2016. Mezad Zohar: A Medieval Fort Near the Dead Sea STRATA Bulletin of the Anglo-Israel Archaeological Society vol. 34: 125 -149

STRATA Bulletin of the Anglo-Israel Archaeological Society Volume 34 2016 The Anglo-Israel Archaeological Society 2nd loor, Supreme House 300 Regents Park Road London N3 2JX The full texts of articles within Strata are available online through Academic Search Premier (EBSCO). This periodical is indexed in the ATLA Religion Database®, published by the American Theological Library Association, 300 S. Wacker Dr., Suite 2100, Chicago IL 60606, Email: atla@atla.com; website: www.atla.com Cover: Beth Alpha synagogue mosaic, NASA image © 2016 The Anglo-Israel Archaeological Society, 2nd loor, Supreme House 300 Regents Park Road London N3 2JX ISSN Series 2042–7867 (Print) Typeset by DMilson, Switzerland Printed and bound in Great Britain by 4word Page & Print Production Ltd. Strata: Bulletin of the Anglo-Israel Archaeological Society Editor: David Milson Reviews Editor: Sandra Jacobs Editorial Advisory Board: Rupert Chapman, Shimon Dar, Yossi Garinkel, Shimon Gibson, Martin Goodman, Sean Kingsley, Amos Kloner, David Milson, Rachael Sparks Please send correspondence and books for review to: The Secretary The Anglo-Israel Archaeological Society 2nd loor, Supreme House 300 Regents Park Road London N3 2JX UK Please format material according to the ‘Notes for Contributors’ found at the back of this volume and submit articles to the editor electronically at: editor@aias.org.uk. Book reviews should be sent to the book reviews editor at strata.reviews@aias.org.uk Strata is published annually. To subscribe, please consult the Society’s website at www.aias.org.uk or use the form at the back of this volume. The Anglo-Israel Archaeological Society HONORARY OFFICERS AND COMMITTEE MEMBERS Honorary Chair Prof. Tessa Rajak Vice-President Prof. Martin Goodman, FBA Mr Mike Sommer Prof. H. G. M. Williamson, DD, FBA Vice-President (Israel) Prof. Amihai Mazar Prof. Ze’ev Weiss Hon. Secretary Dr Nick Slope† Hon. Treasurer Dr Paul Newham Committee Mrs. Barbara Barnett Prof. George Brooke Dr. Rupert Chapman III Dr. Irving Finkel Prof. Shimon Gibson Prof. Martin Goodman Dr. Sandra Jacobs Dr. Sean Kingsley Dr. Mark Merrony Dr. David Milson Mr. Anthony Rabin Dr. Stephen Rosenberg Dr. Rachael Sparks Dr. Guy Steibel Executive Secretary Mrs. Sheila Ford Contents Editorial 5 in MEMoriaM Nick Slope 9 Katharina StrEit The 6th Millennium Cal. BCE Wadi Rabah Culture: Further Excavations at Ein el-Jarba in the Jezreel Valley, Israel (2015–2016) 13 itaMar WEiSSbEin, YoSEf GarfinKEl, MichaEl G. haSEl and Martin G. KlinGbEil Goddesses from Canaanite Lachish 41 Jan GunnEWEG and Marta balla The Provenience of 7th–6th Century BCE Cult Vessels from the Iron Age II Fortress at ‘En Hazeva using Instrumental Neutron Activation Analysis (INAA) 57 EGon h.E. laSS Soil Flotation from the Persian Period at Ashkelon, Israel 73 YotaM tEppEr, Jonathan david and MatthEW J. adaMS The Roman VIth Legion Ferrata at Legio (el-Lajjun), Israel: Preliminary Report of the 2013 Excavation 91 tali EricKSon-Gini, dov nahliEli and robErt Kool Mezad Zohar: A Medieval Fort Near the Dead Sea 125 robErt Kool The Coins from the Medieval Fortress at Mezad Zohar 151 orit ShaMir and naaMa SuKEniK 13th Century CE Textiles from Mezad Zohar 155 YoSEf GarfinKEl The Decauville Light Train at Lachish (1933–1938) 165 Eliot braun Observations on the South Levantine EB1 and the Erani C Horizon: A Rejoinder to Gophna and Paz 191 Book Reviews 213 Books Received 245 Lecture Summaries 247 Reports from Jerusalem 251 Student Grant Reports 259 Notes for Contributors 261 7 Book Reviews 8 Claire Clivaz, Andrew Gregory and David Hamidović (eds.), Digital Humanities in Biblical, Early Jewish and Early Christian Studies. (James Aitken) 209 Erin Darby, Interpreting Judean Pillar Figurines: Gender and Empire in Judean Apotropaic Ritual. (Josef Mario Briffa) 212 Hanan Eshel† (Shani Tzoref and Barnea Levi Selavan, eds.), Exploring the Dead Sea Scrolls: Archaeology and Literature of the Qumran Caves. (George J. Brooke) 214 Eric C. Lapp, Sepphoris II: The Clay Oil Lamps of Ancient Sepphoris: Light Use and Regional Interactions. (Shimon Dar) 216 David M. Jacobson, Nikos Kokkinos (eds.), Judaea and Rome in Coins, 65 BCE - 135 CE. Papers Presented at the International Conference Hosted by Spink, 13th - 14th September 2010. (Yoav Farhi) 217 Reinhard G. Kratz, Historical and Biblical Israel: The History, Tradition, and Archives of Israel and Judah. (Lester L. Grabbe) 221 Jacobson, David M., Antioch and Jerusalem: The Seleucids and Maccabees in Coins. (David F. Graf) 224 David T. Sugimoto, ed., Transformation of a Goddess: Ishtar – Astarte – Aphrodite. (Sandra Jacobs) 226 Gil Gambash, Rome and Provincial Resistance. (Tessa Rajak) 231 Astrid Swenson and Peter Mandler (eds.), From Plunder to Preservation: Britain and the Heritage of Empire, c.1800–1940. (Michael Sommer) 233 Rachel Hachlili, Ancient Synagogues - Archaeology and Art: New Discoveries and Current Research. (Joan Taylor) 236 Strata: bullEtin of thE anGlo-iSraEl archaEolocial SociEtY 2016 voluME 34 Mezad Zohar: A Medieval Fort near the Dead Sea tali EricKSon-Gini, dov nahliEli and robErt Kool Israel Antiquities Authority Excavations at Mezad Zohar, a heavily eroded medieval fort west of the Dead Sea, have revealed that it was constructed in the later decades of the 12th century and occupied until the 14th century CE. During the Mamluk period, the fort was part of the ‘barid’ network between Cisjordan and southern Transjordan. Introduction In January 2004, Erickson-Gini and Nahlieli carried out soundings at Mezad Zohar fortress (Arabic: Qasr al-Zuweira) on behalf of the Israel Antiquities Authority (New Israel Grid 233317/562179; License A-4110/2004). The site is two kilometres west of the Dead Sea (Fig. 1) at the foot