The construction date of the present-day Assyrian Monastery is unknown (Ben-Arieh 1977:277). Below the Church of St. Mark in the monastery compound are several underground halls and rooms; one of these rooms is identified by Assyrian Christians as the place of the Last Supper of Jesus (St. Mary’s Old Church). The unauthorized excavation work that damaged the remains sought to open a passage between two underground halls and to remove earth from one of these halls prior to turning it into a chapel. In the current excavation, two squares were opened: in a roofed courtyard of the monastery compound (Sq A; 4 × 5 m); and in the underground hall intended as a chapel (B; 3 × 7 m), c. 9 m southeast of Sq A. The two squares were excavated to a depth of c. 4.5 m below the level of the courtyard without reaching the bottom of the archaeological sequence at the site. The excavation revealed remains of structures, floors and installations dating from the first century BCE to modern times. The archaeological sequence that was documented in each of the squares was not identical. Thus, the remains from the two squares will be described separately, and no attempt will be made to associate them.
Square A

This square was opened in the roofed courtyard north of the church, alongside the northern wall of the church (W01, below), and the eastern wall of the roofed courtyard. The lower part of the northern wall was built on an earlier wall (W16, below; Fig. 2), which is the northern wall of the underground hall identified as the Room of the Last Supper. The manual excavation in this square began at a depth of 0.6 m below the modern-day paving, and it uncovered a sequence of eight strata (VIII–I), dating from the first century BCE–first century CE through modern times. Wall 16, alongside which the excavation square was opened, extended down to below the base of the excavation, and its foundations were not discovered. In the 1970s, liquid concrete was injected into the foundations of the church, including into W16. The concrete congealed not only on the surface of the wall but also spread into the accumulations and the fill alongside it, trapping sherds, bones and stones. The concrete, which was found down to the bottom of the excavation, covered about half the area of the square. This disturbance severely impeded our ability to distinguish among the various accumulations and fills and to fully understand their stratigraphic relationship to W16.


Stratum VIII (first century BCE–first century CE; Fig. 3). Remains of this stratum were discovered in a probe (L167; Fig. 4) excavated under the floor of Stratum VII (L161, below); the excavation ended with the uncovering of a layer of soil (L172), apparently a floor, which abutted the southern face of a fieldstone wall (W15). The rather compact soil fill (L167; thickness 0.8 m) that covered the floor contained a pottery assemblage which included cooking pots (Fig. 5:1, 2), a jar (Fig. 5:3), a juglet (Fig. 5:4), a flask (Fig. 5:5) and a pared lamp (Fig. 5:6) dating from the first century BCE to the first century CE. A coin from the reign of Agrippa I (41/2 CE; IAA 138664) was also found on Floor 172.


Stratum VII (late first century BCE–70 CE; Figs. 3, 4). A compact earthen floor (L161), uncovered c. 0.7 m above Floor 172, was attributed to this stratum. Floor 161 abutted walls on the west (W06) and north (W11), which created a clear corner; its stratigraphic relation to W16 could not be determined because of the liquid concrete. Wall 11 was built directly on top of W15 from the previous stratum, and like W06 it was preserved to the height of one course of medium-sized limestone fieldstones. A coin from the time of Herod’s reign (37–4 BCE; IAA 138661) was found in the makeup of the floor. It therefore seems that the floor and the two walls abutted by the floor belong to a building from the late first century BCE or the beginning of the first century CE.

Although no clear signs of fire were found on the floor, it was covered by a friable fill (L160; thickness 0.55 m) that included a very large quantity of ash and a great many sherds, all smashed and burned. Many fragments of stone measuring vessels (not drawn) were also found in this fill, as well as a coin of a procurator under Nero (58/9 CE; IAA 138660) and burnt bones. The entire pottery assemblage is dated to the first century BCE–first century CE. Among these vessels are a krater (Fig. 6:1), a bowl (Fig. 6:2), a basin (Fig. 6:3), a cooking pot (Fig. 6:4), jars (Figs. 6:5, 6), a juglet (Fig. 6:7), flasks (Fig. 6:8–11) and oil lamps (Fig. 6:12, 13). The burned remains seem to have originated in a major conflagration, apparently the destruction of Jerusalem in 70 CE.


Stratum VI (the Late Roman period, second–fourth centuries CE; Fig. 3). No architectural remains were found in this stratum. It consisted of a friable fill (L136/L139; thickness 0.43 m) with a few building stones and numerous pottery sherds, and above it—two superimposed accumulations of brown soil (L129, L135) containing fallen building stones and numerous roof tiles.

