Salvage Excavation Reports Salvage Excavation Reports No. 6 Sonia and Marco Nadler Institute of Archaeology Tel Aviv University 6 Emery and Claire Yass Publications in Archaeology P.O.B. 39040, Tel Aviv 69978 Salvage Itamar Taksel.indd 1 Israel 01.12.2009 11:10:21 Salvage Excavation Reports No. 6 Sonia and Marco Nadler Institute of Archaeology Tel Aviv University KHiRBet eS-SUYYaGH a BYZaNtiNe MONaSteRY iN tHe JUDaeaN SHePHelaH itamar taxel Contributions by Nitzan amitai-Preiss, Zvi Gur, arieh Kindler, Shmuel Marco, aharon Sasson tel aviv - 2009 Published by the emery and Claire Yass Publications in archaeology (Bequeathed by the Yass estate, Sydney, australia) of the institute of archaeology, tel aviv University Editors Ze<ev Herzog Moshe Fischer Assistant Editor Na>ama Scheftelowitz Manuscript & Production Editor: Assistant Manuscript Editor: Shirley Gassner Sharon Napchan iSSN 1565-5407 © Copyright 2009 all rights reserved Printed in israel by Kedem Press CONTENTS Foreword Chapter 1 6 iNtRODUCtiON 7 itamar taxel Chapter 2 StRatiGRaPHY aND aRCHiteCtURe 14 itamar taxel PHaSe i: tHe late HelleNiStiC/eaRlY ROMaN aND late ROMaN/eaRlY BYZaNtiNe PeRiODS PHaSe ii: tHe late BYZaNtiNe/eaRlY UMaYYaD PeRiODS PHaSe iii: tHe late UMaYYaD/>aBBaSiD PeRiODS PHaSe iv: tHe MaMlUK aND late OttOMaN PeRiODS Chapter 3 POtteRY 14 24 71 83 84 itamar taxel tHe late HelleNiStiC/eaRlY ROMaN PeRiODS tHe late ROMaN/eaRlY BYZaNtiNe PeRiODS tHe late BYZaNtiNe/UMaYYaD PeRiODS tHe >aBBaSiD PeRiOD tHe MaMlUK aND late OttOMaN PeRiODS Chapter 4 GlaSS 84 89 92 139 144 145 itamar taxel Chapter 5 StONe, BONe aND Metal OBJeCtS 152 itamar taxel Chapter 6 MaRBle FURNiSHiNGS 165 itamar taxel Chapter 7 COiNS ROMaN aND BYZaNtiNe COiNS 172 172 arieh Kindler and Zvi Gur eaRlY iSlaMiC COiNS 180 Nitzan amitai-Preiss Chapter 8 FaUNal ReMaiNS 183 aharon Sasson Chapter 9 eaRtHQUaKe-RelateD DaMaGe 186 Shmuel Marco 3 Chapter 10 KHiRBet eS-SUYYaGH iN CONteXt 188 itamar taxel tHe late HelleNiStiC/eaRlY ROMaN PeRiODS tHe late ROMaN/eaRlY BYZaNtiNe PeRiODS tHe late BYZaNtiNe/eaRlY UMaYYaD PeRiODS tHe late UMaYYaD/>aBBaSiD PeRiODS SUMMaRY Historical Sources References list of loci list of Walls 4 188 190 193 222 228 229 229 253 259 iN MeMORiaM Dedicated to the late Professor Zeev Rubin in grateful appreciation for his encouragement and learned contribution to this study 5 FOReWORD During June-July 2004 and several additional days in March and September 2005, large-scale salvage excavations were conducted by the Sonia and Marco Nadler institute of archaeology of tel aviv University at Khirbet es-Suyyagh (Map Ref. 15012/12842), a site on the north-eastern fringes of modern Beth Shemesh. the excavations (license Nos. B-289/2004 and B-302/2005) were directed by i. taxel, and assisted by S. Divon, a. tass, O. P. Sa‘ar, S. Ben-Dor (area supervisors) and R. Chen (administrator and metal detector). the excavations were financed by the construction firm Damibo international ltd. area photographs were taken by P. Shrago, S. Divon and i. taxel. Plans and architectural elements were drawn by D. Porotsky and i. taxel and prepared for publication by a. Brauner. Computer reconstructions were prepared by Y. Smertenko. Pottery was restored by R. Pelta and finds were drawn by N. Mesika and R. Penhas. Metals and coins were cleaned by N. Halperin and coins were identified by a. Kindler, Z. Gur and N. amitaiPreiss. Petrographic analysis of pottery vessels was carried out by Y. Goren and M. iserlis. Faunal remains were identified by a. Sasson; and botanical remains by N. liphshitz. i am grateful to all of them for the generous contribution of their skills and dedication. additional advice and professional support during and after the excavations was given by a. Shavit (israeli institute of archaeology), D. amit, Y. Dagan, U. Dahari, G. Finkielsztejn, e. Kogan-Zehavi and a. Nagorski (israel antiquities authority; hereinafter iaa), Y. erder and the late 1. Z. Rubin (tel aviv University), e. ayalon (eretzisrael Museum, tel aviv) and the late Y. Hirschfeld (the Hebrew University of Jerusalem). i am most appreciative for their input. i would also like to thank Z. Herzog and M. Fischer, co-editors of this series, for their painstaking reading of the manuscript and very helpful comments. the current publication provides two general site plans on a scale of approximately 1:300 for the major occupation phases in the site – Phases i and ii (Fig. 2.1) and Phase iii (Fig. 2.75). Defined architectural units appear selectively on a scale of approximately 1:100 or 1:150 (Figs. 2.2, 2.6, 2.15, 2.31, 2.41, 2.57, 2.76). Occupation layers (with architectural remains) and some fill layers are marked on the plans, whereas debris, disturbed fills and surface loci are not indicated yet appear in the list of loci at the end of this volume. this study also revises the results of the first small-scale excavation conducted at the site by e. Kogan-Zehavi on behalf of the iaa in March 2004 (Kogan-Zehavi 2008), by re-examining the excavation diaries and the finds unearthed.1 Some of the remains and small finds discovered in this previous excavation are published here. Since the loci and wall numbers of Kogan-Zehavi’s excavation are the same as some of the loci and wall numbers of the present excavations, those loci and walls from the first excavation which are mentioned in this report or marked on the site-plans are marked with an asterisk (*). itamar taxel tel aviv 2009 i wish to thank e. Kogan-Zehavi for permission to use the diaries, photographs and finds from her excavation (license No. a-4121/2004), and to a. Rochman-Halperin (iaa archive), Y. Barschak (iaa photographs archive) and Y. Ben-Michael (iaa storerooms, Har Hozvim) who made this material available. 6 CHaPteR 1 iNtRODUCtiON itamar taxel eNviRONMeNtal BaCKGROUND Khirbet es-Suyyagh (arabic: the ruin of the goldsmiths) is located on the eastern edge of a raised spur, northeast of the modern town of Beth Shemesh (Figs. 1.1, 1.2). the highest point in the site is 258 m above sea level, and ca. 35 m above the bed of Naúal Zanoaú which runs east of the spur. according to British mandatory maps from 1931 (1:20,000; Sheets 12/14 and 12/15), Naúal Zanoaú had different arabic names for each of its sections. the section southeast of Khirbet es-Suyyagh was called Wadi >illin and that which runs northeast of the site was called Wadi abu Khashaba. Fig. 1.1. location map of Khirbet es-Suyyagh. the site itself is located on the northeastern fringe of the Judaean Shephelah (foothills), just at the junction between this region and the higher Judaean Hills (Dagan 2006:9*; Nir 1975:89), delineated by Naúal Zanoaú. the immediate vicinity of Khirbet es-Suyyagh is located, according to Dagan’s division, within the geographic unit of Naúal Soreq-Naúal Yarmut, and the sub-unit of Beth Shemesh. this sub-unit, which is delineated by Naúal Soreq in the north, Naúal ish>i in the south and Naúal Zanoaú in the east, is characterised by wide hills with moderate slopes, excluding the eastern fringes of the area, that lower steeply towards Naúal Zanoaú (Dagan 2006:9*-10*). the landscape in large parts of the northern Shephelah, including in the immediate vicinity of Khirbet esSuyyagh, is composed of hills which combine two types of bedrock, both belonging to the Shephelah group. the dominant bedrock is the brittle chalky limestone (Hebrew: kirton) – a Senonian-eocene formation which is not suitable for use as a building stone. the other comes from the chalk hills of the northern Shephelah which are usually covered with a harder, though still relatively brittle, limestone crust known in arabic as nari. the nari cover of the hills made the creation of terraces, as in the Judaean Hills, impossible and therefore the slopes of the Shephelah hills are more gradual. the nari was used in antiquity as a common, though inferior, building stone (Nir 1975:93-94; Perath 1984:47, 49), as can be seen at Khirbet es-Suyyagh (Chapter 2). exposed nari exists at the spur of the site and in a few spots on its summit and slopes, but most of the surface is comprised of chalk covered by soil. the type of soil which characterises the chalky hills of the Shephelah is a pale grey rendzina, while brown rendzina soils characterise the areas that are covered with nari. at the foot of the hills dark brown alluvial soil typical of valleys 7 Itamar taxel Fig. 1.2: General plan of the site. has accumulated. the pale grey rendzina derived from soft chalky limestone is less fertile and has a lower agricultural worth than the brown rendzina created from a harder chalky limestone. the rendzina is usually used for growing fruit trees, vines, and field crops, while the shallower soils on the hill slopes are used for pasture (Dan 1984:29-30; Ravikovitch 1992:87, 100-101; Waisel 1984:187). the northern Judaean Shephelah is characterised by a semi-humid climate, with an annual rainfall of ca. 450 mm. it is a relatively hot region, with higher temperatures than those on the coastal plain and the Judaean Hills. While the average annual temperature in the area of tel aviv is 19.1°C, and that of Jerusalem is 17°C, the average annual temperature of the area of Beth Shemesh is 22.2°C (Waisel 1984:187). 8 today the natural vegetation of the northern Shephelah includes remains of Mediterranean thicket, dominated by carob (Ceratonia siliqua) and mastic (Pistacia lentiscus) trees. in the transitional area between the high Shephelah and the Judaean Hills the Mediterranean thicket gradually becomes dominated by common oak (Quercus calliprinos) and Palestinian terebinth (Pistacia palaestina) trees (Waisel 1984:188; liphschitz 2007:43). During the time in which Khirbet es-Suyyagh was inhabited, much larger parts of the region were certainly covered by natural thickets and maybe even by small forests of aleppo pine (Pinus halepensis). Most of the region is no longer covered by the native climax vegetation due to a continuous process of deforestation to make way for modern settlements, agricultural plots or planted pine forests. the natural vegetation was, however, partially deforested Chapter 1: IntroduCtIon during antiquity, to enable agricultural activity on the hills. the most dominant crops in the Judaean Shephelah in antiquity were the vine and the olive. it is therefore logical to assume that extensive parts of the vicinity of Khirbet es-Suyyagh were, as they are today, planted with vineyards and olive groves (Fig. 1.3). according to the British mandatory maps from 1931, the immediate vicinity of the site was not cultivated, excluding a small plot of olive groves to its southwest. Natural water sources in the vicinity of the site are scarce. the closest water source – Naúal Zanoaú – is seasonal and flows only in the winter. according to a British mandatory map from 1931, a spring called >Ein al-Bayiāra existed near the bed of Naúal Zanoaú, to the east of the site. Near the spring were also a well and a flour mill. it is not possible to know if this spring was active in antiquity, and if so what the volume of its flow was. it seems, therefore, that the main sources of water for the ancient sites in the vicinity were wells dug in the valleys, and cisterns hewn in the bedrock of the hills, as can be seen at Khirbet es-Suyyagh (Chapter 2). HiStORiCal BaCKGROUND the excavations at Khirbet es-Suyyagh show that the site was inhabited almost consecutively over a long period of time, from the late Hellenistic/ early Roman period to the >abbasid period (Chapter 2). throughout this time span the nature and intensity of the settlement and population of the Judaean Shephelah gradually changed. However, settlement in the region, including in the vicinity of Khirbet es-Suyyagh, remained Fig. 1.3: the view from the site to the northeast: olive groves (foreground) and vineyards. 9 Itamar taxel typically rural, being occupied by farms and villages of various sizes. the closest urban settlements to the site were emmaus (Nicopolis), 10 km to the north, and Beth Guvrin (eleutheropolis), 17 km to the southwest. Jerusalem, the major city in Judaea, is 22 km from Khirbet es-Suyyagh. During its history, Khirbet es-Suyyagh was included within the territory and/ or under the religious authority of either Jerusalem or Beth Guvrin (see Chapter 10 for a detailed discussion). in the last centuries the closest settlement to Khirbet es-Suyyagh was the village of Deiraban (or Dayr aban), located 1 km to the southeast, on the western slopes of the Judaean Hills (Conder and Kitchener 1883:24). this village was in existence at least since the late 16th century Ce, when it was mentioned in Ottoman tax lists (Hütteroth and abdulfatah 1977:119). it grew in size and population until the mid-20th century Ce at which time its population was Muslim, excluding a few Christians. Five roads ran through the village, including a main road which led to Beth Guvrin. in July 1948, during israel’s War of independence, the hill range facing Deiraban, on which also sits Khirbet es-Suyyagh, was occupied by the israel Defence Forces (Orren 1976:205). it is possible that Khirbet es-Suyyagh itself was used as a military position or that a battle took place in its area during the War of independence. this is apparent by the fact that a large amount of ammunition and tracks of damage caused by explosions were found during the excavations, mainly at the northern, higher part of the site. the village was finally abandoned in October 1948 (Khalidi 1992:282-283). the nearest road to Khirbet es-Suyyagh was that which passed through the bed of Naúal Zanoaú to the east and connected the elah valley with the northern Shephelah. in the south this road (known also as the >emeq Ha-telem [valley of Furrows road]) joined the Jerusalem-Beth Guvrin-ashkelon road, which passed through the elah valley and continued towards the coastal plain. Other important nearby roads were the lod10 Beth Guvrin road and the emmaus-Beth Guvrin road (joined at its southern end to the previous road) (Beyer 1933:229; Dorsey 1991:151-155, 189, Map 9; Roll 1995:1168, Figs. 1, 2; Roll and Dagan 1988; Shallev 1994:33; thomsen 1917:84; Weiss, Zissu and Solimany 2004: 11*-12*). these known ancient roads all followed through natural routes (Karmon 1957:144-146; Roll 1976:38) which were in use over a long period of time, up to the present day. the history of Khirbet es-Suyyagh, mainly in its heyday during the Byzantine period, was appreciably dictated by these roads (Chapter 10). HiStORY OF ReSeaRCH Khirbet es-Suyyagh was first mentioned by the surveyors of the British Palestine exploration Fund (PeF), who visited the place (which they called Khǔrbet es-Sîâgh) in 1872. They described it as “foundations and cisterns. apparently an ancient site” (Conder and Kitchener 1883:125). During the late 20th century the site was surveyed at least three more times by scholars from the iaa. the surveyors of the map of Nes Harim described the site as follows: “Remains of Structures: remains of an oil press and a wine press, rock-hewn steps and cuttings in rock outcrops, on slope (the site was destroyed by development works). Pottery: Roman, Byzantine and early islamic” (Weiss, Zissu and Solimany 2004:30*). the site was included also within the survey of Ramat Beth Shemesh and was described in detail. Within its area the surveyor identified foundations of buildings, architectural elements, remains of an oil press, a wine press, a cistern and rock cuttings. He also estimated the total area of the site at 1 hectare – a much larger area than was apparent from the present excavation. this high estimate may have resulted from the fact that the slopes of the spur were also included within the site’s area. according to the surveyor, the pottery found at the site was from the early Bronze age ii-iii (a few sherds on the eastern slope), iron age ii (a few sherds throughout the area), Roman and Byzantine (a large quantity) and early islamic (a few sherds). 150 m south of Chapter 1: IntroduCtIon the site four rock-cut basins and two cup-marks, an opening of a silt-filled cave and an agricultural enclosure wall, were identified. associated with these remains were a few sherds from the iron age ii and many sherds from the Byzantine and Ottoman periods (Dagan, forthcoming). 2 in 1998, the site and its immediate vicinity were surveyed, and a few more spots with remains of agricultural installations and Byzantine-period pottery were identified (Greenhut, Weiss and Solimany 2000). Following this survey, a smallscale salvage excavation was conducted to the northwest and south of the site (Greenhut 2001). in 2004, prior to the present excavation, a limited salvage excavation was conducted at the site by e. Kogan-Zehavi, on behalf of the iaa. the excavation included nine squares and several trenches which were dug in different places throughout the area. these probes yielded parts of a late Byzantine and early islamic complex, to which three architectural phases were attributed. Based on these finds, the excavator was the first to suggest that the complex was a monastery (Kogan-Zehavi 2008). the excavations conducted at Khirbet esSuyyagh by the author in 2004 and 2005 included an area of more than 0.2 hectares, within which a built complex was almost fully unearthed. the lion’s share of the remains and finds belong to the late Byzantine and Umayyad periods, reinforcing Kogan-Zehavi’s suggestion that the complex was a monastery.3 this monastery was part of a small group of Byzantine rural monasteries known so far in the Judaean Shephelah, and one of the few that has been excavated. the present excavations, therefore, provide important additional knowledge not only about the history of the Judaean Shephelah during the discussed periods (and others), but also about the general phenomenon of rural monasticism in Palestine and different aspects related to it. BUilDiNG teCHNiQUeS aND MateRialS the remains of the late Byzantine monastery, as well as the few older and later remains which were integrated within its frame, reflect some of the most common building techniques and materials of the discussed periods. almost all the walls unearthed at the site belong to the late Byzantine monastery. Most of them were built according to a well-known technique, typical of the period, in which the exterior face is made of ashlars and the interior face is made of small fieldstones bonded with whitish mortar. the inner face was roughly smoothed, but not plastered. this construction method was identified in many Byzantine and early islamic sites, including some monasteries, such as those of Khirbet ed-Deir (Hirschfeld 1999:11) and Deir Qal>a (Hirschfeld 2002a:165). the foundation courses of the walls, on the contrary, were built of fieldstones or coarsely-dressed stones in various sizes, usually without using mortar. in some cases, large well-dressed ashlars were set into the foundations, as known from the Monastery of Gabriel in the Judaean desert (Hirschfeld 2002b:170). the foundations were built into the soft bedrock of the site, to a depth of no more than one metre but usually much less. the known uses of plaster at the site were for covering the inner faces of cisterns, collecting vats of wine presses and a small Romanperiod installation. Plaster was also used in the construction of flat ceilings and for covering the interior walls of the church. Floors at the site were of various kinds: some were made of ashlars; some of flat irregular slabs; some of small pebbles; some of mosaic and some of levelled soft bedrock. Most of the roofs were probably flat, as was usually the case in antiquity, including in monasteries (Hirschfeld 1992:68). this is also indicated by the fragment of a stone roller which was used for levelling flat roofs as well as soft bedrock floors (Chapter 5). 2. i wish to thank Y. Dagan (iaa) for the permission to quote yet unpublished details from the final report of the Ramat Beth Shemesh project. 3. For a preliminary report on the 2004 season of excavations, see taxel 2006. 