of the Naqb Zuweira pass, an ancient road leading west to Arad and from there towards Jerusalem. The fort was constructed on a steep chalk and marl hill, located at the confluence of Nahal Zohar and a tributary streambed (Fig. 2). Subsequently, the site has been heavily eroded, particularly along the eastern side of the hill with only a few walls remaining. The fort is made up of a lower area (Areas A and B) including a gate on the western wall and a number of rooms, and an upper area (Area C) approximately 20 m higher (Fig. 3) that appears to have been used as a signal tower. An examination of the architecture indicates at least two phases of construction. Previous Research In 1934, Mezad Zohar was first examined by Frank who dated it to the Roman period (Frank 1934: 201, 256). Two years later, a British inspector working for the Palestine Department of Antiquities arrived at the site on 12 November, walking from Umm Baghagh (‘En Boqeq). He correctly dated the structure to the medieval period, and provided notes describing the architecture and layout of the ‘castle’, as well as photographs and sketches. He identified medieval pottery, a ‘two-centered’ arch over the entrance to the lower fort near a cistern and mentioned rain water present in two of the open cisterns near the 125 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 1. Location map of Mezad Zohar. 126 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 2. Photo of Mezad Zohar, facing south (photo by G. Atkes). Fig. 3. Mezad Zohar, facing northeast. 127 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 4. General plan of Mezad Zohar. site.1 In 1967 Rothenberg’s Arabah survey erroneously reported that the site dated to the Roman period (Rothenberg 1967: 112). In Pringle’s gazetteer of secular buildings in the Crusader Kingdom of Jerusalem it appears (Pringle 1997:114, site P21; Qasr az-Zuwaira) as a possible Crusader site, noting that a coin of Amalric I found there raises the possibility that the earlier fort was reoccupied by the Franks in the 12th century to protect the Hebron-Kerak road (see Kool, below). However, Pringle was skeptical about the dating due to a lack of supporting evidence from controlled excavations. Petersen also included the site in his Gazetteer of Buildings in Muslim Palestine (Part 1), (2001a: 252–254; Pls. 275–277). 128 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 5. Photograph of the gate on the interior of Mezad Zohar, facing west, from 1936 (courtesy of the Israel Antiquities Authority, Archive Dept.). The Architecture of the Fort (Fig. 5) The Lower Level The lower level of the fort (Areas A and B) was originally constructed as a rectangular structure abutting the north hillside. The western side is best preserved near the gate (Fig. 5). A large section between Area A on the west and Area B on the east was heavily eroded with few remains (Fig. 6). Merely 6 m of the long exterior wall (W6) bounding the southern side of the fort still remains. Sections of this wall have survived eight to ten courses high in the southwest corner. The lower courses of a 129 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 6. Photograph of Area B, the slight remains of the eastern edge of the fort, facing northeast. support wall (W7), also 6.0 m in length, were found just outside the corner of W5 and W6. Some sections of this wall collapsed into the Nahal Zohar wadi bed. These exterior walls are nearly a metre in width. Apparently, an interior wall (W3) ran parallel to the southern wall with some small sections still visible. The fort was built with dressed dolomite stones that average 0.30 × 0.30 × 0.35 m in size with somewhat larger stones of a higher quality that appear in the gate, possibly relecting a secondary period of construction. These stones must have been brought to the site from quarries in the limestone/dolomite formations nearby. Several bossed stones with smoothed edges were incorporated into western wall next to the entrance to the lower level (Area A). These stones probably came from Late Hellenistic to Early Roman buildings located a kilometre to the east. The quality of the construction is rather simple, with large gaps (up to 0.23 m) between stones. Beyond the gate, much of the construction is characterized by the use of both regular and irregularly cut stones with rows of galleting, small stones used as inill (Fig. 7). This technique was evident throughout the site, including the built plinths in Area B. The walls of the lower level appear to have been lined with plaster both internally and externally in a later phase. Four shallow niches were carved into the hillside along the northern face of the lower level. Their function is unclear and further excavation is required in this area in order to determine their use. 130 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 7. Photograph facing northwest showing wall construction north of the gate with irregularly shaped stones and galletting. The Gate The photograph of the gate shows that it had an intact arch in the medieval pointed “two-centred” style. By 1956, photographs show that the arch has disappeared, as can be seen at present (Fig. 8). The gate is 1.80 m wide and preserved to a height of eight courses. The jambs supporting the arch have survived to a height of six-seven courses. Several stones forming a low platform have been placed outside of the gate in order to provide a step up into the structure. This platform apparently covers an earlier step, but this was not investigated during the excavation. The interior of the gate opens into a space, 3.5 m in width, bounded on either side with low, terrace walls. 