ARAM, 8 (1996), 253-267 253 HEGRA AND PETRA: SOME DIFFERENCES ROBERT WENNING Mada'in Salih, the Nabataean Hegra, is famous for its rock-cut tomb façades and the large Nabataean inscriptions on these façades. Most of the articles deal- ing with Hegra in recent years discuss just these monuments, as does the excel- lent book published by John F. Healey in 1993.1 Healey gives a good treatment and summary of the other archaeological evidence of the site and its historical and cultural setting as well. It seems that almost everything that can be said about Hegra has been said. There is only one group of monuments which has not received the same attention as the tombs, the votive niches and other religious installations of the Jabal Ithlib and its surroundings. I shall give some detailed interpretations of these. In addition we may be able to learn something more from a comparison of Hegra and Petra. There is no need to demonstrate the strong rela- tions between the two sites in great detail, but I would like to point especially to some differences. 1. NATURAL CONDITIONS In an article published in 19712 Meda'in Salih is described as follows: “Meda'in Salih…lies in a broad basin rather than a valley; the mountains in the immediate vicinity are much broken into isolated stacks and peaks, often weathered into fantastic shapes. It thus presents an appearance very different from that of Petra, despite the comparable architecture. The Meda'in Salih basin is low-lying and sand-filled: it is a collecting area for rain and floods,… Erosion has removed practially all traces of the free-standing town which must have existed in Nabataean times; there are no monumental remains visible,… a few traces of structures remain to testify to the ancient occupation.” I do not want to stress the natural conditions producing differences. Of course, some pecularities of Hegra are to be explained by them. The most serious fact was the lack of rocks around the centre of the settlement. Every larger rock seems to have been reserved for tomb façades. With two exceptions we do not have a triclinium, cultic cella or living room cut into these rocks. At Petra 115 triclinia and about 780 chambers of domestic function have been counted.3 The 1 J.F. Healey, The Nabataean Tomb Inscriptions of Mada'in Salih (= TIMS), (Journal of Semitic Studies Supplement, 1, Oxford, 1993). I would like to thank John F. Healey for proof- reading of the manuscript and for his valuable comments. Any remaining mistakes are mine. 2 P.J. Parr, G.L. Harding, and J.E. Dayton, BIAL, 10 (1971), 26. 3 L. Nehmé, Occident & Orient, 2 No. 1 (1997), 3. 254 HEGRA AND PETRA: SOME DIFFERENCES lack of nearby gorges has the result that there is only one sacred area with votive niches and graffiti, while are so common in the rocky surroundings of Petra. 2. SETTLEMENT PATTERN In the Hellenistic period Petra became the capital of the Nabataean kingdom in the sense that it was the seat of the tribe and its leaders. For the Greek world Petra was a metropolis4, but we are dealing with Arabs and the settlement pat- tern fits with their tradition. Careful consideration of the various pieces of stratigraphical evidence demonstrates that Petra developed from about the mid- dle of the 1st century BC to a typical Semitic settlement, open, with living quar- ters spread over a large area, and dominated by a central sanctuary with a via sacra along a Wadi, and surrounded by cemeteries. The great theatre was the only public urban building of classical type (cf. Gerasa) and is probably to be dated later as often suggested. Not before the beginning of the 2nd century, that means with the establishment of the new provincia Arabia in AD 106 was the character of the site transformed into a more Roman shape with a colonnade street, arches and public buildings.5 Concerning town planning, the changes remain limited. The settlement still existed and the population remained. Although a map of the central area of Hegra is not yet published the general layout of the site is clear and can be compared with that of Petra very well: a central settlement area surrounded by a ring of tombs and behind that a sacred area in the rocks (Fig. 1). The size of the central area at Hegra seems even to exceed that of Petra, but is also situated in the plain without an acropolis or tell. The central area was probably not only covered with living quarters but also contained the cultic centre of the site. Inscription H 36 from the tomb of Halafu mentions a temple of the god Qaysha.6 So far, neither the temple nor a via sacra have been found. Since 1986 regular excavations have been carried out at the site.7 They have revealed building structures and parts of the city wall.8 There seems to be 4 Strabo, Geogr. XVI 4, 21. 