11 Itamar taxel Most of the ashlars used for building the monastery were dressed in a manner that left a more or less uniform surface on the outer face. Some ashlars, however, had a marginal dressing which is different from the familiar combination of margins and protruding bosses. these ashlars have regular relatively wide margins (7-10 cm), but have a roughly-dressed rectangle that remained in the centre of the stone rather than a true boss. except for one stone with two bosses, all the other stones of this type found in the excavation have a single boss. the technique of marginal dressing appeared in the region in the iron age and continued into the Byzantine period. the Byzantine marginal-dressed stones differ from the earlier stones in their bosses, which are much rougher and usually also more protruding. the dressing of two bosses on a single stone is also a Byzantine characteristic (tsafrir 1984:301, 303; tsafrir and Hirschfeld 1979:296). thus, although remains from the early Roman period do exist at the site and re-use of older architectural elements was identified in the construction of the late Byzantine monastery, it seems that the marginal-dressed stones under discussion date to the Byzantine-period. in this period examples of marginal dressing, including stones with two bosses, exist mainly in structures of an ecclesiastical or defensive nature. Sometimes, as in the church at îorvat Berachot (tsafrir and Hirschfeld 1979:296, Figs. 2, 3) and the monasteries of Deir Qal>a and Deir Sam>an (Hirschfeld 2002a:165, 185, Figs, 13-18, 41), the marginal-dressed ashlars represent most of the building stones of a given structure. However, in most cases, such as in the Byzantine city wall of Jerusalem (Bliss and Dickie 1898, Pl. 3), the church at the monastery of Siyar el-Ghanam (Corbo 1955:19-21, Figs. 2, 3, tav. 6:11, 7), the church at Bethany (Saller 1957:9-13, Pls. 11:b, 12-13, 73), the fortress of >en Boqeq (Gichon 1993:53, taf. 2: 2, 3) and the farmhouse at nearby Khirbet el-Jiljil (Strus and Gibson 2005:59, Figs. 41:7, 43), the use of marginal-dressed stones is random, similarly to the situation at Khirbet es-Suyyagh. 12 the remains attributed to the late Byzantine monastery, and to the later settlement that was founded within its remains, reflect a relatively widespread re-use of older building stones and architectural elements – spolia. the builders of the monastery in its first phase made use of the older remains at the site. in the renovations made to the monastery in its later phase, however, it seems that the spolia included also re-use of elements originally belonging to the monastery itself. the new inhabitants who re-settled the site after the abandonment of the monastery used everything they could lay their hands on – mainly building stones, architectural elements and marble furnishings which were taken from the monastery. the phenomenon of spolia is well-known in the Byzantine period and continued into the islamic and Mediaeval periods. the use of spolia, mainly in cities, was even approved in Byzantine legislation, and interpreted as having a propagandistic meaning, reflecting the triumph of Christianity over paganism and Judaism (alchermes 1994; CoatesStephens 2003; Weksler-Bdolah 2006-2007:95-97). the latter scholar suggested seeing the re-use of building stones from the Herodian temple in Jerusalem in the Byzantine city wall as a Christian struggle against the Jewish memory and as a victory of the former over the latter (2006-2007:95-96). it is tempting to interpret the use of spolia taken from the older non-Christian (Jewish and probably pagan) settlements at Khirbet es-Suyyagh in the monastery as Christian propaganda. theoretically, we may regard in a similar way the widespread use of spolia taken from the monastery in the later non-Christian (Muslim/Jewish; see Chapter 10) settlement. However, this conceptual interpretation of the use of spolia at Khirbet es-Suyyagh cannot be proved, and for the moment, i intend to see it mainly as a practical use of available building materials. examples for the use of spolia in other Byzantine monasteries in Palestine can be found in Khirbet abu Rish (Magen and Baruch 1997:138), Yattir (eshel, Magness and Shenhav 2000:158), Hyrcania (Castellion; Patrich 1994:77) and Masada (Marda; Hirschfeld 2001-2002:122-123). Chapter 1: IntroduCtIon the only kind of stone used for construction at the site throughout its history was the local limestone. Most of the building stones and architectural elements (thresholds, lintels, paving slabs, columns, bases, capitals and some of the doorjambs) as well as most of the industrial stone-made elements (such as the oil press’ collecting vats and screw weights), were carved from the relatively soft and brittle nari (Chapter 1) so very common among the Shephelah group of limestones. its closest exposures appear on the slopes of the spur on which the site is situated and on the nearby hills. it seems, therefore, that the stones used for the site’s building were hewn in the immediate vicinity. indeed, a small quarry was identified at the top of the spur south of the site (see below). although the nari is an inferior limestone, it was used as one of the main building stones in the hilly regions of Palestine in antiquity (Perath 1984:49-50). Some other architectural elements, such as some paving slabs and two large beam weights of the Roman phase of the oil press, as well as mosaic tesserae, were carved from a harder dolomitic limestone. this stone is typical of the Judaea group, which represents the limestone forms of the Judaean Hills (Nir 1975:90-91; Perath 1984:39-40). thus, the hard limestone used for the carving of the elements from Khirbet es-Suyyagh did not originate in the immediate vicinity of the site, but still not necessarily far away from it. Wood, of course, was also widely used as a building material at the site, although only a few charred wood remains, probably belonging to the monastery complex, were found during the excavations. Four other kinds of building elements were identified in the excavations. the first three were used only in the construction of the late Byzantine monastery and the fourth could have been used throughout the history of the site. imported marble furniture and liturgical objects were used in the monastery’s church (Chapter 6), and a marble paving slab and a marble pilaster capital or cornice part were also found. Ceramic roof tiles, probably originating from workshops in the vicinity of Jerusalem, were used in the roofing of various parts of the monastery, including the church (Chapter 3). Glass window panes were also used in the building of the church (Chapter 4). iron nails were most probably used for nailing wooden parts, such as roof beams (Chapter 5). 13 CHaPteR 2 StRatiGRaPHY aND aRCHiteCtURe itamar taxel the archaeological remains and finds unearthed at the site belong to seven periods – the late Hellenistic/ early Roman, the late Roman/early Byzantine, the late Byzantine, the Umayyad, the >abbasid, the Mamluk and the late Ottoman. However, the great majority of the architectural remains, as well as the small finds, belong to the late Byzantine and Umayyad periods. the other periods are represented mainly by pottery and small finds, in addition to a few architectural remains. almost no architectural remains were seen above the ground prior to excavation. the modern activity that took place at the site – the military events in the War of independent and agriculturalmechanical works in the 1950’s – caused severe damage to parts of the site, mainly in the north and west. these events not only demolished walls and floors, but also caused some dating difficulties. the surviving architectural remains were covered with a relatively thin layer of topsoil, and/or with fills that usually contained mixed finds from various periods.4 the more homogeneous loci were assemblages of in situ vessels, fills accumulated on floors and inside installations or fills in foundations of built or rock-cut features. Most of these loci will be described in detail below. From a behavioural perspective the finds retrieved from the floor assemblages at the site can be divided into ‘primary refuse’, ‘secondary refuse’ and ‘de facto refuse’ following Schiffer’s terminology (1995:206-211; see also laMotta and Schiffer 1999:19-21). However, the distinction between the formation processes of floor assemblages of these three types is not always clear. this is, inter alia, due to the continuous occupation of the site on the one hand and the relatively shallow accumulation of debris on the other. another factor which should be taken into account when dealing with floor assemblages is materials belonging to roof makeup. as commonly seen in the levant, flat roofs made of beaten earth, mud plaster and perishable materials frequently contained small artefactual finds, usually pottery sherds. the latter could have belonged to various periods, earlier or contemporaneous to the construction of a given building/roof. When collapsed, such rooforiginated material deposited onto the building’s floor/s and could have mixed with the floor assemblages (for a similar situation in an Umayyad house at Bosra, see Wilson and Sa>d 1984:41). thus, the dating and interpreting of a given floor assemblage should be made with caution, and with an attempt to identify any possible roof-originated material and distinguish between it and the true floor assemblage. PHaSe i: tHe late HelleNiStiC/eaRlY ROMaN aND late ROMaN/eaRlY BYZaNtiNe PeRiODS Some find spots and a few architectural remains that pre-dated the foundation of the Christian monastery were unearthed. the only feature which can be securely dated to the late Hellenistic/early Roman period (i.e., the late Second temple period) is a cistern unearthed in Squares D-e/9-10 (Figs. 2.1:24, 2.2, 2.3). the cistern was hewn into the soft chalky bedrock. Since neither mortar nor plaster adheres to this type of crumbling bedrock, the cistern’s walls had to be built of stone. these stone walls, made of one face of medium-sized and large 4. loci 100, 104-106, 109, 110-113, 115, 118, 119, 123, 126, 136, 140, 146, 149, 150, 153-156, 159, 162, 166, 169, 177, 200-207, 211-214, 217, 218, 225, 226, 230, 232, 235, 238, 249, 251, 253, 254, 256, 261, 262, 265, 266, 272, 280, 283, 284, 300, 305, 311, 315, 319, 323, 334-338, 347, 350, 355, 357, 360, 363, 369, 372, 378, 101*-105*, 107*, 108*, 110*, 112*, 115*, 135*, 148*. 14 Fig. 2.1: Plan of Phase i and Phase ii. 15 Fig. 2.2: Plan of Cisterns 24 and 25. fieldstones bonded with mortar, were covered with a layer of mortar and hydraulic plaster. the exposure of the cistern was made in two stages. it was found that very large pieces of nari (which covered the chalky layer in this part of the site), in addition to silt, had slid down, probably in antiquity, totally covering the southern and eastern walls of the cistern. to avoid collapse, only part of the cistern was excavated down to its floor, in a section cut beside Wall 32. the cistern was only completely exposed during the demolition of the site more than a year later when, after removal of the upper nari layer, important information about its plan and time of use was obtained. Firstly, the southern and eastern walls of the cistern were unearthed. it was then seen, that the cistern was square in plan (3.8×3.8 m). Secondly, it was found that the floor unearthed during the excavations belonged to a later phase in the cistern’s usage. Below this floor 16 (0.1-0.15 m thickness), under a thick layer of fill (locus 518; 0.7-1 m thickness), part of the original floor of the cistern was unearthed (Fig. 2.4). this floor (locus 519) lay directly on the bedrock, and was composed of a bed (8 cm thickness) of small fieldstones which was covered with two layers of plaster (2 cm thickness) – the lower greyish and the upper whitish. the lower part of the cistern’s walls, below the level of the later floor, was similarly plastered. this kind of plaster is typical of the early Roman period (Porath 2002:27, 35-36), thus dating the construction of the cistern to that time. the cistern’s floor lay 3.7 m below the preserved height of the walls. the minimal capacity of the cistern was, therefore, 53.5 m³. the fill which covered the floor and was partly excavated was composed of small to very large fieldstones and earth mixed with various finds, mainly pottery sherds. the great majority of the Chapter 2: StratIgraphy and arChIteCture Fig. 2.3: Cistern 24. Above) looking northwest. Below) Section a-a (see Fig. 2.2) through cistern, looking west. 17 Itamar taxel Fig. 2.4: Cistern 24, looking south. Note the two floor levels. pottery included types dated to the early Roman period (Fig. 3.2:8, 9). One of the most important finds from this fill is a fragment of a soft limestone bowl (Fig. 5.2:5), typical of the Jewish culture of the province of Judaea (Iudaea) in the early Roman period. these finds date the cessation of use of the cistern to the 1st century Ce. the richest pottery assemblage of the early Roman period was retrieved from Squares C-D/10 where intensive building activity took place during the construction of the later monastery. the space between three of the massive foundation walls (W13, W14 and W20; Fig. 2.1), built on the bedrock in order to level the slope and support the upper structure of this part of the monastery, was filled with earth containing a large amount of pottery sherds (loci 135, 137, 138, 141-144). Similar, though more disturbed, fills were found also outside the northwestern corner of the monastery (loci 155, 156). a few late Byzantine sherds date the construction and filling to the Byzantine period. Some pottery sherds and one fragment of a glass vessel from the late Roman/early Byzantine period, in addition 18 to three 4th and 5th century Ce coins (Chapter 7: Nos. 5, 14, 15), were also found here. However, most of the pottery, which contained also some partially restorable vessels (mainly various types of storage jars), was dated to the late Hellenistic/ early Roman periods (the 1st century BCe-1st century Ce). apart from storage jars, the ceramic assemblage includes bowls, cooking-pots, jugs and lamps, all characteristic of Judaea in the discussed periods (Figs. 3.1, 3.2:1-7). Some early Roman pottery sherds were also found in late Byzantine-Umayyad (or mixed) loci, mainly in the same area. two early Roman coins were also found. However, these coins came from loci which contained finds dated to the late Byzantine period onwards, and therefore cannot help in dating specific remains or assemblages. One coin (Chapter 7: No. 1) is from the time of augustus, and was minted under one of his procurators (6-14 Ce). the other (Chapter 7: No. 2) dates to the second year of the First Jewish Revolt against the Romans (67/68 Ce) . this is the more important coin of the two, since it points to the Jewish identity Chapter 2: StratIgraphy and arChIteCture of the inhabitants of the place during the early Roman period and strengthens our hypothesis that this small settlement was abandoned during or just after the First Jewish Revolt (Chapter 10). the relatively small number of finds belonging to this period and their concentration in the northern part of the site, points to small-scale activity taking place here during that time. Based on the finding of the cistern and the relatively varied pottery assemblage, it seems that there was here a small farmhouse, perhaps inhabited by no more than one family (Chapter 10). the next stage of occupation at the site can be dated, on the basis of pottery and coins, to the late Roman and/or early Byzantine period (the 4th[early?] 5th centuries Ce). as in the early Roman period, the architectural remains and most of the pottery that belong to this span of time were also concentrated in the north of the site. On the western fringe of the late Byzantine oil press (Square e8) an oval installation cut by the western wall of the oil press was uncovered (Fig. 2.5). this installation (locus 503; 1.45 m preserved length, 1.1 m width, 0.45 m depth) was hewn into the soft chalky bedrock and was plastered with a layer of rough greyish plaster (5 cm thickness) mixed with many tiny stone grits, charcoal grits and charred olive pits. the nature of this installation is unknown, but it could be part of a hip bath (Chapter 10). in that case, the installation might be an older feature, probably from the early Roman period. However, a terminus ante quem for this installation can be established in that it was sealed under a fill (loci 160, 161, 164) containing a large and varied amount of pottery sherds. these included bowls, cooking vessels, storage jars, jugs, juglets and a complete lamp, generally dated to the 3rd-5th centuries Ce (Fig. 3.3:1-11, 14). However, since this assemblage contained some types of jars which are dated no later than the 4th century Ce, it seems that the plastered installation was used only until the 4th/ early 5th century Ce. another architectural feature which was used during the late Roman/early Byzantine period is Cistern 24. as previously noted, this cistern Fig. 2.5: the late Roman/early Byzantine(?) plastered installation (l503), looking northeast. had two floors, the lower representing its original construction phase and the upper denoting a later phase of use. the upper floor was founded over a thin layer of earth which was added in order to level the thick layer (collapse?) of stones and earth that covered the lower floor of the cistern. this later floor was made of mortar and pinkish plaster (5 cm thickness), and slants to the west with a height differential of 0.3 m. in addition to the construction of the new floor, the cistern’s walls were also replastered during that time. the new layer of pinkish, well-smoothed hydraulic plaster contained many tiny pieces of crushed pottery. it was put over a layer of greyish-whitish mortar, which contained tiny stone grits and tiny pieces of crushed pottery. into this mortar were put many body sherds of ribbed bag-shaped jars of a type dated to the 3rd-5th centuries Ce. this method of plastering appeared indeed sometimes in the late Roman or early Byzantine period. in the southwestern corner of the cistern there are two steps, built adjacent to the southern wall (Fig. 2.4). the lower step (0.55×0.45-0.65 m) is placed 0.55 m above the later floor of the cistern. the upper step (0.65×0.15-0.4 m) is placed 0.35 m above the lower one. the distance between the upper step and the preserved height of the southern wall is more 19 Itamar taxel than 1 m, what makes the access to the steps and down into the cistern almost impossible. this fact can be explained in two ways. according to the first explanation, the steps belong to the earlier phase in the cistern’s usage, from the early Roman period, which implies that the upper courses of the cistern’s walls were added in the late Roman/ early Byzantine period. according to the second explanation, the steps were used for a purpose other than allowing access into the cistern. apparently the new inhabitants in the late Roman/early Byzantine period preferred to use the older cistern which remained intact since the abandonment of the site in the early Roman period rather than build a new one (for a similar phenomenon at nearby Khirbet el-Jiljil, see Strus and Gibson 2005:47-48). Moreover, they even did not clean (or totally clean) the cistern from the collapse that covered its floor, but levelled it and put a new floor at a higher level. this decision is somewhat surprising, since it did not save resources, labour or time, and also reduced the capacity of the cistern, which became ca. 10 m³ less than its capacity in the early Roman period. the third place in which remains and finds from the late Roman/early Byzantine period were found was in Squares B/9-10. the construction activities which took place at the site in 2005 partially demolished what seems to be a wine press (Fig. 2.1). the un-destroyed parts of this wine press were excavated, and found to have belonged to more than one period, from the late Roman/early Byzantine period to the late Byzantine/Umayyad period. to the earlier period were attributed the southern fringes of a large treading surface (preserved length 7 m, maximal preserved width 2.2 m; Figs. 2.6, 2.7). the treading surface was cut into the chalky bedrock, and delineated by built walls, from which were preserved parts of the southern and eastern walls (W48 and W49, respectively; 0.75 m height). the southern wall was built of small fieldstones bonded with greyish mortar, and the eastern wall was built of fieldstones and ashlars, at least one of the latter is probably an older reused stone. among the small stones of the eastern wall was found also a fragment of a basalt mill. Both of the walls were 20 covered on their inner face with a layer of mortar and white plaster (4-6 cm thickness). the floor of the installation was of white mosaic (locus 516), made of medium-sized tesserae (2×2 cm average measurements, 25 stones per dm²). the tesserae were set in diagonal lines, with no surrounding frame. the floor slants to the west, probably in the direction of the missing collecting vat, with a height differentiation of 0.13 m. the mosaic was laid over a foundation layer of white mortar (0.1-0.2 m thickness). the foundation layer itself was laid directly on the bedrock over most of the area, but in the western edge of the floor a thin layer of earth divided between the mortar layer and the bedrock. this layer contained hard limestone cut in various shapes and sizes and irregular stone chips, which represent the waste of mosaic production. the treading surface was found completely sealed by a fill of earth (locus 512), which contained a huge amount of pottery sherds, a few glass fragments, fragments of mosaic floor, numerous mosaic tesserae and mosaic production waste (Fig. 2.8). the latter was identical to the waste found beneath the floor foundations. the pottery sherds found in the fill (Figs. 3.3:15, 16, 3.4:1-10) represent a homogeneous assemblage of the late Roman/early Byzantine period. it is composed mainly of bag-shaped jars, although bowls, cooking-pots, jugs and juglets are also represented. like the pottery from the fill which covered the small plastered installation mentioned above, the present ceramic assemblage is also dated generally to the 3rd-5th centuries Ce (excluding one type of bowl which continued into the 6th century Ce). However, in this case, since some of the types are dated no later than the 4th century Ce, it seems that it is reasonable to assume that the treading surface ceased to be used during that century. Beside the eastern foundation (locus 514) of Wall 49 a fragment of a local rouletted bowl dated to between ca. 200 Ce and the 5th century Ce was found. During the removal of this wall, a nearly complete bowl was found built into the mortar of its foundations. this bowl is dated (according to several parallels) to the 4th-early 5th centuries Fig. 2.6: Plan and Section B-B (looking east) of the remains of the two wine presses outside the northwestern corner of the monastery. 