One of the structures that provided the most intelligible results during the 2010 survey was Structure 20 (ST 20 on fig. 2; fig. 11), standing on the NE edge of the plateau, prominently overlooking the city centre. The structure consists of several clearly visible rooms. From the S end of the building, a partially rock-cut and partially built water channel brings in water to the structure. The water is collected in a substantial basin, identified thanks to the greyish hydraulic mortar containing charcoal fragments which improved its waterproofing qualities. On a slightly lower level, but on the same alignment within the building, is a huge room from which came a large number of hypocaust and tubuli fragments during the 2010 survey.
Fig. 11: Umm al-Biyara, detail of plan showing ST 20 (plan: G. Teltsch and W. Kennedy)
In order to find out more about this and the adjacent rooms, all oriented more or less in N-S-direction, a lengthy trench was put down in E-W-direction. It became apparent very soon that the supposed hypocaust room indeed was built with both a floor- and a wall-heating system. Some of the suspensurae, forming small pillars usually with a rectangular base and then disk-shaped elements, still were standing in situ and in the collapsed debris of the walls, many fragments of tubuli mixed with plaster were found (fig. 12). The so far excavated part of the room measures 3.5 m x 2.00 m but, as becomes clear from the visible walls at the surface, must have been at least twice that size, not taking into account the possibility that other parts may have been collapsed down the steep cliff. The hypocaust pillars are built on a layer of big, uncut stones, the space between the pillars was covered with a dense layer of black ashes. Between the big stones some pottery belonging to the last quarter of the 1st c. CE was found. This together with the abundantly used grayish mortar would indicate a construction date for the hypocaust room towards the late 1st c. CE.
Fig. 12: Umm al-Biyara, ST 20. Hypocaust room during excavation (photo: S. G. Schmid)
Adjacent to the W of the hypocaust room follows a huge hall with 7.5 m of width and a yet unknown length, out of which 7.5 m have been excavated so far (fig. 13). This major hall is subdivided by a kind of pillars that clearly show two phases. Directly put on the floor of the hall, at the spot that could be the middle of it, is a small wall of one only course of stones/rock. While in the first phase the two (probably central) pillars were directly incorporated into the small wall, in a second phase the pillars were enlarged and built against the wall, as can be clearly seen by the fact, that one layer of plaster is applied beyond the pillar and the next one around it (fig. 14).
Fig. 13: Umm al-Biyara, ST 20. Big hall with collapsed pillar (photo: S. G. Schmid)Fig. 14: Umm al-Biyara, ST 20. Detail of pillar, later attached top re-existing low wall (photo: S. G. Schmid)
In this room as well, massive collapsed debris indicates a violent destruction that most probably can be assigned to an earthquake. And as in Structure 26 (cf. above), in this room as well fragments of late Roman – early Byzantine lamps were found directly on the floor slabs (fig. 15), indicating a destruction most probably related to the 363 CE earthquake.
Fig. 15: Umm al-Biyara, ST 20, fragment of late Roman – early Byzantine lamp from the big hall of the bathing complex (photo: S. G. Schmid)
During the final destruction of this huge hall, what seems to be the original interior decoration collapsed as well. The torso of a naked male marble statuette, measuring 54 cm in height, was stuck head down directly on the floor slabs and squeezed in by the collapsed ashlars of one of the pillars (fig. 16). The quite prominent belly and generally rounded forms of the sculpture (fig. 17) identify it as a young boy. Similar figures are well known from the Hellenistic and Roman imperial period. Our torso corresponds exactly to a specific type of fountain decoration, a young boy holding a huge vase on his left shoulder out of which the water once sparkled (cf. the list of comparable statues in Kapossy 1969: 41–42). The same type of fountain decoration also occurs with small wings and, therefore, is to be identified as Eros, the child of Aphrodite. In any case, the specific type of sculpture always belongs to the context of fountains with sparkling water. We therefore have yet another indication of the spectacular use of the precious liquid on top of Umm al-Biyara.
Fig. 16: Umm al-Biyara, ST 20. Big hall with marble statuette stuck between collapsed architectural members (photo: S. G. Schmid)Fig. 17: Umm al-Biyara, ST 20. Marble statuette of statuette of nude male figure wearing a cloak (photo: S. G. Schmid)
A second fragmented sculpture could belong to the Dionysiac world. The small statuette, measuring 28 cm in height in its actual state of preservation, was found broken in several fragments in the same debris as the previously described (fig. 18). The small pad of a feline – most probably a panther – belonging to the cloak of the person, would definitely make that person a follower of the god of wine. What is interesting is that the two sculptures cannot be dated later than the 1st c. CE and must, therefore, have been exposed during roughly 300 years as splendid decoration of the huge hall.
Fig. 18: Umm al-Biyara, ST 20. Marble statuette of male figure wearing a panther skin (photo: S. G. Schmid)
At the N end of the building, two rectangular structures measuring 90 x 140 cm and 160 x 220 cm respectively were clearly visible on the surface already in 2010 and during the 2010 season we did clean the smaller one that turned out to be a bathtub (cf. fig. 19). During the 2011 season, it turned out that within the W wall of the huge hall dealt with above, in about 1.40 m height, a water pipe consisting of clay tubes was found that is directly led into the small bathtub all along the wall (fig. 13, in the background).