131 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 8. The gate and the western exterior W5 of the fort in Area A, facing east. The South-western Room The southwestern room (5.8 × 6.5 m) of the lower level appears to have been used as a kind of guard room. The loor is two metres lower than that of the entrance level. Four steps leads downwards to this room where there are two vertical loopholes (arrow slits) in the southern wall. The western loophole is 1.97 deep, 0.78 m high and 1.73 wide in the interior of the room and narrows on the exterior to 0.55 m The eastern loophole is above the lowest step and slants horizontally it is 2.00 m deep, 0.85 m high and 1.75 wide in the interior. Two phases are present, as windows in the southern wall from the earliest phase were covered when the wall was lined. The Cistern A bell-shaped cistern exists ten meters northwest of the gate on the exterior of the fort, and was not excavated. A line of stones may have formed some kind of wall or channel leading into the cistern from the south. The original opening into the cistern has collapsed, leaving an opening 1.5 m in diameter. The interior of the cistern is covered with a well-preserved layer of white plaster. The Plinths Three plinths, probably for a rope bridge, are in Area B, on the far eastern end of the fort below the entrance to the upper level, on a steep cliff overlooking Nahal 132 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 9. The oven in Area B facing north. Zohar. The tallest surviving plinth is 1.5 × 4 × 2.80 m high. It was carved out of bedrock along the cliff edge. A two-metre deep niche was carved into the side facing the intermediate plinth below. The second plinth is just over six metres away, below the upper one. This one was constructed of roughly worked stones and lines of galleting and it survives to a height of 2.38 m. This plinth is ive metres from a third plinth located next to the remains of a medieval oven. The third plinth (1.80 × 1.50 m.) survives to a height of 2.50 m. The Oven The oven was built close to the third plinth and it is bounded on the east by a wall (W8). It measures 2.30 × 3.0 m in size and has survived to a height of six courses along its eastern face and three to four courses on its southern face. The opening to the oven faces south (Fig. 9). The Upper Level Although the upper level, 20 m above the lower level is largely intact, it is dificult to access. Only a small section of the original stairway and a wall to the east has survived. This wall (ca. 3.5 m long) abuts the northeast side of the hilltop, leading to a lat, enclosed platform (Fig. 10). The platform is irregularly shaped, measuring ca. 8 × 8 m in size. It is supported by high revetment walls lining the curving hillside. On its western side, the outer stones have largely disappeared. 133 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 10. The upper level of the fort and the remains of a staircase, facing north. The remains of two other revetment walls to the northwest and the northeast are visible from above. The platform was covered with an easily discerned layer of thick ash. The top of the wall enclosing the platform are all uniform in height and the platform appears to have functioned as a signal tower. Access to the upper level may have been provided by way of a rope bridge supported by a row of three plinths (described above) located below and southeast of the upper level in Area B. A fourth plinth may have been located on the opposite bank of the wadi bed of Nahal Zohar facing Area B. The Excavation Area A Most of the soundings were in Area A, along the southern side of the fort. The well preserved walls of the fort (W5, W6) include the entrance and a cistern outside of the fort, to the west. Three soundings at intervals along the interior of the southern wall of the fort were excavated. These probes showed that the builders used a ill made of ground local clayish soil, producing a series of levelled terraces that extended down the hillside. The loors of the rooms built on these terraces were constructed over the ill. The area west of the gate was not quarried. A thin layer 134 MEzad zohar: a MEdiEval fort nEar thE dEad SEa of soil covered the sloping bedrock. The southern side of this probe was bound by a low terrace wall (W1). Just below topsoil, a campire, several pieces of rope and other organic material were found together with pottery sherds dated to the late 12th and the 13th centuries AD. In particular, we uncovered sections of a bowl with slip-painted decoration (Fig. 11:1; Avissar and Stern 2005:19, Types I.1.6.1–2, Fig. 7: 1–8) and a bowl base with green glaze (Fig. 11:2). A second probe next to the entrance on the eastern side was bounded on the east and north with a low partition wall (W2). Large fragments of an undated handmade storage jar (not illustrated) were found under the corner of this partition wall. A layer of soil with ash extended down to a layer of light colored ill. This ashy layer contained several pottery sherds as well as organic material. A third probe below one of the windows by the southern side of the fort uncovered a wall (W4). A stairway consisting of four steps was found along the western wall of the fort extending downwards (Fig. 13). In order to obtain a better view of this structure the probe was extended 1 m further east (L902, 903). Throughout the probe a layer of soil mixed with ash was evident from the surface down to the ill level. This layer contained numerous pottery sherds, mainly geometric-painted handmade pottery (Fig. 12.11; Avissar and Stern 2005:88, 113, Types II.1.4.2, II.4.4, Figs. 38:6–10; 47, 48), examples of soft-paste wares (Figs. 11:6; 12:12, 12:14; Avissar and Stern 2005:25–29, Type I.2) and fragments of Cypriot slippainted bowls (Fig. 11: 2, 3; Avissar and Stern 2005:58, Type I.8.1, Fig. 23). These pottery types date generally from the end of the 12th to the 13th centuries. However, the geometric painted handmade wares have a longer date range, extending to later centuries as well, while the Cypriot Slip-painted bowls