During November 2011 and January 2012, two plastered rock-hewn miqwa’ot (ritual baths)were documented at Khirbat Kafr Sum (map ref. 20877–80/626605), in the wake of routine activity by inspectors of the Unit for the Prevention of Antiquities Robbery of the Israel Antiquities Authority. The documentation, on behalf of the Israel Antiquities Authority, was directed by E. Klein (photography and drafting), with the assistance of A. Ganor, S. Ganor, S. Ya‘aran and Y. Vashna (surveying and drafting).
Khirbat Kafr Sum is situated in the Jerusalem hills, on a hilltop that rises to an elevation of 735 m above sea level and is bounded on the north and south by deep ravines that empty into Nahal Soreq and Nahal Refa’im. To the west, the ruin is adjacent to Horvot Betar (Khirbet al-Yahudi) and Khirbat Qove, and is located c. 1 km north of the route of the Roman road that connected Jerusalem to Bet Gurvin. Remains of an Arab village that is built on ancient settlement ruins is visible at the site, and in its southern part is an ashlar-built structure that was used as a sheikh’s tomb. The ancient settlement remains include buildings, pits, rock-hewn installations and a large pool. The site has been a prime target for antiquities robbers for many years.
Miqwe 1 (map ref. 208777/626653; Fig. 1) is hewn in a bedrock outcrop c. 2 m high; it is located in a forest planted along the fringes of the site in its northwestern part. A rock-hewn passageway (width 2.7 m) leads to the bath. The passageway is currently filled with alluvium and rocks and it was therefore impossible to ascertain its exact shape. However, it is clear that the passageway served as a drainage surface for the rainwater that filled the miqwe. The miqwe had two adjacent arched openings (a double opening), separated by a hewn central pillar of irregular shape (Fig. 2). The dimensions of the openings (west—width 1, height c. 1.5 m; east—width 1.5 m, height 1.4 m) necessitate crouching to enter the bath. The descent to the immersion pool is by way of two rock-hewn steps that run the entire width of the installation (Fig. 3). The pool is rectangular (2.8×3.8 m, depth c. 1.75 m) and its floor, ceiling and sides were coated with two layers of gray hydraulic plaster that contained small pebbles. This kind of plaster is typical of the Late Second Temple period. The rainwater that was collected in the immersion pool flowed through a channel (width 0.1 m, depth 0.1 m) that was installed on the upper step, in the center of the eastern entrance threshold.
Miqwe 2 (map ref. 208807/626603; Fig. 4) was installed in a low bedrock outcrop, located c. 50 m east of the Miqwe 1. A rock-hewn entrance (width c. 2 m) that is currently covered with soil and alluvium leads to the miqwe. This bath was also provided with two openings (west— width 0.8 m, east—width 0.7 m), separated by a rectangular bedrock pillar (length 0.9 m, width 0.3 m; Fig. 5). The height of the openings is unknown because of the alluvium that has accumulated at their bottom. The two openings were blocked by a stone wall after the installation was no longer in use. The upper part of the stone wall blocking the western opening was dismantled by antiquities robbers during plundering. It was possible to discern at several spots a single bedrock step (height c. 0.35 m) that was installed in a straight line with the southern end of the bedrock pillar and facilitated the descent to the immersion pool. The immersion pool is rectangular (1.3×2.2 m, depth c. 2 m) and its floor, ceiling and sides were coated with a layer of gray hydraulic plaster, identical to the plaster in Miqwe 1.
Although no finds that can date the installations were discovered, the characteristic architecture and the plaster covering the sides indicate they were hewn and used during the Early Roman period. These ritual baths constitute the first archaeological evidence that the site was inhabited in the Second Temple period by Jews who adhered to the laws of purity. The ritual baths that have a double opening enable to separate the unclean coming in from the clean going out. This separation is also mentioned in the rabbinic sources: "All vessels found in Jerusalem on the path leading down to the house of immersion are assumed to be defiled but [those found] on the way up are deemed pure. For [there were two paths and] the path [used by those defiled] leading down [to the miqwe] was not the [same] path leading up [used by those who immersed]" (Sheqalim 8, 2) and in Rashi’s commentary, "one path leads down to the house of immersion and another path leads up".
Similar installations equipped with a double opening were mainly discovered in Jerusalem and its environs, for example in excavations in the Jewish Quarter in Jerusalem, the Temple Mount, Tomb of the Kings, Bethany (Reich 1980), at Khirbat Hilal (ESI 10:150–151) and near Alon Shevut (Amit 1999