Jabal Qurma Archaeological Landscape Project 2013-2014

Volume 118, Number 4 October 2014

AMERICAN JOURNAL OF

ARCHAEOLOGY The Journal of the Archaeological Institute of America

www.ajaonline.org

AMERICAN JOURNAL OF

ARCHAEOLOGY This article is © The Archaeological Institute of America and was originally published in AJA 118(4):627–76. This e-print is supplied to the author for non-commercial use only, following the terms outlined in the accompanying cover letter. The definitive electronic version of the article can be found at: www.jstor.org/stable/10.3764/aja.118.4.0627

NEWSLETTER

Archaeology in Jordan, 2012 and 2013 Seasons GLENN J. CORBETT, DONALD R. KELLER, BARBARA A. PORTER, AND CHRISTOPHER A. TUTTLE

seams that outcrop approximately 800 masl, whereas equally accessible seams at higher and lower elevations were not utilized. The northern border of the mining area is defined by gentler slopes with no visible flint outcrops, while the southern border features relatively steep slopes covered with reddish aeolian sediments that likely conceal such outcrops. Like the mines discovered in 2010, the newly identified mines are situated on promontories with steep escarpments. However, unlike the smaller outcrops and pit mines discovered in 2010, the recently documented mining districts feature large trench mines. With the new discoveries, the total flint mining and workshop area identified by the survey increased to approximately 38 ha. A workshop of that size would be capable of producing nearly 2 million cortical tool blanks. In 2013, the survey focused on wadi terraces in Wadi Rajjil in the Jawa region, where earlier survey work had identified the remains of terraced wadi gardens. While the artificially irrigated terrace gardens consisting of low walls, deflection dams, and channels could date to the fourth or early third millennium B.C.E., our preliminary observations suggest that they date to the Late Roman/Byzantine and Early Islamic periods. During the 2013 season, excavators also inspected and documented two newly discovered Chalcolithic/ Early Bronze Age settlements in the harra desert east of Jawa. The first, Khirbat Abu al-Husayn, is a hillfortlike settlement strategically positioned atop a small volcanic peak near the eastern border of the harra. The site includes massive double-faced fortification walls, small gates, and two possible massive tower structures. The few surface finds suggest a Chalcolithic/Early Bronze Age date. The second, Tulul al-Ghusayn, is a

The 2014 edition of the “Archaeology in Jordan” newsletter presents short reports on selected archaeological projects conducted during 2012 and 2013 in the Hashemite Kingdom of Jordan. Reports are generally organized geographically, starting with the eastern panhandle and then moving from north to south (fig. 1). Bibliographic references and footnotes have been omitted to make space for more entries; please contact the authors directly for more information. The authors’ email addresses are listed at the end of the report.*

northern badia Bernd Müller-Neuhof, Orient Department, Deutsches Archäologisches Institut, reports: Since 2010, the “Arid Habitats in the Fifth to the Early Third Millennium B.C.E.” project has focused on identifying Chalcolithic and Early Bronze Age socioeconomic activities in the northern Badia of northeastern Jordan, which is defined by the harra basalt desert in the west and the hamad limestone desert in the east. Following four surveys in 2010 and 2011, two new surveys were carried out in 2012 and 2013. The first was a continuation of the survey of the Chalcolithic/ Early Bronze Age flint mines on the western escarpment of the al-Risha Plateau (Wadi Ruweishid region); the second was in the Jawa region, especially around two newly discovered harra settlements east of Jawa. Additional flint mines located in the southern part of the Wadi Ruweishid region were investigated. The extent of the entire mining region was defined and can be differentiated into three districts located on the western escarpment of the al-Risha Plateau. Intentional mining activity seems to have been restricted to flint

the exception of codirected joint projects. Publication of this year’s newsletter was once again made possible in part by a contribution from the American Center of Oriental Research. All figures are courtesy of the individual project directors unless otherwise noted. Additional figures can be found under this article’s abstract on the AJA website (www.ajaonline.org).

* We would like to thank the staff of the Department of Antiquities (DOA) of Jordan and the director general, Monther Jamhawi, who took up his position in August 2013, for their continued efforts on behalf of all those committed to understanding and preserving the cultural heritage of Jordan. Each project gratefully acknowledges the role of the DOA, but for the sake of brevity the statements have been removed with

American Journal of Archaeology 118 (2014) 627–76

627 © 2014 Archaeological Institute of America

628

G.J. CORBETT ET AL.

[AJA 118

Fig. 1. Sites and projects in Jordan in the 2012 and 2013 seasons (drawing by T. Paradise).

Chalcolithic/Early Bronze Age site located on a volcanic peak east of Jawa and approximately 20 km west of the eastern border of the harra. It was first discovered from the air by D. Kennedy and R. Bewley in 2011. It is a fortified settlement that extends along the rim of the volcanic crater, and numerous small dwellings are situated both within the crater and along the southern slope of the volcano (fig. 2). Visible inside the crater and along the southern slope are the remains of terraced gardens showing clear similarities to the Chalcolithic/Early Bronze Age terraced gardens discovered in the vicinity of Jawa. At Tulul al-Ghusayn, there is no evidence for artificial irrigation. The few surface finds here are mostly lithic artifacts, some pottery, and grinding stones, suggesting a Chalcolithic/Early Bronze Age date. Future survey and excavation work will focus especially on obtaining carbonized material

© 2014 Archaeological Institute of America

from the newly identified settlements that might be able to provide better dating evidence.

jabal qurma archaeological landscape project Peter M.M.G. Akkermans, Leiden University, reports: The Jabal Qurma Archaeological Landscape Project is a new research initiative in the northeastern Badia, conducted under the auspices of Leiden University and in close cooperation with the Department of Antiquities of Jordan. This survey and excavation project is focused on the highly arid and barren basalt region of Jabal Qurma, some 30 km east of Azraq, near the Saudi border. It seeks to examine local settlement patterns, subsistence practices, social strategies, and the treatment of the dead from a diachronic perspective.

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

629

Fig. 2. Aerial view of Tulul al-Ghusayn. The fortified settlement perched on the rim of the volcanic crater is visible in the center of the image; the dwellings and terraced garden structures are visible both within the crater itself (right side of the image) and on the southern slope of the volcano (lower half of the image).

Survey seasons were conducted in 2012 and 2013, while an excavation season took place in 2014. The study area covers about 300 km2 bounded by Wadi Rajjil in the west and Wadi al-Qattafi in the east. Situated close to Wadi Rajjil is the prominent landmark of Jabal Qurma, a steep-sided, basalt-covered promontory at the very edge of the rugged basalt massif that marks the beginning of the harra desert in this region. Extensive gravel plains extend to the north and south, alternating with mudflats of varying size and low ranges of limestone hillocks. Still farther to the south is the Wadi Sirhan, one of several major caravan tracks connecting this region with Arabia. Detailed study of aerial photographs and CORONA, IKONOS, and TerraSar-X satellite imagery revealed that this desolate, basalt-strewn land is rich with various archaeological features, including occupation sites, hunting installations, animal pens, and burial monuments. Our subsequent intensive pedestrian surveys in

1

See www.ajaonline.org.

© 2014 Archaeological Institute of America

2012 and 2013 identified many more sites, especially small and ephemeral features that were invisible from the air. Altogether, many hundreds of stone-built dwellings and burial cairns have been documented, as well as thousands of rock carvings and Safaitic and Arabic inscriptions (online fig. 1 on the AJA website).1 Stray flint tools, mostly dated to the Middle Paleolithic, were often found on the lower slopes of the Jabal Qurma range. Small numbers of Epipaleolithic sites characterized by small circular enclosures were found near both the lower wadi courses and the higher elevations of the basalt slopes. Neolithic occupation in the region seems to have begun with the construction of several large desert kites in the early seventh millennium B.C.E., whereas a second phase of Neolithic habitation started ca. 6400 B.C.E. with the building of large clustered groups of walled enclosures that together form irregular, honeycomb-like structures measuring more than 2,000 m2. These grouped enclosures

630

G.J. CORBETT ET AL.

often have dense lithic scatters, including many concave truncation burins, and therefore closely resemble other “burin Neolithic” sites known across the Badia. Another major phase of Late Neolithic settlement, dated to the sixth millennium B.C.E., is characterized by sites up to 12 ha in extent and comprising hundreds of freestanding, stone-walled structures of various form and size. These rank among the largest prehistoric sites known in Jordan, and they may have been (semi-)permanent villages with populations of perhaps several hundred. These substantial prehistoric sites belie any simplistic notion that local environmental constraints limited cultural development in this region. There may have been yet another Late Neolithic (or perhaps Chalcolithic) habitation phase in the area associated with the so-called wheel structures, roughly circular stone features (up to 70 m in diameter) with extensive enclosures surrounded by an outer ring of small “huts.” These wheels yielded what appear to be the earliest ceramics identified in the Jabal Qurma area, handmade holemouth pots and small jars, some with loop or ledge handles. A few sherds were painted black, while one fragment was both painted and incised. There is at present no evidence for settlement in the Jabal Qurma region from the Bronze Age to the Iron Age; the area seems to have been abandoned for several thousand years. Local habitation increased significantly, however, between the first century B.C.E. and the fourth century C.E., when large numbers of settlements, burials, and various installations were constructed and thousands of Safaitic inscriptions and rock carvings were made. The region appears to have subsequently been abandoned until the late 19th century, except during the Mamluk period (14th century), when a series of campsites were built and some Arabic stone inscriptions were carved.

wisad pools Gary O. Rollefson, Whitman College, Yorke M. Rowan, The University of Chicago, Alex Wasse, University of East Anglia, Morag M. Kersel, DePaul University, and Austin C. Hill, University of Connecticut, report: The goal of the 2013 season at Wisad Pools, a large late prehistoric site in Jordan’s eastern basalt desert, was to excavate a small circular tower tomb (W-85) that is surrounded by a low wall and 13 small stone piles that extend in a chain toward the west. Unfortunately, in the two years since the 2011 season, looters had destroyed W-85, so attention was turned to another tomb (W-80) situated on an elevated mound about 80 m to the west. The tomb consists of a circular wall five to six courses high and encloses an open chamber approximately 5 m

© 2014 Archaeological Institute of America

[AJA 118

in diameter, which was evidently never roofed. The body was placed inside the enclosure and then covered with medium-sized basalt boulders to a height of about 1 m. There was only enough time to excavate the southern half of the tomb. The body inside the tomb had been badly disturbed, possibly by burrowing animals; only a small portion of the skeleton was excavated in 2013, but the remainder was recovered in 2014. Investigation of the mound on which the tomb was placed revealed that it was a collapsed Late Neolithic house (House W-80), one that turned out to be more complex than the one excavated in 2011. As was the case in 2011, the roof of the original house had collapsed, but the structure was still used after it ceased to function as a residence, since the standing walls offered a protective windbreak during episodes of butchering, tool manufacture, and bead production. The floor in the southern half of the original house was not reached in 2013. The postabandonment layers were rich in faunal remains. Gazelle and hare were dominant, although bird species were also quite abundant and varied; goat and sheep were present although relatively rare, probably making up less than 10–15% of the total. Carnivores were represented by numerous foxes, other canids (including domestic dog), large and small felids, and hyenas. Small mammals and reptiles were also part of the inventory. The hunting of wild animals is supported by the presence of more than 300 projectile points, including transverse arrowheads (more than 85% of the total number of projectile points) as well as Haparsa, Nizzanim, and Herzliya types (online fig. 2). Scrapers, notches, and denticulates accounted for more than a third of the stone tools, and cortical knives and cortical scrapers occurred in appreciable numbers. Borers and drills made up more than 20% of the assemblage and matched the importance of stone, shell, and bone beads, which numbered more than 50 pieces. Basalt groundstone artifacts included nine grinding slabs and 35 handstones; pestles were few but included a cache of three large specimens ranging in weight from 3 to 7 kg. While there may have been local vegetation that was processed, the amount of groundstone material may also indicate that opportunistic planting of cereals was practiced during years of sufficient rainfall. House W-80 displays a complexity that is rare in Late Neolithic arid land dwellings (fig. 3). West of the house is a courtyard offset by basalt slabs set on edge. At the southeastern corner of the courtyard is a smaller activity area (the “porch”), at the center of which is a large grinding slab with a central depression; this might represent an outside work area, although it could also reflect the position of a tent pole after the

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

631

Fig. 3. The half-excavated House W-80 at Wisad Pools, showing the alcove (A), bench or platform (B), doorway (D), “porch” (P), and western forecourt (WF).

house was abandoned. A doorway leads from the activity area into what might be a vestibule. To the south is a small alcove occasionally paved with flat basalt slabs alternating with surfaces of light-colored sediment, perhaps a mixture of gypsum plaster and silt. The main room to the west measures approximately 2.8 m from the southern wall to the unexcavated section to the north. A paved bench or platform constructed in postabandonment times is located along the southern wall of the room. Three radiocarbon dates were obtained from charcoal samples from the main room: cal B.C.E. 6590–6580 (2σ), 6000–5840 (2σ), and 5710–5610 (2σ). Excavation of House W-80 continued in 2014. The work at Wisad Pools is revealing a landscape that was much less xeric than the forbidding vistas in today’s basalt desert. It is possible that from 8500 to 4000 BP the area was a dry grassland with a topsoil that absorbed seasonal rainwater, supporting a more luxurious and longer-lasting vegetation cover for both wild and domesticated animals exploited by the region’s hunter-herders.

maitland’s mesa Yorke M. Rowan, The University of Chicago, Gary O. Rollefson, Whitman College, Alex Wasse, University

© 2014 Archaeological Institute of America

of East Anglia, Austin C. Hill, University of Connecticut, and Morag M. Kersel, DePaul University, report: The Eastern Badia Archaeological Project examines late prehistoric land and water use through two primary study areas in Jordan’s eastern desert, Wadi al-Qattafi and Wisad Pools, both located in the Badia, or Black Desert. One of about 30 basalt-capped mesas running north–south along Wadi al-Qattafi, Maitland’s Mesa (M-4) rises approximately 50 m above the surrounding wadi. Maitland’s Mesa stands out among these mesas because of the many structures on top and along the slopes (online fig. 3). We have identified 478 structures as well as additional evidence for flint tools and debitage dating to the Epipaleolithic, PrePottery Neolithic, Late Neolithic, and Chalcolithic/ Early Bronze Age. Most notable is a line of approximately 50 rectangular to oval cairns along the southern edge of the mesa, which are apparently linked to a large tower tomb on the east end. During the 2012 season, our first excavations at the site included two circular stone features on the summit; Structure 165 was a single-cell building (1.68 x 1.33 m) built of basalt cobbles with an entrance marked by two standing stones, while Structure 14 was a slightly larger, double-cell construction. Despite