21 Itamar taxel Fig. 2.7: the treading surface of the late Roman/early Byzantine wine press (locus 516), looking west. Fig. 2.8. Mosaic production waste found in the fill of the late Roman/early Byzantine wine press (locus 512). 22 Ce (Fig. 3.3:13). another complete bowl of the same type, embedded in a piece of mortar (0.13 m thickness), was found on the bedrock above the treading surface (locus 511; Figs. 2.9, 3.3:12). this bowl was probably originally embedded into another part of the wine press. One of the mortar layers of the installation’s walls was composed of body sherds of ribbed bag-shaped jars of the same type represented in the fill which sealed the installation. the fill itself contained some pottery sherds bound with mortar, which probably originated in the walls and/or the floor of another part of the wine press that was not preserved. these fragments include not only body sherds of jars, but also some more diagnostic types dated to the 3rd-5th centuries Ce (Figs. 3.3:16, 3.4:5). the overall ceramic evidence provides a 4th century Ce terminus post quem for the construction of the treading surface. Pottery sherds from the 3rd-5th centuries Ce were found sporadically, mainly in mixed contexts, in other parts of the site too, thus none of them can date the construction time of any architectural features. Coins dated to the 4th and 5th centuries Ce, in addition to two coins from the 3rd century Ce, were also found at the site (Chapter 7: Nos. 3-23). it seems reasonable to assume that the 3rd century Ce and some of the 4th and 5th century Ce coins belong to the Fig. 2.9. Ceramic bowl embedded in mortar (locus 511). Chapter 2: StratIgraphy and arChIteCture late Roman/early Byzantine phase at the site. However, 4th and 5th century Ce coins are known to remain in circulation until the late Byzantine period and even later, and some of those found in the site indeed originated from clearly 6th-7th century Ce contexts. excluding the above-mentioned remains, there are a few others that can be attributed to the Roman or early Byzantine phases at the site. Some elements that belonged to an oil press were found in various secondary uses in Squares D-e/8-9 and in other later parts of the complex. these elements include two rectangular beam weights of the reversed-t type, which was the main type of beam weight in the levant from the Hellenistic period onwards (Frankel 1999:101, and see Chapter 10). the weights are very large, and made of hard dolomite limestone. the first weight (1 m height; 0.73 m width; 0.58 m thick; Fig. 2.10) was built into a wall of the late Byzantine oil press. the upper and side openings of the reversed-t bore are square in shape (0.23×0.23 m), and the upper fringes of the side opening are convex in the middle. its estimated weight – ca. 570 kg – has been calculated using an average density of 1500 kg/m³ for nari (Syon 2004:162). the second weight (1.06 m height; 1.04 m width; 0.58 m thick; Fig. 2.11) was converted into a press base in the late Byzantine oil press (see below), and was found beside one of the collecting vats of the latter. the upper and side openings of the reversed-t bore are rectangular in shape – the first is measuring 0.27×0.20 m, and the other measures 0.27×0.25 m. Here too, the upper fringes of the side openings are convex in the middle, and its estimated weight is ca. 870 kg. as to the rest of the elements of the oil press to which these weights belonged, such as the olive crushing basin, press base and collecting vat/s, we cannot say much. the excavation of the late Byzantine and Umayyad oil press did indeed yield a crushing basin and two collecting vats, but at least one vat was not hewn before the late Byzantine period. However, the second vat, which is quite different from the first, and also the crushing Fig. 2.10. Beam weight from Phase i oil press. Fig. 2.11. Beam weight of the Phase i oil press, reused as a pressing base in the Phase ii oil press. 23 Itamar taxel basin, may originate in the Roman period. if these elements remained intact until the late Byzantine period, it is very possible that they were reused in their original functions in the later oil press. Since this free-standing collecting vat is very heavy, it can be assumed that it remained in the same place since the Roman period. the weights represent an oil pressing technique which is very different from that used at the site during the late Byzantine and Umayyad periods (i.e., the lever and screw). the pressing technology using beam weights continued throughout the country well into the Byzantine period. However, the fact that the two large weights described above were found in secondary use in an oil press which used a different technique, and despite the absence of Roman period finds that can be directly connected to them, it is quite clear that these weights should be dated prior to the late Byzantine period. their exact date, however, cannot be determined based only on their form and measurements since parallels for similar weights are known mainly from early Roman period assemblages and some later assemblages (Chapter 10). it is not impossible to assume that an oil press which existed here during the late Hellenistic/early Roman period continued to function under new inhabitants that resettled the site in the late Roman/early Byzantine period. Only during the massive construction activities which occurred here in the late Byzantine period was the old oil press partially dismantled and its beam weights removed for the building of the new, improved installation. PHaSe ii: tHe late BYZaNtiNe/eaRlY UMaYYaD PeRiOD the architectural picture of Khirbet es-Suyyagh, as known from the excavations, originated mainly in the late Byzantine to >abbasid periods. two major stages of site formation can be identified within Phase ii, which represent the time of the monastery’s heyday during the late Byzantine and early Umayyad periods (the 6th century Ce to late 7th/early 8th century Ce). attributed to 24 the first stage (Phase iia) is the foundation of the monastery and its main time of existence, which ended in the destruction of parts of the monastery sometime around the mid-7th century Ce. the second and shorter stage (Phase iiB) represents the time between its reconstruction and its abandonment by the monastic community. it must be noted that there were sometimes difficulties in determining whether a certain remnant or feature belonged to Phase iia or Phase iiB, or to Phase iiB or Phase iii. the built area of the monastery covers ca. 2030 m². it is a roughly rectangular structure, oriented north-south (Figs. 2.1, 2.12, 2.13). it is composed of two units: the living/industrial complex with the tower and the church complex. the domestic/industrial complex is rectangular (50×17.5 m; ca. 880 m²). the massive eastern wall (W33; Fig. 2.14) was built with an outer face of large rectangular ashlars and an inner face made of small fieldstones bonded with whitish mortar. the average width of the wall is 1 m, except for the last 10 m of its northern section, where it is 2.5 m wide and built in a somewhat different method. it is preserved to a maximum height of four courses (1.2 m), mainly in its northern part. the foundation course was in some places slightly wider than the upper courses. the line of the wall is not exactly straight, and ca. 30 m from its southern end it makes a slight turn to the east. the seam between the two sections and their identical characters indicate that the wall was built in two technical stages. it is possible that the northern part of the monastery was built first, with its massive walls, tower and cisterns. Only after the work there reached a certain point was the rest of the monastery constructed, and the southern section of the eastern wall was built on a somewhat different line. another explanation for the change in the wall’s course, could be the older remains in the northeastern corner of the monastery (such as parts of the Roman oil press), that obstructed the late Byzantine period builders, and were finally included within the enclosed walls. the southern wall of the monastery (W139; 18.7 m length; 1 m average width) encloses the central unit (Rooms 2, 4-6, 8-21). it was preserved to a Chapter 2: StratIgraphy and arChIteCture Fig. 2.12: General view of the site, looking north. maximum height of 3 foundation courses (1.2 m). in the west it ends at the gatehouse. the eastern and southern walls of the monastery are massive and have relatively deep foundations. these walls not only served as peripheral (and therefore defensive) walls for the monastic community, but also, being built on the moderate eastern and southern slopes of the site, supported the mass of construction built behind them on higher levels. On the west it is bounded by three consecutive walls (W8, W100 and W129), with a total length of 43 m (Fig. 2.1). the two southern sections (W100 and W129) were less massive (0.7 m width) than the eastern and southern external walls of the complex. Only the northernmost of the three, which belongs to the tower, was thicker and built of very large ashlars. the northern part of W100 was destroyed sometime later and rebuilt. However, the newly built section did not exactly match the course of the original wall, and digressed slightly eastward. the central unit is enclosed on the north by W38 (8 m original length; 0.9 m width), running from the tower in the west to the levelled bedrock above the cisterns in the east (Fig. 2.1). However, since this part of the site was one of the most severely damaged by modern activity, only the western part the foundations of W38 and a few stones of its first course were preserved. 25 Itamar taxel 26 Fig. 2.13. General reconstruction of the monastery, looking northwest. (Drawing by Yura Smertenko) Chapter 2: StratIgraphy and arChIteCture Fig. 2.14. the eastern enclosure wall of the monastery (W33), looking south. THE GATEHOUSE the main gate of the monastery was located at the southwestern corner of the central unit (Square D5). However, the gate itself was only one part of the more complex unit of the gatehouse (Figs. 2.15, 2.16). the southern wall of the monastery (W139)ends on the west in a large monolithic threshold (2.05×0.8 m) carved of a relatively soft limestone (Fig. 2.17). three sockets carved in the threshold indicate that the gate had two doors that opened to the inside – one was 0.75 m wide and the other 0.6 m wide. Four parts of the gate’s finely-dressed doorjambs were found embedded in secondary use in a later, nearby wall dated to Phase iii (W210; Fig. 2.18). another architectural item that can be attributed to the main gate area is a rectangular ashlar with a depiction of a cross on it, found inside the room immediately to the northeast of the gate. the stone (0.6×0.4×0.29 m) was decorated near one of its edges with a shallow incision of a simple cross within a circle (Fig. 2.19). it must be noted that this stone is the only architectural item found in the excavations that bears the sign of the cross. the simplicity of the cross decoration suggests that it was made by one of the monks, and not necessarily by a skilled artisan. a similar cross-decorated stone was found at the monastery of Khirbet ed-Deir, and the excavator suggested that it was set in the wall above the gate’s lintel (Hirschfeld 1999:19, Fig. 15). to the east of the main gate a rectangular room (Room 3; 4.1×2.5 m) projects southward from W139. the entrance to this room was through a doorway with a threshold and a socket in the middle of its northern wall. its floor (locus 359) was built of small to large irregular slabs, mostly made of soft nari limestone but a few made of hard dolomite (Fig. 2.20). along the southern part of the eastern wall (W216) was a line of four building stones that may have served as a bench. this feature, and the location of the room, suggests that this was a guardroom protecting the main gate and controlling ingress into the monastery. the room’s floor was laid over an earth fill (locus 393; 0.1-0.2 m thick) which levelled the surface. the fill contained pottery sherds, mainly belonging to cooking vessels and jars, which are dated to the mid-6th-7th century Ce. these finds indicate that the room was repaved sometime after its original construction, maybe during Phase iiB. another possibility is that the whole room was added during that later phase. From the gate a path, bounded by walls W218, W220 and W230 on the west and the western wall of Room 3 (W214) on the east, led to the south (11.5 m known length; 3 m width). its continuation to the south was not excavated, so its total length is unknown. the destruction caused to the monastery by the earthquake that marked the end of Phase iia is very apparent in the area of the main gate. the massive threshold was broken into two large pieces (see Chapter 9). the remains which point to the architectural changes that took place here during Phase iiB are: (1) the western part of the monastery’s southern wall, which included the main gate, is not built exactly on the same line as the rest of the wall. in addition, the level of the main gate’s threshold itself is 0.5-0.6 m higher than that of the path which leads to it from the south. this height 27 Itamar taxel difference did not, of course, allow comfortable passage from the path into the monastery; (2) the western edge of the gate ended without being connected to any of the other nearby walls, e.g., those of the church complex (Fig. 2.15). actually, only a thin wall (W218; whose first course is built of three small coarsely-carved stones) abutted on it from the south and connected it to the church’s apse, while creating a small trapezoidal room (Room 33; 2.3-3.3×2.2 m); (3) the path and the entrance corridor to its north was found covered with a fill rich in pottery sherds and other finds. at the time of the reconstruction, the gate was not only rebuilt along a somewhat different line than the rest of the southern wall, but also its threshold was raised. the original, broken, threshold remained in use in the new gateway. it is possible that the continuation of this wall toward the church complex was also destroyed, and the present thin wall was built instead. the problem of height difference in the gate area was solved by laying an earth fill (loci 341, 358, 362, 370, 379) which covered the outer path and created a ramp that led to the new threshold level. the same fill continues north of the gate – all along the entrance corridor, up to the height of the second stair from the bottom (loci 240, 281, 325, 330, 342, 368, 374). it was found that W207 was founded on the same fill, 0.45 m higher than the bedrock surface. this wall was indeed built from ashlars, but rather crudely without any real foundation course. it seems, therefore, that the original wall was also destroyed or damaged by the earthquake, and rebuilt later at a higher level than before. Room 33, west of the gate was paved with a thin layer of small pebbles (locus 385). this may have been the floor itself or only its base, although no signs of other paving, such as mosaic tesserae, were found here. Only the southwestern corner of this pebble layer was preserved. When dismantling this layer we found that it was laid over an earth fill (locus 386; 0.3 m thick) which contained pottery sherds dated to the first half of 7th century Ce (Fig. 3.7:14, 15). a section cut in the northwestern corner of the floor, through its earth foundations and into the bedrock (locus 28 356), revealed more pottery sherds (Fig. 3.7:16-18) which also provide a first half of 7th century Ce terminus post quem for the construction of the floor, thus dating it to Phase iiB. another interesting detail regarding this room, is that one of the stones of its thin eastern wall has a marginal dressing with two bosses (Fig. 2.22). the fill on both sides of the gateway contained large amounts of pottery sherds (Figs. 3.5, 3.6. 3.7:1-13) and other finds. the latest pottery types which were found on the bedrock give a terminus post quem of the first half of the 7th century Ce for its accumulation. Four coins were found in the fill north of the gate: one of Crispus (324-326 Ce; Chapter 7: No. 8), one dated generally to the 5th century Ce (Chapter 7: No. 19), one of Justinian (536/7 Ce; Chapter 7: No. 28) and one of Constans ii (641-648 Ce; Chapter 7: No. 36). the latter confirms the date given by the pottery. an interesting artefact found in the northern fill is a small bronze weight (Fig. 5.4:17). the contents of this fill probably originated from the cleaning out the monastery after the earthquake, and/or in the refuse accumulated during the first century or so of its existence. Nevertheless, we cannot rule out the possibility that some of the finds from it were accumulated during Phase iii, after construction of the later blocking wall that runs north of the gateway. there were two ways to enter the central unit of the monastery from the main gate. the indirect route leads to the north through Corridor 7 (Fig. 2.15; 15 m length; 2.1 m width). that opened into a courtyard (Room 28) outside the central unit. the corridor was delineated on the west by W207 that belonged to the church complex and on the east by walls W129 and W100. From this area one could enter the central unit through a subsidiary gate in its western wall (see below). the builders of the monastery overcame the height differences between the southern end of the corridor and the area of the outer courtyard (on top of the hill) by constructing a staircase of four steps (4.1 m total length) in the northern part of the corridor (Fig. 2.21). Chapter 2: StratIgraphy and arChIteCture Fig. 2.15. Plan of the gatehouse complex and the southern area of the monastery. 29 Itamar taxel Fig. 2.16. Reconstruction of the gateway, looking north. (Drawing by Yura Smertenko) Fig. 2.17. the threshold of the main gate, looking south. 30 Fig. 2.18. Parts of the door-jambs of the main gate embedded in a Phase iii wall (W210), looking south. Chapter 2: StratIgraphy and arChIteCture Fig. 2.19. ashlar decorated with cross. Fig. 2.20. the floor of Room 3, looking south. Fig. 2.21. the staircase of the entrance corridor, looking north. 