BMCR 2007.06.18

How Petra was built: an analysis of the construction techniques of the Nabataean freestanding buildings and rock-cut monuments in Petra, Jordan

, How Petra was built: an analysis of the construction techniques of the Nabataean freestanding buildings and rock-cut monuments in Petra, Jordan. BAR international series 1460. Oxford: Archaeopress, 2005. xii, 237 pages. ISBN 9781841718989 £36.00.

The past 25 years have seen a renaissance in the study of Petra and the Nabataeans. Numerous Jordanian and international teams have carried out extensive survey and excavation within the settlement centre, the Siq, and in outlying areas such as Jebel Haroun, Wadi Sabra, and Beidha. Important field research has also been conducted elsewhere in the region formerly occupied by the Nabataean kingdom, which spread across portions of present-day Israel, Palestine, Saudia Arabia, Jordan, and Syria. Much of this work is reflected and cited in the excellent survey Petra Rediscovered, issued in connection with the exhibition “Petra: Lost City of Stone” that toured North America in 2004-5.1 Such new information constantly updates and refines our understanding of the history, religion, languages, and material culture of the Nabataeans, making references to “lost” cities and “mysterious” desert dwellers seem quaint and old-fashioned. Our understanding of Nabataean architecture naturally has also benefited from the intensified field research, and numerous recent publications deal with the sacred, domestic, and hydraulic structures constructed by the Nabataeans, along with the striking tomb architecture carved into the brightly coloured sandstones of Arabia Petraea.

Given the large amount of material now available for study, Rababeh (hereafter R.)’s book is timely and important. Despite the general ease of passage and consequent movement of people and ideas around the Mediterranean world in antiquity, each culture developed an architectural tradition that was distinct from that of its neighbours. Given the inevitable local variations in available materials, topography, population density, and technological expertise, the techniques of construction can vary as much as the types and plans of the structures built. Although R.’s book focuses on one site, some evidence is drawn from elsewhere, and the result is a useful survey of Nabataean construction technology, well documented and well illustrated. The book shows some signs of its genesis as an Oxford DPhil thesis (2005?)—the potted history of the Nabataeans, for example, that occupies much of Chapter 1, comments such as “This dissertation is divided into six chapters” (p. 29), and the unfortunate absence of an index—but it is well written and tightly constructed, and the author’s training as an architect serves him well.

The book aims “to collect and document the technical features of the construction of the rock-cut and the freestanding monuments in Petra during the Nabataean period” (interpreted as 312 B.C. to A.D. 106) and to determine “precisely when and why these features appeared” (p. 29). R. is correct that this is the first extensive investigation of the subject. He cites most of the articles and short sections of books that have already dealt with the topic (p. 29 n. 219), but he asserts that most of the evidence cited results from his own fieldwork. This autopsy was made possible by the focus on a single site. Nearly all of the photographs were taken by R., and most of the drawings executed by him; they are uniformly of high quality and form an essential and effective complement to the text. The arrangement of the subject matter after the introductory chapter is logical: Chap. II: “Building Materials” (stone, wood, metals, and “other”); Chap. III: “Quarries and Quarrying in Petra” (location, techniques, transportation, relation to landscape); Chap. IV: “Stone Dressing and Lifting” (block preparation, dressing of tomb facades, measuring tools and modules, lifting methods); Chap. V: “Construction of Walls, Columns and Floors” (including anti-seismic and stabilizing techniques); Chap. VI: “Construction of Roofs” (forces involved, arched and vaulted structures, the complex roofing of larger structures involving both tension and compression). There is a short “Conclusions” section and a full bibliography.