632

G.J. CORBETT ET AL.

thorough sieving, no diagnostic artifacts to clarify the dating were recovered. On the southern slope, where we identified 26 structures, we excavated one of the large dry-masonry circular structures. Although large basalt slabs had collapsed, House 11 (SS-11) retained a doorway and some intact walls. Two building phases were identified. First, the building interior (2 x 3 m) was built on level ground; a roof was probably supported by a long basalt slab found in the interior fill. Unlike the exterior, the interior wall construction included large slabs placed vertically on edge, with smaller stones inserted in between for stability. Two doorways on opposite sides were framed by vertical slabs and topped by lintels less than 1 m high. Dark ash from an interior hearth provided a single calibrated date between cal B.C.E. 5480 and 5320 (2σ). In the second phase, the oval interior was altered; after leveling, the floor was carefully paved with flat basalt blocks (30–40 x 50 cm), and small basalt chips and stones were wedged in between the slabs. This paving raised the floor level significantly. The gap in the center of the paving may indicate where a pillar was set as a roof support. Preservation of exterior features was unanticipated. A courtyard on the east included a small, ashy, stone-lined fireplace similar to one found at Wisad Pools House 80. A substantial ashy layer was found to the southeast of the courtyard entrance, below an enclosure wall. On the southern side of the wall, built against the structure’s exterior, a low roofed room (ca. 1.9 x 1.5 m) with a single pillar about 45 cm high was probably a storage area. Another storage bin was built on the northern side, next to the courtyard entrance, and both had apparent openings to the courtyard. Our initial assumption that these were burial structures possibly dating to the Chalcolithic/Early Bronze Age was proven incorrect, at least in the example of House 11. The radiocarbon date is supported by Late Neolithic artifacts, including burin spalls, a drill on a burin spall, a Haparsa point, and a Yarmouk point. In addition, tabular knives, bifacial and unifacial knives, scrapers, denticulates, and notches were recovered. Little grinding equipment was found, in contrast to the many grinding slabs, handstones, and pestles found at Wisad Pools. Coupled with the results from Wisad Pools, the results from Maitland’s Mesa have led us to reconsider the nature of buildings on these slopes and the slopes of other mesas along Wadi al-Qattafi. If House 11 is typical, we must reassess our earlier assumptions that the buildings served as burial chambers. These and apparently similar collapsed basaltic structures indicate greater investment and semipermanent occupation in this arid zone. Perhaps the substantial nature

© 2014 Archaeological Institute of America

[AJA 118

of these buildings reflects anticipated seasonal visits to the same locations, allowing greater population clusters during the Late Neolithic than previously recognized for the area. Future research will investigate similar structures, try to date other structural types, and carry out geomorphological investigations to determine whether a wetter climate with better soil conditions once existed.

azraq: ‘ayn sawda reservoir Lorraine Abu-Azizeh, Institut Français du ProcheOrient (Amman), and Claude Vibert-Guigue, Centre National de la Recherche Scientifique, École Normale Supérieure, report: Located in the Azraq Wetland Reserve, the ‘Ayn Sawda Reservoir is of both ecological and archaeological interest. In 1983, when intensive pumping began to lower the wetland’s water level, the Department of Antiquities of Jordan discovered 36 carved basalt stones associated with the so-called Umayyad Reservoir. It was in this context that the ‘Ayn Sawda Reservoir research program began. In 2002, the ‘Ayn Sawda spring dried up completely, allowing a Jordanian-French team to survey the former wetland. Since then, five survey seasons led by C. Vibert-Guigue have identified, studied, and preserved many more of the reservoir’s carved stones, most found in their original positions. Of the 106 carved blocks, almost all have been found on the west side of a large buttressed platform (30.5 x 6.0 m) that is part of the monumental reservoir. Previous researchers suggested this structure could have been built as a viewing platform, perhaps related to nautical games held in the reservoir, or that it could have been used for fishing, hunting, or hosting banquets and celebrations. Our surveys have tended to confirm these suggestions. We began a new survey phase in 2013 with the aim of understanding the remains from an architectural perspective. In addition to conducting a general archaeological survey, we completed a topographic survey of the reservoir to produce a comprehensive plan and, for the first time, have elevation measurements for every part of the reservoir. We also began assessing the status and preservation of the site’s archaeological structures, both inside and outside the reserve. While many parts of the reservoir have already been conserved, the gradual drying out of the former wetland soil continues to degrade the masonry of other features.

the azraq marshes archaeological and paleoecological project April Nowell, University of Victoria, Carlos E. Cordova, Oklahoma State University, James Pokines,

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

Boston University, and Christopher J.H. Ames, University of California, Berkeley, report: The Azraq Marshes Archaeological and Paleoecological Project (AMAPP) seeks to recover data from stratified sites in the former wetlands of the Azraq Oasis, particularly the Druze and Shishan Marshes. The stratigraphic record of sites here consists of marsh, lacustrine, shoreline, and aeolian sediments in association with Lower, Middle, Upper, and Epipaleolithic occupations and thus is an important resource for studying hominin occupation and subsistence in the Jordanian desert through the Middle to Late Pleistocene glacial and interglacial periods. Because of its location between the Levant and the Arabian Peninsula, the Azraq Oasis may have been one of the desert refugia visited by modern humans as they spread into southwest Asia. No sites were previously reported around the Druze and Shishan Marshes because these areas were still wetlands. In the late 1980s, when water was pumped to satisfy urban demand, these areas dried out. Jordan’s Royal Society for the Conservation of Nature (RSCN) preserved 10% of the former Shishan Marsh as the Azraq Wetland Reserve (online fig. 4), but the Druze Marsh was never restored. The drying of the marshes led the local population to dig wells and dredge the area in attempts to catch up with the falling water table. This exposed the underlying sediments and paleosols, as well as evidence of stratified Paleolithic sites, often with abundant faunal remains and lithics. The sites in the Shishan Marsh were studied in the 1990s and those in the Druze Marsh more recently. AMAPP initiated excavations at Druze Marsh in 2008; in 2013, the project expanded to include a survey of sites in the areas surrounding the Azraq Marshes, including the harra basalt plateau and Wadi Rattam, and excavations of sites in the Shishan Marsh. Our primary goal is to investigate whether the Druze and Shishan Marshes functioned as a desert refugium for hominins in the Levantine interior during the Late Pleistocene, and if so, when and in what ways. Out of this primary research goal emerge numerous avenues of inquiry regarding local environmental conditions in Azraq during Middle to Late Pleistocene glacial and interglacial cycles and how these changing conditions may have affected hominin survivorship. In 2013, we excavated four units at site SM1, conducted a study of the stratigraphic profile of a second location (site SM2), and excavated two test pits. At site SM1, we uncovered two cultural layers—layers 7b and 8. Based on associations with other sites in the area, layer 7b is likely approximately 200,000 years old, while layer 8 is older. We are currently awaiting optically stimulated luminescence (OSL) results to

© 2014 Archaeological Institute of America

633

refine our estimates, but they are supported by the archaeological materials recovered from each of these layers. Layers 7b and 8 have been tentatively assigned to the Lower Paleolithic. The lithics are in remarkably pristine condition—most artifacts are characterized by fresh edges with only minimal damage, which suggests they originate largely from in situ deposits. Some artifacts exhibit thermal alteration, and we are in the process of determining whether this is due to exposure to hearths or flash burns. An unusual component of this assemblage is the large number of borers (perçoires) and scrapers on small (2–3 cm) pebbles. The results of use-wear and residue analyses are expected soon. Fragmentary faunal remains recovered from layers 7b and 8 attest to the presence of a variety of animals—including camels, aurochs, horses, and rhinoceroses—indicative of a desert steppe environment. Pollen, phytolith, ostracod, diatom, and soil micromorphological analyses are currently underway, the results of which will provide a clearer picture of the environmental conditions experienced by Lower and Middle Paleolithic hominins.

quseir ‘amra project Gaetano Palumbo, World Monuments Fund, and Giovanna De Palma, Istituto Superiore per la Conservazione ed il Restauro, report: Quseir ‘Amra, an Umayyad “desert castle” and, since 1985, a World Heritage site, is located in the eastern steppe 85 km east of Amman. The site extends over an area of approximately 2 km2 and includes a variety of archaeological remains. The most renowned is the bath complex featuring the largest extant Early Islamic figurative paintings. The Quseir ‘Amra Project began in 2010 as a joint effort of the Department of Antiquities of Jordan, the World Monuments Fund, and the Istituto Superiore per la Conservazione ed il Restauro in Rome. The project aims to integrate conservation efforts at Quseir ‘Amra through the combination of conservation interventions, archaeological investigations, sitemanagement planning, and tourism development. We report here on key work executed in 2012 and 2013. Conservation interventions focused on the main monument building and the adjacent saqiya (a well with a water-lifting device). Focus inside the building was primarily on the western aisle and included cleaning and consolidation of its walls, vault, and paintings. The condition of the paintings is poor because of a combination of factors, including graffiti, soot, dirt, and animal deposits. The progressive aging of the shellac applied as a consolidant in the 1970s also had to be addressed. Cleaning involved progressively removing the shellac, the subsequent layers of repainting, and

634

G.J. CORBETT ET AL.

accumulated soot. Gaps in the preparatory layers were plastered and leveled. The current intervention brought to light new, remarkable scenes and iconographic details within the paintings (fig. 4). On the western vault, previously unknown images were revealed, including a three-headed dog (possibly a Cerberus) and two standing figures— one (possibly a Dionysos) holds a basket of grapes and is flanked by a panther. On the south wall, an Arabic inscription was found bearing the name of Walid ibn Yazid, one of the last Umayyad caliphs (743–744 C.E.). The inscription, along with a radiocarbon date from an ash deposit in the caldarium (cal C.E. 750–710 [1σ]), allows us to date the building to the reign of Hisham (723–743 C.E.), when Walid was crown prince. Cleaning and consolidation work was also conducted on the eastern aisle’s northern wall and vault, as well as inside the bath area—specifically, the hypocausts of the tepidarium and caldarium and the lower walls of the apodyterium. In the apodyterium, conservation work included the removal of concrete that had been applied over masonry during previous conservation campaigns. This intervention, financed by the U.S. Ambassadors Fund for Cultural Preservation, allowed the paintings decorating the vault and the lunettes of the apodyterium to be cleaned in the spring of 2014. Additional conservation efforts included the cleaning and consolidation of the main building’s external walls and vaults, as well as the removal of a recent cistern (built in 1964) from the archaeological remains of the praefurnium. Archaeological investigations and soundings targeted the praefurnium, the saqiya, and two alcove rooms, as well as a new feature—perhaps a service building used during the construction of the bath complex— identified 50 m south of the visitors’ center. Archaeological surveys conducted within an approximately 1 km radius around Quseir ‘Amra provided evidence of human occupation dating back to the Paleolithic period. As a key component of the sustainable conservation and development of the site, the project began developing a long-term site-management plan. Other components of the project included an extensive documentation survey that incorporated topographic mapping and three-dimensional laser scanning of the site. As part of a recent tourism survey, more than 800 visitors were interviewed about tourist expectations and demands. The survey results represent a unique, field-based study for the strategic development of a valuable tourism product in the Badia region.

kharaneh iv excavation project Lisa Maher, University of California, Berkeley, and Danielle A. Macdonald, Université Nice Sophia Antipolis, report: © 2014 Archaeological Institute of America

[AJA 118

In 2013, the Epipaleolithic Foragers in Azraq Project (EFAP) conducted excavations at Kharaneh IV, which is located approximately 1 km south of Qasr Kharaneh. The archaeological sequence spans the Early and Middle Epipaleolithic (ca. 20–16 ka BP) and contains dense and deep deposits of lithic and faunal material. It covers an area of approximately 21,000 m2, making it one of the largest Epipaleolithic sites in the Levant. The site was originally surveyed in the 1970s, and excavations in the 1980s identified a complex depositional sequence of hearths, postholes, and occupation surfaces dating to the Middle Epipaleolithic, in addition to two Early Epipaleolithic burials. Since 2007, EFAP has sought to explore further the earlier-excavated deposits, extend excavations to new areas of the large prehistoric mound, and investigate the use of the site within the larger paleoenvironmental context of the Azraq Basin. EFAP excavations in 2008–2010 began reconstructing the nature of the Late Pleistocene occupation. Two hut structures were discovered and mapped in 2010 but not excavated. Hut structures are extremely rare for this period, and the discovery of these features represents a unique opportunity to explore the organization of hunter-gatherer activities at this aggregation site. The project team excavated one hut feature in 2013 to try to understand the nature of occupation and the activities conducted within and outside the structure (online fig. 5). Structure 1 contained several overlying in situ deposits. The upper deposits contained fine orange sand, under which lay three caches of pierced marine shell and red ochre. The sediment beneath was burned and contained charcoal, burned lithics, faunal remains, perforated marine shells, and an articulated fox paw. These deposits may be the remains of the hut’s superstructure, which was burned after abandonment; however, the shell and ochre caches in the upper hut layers have no evidence of thermal alteration, which suggests they were placed there afterward. Underlying the burned deposits was a compact, light-colored, clayey sediment representing either fill that accumulated within the structure after abandonment (but prior to burning) or a floor deposit. On the surface of this deposit are large artifacts including lithics, fauna, perforated shells, and worked bone, as well as clusters of rocks. Beneath this deposit were two more compact, artifact-rich surfaces. The compact nature of these deposits and the flat-lying artifacts suggest that the hut may have had a sequence of overlying floor deposits. Dense cultural material is present throughout the Structure 1 deposits, suggesting that the hut was not cleaned prior to abandonment. The artifacts recovered from the structure include lithics characteristic of the Early Epipaleolithic, such as narrow-faced cores, endscrapers, and nongeometric microliths.

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

Fig. 4. A section of a mural painting on the west wall of the main monument building at Quseir ‘Amra, before (top) and after (bottom) conservation efforts.

© 2014 Archaeological Institute of America

635

636

G.J. CORBETT ET AL.