Most of the pottery in Fill 136/139 is dated from the first century BCE–first century CE to the second century CE (Fig. 7). These included basins (Fig. 7:1, 2), cooking pots (Fig. 7:3, 5), an amphora (Fig. 7:5), jars (Fig. 7:6, 7), a jug (Fig. 7:8), a juglet (Fig. 7:9) and flasks (Fig. 7:10, 11). This fill apparently consisted of remain of the destruction of 70 CE, along with vessels from the second century CE, which indicate when the fill was laid, apparently in preparation for renewed activity. 

Accumulations 129 and 135 were severely damaged by the liquid concrete, and pits that penetrated from Stratum II caused the mixture of the finds from the two accumulations. The lower accumulation (L135; thickness c. 0.9 m) was very friable and contained numerous sherds dating from the second–fourth centuries CE, including bowls (Fig. 8:1–3), a krater (Fig. 8:4), a lid (Fig. 8:5), a cooking pot (Fig. 8:6), a jar (Fig. 8:7), a cooking jug (Fig. 8:8) and a discus lamp (Fig. 8:9). The accumulation also contained numerous roof tiles, one of which bears the stamp of the Tenth Roman Legion (Fig. 8:10). A mixed assemblage of glass vessels was also found in the accumulation, dating from the Early Roman to the Mamluk periods (Gorin-Rosen, below), apparently due the later pits that penetrated the accumulations. A disjointed, hard-to-identify layer of compact soil (L130; 761.71 m asl) separated this Accumulation 135 from Accumulation 129 (thickness c. 0.2 m) above it. These accumulations may be associated with the Late Roman period (second–fourth centuries CE).


Stratum V (the Byzantine period, fifth–sixth centuries CE; Figs. 9, 10). A thick, hard plaster floor (L119; 762.71–763.06 m asl), which slopes northeast, covered the entire area of the excavation except where the pits from Stratum II (L108) damaged it. The floor was laid over a fill (L126; thickness 0.45–0.60 m) of crushed chalk, which also sloped northeast, and contained pockets of ash with sherds from the Byzantine period (fifth–sixth centuries CE). The sherds included bowls (Fig. 11:1, 2), a jar (Fig. 11:3) and a jug (Fig. 11:4). These finds dated the fill and the floor to the fifth–sixth centuries CE.


Stratum IV (eighth–ninth centuries CE; Figs. 12, 13). In the western half of the square, above a thin fill that covered Floor 119, was the eastern part of a space (L117), which extended westward beyond the excavation area. Space 117 was delineated by three walls (W03–W05), of which only one course, possibly a foundation course, was preserved. No floor was found in the space; it may have been higher than the preserved foundation course. A rectangular cesspit (L121) installed in this area was built of small stones; it too was only partially uncovered, as its western end was beyond the excavation area. It is difficult to determine the use of this space, but it may have been an installation or part of a courtyard alongside a dwelling. The cesspit and the pottery (not drawn) retrieved from the fill in Space 117 point to a date in the Abbasid period (eighth–ninth centuries CE).


Stratum III (tenth century CE; Figs. 14, 15). This stratum comprised disjointed remains of an open courtyard, which was damaged by the pits and foundation trenches of Stratum II. The remains included patches of a plaster floor (L109); a plastered water channel (L110), in which two construction phases were identified, that runs across the courtyard in a general north–south direction; and a rectangular installation (L107) with stone paving (L112), of unclear use. The plaster floor abutted the installation, and apparently abutted the channel as well. In its first phase, the channel ran due south, as evident in the southern part of the excavation square; in its second phase it was diverted eastward. The sherds found under the plaster floor (L113, L114, L116) included bowls (Fig. 11:5, 6) and a jug with a pinched rim (Fig.11:7) from the tenth century CE, dating the remains of the courtyard to that century.


Stratum II (fifteenth century CE; Fig. 16). The foundation trench of the northern wall of the church (W01) could be discerned beside the wall, in the southern part of the square. The trench (L105) was narrow and cut through the remains of Strata III–V. The western face of a stone structure, built of stepped courses (W02), was uncovered in the northeastern part of the square, along the eastern balk. It is probably a pier that supports a column standing in the courtyard to the north of the church and supporting the second story. Prior to construction of the pier, a large pit was dug. It covered about one quarter of the square (L104, L108; Figs. 10, 13) and damaged the earlier strata. The foundation trench of W01 and the foundation pit of the pier were filled with a very friable fill, rich in sherds, bones and stones. The latest pottery in these fills included bowls (Fig. 11:8, 9), one of which (No. 8) is glazed, an ‘elephant ear’ handle of a bowl (Fig. 11:10) and a jar (Fig. 11:11), all of which date from the late Mamluk and early Ottoman periods.