have a more restricted date, only in the 13th century.2 Well-preserved textiles and organic matter were also found. Sherds of molasses jars were found in this and other probes in Area A (Fig. 11:9; Avissar and Stern 2005:103, 104, Types II.3.1.5–6, Fig. 43). This type of handless jar was typically used in the industrial process of producing sugar from sugar cane. The liquid molasses was drained to the jar while the sugar loaf dried and crystallized in a sugar pot (Avissar and Stern 2005:86, type II.1.3, Fig. 37; Stern 2001: 279–281, Fig. 19). Molasses was possibly marketed in this type of vessel, as it has been found in sites that were not sugar production sites. Large quantities of molasses jars were found in a store room at Tall Hisbān (Walker and LaBianca 2003:450– 451, Figs. 5, 9). Walker suggested that the presence of the molasses jars indicated that Hisbān was a transport point for sugar in the 14th century (Walker 2004:133). Furthermore, nine fragments of molasses jars with no evidence for sugar production were found at Tel Yoqne’am (Tall Qaymun) in the northern part of Israel (Avissar 1996:155–156, Type 16, Fig. XIII.123:1) and molasses jars in small quantities were also found at few sites in Jordan, such as: Mugharat el Wardeh in the ‘Ajlun 135 tali EricKSon-Gini, dov nahliEli and robErt Kool 1. 3. 2. 4. 5. 6. 7. 8. 9. Fig. 11. Pottery. 136 MEzad zohar: a MEdiEval fort nEar thE dEad SEa No. Vessel Basket Locus Description 1 Bowl 4 L301 Reddish brown ware 5YR4/4; occasional large white inclusions and numerous tiny white inclusions; black 5YR2.5/1 to dark gray 5Y4/1 glaze on interior rim; bright green glaze towards center. Parallels 2 Bowl 1 L101 Dark gray ware 7.5YRN3/; Cypriot slipped painted ware; very dark grayish brown Yoqneam (Avissar 2005: Fig. glaze 10YR3/2; olive 2.15.5). yellow glazed decoration 5Y6/6. 3 Bowl 5 L401 Red ware 2.5YR4/8; occasional Cypriot slip-painted ware; Acco (Stern 1997: Fig. 8:71); medium white inclusions; Yoqne’am (Avissar 2005: Color reddish brown glaze on interior 5YR4/4; pale plate 1:1). yellow glaze swirled decoration, 5Y7/4. 4 C o o k i n g 11 pot L602 Red ware 2.5YR4/6; numerous Yoqne’am, Avissar’s CP-Type 7 tiny white inclusions; (Avissar 2005: Fig. 2.18:1; traces of dark brown glaze. Avissar and Stern 2005: Fig. 39.2). Dated to the second half of the 12th c. and irst half of the 13th c. 5 Cooking 1 pot L101 Red ware 2.YR4/8; occasional Yoqne’am, Avissar’s CP-Type 7 tiny white inclusions; dark (Avissar 2005: Fig. 2.18:1; reddish brown glaze on Avissar and Stern 2005: Fig. shoulder 5YR2.5/2. 39.2; Pl. XXVI:1). Dated to the second half of the 12th c. and irst half of the 13th c. 6 Ring base 9 L902 Very pale brown 10YR8/3; Avissar and Stern’s Type I.2.2, softoccasional medium to paste monochrome glazed ware large white inclusions; (Avissar and Stern 2005: Fig. turquoise glaze on interior 1–3); Dated 12th-13th c. and exterior. 7 Jug base 1 L101 Light brownish gray ware 2.5Y6/2; charred exterior; traces of pale green glaze on base. 8 Flask 9 L902 Light gray 2.5Y7/2; numerous Acco (Stern 1997: 4:18); Dated 13thtiny to medium white 14th c. inclusions; light yellowish brown slip 2.5Y6/4. 9 Molasses 16 jar L903 Very pale brown 10YR7/4; Avissar and Stern’s molasses jar type dark gray inclusions; II.3.1 (Avissar and Stern 2005: light yellowish brown slip Fig. 43.10). Dated 14th -15th c. 10YR6/4. 137 tali EricKSon-Gini, dov nahliEli and robErt Kool 10. 11. 12. 14. 13. Fig. 12. Painted pottery. 10 Painted jug 14 L1202 11 Sherds 16 L903 12 Sherds 4 L301 13 Sherds 8 L802 14 Sherds 4 L301 138 Yellowish red ware 5YR5/6; Avissar and Stern’s handmade small white inclusions; jug with geometric painted light brown slip 7.5YR6/4; decoration type II.4.4 (Avissar gray 10YR5/1 to dark gray and Stern 2005: Fig. 47: 4–5, 7). 10YR4/1 deco. Dated from the 12th c. throughout the Mamluk period. MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 13. The interior stairs abutting W5 in Area A. district (Coughenour 1976: 72–74, Pl. XXXII) and from er-Rabbah and Qumeir on the Kerak Plateau (Miller 1991:65–66; sites 94, 108, and 215; Brown 1991:235, pl. 8:446, 447). These jars, with no evidence of sugar production, strengthen the assumption that the molasses jar fragments found at Mezad Zohar indicate transportation of molasses and not production. Area B Directly in front of the stairway leading into the keep in Area B are three tall plinths, one carved from bedrock and the others constructed using stones. These may have been used to support a rope or wooden bridge to facilitate access between the lower and upper levels. A probe (L401) between the upper plinth and the hillside revealed collapse from the stair and entrance into the keep. This material appears to have rolled down from the entranceway of the keep. The collapse layer sealed a thick layer of ash, organic material and some textiles. Several pieces of wood found in this area may have been used as cups or utensils. No building remains were found in the probe. Soundings were made around the lower plinths where a medieval oven was uncovered. Although its domed roof had collapsed, the southern half of the oven was excavated (Fig. 9; L701). A few animal bones and unidentiiable pottery sherds were 139 tali EricKSon-Gini, dov nahliEli and robErt Kool found in the rubble above the loor of the oven. Near the oven is a low stone wall, one row high (W8). The area in front of the oven (L601, L602) contained a thick layer of ash with debris from the oven. Dating the ceramic material indicates the earliest phase of the fort from the late 12th or early 13th century. Pottery sherds included globular cooking pots with dripped brown glaze dating from the second half of the 12th century to the early 13th century (Fig. 11: 4, 5; Avissar and Stern 2005:91, Type II.2.1.2, Fig. 39:2). Small pieces of wood and wood dust discovered in the