The chapter on building materials begins, of necessity, with a discussion of the geology of the Petra region. Within Petra itself the bedrock consists of three sandstone formations: Disi above the Umm Ishrin above the Salib Arkosic deposits. The tombs in the city centre are carved in various sub-levels of the Umm Ishrin formation, and the quarries for blocks were cut from this formation as well. Harder limestones were imported from the area of Wadi Musa, above Petra, and from the area around modern Ma’an, 50 km to the southeast. White and white-grey marbles were used sparingly as well in Nabataean structures, but imported from far outside the region. R. notes that analyses have not been carried out for the early marbles, mentions some unlikely regional sources, but concludes—without any real documentation—that the Luna quarries of Italy were the source. This seems very unlikely, since extensive production and exportation from the Luna quarries did not begin until ca. 40 B.C., and there were many pre-existing sources of white marble in the eastern Mediterranean. I have not found any references to veneers and architectural elements of Luna marble in the Levant, and Ward-Perkins implies that Rome soaked up most of the production.2 The source of the granite columns found at Petra is also in doubt, but they definitely came from outside the Nabataean kingdom. Although the region around Petra has few trees today, wood was an important building material. The Nabataeans used it to give tensile reinforcement to masonry walls, to support roof structures, to construct scaffolding and formwork for arches and vaults, and for wedges and dowels. The author concludes that the Nabataeans imported their construction timber from Lebanon, but the topic of the timber resources of this region in antiquity requires much more research before such dramatic measures can be assumed. The Nabataeans used bronze and iron fixtures, fittings, hooks, and rods, but did not use metal clamps, probably because the sandstone was too friable to make this measure useful. Various mortars were used as well, mainly in wall cores, arch spandrels, and the linings of cisterns.

In Chapter III R. deals with the important issue of the quarries at Petra. Nearly all the landscape of Petra is sandstone bedrock, so why were some locations selected for quarries, and others not? He isolates as factors the distance from the building to be constructed, geomorphology, landscape, and the quality and size of blocks required. R.’s classification of the quarries as “primary, levelling, and tomb quarries” seems appropriate. The primary quarries were opened simply to produce sandstone blocks. He calculates the probable yield of each quarry, assuming wastage at about 30%, and comes to a useable total of 78,900 cubic metres. Each cubic metre of useable stone representing at least four large blocks of 0.50 x 0.50 x 1.0 m, resulting (by my calculation) in a total of 315,600 blocks. This represents an enormous quantity of building material, and the wastage could have been used as fill for buildings and streets. Levelling site quarries occur where it was necessary to provide a flat space for construction of a large building, as around the “Great Temple,” the “pool complex,” and the “upper and lower markets.” R. calculates a total volume of useable blocks removed at 27,500 cubic metres, excluding wastage, representing (by my calculation) a further 110,000 blocks. He terms the final category tomb quarries. Production of the enormous rock-cut tomb facades at Petra had to begin with reduction of the selected cliff face to a flat vertical surface, often with a level platform on top, set in from the original surface. It was both more efficient and more practical to remove the unwanted stone in the form of blocks rather than just hacking it away in bulk. R. estimates that 63,000 cubic metres of stone were extracted in this fashion, adding another 252,000 blocks for a grand total of 677,600 blocks.

The Nabataeans certainly made the best of the topographical situation fate allotted to them, but where did all these blocks go? R. makes an estimate of the total volume of stone required to furnish three well-documented buildings in the urban centre of Petra: 7,000 cubic metres for the Qasr Bint, “Great Temple” 15,000 cubic metres, and “Temple of Winged Lions,” 3,000 cubic metres, for a total of 25,000 cubic metres. Since these structures cover only about “10 percent of the total area of the city centre and az-Zantur” (p. 58), R. concludes that the quarry sources mentioned above would have had to be supplemented by the quarrying of stone to level the various building sites. The result is very hypothetical, but a useful statistic for further discussion.

Surviving quarry faces indicate that the Nabataean workers extracted the stone blocks by means of the trench and wedge method, which has been documented throughout the ancient Mediterranean world. Picks of appropriate shape are represented in quarry marks, and one iron example was found in the unfinished el-Habis tomb (figs. 3.14-15). R. works out logical methods for efficiently removing the stone from the tomb faces. He also argues persuasively against the use of wooden scaffolding during the production of the approximately 800 rock-cut tomb facades. The cuttings supposedly used by such scaffolding would have been inappropriate for that function, and it would have been very difficult to obtain the amount of wood required. The author proposes instead a relatively simple and efficient method for carving the facades from top to bottom by means of a narrow working surface of stone left in front of the desired tomb faade after quarrying away the stone of the sloping natural surface of the cliff. Logic and several unfinished tomb facades support this proposal. On the basis of discussions with Bedoul stone workers, and his own experience as an architect, R. calculates that “it would have taken approximately three years to complete el-Khazneh” (p. 71), one of the largest and certainly the most elaborate of the tomb facades at Petra. The author proposes as well reasonable procedures for quarrying out the interior chambers and for overall organization of the work. He seems to have missed Healey’s useful discussion of tomb construction and masons’ names.3