A very deep round pit (L122; diam. 1.2 m, depth c. 2 m; Figs. 10, 13, 16) discovered in the center of the excavation area cut through the remains of Strata V–III. The pit was packed with large building stones interspersed with small stones. In the southeastern corner of the square, adjacent to W01, was a pit grave (L120), part of which also extended into Stratum V. The grave contained the skeletal remains of a woman older than 50. The sherds found in Pit 122 and in Grave 120 (not drawn) were also dated to the Mamluk and early Ottoman periods. It can thus be concluded that the construction of the current church along with the pier in the courtyard can be dated to the fifteenth–sixteenth centuries.


Stratum I (twentieth century CE). A stone-built and plastered water channel (L102) was found under the paving of the current courtyard. It ran along an east–west axis under the piers of the columns that support the open courtyard north of the church. This is apparently modern construction, from the twentieth century, which also included the piers that currently support the roof of the western part of the monastery.

Square B

The excavation of the underground hall was undertaken after part of the earth that filled it, which included ancient archaeological remains, accumulations and fills, was removed without archaeological supervision; the northern wall of the hall was broken through during this illegal work. The center of the room was damaged during removal of the soil, and therefore mainly its western and eastern margins were excavated. The excavation uncovered a sequence of five strata (V–I) dating from the second–third centuries CE to modern times.


Stratum V (early Byzantine period[?]; Fig. 17) comprised a heap of fallen stones (L164, L168) containing mixed finds (not drawn): two coins from the first century CE—one possibly of Agrippa I (41/2 CE; IAA 138662) and one of a procurator under Claudius (54 CE; IAA 138663)—as well as pottery sherds from the Late Roman period (second–third centuries CE), including roof tiles bearing the stamp of the Roman Legion, and from the Byzantine period (not drawn); also found in this heap was a coin from the early seventh century CE—a follis of Maurice (601/2 CE, Antioch mint; IAA 138665). The tops of three walls (W17–W19) could be seen protruding from the heap of stones, but these were not excavated.


Stratum IV (Figs. 18–20; the Byzantine period). Above the Stratum V heap of stones were the foundations of two walls (W09, W10), which formed a corner of a structure; the walls were carelessly built of two rows of fieldstones. Sherds from the Byzantine period were found among the stones of the wall, including a bowl and a krater (Fig. 21:3, 4). A floor of compacted earth mixed with plaster abutted W09 from the west (L156; 760.95 m asl). One piece of plaster contained a nummus from reign of Justinian I (552–565 CE; IAA 138657), indicating that the building was constructed in the sixth century CE.


Stratum III (late Byzantine–early Umayyad periods; Fig. 20). Remains of two superimposed plaster floors (L151—761.81 m asl; L153—761.26 m asl) were uncovered in the eastern third of the hall. A coin from the First Jewish Revolt against the Romans (67/8 CE; IAA 138655) was found in Floor 153, the lower of the two. A probe below this floor revealed a fill of earth and stones (L163) containing a mix of ceramic finds, including a roof tile bearing the stamp of the Roman legion (Fig. 21:1) and fragments of pottery vessels from the Byzantine period (fifth–seventh centuries CE; not drawn). Thus, is seems that the floor dates from the late Byzantine or the beginning of the Umayyad period; the upper floor resembled the lower one and should probably be dated similarly.

A few walls whose tops were seen on the western edges of the hall may possibly belong to this period as well, although they may be earlier in date. They are situated at the foot of the walls of the hall, which are very high (Fig. 22). The stratigraphic attribution of these walls is not clear, as it is impossible to dismantle them, and next to them are only small segments of the plaster floors (L146, L148). Nevertheless, it is clear that the walls were covered with debris placed there in the Abbasid period (Stratum II, see below), and are therefore earlier.


Stratum II (Abbasid period; Fig. 23) comprises a series of earthen layers that were spilled from the southeast and west into the depression in the center of the hall (L143, L149, L152). They covered Floor 156 from Stratum IV and the Stratum III remains that were preserved on the eastern and western ends of the square. The pottery in these earthen layers was mixed; it included an imbrice bearing the stamp of the Roman Legion (Fig. 21:2), but the latest fragments in the layers are from the Abbasid period and include lamps (Fig. 21:5, 6; tenth–eleventh centuries CE).