far east corner indicate that this particular area was used to store irewood. This type of oven, which has a low domed roof constructed over a square platform, is a well-known medieval type in the Levant. Baking ovens with raised platforms and domed roof are a known European type (Yehuda 2012:54). Parallels have been found at Vadum Jacob (Mezad Ataret, Qasr al-‘Atra), ‘Atlit, Jaffa, Nebi Samwil, Crac des Chevaliers and Kerak (Boas 1999:77). Additional probes near the intermediate plinth uncovered two bronze coins close to the surface on the slope. These were the only coins discovered anywhere in the fort. The irst is a coin of the Ayyubid al-‘Ādil Sayf al-Dīn (d. 1218), dating to 1202 CE.3 The second is a coin of the Mamluk Sultan Baybars (1260–77 CE).4 Nearby, two adult skeletons (L1201) were found less than half a metre below the steeply sloping surface. The skeletons were partially excavated and subsequently covered and left in situ. Well-preserved textiles and part of a leather pouch were found near the remains. Along the northern face of the plinth (L1202) a complete handmade jug with painted brown geometric decoration was discovered in an inverted position (Fig. 12:10). Area C In the area of the keep located over 20 m above the fort (Area C), a 2 × 2.5 m probe was excavated in the middle of the platform. A layer of ashy soil extended about half a metre in depth to a ill layer. This area was probably used as a signal tower, as a thick layer of ash and organic material was found all over the platform. Pottery sherds dating to the 13th century CE with geometric painted decoration similar to those discovered in Areas A and B were discovered.5 Investigations Outside of the Fort The area beyond the southwest exterior of the fort appears to have been used as a midden (L1100, L1101). No diagnostic pottery sherds were found on the surface of the rocky slope in this area, where there is a thin layer of soil over a hard, sterile surface. Another probe (L1000, L1001) east of the southwestern corner of the fort along W7 was sterile. Roi Porat and Uri Davidovitz carried out an investigation of the caves in the vicinity of the fort and the fort itself using a metal detector. In the small, man- 140 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Fig. 14. The guard cave above the road leading past the fort, facing southeast. made cave southeast of the fort, which overlooks the road leading eastward (Fig. 14), six Mamluk bronze coins and some well-preserved textiles were discovered approximately 0.40 m below the loor surface. The cave is open to east and the west and a round seat was carved into the loor. This room probably served as a watchtower overlooking the bend in the road east of the fort. The cache of coins discovered here date to the 14th century.6 Of these, one coin dates to the second reign of al-Ṣāliḥ Ṣalāḥ al-Dīn Ḥājjī II Hajji (1389– 1390 CE)7 and two coins were issued during the second reign of the Circassian Mamluk Sultan, al-Ẓāhir Sayf al-Dīn Barqūq (1390–1399 CE).8 The Water Supply At least three dams constructed in narrow passages collected water from the streambeds below the fort. Two dams are still extant north and northwest of the fort and traces of a third, together with evidence of hydraulic plaster, have been found southwest of the fort. One pool still collects water during loods. Water collected in loods in these pools was stored in the cistern located next to the entrance of the fort. 141 tali EricKSon-Gini, dov nahliEli and robErt Kool Fig. 15. Map of sites along the Mamluk ‘Barid’ road between Gaza – Hebron and Transjordan. The Gaza – Transjordan Road (Dov Nahlieli) In pre-modern times, Mezad Zohar was on the main artery between Gaza and southern Transjordan (Fig. 15). One branch of the road led eastward from Mezad Zohar. This branch may have led to anchorages along the southern shore of the Dead Sea. There is evidence for a limited amount of shipping on the lake during the Crusader period, and it was an important source of salt and bitumen (Mayer 1987, 201).9 The road was the primary route between Transjordan and the Mediterranean, by way of Ghor es-Sai (Zoar) southeast of the Dead Sea and the regions of southern Moab and Edom. Westward the road led to Hebron and Gaza, passing through a series of waypoints in southern Judaea and the western Negev. This road was probably used by Baldwin I of Jerusalem on his expedition into Edom when he stopped Muslim attempts to construct a fort at Wadi Musa in 1107 CE (ibid). In this period Baldwin I initiated the construction of the important fortress of Shaubak in Transjordan.10 Furthermore, Crusader remains have been discovered along the route in southern Judaea. These include a Crusader tower located at Khirbat al-Karmil (Horvat Carmel). Prawer noted in his discussion concerning roads in the Latin Kingdom of Jerusalem that a fortiied station was located at Carmel (Kh. al-Karmil) along the road leading from Hebron to Segor (Palmarea – Zoar) at the southern tip of the Dead Sea. Crusader sources refer to an ancient water reservoir that was still functioning at Carmel (ibid., n. 14). In the Mamluk period, the Zuweira/Zohar road appears to have been used by the imperial post (barīd), (Fig.14). For instance, in the Haram archive a sawwaq (imperial courier) ‘of Kerak and Gaza’ via el-Khalil (Hebron) is mentioned (Richards 1987: 206, n. 10). According to Qalqashandi “in regard to the road to Kerak: and from Gaza to Mulakas which is the centre of the ‘barid’, and from there to the city (balad) of el-Khalil (Hebron), may peace be upon it, and later from there to el-Janba and from there to el-Saiyah and later from there to Kerak.” (al142 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Qalqashandi 1913: 379).11 A number of sites described by Qalqashandi as Mulakas may be identiied. The most likely of these is the site of Khirbet Umm