Preliminary Report on the 1999 Swiss-Liechtenstein excavations at ez Zantur
by Bernhard Kolb and Daniel Keller
IV. Ez Zantur IV: Rooms 14, and 29 (PQ 91–2/AQ–AP)
Complete exposure of the floor-heated room 14 at the southern edge of the terrace gave further clues to the history of the building and, thanks to the good state of preservation, it offered interesting insights into the method of construction of a hypocaust in the centre of the Nabataean realm. The room measures 5.1 (NS) x 3.4 m and is paved with tiny hexagonal tiles (Fig. 6).
Fig. 6: EZ IV. Room 14 from the south (photo: D. Keller)
A row of rectangular flags (Diam. 60 cm) divides the floor into a southern section of 3 m length and a northern one of 1.5 m. The rectangular flags run on the axis of two pilasters. Originally, they supported an arch which was part of a vaulted ceiling (infra). The battered Attic base of the eastern pilaster still is in situ while the matching capital was found in the débris. A blocked doorway between rooms 4 and 14 as well as two members of a door post at the southern end of wall U prove that rooms 14, 12 and 4 were interconnected initially and shared the same floor level (923.80 m). At a date yet unknown, the floor level of room 14 was lowered by 1.5 m into the bedrock, and concurrently, the fire channel to the praefurnium 29 as well as the door to corridor 11 were cut. Moreover, closely spaced pilae or pillars of 60 cm height made of square (Diam. 19. 5 cm) and round tiles (Diam. 16 cm) respectively were installed on the levelled bedrock surface of Phase II (Fig. 7). Bipedales and cut back tegulae were placed onto the pillars – most of them are, unfortunately, cracked by now. The floor itself consists of a 10 cm thick layer of mortar with crushed pottery aggregate and of a paving of hexagonal tiles. By remodelling room 14 as a heated room, the architect followed the Vitruvian instruction (de architectura 5, 10) prescribing the placing of heated rooms in the south-west of a building. It is quite remarkable that not only the location of the hypocaust, but also such technical details of construction as measurements and proportions, seem to be based on the description of the famous Roman architect Vitruvius, de architectura, 5. 10, 2–3. Yegül 1992: 257 notes that the instructions described by Vitruvius „remained remarkably uniform in its main lines from Syria to Scotland“. .
Fig. 7: EZ IV. Room 14: view through the fire channel into the hypocaust (photo: D. Keller)
The floor was heated by the circulation of hot gasses produced by the furnace in room 29. Usually, a chimney or, more commonly, a few terracotta pipes consisting of box tiles (tubuli), which were embedded in the wall, were sufficient for creating the necessary draft and thus for maintaining an adequate circulation of hot air. Three ducts for the tubuli were carved out of the northern wall AF of room 14 (Fig. 6). Also, many fragments of box tiles (18 x 14 x 11 cm) were found scattered on the northern part of the floor (Fig. 8). The findings on the pavement indicate that the tubuli were broken out of wall AF by the tremors during the earthquake of 363 and thrown together on the floor with fragmented wall-decoration. Since the carved tubuli-leads are confined to wall AF, we may conclude that they were intended as chimneys rather than wall-heating installations. Questions regarding the function of the floor heated room remain open for the time being. Conceivable, though, is a caldarium of a private bath, or some other type of room which could be heated during the winter months. The absence of the water basin – an integral part of a Roman caldarium – is puzzling, especially in regard to the discussed background of close references to Roman models in matters of construction. Further excavations in the SW will be necessary for elucidating remaining uncertainties. The later remodelling of the narrow room 29, the praefurnium of Phase II, left hardly a trace of its former function.
Fig. 8: EZ IV. Tubulus from room 14 (photo: O. Jäggi)
As during the preceding campaign considerable quantities of the fragmented wall-decoration came to light in room 14 (locus 3215). The findings indicate that most of the painted stucco elements originally belonged to wall AF. A small section of the decoration is illustrated in Fig. 9: The main zone is set below a dentil corona and a frieze of 23 cm height. It consists of a panel, framed by a stuccoed filet with a lesbian cyma on the inner side. Around this inner frame, there are four painted decorative bands. The majority of colours applied to the corona (yellow, purple, pink, white and green) is repeated in the frieze, where by way of coloured, overlapping zigzag-bands, a gradual transition of the colours was achieved. The colours change, top to bottom, from black to purple, to pink, white, light- and dark-blue, giving the impression of a rainbow. A black zigzag-band accentuates the green zone at the bottom of the frieze. Besides the painted panels, fragments of a dark-blue frieze with figured paintings were part of the decoration. Most of the figure scenes have vanished, but a female playing the lyre, facing to the right, is preserved. The frieze is framed horizontally by a simple moulding at the bottom and a moulded and painted ovolo at the upper end respectively. The zone above the projecting ovolo was originally divided by moulded pilasters (14 x 2.5 cm). Above the pink bases, the pilasters were painted with at least two different types of decoration, one consisting of white octagons and diamonds painted, on a red background, the other of white flowers and fruit painted on a green background. On the basis of several corona fragments belonging to a lunette with preserved remains of the ceiling-decoration, we can reconstruct a barrel vault covering the room.
Fig. 9: EZ IV. Room 14: fragmented decoration from wall AF (photo: O. Jäggi)
All small finds from room 14 can be dated to the fourth century. They lay under a massive deposit of fragmented wall and ceiling decoration. The stratigraphy shows that, firstly, the room was still in use during the fourth century and, secondly, that the murals were only destroyed during the earthquake of 363.