Stratum I.In the southern section of the square (Fig. 21; Section 1–1) stood a pier, which supports the modern southern wall of the monastery. The pier was built on a sloping plaster surface (L132; Fig. 23), apparently intended to seal the layers of earth laid in Stratum II and stabilize them. Only a small part of Surface 132 survived near the pier, to its north, but its continuation was clearly identified in the balk. The pier’s construction cannot be dated with certainty, but it post-dated the tenth century CE, the date of the latest fragments found in the earthen layers.

Glass Finds
Yael Gorin-Rosen

The excavation yielded twenty-seven glass fragments that could be identified and dated. They represent several of the important periods in Jerusalem’s history: the Early and Late Roman periods, the Byzantine and the beginning of the Umayyad periods and the Mamluk period, to which only one vessel could be ascribed. In every chronological group are vessels worth publishing, although at least some of them were found in fills. For example, in a mixed basket from Accumulation 135 (Area A, Stratum VI), were bowls from the Early Roman period, a Byzantine-period wineglass and the neck of a jar from the Mamluk period. Sixteen vessels will be described below, 12 of which are presented in Fig. 24. The two first bowls described here were cast, whereas all the rest were free-blown.


Early Roman period. The earliest glass vessel found in the excavation is represented by a very small fragment. It belongs to a cast bowl, decorated on the interior under the rim with a horizontal grooved line (not drawn; L135, B13502). The glass, which is yellowish to olive-green, is covered with thick silvery and golden weathering. Many bowls of this type were found in excavations in the Old City of Jerusalem, and they are dated to the second half of the first century BCE and the beginning of the first century CE.

The most important fragment found in the excavation is the rim of a cast and polished bowl made of colorless glass (Fig. 24:1). The bowl’s rim is everted to a broad ledge, thickened and pulled downward at the end; it is surrounded by an additional thickening where the rim is connected to the wall. This bowl belongs to a group of very high-quality bowls made of colorless and well-polished glass. Bowls of this type were found in caves in assemblages from the Judean Desert dated to the Bar Kokhba Revolt. A very similar bowl was found in the Cave of Horror (Barag 1962: Fig. 6), while in the Cave of Letters were two small plates with a very similar rim, although their body was different (Barag 1963: Fig. 39).

Another bowl (Fig. 24:2) features a double hollow tube below the everted and rounded rim. The bowl is made of colorless glass with a light greenish yellow tinge, and it is covered with silvery-black weathering. The date of this bowl is similar to that of Bowl No. 1; similar bowls were found in caves and sites dating to the Bar Kokhba Revolt, such as the Cave of Letters (Barag 1963:108, Fig. 38:7-52.כ).

Another fragment that can be attributed to an Early Roman period bowl is part of a base with a double hollow fold in the connection between the wall and the base (Fig. 24:3). The object is made of colorless glass with a light bluish tinge, covered with silvery iridescent weathering that penetrated the folds. Bases of this type are known from assemblages dated to the end of the late first century and the beginning of the second century CE.

The lower part of a beaker with a low, concave base and a cylindrical body (Fig. 24:4) was found. The beaker is made of greenish bluish glass covered with brown-black weathering that corroded the wall, and it does not bear a pontil scar. It resembles beakers with cut decorations, similar to those found in previous excavations in Jerusalem, in assemblages dated to the Early Roman period, e.g. at Akeldama Tombs (Winter 1996: Fig. 5.1:1, 2).

The identification of another fragment (Fig. 24:5) is uncertain. The base is thickened compared to the base of Beaker 4, but it too, has no pontil scar. The base is made of bluish-green glass and is covered with an enamel-like silverfish black weathering that corroded the wall. The base may have belonged to a vessel similar to Beaker 4 or to a bottle, either piriform or candlestick-type, both typical of the Early Roman period (for an example from Akeldama, see Winter 1996: Fig. 5.2:2).


Late Roman period. The most typical vessel in assemblages from the end of the Roman and the beginning of the Byzantine periods in Israel is a beaker with a solid base (Fig. 24:6). An almost complete base made of light bluish glass covered with silvery weathering was found in the excavation. A pontil scar with a remnant of glass from the pontil could be discerned on the base. This was the standard drinking vessel during the fourth and early fifth centuries CE. This date fits well the date of the locus in which the vessel was found: Fill 126 under Floor 119, which was dated to the fourth–fifth centuries CE.


Byzantine and Early Islamic periods. One of the several wineglasses found in the excavation is made of light bluish green glass and has a hollow ring base (Fig. 24:7). It was found in a conglomerate chunk of concrete and debris, and like other finds in the excavation, it was probably trapped in the liquid concrete that was poured into the foundations of the church in the 1970s.

