The Ayl to Ras an-Naqab Archaeological Survey, Southern Jordan 2005–2007

The Ayl to Ras an-Naqab Archaeological Survey, Southern Jordan 2005–2007  is the latest contribution of MacDonald’s archaeological survey (ARNAS) of the “Edomite plateau” in southern Jordan, the area between the Wadi el-Hasa and the Ras en-Naqab Plateau, overlooking the Hisma Desert to the south (where Edomite remains are minimal). The previous contributions covered substantial blocks of this landscape: The Wadi el Ḥasā Archaeological Survey, 1979–1983, West-Central Jordan (Waterloo, Ontario 1988), The Southern Ghors and Northeast ‘Arabah Archaeological Survey (Sheffield 1992), and The Tafila-Busayra Archaeological Survey 1999–2001, West-Central Jordan (Boston 2004). These volumes cover the 860 km² of the southernmost stretch of this landscape, between Ayl (southeast of Petra) to Ras en-Naqab in the south, and from the al-Shera escarpment in the west to the desert in the east approaching Ma῾ān (an area 25.5 km north–south x 39 km east–west). The area between Busayra and Ayl still remains outside the focus of the extensive surveys. The primary objective of the ARNAS of 2005–2007 was to “discover, record, and interpret archaeological sites” (1) in the region, disclosing the settlement patterns from the Lower Paleolithic to the modern era, including lithic and camping sites that lack any architectural remains. The region is divided into three zones: zone 1 in the west is cut by wadis from the drainage of the plateau; zone 2 is the central highlands with the major north–south routes; zone 3 in the east is the steppe on the edge of the “desert and sown.” The ARNAS is represented as the first “comprehensive and systematic survey of the area” (3).

After an introduction by MacDonald (ch. 1), the detailed catalogue of 389 sites follows (ch. 2), with more data and pictures in the enclosed CD-ROM. Most sites are in zone 2 (sites 1–209) and zone 1 (sites 210–324). Chapter 3 discusses the more intensive survey of random squares (500 x 500 m) in the three zones: 28 in zone 1, 25 in zone 2, and 88 in zone 3. These represent only 5% of the territory covered by the survey, but they yielded 115 additional sites (30% of the sites recorded). Why the most populous and settled region (zone 2) had the least random squares is left unexplained. Chapter 4 concerns the evidence for the Paleolithic period in the region, discussed by Clark, who offers a candid appraisal of the limitations of the investigation. The survey in 2005 produced 13 contextually impoverished Pleistocene artifact scatters, 14 in zones 1 and 2 and 29 in zone 3, but only one Epipaleolithic site and one Pre-Pottery Neolithic site.

In chapter 5, MacDonald discusses the settlement patterns. The result is 389 sites (11–21 [listed]), primarily dependent on the collection of 17,948 sherds, of which only 4,401 are considered “diagnostic” (24.52%), but only 2,089 are described. More than half of the sites are in the highlands of zone 2, which has the highest elevation and the most rainfall. The number for each period are: Chalcolithic–Early Bronze Age (three pottery sites; five lithic), Late Bronze Age (one site), and Iron I (seven sites). The upsurge comes in Iron II (111 sites), with a dramatic decrease in the “preclassical” (four sites) and Hellenistic (three sites) periods, and then a steady revival in the Nabataean (77 sites), Roman (192 sites), and Byzantine periods (216 sites). The different chronological phases of Nabataean painted fine ware could have provided more chronological refinement, as the half-dozen distinctive stages from ca. 150 B.C.E. to several centuries after the Roman annexation in 106 C.E. The few Early Islamic (five sites) and medieval Islamic sites (five sites) must not reflect the reality of settlement in the period, but rather the failure to discern the regional types of southern pottery. Larry Herr, the ARNAS ceramicist, is new to the region, a specialist in northern Jordan. Chapter 6 concerns the “inscriptions, rock drawings and wusūm (tribal marks)” (433–65) in the region, edited and interpreted by MacDonald. More than 60 graffiti in the so-called Hismaic script are discussed, most in zone 3, but with scattered finds elsewhere. The exception is a two-line Nabataean Aramaic graffito (site 356.1; the second line surely is the traditional šlm, “peace,” not the meaningless “(q/š)(n/l)t” in MacDonald’s text). The discussion would have been aided by a distribution map and an index of the names.