Faunal remains include gazelle, equid, tortoise, and aurochs. Other artifacts include pierced marine shell (from the Mediterranean and Red Seas), ochre, and worked-bone tools and suggest that the inhabitants participated in a variety of economic, technological, and social activities. Further analyses of the material culture and the use of space at this site will help illuminate how hunter-gatherer groups organized their built environment and structured tasks in large group scenarios.

qasr mushash survey Karin Bartl, Orient Department, Deutsches Archäologisches Institut, and Ghazi Bisheh, Department of Antiquities of Jordan, report: Qasr Mushash, commonly considered to be one of the Early Islamic “desert castles,” has been the focus since 2011 of a project with two major research objectives. One is the complete documentation of all visible architectural remains at the site of Qasr Mushash itself, and the other is the recording of all archaeological remains within a 10 km radius around the site. The overall goals of the study are to determine the site’s function in the Early Islamic period and to understand the history of settlement, water supply, and land use within the immediate vicinity. Qasr Mushash is located in the desert steppelands about 40 km east of Amman. It comprises two areas of settlement situated along two wadi courses located about 1.5 km apart. Qasr Mushash West is positioned directly beside the Wadi Mushash and consists of several quarters, including the fort (qasr) located at the eastern edge of the settlement (fig. 5). The central quarter forms a building complex that comprises several architectural units, including a 1,600 m2 building with a large courtyard, an adjacent bath, and an associated water reservoir. Small houses of one or two rooms lie to the northwest of this area. To the northeast are three large, multiroom buildings, and to the west lies another large water reservoir. Several long walls recorded along the Wadi Mushash apparently delimited a water-retention area associated with a now-destroyed barrage wall. Additional structures, some quite large, are found along a tributary wadi, including another very large reservoir with a capacity of approximately 2,000 m3. Whereas the settlement probably relied on the reservoirs located in the tributary wadis, the waters held behind the Wadi Mushash dam likely served to irrigate areas immediately south of the wadi. Qasr Mushash East is located next to a north–south running tributary of Wadi Mushash. This part of the settlement consists of a small square structure with a central courtyard, several adjacent buildings with either one or two rooms, and a small dam barring a

© 2014 Archaeological Institute of America

[AJA 118

narrow secondary wadi. More architectural remains are visible in the immediate and distant surroundings, including a massive wall blocking the main wadi that likely functioned to capture winter runoff. To the north of the relatively small settled area there are several cisterns cut into the bedrock and lined with water-resistant mortar. The collected surface pottery indicates that both settlement quarters were occupied during the Umayyad period. It is hoped that laboratory analysis of mortar samples collected from the cisterns, bath, and fort will not only furnish valuable information about the mortar’s composition but also provide new data concerning the dating. Furthermore, X-ray analysis on the bath’s marble covering will provide evidence regarding the origin of the stone.

umm el-jimal project Bert de Vries, Calvin College, reports: Preservation and presentation fieldwork at the complex known as House XVII–XVIII at Umm el-Jimal took place in 2012, enabled by a grant from the U.S. Ambassador’s Fund for Cultural Preservation. Following preliminary studies by Calvin College staff in January, the Umm el-Jimal Project (UJP) and the Department of Antiquities of Jordan carried out the actual preservation work, as well as several soundings, from April to June. In January 2013, the UJP also held a workshop that outlined plans for future project publications. Nicknamed “Beit Sheik,” Houses XVII and XVIII constitute a mixed-use complex from the Byzantine/ Umayyad period. Howard Crosby Butler, the Princeton University archaeologist who first mapped the site, named these structures on the premise that they were two distinct domestic buildings, and his interpretation prevailed until 2012, when project archaeologists remapping the structure discovered a doorway connecting both buildings and concluded that it was, at least in its latest phase, a single complex. Among the building’s features is the precariously perched double-arched window that has long been the icon of the UJP. Detailed mapping and photogrammetric documentation of the structure led to the discovery of new architectural features, including interconnecting doorways between the two buildings and six entries into the 400 m2 courtyard of House XVIII (fig. 6). In antiquity, a massive tower was located in the northwest corner, while cantilevered stairwells provided access to the building’s second and third stories. A large reservoir located under the eastern wall provided ample water for its occupants. Structural analysis has revealed the building methods and materials that were used, especially the types of plasters and mortars.

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

637

Fig. 5. Aerial view of Qasr Mushash (D. Kennedy; © Aerial Photographic Archive for Archaeology in the Middle East).

Work on the masonry, carried out by a talented crew of local Umm el-Jimal preservation archaeologists, has followed a minimalist strategy. Our goals were to make the complex understandable to visitors and also safe and passable for further exploration. We cleared the large courtyard of House XVIII and the eastern passage that links to the water reservoir. From the reservoir, we retrieved significant architectural components belonging to the upper levels of the eastern wall. In House XVIII, we preserved the eastern exit passage to the level of the ceiling corbels of the lower floor and stabilized the famous double windows, using two cranes and clever planning. In House XVII, we opened the main entry, a Byzantine stable, and access to the mangers in the courtyard, and we also repaired a collapsing roof that later Mamluk or Druze inhabitants had reinstalled over a second Byzantine stable. Our excavation below the layers of cleared debris has complemented our understanding of the construction and occupation sequences first analyzed by Robin Brown in 1977. It became clear that Houses XVII and XVIII were new Byzantine constructions founded on carefully leveled Roman building remains. Subsequent remodeling in the Umayyad period included the integration of the two buildings into a single large

2

See www.ummeljimal.org.

© 2014 Archaeological Institute of America

complex. Although some remodeling was done in the Mamluk/Ottoman period, including reuse of the water reservoir, it is difficult to distinguish this phase from that of the Late Ottoman/Mandate reoccupation of the site. Putting all this together in the context of the historic landscape gives us a picture of a major structure that over time became much more than just a pair of typical family dwellings. While final interpretation is ongoing, the data suggest that House XVII–XVIII began its centuries-long occupation as a pair of homes but was later integrated for public use, perhaps as an administrative center or possibly a hotel for caravans commuting through the region. Much of the project’s planning and presentation work was done in partnership with Open Hand Studios, the designer of the UJP website,2 where much more information on our work and the new local cooperatives dedicated to preserving heritage and antiquities is presented.

abila of the decapolis David Vila, John Brown University, reports: The 2012 excavation season at Abila of the Decapolis in northwest Jordan focused on two Byzantine

638

G.J. CORBETT ET AL.

[AJA 118

Fig. 6. Plan of the first floor of House XVII–XVIII complex, Umm el-Jimal (drawing by M. al-Bataineh).

churches (Areas E and G) and the Early Bronze Age occupation of the north tall (Area AA). Excavations in Area AA aim to uncover the settlement history of the northernmost tall, Tall Abil, from the Early Bronze Age through the Umayyad period. The upper portion of the tall features Hellenistic/ Roman layers and Byzantine architecture. In opening a square to the east of the upper tall, we hoped to avoid these time periods. Work in the square excavated much of the topsoil. Pottery was the expected mix of Byzantine/Umayyad forms. Excavation in Area AA’s other squares continued to work through the Iron Age and into Late Bronze Age levels. Excavations in the five-aisled church in Area E to the east of Tall Abil have helped us understand the structure’s phased development. Excavation of this structure began in 1990, and our work since has uncovered a church dated to the Umayyad period (seventh to eighth centuries C.E.). Probes in the north and south aisles during the 2010 season uncovered large segments of mosaic flooring in situ, approximately 1 m below the floor surface of the Upper Church. The mosaic contained several inscribed crosses, and an iconostasis was found in the north aisle, indicating that the mosaic belonged to an earlier church. Work in the 2012 season focused on the southwest corner of © 2014 Archaeological Institute of America

the upper structure, where a chapel had been uncovered. Two new squares were opened and several balks removed. An altar screen was located in situ along with a well-preserved opus sectile floor (fig. 7). One of the more important features of this Umayyad-period church is that it seems to have functioned as a place of worship for both Christians and Muslims before it was destroyed in a major earthquake in 749 C.E. Work in Area G—a single-apse church east of Tall Umm al-Amud (the southern tall)—focused on conservation/restoration and excavation. Previous seasons had uncovered beautiful opus sectile floors in the north and south aisles. Our current conservation and restoration work included removing earlier excavation balks and covering the excavated floors with sand. Our excavations opened two new squares, a full square at the east end of the diakonikon and a half square at the structure’s west end.

umm qeis/gadara and its urban hinterland Claudia Bührig, Orient Department, Deutsches Archäologisches Institut, reports: The hinterland survey of Umm Qeis/Gadara provides new insights into the transformation process of the city and its surroundings. The survey was con-

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

639

Fig. 7. Opus sectile floor and altar screen discovered during the 2012 excavations of the Area E church at Abila of the Decapolis.

ducted by the Orient Department of the Deutsches Archäologisches Institut, Berlin, in cooperation with the Archaeological Institute of the University of Hamburg and the Geomatics lab of HafenCity University, Hamburg. The survey area encompasses the Yarmouk Valley in the north, the village of al-Mansura in the east, the slopes of Wadi al-‘Arab in the south, and the high plateau of Ard al-Ala in the west. During the last two years, the project has focused on exploring the hinterland and its relation to the ancient city, as well as regional settlement and climatic developments since prehistory. Some initial findings regarding the region’s climatic and environmental history have resulted from dendrochronological, archaeobotanical, and pedological analyses. The project has also placed special emphasis on addressing questions related to the preservation, restoration, and presentation of the ancient remains. The survey has provided evidence of Paleolithic, Neolithic, and Early Islamic settlement and identified two large extra-urban sanctuaries from the Late Augustan/Early Roman era. We have also documented agricultural installations, farmsteads, a villa, small settlements, tombs, and a cemetery, as well as numerous watchtowers. During the 2012 season, we mapped, among other features, a large necropolis of © 2014 Archaeological Institute of America

shaft graves on a hilltop south of Gadara and a small settlement with many water installations overlooking the Wadi Mintanra. Finds from this remarkable settlement include fine wares (i.e., fragments of unguentaria, relief-decorated beakers, and fish plates) and numerous buildings exposed by illicit digging. The survey has recorded the remains of ancient agriculture, sophisticated irrigation systems, and limestone quarries spreading across an area of 5 ha. We investigated one of the quarries to understand better how the area’s geological resources were developed and how quarried materials were transported. In 2012, we made an initial tachymetric and photogrammetric record of the quarry with our colleagues from HafenCity University, Hamburg. The survey area is part of the Yarmouk Nature Reserve, opened by the RSCN in 2012, and the findings from our ongoing research are now being integrated with the development of the reserve’s attractions and guided tours. This groundbreaking collaboration brings together nature conservancy and archaeology in the attempt to convey different aspects of the region’s natural and cultural landscape to visitors (both locals and tourists). Training workshops on these initiatives were held in Umm Qeis in 2012 and 2013, and we hope to encourage the participation of local

640

G.J. CORBETT ET AL.

schools as important partners in the sustainable development of cultural tourism.

tall zira‘a/wadi al-‘arab Jutta Häser, German Protestant Institute of Archaeology, and Dieter Vieweger, German Protestant Institute of Archaeology and Biblical Archaeological Institute Wuppertal, report: The Gadara Region Project has conducted surveys in the Wadi al-‘Arab since 2001 and excavations at Tall Zira‘a since 2003. The focus of the team specialists during the 2012 and 2013 seasons was on preparing the final publication. The work included analyzing all material culture remains, conducting experiments on early pottery and glass production, and completing the threedimensional modeling of Tall Zira‘a and its excavated artifacts. The final publication is expected in 2015.

northern jordan project Bethany J. Walker, University of Bonn, reports: The Northern Jordan Project (NJP) conducted a multidisciplinary survey in the villages of Shajrah and Kharja in 2012 as part of an investigation of the Upper Wadi Shellaleh and its drainage system. The project was launched in 2003 with the initial objective of better understanding the factors behind the settlement fluctuations of the Middle and Late Islamic periods in the well-watered region between Irbid and the Yarmouk River. Each season a different village has been the focus of fieldwork, which is combined with archival, ethnographic, architectural, and environmental analyses. The team consisted of faculty and students from Missouri State University (the primary financial sponsor), Yarmouk University, University of Bonn, and Friedrich-Alexander-Universität Erlangen-Nürnberg, as well as a geomorphologist, phytolith specialist, and palynologist. The villages of Shajrah and Kharja are located some 15 and 12 km north of Irbid, respectively, facing each other on hilltops across the Upper Wadi Shellaleh, near the Syrian border. This region was chosen to complete a geographic transect of the northern highlands stretching from Umm Qeis in the west to al-Turra in the east. Four simultaneous surveys in two villages and their hinterlands were undertaken. Two survey teams were devoted to a pedestrian archaeological survey (with a goal of 10–25% coverage), which investigated different zones of each village’s lands and off-site locations (online fig. 6). Lands accessible this time of year (olive groves, fallow land, rocky outcrops) were sample surveyed, with the aim of targeting each distinctive land-use, environmental, and topographical zone. An environmental team of soil and plant scientists constituted the third survey team, employing vertical and horizontal sampling strategies to obtain culturally © 2014 Archaeological Institute of America

[AJA 118

and environmentally relevant soil samples (for pollen, phytolith, and geomorphological analyses) and those amenable to magnetic dating. Sampling locations were selected, in part, on the basis of aerial photographs, and they were scrutinized for anomalies potentially indicating ancient field lines, terraces, waterworks, and settlement traces. The fourth team included the project ethnographer and architectural drafter, and they devoted their time to studying the “living” villages, interviewing local residents, and documenting historical buildings through architectural drawings, three-dimensional reconstructions, and photography. The preliminary results of the combined surveys suggest compelling patterns of settlement and land use in the Islamic period. A section of the western watershed of the Upper Wadi Shellaleh was likely abandoned for the opposite bank of the wadi by the Abbasid period, in a settlement shift possibly related to differences in the drainage systems and soil qualities of the two watersheds and changes in land use. The settlement clusters that then developed on the east bank of the Upper Wadi Shellaleh during the Early and Middle Islamic periods were characterized by large villages and their satellites and gave way to a more dispersed pattern of settlement and land use after the 14th century C.E., similar to that identified in other regions of Jordan. Soil properties and sherd collection also suggested that the environs of the Byzantine necropolis of Khirbat Majid on the east bank subsequently hosted settlement in the Umayyad and Abbasid periods. Fields associated with all these sites, many naturally terraced, were watered by a complex runoff-irrigation system that made use of springs, natural caves (for water collection), tunnels and channels, and aqueducts and qanats. The archaeological survey identified a previously undocumented site in the lands to the north of Kharja, which are known locally as ‘Ayn Ghazal. The site includes Roman-era tombs of columbaria form and a large-scale water storage and harvesting system. The survey also recorded Byzantine-era architecture in the northern and westernmost portions of Tall ‘Abduh and identified a larger range of Islamic-era wares than was previously known in the area.

tall abu al-kharaz Peter M. Fischer, University of Gothenburg, Sweden, and Teresa Bürge, Oriental Institute of the University of Vienna, report: Excavation at Tall Abu al-Kharaz in the central Jordan Valley in 2012–2013 focused on exposing the eastern portion of the unique Iron Age I compound on the southern part of the tall (Area 9) and investigating the area east of it (Area 10). A further aim was to explore the eastern edge of the upper plateau of the tall by means of several test trenches (Area 11).