Another wineglass (Fig. 24:8) found in the excavation is made of bluish green glass covered with silvery weathering. An unusual foot rises from its hollow ring base: it is hollow and stepped, like a column drum which protrudes slightly in the center. The wineglass belongs to a local, Byzantine-era tradition, but presents a variation that is not known from previous findings in Jerusalem or elsewhere in the country. It may be the result of an error in production and not of intentional design.

A complete handle preserved with a small part of the vessel’s wall belonged to a three-handled bowl lamp (not drawn; L143, B14309) made of bluish green glass; this object points to a Byzantine-period presence at the site.

Another complete handle with a small part of the vessel wall (Fig. 24:9) is made of bluish glass with greenish veins. The quality and translucence of the fabric seem typical of Umayyad-period glass. The straight wall and the manner by which the handle was fashioned are typical of bowl-lamps that first appeared in the Umayyad period.

The vessel in Fig. 24:10 is somewhat unusual; it features a wide mouth and a thickened rim, partially rolled inward, and is made of bluish green glass with a yellow-greenish vein that bears slight silvery weathering. The quality of the fabric, its color and the manner by which the rim was fashioned attest to its Umayyad-period date. This fragment was found in Pit 122, which was dug in the fifteenth century CE but cut into earlier strata from the eighth–tenth centuries CE and from the Byzantine period (fourth–fifth centuries CE). In the same basket as this fragment was a small fragment of an upright rim made of light bluish-greenish glass, decorated with thin trails of turquoise glass mixed with opaque red glass (not drawn); it may have belonged to a wineglass or a beaker bowl from the late Byzantine and the early Umayyad periods.

A base of a small bottle, square in section also dates from the Early Islamic period. It is made of greenish blue glass and is covered with golden weathering (Fig. 24:11). The fabric and the weathering are typical of the Umayyad and the early Abbasid periods. The bottle belongs to a group of medium-sized bottles, square in section, that have a slightly thickened base. This group should not be confused with the narrow bottles with a thick base, which are also square in section and were more common in the Abbasid period. The bottle was found in Channel 110 in Area A, which according to the pottery, dated to the tenth century CE, was ascribed to Stratum III.

A large fragment of a rectangular windowpane was also unearthed (Fig. 24:12). It is made of bluish green glass and is covered with enamel-like brown and silvery weathering. The thickness of pane is not even, and it thins toward a rounded rim at the edge. The windowpane is of a type which was very common in the Byzantine period and is usually found in association with public buildings. This glass fragment was found in Cesspit 121, which was dated to the Early Islamic period.


Mamluk period. The latest vessel found in the assemblage is a broad neck made of colorless glass with a light brown tinge (not drawn; L135, B13502). It belonged to a large, thick-walled bottle—a very large vessel that was extensively uses during the Mamluk period throughout Israel.


The importance of this glass assemblage lies in the two main groups of vessels it contains. The first group, from the Roman period, dates from the first century and first third of the second century CE and resembles in its composition assemblages from the time of the Bar Kokhba Revolt. The second group dates from the Byzantine and Early Islamic periods.


In addition to the abovementioned coins, the excavation yielded two coins from the First Jewish Revolt (67/8 CE, IAA 138658; 68/9 CE, IAA 138656) and two half follis, one of Justin II (565–578 CE; IAA 138653), and the other of Heraclius I (629–631 CE; IAA 138659).


Although the remains uncovered in the excavation represent most of the periods from the Second Temple period onward, the small excavation area and late disturbances allow only a limited glimpse into the varied history of the area in antiquity. It seems that the modern structure of the church post-dates the fifteenth century CE, but W16, on which W01 was built, is much earlier in date. Because the excavation did not extend down to the foundations of W16 and due to the later damage sustained by the face of the wall and the remains beside it, that wall cannot be dated with any certainty. Nevertheless, it may be adduced that W16 dates from the Byzantine period at the latest. This wall encloses the ‘Room of the Last Supper’ on the north, indicating that the room may be part of an ancient structure. Wall 16, the walls from the Early Roman period discovered at the bottom of the excavation in Sq A and the Byzantine-period walls in this square were all built in a similar orientation: east–west and north–south. It seems that this arrangement of walls was preserved over the generations, and eventually dictated the plan of the Assyrian Monastery, including that of the Church of St. Mark, whose walls were also built with the same orientation. No built remains were located from the period between the destruction of 70 CE and the Byzantine period; however, it is possible that the fill within the remains of the building from the first century BCE–first century CE, which contains material originating in the destruction of 70 CE, was laid in preparation for the construction of the Roman Legion camp, if it indeed was built on the western hill (Weksler-Bdolah 2015).