Two wells are located in the vicinity of the site: Bir Na‘ar (Ne‘ora) and Bir Rafat (Bir Qoubar), in addition to numerous cisterns. In 1838, E. Robinson reported seeing the site—“the wide-spread ruins of Râfât”—from afar (Robinson 1841:20). In 1868, V. Guérin described the ruins at Rafat as comprising a temporary encampment used by local farmers during the harvest season, the two wells and the Muslim sanctuary (maqam) of Sheikh Hassan (Guérin 1982:20). In 1883, building remains, cisterns, wine presses, fourteen tombs and columns protruding from the ground were documented at the site that was surveyed by the PEF (SWP; Conder and Kitchener 1883:154–155). In 1866, the land was acquired by the Latin patriarchate and in 1925 the Monastery of Deir Rafat was opened on the hill west of the excavation. In 1929 two Dominican priests, F. M. Abel and A. Barrois, published a burial cave that was discovered not far from the monastery (Abel and Barrois 1929). Two columbarium caves were exposed in 1934 and 1937 while building the wall around the monastery and pottery sherds dating to the Roman period were discovered in one of them (Bagatti 1990). Rock-hewn installations, cave dwellings, a well, numerous building remains, many burial caves and cisterns were documented in a survey of the monastery’s grounds performed in 1977 (Bagatti 2002). In a survey carried out east and west of the monastery in 2007 architectural remains, burial caves, rock-cut installations, quarries, columbarium remains, a well, cisterns and agricultural terraces were documented (Nagorsky 2010). In 2009 and 2011 two seasons of excavations were conducted slightly east of the current excavation area and building remains were exposed dating to the Roman and Early Islamic periods, as well as ceramic finds from the Byzantine period (Permit No. A-5598, A-6148). The architectural remains that were exposed in the current excavation are the western continuation of those remains.
Ritual Bath (miqveh; Fig. 3). The upper part of the miqveh was built of stones whereas its bottom was hewn in bedrock. Three built walls of a square structure (W31, W32, W36) were exposed of the upper part of the bath. Two parts of a threshold stone were installed in the northern wall (W31). These were probably in secondary use and served as an entrance for the opening to the building (Fig. 4). A plastered channel (length 0.6 m, width 0.1–0.2 m) between the two stone parts conveyed water into the building. The eastern wall (W32) was built of a single row of ashlar stones and coarsely hewn stones above it. The western wall (W36) was constructed of two to four rows of small and medium, roughly hewn stones; it was preserved to a height of two to three courses. The walls in the upper part of the miqwe were treated with a thin layer of light gray plaster (thickness 1.5 cm). The three stone walls were founded on rock-cut walls that constituted the bottom part of the miqveh. A plaster floor (L110, L116) from which three rock-cut steps descended toward the west was exposed in the eastern part of the miqveh, near W32. The steps led to two openings hewn in the western bedrock wall of the lower portion of the miqveh; one opening was rectangular (width 0.35 m) while the other was arched (width 1.3 m). The part of the bath beyond the rock-hewn openings was not excavated. A plastered elliptical basin hewn at a lower level than that of the plaster floor was exposed south of the hewn steps. A level of tamped soil and lime (L108, L111), possibly a floor, abutted the outer face of the northern and eastern walls. The excavation of the building was not completed, but in view of the components that were exposed it was determined the structure was used as a ritual bath. The ceramic finds discovered on the floor outside the miqveh and in the soil fill inside it date the facility to the Second Temple period (first century BCE – first century CE).
Rock-hewn Pit. A pit (L106; diam. 1.3–1.4 m, depth 1.4 m) hewn in chalk was exposed north of the miqveh. A hewn surface was revealed slightly south of the pit (L105; 0.9–1.5 × 1.3 m; it was discovered while overseeing the covering of the excavation and therefore does not appear on the plan). Its shape might indicate that it was used as a granary. Soil fill mixed with fragments of pottery vessels from the Second Temple period (first century BCE–first century CE) was discovered in the pit.
Architectural Remains (Fig. 5). After the ritual bath went out of use and was covered with soil, a large building was constructed nearby to its east, slightly further up the slope. Two of the building’s walls (W34, W35), which formed a corner, were unearthed. The structure extended mainly south of the excavation area. A round plastered installation (L118) built of stones was exposed inside the corner of the building. Soil fill containing pottery sherds dating to the Byzantine period (fifth–seventh century CE) was discovered in the installation. It was not possible to date the building on the basis of the finds in the installation because the latter might have been constructed following the use of the building.
Terrace Wall. A section of an agricultural terrace retaining wall was exposed that was built in an east–west direction (W33; length c. 36 m; only one of the wall’s stones appears on the plan). The wall was visible on the surface to the east and west of the excavation area. The wall was constructed on soil fill that negated the corner of the building from the second phase, and it is therefore clear that it postdates that structure. Presumably the wall was erected in the modern era.

Cave. An irregular shaped rock-cut cave (M2; c. 5 × 6 m, depth to the soil fill at the bottom of the cave 1.0–1.5 m) was documented southwest of the ritual bath. The opening of the cave was arched and built of stones (Fig. 6). A wall was constructed on either side of the opening; it closed off the front of the cave from the north. Three small arched niches hewn in the bedrock (c. 0.5 × 0.6 m each) were discerned opposite the opening, in the cave’s southern wall. The beginning of a tunnel was hewn in the wall of the cave east of these niches. A small rock-cut recess (depth 0.2 m) was documented in the cave’s western wall, 1.5 m above the soil fill. Lying on the soil fill in the cave was a large trapezoid stone that had a hole in it, probably a weight (Fig. 7). The cave might have been used as an underground oil press. The recess in the cave’s western wall secured a horizontal wooden bean in a press installation, and the stone weight was used as part of the press installation. Storage jars were placed in the niches in the southern wall of the cave.