Fig. 19: Small bathtub within ST 20 after cleaning with outflow (photo: S. G. Schmid)
To the W of the smaller bathtub is a narrow room showing one identifiable doorway in situ. Next to this doorway, the fragment of a twisted stand of very fine craftsmanship was collected in 2010, made of hard dark gray, almost black soft stone and showing traces of fine whitish plaster. During the 2011 excavation season, the upper part of the same stand was discovered just inside the small room at the N end of the bathing complex (fig. 20) and, therefore, basically at the same spot as the lower part from 2010. In the middle of the stand a vertical rounded hole goes through the entire preserved length of it, posing problems of interpretation but maybe also offering a possible solution. Given the specific context, one could imagine the object as a stand of a table or a basin In general terms on basin stands see Pimpl 1997, although that author does not list any twisted basin stand nor stands with holes. . In that case, the round hole would have been used to pump water from the bottom into the basin, simply using pressure provided by gravity flow; all that is needed is a water tank on a slightly higher level than the basin. That in general terms basin stands in the shape of small columns were a common feature by the 1st c. CE can be shown by several such objects from Pompeii, although there too no twisted stands seem to be known Pernice 1932: 38–54. . Also from several Pompeian examples comes the confirmation that spectacular water installations such as fountains and pools belonged to the usual features for decorating gardens and thermal installations of the contemporary upper class See some examples in Farrar 1998: 64–96. . Finally, in the Casa del Camillo at Pompeii, a table shaped fountain on a column stand is used in exactly such a way, forcing the water through a central hole of the stand Cf. Andersson 1990: 234f. with fig. 19. .
Fig. 20: Umm al-Biyara, ST 20. Upper end of a twisted basin stand (photo: S. G. Schmid)
The small room that seems to form the end of the built structures at the N side of the bathing complex deserves special attention. The room with maximal dimensions of 4.0 m x 5.3 m is built around and against the small and the big bathtub respectively (fig. 21). Parallel to the two steps that are leading into the room at its SW corner, a small rock cut water channel brings water into the room that initially was paved with rather big floor slabs of good workmanship. At the level of the floor slabs the water is – at least partially – diverted into a partially rock cut and partially stone built water channel that runs across the room in a Pi-shaped alignment. The stones of that channel and the floor slabs of the room are at exactly the same level; it is, therefore, clear that the water channel was not covered and must have been directly visible and reachable by whom ever used the room. At the end of that water channel, the water flows into a larger channel of 40 cm of width that is plastered with huge stones. Since the stones are not set very closely and since there is no mortar used to fix them, this channel most likely was for used (waste) water. Exactly opposite from the spot where the water from the built channels flows into the larger waste water channel, there is the outflow of the huge bathtub that equally flows into the same wastewater channel. So, there is a quite substantial volume of water flowing through that channel. All the water that runs through this room leaves it finally through a large opening at its NW corner.
Fig. 21: Umm al-Biyara, ST 20. Last room of the complex, presumably a toilet (photo: S. G. Schmid)
This particular organisation of water requires an explanation, since it would have been easily possible to direct the water from the huge bathtub directly outside the building into the slope, as the entire bathing complex apparently is the last building before the steep rocks at that point of the plateau of Umm al-Biyara. The outlet of the room with the Pi-shaped water channel directly included into its surface and the massive waste water channel at a deeper level strongly recalls the plan of Roman type toilets. The only thing that is missing so far from the structure at Umm al-Biyara are the seats. Since these most probably were leaning at or even directly included in the outer N wall of the room, it is most likely that they collapsed together with that wall. The construction of a toilet at this spot makes perfectly sense: It is at the lowest point and at the end of the entire complex. Since there is anyway a lot of water used in this complex which, as we shall recall, was a bathing installation, the water used in the bathtubs etc. could be used afterwards in order to flush the toilet. Also the rather big dimensions of the waste water channel and the final outflow from the room (and therefore the entire building) that measure both 40 cm in width would fit that explanation.
Fig. 22: Umm al-Biyara, ST 20. Last room of the complex, presumably a toilet, with destruction debris (photo: S. G. Schmid)
The chronological elements and information related to ST 20 likely confirm and sharpen the picture as it is drawn from the results of the other structures. According to the massive debris within in the room, with all blocks fallen in a common direction (fig. 22), it is quite likely to suppose that an earthquake is responsible for its destruction. At the very bottom of the sewage channel fragments of late Roman – early Byzantine lamps were found (fig. 23), deposited there together with an important amount of pottery. It seems, therefore, logic to suggest a destruction by the earthquake of 363 CE that most likely destroyed most if not all of the buildings on Umm al-Biyara.
Fig. 23: Umm al-Biyara, ST 20, fragment of late Roman – early Byzantine lamp from the sewage channel in the toilet of the bathing complex (photo: S. G. Schmid)
In a small sounding beneath one of the floor slabs of the presumed toilet, some important hints as to the construction of the room were obtained. Although put almost directly on the bedrock, enough fill consisting of layers of smaller and flat stones and clay containing earth was found that contained a small amount of pottery sherds (fig. 24). The pottery from beneath the floor slab is homogeneously Nabataean and the painted fragments all belong to phases 3a and 3b respectively, i. e. to the third and forth quarter of the 1st c. CE (fig. 25). We can, therefore, reckon a construction of the presumed toilet within the forth quarter of the 1st c. CE. Since that room cannot be explained without the entire bathing installation connected to it, this is most likely the date of the construction of the entire complex.
Fig. 24: Umm al-Biyara, ST 20. Small sounding beneath floorslab in the toilet (photo: S. G. Schmid)Fig. 25: Umm al-Biyara, ST 20. Rim sherds and painted Nabataean pottery from small sounding beneath floor slab of the toilet (photo: S. G. Schmid)