5 R. Wenning, in Th. Weber – R. Wenning, (eds.), Petra. Antike Felsstadt zwischen arabis- cher Tradition und griechischer Norm, (Mainz, 1997), 56-62. 6 J. Starcky, SDB, VII, (Paris, 1966), 1001 translates byt qys' as “la maison de la mesure” according to Arabic qa}is = measure. Refuted by Healey, TIMS 120; S. Krone, Die altarabische Gottheit al-Lat, (Heidelberger Orientalistische Studien, 23, Frankfurt, 1992), 536. 7 They are directed by Dhaifalla Al-Talhi for the Department of Antiquities in Riyadh. Pre- liminary reports can be found in Atlal, 11 (1987), 47-57; 12 (1989), 21-28; 13 (1990), 21-32; 14 (1996), 25-42. 8 I am not convinced by the thesis of E.A. Knauf, in Arabian Studies in Honour of Mahmoud Ghul, (ed. M. M. Ibrahim), (Wiesbaden, 1989), 57; E.A. Knauf, in Weber-Wenning, Petra, (1997), 17, 21, portraying Hegra as a non-Nabataean central cultic site (reqem) of the Shalamu tribe. This is also disputed by Healey, TIMS, 25 note 133. Concerning the Salamians: we cannot localize this tribe precisely, other than noting that they were to be found somewhere in the region and in the vicinity of the Nabataeans (Tayma'?). R. WENNING 255 Fig. 1: Mada'in Salih/Hegra, Map of the Sit (Courtesy J. F. Healey, TIMS Map II) nothing of a Nabataean settlement that can be dated before the middle of the 1st century BC when Hegra was taken over by the Nabataeans from the Lihyanites.9 It is not yet possible to compare the houses from both sites exactly; there seem to be differences. 3. THE TOMBS There are various differences between the Hegra and Petra tombs. These are discussed often and I can limit myself to some remarks. At Petra 619 tomb façades have been counted, at Hegra 82, of which 36 bear large inscriptions, while at Petra only the Turkmaniye tomb and in a different category the Flo- rentinus tomb (and possibly the Obelisk tomb) do. We can only speculate 9 Cf. Healey, TIMS, 25f. F.V. Sherd Winnett, – W. L. Reed, Ancient Records from North Ara- bia (= WR), (Toronto, 1970), Fig. 81,11 belongs to phase 2a according to the typology of S.G. Schmid, SHAJ, V, (1995), 637-647, which is dated to 50-30/20 BC. A coin published by A. Jaussen – R. Savignac, Mission archéologique en Arabie I, (= JS), (Paris, 1909), 441 as that of Obodas III (followed by R. Wenning, Die Nabatäer [NTOA, 3, Göttingen-Fribourg, 1987], 119) is to be attributed to Aretas IV. It corresponds to coin type No. 47 in the catalogue of Y. Meshorer, Nabatean Coins, (Qedem, 3, Jerusalem, 1975) dated to 9 BC. Even so, the coin belongs to the earliest dated artefacts from the site. The first dated tomb is from 1 BC/AD. 256 HEGRA AND PETRA: SOME DIFFERENCES about this phenomenon.10 The wish to mark the tomb clearly visible as belong- ing to a particular family for posterity by writing the legal agreements about the use of the tomb on the façade in addition to the original document deposited in the temple or a civic registry (note that the façades are directed towards the settlement/temple) could find an explanation when we consider that owners of the tombs may have been at Hegra only for a shorter period, so that there was nobody to safeguard the family tomb after their death. Other differences are found inside the tomb chambers. In general the cham- bers are less well carved at Hegra and there seems to be a greater use of bur- ial niches high in the walls, while at Petra loculi at the ground level are pre- ferred. The average number of burials in one tomb is greater at Hegra. This might be explained by the limited number of available chambers; the inscrip- tions point to the fact that different sections of one family had to share a tomb. On the other hand, the insistent prohibitions in the inscriptions could easily reflect just a normal situation in that owners like to prevent through their doc- uments and inscriptions the selling or leasing of burial places and tombs. An important consideration in the classification of the tombs was to relate the different types of façades to the social background of the owners.11 Ten of the tombs belonged to families of high ranking officers. This should not be surprising for a frontier district of such strategic and commercial importance. The military aspect seems to be overstressed in scholarly discussion.12 It should be bore in mind that the inscriptions reflect a very complex society at Hegra.