124), which included layers of brown-black soil containing charcoal and a large quantity of small stones. This layer apparently served as a bedding for floors that did not survive; it was sealed by the construction of Phase A.
Phase A (Figs. 3, 4). This is the phase in which the present-day building was constructed. Above the fills from Phase B, a thick layer of brown soil was set with a few small stones (L111). Above that layer, the floor of the building was set; this floor had been removed before the excavation. Four walls from this phase were revealed (W101–W104). Over time, several changes were made in the building. Walls 101 and 103 belong to an early sub-phase, which apparently included an open courtyard at the entrance to the building. It is possible that in this phase, the space of Vault 120/125 was used as a cesspit. It seems that during this sub-phase, a small, round installation (L107; Fig. 11) was dug into the fills and lined with small stones. In a later sub-phase, Walls 102 and 104 were added, which enclosed the courtyard. At that time, a new entrance to the building was made through W102. Wall 102 was built over the southeastern part of Installation 107, indicating that the installation predates the wall. A poorly constructed wall (W105) was built perpendicular to W102.
In this phase, a pit was dug down at the eastern end of the excavation square; it cut through the layers of fill from Phases B and A, down to the breach at the top of the vault and blocked it with a fill of stones and soil (L114). A sewage channel (L108) built above this blockage ran from the east and turned south, entering an opening in the lower part of W101. The channel was destroyed by later construction activity in the building, but stones from its walls and cover slabs were found lying along its course (Fig. 12). A segment of the channel survived in the foundations of W101, where its rectangular profile can be seen, without plastering. One asbestos cover slab could be seen (Fig. 13), indicating that the channel was used in modern times.
In each phase of construction, the excavation revealed a great many potsherds (see Appendix). Two fragments of glass were also found: a vessel dating from the late Byzantine or Early Islamic period, found in the drainage opening in Flooring 127, and a worked chunk of raw glass, which was discovered in Fill 118; this worked chunk is especially rare (Katsnelson, below).
Two coins were found in the excavation, both in an accumulation of soil in the part of Sewage Channel 108, where it is built into the foundations of W101. One is a gold Tunisian coin minted under the French colonial government in 1898, and the other is an Ottoman coin of Mehmed V (1909–1918).
The Glass Finds
Natalya Katsnelson

The scant glass finds consist of a blown vessel and an elongated glass chunk (Fig. 14).

The vessel (Fig. 14:1; L127, B1024) displays a fire-rounded funnel-shaped rim. It is made of greenish-blue glass that is covered with silver patches of weathering. A precise reconstruction of the vessel’s complete form is difficult, but it could have been an upper part of a wineglass or a bottle. Such a vessel types that share similar simple shapes is particularly characteristic of local glass wares at the end of the Byzantine period and the beginning of the Early Islamic period (Gorin-Rosen and Winter 2010). The quality of the fragment’s fabric supports this date.

The elongated glass chunk (Fig. 14:2; L118, B1012; length 6.5, width 2.6 cm, thickness 0.5 cm), which is mended from three fragments, was purposely shaped into a blade-like tool, making it an unusual find. It is made of greenish glass, the surface of which is covered with silver iridescence; small, rounded bubbles and black impurities are visible. The piece has the shape of a blade with a triangular cross-section. It may have flaked from a larger chunk of raw glass intended for re-melting prior to the production of glass vessels or objects. However, it’s retouched diagonal truncation on the upper surface and reworked breaks at the bottom edges resemble a traditional cutting tool made of obsidian or flint. The exact function of this specimen is unclear; it could have been a knife or a scraper. It may have served as tool for some craft, such as decorating pottery (Fünfschilling 2015: Figs.16.7, 8). Another possible use for such a tool made of glass is mentioned in the Babylonian Talmud: cutting and scraping the top of a weaver’s shuttle and rod to beat and straighten the warp threads of a loom (רַבִּי יְהוּדָה אוֹמֵר ... זְכוּכִית, כְּדֵי לִגְרֹר בּוֹ רֹאשׁ הַכַּרְכָּר.; BT, Mishna Shabbat 8:6; Grossmark 1989:155, and see references therein). To date, only a few examples of such reworked glass fragments have been published; this may not necessarily reflect their scarcity at sites, but rather the limited attention given to such finds. A similar glass blade was found at Jerash in a context dated to the sixth century CE, alongside similar tools made of pottery and flint (Kehrberg 1992:460, Fig. 9).

To conclude, although the excavation yielded only two glass finds, one of them—the tooled, raw glass chunk—provides new and interesting data that furthers our understanding of glass usage in Byzantine Jerusalem.


The remains of Paving 119 from Phase C cannot be securely dated because no finds were discovered in its bedding. The wear and tear on the flooring shows that it was part of some public space, perhaps a street or an alleyway. The southern stone slab, which is higher than the northern one, may have been part of a sidewalk or the curb of a street which lay along a northwest–southeast axis, diagonal to the layout of the present-day streets. The glass fragment discovered in Drain 127 is dated to the end of the Byzantine and the Early Islamic periods. All the pottery sherds from between the stones of the Phase B vault date from the Early Islamic period, while in the fill above the vault were potsherds from the Early Islamic period and the Middle Ages, up to the Mamluk period. This fill certainly belonged to a floor that existed along with the vault but did not survive. Therefore, the Phase B vault should be dated to the Mamluk period. The pottery from the Early Islamic period found among the stones of the vault, like the architectural elements found in secondary use at the top of the vault, seem to have originated in the Phase C flooring, which was dismantled to build the vault. If so, it may be assumed that the flooring of Phase C dates from the Early Islamic period.
Phase A had several sub-phases of construction, the main one being the building that still stands on the site. The pottery finds from Phase A (L111) date the building to the early Ottoman period (sixteenth–seventeenth centuries CE). All of the floors from this phase were removed prior to the excavation, making it impossible to date the construction phases of the building. The two coins from Sewage Channel 108 apparently belong to the end of its use, at the beginning of the twentieth century.