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

In Areas 9 and 10, phase X (Iron Age IB/IIA, 930–850 B.C.E.), which lies below the structures from phase XI (Iron Age IIA), contained 10 more rooms of a compound. The layout resembles that of phase IX (Iron Age I) because of the reuse of the older walls. Finds from this building include a “Philistine-style” jug, storage jars, a cooking jug, a spool-shaped loomweight, and a bronze arrowhead. Unfired ceramic material from several rooms suggests pottery production. Two skeletons were found in one of the phase X rooms: a male about 30 years old and a female about 25–30 years old. Both were buried when the building collapsed. Seven more rooms of the 46 m long, two-story, cellplan compound from phase IX were exposed. The total number of rooms is 21, of which 18 are arranged in pairs. The remaining three are single rooms because of the superficial bedrock, which rises to the north. The finds include more than 200 intact or complete ceramic vessels; metal, bone, and stone objects; textileproduction tools; several exquisite alabaster vessels; beads; a scarab; and a scaraboid. Vessel contents include wheat, barley and barley flour, millet, lentils, chickpeas, olive pits, and the dried remains of olive oil. Several clay ovens (tananeer) were found. Together with the standard Canaanite open-shaped cooking pot, a new type appears: the closed-shape cooking jugs with rounded bases and two handles that resemble contemporary types from sites in the Aegean, Cyprus, and Philistia. There are imported objects that reflect influences from all over the eastern Mediterranean and indicate this was a fairly wealthy society with extensive contacts. We suggest contacts with Philistia, Cyprus, and the Aegean through offshoots of the Sea Peoples. East of the phase IX compound is a defense system that was originally built in Early Bronze (EB) IB and II (local phases IB and II—i.e., ca. 3100 B.C.E.) but that was reused as a glacis as part of the Iron Age I defenses. The builders of the phase IX compound cut through the entire Early Bronze Age defense system and deposited the foundation walls of their structures on the earlier defenses. At the southeasternmost corner of the upper plateau of the tall, the glacis turns toward the north, leaving a 2 m wide gap, presumably where one of the city gates once stood. The gate faces Wadi al-Yabis, the site’s most important water source. In phase IB/II, an intramural burial of an infant, facing northwest, was unearthed next to the EB II city wall. The estimated age of the infant is 6–8 years old. The only burial gift was a Grain-Wash Ware jar. New test trenches in Area 11 revealed remains from the Late Roman and Early Islamic periods, including three Late Roman skeletons: an approximately 60-year-old female and two infants 3–5 years old. In this area, parts of another four-room house from phase XIV (Iron Age IIB) were exposed below the © 2014 Archaeological Institute of America

641

Late Roman burials. Finds include complete ceramic containers (juglets, a cooking pot, and a holemouth jar), spindlewhorls, loomweights, and bone shuttles. An extraordinary find is an intact limestone cosmetic palette (fig. 8). The remains from phase XIII (Iron Age IIB) comprise an unusual bichrome decorated juglet and a pierced limestone stamp with an incised sign. In the northernmost part of Area 11, a domestic context from phase V (Late Bronze [LB] IA) appeared below the Late Iron Age phases, and the finds included a Chocolate-on-White Ware I juglet, a frying pan, a cooking pot, a storage jar, and a glass bead.

tall abu sarbut Margreet Steiner, Noor Mulder-Hymans, and Jeannette Boertien, University of Groningen, report: Tall Abu Sarbut is located in the central Jordan Valley, approximately 3 km west of Tall Deir Alla. It measures about 250 m east–west, 125 m north–south, and only 7 m high and is surrounded by agricultural fields. It features two “summits” (at the eastern and western sides) separated by a “valley.” Between 1988 and 1992, a team from Leiden University, under the direction of E. LaGro and M. Steiner, conducted four seasons of excavation. As the aim of the project was to excavate a rural site from the Islamic period, only the upper layers of the tall were exposed, and Islamic remains were found mainly on the western part. The amount of material excavated in four short seasons was astounding. More than 150,000 pottery sherds were recovered, as well as 60 complete or nearly complete pots, approximately 1,850 glass fragments (including 400 glass bangles), botanical remains, animal bones, iron and bronze objects, coins, stone pestles, and several ostraca. In one of the test trenches, remains from the Roman period were found, including three rooms of a heavily burned mudbrick building. The building’s preservation was exceptional, with walls still standing to a height of 1 m and rooms filled with objects. The pottery consisted of ordinary household wares dating to the Late Roman period (third to fourth centuries C.E.). In addition, sherds of Roman luxury wares (red-slipped wares and bowls with applied relief) were found on the east side of the tall. In 2012, a new archaeological campaign was launched, in cooperation with the University of Groningen, by N. Mulder-Hymans, M. Steiner, and J. Boertien. The goals of the renewed excavations are to (1) establish the layout and context of the Roman-period burned building; (2) conduct spatial and functional analyses of the finds; and (3) excavate eastern portions of the tall, where a longer sequence of occupation from the Roman to Byzantine periods is expected. Square F is located atop the eastern part of the tall. On the west side of the square, two mudbrick walls

642

G.J. CORBETT ET AL.

[AJA 118

Fig. 8. Cosmetic palette from four-room house in phase XIV of Area 11 at Tall Abu al-Kharaz.

Fig. 9. Late Roman cooking pot from square H at Tall Abu Sarbut, as restored by the ACOR Conservation Cooperative.

(80 cm wide) running east–west were associated with two floors made of cemented sherds and covered with pottery. On top of the floor, several nearly complete vessels were excavated, including an Early Islamic jar decorated with painted white lines and swirls. In addition, a trench excavated on the south side of square F revealed several mudbrick walls and three consecutive floor layers containing Late Roman pottery. Square H is located in the lower “dip” in the tall, just to the west of where the remains of the Late Roman building had been excavated. The renewed excavations uncovered parts of the building’s courtyard. On the east side of the square, a nearly complete bread oven was found, underneath which were several earlier ovens. A deep test trench showed five superimposed surfaces of packed mud, all containing much pottery from the Early to Late Roman period, mainly cooking pots (fig. 9), jars, bowls, and juglets. In addition, 17 limestone cup fragments were recovered from several excavation contexts. These cups, used in the first centuries B.C.E. and C.E., are well known from excavations at Qumran and Jerusalem. At Tall Abu Sarbut, these objects may signify contacts with Jewish communities west of the Jordan River, and this should come as no surprise, as the site is located in ancient Peraea.

backfill was a pit with metal workshop debris, including crucibles, nozzles, molds, fragments, and a poorly preserved clay mold that was in situ at the bottom. Although the mold is clearly for a large circular object, its exact purpose was not determined because only about 10 cm of the sides were preserved. The central and lower parts of the mold, made of loose soil mixed with tile and brick fragments, were even more poorly preserved. This fragile artifact could not be properly conserved at the time because of limited resources. Even though the mold was an exciting and rare discovery, it was not deemed to be a research priority; furthermore, the mission did not include specialists in mold conservation or metal casting. In 2012, with the support of the German and French embassies in Amman, the University of Jordan, Yarmouk University, and the Department of Antiquities of Jordan, a new effort began to restore and reconstruct the mold fragments and to excavate new squares with the hope of finding additional molds; fragments of a second large circular mold were found at the bottom of another pit adjacent to the first (online fig. 7). Excavation and conservation of these mold fragments in situ revealed that they were part of a large bronze kiln capable of producing more than 300 kg of metal alloy. Analysis of copper slag deposits also found in association with the complex will provide further information about bronze manufacture at the site. Initial study suggests the in situ mold was used to cast a large metal basin (diam. 1.1 m; depth 1.2–1.4 m), while analysis of the original 1993 fragments indicates that not one but several molds were most likely used to cast garment pieces for large statuary. No absolute date for the molds was determined by the recent excavations. The stratigraphic position of the pits makes it clear that they were in use after the

jarash: bronze workshop in the courtyard of the sanctuary of zeus Lutfi Khalil and Thomas Weber, University of Jordan, and Jacques Seigne, Université de Tours and French Archaeological Mission in Jarash, report: Excavation beneath the paved courtyard of the Sanctuary of Zeus at Jarash in 1993 identified the leveled and backfilled remains of an early shrine erected by Theon, son of Demetrius (130/140 C.E.). Dug into the

© 2014 Archaeological Institute of America

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

destruction and subsequent backfilling of Theon’s shrine in 135/40 C.E., and they certainly date before the earliest courtyard pavement was built over the abandoned pagan sanctuary in the middle of the fifth century. It can thus be suggested that the bronze molds were installed in connection with the construction of the great peripteral temple dedicated in 162/3 C.E., a monument that would have necessitated the manufacture of large bronze statues and basins for display.

jarash northwest quarter project Achim Lichtenberger, University of Bochum, and Rubina Raja, University of Aarhus, report: The Danish-German Jarash Northwest Quarter Project (JNQP) investigates the settlement history of Jarash’s northwest quarter near the monumental Artemision. This area, the highest within the walled city, went largely unexplored by earlier projects. In 2011, the JNQP conducted a geodetic and architectural survey of the area, along with geomagnetic investigation of subsurface remains. In 2012 and 2013, eight excavation trenches were finally opened. The project produced a detailed plan of the area’s surface features, extending across an area of approximately 4 ha, including several thoroughfares and major building complexes, and showed that these features were built along terraces that were aligned with the east–west orientation of the hill (online fig. 8). It was already apparent during the survey that most of the visible structures were of Byzantine and Islamic date. Excavations detected an Ayyubid/Mamluk settlement on top of the hill. The settlements consists of several small and associated complexes positioned along a large terrace in front of the so-called Ionic Building, which occupies the summit. Jarash’s northwest quarter was used initially as a quarry, most likely providing stones for nearby constructions, and was most densely occupied during the Byzantine/Early Islamic period. An important feature of that period was a complete oil press excavated in the northern part of the quarter in 2012. Found in secondary use within the press was a large stone block with altar iconography. This block, which originally stood nearly 3 m high, must have been part of the entrance to a building (perhaps a sanctuary), since a clear anathyrosis is visible on its right side, while the front and left sides are worked. The front features horns at the corners and a deep niche flanked by pilasters. The short side carries a stylized offering bowl flanked by two horns, a typical decorative element on Jarash’s Roman altars. A large rectangular cistern on the southern slope of the hill—the largest cistern within the city—dates to the Roman period. This cistern was used until Byzan-

© 2014 Archaeological Institute of America

643

tine times and includes several phases. After the collapse of a natural cave located beneath the cistern, the cistern was reused as a production and habitation area, while the cave below functioned as a water reservoir. In 2013, the project opened trench G north of the hill in the hopes of identifying additional remains of the north decumanus, which is visible farther to the east near the North Theater. An earlier AmericanBritish excavation failed to find conclusive evidence of the street in this area. Trench G was excavated to virgin soil and unfortunately yielded no traces of a Roman or later street. A fair amount of organic material and traces of terrace walls were found, suggesting the area was used for agriculture at some time. Pressurized water pipes were excavated in trenches E and H. AMS analysis of the mortar fittings of the pipes indicates a second-century C.E. date. This result is further supported by analysis of mortar from the earliest phases of the large cistern, which provided a date between the late first and second century C.E. So far, Hellenistic finds from the northwest quarter have been sparse, consisting of only pottery and glass.

al-hsayyah/eh-sayyeh Karin Bartl, Orient Department, Deutsches Archäologisches Institut, and Zeidan Kafafi, Yarmouk University, report: The Neolithic settlement site of al-Hsayyah (previously spelled “eh-Sayyeh”) is located about 10 km to the northwest of the city of Zarqa, on the slope of a ridge, close to the confluence of Wadi Zarqa and one of its small, nearly perennial tributaries, Wadi edhDhuayl. While the modern landscape along both the lower ridge slope and within the Zarqa Valley itself is marked by intensive agricultural exploitation, it is possible that in prehistoric periods the valley harbored dense forest and wildlife. Initial investigations at the site were carried out under the direction of Z. Kafafi and G. Palumbo in the 1990s and furnished substantial data on the human occupation of northern Jordan during the eighth to sixth millennia B.C.E. The site’s crucial importance for our understanding of the chronology of the otherwise poorly documented phase of settlement development in Jordan between the Early and Late Neolithic led to the resumption of investigations in 2013. The project, launched in cooperation with Hashemite University, included a training program for university students who were taught new documentation and field techniques. Excavations focused on five areas in different parts of the settlement. The results generally reflect a dense sequence of Neolithic layers in each settlement area, although a continuous sequence of layers spanning

644

G.J. CORBETT ET AL.

the eighth to sixth millennia B.C.E. still needs to be substantiated. In all areas, the architecture is built from either unworked or only slightly worked stone blocks of varying size gathered from around the site. In three of the excavation areas, no ceramic remains were discovered, which may suggest an Early Neolithic date. Pottery was found in the two remaining areas, although in only one case could the pottery be identified as Late Neolithic Yarmoukian. Early Neolithic layers of the eighth to early seventh millennia B.C.E. were found in the western and eastern areas of the settlement, and the latter revealed portions of a large building with three parallel walls and a solid floor of lime mortar. The Late Neolithic layers of the seventh millennium B.C.E. are characterized by large, dispersed buildings that are rather poor in finds. In addition to irregularly shaped walls, the identified structures display a trough-like depression, which likely functioned either as a reservoir or a central posthole. The radiocarbon results obtained from seeds of grain in the Yarmoukian layers provided the dates cal B.C.E. 6396–6230 (2σ) and 6416–6241 (2σ), both of which corroborate the date generally given for the beginnings of pottery development in the southern Levant. The ceramic evidence consists primarily of typical plant-tempered Yarmoukian type, though it is mostly undecorated except for occasional examples with red slip. Other types of decoration include slightly protruding bands with carved herringbone patterns. The modest array of forms consists of pots, cups, and bowls. The relatively small number of sherds may indicate that early Yarmoukian pottery was still very rare in the middle of the seventh millennium B.C.E. The site’s lithic industry is represented mainly by a large number of flakes, as well as a few well-defined tools, such as leaf-shaped arrowheads and several sickle blades. The broader range of finds consists of different artifact groups related to everyday needs. It includes numerous bone tools, such as awls and needles for leather and textile working, as well as heavier tools, such as stone axes for working wood and grindstones for processing vegetable foodstuffs. Jewelry consists mainly of pierced shells and bone beads.

khirbat al-batrawy Lorenzo Nigro, Rome “La Sapienza” University, reports: Khirbat al-Batrawy was a major city of EB II–III, controlling the ford through Wadi Zarqa and the routes passing from the desert to the Jordan Valley. The project’s eighth (2012) and ninth (2013) seasons were devoted to excavating and restoring the city walls and the so-called Palace of the Copper Axes (Palace B).