31 Itamar taxel Fig. 2.22. ashlar with marginal dressing and two bosses from W218. threshold) were laid above large, coarsely dressed stones. the paving stones themselves were placed directly above the levelled soft bedrock, and the few ceramic body sherds found here can not give a more accurate date to the construction than the Byzantine period. this unit had no visible access from other directions, but it may have had another doorway in the southern part of its eastern wall. this courtyard probably functioned as a gathering place for guests and pilgrims who visited the monastery. the rooms to the northeast may have served as a hostel. Parts of the floor were found covered by a thin layer of topsoil that contained a few pottery sherds dated to the late 6th/7th-8th century Ce, which seem as represent later occupation and/or post-abandonment deposits. THE dOMESTIC UNIT roomS 4-6 a direct route which led from the main gate turned to the right into a small rectangular room (Room 1; 4.1×3.8 m). this entrance room led to other rooms in the north and to a small inner Courtyard 2 (locus 367; 5×4.5 m) to the east (Figs. 2.1:2, 2.15, 2.23). the Southern Courtyard the doorway into Courtyard 2, in the southern part of W215, also had a monolithic threshold like that of the main gate (Fig. 2.24; 1.6×0.55 m) with two doors – one 0.65 m wide and another 0.4 m wide. the courtyard was paved with ten rows of small and medium-sized nari ashlars, smoothed on their upper side. the northern part of this courtyard was narrower than the southern part (3.5 m width), due to the existence of a short wall (W233; 2 m length) which abutted on the courtyard’s northern wall (W149). the function of this wall is uncertain, and it may be a later addition which represents part of the reconstruction activities carried out in the monastery during Phase iiB. the surrounding walls of the courtyard were preserved to the height of the foundation or the first course only, due to robbing. Some of the paving stones were also robbed. in a section (locus 392) made through the courtyard’s floor in its southwestern corner, we found that the stones of the western wall (including the 32 Room 4, north of the small courtyard was large and rectangular (9.1×3.3 m). like most of the rooms in the monastery, its floor was the soft bedrock itself, probably roller-levelled. Most of its area was found occupied by remains attributed to Phase iii – poor walls and a small lime kiln. therefore, its nature during Phase ii is unclear. an entrance with two small steps led to the room to its north (Room 5; 4.3×3.8 m). this room (Fig. 2.25) is unique in that its ceiling was supported by an arch of which only two rectangular pilasters (0.5×0.6 m) built beside its eastern and western walls were preserved. the first stone of the eastern pilaster has a marginal dressing (Fig. 2.26). Despite the relatively elegant appearance of this room, its floor was simply the levelled bedrock. the pottery found on it (loci 161*, 260) included some restorable vessels which can be dated to the late 6th/early 7th to end of 7th century Ce (Figs. 2.27, 3.8:1-4). in addition, a coin, maybe of Justin i (518-527 Ce; Chapter 7: No. 27), was also found here. therefore, these finds have been attributed to Phase ii, and most probably to its later stage. a doorway in the southern end of the eastern wall of Room 5 led into Room 6, a smaller room (4.1×2.8 m) which is not accessible from any other rooms. in its northeastern corner was a raised rectangular platform (0.9×0.7 m, 0.3 m height), Fig. 2.23: Courtyard 2 (locus 367), looking northwest. Fig. 2.24: the threshold of Courtyard 2, looking west. which may have served as a low shelf. this room also had a levelled bedrock floor. the dividing wall between the two rooms was thickened by adding a new row of stones to its eastern face. these stones were probably added in order to strengthen the wall, which may have been damaged in the earthquake. thus, i intend to attribute the finds which were unearthed on the bedrock floor of the backroom (locus 259) to the same period as the assemblage in Room 5. these finds include some pottery sherds, two complete juglets (Fig. 3.8:5-7), and a coin of Maurice (582-602 Ce; Chapter 7: No. 30). the latest pottery types point to a 7th-mid 8th century Ce date for this assemblage, which corresponds also to the date given by the coin. Rooms 5 and 6 seem to be the quarters of a senior personality, maybe even the head of the monastery ( /  ) itself. this is indicated by the relative splendour of the room, by its separation from the main living quarter of the monastery and by its proximity to the church. a threshold in the doorway of the later phase. the lintel is large (1.6 m preserved length, 0.5 m width, 0.45 m height) and well dressed (Fig. 2.28). From a stratigraphic point of view, the area immediately behind the subsidiary gate (Rooms 8-10; Squares D/7-9) is the most complex of the site. this wide area (21×7.5 m, ca. 157 m²) was paved with an impressive ashlar floor (locus 289; Fig. 2.29). about one third of the floor in this area was not preserved, due to modern activity that demolished almost the entire northern wall of the central unit. at first it was thought that the entire floor belonged to the original stage of the monastery’s construction (Phase iia). it seemed that we were dealing with a wide courtyard, which connected all the units in the northern part of the monastery. this floor looked very different from that of the smaller courtyard in the south. the latter was paved with ashlars more or less uniform SubSIdIary gate and Central Courtyard the subsidiary gate of the central unit from the west in Square D8 was less impressive than the main gate in the south. However, the damage caused by the earthquake which occurred at the end of Phase iia, and the changes following it (during Phase iiB or Phase iii) left nothing of the original entrance in situ. Some of the architectural items belonging to the doorway were found in different uses very close to their original location. the width of the original subsidiary gate can be assessed only by its partly-broken lintel, found in secondary use as Fig. 2.25: Room 5, looking northeast. 33 Itamar taxel Fig. 2.26: ashlar with a marginal dressing from the eastern pilaster of Room 5. in size, which were arranged in straight lines. On the other hand, the northern courtyard’s floor was made of stones of different sizes, which were usually placed randomly, sometimes with small fieldstones between them. Under the eastern side of the floor was unearthed part of a plastered drainage channel (locus 269), whose southern section was still partly sealed by the paving stones. it seems that the channel ran from the southeast to the north, where it was joined to a ceramic pipe which apparently led into the Cistern 24. the channel therefore drained rain water from the nearby roofs and the courtyard’s floor to the cisterns. However, only its first 6 m were preserved, in addition to two complete plastered ceramic sections (Fig. 3.8:8) found in situ beside the carved bedrock south of Cistern 24. the assumption that not all of the courtyard's floor belonged to the initial phase of construction was verified after large parts of it were removed, mainly in the middle and along the eastern wall of the tower (W8). it was found that the bedrock surface here had not been levelled as in the northern part of the courtyard, where the floor was almost entirely demolished, but lay about 0.5 m lower. in this lower part were found remains of two ashlar-built walls, oriented east-west and placed at a distance of 2.5 m from each other (Fig. Fig. 2.27: Crushed pottery on the floor of room 5 (l260). Fig. 2.28: the subsidiary gate of the monastery, with the lintel reused as a threshold, looking west. 34 Chapter 2: StratIgraphy and arChIteCture Fig. 2.29: General view of the northern part of the site, looking northeast. 2.1:9). along the inner face of the northern of these walls was a raised platform (0.9 m wide). Between it and the southern wall were two steps(?) (1.6 m wide), also built of ashlars, which led to the east, to the current level of the courtyard’s floor. it seems that these are the remains of a small entrance room or corridor, which led from the subsidiary gate into the courtyard. the Phase iia courtyard would, therefore, have been smaller than during Phases iiB and iii. actually, it can be assumed that the central unit of Phase iia contained two courtyards: Room 8 (12.8×7.5 m) to the north of the entrance corridor, and Room 10 (7.5×4.2 m) to its south. alternatively, it is possible that the original courtyard included only the area north of the corridor (i.e., 8), while the southern area (10) was a regular room. it is possible that access to the northern courtyard could have been from the point between the subsidiary gate and the entrance corridor, but since all of this area was renovated during Phase iiB and/or Phase iii, this remains only a hypothesis. the original arrangement of the courtyard was changed, probably due to the earthquake that struck the monastery at the end of Phase iia. the walls of the entrance room or corridor were demolished down to their foundation course. the floor of this room/corridor, which was lower than those of the courtyards themselves, was raised to create a uniform-level platform in which were embedded the remains of the entrance room. this change was identified when parts of the floor were removed (loci 182, 287, 293, 294) and the remains and finds below were unearthed. it 35 Itamar taxel was found that the area of the entrance corridor was filled with loose soil which contained a large amount of pottery sherds (including a complete lamp), fragments of roof tiles and glass vessels, a fragment of a marble bowl and two pieces of a large limestone basin (Fig. 2.30). a few sherds, including a lamp dated to the Umayyad period, were also found in the silt inside the part of the drainage channel which was sealed by the floor pavement (locus 296). a short distance to the north, an irregular pit (locus 600) hewn into the soft bedrock was unearthed below the floor. the pit was full of pottery sherds and glass fragments, in addition to some nearly complete pottery vessels and a complete glass bottle. another part of the floor was removed in the south of the courtyard (Room 10; locus 404). it was found that it had been laid over a thin layer of earth (0.1 m thick) which contained a small amount of pottery and roof tile fragments, some of which bear traces of fire. this strengthens the above-mentioned suggestion that this area was also rebuilt at a later stage and corresponds to the rebuilding of the northern section of the complex’s western wall (W100). the ceramic assemblage in all of the above-mentioned loci is dated, based on the latest types of fine ware found in it, to the 7th century Ce (Figs. 3.9-3.12). these finds, especially the fragments of roof tiles and the marble bowl, point to the floor being renovated later than the foundation of the monastery, probably during Phase iiB, around the mid-7th century Ce. the alterations made to the subsidiary gate also point to the later construction of the floor. as mentioned before, the large lintel of this gate was found in secondary use as a threshold, half buried under the floor level of the courtyard. the socket of the doorway was not carved in the lintel itself, but in a small flat stone, which was put against its northeastern corner, on the courtyard’s pavement (Fig. 2.28). None of the gate’s doorjambs were found and it is not known if the original doorjambs were still used in the later doorway. Part of a threshold with a socket, maybe that of the original gate, was found in secondary use as a building stone in a room built during the same time outside 36 the western wall of the complex. the relation between this room and the renovated doorway, and the pottery found in the joining point between the two also help us to attribute the new doorway to Phase iiB. roomS 11-21 to the south and southeast of the courtyard (Squares D-e/6-8) were one large room and nine smaller rooms arranged in two distinct groups. the northernmost, Room 11 (5.5×4.5 m), was bounded by the oil press in the north. it served only as the main access into the oil press complex, since it was not connected by doorways to any other room. entry into it was directly from the courtyard. the floor of this room, which was only partially preserved, was 0.5 m lower than that of the courtyard. it was also built of medium-sized ashlars in secondary Fig. 2.30: Broken stone basin from the fill beneath the floor of the large courtyard. Chapter 2: StratIgraphy and arChIteCture use, some of which were smoothly-dressed. Some building stones and a large fragment of a threshold found in the fill between this floor and a later floor built above it during Phase iii, indicate that originally there was a wall with a doorway which separated Room 11 from the courtyard. this wall was demolished during the construction of the Phase iii floor. another indication of this is the drainage channel which ends directly at the approximate line of the missing wall. it is most probable that a trough which was built adjacent to the wall was joined to the channel that runs under the courtyard’s floor toward the cisterns. Between the floor of Room 11 and the bedrock was a thin layer of fill (locus 403; 0.1-0.2 m thick) which contained a small amount of pottery including one fragment of a roof tile. the fact that the floor was higher than the threshold level of the oil press indicates that it does not belong to the original construction phase of the monastery, but was built (re-built?) during Phase iiB. the eastern group of rooms are arranged in a line which runs between the oil press in the north and the southeastern corner of the monastery. the walls of Room 12 (3.6×3 m) were not preserved to a height of more than three courses (0.85 m). Beside the foundations of its eastern wall (W33) and part of the southern wall (W126) a line of large coarsely carved stones was unearthed. these stones were built into the soft bedrock in a way which left them covered entirely by the room’s floor, but since no finds earlier than the Byzantine period were found here, they presumably do not represent an earlier phase of construction. i suggest seeing them as a strengthening ‘belt’ added to the foundations of the Phase ii walls. No signs of a doorway were found in this room due to the later looting of stones, but it seems that it must have been located in the western part of the northern wall or in the southern wall. the western half of this room was unearthed already in the iaa excavation at the site. On the bedrock level of this part of the room, which was identified as the room’s original floor (locus 114*), pottery sherds and two coins dated to the first years of Justinian’s reign (527-536 Ce; not published here) were found. the latest pottery types, however, give a 7th century terminus post quem for this assemblage (Fig. 3.13:1) which apparently represents the abandonment of the room at the end of Phase iiB. However, there are some features that were built inside the room during Phase iii, so these may well have disturbed the earlier assemblage. the next room to the south, Room 13, has similar measurements (3.8×3.3 m), and here also no traces of a doorway were found. the floor, as in the previous room, was the levelled bedrock itself. in this room the foundation trenches of the eastern and northern walls were unearthed. the few pottery sherds found inside the trenches have been dated to the late Roman/early Byzantine period (3rd-5th century Ce). However, since except for the late Roman/early Byzantine wine press this is the only place in the site where finds earlier than the late Byzantine period were found in the foundations of architectural remains, these can be considered to be coincidental and not indicating the date of construction. in the middle of Room 14 (4.1×2.8 m) part of a stone paving made of small and medium-sized fieldstones (locus 298) remained. in order to level the uneven surface, another layer of stones was laid below the paving in the eastern part of the room. a section cut through the floor (locus 191) unearthed neither finds nor older remains. therefore, although this paving cannot be securely dated, it is attributed to Phase iia. the unit comprising Rooms 15 and 16, to its south, was more than twice as large as any of the other rooms. it consisted of a long hall (10×4 m) and a smaller room (4×2.5 m) in the southeastern corner of the monastery. the location of the entrance to this unit is unknown, since no threshold was preserved. Probably access was through the small courtyard in the southwest (Room 2). thus the hall and adjacent room were part of a separate unit in the south of the monastery, and not part of the above-mentioned row of rooms. a somewhat curious fact is the space between the northern part of the hall’s western wall (W131) and the eastern wall (W145) of Room 6. this narrow space was divided in the middle by two large ashlars 37 Itamar taxel protruding from Walls 131 and 145, which created two small chambers between the two larger rooms although they have no visible access (For a similar feature in the Byzantine farmhouse of Ramat Hanadiv, see Hirschfeld 2000a:55-56). Given the location and size of Room 15, i suggest identifying it together with Room 16 as a hostel for guests and pilgrims who visited the monastery, whereas Rooms 12-14 were probably used as living rooms for the monks. visitors would have entered the monastery through the main gate, and immediately turned to the right towards the small courtyard where they were received by the monk on duty. access to Rooms 17-21 was through the northern courtyard. it seems that the entrance to Room 17 (3.8×2.3 m) was in its western wall (W127), directly from the courtyard. a small patch of paving made of small, flat fieldstones survived in the middle of the room. a doorway in the southern wall of the courtyard, with a threshold partly preserved in situ, led into a small entrance room (2.5×2.1 m), Room 18. it opened to the west into Room 19 (7.5×2.6-3.5 m) that was paved with small flat fieldstones, preserved only in its middle and north (loci 268, 290). it seems that the rest of the paving here and in the entrance room was uprooted during Phase iii, to which we can attribute the finds found in these rooms. as already noted, the western wall of the domestic complex, W100, (which was also the western wall of Room 19) was partially rebuilt sometime later than its initial construction. this rebuilding seems to have caused the northern part of Room 19 to be a little bit narrower than it was originally. Room 18 also opened into a short corridor (3×0.85 m), which led into Room 20 (7×4.3 m), which had an indirect entrance. the few pottery sherds found in the foundation trenches of its northern and eastern walls (locus 258) date their construction to the 6th century. Some changes were made in the plan and use of this room during the two later phases. the latest change can be related without doubt to Phase iii (see below), but the earlier one is also the more difficult to date. this is a wall (W107) which was built in the middle of the room, dividing it into two equal rooms (2.2 m width each). in the 38 western room another line of stones (a bench?, a shelf?) was built along the northern wall. On the levelled bedrock floor of the room (locus 242) was found a living layer, which contained some crushed ceramic vessels – two storage jars, a jar lid and a basin, in addition to sherds of other vessels (Fig. 3.13:2-4). this assemblage can be generally dated to the mid-6th/mid-8th century, so in this case it can be dated either to the end of Phase iiB or the beginning of Phase iii. east of Room 18 was Room 21 (4.1×2.3 m) that seems to have been entered through the former. THE TOWER the last 10 m of the northern wall in the west (Squares C/8-10) were part of a massive tower (Figs. 2.1:22, 2.29, 2.31), which soared above the northwestern corner of the monastery. the total external measurements of the tower are 10×15.8 m. in the mid-20th century the tower was severely damaged by bulldozing at the site. Parts of its western wall (W11) and one of its inner walls were demolished and the other walls were preserved to the height of the foundation course only. its foundations were built of very large, coarsely carved stones, and between them smaller fieldstones which were bound with mortar. the foundations were not deep – between one or two courses below the surface. in the north they were founded on the nari layer and in the south they were dug into the softer chalk layer. Usually, the thickness of the walls was between 1 to 1.3 m, but in the southeastern corner the foundations were between 1.7 to 2 m wide, since they had another row of medium-sized and large fieldstones. these stones were probably added because of the brittleness of the bedrock in this area. the upper structure of the tower was built of two faces of somewhat smaller but smoothly dressed ashlars. this is indicated by parts of the second and third courses preserved in the southern and eastern walls, and from other stones found in the debris and later fill which covered the area. the tower had four external walls (W7, W8, W11, W16), and four internal walls – two oriented east-west (W9, W37) and two oriented Chapter 2: StratIgraphy and arChIteCture north-south (W14, W39). W9 (5 m long, 1.5 m wide) was preserved only to the height of its foundation course, which was made of small to large fieldstones bound with whitish mortar that contained small pottery sherds. this wall perhaps supported the staircase which led up from the ground floor. W39 (3 m long, 1 m wide) connected W7 and W9 and was built of two rows of large coarse ashlars. Only part of the foundation course (1.3 m wide) of W37, north of W9, was preserved. it was composed of two rows of large, coarsely carved stones, and the bedrock itself. this wall divided the tower’s