The final essay (ch. 7) provides a summary and conclusions of the ARNAS project by MacDonald. Regarding the objectives of the project, it can be termed only mildly successful. One focus was an attempt to achieve a date for the lengthy stone wall called the Khatt Shabib that runs from the Wadi al-Hasa in the north to Ras en-Naqab in the south, but the modest conclusion is that it must date to the “Nabataean [period] or earlier” (469), based on earlier proposals. A second pursuit was the transportation lattice of the region (“Roads in the Survey Territory”). The discussion is marred by misunderstandings and misrepresentations. Glueck did not “discover a painted Roman milestone” at Ayl (469); it was rather discerned as a painted text by the reviewer and published only in “Milestones with Painted Latin Texts” (D.F. Graf, Studies in the History and Archaeology of Jordan 5 [1995] 417–25). The “remnants of a ‘paved’ Roman road south of Ayl” (469) is undoubtedly the Turkish military road that led from Ma῾ān (through Ayl) to Gharandal in the Wadi Arabah observed by Glueck (Explorations in Eastern Palestine. Vol. 2. AASOR 15 [New Haven 1935] 71) and reported by Parker (Romans and Saracens: A History of the Arabian Frontier [Philadelphia 1986] 99). It is well known now that roads were paved in the region by the Ottomans and under the British Mandate. Aydini Abdullah Pasha, the governor of Damascus (1730–1733), paved roads in Transjordan, at Qal῾at al-Hasa and Mudawara, and the route to and from Ma῾ān was particularly perilous and difficult to traverse in times of rain and surely the object of other constructions (see A. Petersen, The Medieval and Ottoman Hajj Route in Jordan [Oxford 2012] 32). The ARNAS did not “discover” milestones at Kh. ad-Dur (site 064) and Kh. Qurayn (site 099), but I did during my 1986–1989 survey of the Via Nova Traiana in the region (pace 469), and they were published in Roman and Byzantine Near East (J. Humphrey, ed. JRA Suppl. 14 [Ann Arbor 1995] 250–51).

Of course, the major objective of the ARNAS was to “discover, record, and interpret archaeological sites” in the region, and MacDonald admits “we have no allusions [sic] that we recorded all the archaeological sites in the region” (467). But here the failures are both in the horizontal and vertical dimensions of the landscape. For example, site 007, Kh. as-Sadaqa (35), is listed as the second candidate of the 389 sites for future investigation, with evidence of “Iron II, poss.; Roman, Byzantine; Late Islamic” (21) periods. But Sadaqa is clearly a major Nabataean site, located at the nexus of two major roads leading to Petra. Glueck ([1935] 70–2) earlier reported “considerable amounts of Nabataean pottery” at the site, and a Nabataean tomb was excavated on the tell in 1971 that yielded Nabataean pottery of the first century C.E., including three lamps inscribed in Nabataean Aramaic; a 1973 survey of the site also indicated a Nabataean occupation (H. Kurdi, "A New Nabataean Tomb at Sadagah," ADAJ 17 [1972] 85–7). The ARNAS seems unaware of this earlier work—and the several sondages conducted in 1989 at the site that confirmed these earlier results and produced Bronze and Iron Age evidence as well; it should be noted the south and west walls of Castra Zodacatha can still be clearly traced through the jumbled ruins (cf. A. Negev, The Inscriptions of Wadi Haggag, Sinai. Qedem 5 [Jerusalem 1977] nos. 72, 104). Such a serious omission of a major Nabataean site raises questions about the methodology of the ARNAS and other “gaps” in occupation at other sites.

As a consequence, MacDonald’s effort to integrate the results of the ARNAS into the data of previous surveys must be viewed with caution. The periods of “emptying out” and “filling up” (474) in his demographic analysis may be somewhat artificial. There is also no indication of any awareness of current trends in landscape archaeology. The settlement maps may just be what Witcher has called “broken pots and meaningless dots” (“Broken Pots and Meaningless Dots? Surveying the Rural Landscapes of Roman Italy,” PBSR 74 [2006] 39–72), without any penetration of the landscape by test trenches or an attempt to “connect the dots” by intrasite connections or “network analysis” (T. Brughmans, “Connecting the Dots: Towards Archaeological Network Analysis,” OJA 29 [2010] 227–303). As a guide to the landscape of southern Jordan, the ARNAS will be useful for future archaeology in the region, but it should be regarded as impressionistic, not “comprehensive.”

David F. Graf
Department of Religious Studies
University of Miami
dgraf@miami.edu