© 2014 Archaeological Institute of America

[AJA 118

Excavation of the fortifications on the northern side of the city exposed Tower T.830, a large rectangular tower protruding from the main inner wall. The structure, preserved to a height of more than 2 m, was added to the main inner wall in EB II and was intended to defend the city at its most vulnerable point of attack. The tower, along with the main inner wall, was rebuilt in EB III, when it was encircled by an outer wall that protected the new city gate located to the west. The tower remained in use until the final destruction of the city at the end of EB III. Carbonized fragments of collapsed wooden beams from the tower’s superstructure were uncovered, along with burned bricks and other tumbled material from the collapse. The monument likely reached a height of approximately 8 m. Exploration of the Palace of the Copper Axes continued toward the west, east, and south (online fig. 9). We excavated Courtyard L.1046 in the palace’s eastern pavilion, defining its southern limit and determining that it was entered through a double-pillared passage from Corridor L.1050, the central passageway that separates the two wings of the palace. From Courtyard L.1046, a door in the northern wall gave access to Hall L.430, a huge rectangular room with a staircase on its eastern side. This wing of the palace was presumably used for gatherings and gift exchange, as we found here a copper fragment and a carnelian bead. In the palace’s western pavilion, work focused on Hall L.1250, a slightly elevated 4.5 m wide area located on a rock terrace near Hall L.1110. Various rock-cut steps and features, including a ramp and a series of installations, define the hall’s southern limit. Near the middle of the room, a series of cup marks suggests that this space was devoted to food processing and storage. On another upper terrace to the south, L.1230, large ceramic vats and grinding stones were found. The continued excavations in Storeroom L.1120 led to the discovery of a second upper disk of a potter’s wheel (the first one was found in Pillared Hall L.1040 in 2010), which may indicate palatial involvement in promoting such technological innovations. In Pillared Hall L.1040, a fragment of an Egyptian palette was discovered, which, together with the earlier discovery of a lotus vase, highlights Egypt’s connection to early urbanism in Jordan. As yet, no definitive absolute-dating evidence can be derived from either these finds or radiocarbon dates obtained from the palace’s destruction layer. Restoration work was carried out both in the palace and along the northern city walls. Laboratory analyses were performed on the previously discovered copper axes. The results suggest the copper ore originated from Wadi Faynan or perhaps Timna. Finally, the beads used to make the necklace discovered in Hall

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

L.1110 (online fig. 10) were found to include semiprecious stones originating from both Egypt (amethyst, carnelian) and the Arabian Peninsula (olivine, rock crystal).

tall al-‘umayri Douglas R. Clark and Kent V. Bramlett, La Sierra University, report: As part of the larger Madaba Plains Project, excavations at Tall al-‘Umayri have sought to understand the survival strategies of the site’s occupants over the past several millennia. Recent research has focused primarily on recovering cultural remains from the Bronze and Iron Ages. A successful lawsuit against the Jordanian government by one of the site’s two landowners meant, however, that 2012 could have been our final excavation season (although continued negotiations have allowed excavations to continue for the time being). The resulting adjusted research objectives focused on bringing excavation in fields A, H, and L into phase; implementing a virtually paperless recording system using iPads and iPods for data retrieval; and recording absolutely everything with as much digital technology as possible. The primary objective in field A was to expose the fourth and fifth of five Late Bronze/Early Iron I buildings along the western perimeter wall of the site. The first three structures, Buildings B, A, and C (from north to south), were excavated in previous seasons. Building D was almost entirely exposed down to its two use surfaces, both of which are associated with a destruction of some kind and separated from each other by only 10–15 cm. Because the lower surface was in use when a major earthquake struck, requiring repairs, and because the upper surface was in use when the entire town was destroyed by external forces, the two surfaces mark the beginning and end points of a newly identified subphase. Along with bronze points that may indicate the cause of the destruction, household assemblages emerged, suggesting that all these adjoining buildings served domestic functions. In field H, the intention was to bring the remainder of the field into phase with its Early Iron I domestic structure (four rooms with quoin-and-pier pillars) and to excavate several key probes to establish more firmly the date of the structure and the origins of its large exterior walls, as well as stratigraphic connections with the destruction apparent in fields A and B. With the entire structure now fully excavated, it is possible to reconstruct the sequence of construction, use, and abandonment, beginning with the building of three large exterior walls against a preexisting perimeter wall to the west. Inside this larger rectangular structure several large pillars joined by smaller, less substantial walls were constructed to divide the inner rooms.

© 2014 Archaeological Institute of America

645

Artifactual remains suggested food-preparation and textile activities, as well as personal sealing practices. The probes demonstrated that the Late Bronze/Early Iron I occupational phase in this part of the site is earlier than the excavated building. The two phases are separated by a short-lived, but clearly distinct, occupational phase that dates closer to the Early Iron I of the domestic structure than to the Late Bronze/Early Iron I settlement. Work in field L continued to record and remove Hellenistic occupation material to reveal the Iron Age structures beneath and to determine the date of the large east–west wall that runs across the breadth of the area. Also investigated was the relationship of this wall to a north–south line of monumental stones to the north. The wall was founded in the Iron Age I period and rebuilt in Iron Age II, when massive stones were part of a huge rebuilding and fortification phase. Excavation also uncovered domestic and food-storage areas. The 2013 season was extremely brief and involved only two staff members along with two professors from Hashemite University. After observing what appeared to be the foundation of a second dolmen mortuary installation, the project team undertook a subsurface mapping strategy to locate additional dolmens (fig. 10). The data, which are still being analyzed, suggest the need for continuing the search for dolmens.

tall hisban cultural heritage project Bethany J. Walker, University of Bonn, and Øystein S. LaBianca, Andrews University, report: The Tall Hisban Cultural Heritage Project returned to the field in 2013 for the third phase of the excavations, which focused on scientific investigations of the medieval village on the slopes of the tall, the development and application of new technologies, and site presentation. Excavations were designed to address very specific questions related to the history and development of the medieval Islamic settlement and the site’s water systems. Fieldwork resumed in areas in which excavation had begun in previous seasons. The project team cleaned the Iron Age reservoir (field B), which was excavated in the 1970s, to clarify and reinvestigate water channels and map their functional relationships to one another and to the adjacent caves. Different phases of water harvesting and use were revealed as a result, suggesting that at different times the reservoir was used as a quarry and then a reservoir, as part of a larger complex system of irrigating terraces, fields, and gardens. When the reservoir was being cleaned, a stone seal was recovered from interseasonal fill. The seal bears a pattern (ibex and kid) similar to that of a bulla excavated in 1973;

646

G.J. CORBETT ET AL.

[AJA 118

Fig. 10. Photoboom image of dolmen remains in field K at Tall al-‘Umayri.

both can be dated stylistically to the seventh to sixth centuries B.C.E. and may be Ammonite in origin. To locate one corner of the reservoir, excavators opened a single excavation unit above it in 2011. Continued excavation in that square revealed the walls of a Byzantine house with three pits, which was rebuilt and reoccupied in the Mamluk period. One stone-lined pit, dated to the 16th century C.E., produced a nearly complete handmade jar, which was found among refuse that included pottery and glass dated to the Late Mamluk/Early Ottoman period. To make visible the fortification wall and to understand better its history of construction, additional sections of the wall exterior in fields M and R were excavated in 2012. This work revealed a system of boulderconstructed towers built on bedrock (which may have been part of the original fortification system), a series of possible terrace walls of similar construction (a system that continues down much of the slopes and appears to give the tall its present form), and rows of Mamluk barrel-vaulted structures used at different times as houses and stables. Excavation of barrel-vaulted buildings on the southwestern slopes (field O) continued from the 2007 season and revealed two adjacent Mamluk-era farmhouses with well-plastered floors and walls. They seem to have been part of the larger village settlement.

© 2014 Archaeological Institute of America

Houses of the same size and structure shared cisterns and common walls. Although foundation levels were not reached this season, it is likely that these structures were constructed earlier and reused in the 14th and 15th centuries C.E., as elsewhere at the site. Unlike the domestic structures of the upper slopes, those on the lower slopes and beyond continued to be occupied into the Ottoman period and reveal a distinctly different spatial organization in relation to one another, the possible result of different patronage patterns in the Mamluk period. In addition to excavation, and as part of continued efforts at site development, a new welcoming area— with seating for lectures—was built adjacent to the site entrance using abandoned building materials from the site. Plans were also drafted for rehabilitating part of a Late Ottoman/British Mandate–era farmhouse (locally known as the “Nabulsi qasr”), one of the oldest buildings in the village, and putting it to use as an offsite visitors’ center.

tall madaba archaeological project Debra Foran, Wilfrid Laurier University, reports: The 2012 season of the Tall Madaba Archaeological Project (TMAP) was the 12th season of excavation on the western slope of Madaba’s acropolis (field B) (fig. 11). Excavations began in 1998 and have uncovered

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

647

Fig. 11. Western slope of Tall Madaba’s acropolis.

remains dating to the Late Ottoman, Late Byzantine/ Early Islamic, Early Roman/Nabataean, Late Hellenistic, Iron Age II, and Iron Age I periods. The primary objectives of this season were to investigate further the Iron Age I remains, to expose the southern extent of the city’s fortification wall, and to clarify the nature of the previously excavated architecture in the southern portion of the site. The Iron Age remains at Tall Madaba are sealed in some areas by a thick layer of sheetwash. The latest Iron Age phase (late in Iron Age IIB) consists of what has been termed “squatter” occupation. The early part of the Iron Age IIB phase is represented by a series of pillared buildings, and Iron Age IIA remains consist of a large multiroom building. Excavations at the northwestern end of the site continued to expose these features. More of the Iron Age IIA complex was uncovered, and the remains of a second Iron Age IIB pillared building were also exposed. The one feature that dominates the site is the massive preclassical fortification wall that runs north– south through the site. Previous excavations had exposed the top of the wall and its inner and outer faces. The original construction of the wall has been dated to the Early Bronze Age, although it was reused throughout the Iron Age. This season’s excavations at the southern end of field B succeeded in exposing more of the top and inner face of the wall. In this area, the wall was built with a narrow ledge at a lower level along its inner face, although the rest of the characteristic construction techniques of Madaba’s fortification wall are also preserved in this area. Excavation efforts at the southern end of field B also focused on elucidating the nature of the ashlar-built wall that currently supports the western wall of a large

© 2014 Archaeological Institute of America

Late Ottoman house. The wall sits on top of a larger construction that dates to the Hellenistic period and possibly the Iron Age. This foundation indicates that the ashlar wall may have been part of a monumental Byzantine-period structure that once stood atop Madaba’s acropolis. The remains of a series of structures dating to the Late Hellenistic period have also been unearthed. Excavation at the southern end of field B focused on clarifying the relationship between the Hellenistic architecture in this area and the city’s fortification wall. It appears that the defensive walls dating to the second century B.C.E. were built over the fortification wall; however, the city’s defenses were surely incorporated into the overall building program initiated in this area during the Hellenistic period. Excavations also exposed the remains of a large drain associated with the Hellenistic occupation. Now that excavations on Madaba’s west acropolis have come to a close, issues of site maintenance and presentation have become paramount. In the future, resources will be allocated to preserve the excavated remains and prepare the site for presentation to the public.

wadi ath-thamad project Steven Edwards, University of Toronto, reports: The 2012 excavations at the Iron Age site of Khirbat al-Mudayna on the Wadi ath-Thamad were focused primarily on the areas to the east and west of Gate 100. In field A, two rooms were uncovered immediately east of the gate, but within the outer casemate wall. To the west of Gate 100, in field C, five additional rooms were exposed. These rooms are located between the western wall of the gate and a large circular depression in

648

G.J. CORBETT ET AL.

the northwest corner of the site. A small excavation unit was opened in the depression in an attempt to determine its function. A salvage operation was also carried out at Wadi ath-Thamad site 200 (WT 200), a recently identified Iron Age rock-cut tomb located about 350 m west of Khirbat al-Mudayna. In field A, east of Gate 100, the northern Room 135 is triangular in plan. Gate 100 and the outer casemate wall form its western and northeastern walls, respectively. The contents of the room included four intact vessels and at least three smashed vessels, all of which were recovered in a debris layer above the only apparent floor in the room. The seemingly random deposition of the finds within debris layers suggests that the room functioned as a dump. No doorways were recorded, indicating that access into the room must have come from above, possibly in the northwest corner through an opening in the east wall of the gate. Room 134, directly south of Room 135, was only partially excavated, but the limited exposures have already revealed four occupation layers that all evince cooking and other domestic activities. At least one fire pit and several ovens were unearthed, and each occupation layer contained oven fragments. A doorway was located in the southeast corner of the room, but it was partially blocked by a stone-built bin installation. In field C, four rooms were excavated in the area to the west of Gate 100. The northernmost casemate room consisted of two small rectangular units (Rooms 154 and 158) divided by a low partition wall. A doorway through the inner casemate wall provided access to Room 156 to the south, though this was later blocked. This rectangular room was bounded on the west by a well-preserved mudbrick wall. Farther south, the elongated Room 157 contained a doorway in the southwest corner that may have provided access to the large circular depression just to the west. Burned beams and considerable deposits of ash point to a significant destruction event. The contents of these rooms included several millstones and 11 loomweights, which suggests that cooking and weaving activities took place in this area. The excavators opened a small north–south oriented trench to the west of Room 157 to determine whether the large depression functioned as a water system. Bedrock was exposed along the southern edge of the feature, and it was shown to slope steeply from south to north. The bottom of the depression was not reached, and so far no evidence for the plastering of the bedrock has been detected. Salvage excavations were carried out at Wadi athThamad site 200, a previously unknown Iron Age rock-cut tomb located on a gently sloping hill on the left bank of the wadi, roughly 350 m west of the main mound. A. al-Bwareed of the Department of Antiquities © 2014 Archaeological Institute of America