ground floor ino a northern and southern wing. the fourth internal wall (W14) was built between the northern wall of the tower and the steeply sloping bedrock in the south in order to support the massive superstructure of the tower. its foundations were made of medium-sized and large fieldstones bonded with a mixture of small fieldstones and whitish mortar. its upper course was made of large coarse ashlars and the same mixture of small fieldstones and mortar. Between this wall and the eastern wall of the tower was a deep natural depression in the bedrock, which was partly filled with small to large fieldstones that give the illusion of another supporting wall. the spaces on both sides of this ‘wall’ were filled with earth that contained a huge amount of early Roman period pottery that seems to reflect the cleaning of the area during the massive construction in the Byzantine period. the floor of the southern wing of the tower was made of small and medium-sized fieldstones (locus 501). in the space between the two internal walls the paving stones levelled natural depressions in the nari bedrock. However, in the smaller space between the southern internal walls of the tower and its southern external wall, the floor was laid over a thin layer of earth (locus 520) lying above the bedrock. the southern edge of the floor was joined to the southern wall of the tower by a fine greyish mortar. the few pottery sherds which were found in the earth fill sealed by the floor included a rim of a late Roman C/ Phocaean Red Slip Ware bowl, giving a terminus post quem of the 6th century for the construction of the tower. two doorways led into the tower, both in its eastern wall. the northern doorway, beyond all doubt, belongs to the tower’s initial stage, i.e., to Phase iia. access to it was from the northwestern corner of the large courtyard. in front of it was a roughly-levelled bedrock surface, which was integrated in the courtyard’s paving. this was a relatively impressive doorway (1.3 m wide), partly hewn in the nari bedrock and partly built (Fig. 2.32). the lower part of the southern doorjamb (0.55 m high) and the southern part of the threshold were hewn as a single unit. the northern doorjamb (which was not preserved) and the northern part of the threshold were built from ashlars which were bound with whitish mortar. two sockets were carved in the threshold, indicating that the entrance was equipped with a double door. this doorway led into the northern wing of the tower. the southern part of the floor in this wing was probably the nari layer itself, which is very cracked at this point. On its northern edge was a carved step (2.2 m long, 0.2-0.28 m wide, 0.15-0.25 m high), with only its two ends preserved: the eastern one was hewn in the bedrock itself, while the western was carved in one of the stones which were used in the filling of the deep natural depression at this point. it seems that the step was designed to help bear the floor which covered the rest of the northern wing. this floor was apparently built of perishable materials, such as wooden logs, covered by mud and/ or plaster, and therefore was not preserved. about two thirds of the area covered by the floor, between the eastern wall of the tower and the internal wall in the west, were totally sealed by an earth fill. the westernmost part of the northern wing had a different use and appearance. it was lower by ca. 2 m than the ground floor level of the tower and therefore should be treated as a basement (Fig. 2.33). its southern half (locus 196) was partly built and partly hewn in the nari bedrock. the hewn part is oval (1.5×1.8 m, 1.5 m deep), and above it is a carved ledge or step (1.1 m long, 0.27-0.33 m wide, 0.3 m high). the curved wall (W19) that 39 Itamar taxel Fig. 2.31. Plan of the tower. encloses it was built of two faces of ashlars bound with mortar, and founded on a low protrusion in the bedrock. it was preserved to a height of one course only, and probably did not include more than one or two courses. On the bedrock walls of this feature and on the inner face of the built-up wall were found remains of a thin layer of pinkish plaster mixed with many tiny pottery grits – a technique characterising the Byzantine period. another, thinner wall (W13*; excavated by the iaa) made of small ashlars and fieldstones bound with a whitish mortar was built inside a shallow depression in the bedrock surface to its south. its 40 nature indicates that it is contemporary with W19. the above feature was found full of fallen stones and earth (loci 133*, 149*, 162*) which contained some pottery sherds. two of the sherds (a basin and a jar lid) were bound with mortar and probably originated in the make-up of the floor or walls built on a higher level. they give additional support for dating the tower’s construction to the 6th century. the rest of the basement is a larger, roughly trapezoidal, room (locus 148; 2.3×1.6-3.3 m). in its northeastern corner four crushed storage jars were found in situ (Fig. 2.34). the jars belong to a type dated to the late 6th-early 8th century. Chapter 2: StratIgraphy and arChIteCture together with the jars were some ceramic lids (one of them was still stuck in a jar’s mouth), one stone lid and a coin from the 4th century (Figs. 2.35, 3.13:5-8, 5.3:1, Chapter 7: No. 12). all these finds were covered with a thick layer of yellowish earth. this material seems to be a finely-crushed chalk, of the same type as the site's bedrock. the room’s contents were sealed by a different layer of greyish earth (locus 134) which contained fragments of a mosaic floor made of white medium-sized tesserae (average measurements 1.5×1.5 cm, 56 stones per cm²), pottery sherds and fragments of roof tiles. Sherds of a bowl, a basin and a jar were bound with mortar like those described above, and probably belong to the make-up of the floor or walls. the latest types represented by these sherds are dated to the mid-6th to 8th century, and therefore push-up the terminus post quem of the tower’s construction to the mid-6th century. in this locus, and in the upper layer of locus 148, some regular pottery sherds were also found, the latest being dated to the 7th century (Fig. 3.14:1-6). the above-mentioned assemblage seems to represent a storage basement, which in its last phase of use contained four storage jars. the jars were crushed when the ground floor above them collapsed. this basement’s ceiling carried the mosaic pavement of the ground floor. the scanty remains of this mosaic, the few sherds and the Fig. 2.32: the northern doorway of the tower, looking east. roof tiles found together with it are the only finds remaining from the ground floor, and they can date its collapse to sometime in the 7th century. However, we cannot say for sure if this collapse occurred at the end of Phase iia or sometime later. the yellowish material that surrounded and covered the jars that were stored in the basement could be the bed into which the jars were put, maybe in order to keep the temperature of their contents (wine or olive oil?) stable. the existence of roof tiles within the ground floor collapse indicates that the northern wing of the tower consisted of no more than one floor. the lack of stairs in the basement indicates that the access into it from the ground floor was probably by a wooden ladder (Fig. 2.36). the second doorway of the tower (1.2 m width), located in its southern wing (Fig. 2.37), was completely different from the northern doorway. its nature indicates that it was built or rebuilt during Phase iiB or Phase iii. First of all, it can be clearly seen that a breach was made in the tower’s eastern wall in order to build this doorway. the builders even uprooted some of the paving stones of the tower, in a small area against the doorway, which remained paved only by the natural bedrock. the doorway does not have elaborate doorjambs, except for the wall’s ashlars themselves. its threshold was not made of a flat Fig. 2.33: the northwestern corner of the tower (the basement), looking east. 41 Fig. 2.34: Crushed storage jars in the northeastern corner Fig. 2.35: the repaired storage jars and some lids from the of the basement (l148). basement. monolithic stone, as was usually the case, but of two ashlars in secondary use. these do not even have the same measurements. the door’s socket was carved in the northern ashlar. the threshold stones were founded over a thin layer of whitish mortar in which was found a coin of Maximinus trax (235-238 Ce; Chapter 7: No. 4). However, this coin makes no contribution to the dating of the doorway. there is not enough evidence to determine whether the southern doorway was built during Phase iiB (maybe after the northern wing of the tower collapsed in the earthquake) or during Phase iii. the tower, with its massive walls made of large well-dressed ashlars, was without doubt one of the most impressive parts of the monastery. its thick walls indicate that its southern wing was raised to a height of two or three stories (ca. 6 to 10 m high), while the northern wing included probably only a basement and a ground floor (Fig. 2.38). Fig. 2.36: Reconstruction of the basement. (Drawing by Yura Smertenko) 42 Chapter 2: StratIgraphy and arChIteCture ROOM 23 east of the tower in Square D10, between it and Cistern 24, was a trapezoidal space (Fig. 2.1:23; locus 167; 3.3×5.8-6.6 m), which was poorly preserved. Only some small carved stones, probably belonging to a floor, were found in its southern part. in addition to these scanty remains were found parts of a crushed jar, identical to those found in the basement to the east. From these remains it cannot be said what was the nature of this small space, whether a living room or another storeroom. Fig. 2.37: the southern doorway of the tower, looking east. Note the Phase iii wall (W1) on the right, with a reused oil press beam weight. Fig. 2.38: Reconstruction of the tower, looking southeast. (Drawing by Yura Smertenko) 43 Itamar taxel THE INdUSTRIAl UNIT the domestic unit was bordered on the north (Squares B-D/9-10) by complexes of a different nature. in this area a relatively hard layer of nari limestone covered the much softer chalky bedrock. the enclosure walls of the monastery surrounded it on the north and east. the northern wall (W16) is the most massive and impressive, being 1.3 m wide and originally ca. 29 m in length (its eastern end was found destroyed). it was built on the lowest part of the site, where the bedrock gently slanted westward. the foundation, and the first courses of W16 were built of large, coarsely carved ashlars, sometimes replaced with smaller fieldstones. However, its upper courses were not uniform all along its route. along the first 19 m from the east, where the wall, up to a certain height, was not freestanding but leaned against the soft bedrock, the upper courses were built of an outer face of ashlars and an inner face of small fieldstones bound with whitish mortar. Since the wall was not preserved above its third course, we do not know how it was built in its free-standing part. its total height is also unknown, but it had to pass over the level of a floor which was found ca. 3 m above its current level (see below). the CISternS the northeastern corner of the monastery (Squares D-e/9-10) was occupied by two water cisterns (Figs. 2.1:24, 25; 2.2). Cistern 24 was hewn in the early Roman period and renovated in the late Roman/ early Byzantine period. it continued to be used also in the late Byzantine and Umayyad periods, and was in good condition. this is attested by the fact that no silt layer was found on its floor and no new layer of plaster was added to its walls since the late Roman/early Byzantine period. the cistern was probably constantly cleaned and maintained by the inhabitants of the site, until its complete collapse. it was found filled with large ashlars and smaller stones, which belonged to its upper courses and ceiling. this massive collapse (locus 175) contained some pottery sherds, a complete lamp and other finds, mainly from the late Byzantine to the >abbasid periods (Figs. 3.14:7, 3.29:1). However, 44 which of these artefacts fell down when the cistern collapsed and which were washed/dumped inside later could not be determined. therefore, whether the collapse was caused by the earthquake around the mid-7th century or occurred later in the >abbasid period remains unsolved. Cistern 25 (Fig. 2.39) was built perpendicular to Cistern 24, adjacent to W33. it was hewn into the soft limestone and has stone-built walls. this cistern, however, was rectangular in plan (internal measurements 4.1×2.6 m). it was not excavated down to its floor, due to technical difficulties and danger of collapse. its known depth is 4.3 m, but it seems that its floor is not much below the excavated level. its capacity, therefore, was at least 48 m³. the construction date of this cistern can be traced by the pottery sherds found in the seam between its eastern wall and the eastern wall of the monastery (locus 505), which is very thick at this point. this assemblage included, indeed, only body fragments of storage jar types which did not appear before the mid-6th century. it was, therefore, was built much later than the more northerly cistern. the fact that it was built parallel to the complex’s eastern wall is another indication of its construction being part of the planned late Byzantine monastery. the picture reflected from this cistern is quite different from that seen in the northern one. the nature of its inner stratigraphy, the finds connected to it and Fig. 2.39: the eastern cistern (25), looking east. Chapter 2: StratIgraphy and arChIteCture to its latest layer of plaster belong to Phase iii (and partially also to Phase iiB), and therefore will be described below. the two cisterns were separated by a ridge in the bedrock. its southwestern end was carved to form a low flat surface to the south of which was a small square plastered basin (0.5×0.5 m, 0.1 m deep). it is not certain whether this shallow basin was connected to the oil press complex or whether it was part of the drainage system which carried rain water from the large courtyard area (8) to the cisterns. the latter option may be inferred from the two pipes found in situ to its west, which were probably connected to the drainage channel in the south. the southeastern part of the bedrock surface remained rough and at some stage it cracked into three large pieces, two of which slid down to the north until they covered the southern wall of Cistern 24. On the north of this bedrock surface was a large boulder which had five steps (1 m long, 0.3-0.6 m wide, 0.15-0.2 m high) carved into its northern side (Fig. 2.40). Sometime later, parts of the two lower steps broke away. at the top of the staircase was an oval cup mark (0.2×0.15 m in diameter, 0.1 m deep). the top of the southern part of the boulder was crudely worked to form a f lat surface on the edge of which a rectangular niche, whose purpose is unknown, was carved. a step (1 m preserved length, 0.28 m wide, 0.15-0.3 m high) was carved into the western edge of the boulder, just above Cistern 24, similar to that carved in the bedrock surface in the tower’s northern wing. this step could perhaps have carried part of the roofing of the cistern, and therefore was probably hewn already in the early Roman period. the position of the staircase suggests that it led directly to the cistern’s opening. the staircase was probably hewn at the same time as Cistern 24 and provided access from all directions. after the foundation of the late Byzantine monastery and the construction of the enclosing walls and Cistern 25, the staircase probably went out of use. the access to the cisterns’ opening seems to have been from the south and/or west. Fig. 2.40: the rock-cut staircase near the cisterns, looking southwest. the oIl preSS the oil press in Squares D-e/8-10 was built on at least two levels, in a sloping part of the site (Fig. 2.41). it is bounded on the north by Cistern 25, on the east by the enclosure wall of the monastery, on the south by Rooms 11 and 12 and on the west by Courtyard 8 and Room 11. it comprises a central, roughly rectangular unit (9.2×6.7 m) and another narrow long room (7.7×1.8 m) adjacent to its southeastern corner (27). the total area of the oil press, therefore, is ca. 75 m². Unfortunately, the central space of the oil press area is one of the mostly severely damaged areas in the site, probably due to a combination of natural erosion and modern damage. However, the preserved remains allowed us to restore the original plan of the oil press almost entirely. Nothing was preserved of the northern wall of the oil press, although it may have somehow been combined with the southern wall (W34) of Cistern 25. 45 Itamar taxel Fig. 2.41: Plan of the oil press. Fig. 2.42: the entrance staircase of the oil press, looking west. 46 Chapter 2: StratIgraphy and arChIteCture the southern wall of the main unit (W24) was 5.5 m long. it was built of two rows of ashlars, and preserved to a height of one course (0.45 m). exactly in its middle was a doorway, of which only the threshold (1.2 m wide) was preserved. From this three shallow steps led down to the oil press, the uppermost being much wider than the two below it (Fig. 2.42). the staircase (1.5 m long) was built along a short wall (W26; 3 m long) faced with ashlars, some without a doubt in secondary use. the western wall (W22) was built of one row of well-dressed and smoothed ashlars. among these were also some architectural elements in secondary use, such as part of a doorjamb with a hole for a bar and maybe a threshold. the wall is 6.7 m long, and curves slightly towards the southeast. it was built against a cut made in the soft bedrock of the higher part of the slope. the narrow space between them was filled with large fieldstones and reddish-brown soil. Wall W22 was preserved to a maximum height of three courses (1.4 m). a floor made of well-dressed ashlars abuts on the lower course of the northern half of the wall (locus 184). Here too, some of the paving stones were architectural elements in secondary use. the ashlar-paved area measures 2.75×1.7 m. On the north it is delineated by a cut bedrock step (1 m high), which was covered with thin white plaster. On the east the floor is delineated by the upper collecting vat of the oil press and the pavement which abuts on it. the level of the pavement around the collecting vat (locus 263) is 0.2 m higher than the ashlar floor. a line of small ashlars separates these two floors. On the south, the ashlar floor ends in a wall (W23) made of four ashlars similar to those used in the construction of the floor and the western wall. it is preserved to a height of one course (0.5 m), but may have had more. two additional ashlars joined the eastern end of this wall to the more massive W22. the area south of the ashlar floor, which was surrounded by W22, W24 and W26, was unpaved. in the middle of this area a Roman plastered installation, which was cut by W22, was unearthed at a lower level. On the fill that sealed this installation, between W22 and W26, was a thin wall (W25) made of ashlars and small fieldstones, preserved to a height of two courses (0.8 m). the westernmost part of the oil press was thus divided into three compartments, with the two south of W23 being unpaved and having only a levelled earth floor. the pottery sherds found in the fill (locus 158) which covered the area south of the ashlar floor are generally dated to the 6th-mid 8th centuries. therefore, it cannot be said if W23 and W25 were built during Phase ii or only in Phase iii. the dating of the ashlar-paved floor is even more problematic. When removing this floor during the excavations, it was found that the paving stones were founded on a levelled layer of crushed chalk and stone chips. Unfortunately, no dateable finds at all were found below the floor. the finds found above the floor, as a later wall built over it, belong to Phase iii (see below). the dismantling of W26, however, yielded some pottery sherds dated generally to the mid-6thmid-8th centuries that were found in the earth which filled the gaps between the wall’s stones (locus 504). these sherds give a mid-6th century terminus post quem for the construction of this wall and most probably other features, such as the southern wall (W24) and the staircase built in the junction between it and W26. the oil press included one crushing basin and two pressing systems, built on two different levels, which probably functioned contemporaneously. the crushing mill was located in the southern part of the main unit of the oil press, ca. 1 m from its southern wall. it included a low oval base (locus 165) made of small and medium-sized fieldstones and flat dressed stones (1.8×1.3 m in diameter; 0.3 m height) filled with