Lāqis (H. Leqesh, New Israel Grid 17038/60902) near the modern community of Sade David in the western Negev. Sir Flinders Petrie partially excavated the site in 1890 (Lamdan, et al., 1977: 188) and subsequent surveys carried out under the direction of Lamdan found a site covering ifteen dunams, with pottery from the Crusader and Mamluk periods (New Israel Grid 17065/60900; ibid). According to a list of villages in belonging to the county of Ascalon, Umm Lāqis appears under the name Malakus in the year 1256 CE (Prawer 1958: 235). The site of Janba is that of Khirbet Janba (or Jinba, New Israel Grid 21330/58557) located east of Mezudat Yehuda and as-Safiya refers to es-Sai southeast of the Dead Sea. The site of Mezad Zohar (az-Zuweira) appears in an account of the barid system by al-Tahari. Al-Tahari provides a detailed description of the barid road between Egypt and Gaza by way of North Sinai and the various roads that radiated from Gaza: “and the road to Kerak from Gaza to Balakas and then to Hebron and then to Janba and then to az-Zuweir and then to as-Saiyeh and then to el-Hafer and then to Kerak.” (al-Zahiri 1894: 119) It should be noted that the barid usually took the shortest and most direct route from one point to another and that this particular route was lengthened since it bypassed the Beer Sheba basin and the northern Negev. Prawer notes the absence of settlement south of Hebron in this period: “the most striking phenomenon is the utter desolation of the semi-arid areas in the southern part of the country,” (Prawer 1972a: 354). It appears that the reason for this may have been a desire to pass between larger, more central towns on the one hand, and on the other hand, this may point to security concerns in the periphery area of the Beer Sheba basin, in spite of the fact that the Mamluk regime generally ensured the safety of roads under its jurisdiction. Here, the Mamluks may not have been able to control Bedouin tribes completely and the roads leading through this area may have been too dangerous to travel. Conclusions The indings of the archaeological investigation of the site of Mezad Zohar, which include the numismatic, ceramic and architectural evidence, indicate that it was established late in the 12th century CE, possibly by the Ayyubids, and it continued to be occupied in the 13th–14th centuries during the Mamluk administration. The possibility that the fort was originally constructed by the Franks is countered by narrow chronological constraints: it is unlikely that it would have been built anytime before 1142 CE when the construction of the Kerak fortress began whereas the Franks lost control of the area in wake of the Battle of Hittin in 1187 CE (M. Sinibaldi, pers. comm.). Notably, the new borders established in the Treaty of 143 tali EricKSon-Gini, dov nahliEli and robErt Kool 1240–1241 CE did not include the area of Mezad Zohar (Prawer 1972a: 31; 1975: 287; 284). In Petersen’s discussion of the site of Mezad Zohar (Qasr az-Zuwayr), he notes that it bears similarities to the manner in which some Crusader castles are built upon spurs, taking advantage of the natural topography for defensive purposes (2001b: 393). However, he also notes the similarity to Assassin castles built in remote locations (ibid). The fort contains two or more architectural phases and features that can be attributed to the Crusader and the Ayyubid periods. The ceramic evidence from the 2004 excavation of the site of Mezad Zohar is not deinitive enough to determine whether the site was irst occupied by the Franks or the Ayyubids and the assemblage is largely made up of types that were in use in the 13th century. The fort was probably constructed in order to protect the important Zohar Pass and to serve as a station on the main road between the Cisjordan and southern Transjordan. The platform of the keep (the upper level, Area C) appears to have been used a signal tower that would have been visible on the Kerak plain east of the Dead Sea. This plain, at an elevation of over 1000 m above sea level on the southern edge of the Dead Sea, is clearly visible from both levels of the site. The lack of collapsed building stones between the plinths on the eastern side of the fort in Area B point to the possibility that the keep was accessed by way of a rope or wooden bridge strung across the three plinths and across Nahal Zohar. These plinths are located more or less in a line under the stairway and entrance into the tower on the north side of the hilltop. The streambed of Nahal Zohar apparently served as a kind of dry ditch and part of the fort’s defences. The presence of deep ditches around Crusader fortresses prevented an enemy’s use of siege equipment such as a ram or towers against the walls (Prawer 1972b: 122). Timber or rope bridges were often used to span ditches of this sort, enabling the Crusader defenders to burn the bridge during an attack (ibid. 123). The choice of a steep hill in the dramatic terrain of a dry riverbed, recalls the location of the Crusader fort of al-Wu’ayra near Petra. Like al-Wu’ayra, Mezad Zohar is seriously eroded with both forts displaying signs of features carved into the bedrock (Vannini and Tonghini 1997: 383, n. 46). Pottery dated to the second half of the 12th and early 13th centuries was found in association with the oven located on the eastern side of the site in Area B. Unlike al-Wu’ayra in southern Jordan, Mezad Zohar was located on an important artery, the main road between the Ghor es-Sai (Zoar)/Transjordan and the Mediterranean coast. The domed baking oven with a platform in this area points to European inluences. According to Yehuda, this type was commonly found in private houses and bakeries in medieval Europe (Yehuda 2012: 54). The numismatic and ceramic evidence indicates that it was probably re-garrisoned by 144 MEzad zohar: a MEdiEval fort nEar thE dEad SEa the