[AJA 118

of Jordan’s Madaba office notified the project director of looting activity at the site, and a salvage operation was subsequently organized. The tomb is 6 x 4 m and approximately 2.7 m in depth. The limestone ceiling of the tomb had collapsed, exposing the rock-cut chamber. Looting activity had partially emptied the tomb, but the floor had not yet been reached. The remainder of the tomb was systematically excavated and documented, and the debris piles produced by the illicit excavations were resifted. The contents of the tomb included several smashed pots with painted decoration, along with several lamp fragments. The assemblage suggests an Iron Age II date for the tomb. Human remains were also recovered both within the tomb and in the debris piles adjacent to the chamber.

khirbat iskandar Jesse C. Long, Jr., Lubbock Christian University, Suzanne Richard, Gannon University, Rikke Wulff Krabbenhöft, Uppsala University, and Susan Ellis, Wayne State University, report: The objectives of the 2013 season in Khirbat Iskandar’s Area B were to (1) expose more of the EB IV settlement and investigate connections to the fortifications; (2) expose more of the EB III phase C settlement; and (3) uncover occupation associated with the earlier phase D fortifications. The EB IV phase A settlement—multiroom structures, pillared buildings, and domestic remains—is the latest occupation on-site. Excavation in new square B22 exposed more of phase A, including the corner of a domestic structure. The hard-packed earthen surface associated with phase A was, predictably, the top of the phase B roof collapse. Phase B is known for its well-preserved EB IV public building/storeroom. This season of work achieved more horizontal exposure of phase B to the south in squares B22 and B21. The typical phase B level of roof collapse contained restorable vessels and broken pottery. In reopened square B21, a domestic context in which several whole vessels were excavated in 2010, we discovered at least 14 more whole and restorable vessels in the roof collapse. The fortified phase C (EB III) settlement comprises an upper (phase C1) destruction level and a lower (phase C2) foundation level. The phase C1 settlement includes a central room, an adjoining courtyard, and a contiguous workroom. The stratigraphic removal of portions of the square B9 balks revealed more evidence for the phase C1 destruction and the extension of the phase C1 courtyard. Another objective of the 2013 season was to expose more of the phase D settlement associated with the curvilinear structures discovered on either side of a threshold, an apparent portal in the earliest fortifications at

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

the site. Excavation below phase C levels in squares B2 and B1 brought to light new information about the phase D settlement. In square B1, we uncovered the corner of a substantial stone structure (ht. 1 m), which ran under the phase C tower platform. We are tentatively assigning this new building to phases D and E. Although the upper surfaces were reused in phase D, the bottom course of the wall lay below the founding level of the curvilinear structure. Notably, the ceramics embedded in several surfaces dated the structure to EB III, thus clarifying the date of the phase D settlement (originally thought to be EB II). The presence of this substantial new structure in early EB III suggests that there may be an earlier fortification system. The site’s fortifications consist of (1) a phase C northern (outer) line and square tower; (2) a phase D earlier (inner) mudbrick and stone wall; and (3) a phase C western perimeter line with a rubble interior that abuts the southwestern corner of the tower. The most significant find of the 2013 season was a new fortification wall, a major phase C defensive line that runs parallel to and along the exterior of the previously excavated phase C western perimeter wall (fig. 12). The new wall was discovered during excavation of a section of the previously known wall. The section revealed more of the rounded tower feature but, more importantly, it exposed for the first time the fortification wall (Wall B4A006) that abuts the northwestern corner of the tower. This substantial (2 m wide) wall is the original phase C western perimeter wall built contemporaneously with the square tower and northern outer wall. The fact that this phase C wall cut phase D Wall B2A077 adds to a wealth of data that confirms the phase D defenses are earlier than the phase C outer line. As noted above, we now date phase D to an earlier stage of EB III than phase C. The discovery of the new fortification wall places the previously discovered rubble-interior western wall in new light. The latter is late, and based on an accumulating set of data it appears to have been reused, or perhaps even first constructed, in the EB IV. To further clarify the relationship between EB IV phase B and the rubble-filled western wall line, we removed part of a phase A wall that obscured the link between the two. This investigation confirmed that the threecourse phase B wall (Wall B19A019) continued to the western perimeter wall as a six-course transverse wall (Wall B19A043). This shows definitively that the site’s phase B inhabitants built up against and reused the outer wall line.

balu’a regional archaeology project Monique D. Vincent, The University of Chicago, Matthew L. Vincent, University of California, San Diego,

© 2014 Archaeological Institute of America

649

and Friedbert Ninow, Theologische Hochschule Friedensau, report: The Balu’a Regional Archaeology Project team resumed excavations in 2012 to understand better the full extent of occupation, including not only the Iron Age but also the Bronze Age, the Roman period, and a later Mamluk-era village discovered in the western half of the site. In 2010 and 2012, an intensive survey was undertaken to define the horizontal extent of occupation at the site. Using a GPS system, the team mapped most identifiable exposed architectural features in the immediate vicinity of the khirbat. The resulting data were incorporated into ArcMap and used to determine the overall layout of the site. The predominant ruins belong to the Iron Age settlement, as confirmed by previous excavations, and could be as large as 16 ha. A Middle Islamic complex (dated by surface sherds) dominates the western end of the site and was possibly built within preexisting Iron Age ruins. Farther to the west there is a possible Roman settlement, close to 5 ha in size (also dated by surface sherds). As the site is excavated, the map will be refined to include orthophotography and dating information for the architecture. Relying on the architectural information gathered during the 2010 survey, the excavators selected two adjacent buildings for excavation in the southeastern area of the settlement. Both buildings consist of rectangular north–south structures with small southern rooms formed by a casemate wall, and larger, open northern rooms. Probes were opened in each building and excavated to sterile soil. The goal was to determine the date and occupational phases of the layers associated with the casemate wall system. These probes revealed a single phase with very scant evidence of occupation, likely resulting from the site’s abandonment toward the end of the Iron Age. The few securely dated sherds are Late Iron Age II. Two other probes were placed toward the western end of the khirbat and were designed to detect the earliest periods of occupation and create a diachronic picture of the site’s occupation. Each probe was marked out against visible wall lines north of the qasr. Unfortunately, the first probe was halted by vandalism, but not before a previously unknown Hellenistic phase was detected. In the second probe, an Iron Age phase was excavated to a surprising depth of 2 m, after which excavation was halted for the season. A portion of a doorway was revealed, and a smashed jar was found on an earth surface before it. A subprobe in the surface of this phase indicated that more occupational phases continue below. Sherds found with the Iron Age pottery of this phase point to a possible earlier Bronze Age

650

G.J. CORBETT ET AL.

[AJA 118

Fig. 12. Khirbat Iskandar: the northwest square tower with two abutting walls, newly discovered western wall (Wall B4A006, under photograph board), segment of the western rubble wall on right, and curvilinear (phase D) wall below.

occupation. A heavily burned mudbrick layer sealed this building, indicating the Iron Age phase ended with a fiery destruction. While our search for pre–Iron Age stratified remains was unfulfilled because of vandalism and the surprising depth of the Iron Age deposits, we did discover a previously undetected Hellenistic phase. We also discovered a destruction layer in the central part of the Iron Age settlement, which is notably absent from the eastern extension of the settlement. There is still a great deal of the Iron Age settlement to explore, but from the two eastern probes, we can argue that the expansion and settlement here lasted only a short time before it was abandoned.

al-lajjun survey Jennifer E. Jones, University of Minnesota Duluth, reports: The broad purpose of the survey of al-Lajjun on the Karak Plateau is to understand social organization and craft production during the period of population aggregation in the third millennium B.C.E. The Early Bronze Age witnessed a change in land use toward the occupation of larger, fortified settlements and the intensive cultivation of grains and orchard crops.

© 2014 Archaeological Institute of America

The nearly 12 ha EB II settlement of al-Lajjun is located in south-central Jordan between Karak and Qatrana. A fortification wall, wide internal walls, and a large circular feature are visible on aerial photographs and Google Earth imagery. Since there was no subsequent occupation at the site, even the foundations of the Early Bronze Age structures and walls are visible on the surface. Al-Lajjun thus provides a rare chance to research the internal layout and activity areas of an Early Bronze Age settlement. Nearby the site is a row of large standing stones (menhirs) well known to both the archaeological and local communities. Earlier test excavations identified 0.5 m of soil deposition at the site, along with floral and faunal remains that included grass pea, common pea, large legume, olive, grape, barley, einkorn, emmer, sheep, goat, gazelle, cattle, and donkey. During the initial season, in 2003, we mapped concentrations of sherds and chipped-stone artifacts in four sample areas to begin understanding intrasite craft production and artifact deposition. We focused in 2012 on identifying the site’s main architectural features. Among the 100 complete and/or partial features, we discovered various rectilinear, multiroom structures; a long, narrow building at the summit; two

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

large stone circles (one inside the fortification wall and another outside); and several small tombs. Rectilinear, multiroom structures were identified across the site, with the best-preserved examples measuring 6–11 m long and 3.0–5.5 m wide. We also identified a long, narrow structure that may be a temple. The structure’s interior measures 10.7 x 2.8 m. The walls are built of stones as large as any of the stones used in the fortification wall and much larger than those used in the rectilinear buildings. Sections of what may be an external wall surrounding the possible temple were also discovered, measuring at least 27.5 x 15.3 m, although the wall’s full dimensions are obscured by dense rubble. Close examination of the large circular feature visible on aerial photographs revealed a thick circular wall 32 m in diameter enclosing a 12–15 m wide area of stone rubble with a small square feature at its center. Another 15 m wide stone circle located outside the fortification wall also has a small rectangular feature at its center. These enclosed features may be tombs but are much larger than other tombs scattered across the site, which are generally marked only by two standing stone slabs. On a sadder note, the famous line of menhirs appears to have been damaged in the months leading up to the 2012 season. A hole was dug at the base of the stones, causing two of the four to collapse. This is an unfortunate loss, as the menhir line was a unique and, until then, largely undamaged ancient monument.

ghor es-safi Konstantinos D. Politis, Hellenic Society for Near Eastern Studies, reports: The 2013 excavations at Ghor es-Safi focused on exploring the origins and development of the sugarcane factory complex at Tawahin as-Sukkar and the nearby urban center of Khirbat esh-Sheik ‘Isa. Investigations were also made into the earlier Roman and Byzantine phases of the two sites, as well as the environs. Previous excavations at Tawahin as-Sukkar uncovered lower chambers with arched entrances below the pressing rooms and several external stone walls. It was presumed that one of the walls leading northward connected with an underground channel that carried water onto adjacent fields. This season, a stone stairway was uncovered, and it apparently gave access to the pressing rooms. A pivot stone was found, providing good evidence for the pressing mechanism. A small sounding was made at the entrance to the arched building farther north, revealing several courses of well-hewn sandstone blocks. The construction methods and materials were different from those used for the upper courses of mudricks from a later period

© 2014 Archaeological Institute of America

651

(probably Late Ottoman). The building’s architecture and lower elevation indicate a close association with the pressing rooms and their related chambers to the south. A close examination of Tawahin as-Sukkar’s mudbrick walls and finds (particularly cartridge shells) indicates a final occupation phase during the early 20th century, when the abandoned sugar factory was reused during hostilities associated with the Arab Revolt. At Khirbat esh-Sheik ‘Isa, excavations resumed in trenches II and VIII, while three new trenches (IX, X, and XI) were opened to the west to follow the line of a long, well-built, 3 m high wall. A deep sounding below the mosaic pavement in trench VIII revealed Late Roman/Early Byzantine sherds resting on wadi sediments, thereby confirming the date of the site’s earliest occupation. The date of the wall, however, is still uncertain: it slightly overlies the mosaic in some areas, but trench IX also showed that the mosaic clearly ends at the wall. Additional finds, including chancel fragments of fine white marble, provide clear evidence that a church was built in the immediate vicinity. The wall continued into trenches X and XI. The upper levels of trenches IX, X, and XI mirrored the architectural remains from trench VIII and included small, roughly built stone rooms and finds dating to the Ayyubid/Mamluk period (ca. 12th–15th centuries). At the lower levels, work uncovered larger mudbrick rooms that were apparently peacefully abandoned in the Abbasid period (ninth and 10th centuries). Excavations in trench II to the north revealed a similar sequence of Abbasid and Ayyubid/Mamluk occupation. In this area, excavations also reached the level of the paving slabs, thereby providing a clear view of the Byzantine-period occupation (fig. 13). In light of the 2013 investigations, the following preliminary conclusions regarding the main occupational phases of Khirbat esh-Sheik ‘Isa can be drawn. The area of the Khirbat esh-Sheik ‘Isa excavations was presumably the original location of the main church of Byzantine Zoara. The succeeding settlement in the 10th and 11th centuries represents the occupation of the then-abandoned church building. This was followed by a substantial period of occupation during the 12th–15th centuries closely associated with the nearby sugar factory and related industries. In addition to containing workshops and profiting from the sale of sugar, Khirbat esh-Sheik ‘Isa also likely served as a village for local workers, since tabunat (bread ovens) and other domestic items were discovered inside its houses. Finally, the project worked to conserve several of the in situ mosaics discovered in trenches VIII and IX, as well as pottery, stone, and metal finds. Some of these finds were put on display in the Museum at the

652

G.J. CORBETT ET AL.

[AJA 118

Fig. 13. Paving slabs of the Byzantine complex in trench II at Khirbat esh-Sheik ‘Isa.