earth. On this base was originally laid the monolithic crushing basin, which was found broken into four pieces and discarded ca. 1 m away from it, in the fill covering the entrance staircase (Fig. 2.43). the concave crushing basin (1.3 m in diameter; 0.25 m thickness) was made of hard dolomite limestone (Fig. 2.44). its rather thin measurements indicate that it was used for a long time, during which it became slowly worn. it is 47 Itamar taxel very reasonable, therefore, that the crushing basin originated in the earlier Roman oil press. in the middle of the basin was a central square sunken socket (0.15×0.15 m), into which was affixed the vertical pivot attached by a horizontal shaft to the crushing stone. the crushing stone itself was not found, probably because it was robbed later, after the oil press was turned out of use. the floor around the crushing mill was probably made of medium-sized ashlars, as indicated by one stone which remained in situ south of the oval stone base. the pressing technique used in the late Byzantine/Umayyad oil press was that of the lever and screw. the cylindrical stone screw weight (into which was affixed the wooden screw) of the upper pressing system was not found in situ, but discarded on its side outside the southwestern corner of the oil press (Fig. 2.45). this weight (1.2 m high; 1.3 m in diameter; ca. 3.8 tons in weight), which was carved from nari limestone, belongs to a subtype of Frankel’s 'Samaria' screw weight (1999:111-113, and see Chapter 10). it has a central round socket (0.2 m in diameter) and two external dovetail mortises, of which only one remained undamaged (0.55 m long; 0.15 m minimum width; 0.3 m maximum width). another round socket (0.15 m in diameter) was hewn also in the middle of the bottom end of the weight, but its function is unknown. the original position of the weight was, in our opinion, in the southernmost cell of the western ‘corridor’. it probably stood on top of the earth fill which covered the older plastered installation, adjacent to W25. the screw weight was uprooted from its original position during Phase iii, when at least part of the oil press ceased to be used. this can be inferred from the fact that the weight was embedded where the wall which originally divided between the large courtyard and Room 11, and is now missing, ran. almost half of the weight, as well as one of the dovetail mortises and the surface around the upper socket, was intentionally broken before or after it was uprooted from its original place. the collecting vat of the upper pressing system (locus 194) is placed 2.2 m northwest of the 48 crushing mill. this rounded vat (height 1.2 m; depth 0.6 m; external diameter 1.5 m; internal diameter 1 m) carved from nari limestone was inserted almost entirely (excluding its upper 10 cm) into the chalky bedrock (Fig. 2.46). its volume is 0.47 m³ (470 litres). the vat has a barrel-shaped section, and a rounded settling pit (0.3 m in diameter; 10 cm deep) was hewn in its base. the rim of the vat is stepped, and bears eight radial grooves to channel the expressed liquid into the vat. a well-dressed, sloping ashlar paving abuts on the vat’s rim on the south. it seems that the paving originally surrounded the vat also from the east and maybe even from the north, however most of its stones were extracted after the oil press went out of use. No pressing base was found together with this vat, so presumably it was robbed. alternatively, it is possible that in this instance the baskets holding the crushed olives were placed on a perforated wooden board over the rim of the vat, as is known from other Palestinian oil presses. the wooden beam used in the pressing process must have been affixed inside a niche in the northern, unpreserved wall of the oil press, behind the collecting vat. if our assumption about the original position of the screw weight of this pressing system is correct, then the beam would have been ca. 7 m long. Some signs which connect the western ashlar paving and the sloping paving adjacent to it might indicate that these floors were built only during Phase iiB. First, the sloping paving is partially laid over some horizontal ashlars. When this was dismantled, an Umayyad-period oil lamp fragment was found in the shallow fill beneath them (locus 187). it seems, therefore, that sometime between the late 7th to early 8th century, part of the horizontal floor which surrounded the collecting vat was raised to form a slightly sloping paving. a more problematic find is the large moulded stone which was reused as a paving slab in the western floor of the oil press. this element (0.72×0.6×0.3 m) has a moulded corona-like profile with a cavetto on one of its narrow ends. the profile is divided into three panels – the topmost is vertical and straight, the middle one is slightly concave and the bottom one has a protruding obliquely- Chapter 2: StratIgraphy and arChIteCture Fig. 2.43: the western part of the oil press: Above) looking northeast. Below) Section D-D (see Fig. 2.41), looking west. 49 Itamar taxel Fig. 2.44. the crushing basin of the oil press. Fig. 2.45: the screw weight of the upper pressing system. 50 Fig. 2.46: the collecting vat of the upper pressing system, looking east. Chapter 2: StratIgraphy and arChIteCture pointed cross-section (Fig. 2.47). this stone was originally either part of a cornice or architrave, or a column/pilaster capital. although an almost identical element, identified as part of a cornice, was found in relation to a public building from the late 1st-early 2nd centuries unearthed at îorvat >etri (Zissu 2001:Fig. 88:3), there is no indication that such a building existed at Khirbet es-Suyyagh during the Roman period. in the present case, the element was identified as characteristic of the Byzantine period. Pilaster capitals, cornices and architraves with similar profiles were found, indeed, in several late Roman/Byzantine public and private buildings in Palestine and transjordan, including churches (alliata 1994: No. 11; Figueras 2004a: Fig. 18:6; Negev 1988: Ph. 91). it is unlikely that this heavy element was brought to the site to be reused as a paving stone. therefore it originated in the site itself. the most solid evidence for the use of pilasters of any kind was found in Room 5 in the southern part of the domestic unit. the width of the pilasters in this room fits that of the discussed element, so it is not impossible that it originated in that room. However, it is not less reasonable that this stone, whatever its prior use, was taken from the church building. if so, we should date the fine paving of the oil press no earlier than Phase iiB, when the church, like other parts of the monastery, were renovated. the lower pressing system was built along the eastern wall of the oil press, ca. 1.5 m lower than that of the upper pressing system. the central part of the oil press complex was severely damaged, and the junction between the two pressing systems was not preserved. Most of the elements which belong to the lower pressing system were housed in a narrow room (27) projecting southward from the southeastern corner of the main unit of the oil press. in the middle of the western wall of this room (W30) was embedded a large beam weight which probably belonged to the oil press that existed here in the Roman period. in the northern Fig. 2.47. the moulded stone found in the pavement at the west of the oil press (photographed upside down). 51 Itamar taxel part of the room, on a crushed limestone floor (locus 140*; 4-6 cm thickness) stood the screw weight, of the lower press which was found in situ (Fig. 2.48), placed up against a thin wall (W29). although the weight was not really integrated in the wall, it was attached to it by small fieldstones. this weight, like that of the upper pressing system, was also carved from nari and is cylindrical (1.1 m high; 1.25 m in diameter; ca. 3.2 tons weight). However, it belongs to a much less common type. it has an internal dovetail mortise (0.55 m length; 0.36 m width; 0.25 m depth) surrounded by a rectangular frame (0.73×0.43 m; 7 cm deep), in addition to four more external dovetail mortises (0.46-0.56 m length; 0.12 m maximum width) (Fig. 2.49). according to Frankel’s typology, this weight represents a combination of two types – a sub-type of the 'Samaria' screw weight with a central round socket and four external mortises (1999:112), and a subtype of the 'Kasfa' screw weight with internal dovetail mortise surrounded by a rectangular frame and two external mortises (ibid.:114, and see Chapter 10). the thin earth fill below the crushed limestone floor (locus 180) contained few pottery sherds from the late Roman and the late Byzantine periods. the latest most dateable sherd belongs to a late Roman C/Phocaean Red Slip Ware bowl from a type dated to the late 6th-early 7th century. therefore, the terminus post quem for the construction of this floor, and maybe also the related walls, is the late 6th century. to the north of this cubicle was the pressing base and the collecting vat (locus 195). the vat (Figs. 2.50, 2.51), like that of the upper pressing system, was carved from a nari limestone (1.1 m high; 0.85 m deep; 1.3 m external diameter; 1 m internal diameter; volume 0.66 m³ = 660 litres). the vat has a cylindrical section, with a round settling pit (0.2 m in diameter; 0.1 m deep) gouged out of its base. However, the internal shape of the vat is not the only characteristic which differs from the vat of the upper pressing system. its rim is different, being plain and narrow instead of stepped and wide. Here too, three radial grooves were carved on the rim, and since part of the vat was broken, it is probable that the original number of grooves 52 was larger. Only the base of the vat was buried in a bedding of earth and small fieldstones. On the east, north and west it was surrounded by walls (W30, W33, W40). the northern, short wall (W40) was originally the niche for the press’ beam. Based on the distance between the centre of the screw weight and the northern wall, the beam’s length is estimated at ca. 5 m. the pressing base of this system is quite an important element. it was carved on one of the broad sides of a large beam weight which belonged to the earlier Roman oil press (Fig. 2.11). the rough surface of the weight was smoothed and polished, and a circular channel (0.95 m in diameter; 6 cm wide; 5 cm deep) was carved almost around its entire width. a short groove ran from the channel towards the weight’s edge, apparently, to direct the oil into the collecting vat. However, since the original large hole of the weight remained open, and the channel cut its top, it is more likely that this hole served as a much more efficient means of draining off the oil from the pressing base. the base itself was originally put on top of the collecting vat, but it was not found in situ, but fallen between the vat and W29. another element which probably belonged to the late Byzantine oil press is a beam weight of the reversed-t type, which was found embedded in a later wall (W1) built during Phase iii in the southern part of the tower (Figs. 2.37, 2.52). this weight (0.8 m long; 0.5 m wide; 0.55 m high) was carved of nari limestone. it has a reversed-t bore with rounded openings, the upper one is 0.19 m in diameter, and the side openings have a conical cross-section, 0.15 m external diameter and 8 cm internal diameter. as can be seen, this weight is different from the two large, hard dolomite weights which were found re-used in the late Byzantine oil press and attributed to an earlier, Roman oil press that existed here. the fact, that it was re-used in the building of a wall dated to Phase iii and not earlier, strengthens the hypothesis that this weight is Byzantine and not Roman in date. examples of Byzantine oil presses which present a combination of screw weight and beam weights in the same pressing system are known from sites in Palestine Chapter 2: StratIgraphy and arChIteCture Fig. 2.48: the screw weight of the lower pressing system, looking southwest (courtesy of the israel antiquities authority). Fig. 2.49: the screw weight of the lower pressing system. 53 Itamar taxel Fig. 2.50: the collecting vat of the lower pressing system (front) and the older beam weight, looking south. (Chapter 10), so it is not unacceptable that this was the case also here. the roofing method of the oil press, and mainly of its central unit, is unclear. the few roof tile fragments found within its area do not necessarily attest to a gabled roof, although we should not rule out this possibility. For instance, at nearby Khirbet Fattir, a few roof tiles found in one of the oil presses led the excavator to reconstruct the building with a gabled roof (Strus 2003:140, Fig. 4.9) while at Sumaqa, the excavator preferred to reconstruct it with a flat roof (Dar 1999:97, Fig. 62). However, the smaller and narrower space of the lower pressing system probably had a flat roof. 54 Fig. 2.51: the collecting vat of the lower pressing system. the excavation of the oil press yielded three uncommon architectural elements made of limestone. these elements have the shape of small rectangular ashlars (0.45-0.6 m long; 0.23-0.3 m wide; 0.28-0.32 m high) with one of their narrower faces concave with straight edges (Fig. 2.53). Since all of these stones were found out of context, their original position and function in the building is uncertain. Similar elements were found in the monastery at tel Masos (Fritz 1983: Pl. 114:B), in two churches at Umm al-Rasas (alliata 1994: No. 15; Piccirillo 2003: Fig. 12), in the church on the summit of Mount Sinai (Dahari 2000: Plan 6:9), and in dwellings at Rafid in the Golan Heights Chapter 2: StratIgraphy and arChIteCture (Hartal 2005: Figs. 184, 185), where they were identified as window frames. it is possible that this was also the function of the present stones, but it cannot be said whether their original place was the oil press or the church complex, from which they were maybe taken. THE WESTERN UNIT in Squares B-D/6-8, between the tower and the church complex, was a rectangular complex, (Room 28; external measurements 18.3×10.8 m). it was surrounded by three thick walls, but only the foundations of the northern and western walls were partially preserved (W12 and W15; 1.3 m wide; 16 m total preserved length; Fig. 2.54). the original height of the walls is unknown, but their width and their clear role in the protection of the monastery indicate that they rose up to at least 2-3 m. Due to the bulldozing of the area in the mid-20th century. the Fig. 2.52: the small beam weight of the oil press. southern wall was not preserved, except for a few stones at its eastern end at the top of the staircase north of the main gate. it can be estimated that from this point the wall continued west for ca. 9 m to join the western wall. the foundation course of the western wall and the shorter northern wall were built of two rows of large fieldstones with smaller stones between them. the western wall runs along ca. 19 m, when it joined the northern wall that ended after 2.8 m against the southwestern corner of the tower. at this point a 1 m wide passage, controlled by the massive tower, led into the monastery. the StoreroomS Since modern bulldozing caused severe damage to this part of the site, only some of the remains in the western unit were preserved. in the northeastern corner of this complex was a small room (Figs. 2.1:29, 2.55; internal measurements 2.5×2.3 m), Fig. 2.53. three stone window frames (?) found in the oil press complex. 55 Itamar taxel with a doorway (1.15 m wide) in its southern wall. the doorway was slightly higher than the room’s floor, which was made of small and medium-sized flat stones. the floor was founded directly on the soft bedrock, and no finds were found below it in a section which was made through its eastern fringes. On the floor (loci 107, 120, 128, 131), together with some collapsed stones, were fragments of 16 storage jars, of the same type as those found in the tower’s basement. in addition to the jars, a jar lid and a glass oil lamp (Figs. 3.14:15-18, 4.2:14) Fig. 2.54. the enclosure walls of the western unit (W12 and W15), looking southwest. were found here. all these finds were covered by a thin layer of yellowish earth, identical to that which covered the jars in the tower’s basement. Here, too, this material (crushed limestone?) was probably used as the bed into which the jars were put. However, since these jars were only partially restorable, it cannot be said whether they represent an in situ assemblage or later dumping. at any rate, the destruction and abandonment of this room can be dated by the latest pottery sherds which accumulated above the collapse or penetrated into it, to not earlier than the mid-7th century (Fig. 3.14:8-14). the finds in the storeroom itself should be dated, therefore, somewhat earlier in the 7th century. the construction of this storeroom can be attributed to Phase iia, and it probably collapsed in the earthquake that ended this phase. 56 Outside the room, near its southwestern corner, were remains of a floor made of small and mediumsized slabs (locus 197), which was founded directly on the soft bedrock. One of the stones was a fragment of a basalt millstone in secondary use. On the floor, just outside the storeroom, were parts of another crushed storage jar. the northern wall of Storeroom 29 was built a small distance away from the southern wall of the tower and not attached to it. Only its northwestern corner was attached to the tower’s foundations. it seems, therefore, that the storeroom and the floor (whose total area is unknown) were built somewhat later than the construction of the tower, but contemporarily to the construction of the monastery’s western wall. the nature of the architectural remains and the finds which were unearthed in the western unit indicate that during Phase iia there was here at least one storeroom for food commodities, and a paved courtyard which was apparently used for various daily tasks and maybe even for housing the monastery’s livestock. Sometime after the destruction of this storeroom, a new room was built 0.7 m to its south (Fig. 2.1:34; known length 3.3 m, width 2 m). the floor level of this room (locus 117) was ca. 0.3 m higher than that of the former. its walls were crudely built of large ashlars in secondary use, including the threshold fragment which probably belonged to the original subsidiary gate of the domestic unit. its floor was also made of coarse slabs of various sizes, clearly different from the smoother slabs used in the older storeroom. it is possible that Storeroom 34 was built to replace the collapsed Storeroom 29. Rather than remove the ruins and restore the storeroom, the inhabitants of the monastery preferred to build a new room from materials which could be pickedup all around. the fact that half the length of the secondary-used lintel of the subsidiary gate was hidden behind the northern wall of the new room, indicates that the new gate was much narrower than the original one – ca. 0.7 m wide. the entrance into the central unit was now made through a narrow corridor, located between the old and the new storerooms. the eastern end of this corridor was paved with small fieldstones. Chapter 2: StratIgraphy and arChIteCture Since Storeroom 34 was discovered very close to the surface, none of the few pottery sherds that were found above its floor can be helpful in its dating. this data came from another place – the rough fieldstone paving at the eastern end of the entrance corridor (locus 181), and a small earth pocket (locus 183) sealed below it. the fieldstone paving was founded over a thin earth layer which contained some pottery sherds. the latest type represented gives a late 7th century terminus post quem for the construction of the floor (Fig. 3.15:1, 2). Below the foundation layer a small pocket of reddish-brown soil was found against the foundation level of the new gate and the new room. the earth in this pocket was mixed with a large amount of pottery sherds, mainly partially restorable cooking vessels. the rest of the pottery is represented by single sherds, including rims of african Red Slip Ware and late Roman C/Phocaean Red Slip Ware bowls, which give a mid-7th century terminus post quem for the construction of the new doorway and probably also of the new storeroom (Fig. 3.15:3, 5). therefore, the construction of this room can be attributed to Phase iiB, and is connected with the large-scale renovation of the large courtyard and the subsidiary gate. THE WINE PRESS a wine press (35) was positioned 5 m west of the tower in Square B9. excavation of this area revealed the remains of two wine presses – one from the late Roman/early Byzantine period (Fig. 2.6) and another from the late Byzantine period. to the latter