Transport of the blocks produced during quarrying and tomb production is an important issue in such a hilly landscape. It is likely that the structures near the isolated ed-Deir tomb were produced during the quarrying of that large facade. Elsewhere, blocks were moved longer distances. The author suggests that blocks were trimmed as much as possible in the quarry, then moved on sledges. I am less convinced by the fascinating suggestion that the “landscape effect [of the quarries] was aesthetically driven” (p. 83); that is, the vertical faces and step-like remains of block removal visible around the city communicated some positive message to the inhabitants of the city. R. touches on an important issue, however, when he say it follows that “there may have been an authority to assign plots for the quarries and for the tombs,” the quarries may have been “a useful source of income for the city,” and therefore officials would have wanted to control them. It is likely that there was in fact some control over the siting of tomb facades and quarries, but as the author admits (p. 83 n. 105) we have absolutely no evidence that might indicate who was in charge. The king might well have had final authority, but the main decisions could have been in the hands of heads of clans, all jousting for the best locations.

Chapter IV concerns the methods for finishing the blocks and lifting them into position. For the most part these techniques fit easily into the Greco-Roman tradition. The softness of the sandstone, however, and its layering meant that splitting blocks off along natural planes in the rock was an important procedure, as was the occasional finishing of blocks by sawing. The claw chisel was used to produce the famous “Nabataean diagonal dressing” on blocks, and drills were used to prepare and finish some details. Finishing of blocks and facades could be carried out by pecking, or with stone pounders or polishers. The masons clearly knew how to use the straightedge, setsquare, template, and compass. The finer buildings were finished off with several layers of lime plaster for the sake of weatherproofing and decoration. Although the details are uncertain, it seems that some sort of cubit was the module used. Where necessary, blocks were lifted with bipod cranes fitted with compound pulleys.

In Chapter V R. compiles evidence for the methods of construction of walls, columns, and floors. These techniques are again fairly standard for the Mediterranean region, with some allowances for the softer stones used. Header and stretcher coursing was common in the walls of larger or finer structures, but two-skinned construction with a core of rubble was used frequently as well. Wooden beams were embedded in walls to stabilize them during earthquakes. R. asserts that such tie beams supporting arches and walls, found in the Winged Lions, were “of seminal importance for the history of Nabataean architecture” and “form a precedent for similar structures found later in early Muslim architecture” (p. 146). Roof construction, discussed in Chapter VI, is more complex. The friable character of the local bedrock did not allow long unsupported spans by single blocks, and the Nabataean architects learned at some point in the first century B.C. to roof interior spaces with cross arches carrying wooden beams or short sandstone slabs. Although R. mentions the use of this technique at Delos, he does not comment on the possibility that Nabataean merchants brought this technique back to Nabataea from the Aegean.4 He sees Alexandrian architecture as the main outside influence on Nabataean architects (pp. 23-4). For the larger temples he reasonably restores the prop-and-lintel technique.

R. has produced a very interesting and useful study of Nabataean engineering at Petra. The main drawback of this pioneering book is the nearly complete focus on Petra, excluding the growing body of evidence for Nabataean engineering elsewhere in the region. While this restriction makes sense for a dissertation, one hopes that the author will continue his research with the study of the other Nabataean settlement sites. In the “Conclusions” section the author indicates that this expansion of his study is one of his goals. He would also like to focus in more detail on the manpower requirements and procedures for carving the rock-cut tombs. Even with the focus on Petra some topics are neglected. More should be said about the composition and the role of mortars and plasters (pp. 47-8, 120-24), and the role of Hellenistic architectural ideas and techniques in the development of Nabataean architecture. The discussion of modules might also have been more detailed. Nevertheless, we must remain grateful to the author for his clear and very well illustrated review of the topic.


1. G. Markoe, ed., Petra Rediscovered: Lost City of the Nabataeans. New York: Thames and Hudson, 2003.

2. H. Dodge and B. Ward-Perkins, Marble in Antiquity: Collected Papers of J.B. Ward-Perkins. London: British School at Rome, 1992, 21 n. 30, 23-4.

3. J.F. Healey, The Nabataean Tomb Inscriptions of Medai’in Salih. Oxford: Oxford University Press, 1994, 6, 93, 290.

4. J. P. Oleson, “The Origins and Design of Nabataean Water-Supply Systems,” Studies in the History and Archaeology of Jordan, 5 (1995) 707-19; “Water-Supply in Jordan,” pp. 603-14 in B. MacDonald et al. ed., The Archaeology of Jordan. Sheffield: Sheffield Academic Press, 2001.