Lowest Place on Earth in Safi. The project also made plans to consolidate and conserve the site’s various ancient structures.

follow the pots: mapping, surveying, and site monitoring at fifa Morag M. Kersel, DePaul University, Austin C. Hill, University of Connecticut, and Meredith S. Chesson, University of Notre Dame, report: The Follow the Pots (FTP) project explores two interconnected sides of a looted landscape: the conventional archaeological mapping, surveying, and site monitoring of an Early Bronze Age cemetery; and the multiple and contested values of this archaeological heritage to various stakeholders today. FTP draws on this comparative data to rewrite the traditional archaeological looting story by focusing on materiality. It considers how Early Bronze Age peoples deployed material culture in graves in the past and how archaeologists, looters, and collectors reuse and revalue this same material culture in the present.3

3

The EB IA (ca. 3500–3200 B.C.E.) cemetery of Fifa contains thousands of cist tombs whose use coincides with the emergence of the first walled, urban settlements in the region. In the archaeology of the southern Levant, this site represents an incredibly important resource: it is one of only four known large Early Bronze Age cemeteries (the others being Bab edh-Dhra‘, Jericho, and Naqa). Unfortunately, since the 1980s the site has been the target of systematic illegal excavation in search of artifacts destined for the antiquities market. While there have been two small seasons of excavation conducted by W. Rast and R. Thomas Schaub in 1989–1990 and M. Najjar in 2001, until recently there was no complete plan of the site and very little record of the threatened landscape. In 2011, FTP surveyed and mapped the extent of the cemetery using a handheld GPS unit and a total station. The initial mapping provided an estimated area of 64,000 m2 for the cemetery. We recorded more than 800 graves in the looted landscape, although thousands more exist, yet to be looted or excavated.

For further information, see the Arabic and English project website, http://followthepotsproject.org.

© 2014 Archaeological Institute of America

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

In examining the upcast from looting, we were able to determine that there were different looting episodes. As part of the mapping, we included the Iron Age structures (at the western edge of the site) and modern military trenches and structures in addition to the EB I tombs. The 2001 excavation area was also included in the FTP map, providing a comprehensive plan of the cemetery (online fig. 11). In surveying tombs in the Fifa cemetery, FTP recorded the two tomb types, slab lined and cobble lined, which were previously identified in the Rast and Schaub excavations, and it also identified a third type that uses large boulders for tomb construction. There is a greater density of slab-lined tombs in the western half of the cemetery, while there are no visible slab-lined burials on the cemetery’s north ridge. In the wadi bottoms there is a higher percentage of slablined burials. We are hesitant to suggest a chronology for the burials, although some evidence suggests that the slab-lined tombs are earlier than the cobble-lined graves: the stratigraphic profile shows the slab-lined burials cutting through an earlier ashy layer thought to be Neolithic. Further statistical analysis is needed to make an assertion with greater confidence. As part of FTP, the Landscapes of the Dead Project was carried out during the 2013 season. One of its primary objectives was to assess changes to this archaeological landscape over time to understand better both the ancient and modern uses of a mortuary site. The main research goals of the Landscapes of the Dead Project were (1) to obtain quantitative data on the scale and intensity of looting at Fifa using innovative aerial photography; and (2) to continue recording and mapping the looted landscape of Fifa. The results of the 2013 season allowed us to establish a baseline from which we will be able to observe and document changes in the looted landscape. A small fixed-wing unmanned aerial vehicle (UAV) provided a platform for stable, low-elevation aerial photography, making it possible to both document looting and destruction at Fifa and generate spatial data for digital mapping. The aerial survey used photogrammetric image processing to produce high-resolution digital elevation models of the site. A chronological assessment of looting will allow identification of new looting pits and determination of whether any of the looters’ holes have been revisited, whether the looting was recent (indicated by fresh earth), and whether there is a discernible difference in looting episodes. Mapping and groundtruthing of the landscape in 2013 allowed us to conclude that looters are revisiting earlier holes, that there is ongoing recent looting, and that there is a difference in looting episodes. The 2013 season of aerial site monitoring was the first of five planned

© 2014 Archaeological Institute of America

653

seasons in which the FTP team will revisit the site at the same time each year to investigate change over time and to assess the potential impact of the Department of Antiquities of Jordan’s anti-looting campaigns and outreach programs.

barqa landscape project Russell B. Adams, University of Waterloo, Hannah A. Friedman, Texas Technical University, David D. Gilbertson, University of Plymouth, and Keith Haylock and John P. Grattan, University of Wales, Aberystwyth, report: The 2013 Barqa Landscape Project (BLP) season continued research begun in the 2009 and 2010 pilot seasons. The project’s primary objective is to assess the archaeological and environmental evidence for the intensification, spread, and impact of copper production in the Faynan Basin of southern Jordan during late prehistory (3000–2000 B.C.E.). The Faynan Basin was one of the world’s earliest industrial landscapes and therefore is important for understanding the historical impact of ancient industry and pollution and the environmental legacy of such activity. During the pilot phase, we developed a quantitative methodology to assess the effects of the expansion of copper production and heavy-metal pollution through time. In two four-week seasons, we used a combination of satellite remote-sensing data, traditional archaeological survey, and limited excavation of test pits to map the spread of pollution both geographically and temporally for a 5 km2 section of the Faynan district at Barqa. Using a portable X-ray fluorescence (p-XRF) analyzer we were able to examine a wide variety of sediments at a range of archaeological sites. Analysis of the anthropogenic pollution observable in the sediments was designed as a reliable and quantitative way of determining the extent and scale of pollution that could also be used to build a diachronic model of the scale and intensity of copper production. The initial season in 2009 was devoted to archaeological reconnaissance and a pedestrian survey of the wider survey zone, while the subsequent 2010 season was largely devoted to excavating test pits and small areas in the northwest quadrant of the Barqa survey zone. The excavations were intended to collect wide-ranging data on pollution from copper smelting during antiquity and to ascertain the diachronic extent and spread of this pollution. The survey’s pilot phase achieved a greater understanding of pollution at one smelting site, as well as the spread of residual pollution in its immediate vicinity. During 2013, expertise developed in the pilot phase was used to expand the pollution analysis across the Faynan Basin. A wide variety of sites (ore-processing

654

G.J. CORBETT ET AL.

locations, waste tips, agricultural sediments, settlement sites) and nonsite localities (open and excavated dunes) were sampled, including open and previously excavated sites. In addition, several pollution transects were carried out across both the Wadi Faynan/Wadi Fidan drainage and its tributaries, the Wadi Shagir and Wadi Ghuweir. In the same year, 760 sample points were taken with the p-XRF analyzer, raising the total collected data points to nearly 5,000. Wherever possible, in situ p-XRF sampling was supplemented by collection of sediment samples for laboratory analysis to confirm the p-XRF data. During a typical analysis of an open excavation, the vertical section was described, drawn, and photographed. Additionally, p-XRF data were collected at various lithofacies and in discrete areas (charcoal lenses, ash middens, and fire pits). Supplementary data for relative and chronometric dating (pottery, stone tools, charcoal) were also collected, and sediment samples from individual lithofacies were collected for laboratory analysis. One of the season’s objectives was the reexamination of the site of Khirbat Hamra Ifdan, a key Early Bronze Age copper-manufacturing center at the eastern end of Wadi Fidan. Our trench excavation aimed to link information gathered from earlier Khirbat Hamra Ifdan excavations with the Barqa Landscape Project’s pollution studies. Since the chronology, pottery, and industrial activities of the site are well understood, what remained was to link the pollution levels of different layers to the known history of the site. Excavation of seven layers of human occupation (including floors, pits, a tabun, and numerous sublayers) were distinguishable, and all were analyzed in situ with the p-XRF device, while 5 ml sediment samples were taken for laboratory analysis. In total, 33 distinct contexts were sampled for pollution analysis. These data will be integrated with the wider pollution analysis of Early Bronze Age copper production to determine changes in the scale and intensity of production throughout the period.

busayra cultural heritage project Stephanie Brown, Benjamin W. Porter, and Christopher Bravo, University of California, Berkeley, and Andrew Wilson, University of Oxford, report: The Busayra Cultural Heritage Project (BCHP) conducted an assessment season in 2013 at the site of Busayra, which is widely believed to be ancient Bozrah, the capital of the Iron Age polity of Edom. Excavations carried out during the 1970s identified monumental architecture, including a possible temple and palace protected by an extensive fortification system used between the late eighth and fifth centuries B.C.E.

© 2014 Archaeological Institute of America

[AJA 118

The BCHP’s principal goals include the investigation of Busayra’s political and economic roles during the Iron Age, as well as the implementation of sitemanagement and development plans with the Department of Antiquities of Jordan. The primary BCHP task was to assess the overall condition of the site, document features using noninvasive techniques, and develop an integrated plan combining scientific research and site management that will guarantee Busayra’s preservation. Several methods of documentation were employed for the assessment. The project created new topographic and architectural maps of Busayra using total-station technology and GIS software. The project also used digital photography, photogrammetry, and threedimensional reconstruction software to document the condition of previously excavated buildings. By comparing the exposed architecture with published drawings and photographs from the earlier excavations, damage was identified and documented. The project found most buildings in remarkably good shape, especially considering that they have been exposed for more than 30 years. Busayra still requires a number of careful interventions to ensure its long-term preservation. These include backfilling select trenches that have experienced subsequent erosion; removing or stabilizing excavated soil heaps and rock piles; and conserving exposed architecture using appropriate restoration materials. Large portions of Busayra remain uninvestigated and need to be documented through geophysical survey and excavation. In preparation for future work, the BCHP performed a random survey of the earlier excavation’s Area D, just north of the acropolis. Twenty-four 5 x 5 m units were surveyed. Ten of the units were selected randomly, while the other 14 were selected strategically to achieve maximal coverage of the area. All artifacts were collected, and sherd-density data were generated for each of the units surveyed. The survey yielded predominately Iron Age IIB and Iron Age IIC ceramic evidence, as well as a limited number of Classical-period sherds. These data will be considered along with other information when the BCHP decides where further sampling should occur. Also in preparation for future excavations, the BCHP sought to develop an appropriate soil-sampling strategy that could yield data to help illuminate the Edomite agricultural and pastoralist economies. Ten 5-liter soil samples were taken from selected exposed sections in previously excavated trenches to determine the density of carbonized botanical remains within the soil. Though the analysis of the collected remains is still ongoing, it is clear that archaeological deposits within buildings possess enough carbonized evidence

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

to make intensive sampling strategies worthwhile. Preliminary analysis has so far identified barley and grape seeds in the sampled materials.

jafr basin Sumio Fujii, Kanazawa University, Japan, reports: The project’s previous investigations confirmed that short-range pastoral transhumance started soon after the domestication or introduction of sheep and goats. This initiated the period that we have termed the Jafr Pastoral Pre-Pottery Neolithic B (PPN-B), which was sustained by an outpost, a cistern, and a basinirrigation barrage system. The research campaigns of the past two years have tried to understand how this PPN-B “triple set” was transformed as the process of pastoral nomadization accelerated during the Late Neolithic period. The site of Hashm al-‘Arfa, a small campsite in the eastern Jafr Basin, offers a glimpse into this process (online fig. 12). The size of this settlement is remarkably reduced, only about 0.02–0.03 ha, which is approximately a tenth of the size of Wadi Abu Tulayha, a type-site of the Jafr Pastoral PPN-B. The site also shows limited structural complexity and consists mainly of a few round, semisubterranean structures. The flint assemblage is made up largely of Late Neolithic artifacts (single-platform cores and bifacial knives) but also includes a few naviform core-and-blade components. The mixture of these two distinct flint assemblages suggests the site belongs to the beginning of the Late Neolithic. Several uncalibrated radiocarbon dates (7900 ± 30 BP, 7730 ± 30 BP) taken from charcoal samples also support this dating. While faunal remains (largely sheep and goat) were abundant, no floral remains were recovered. This fact, coupled with the small size and isolated location of the site, indicates that the small encampment was likely used by pastoral nomads. Also noteworthy is the discovery of a few small bilaterally notched stone weights. Similar artifacts were found around the nearby barrage systems described below, suggesting the site may have been associated with these water-related features. No later settlements have yet been discovered in the basin. This indicates that during the transition from the Jafr Pastoral PPN-B to the Late Neolithic, there was a shift from fixed outposts to tent houses and small encampments. Our recent fieldwork was also concerned with identifying Late Neolithic open-air sanctuaries (or what we have termed “pseudosettlements”). Our previous investigations confirmed two examples in the western Jafr Basin, and the ‘Awja sites in southern Jordan have now added another five. They have much in common with Late Neolithic pseudosettlements in the western Jafr Basin, on the one hand,

© 2014 Archaeological Institute of America

655

and Late Neolithic open-air sanctuaries in the Negev and Sinai, on the other. The presence of such largescale funerary complexes attests to the beginnings of pastoral nomadism in the deserts of the southern Levant. At ‘Awja 1, we also found a group of feline representations outlined on the ground in stone, also attested in the sanctuary sites of the Negev and Sinai. We located a few dozen barrage systems in the eastern Jafr Basin tentatively dated to the Late Neolithic based on the occurrence of small stone weights similar to those found at Hashm al-‘Arfa and the barrage systems from Wadi Nadiya. They differ from the PPN-B barrage systems in many respects. While the PPN-B barrage is larger, has a more open plan, and is located at the lower edge of a semiopen playa, the Late Neolithic barrage is much smaller, more closed, and occupies the center of a small depression. Also, while the PPN-B barrage is focused on the western Jafr Basin closer to contemporary farming communities, the Late Neolithic barrage extends deep into the desert. These contrasts suggest there was a shift from large-scale, basin-irrigation barrage systems during the PPN-B to small-scale, cistern-type barrage systems constructed by pastoral nomads during the Late Neolithic. Our investigations have shown that the Late Neolithic Jafr Basin witnessed (1) the replacement of fixed outposts by tent houses and small encampments such as Hashm al-‘Arfa; (2) the appearance of the open-air sanctuary or pseudosettlement; and (3) a shift toward small, cistern-type barrage systems. The deconstruction process of the Jafr PPN-B triple set seems to mirror the initial stage of pastoral nomadization in southern Jordan.

khirbat edh-dharih Caroline Durand, Institut Français du ProcheOrient (Amman), Pauline Piraud-Fournet, Université Paris, and Laurent Tholbecq, Université Libre de Bruxelles, report: After a five-year hiatus, a new excavation campaign was carried out at Khirbat edh-Dharih in 2013 on behalf of the French-Jordanian Archaeological Mission of Khirbat edh-Dharih (fig. 14). The site, organized around a large sanctuary from the Nabataean/Roman periods, was first excavated between 1984 and 2008. Three operations were undertaken on different buildings to complement the previous excavations. Soundings and cleaning were carried out in the village to determine the phasing and chronology of House V1, which is the largest domestic building at the site. The troughs in the house’s stable were cleaned, and several fallen blocks were returned to their original positions. Soundings also provided a better understanding of the original layout in Room L, which

656

G.J. CORBETT ET AL.