belongs a rectangular plastered installation (internal measurements 2.6×2 m), which was originally surrounded by four walls (Fig. 2.56). However, only parts of the southern and western walls and the inner mortar and plaster face of the eastern wall were preserved. the northern wall was completely robbed or demolished, either in antiquity or during the modern bulldozing of the site. the southern wall was built of an inner row of ashlars and an outer row made of small fieldstones. the western wall was built of two rows of ashlars with a fill of earth and small stones between them. all the walls were covered Fig. 2.55: Storeroom 29, looking north. on the inside with a layer of mortar, plastered with a hydraulic plaster (total thickness 0.1 m). it must be noted that the southwestern and southeastern corners were built at 90°, whereas the northwest and the northeast corners were slightly curved on the inside. the floor of the installation represents two stages of use. in the first stage the floor was a coarse white mosaic (locus 517) of large tesserae (2-4×3-4 cm, 9-12 stones in dcm²). the foundation of the mosaic was made of whitish mortar, which was laid directly on the bedrock (total thickness 8 cm). the tesserae were set diagonally, within a frame of four rows of tesserae laid parallel to the walls. in the northwestern corner of the floor was a small settling pit (0.4 m in diameter; 0.2 m deep), which was coated with plaster instead of with mosaic. the floor slants to the north, towards the settling pit, with a height differential of 5 cm. in the second stage the mosaic floor was completely covered with a layer of smooth hydraulic plaster (locus 515; 3-5 cm thick). the old settling pit was still in use at this stage. it too was replastered, and as a result its became somewhat smaller (0.35 m in diameter, 0.15 m deep). the new plaster floor slants in the same direction as the older mosaic floor, with a height difference of 8 cm. in both stages the original plaster coating of the walls was used. 57 Itamar taxel Fig. 2.56: the collecting vat of Wine press 35, looking south. although two stages of paving were identified in the installation, it cannot be determined whether the first one represents Phase iia and the later represents Phase iiB, or whether both floors were built during Phase iia. the dating of these remains can be determined by the ceramic evidence. into the mortar of the walls were embedded pottery fragments, which included mostly body sherds of bag-shaped jars but also two rims of basins (Fig. 3.15:4). these sherds give a late 6th century terminus post quem for the construction of the installation. the later plaster floor was found covered with large building stones (locus 510), which probably collapsed from the upper courses of the walls. in the earth accumulated between the stones were a few pottery sherds dated to the 6th-7th/8th centuries, which can give a general date to the abandonment and destruction of the installation. the measurements and nature of this installation, and the fact that it is not connected 58 to any other installation, indicate that this is the lower part of a collecting vat of a wine press. the rest of the wine press, including the treading surface, was most probably located to the south of and higher than the collecting vat. However, since this area was severely damaged during the modern bulldozing of the site, nothing was preserved from the southern part of the wine press. the closest built remains to the south of the collecting vat is the thick surrounding wall (W15) of the western unit (7.5 m away) which, as mentioned above, was itself preserved only to the height of the foundation course due to modern destruction. the level of this area is higher in ca. 1 m than that of the walls of the collecting vat. it seems reasonable therefore, that the other parts of the wine press were located somewhere in between these two features. Judging from parallels of other contemporary wine presses, it seems that the depth of the collecting vat was more or less twice the preserved depth, i.e., ca. 1 Chapter 2: StratIgraphy and arChIteCture m. this height was probably also the level of the treading surface. according to the approximate measurements of the collecting vat, its volume was ca. 5 m³ (5000 litres). access into the collecting vat was probably by means of footholds cut into the walls or protruding from them (although no such devices were preserved), since its plan included no steps. it must be noted that the late Byzantine wine press (or at least its collecting vat) was built in almost the same orientation of the nearby treading floor of the late Roman/early Byzantine wine press. Maybe south of that treading floor was another built feature (floor?) which was still visible in the 6th century, and dictated the orientation of the later wine press. another nearby built feature which can be dated to the late Byzantine or to the Umayyad period is a narrow wall (W50) that was built between the western wall of the tower and the eastern wall of the late Roman/early Byzantine wine press. the wall was built of small ashlars, probably in secondary use, and abutted in the south a low step hewn in the bedrock. From both sides of it was found a layer of brown earth (loci 513, 514) mixed with pottery sherds, fragments of roof tiles and glass vessels – all dated to the 6th7th/8th centuries. THE CHURCH COMPlEX adjacent to the southwestern corner of the monastery, in Squares B-D/4-6, was a separate complex. this consisted of a church and a large hall to its north (Fig. 2.57) which was very possibly the monastery’s refectory. the ChurCh the church (Figs. 2.57, 2.58) was built in basilical form. its external measurements, without the narthex, create a nearly perfect square (14×13.4 m, which are equal to 45×43 Byzantine feet [1 foot = ca. 31 cm]). together with the narthex the external measurements are 16.7×13.4 m (= 53×43 Byzantine feet). the nave was a rectangular hall (internal measurements 12×5.3 m), which ended in a projecting polygonal apse. North and south of the nave were two narrow aisles (9.5×2.1 m and 9.5×1.8 m, respectively). the ratio of the nave and the aisles is, thus, almost 1:3:1. West of the nave and aisles was a narrow narthex, divided by a wall into two unequal spaces (7.2×2.1 m and 3.4×2.1 m). the ratio of the internal measurements of the church, excluding the narthex (12×11.4 m, or 38×36 Byzantine feet), is almost 1:1. as a result of modern bulldozing, almost all of the church area was preserved only below floor level. Hence, no doorways were preserved. the plan of the church, however, indicates that it was accessible from outside the monastery’s main complex, which is an unusual plan (see Chapter 10). the only probable remains of the bema are the foundations of the northern end of a wall (W200), which abutted on the northern stylobate from the south, 4 m west of the end of the apse. Other elements of the altar are two fragments of the limestone chancel rails which were found in other parts of the site. One (0.8 m preserved length; 0.4 m wide; 0.35 m thick) was found in the collapsed material which covered the northern part of the large courtyard, just outside the oil press. it has a narrow groove (5 cm wide and deep) all along its broad side (Fig. 2.59). the second fragment (0.9 m preserved length; 0.5 m wide; 0.3 m thick) was found reused as a building stone in a wall built during Phase iii on the ashlar floor west of the upper collecting vat of the oil press (Fig. 2.60; and see below). On one of its broad sides there is a narrow groove (0.1 m width and depth) which ends in a roughly square depression (0.4×0.35 m; 0.1 m depth). these elements were identified as chancel rails based on parallels from numerous other churches. the foundation courses of the church’s walls were built of small to large fieldstones and coarselycarved ashlars bonded with whitish mortar. the upper courses of the walls were surely also built of ashlars, but these were almost all looted after the abandonment of the monastery. the main gate and path into the monastery from the south lay at a lower level than the church floor. therefore, its eastern wall was visible from the outside almost down to the foundations. the apse was better preserved than the northern and southern 59 Fig. 2.57: Plan of the church complex. 60 Chapter 2: StratIgraphy and arChIteCture Fig. 2.58. the church (30): Above) looking north. Below) Section C-C (see Fig. 2.57) through church, looking west. Fig. 2.59: Part of a chancel rail (?). parts of the east wall, to a height of 3 courses (1.3 m high). except for the roughly-dressed stones of the foundation course, the ashlars used for the building of this wall were large and well-dressed (Figs. 2.61, 2.62). among these stones was also a lintel (1.4×0.4×0.45 m) in secondary use, exactly in the middle of the apse (Fig. 2.63). Since the apse wall was later partly destroyed and rebuilt (see below), it cannot be said for sure to which construction phase the lintel belongs. the large amount of ceramic roof tiles which were found in the church area (Fig. 3.16:2-4), mainly covering the southern aisle and 61 Itamar taxel Fig. 2.60. Part of a chancel rail (?) reused in a Phase iii wall (W27) built in the oil press. 62 Chapter 2: StratIgraphy and arChIteCture Fig. 2.61: the central wall of the apse (W220), looking northwest. Fig. 2.62: the southern (repaired) wall of the apse (W223), looking northwest. 63 Itamar taxel the area to its south, indicate that the church had a wooden gabled roof. it must be noted that the stylobates were equal in width to the external walls of the church (ca. 1 m). Only the dividing wall between the narthex and the church and the inner wall of the narthex were slightly thinner. although the stylobates were preserved to the floor level, the method by which the roof was supported can only be partially reconstructed. three fragmentary and complete column bases and two column fragments, all made of nari, were found in different places at the site and are attributed to the church stylobates. No column capitals were found. One column fragment, 0.35 m in diameter, was found in the topsoil between the large paved courtyard and the oil press (Fig. 2.64). the second column fragment is actually an engaged column (0.43 m in diameter, 0.31 m width, 0.5 m preserved length), which was found in the foundations of the lower floor in Room 11 that led to the oil press (Fig. 2.65). One column base was found in the collapse which filled the eastern cistern. it has a square pedestal (0.37×0.37 m) a scotia and two tori (0.35 m in diameter), and its total height is 0.3 m (Fig. 2.66). a second column base was found after the excavations in a heap of earth and stones which were removed by bulldozers from the northern part of the site. its original place, therefore, is unknown. this base belonged to an engaged column. it has a rectangular pedestal (0.6×0.32 m), a scotia and two tori (0.45 m in diameter, 0.32 m wide), and its total height is 0.32 m (Fig. 2.67). a third base is represented by a small fragment, similar in size and measurements to the first base, which was also found only after the excavations. all these bases belong to the attic-ionic type. as can be seen, the two column fragments and the first column base have the same diameter. However, it seems that the base of the engaged column, although a little wider than the above-mentioned engaged column, belongs together with it. engaged columns are pilasters at the ends of stylobate walls, as in the church at îorvat Berachot (tsafrir and Hirschfeld 1979: Fig. B), or in other parts of monumental buildings, e.g. the monastery at Mount Nebo (Saller 1941: Fig. 6:F, 64 Pl. 32:2) and the synagogue at îorvat >anim (amit 2003: Figs. 36.5:9, 37.5:12). a large moulded stone, which might be a pilaster capital may also have been associated with the roofsupporting system. this stone was also found in secondary use in a later floor dated to Phase iiB. another possible pilaster capital or cornice fragment, made of marble, was found on the surface in Square C6. this is a thick slab (9 cm), smoothed on its lower side, with a diagonal, slightly concave profile (Fig. 2.68). Similar elements, made of limestone (Hirschfeld 2000a: Figs. 79, 80; Patrich 1988b: Pl. 6:5, 9, 10) or marble (Hirschfeld 2000a: Figs. 82, 83) are known from other secular and religious Byzantine buildings as cornice stones on the top of walls or pilasters. Other architectural elements which perhaps belonged to the arches supported by the columns are two large voussoirs. One was reused in the later restoration made in the apse wall (see below), and another was found in the collapse of the church’s northern wall. as said before, almost all of the church walls were preserved only below floor level. Some white, red and grey tesserae (1×1 cm average measurements) that were found in some places within the church suggest that a coloured mosaic once covered parts of its floors. in addition, a fragment of a large marble slab (6 cm thick), polished on one side only (Fig. 2.69) which was found in the topsoil layer of the oil press complex (locus 256), perhaps also belonged to the church’s pavement. it cannot be said, however, if these coloured tesserae and marble slab represent the original pavement of the church (of the nave?), or whether they belong to the later renovation of the church during Phase iiB. the only floor preserved in situ is a small section of white mosaic made of medium-sized tesserae (2×2 cm average measurements; 25 stones per dcm²), unearthed in the eastern part of the northern aisle (locus 384; Fig. 2.70). the mosaic was founded over a thin bedding of small pebbles, and abutted on W204. in addition, it was clearly laid over at least part of the northern stylobate, and abutted on W200, the altar’s curtain wall. Fig. 2.63: a lintel reused in the central wall of the apse (W223). Fig. 2.64. a column fragment. Fig. 2.65. Part of an engaged column. Fig. 2.67. Base of an engaged column. Fig. 2.66. a column base. Fig. 2.68. Part of a marble pilaster capital or cornice. 65 Itamar taxel in a section through the mosaic (loci 387, 391) it was found that under the floor’s bedding was an earth layer (0.1-0.2 m thick) which contained some pottery sherds dated to the late 6th to late 7th/early 8th centuries, a few roof tile fragments and a coin dated to the end of the third quarter of Heraclius’ reign (629/630 Ce; Chapter 7: No. 34). Based on this date and the fragments of roof tiles, this floor is dated to Phase iiB. Below this earth layer was a thin layer of crumbly white plaster (locus 389; 5 cm thick) which lay directly on the bedrock and abutted on the foundations of the northern stylobate. it cannot be determined if the plaster layer was part of an older floor (Phase iia) or part of the church’s foundations. in any case, it does not looks like a bedding of a mosaic floor. Since almost none of the floors were preserved, there is difficulty in attributing the finds which were found within its area to a particular phase of use. Most of the pottery sherds found within the walls of the nave and aisles (loci 304, 306, 308, 309, 313, 316, 318, 320, 327, 328, 333, 340) can be dated to a long time span, between the mid6th and the mid-8th century (Fig. 3.15:6-9). it seems, therefore, that modern destruction caused penetration of later finds to below the floor level of the church. More solid evidence for the date of construction of the church was the pottery found in a trench which was excavated adjacent to the inner face of the northern wall of the apse (locus 349). Fig. 2.69: Fragment of a thick marble slab (l256). 66 the trench was excavated down to the foundations of the apse which were deeper than those of the rest of the walls, so pottery from this level was not disturbed or mixed with later finds. the latest sherd found here is a rim of a late Roman C/ Phocaean Red Slip Ware bowl, which dates to the late 6th-early 7th century (Fig. 3.16:5, 6). the late 6th century, therefore, is the terminus post quem for the construction of the church. as mentioned above, the apse reflects a major renovation carried out at some time. the nature of construction of W223 and part of W220 is clearly different from that of W211. Only the uppermost course of W220, or at least its southern and central three stones, appear to be part of the renovation phase. the central stone was, as mentioned above, a lintel in secondary use. the lower courses were apparently not destroyed, since they are similar to those of W211. the upper courses of W223 were rebuilt from ashlars and architectural elements in secondary-use, such as a large vossoire. these were not necessarily arranged in an orderly manner nor bonded with mortar and small fieldstones. it can be clearly seen that about half of the apse wall had collapsed at some time and been hastily rebuilt, using some of the original building stones as well as stones which were roughly cut ad hoc. Due to the bad preservation of the building we cannot know if other walls were also damaged at the same time, but the destruction of the apse alone probably caused damage to other parts of the church, such as the roof above it. However, if the monastery continued to function even for a short time after the earthquake, it is very reasonable that an effort was made to reconstruct the church, even in a poorer quality than before. Since none of the column and pilaster components were found in relation to the church’s original phase of construction, but in later contexts, it is clear that the renovation of the building during Phase iiB applied also to the supporting system of the roof. it is possible that in this phase, the columns and bases were replaced by pilasters, still using the stylobate walls of the previous phase. However, excluding probable poor remains of a pilaster, rising from the northern stylobate (ca. 1.5 m from its western end), no other evidence can confirm or refute this suggestion. Chapter 2: StratIgraphy and arChIteCture Both were preserved to a height of one course. the northern wall, W137, was preserved to a height of 1-2 courses. its foundation course, which was deeper than those of the other walls, was built in the same way as those of the church’s walls – from small to large fieldstones. the second course was built of two rows of ashlars. None of the walls of Hall 31 include any doorways, and since the western wall was preserved only to the height of the foundation course, it is probable that entry to this hall was through Room 32. However, the southern edge of a wall which abutted on W137 from the outside maybe hint at the existence of a doorway at this point. Within Hall 31 were several built elements. Parallel to W137 and W208 were two internal walls (W138 and W205, respectively) . their outer faces (i.e., facing north and west) were made of ashlars and their inner faces of small fieldstones. like some of the walls of Hall 31, these walls were founded above the fill which levelled the floor. the walls marked out an area (7.5×5.2 m) which was probably entered from the northeast, through a gap (1.4 m width) between the end of W138 and W207. in the eastern part of the delineated area, equidistant (1.7 m) from W138 and W201, was a round base (locus 310; 1.65 m in diameter; 0.7 m high) built of one course of ten ashlars surrounding a core of earth and small to large fieldstones (Fig. 2.71). this base was founded on the bedrock which was levelled by a thin fill. its function is uncertain. two similar bases were unearthed in the oil presses at Mevo Modi>im (Umayyad/>abbasid period; eisenberg and Ovadiah 1998:10*, Fig. 3) and nearby Khirbet Fattir (Byzantine period; Strus 2003:143, Fig. 4.1), where they were used to support the crushing basin. However, these are exceptional cases, since in most oil presses the crushing basins were laid directly on the floor or on a much lower base made of small fieldstones. even in the oil press built at the site itself, the base of the crushing basin was only 0.3 m high. On the other hand, such a base could have been used to support a large horizontal basalt mill. Such a feature, which was Fig. 2.70: Remains of a mosaic floor (l384) in the church’s identified indeed as a base for a flour mill, was northern aisle, looking northwest. found in a rural monastery near Naúal Kidron the refeCtory and KItChen another unit stood adjacent to the church on its north. its external measurements create an asymmetrical rectangle (16.7×8.4-9.2 m). its is divided by W208 into a large rectangular Hall 31 (10×7.4-7.6 m) and a smaller rectangular area, Room 32, (7.9-8.3×4.1 m) to its west. Room 32 was badly preserved. Only the foundation courses of its northern, eastern and southern walls and less than half of the foundation course of its western wall remain. No traces of doorway/s were found, but it seems that it could be accessed through Hall 31 and perhaps also from outside through its northern or western walls. its most logical function would have been to serve as the monastery’s refectory (refectorium; ). Hall 31, on its east, is identifiable as the monastery’s kitchen, so the existence of a dining room beside it is appropriate. in addition, the area of Room 32 (34 m²) has enough capacity for the estimated number of monks living in the monastery (see Chapter 10). the area on which Hall 31 was built sloped down slightly towards the east, so that it was levelled with thin fill (up to 0.3 m thickness) composed of earth and a few ashlars. this fill was the continuation of that under the foundations of the corridor (7) running northward from the gate. Wall W207 and the eastern part of W201 were founded on it. 67 Itamar taxel (Hervé and Zelinger 2006:289, Fig. 1). in an early 20th century photograph, a similar structure built of three courses of ashlars (height unknown) is seen as a support for a flat circular basalt millstone laid above a wider limestone base. in this case, the mill was used to produce sesame oil (avitsur 1994: Fig. 111). Fragments of a large basalt mill were found in an >abbasid-period industrial complex at Khirbet Deiran, beside elements which belong to an oil press. therefore, the excavator of that site suggested that here the mill was used for crushing the olives (Buchennino 2007:123) although it could equally have been used for making flour. if indeed the round base carried a mill, and if it and the two inner walls were contemporary with the building of Hall 31 itself, it seems reasonable to identify the hall as the monastery’s kitchen. this lends credence to the above suggestion that Room 32 was the refectory. the building of public dining rooms beside or close to the church in monasteries is a phenomenon which is well documented in the literature as well as archaeologically. the exact arrangement of the hall is unknown, mainly because it continued to be occupied and remodelled during Phase iii. its roofing method is also unknown, since no columns or pilasters were found within its wide area. However, if the two inner walls were built up to ceiling height, it may be assumed that they supported the roof. the area between W138 and W137 (loci 278, Fig. 2.71. Round installation in Hall 31, looking northeast. 