III. First results – f. Parallels and interpretation
Although we can describe and explain how water reached the different buildings, it still remains surprising that somebody could have built an entire bathing installation on top of the most prominent and isolated elevation of the area. An identification as a public bath can presumably be excluded by its location, very difficult to access for just a short bathing experience. If we try to sum up some of the known characteristics of this installation, it becomes clear that it must have been an outstanding one in several respects. It is situated not only at one of the most prominent spots of the hill, offering a splendid view over the city centre of Petra, but it must also have been visible from all over Petra. Further, it is at the most distant point of the entire plateau in relation to the steps giving access to it. Irrespective of whether there were other paths leading up Umm al-Biyara, the one followed by the modern steps, which are a restoration of the Nabataean steps, surely was the most „official“ access to the plateau during the Nabataean period. Therefore, the bathing installation is both very prominent and at the same time very private, since access to it was strongly reduced and controlled. Since that installation made not only use of water, already a luxury item in this specific location, but also of wood or other fuel needed to heat the floor and wall heating systems described above, we are facing an almost provocative display of wealth and luxury. Despite the fact that heated rooms per se were not necessarily considered a specific luxury item by the time of their construction, the fact that they are situated on top of the highest elevation in the region makes them outstanding, since every single twig that was burned in their praefurnia needed to be carried up the hill.
That this has to be a building out of the ordinary is further suggested by the general geo-strategic situation of Umm al-Biyara as described above. It is irrelevant whether Umm al-Biyara is the „rock“ of the Nabataeans reported for the year 312/11 BC by Diodorus: by the late 1st century BC and the 1st century CE, Umm al-Biyara must have been sufficiently important that not everybody was allowed to build there. It is precisely this combination of strategic importance and ostentatious demonstration of wealth that places these Nabataean buildings in close relationship with some of Herod the Great’s hilltop palaces. In Masada, Herodeion, Kypros and Machaerus (Machairous), heated rooms, usually as part of Roman style thermae, are an outstanding characteristic On the hilltop palaces of Masada, Herodeion, Kypros and Machaerus, see Netzer 2001b; Japp 2000; Lichtenberger 1999; Roller 1998; Nielsen 1994, 181–208; especially on their bathing installations see Netzer 1999. . Beside the pools etc. of major bathing installations, individual bathtubs are common to most of the mentioned Hasmonean and Herodian structures Cf. Netzer 1999. ; however, no bathtubs for two or three persons such as the one mentioned above in ST 20 seem to be attested from Herodian buildings.
We can assume that these Herodian installations were not only known to the Nabataean upper class Cf. Schmid 2009. , but especially the palace at Machaerus, situated on the eastern shore of the Dead Sea, must have been in many ways a sort of provocation for the Nabataeans. It seems, therefore, perfectly appropriate to suggest that the building on top of Umm al-Biyara consisted of something like the Nabataean response to the Herodian hilltop palaces. Probably the best overall comparison is offered, for the time being, by the Herodian palaces at Masada Specifically on Masada see Netzer 1991. . The general situation is the same, i. e. the Herodian buildings are displayed all over the plateau of the massif rock elevation that is Masada, and, as on Umm al Biyara, there is no common orientation for all buildings, rather they form smaller groups according to there successive date of construction. There, too, the most luxurious and at the same time the most private structures, the ones known as the North palace, are placed at the spot opposite to the main access to the hill. As at Umm al-Biyara these Herodian structures are playing with visibility, incorporating the splendid panoramic view into the architectural display, as is especially true for the three levels of the northern palace. Likewise, they feature lavishly decorated bathing installations.
Despite the fact the Masada offers the best overall comparisons to our structures from Umm al-Biyara, in details most of the other Herodian residences can be compared as well. For instance, the deliberate playing with visibility and the view is very prominent within Herod’s third palace at Jericho Netzer 2001: 231–286. . The triclinium B70 (ibid. 239) and the courtyard B55 (ibid. 251–254) from Jericho can be compared to our ST 26 with its extreme position built literally over the cliff. Since the southern wall of courtyard B55 in Jericho fell into the Wadi Qelt and cannot be reconstructed securely, as is the case with the eastern wall of our ST 20, in both cases a direct opening to the natural view would be possible.
Fig. 23: Painted Nabataean sherds from ST 26 (left), ST 12 (top right) and ST 20 (bottom right) repectively (photo: S. G. Schmid)
In terms of chronology we can already make a few observations. Of course, a precise chronology and dating of the structures referred to above will only be possible after systematic stratigraphic excavation. For the time being the earliest Nabataean element consists of one single painted rim sherd of a drinking bowl dating to ca. 50–25 BC Phase 2a after Schmid 2000. (fig. 23 top left). It was found on ST 12 (cf. fig. 3), being the highest elevation on the plateau consisting of a outcropping rock that was incorporated into a built structure, as is attested by a few remaining stone blocks from a wall, as well as by the carefully carved stone „steps“ onto which the walls were set. A few metres to the W of the rock a built wall is visible, indicating that the structure was probably bigger than just the part built on the rock. The rest of the pottery related to ST 12 consists of several sherds belonging to the last quarter of the 1st century BC (fig. 23 bottom) and others belonging to the 1st c. CE. The last quarter of the 1st century BC Phase 2c of Nabataean pottery according to Schmid 2000. is the earliest date present within most areas, except of course the Edomite pottery. The latest pottery that is present to some extent in most areas belongs to phase 3c of Nabataean pottery according to Schmid 2000. and dates, therefore, to the early 2nd c. CE. Within the major Nabataean structures sometimes a difference in chronology seems to occur. For instance, on the steep slopes immediately to the N of ST 26, an large amount of very well preserved, although broken Nabataean pottery was found. Beside a few fragments of coarse ware, most of that pottery consists of Nabataean fine ware and especially of fragments of painted bowls. Within these, phases 2b and 2c (last quarter 1st century BC and first quarter 1st c. CE) are the most prominently represented, followed by phase 3a (second and third quarter of the 1st c. CE). Phase 3b (last quarter 1st c. CE) is very discreetly represented, phase 3c (early 2nd c. CE) completely absent.
Fig. 24: Mortar fragments from hypocaust room in ST 20 (photo: S. G. Schmid)
Quite different is the picture in and around ST 20, the bathing installation pointed out above. There, painted pottery strays from phases 2b (last quarter of 1st century BC) to 3c (early 2nd c. CE), the most prominent group being the one of phase 3b (last quarter of the 1st c. CE). Although one should not come to hasty conclusions, it would seem, then, that the major phase of occupation of ST 26 should date to the turning of the eras, while ST 20 probably saw its apogee in the late 1st c. CE. The latter would also fit well both with the greyish hydraulic mortar using charcoal additive On the chronology of the different hydraulic mortars see Graf et al. 2007: 225–227. as well as with the tubuli introduced to the Nabataean realm towards the end of the 1st c. CE. On this see Kolb and Keller 2001; 2000. . Maybe there is even a way for a more precise dating of the heated room within ST 20. In the area of the above mentioned hypocaust construction, several fragments of hydraulic mortar were found, showing the reddish variant using crushed pottery additions on one side, the greyish variant with charcoal fragments on the other side (fig. 24). While the reddish variant was in use from the 1st century BC onwards, the greyish one was introduced towards the end of the 1st c. CE; therefore, the combination of both would indicate a date at the very beginning of the use of the greyish variant.