[AJA 118

Fig. 14. Aerial view of Khirbat edh-Dharih (© C. Bacri and A. Normier /ORA).

contained an installation that is probably a bathtub. This installation, combined with the adjoining heated room built on hypocausts, highlights the luxurious character of this part of the house. Finally, soundings were conducted in the northern and southern parts of the house to date the construction precisely. The pottery suggests the house was founded between the first and second centuries C.E., while the material from the destruction/abandonment layers is no later than the fourth century. Additional work was carried out in Area A, about 150 m south of the sanctuary’s southern entrance, on a Nabataean/Early Roman caravanserai and its associated bath complex. The caravanserai is organized around a courtyard and surrounded by individual rooms. The mission team excavated a 7.5 x 2.0 m square north of Room A11 to get a better understanding of the inner court that was partly identified in 2004 east of the main porch. No destruction from the upper courses of the walls has been discovered in the eastern half of the probe. The lack of destruction debris indicates that the building material was looted in this area. Similarly, the exposed base of a pavement was also looted, most likely during the Byzantine period. So far there is no evidence of an inner portico.

© 2014 Archaeological Institute of America

The earliest use of the building is dated to the end of the first or the beginning of the second century C.E. Where they can be identified, the occupation levels can be dated to the second and third centuries. Ceramic finds associated with the destruction levels are dated to the late third or early fourth century C.E. The excavations in the bath complex focused on the southern part of the building (Rooms F, G, and H), and a general plan was completed. One excavated room was identified as a heated room (caldarium). Along the northern, southern, and eastern walls, several rectangular projections, regularly disposed, were used to flank tubuli panels. At least two semicircular grooves that functioned as roof chimneys were uncovered in the southeastern corner of the room. In the western part of the room, the thickening of the northern and southern walls probably indicates the limits of a large pool built against the western wall. A heating channel is visible in the center of this wall and appears to be embedded in a rectangular structure that could be the base of a support for a heated cistern. To the west, a long rectangular room or corridor helped fuel the furnaces for the two heated rooms. Another room, which included another heating channel opening to the caldarium to the south, was also partly excavated.

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

The identical phasing of the caravanserai and the baths suggests that both belong to a single building program initiated at the turn of the second century C.E.

shawbak Michele Nucciotti and Elisa Pruno, University of Florence, report: The Medieval Petra Project of the University of Florence (under the direction of G. Vannini) included continued archaeological investigation at the site of Shawbak in 2013. The aim of the project is to investigate the medieval settlements in the area of Edom, particularly in the region of Petra-Shawbak. The project’s methods involve an integrated system of noninvasive documentation techniques (including documentation of extant buildings, landscape archaeology, and environmental surveys), excavations, and data management. With the expansion of the Latin Kingdom of Jerusalem and the subsequent Crusader reoccupation of the ancient limes arabicus from 1100 to 1118, a notable shift occurred in the settlement pattern of the Shawbak region. In particular, the rapid development of European-styled fortified villages (generally known from written sources as “castles”) gave the region of Edom a completely new role within Transjordan and Outremer. In occupying the strongholds of Shawbak and Petra (including al-Wu’ayra and al-Habis but also smaller settlements such as Wadi Farasa and Bayda), the Latin kingdom did, in fact, reestablish an eastern frontier that closely resembled (at least geographically) the ancient limes arabicus. The 2013 excavation season focused on Area 35000 (online fig. 13), a large, barrel-vaulted, rectilinear structure located in Shawbak’s “monumental” northern sector, adjacent to the audience hall of the Ayyubid palace. While Crusader masonry and building techniques are evident in the structure’s east, north, and west walls, the north side of the structure also shows typical Ayyubid stone-cutting techniques. The Crusader building was most likely reused in the Ayyubid period, possibly as part of the palace complex of Al-Malik Al-Adil and Al-Mu’azzam ‘Isa. A monumental Ayyubid stairway with concentric semicircular steps was unearthed on the east side of the room (online fig. 14); it is connected to an earlier, presumably Crusader, door but was covered by layers from the Late Ottoman–period occupation. Probably during the Mamluk period, the stairway was partly buried, and a new monumental doorway resting on two side pillars was built. The project inventoried and studied the material remains from Area 35000 in hopes of better understanding the material assemblages of the Crusader/Ayyubid period and Shawbak’s pre-Crusader

© 2014 Archaeological Institute of America

657

occupational phases (namely, the Byzantine and Early Islamic periods) as represented by small finds. The documentation and stratigraphic analysis of the Crusader Upper Church (Building 10) was completed, and the nearby vaulted Building 29 was surveyed and documented; this work was part of a unified effort to determine the functional and stratigraphic relationship between the two structures so as to illuminate the transition between the Crusader and subsequent Ayyubid and Mamluk phases. We also analyzed the stone dressings of these and other structures, including the Lower Church (Building 1) and Building 28, and thereby updated the site’s registry of known stone dressings and techniques. Building 29 is particularly interesting in this regard, as its walls show evidence of two different styles of stone dressing that presumably belong to two different periods—the Crusader period (Stone Dressing S8) and the Ayyubid period (Stone Dressing S22.a). From a stratigraphic point of view, Building 29 appears to be later than the Lower Church, as its main structure abuts the western and northern portions of the church. Our stratigraphic and archaeological surveys have also demonstrated the existence of an earlier building (perhaps from the Crusader period) along the northern walls of Building 29.

wadi arabah earthquake project (qasr et-tilah) John D. Rucker and Tina M. Niemi, University of Missouri–Kansas City, report: The Wadi Arabah Earthquake Project (WAEP) was initiated in 1996 to research the archaeoseismology and paleoseismology of the Dead Sea Transform fault through archaeological, geological, and geophysical excavations and surveys at many sites within the valley. The 2013–2014 excavation season focused on Qasr et-Tilah. Qasr et-Tilah, located just south of the Dead Sea, is a large multiperiod site with ruins from the Nabataean, Roman, Byzantine, and Early Islamic periods, as well as artifactual evidence from earlier and later periods. The site’s complex, multiphased water-management system provided water for immediate use and irrigation for an extensive agricultural field system. The site also lies directly across the alignment of the Wadi Arabah fault, and the architecture reveals significant damage and repair related to past earthquakes. Extant remains include a Late Roman quadriburgium (probably the site “Toloha” referenced in the Notitia Dignitatum) and a Late Byzantine reservoir, while abundant Nabataean and Byzantine pottery attest occupation during those periods as well. The excavators opened a trench encompassing the fort’s outside wall and the alignment of an underly-

658

G.J. CORBETT ET AL.

ing structure (possibly Nabataean) to garner data to inform both the dating and the history of this structure and the earthquakes that had affected the site. Excavation revealed collapse layers related to the fort, as well as evidence for subsequent reuse of the ruins. This reuse phase included ash deposits (which provided charcoal for radiocarbon dates) and at least one postcollapse living surface. A second goal was to clarify the phases of Qasr etTilah’s water-management systems. A trench was dug at the bottom of the site’s main (northern) water reservoir. Beneath the surface soil and several layers of silt deposition, we reached a major plastered floor. This floor rested on a subfloor of mortared large cobbles and boulders nearly 1 m thick, which itself overlay another anthropogenic surface on bedrock. We then investigated and excavated a previously unexplored aqueduct that leads toward the northern reservoir, revealing the first firm evidence for a covered aqueduct at the site. However, the anticipated connection between the aqueduct and the reservoir was not found, suggesting this aqueduct leads to a yet-undiscovered location. A trench was opened in the site’s previously unexcavated southern reservoir. This reservoir, located across Wadi et-Tilah away from the rest of the site’s extant architecture, revealed two phases of a plastered floor and may have been reused later as an area of habitation. The excavation produced samples for radiocarbon dating, which will provide an absolute date for the structure. Possibly the most important result of the expedition is the creation of a precise GPS-based map of the site and its topographic surroundings. We created a plan of the fort, a topographic map of the site, and elevation profiles of the field system. The site map also includes areas of recent looting activity, a major problem for cultural resource management in Jordan. This season we documented the location and subsurface sediment exposed in the illicit excavation pits. The looting map will help Jordanian authorities monitor damage to the site.

bir madhkur Andrew M. Smith II, The George Washington University, reports: The Bir Madhkur Project continued its work in the hinterland of Petra in 2013. It expanded the intensive regional survey to cover the south bank of Wadi Musa through the foothills of Wadi Arabah and conducted salvage excavations at a regional caravan station, Khirbat Sufaysif, an important stop on the ancient incense route. This work supports the primary goal of the project, which is to examine the role of Petra in the ancient economy from a rural, landscape perspective.

© 2014 Archaeological Institute of America

[AJA 118

The excavations at Khirbat Sufaysif, a structure measuring approximately 21 x 26 m, were carried out in response to increased looting at the site and continued erosion of the southwest corner of the structure—at least 5 m of the wall had already collapsed into Wadi Sufaysif. The purpose of the excavation was to salvage as much information as possible from the deteriorating site, with the specific goal of determining the periods during which the site was occupied and the nature of that occupation. We opened five trenches, two of which were fully excavated. Moreover, while one trench spanned a room along the east wall, where a small doorway was revealed facing the wadi, the others were located along undisturbed sections of the north wall. Here, along the north wall, two rooms were excavated, one of which may have served as a kitchen with a small hearth. All these rooms open onto a central courtyard. We were successful in determining the layout of the site as well as the periods of occupation. Based on a preliminary assessment of the data, it appears that the caravan station was established in the first century B.C.E. as a thriving way station along the Nabataean trade route linking Petra and Gaza. While settlement at the site seems to have declined by the third century C.E., there is a clear reoccupation in the fourth century that continued into the fifth. In support of this preliminary phasing, the pottery finds include many fragments of piriform unguentaria, terra sigillata, and Cypriot Red Slip Ware. Six coins were also retrieved during the excavation. Three of the coins are Nabataean (one of Aretas IV and a veiled bust of his first wife, Huldu, and two of Aretas IV and his second wife, Shaqilat); one is of Augustus; another is of Domitian; and another, poorly preserved coin dates to the fourth century. Surface finds also suggest an Early Islamic presence at Khirbat Sufaysif, the nature of which remains unclear. The survey along the south bank of Wadi Musa between Jabal Tayyiba and Wadi Umm Ratam was also successful (the north-bank survey was completed in 2012). In addition to Wadi Musa itself, several tributary wadis were explored, which included, from west to east, Wadis Samra, Sufaysif, Hamra, Tegan, Atawla, and Umm Ratam. Surveying in this area allowed us to contextualize more broadly the excavation data derived from Khirbat Sufaysif. In all, we documented 351 new sites in the region. These sites range from isolated cairns, stone circles, and graves to more complex features such as multiroom structures and small settlements. We documented scores of sites and features indicative of an expansive agricultural regime in classical and late antiquity (fig. 15), which provide significant information about the rural economy of Petra. Among the recorded sites where we found artifacts, the dates range from the prehistoric to modern periods.

2014]

ARCHAEOLOGY IN JORDAN, 2012 AND 2013 SEASONS

659

Fig. 15. Expansive agricultural system in Wadi Musa.

petra sculptures Robert Wenning, Münster University, reports: The first season of a three-year project to document Petra’s ancient sculptures began in 2013. This joint project is led by R. Wenning of Münster University, T.M. Weber of the University of Jordan, and M. alMarahleh of the Department of Antiquities of Jordan, and it is sponsored by the German Research Association. The project is also supported by the directors of current Petra excavations. The project will document all of Petra’s Hellenistic/ Roman figural sculptures in stone, stucco, and bronze, whether they are identified in the field or in museums. This work will provide the Petra authorities with a complete inventory of sculptures to facilitate both further registration and future exhibition in the new Petra Museum. It will also permit development of a critical analysis and classification of the sculptures that will serve as a valuable resource for Nabataean research and related areas of study. The preliminary list of Petra sculptures, identified in both publications and field visits, includes 574 entries. While sculptures recovered from excavations are relatively well documented (excepting those of earlier excavations), other sculptural finds are not. As a result of this project, many previously unknown objects held in museum storerooms in Amman and Petra have been documented for the first time. We have also requested that foreign museums provide inventories of sculptures from Petra held in

© 2014 Archaeological Institute of America

their collections. Unlike Palmyrene sculptures, Nabataean sculptures are rarely found outside Jordan (ca. 25 items have been recorded to date).

petra: nabataean stone dressing and lithology Thomas R. Paradise, University of Arkansas, reports: During the height of the Nabataean kingdom, Nabataean stonemasonry, like the architecture, hydraulic engineering, and theology, shows signs of having been unique. Across ancient Nabataea, chisels, hammers, axes, and surveying devices have been found. As the crafts of masonry and stone dressing date back more than eight millennia, it is now fairly straightforward for researchers to identify the chisels and tools associated with particular types of stone dressings and marks. In examining chisel marks and dressings, together with characteristic features of Petra’s lithology, our project seeks to understand better the various tools that were used to dress, work, and sculpt Petra’s stones during the Nabataean and Roman periods. An examination of these relationships is crucial for understanding Petra’s trade economy as well as the community of craftsmen who created the city’s stone architecture during its heyday (50 B.C.E.–350 C.E.). More than 100 Nabataean/Roman-period dressedstone features and surfaces throughout Petra were identified, examined, and measured. Rock type was determined using microscopic and micrographic

660

G.J. CORBETT ET AL.

analyses, while relative hardness was determined using a Schmidt hammer (resiliometer). Various types of building stone were examined, including ash-Sharah limestone, Umm Ishrin and Disi sandstones, and the Precambrian basalts of Wadi Siyagh and Jabal Aswad. Basic stone-chisel techniques have changed little since the Classical period. Then and now, the shape of the chisel blade varies according to the desired style of dressing, although some rock types require specific types of chisels. There are four primary chisel types: point (moil), flat (straight-edged), rounded (roundel), and toothed (claw) (fig. 16). In addition, ancient masons used large, flat-headed, toothed hammers (called bush hammers) to dress hard, resistant rocks such as basalt and granite. It is likely that axes (doubleedged metal chisels attached to wooden handles) were also widely used (online fig. 15). In Petra, there is evidence that various chisel types were used, along with the mason’s axe and the bush hammer. For Petra’s softer sedimentary rocks, such as the Disi and Umm Ishrin sandstones (Schmidt hammer values R