Ayyubids. This may have taken place following the general abandonment of fortiications in Transjordan by the Crusaders in wake of their defeat in the Battle of Hittin in 1187 CE, including the fall of the fortresses at Kerak and Shaubak to Saladin in 1188 and 1189. In the 13th and 14th centuries the fort was quite likely part of the Mamluk barid, i.e., the imperial postal system established by Sultan Baybars due to its location on the Zohar (Zuweira) Pass. In this period, renovations were apparently introduced, including the lining of the walls. Both Faucherre and Sinibaldi point out that the encasing of older castles in new walls is a well-documented building technique of the Mamluk period (pers. comm.). Sinibaldi also notes that the presence of the cistern in an unprotected position on the outside of the fort suggests that it was constructed after the Crusader period, probably in the Mamluk period, whereas cisterns are usually an internal element in Crusader-period forts. Most of the pottery at the site appears to date to the latest occupation there in the 13th century. This includes fragments of molasses jars, Soft-Paste ware and particularly rather large quantities of handmade vessels with geometric painted decoration as well as some fragments of pottery originating in Cyprus. The molasses jars discovered at the site indicate that the Zohar pass and route was used to transport produce from the Ghor es-Sai area to the Mediterranean coast for further export by sea. The presence of the Cypriot slip painted ware dating to the 13th century indicates that other goods coming from the Mediterranean basin were also traded along this road to sites in Trans-Jordan.12 Excavations and surveys in the region of Ghor es-Sai, in the site of Zoar(a) / Zughar or Sughar, have uncovered kiln wasters of sugar molds and other vessels related to the production of sugar from Ayyubid and Mamluk periods (Photos-Jones et al: 2002, Figs. 17–18; p. 612). The production of cane sugar in the Jordan Valley, including substantial investment in irrigation systems, appears to have been heavily promoted by the Ayyubids and the Mamluks between the 12th and 15th centuries (ibid. 593).13 Investigations at the site of Ghor es-Sai in 2008 and 2009 have revealed the presence of sugar cane processing with at least three presses that operated between the 12th and 15th centuries (Politis 2013). It was a particularly important cash crop in the 14th century and Mamluk rulers held a virtual monopoly on its production and marketing (Walker 1999: 213; 2003:258–259). Mezad Zohar appears to have been abandoned long before the Ottoman period. Local Beduin claim that the site was used by robbers to ambush travelers. The human remains buried next to the plinth in Area B were obviously interred sometime after the abandonment of the site and may have been victims of banditry. 145 tali EricKSon-Gini, dov nahliEli and robErt Kool Acknowledgements The excavation was funded by National Water Company, Mekarot, as part of their drilling project near the site. The work was carried out under the direction of Tali Erickson-Gini and Dov Nahlieli on behalf of the Israel Antiquities Authority. In addition to a small number of hired workers, volunteers from the Blossoming Rose Organization also participated. Other volunteers include Gilad Nahlieli who aided in excavating Area C and Roi Porat and Uri Davidovich who checked the site and the surrounding area twice with metal detectors. This work produced the only coins found in and around the site. We are sincerely grateful for Edna Stern’s contribution towards the identiication and discussion of the ceramic inds. The pottery was drawn by Irina Libsky. The coins were examined by Ariel Berman and the inal report of the coins was prepared by Robert Kool. Donald Zvi Ariel provided photographs of the coins. Nicolas Faucherre visited the site with Haim Barbé (Israel Antiquities Authority) in September 2007 and has kindly provided his observations concerning the architecture of the site. Micaela Sinibaldi (Cardiff University) visited the site in 2010 and viewed the inds and also provided her observations concerning the historical setting and the archaeological context of the architecture and the pottery. We would like to thank each of them, as well as the anonymous referees of this paper, for their valuable suggestions and contributions. Notes 1 Other notations on the inspector’s report include the following: Map Ref. XVI.183.062 – Qasr ez Zuweira. Potsherds in Palestine Antiquities Museum, sherd coll. No. 545. Many thanks to Arieh Rochman-Halperin for providing us with the material in the Israel Antiquities Authority archive in the Rockefeller Museum in Jerusalem. 2 Stern (2012) reviewed this type of imported pottery, while a recent study of the date and distribution of geometric-painted handmade wares can be found in Gabriele et al. (2014). 3 IAA No. 108961. 4 D. Ariel, pers. comm. 5 The ceramic material was examined and identiied by Miriam Avissar† and Edna Stern on behalf of the Excavation and Survey Department of the Israel Antiquities Authority. 6 IAA Nos. 108955–108960. 7 IAA No. 108955. 8 IAA Nos. 109857 and 109858. 9 Rey proposed that the site of Mezad Zohar/Zuweira, as well as Kh. al-Karmil and Samu, formed a defensive line along the southern border of the kingdom of Jerusalem against Egypt (1871:4, 104; cf. Pringle 2013: 227). 10 See Mayer 1987, 199–200; Pringle 1997, 59–60, site 124 (Kerak), 75–76, site 157 (Montreal-Shaubak); 2003, 678. 146 MEzad zohar: a MEdiEval fort nEar thE dEad SEa 11 According to Gaudefroy-Demombynes’ translation Mulakas could be described as a mere postal station (barid) along the route (1923: 242). 12 Pottery may represent other traded goods mainly because it fulils two important conditions: it survives well in the archaeological record, and its provenience can often be traced using typology and analytical studies. Although pottery was not the most important product traded, it is representative of a wide range of medium-priced bulk products such as foodstuffs, cloth, glass or metals that usually do not survive in the archaeological record, or whose provenience is harder to trace (pers. comm. from E.J. Stern). 13 For examples of sugar production in the archaeological record, see: Abu Dalu 1995; Stern 1999; Stern 2001; Strange-Burke 2004. References Primary Sources al-Zahiri, Khalil b. Shahin. (1894) Kitab Zabdat Kashf al Mamlik. Paris. Qalqashandi, Abi al-Abbas Ahmad (d. 823/1418) (1913) Subh al-A’sha i Sina’at al-Insha. 14 vols. (Cairo: Al-Matba’a al-Amiriyya). Secondary Sources Abu Dalu, R., (1975). ‘The Technology of Sugar Mills in the Jordan Valley during the Islamic Periods’. Pp.37–48 in Studies in the History and Archaeology of Jordan V. (Amman, in Arabic). Avissar, M., (1996). ‘The Medieval Pottery’. Pp. 75–172 in A. Ben-Tor, M. Avissar and Y. Portugali. Yoqne‘am I: The Late Periods (Jerusalem). Avissar, M., and Stern, E. J., (2005). Pottery of the Crusader and Mamluk Periods in Israel. (Jerusalem). Boas, A.J., (1999). Crusader Archaeology. The Material Culture of the Latin East. (London). Brown, R.M., (1991). ‘Ceramics from the Kerak Plateau’. Pp. 169–280 in Miller M.J. ed. Archeological Survey of the Kerak Plateau. (Atlanta). Coughenour, R.A., (1976). ‘Preliminary Report on the Exploration and Excavation of Mugharat el Warda and Abu Thawab’. Annual of the Department of Antiquities of Jordan 21:71–78. Frank, F., (1934). ‘Aus der Arabia’. ZDPV 57:191–280. Gabriele, R.S., Ben-Shlomo, D. and Walker, B.J., (2014). ‘Production and Distribution of Hand-Made Geometric-Painted (HMGP) and Plain Hand-Made Wares of the Mamluk Period: A Case Study from Northern Israel, Jerusalem and Tall Hisban’. Journal of Islamic Archaeology 1/2: 193–229 Lamdan, M., et al., (1977). An Archaeological – Prehistoric Survey in Nahal Shikma. (Sha’ar Nanegev.) (Hebrew). Mayer, H.E., (1987). ‘The Crusader Lordship of Kerak and Shaubak. Some Preliminary Remarks’. Pp. 199–203 in A. Hadidi (ed.). Studies in the History and Archaeology of Jordan III. (Amman). 147 tali EricKSon-Gini, dov nahliEli and robErt Kool Miller, M.J., (1991). Archeological Survey of the Kerak Plateau. (Atlanta). Petersen, A., (2001a). A Gazetteer of Medieval and Ottoman Buildings in Muslim Palestine Part 1. (Oxford). Petersen, A. (2001b). Two Medieval Castles and Their Position in the Military Architecture of Muslim Palestine, Pp. 383–406 in U. Vermeulen and J. Van Steenbergen, (eds.), Egypt and Syria in the Fatimid, Ayyubid and Mamluk Eras. (Leuven). Photos-Jones, E., et al., (2002). The Sugar Industry in the Southern Jordan Valley. Annual of the Department of Antiquities of Jordan 46, 591–614. Politis, K.D., (2013). ‘The Sugar Industry in the Ghawr as-Sai, Jordan’. Pp. 467–480 in A. Hadidi (ed.) Studies in the History and Archaeology of Jordan, XI. (Amman). Prawer, J., (1958). The ‘City and Duchy of Ascalon’ in the Crusader Period. Eretz Israel 5: 224–237. Jerusalem. (Hebrew). English Summary. 96*–97*. Prawer, J., (1972a). The Latin Kingdom of Jerusalem. European Colonialism in the Middle Ages. Weidenfeld and Nicolson, London. Prawer, J., (1972b). The World of the Crusaders. (New York). Prawer, J., (1975). Historie du Royaume Latin de Jerusalem, Vol. II. (Paris). Pringle, D., (1997). Secular Buildings in the Crusader Kingdom of Jerusalem. (Cambridge). Pringle, D., (2003). Crusader Castles in Jordan. Pp. 677–684 in B. MacDonald, R. Adams and P. Bienkowski (eds.) The Archaeology of Jordan. (Shefield). Pringle, D., (2013). ‘Castles and Frontiers in the Latin East’, Pp. 227–240 in K.J. Stringer and A. Jotischky, (eds.) Norman Expansion: Connections, Continuities and Contrasts. (Farnham). Richards, D.S., (1987). ‘The Mamluk Barid: Some Evidence from the Haram Documents’. Pp. 205–209 in A. Hadidi (ed.), Studies in the History and Archaeology of Jordan III. (Amman). Rey, E.G., (1871). Étude sur les monuments de l’architecture militaire des croisés en Syrie et dans l’île de Chypre. (Paris). Rothenberg, B., (1967). Negev. Archaeology in the Negev and the Arabah. Masada Publishers, (Givatayim – Ramat Gan; Hebrew). Stern, E.J., (1999). The Sugar Industry in Palestine during the Crusader, Ayyubid and Mamluk Periods in Light of the Archeological Finds. M.A. thesis, Hebrew University. Jerusalem (Hebrew; English summary). Stern, E.J., (2001). ‘The Excavations at Lower Horbat Manot: A Medieval Sugar-Production Site’. ‘Atiqot 42: 277–308. Stern, E.J., (2012). ‘Akko I. The 1991–1998 Excavations. The Crusader-Period Pottery. Parts 1 and 2. (IAA Reports 52). (Jerusalem). Strange-Burke, K., (2004). ‘A Note on Archaeological Evidence for Sugar Production in the Middle Islamic Periods in Bilād al-Shām’. Mamlūk Studies Review 8/2, 109–118. Vannini, G., and Tonghini, C., (1997). ‘Mediaeval Petra. The Stratigraphic Evidence from Recent Archaeological Excavations at al-Wu’ayra’. Pp. 371–384 in A. Hadidi (ed.) Studies in the History and Archaeology of Jordan VI. (Amman). 148 MEzad zohar: a MEdiEval fort nEar thE dEad SEa Walker, B.J., (1999). ‘Militarization to Nomadization: the Middle and Late Islamic Periods’. Near Eastern Archaeology 62/4: 202–232. Walker, B.J., (2003). Mamluk Investment in Southern Bilād al-Shām in the Eighth/Fourteenth Century: the Case of Hisbān, Journal of Near Eastern Studies 62: 241–261 Walker, B.J., (2004). Mamluk Investment in Transjordan: a “Boom and Bust” Economy. Mamlūk Studies Review 8/2:119–147. Walker, B.J., and LaBianca, Ø.S., (2003). The Islamic Qusūr of Tall Hisbān: Preliminary Report of the 1998 and 2001 Seasons. Annual of the Department of Antiquities of Jordan 47: 443–471. Yehuda, L., (2012). ‘Cooking and Food in the Latin Kingdom of Jerusalem’. Pp. 52–61 in The Last Supper at Apollonia. (Tel Aviv). 149