68 282) was found filled with collapsed building stones mixed with earth which contained pottery sherds and other finds, including two coins of Constantius ii (337-361 Ce; Chapter 7: No. 10) and Heraclius (610-641 Ce; Chapter 7: No. 33). the pottery and the latter coin points to a 7th century date for this assemblage. the fill south of W138, above which were found some of the walls and elements of Hall 31 (loci 321, 322, 331) contained a coin dated generally to the 5th century (Chapter 7: No. 20) and pottery sherds dated to the 7th century onwards (Fig. 3.16:7-9). However, since the definition between this fill and the one which was later accumulated above it was not very clear, and since a few sherds from the >abbasid period were also found in it, more accurate information cannot be derived from this assemblage. it seems that parts of Hall 31, namely its eastern wall, the eastern end of its southern wall and its two inner walls, were rebuilt sometime in the 7th century. this activity, therefore, can also be connected to the reconstruction of parts of the monastery during Phase iiB. THE AGRICUlTURAl ENVIRONS Most of the natural landscape which surrounded the site has been extremely changed and damaged by modern development, mainly by construction on the eastern fringes of the town of Beth Shemesh and the foundation of the nearby Moshav Maúseya. at the time of the excavations, only a relatively small area remained untouched. this area includes the northern and eastern slopes of the spur on which the site is located, and the eastern end of another spur, ca. 100 m southeast of the site. in a previous survey conducted north of the site by the iaa, the rock-cut treading floor of a wine press was identified, with two hewn basins nearby (Dagan, forthcoming). in a small-scale survey conducted west of Moshav Maúseya, a few archaeological remains which seem to be connected to the site of Khirbet es-Suyyagh were found. in an area 250-500 m north of the site were cuts in the bedrock (probably parts of agricultural installations, maybe even those reported by Dagan), and scatterings of Byzantine-period pottery sherds, part of a terrace wall and a cistern were found less Chapter 2: StratIgraphy and arChIteCture than 100 m southeast of the site (Greenhut, Weiss and Solimany 2000). Following this survey, a smallscale salvage excavation was conducted on the slope where the above-mentioned terrace wall was found. a segment of the wall, 9 m in length, was unearthed. the wall was built of one course of medium-sized fieldstones, arranged in two rows, and a few pottery sherds from the Byzantine period were found near it (Greenhut 2001). During the present excavations, the spur south to the site was surveyed. the top of the spur is relatively flat, and its steep slopes start almost abruptly. the spur is covered by rendzina soil from which emerge nari surfaces and boulders. the whole area is covered with dense natural vegetation and man-planted pine trees, so the identification of ancient remains is very difficult. Nevertheless, at least five main features were found on the top of the spur and its northeastern slope. the first feature is a bell-shaped cistern (5 m in diameter), probably that which was identified in the iaa survey. it was hewn into the relatively soft bedrock in the flattest area at the top of the spur. the cistern’s ceiling is now collapsed, and it is full of silt and stones to at least half of its original depth. the walls of the cistern were covered with one layer of pinkish hydraulic plaster (2 cm thick) containing a large amount of tiny pottery fragments – the characteristic plaster of the Byzantine period. the second feature is a wine press, located 50 m east of the cistern, at the top of the northeastern slope of the spur (Fig. 2.72). this installation was hewn entirely in the hard bedrock. it includes an almost square treading surface (ca. 5-5.65×5 m, ca. 29 m²) which now is completely covered with silt. Some large white tesserae found on the ground above the treading surface indicate that it was paved with a coarse mosaic, characteristic of wine presses of the Byzantine period. the entrance into the wine press (0.8 m wide) is located in the southeastern corner of the treading surface. On the bedrock step adjacent to the treading floor from its south, were hewn two small compartments (Fig. 2.73). the northern is square (0.5×0.5 m) and opens onto the treading surface. the southern compartment is circular (0.5 m in diameter), and has a small rectangular settling pit at its base and a small opening below its eastern edge. at the eastern end of the treading surface there is a square collecting vat (2.1×2 m), filled with silt to about 0.5-0.7 m down to its edge. its volume, therefore, was at least 3 m³ (3000 litres). in the northeastern corner of the vat a segment of thick plaster layer, embedded with pottery fragments was preserved. the body sherds of bag-shaped storage jars date to the Byzantine period. this plaster originally covered the entire collecting vat. On the southern wall of the vat, 0.5 m below its edge, a vaulted opening, probably of a subterranean space which was placed below the treading surface, can be seen. the nature of this space and its connection to the wine press is unknown. as far as can be judged from the present condition of the wine press, it apparently belongs to one of the simple types with a treading floor and one collecting vat. However, the presence of a subterranean space connected to the collecting vat might indicate a more complex system. However, this can only be clarified by further excavation. Southeast of the wine press, a small cluster of cuttings in the bedrock was identified. the right angles of the cuttings and the presence of severance channels on their fringes indicated that this is a small quarry (Fig. 2.74) from which a limited number of ashlars were hewn. the quarry cannot be dated, but based on the main phases of construction at the site it can be either Roman or Byzantine. in any case, its small scale indicates that this place was not the major source of building stones for the site. Quarries were also found near two monasteries in the Judaean desert – those of Khirbet ed-Deir (Hirschfeld 1999:12, Fig. 5) and Chariton (idem. 2002b:166, Fig. 83). Other features are two terrace walls which were built on the lower part of the northeastern slope – one (22 m long, 0.8-1 m high) at the bottom of the slope and the second (15 m long, 1-1.3 m high) 5 m west of the former. they were both made of small to large fieldstones and the higher terrace 69 Fig. 2.72: Plan of the wine press on the southern spur. Fig. 2.73: Side compartments in the wine press, looking west. Fig. 2.74: Part of a quarry on the southern spur, looking west. 70 Chapter 2: StratIgraphy and arChIteCture PHaSe iii. tHe late UMaYYaD/>aBBaSiD PeRiODS Due to later use in the post-monastery phase and the remodelling which took place at that time, it was not always easy to distinguish between the remains and finds of Phases iiB and iii. However, the nature and location of some of the features which postdate Phase ii clearly indicate that at the time of their construction the site no longer functioned as a monastery, and therefore they can be securely attributed to Phase iii (Fig. 2.75). Phase iiB seems to have ended around the late 7th/early 8th century. later, maybe after a short gap in occupation, the site was resettled. this time the inhabitants were no longer Christian monks, but simple peasants, who’s religious identity is yet unclear (see Chapter 10). Most of the walls attributed to Phase iii were built rather poorly compared to the walls of Phase iia and even to some of those of Phase iiB. these walls can be distinguished from the walls of the previous phases by their rough, sometimes slapdash construction, which used stones of various sizes, including older ashlars and other architectural elements in secondary use. architectural remains and small finds which can be attributed to this phase were found throughout the site. the most prominent characteristics of the building activity of Phase iii are the re-division of rooms and spaces and the blocking of old doorways and passages. THE GATEHOUSE an important clue to the changing nature and function of the site can be seen in the area of the main gate (Fig. 2.76). the stepped corridor which led from the gate to the western and domestic units of the monastery was blocked by construction of a wall (W210; 7.5 m long, 0.9 m wide), which abutted on the northeastern corner of the church, and continued eastwards, directly into the doorway of Courtyard 2 (Figs. 2.77, 2.78). the wall was founded in the bedding fill of the entrance corridor. after the construction of the wall, domestic garbage continued to be thrown in and accumulated on both sides of it, until it covered the remains of the gate and the staircase in the north. to this layer of the debris one can attribute two fragments of a marble altar table and a marble bowl and fragments of a glass window pane which originally belonged to the church building. By building this wall, the new inhabitants prevented not only free access from the main gate northwards, but also cancelled the entrance into the small courtyard in the east. an interesting detail related to this wall is the reuse of four large parts of doorjambs, which undoubtedly previously belonged to the main gate of the monastery, in the construction of its western section. During the dismantling of the eastern end of the wall, in order to unearth the threshold of Courtyard 2, some pottery sherds were found in the wall’s foundations (locus 390). the latest ceramic types give a mid-7th century terminus post quem for the construction of the wall (Fig. 3.17:1). a fragment of a roof tile which was also found here is another indicator that the wall was built after the destruction and collapse of the monastery’s tileroofed units. Signs of activity which can be attributed to Phase iii were found also in the guard room (Room 3). this room had been stone-paved and now most of its slabs were now robbed out. the finds from the fill inside this room (loci 343, 348, 361) include pottery sherds, most of them dated not before the 7th century (Fig. 3.17:2-9). in addition, two coins were found directly above the stone floor (locus 359); one is of Heraclius (610-641 Ce; Chapter 7: No. 32) and the second is of Constans ii (641-668 Ce; Chapter 7: No. 35). these coins give a mid-7th century terminus post quem to the robbing of the stone floor and to the accumulation of finds inside the room. in a collapse outside the southern wall of the room was a shallow stone basin (0.25×0.28 m, 5 cm depth), which was carved into a small ashlar (Fig. 2.79). However, it cannot be said if the basin belonged to Phase iii or to the original use of the room in Phase ii. For some reason, the new inhabitants of Phase iii no longer needed the stepped corridor but they did utilize most of the monastery’s other living rooms. the debris which had accumulated in these would have been cleared out and fertilizing agents such as ashes and manure used as garden soil (see 71 Fig. 2.75: Plan of Phase iii. 72 Chapter 2: StratIgraphy and arChIteCture Fig. 2.76. Plan of the gatehouse area and the southern area of the monastery during Phase iii. Miller and Gleason 1994:37-39). therefore, the closed rectangular space created by the blocking wall was probably now filled with earth which contained rich domestic rubbish and used as a small garden. the phenomenon of converting existing architectural units into gardens is known also from the monastery of Martyrius in the Judaean desert, where a small farm with channel-irrigated gardens was built above parts of the Byzantine complex during the late Umayyad or >abbasid period (Magen and talgam 1990:104-105, Fig. 16). Similar activities took place, in different contexts, also at Beth Shean during the Umayyad period (tsafrir and Foerster 1997:138) and at Caesarea at the very end of the Byzantine period (Patrich 2007:154-163). 73 Itamar taxel Fig. 2.77: the southeastern corner of Hall 31 and the Phase iii blocking wall (W210) of the entrance corridor. Fig. 2.78: the eastern end of W210, built over the threshold of Courtyard 2. Fig. 2.79: Stone basin found outside Room 3. 74 Chapter 2: StratIgraphy and arChIteCture THE dOMESTIC UNIT Some of the architectural features which can be attributed to Phase iii were built on the floor of the large courtyard. a wall (W104) was built against the eastern wall of the tower and at an angle to it. it comprises two rows of large ashlars in secondary use, but due to modern damage to the northern part of the courtyard, only its western end was preserved. the pottery found south of the wall, both in the collapse that covered the floor and on the floor itself (loci 209, 210), can be generally dated to the late Byzantine/Umayyad period (Fig. 3.17:10, 11), although an >abbasid coin dated to 841-842 Ce (Chapter 7: No. 46) was also found on the courtyard’s floor. Other remains that belong to Phase iii are two patches of a floor (locus 291) and foundations of two thin walls (W105, W121), which were built south of W104, along the eastern wall of the tower and against its southern doorway (Fig. 2.80). the floor was built mainly of small fieldstones, in addition to a few older building stones in secondary use, including fragment of a threshold which was turned upside-down. the new floor was set on a thin layer of earth fill which lay directly above the older ashlar floor of the courtyard. the southern part of the new floor was a little higher than the socket stone of the tower’s southern doorway, thereby precluding the existence of a door. like the floor, the new walls were also founded above the older courtyard’s floor. they created a narrow space (1.3 m wide, 1.3 m known length) between them and the eastern wall of the tower, that was probably no more than a small storage chamber or the like. it cannot be said if the later floor covered an area much larger than that unearthed in the excavations. the pottery sherds which were found in the foundations of the floor (locus 295) are dated to the late Byzantine/Umayyad period, and include also roof tile fragments. another wall which can be attributed to this phase was unearthed in the excavations conducted at the site by the iaa, a little bit to the south of the above-mentioned remains (W10*; not marked on plan). the pottery sherds found beside this wall (locus 110*) are dated to the late Byzantine- Umayyad and >abbasid periods. a terminus post quem of the late 7th century can be given to the construction of this wall by a rim of a certain type of basin found within the wall’s building stones. all these above-mentioned remains – W10*, W104, W105, W121 and the late floor, were added to the courtyard not before the late 7th century, if not later. On the one hand, most of the pottery found below the floor is dated not later than the mid-8th century. On the other hand, more than one >abbasid coins were found in the site beside pottery which theoretically should be dated earlier. these features were in use at least until the beginning of the >abbasid period, in the second half of the 8th century, or even built at that time. it should be noted that numerous pottery sherds were found on the bedrock in the northern part of the courtyard, where the pavement was completely demolished (loci 205, 212, 214, 218, 222, 224, 227, 229, 234, 267). Most of this ceramic assemblage, like that found under the floor, is dated to the 7th century. there was also a coin of Justin i, 518-527 Ce; Chapter 7: No. 26). However, since it contained also a few sherds from the >abbasid period, an >abbasid coin dated to 837-838 Ce (Chapter 7: No. 45) and dozens of rifle bullet casings, we can not know if the 7th-century material represents the original fill from below the floor or the accumulation above it. Similar architectural changes took place, probably at the same time, also in Room 11 which led to the oil press. its later floor (locus 189) was actually a continuation of the large courtyard’s floor to the east. the floor was preserved only in some parts of the room. as previously remarked, the construction of this floor probably dismantled the wall which divided between Room 11 from the courtyard. the new floor was also built at a higher level than the old floor of the room. the foundation layer of the new floor (locus 402) was composed of building stones and a threshold of the older wall, and from earth which contained pottery sherds (including one fragment of a roof tile) dated generally to the mid-6th–mid-8th centuries. During the construction of the new floor, the broken screw weight of the upper pressing system in the oil 75 Itamar taxel Fig. 2.80: late floor and walls (Phase iii) built over the floor of the large courtyard, looking south. press was embedded in one of the junction points between the floors of the courtyard and the room (Fig. 2.81). the thin earth layer below the weight (locus 506) contained animal bones and pottery sherds, some bearing fire traces, dating generally to the mid-6th–mid-8th centuries (Fig. 3.17:12). in addition to the above-mentioned blocking wall (W210) built in the gatehouse area, more significant changes took place in Courtyard 2 and the area surrounding it. a short wall (W217) was built between W210 and the corner of the older walls W129 and W225. the northern and western walls of the courtyard were almost completely robbed of their stones, and new walls (W133, W232) were built adjacent to their outer face. another new wall (W146), oriented north-south, was built 76 between the northern wall of the paved space and the southern wall of Room 5. this wall reduced the area of Room 4. in Room 4 itself some remains attributed to Phase iii were also found. in its northeastern part, adjacent to its north wall, was a rounded installation (locus 285; diameter 1.7 m) carved into the bedrock and lined with small fieldstones (Figs. 2.76, 2.82). the total depth of the installation is 1.9 m, but only its upper 1.2 m was lined with stones. the installation was full of ash and pieces of charred wood, as was the area around it. its walls were sooty, and the bedrock in its base had changed from its natural yellowish colour to almost red. these signs indicate a strong fire burning inside the installation. it seems, therefore, that it was Chapter 2: StratIgraphy and arChIteCture used as a lime kiln. Some thinner walls were built west of W146, surrounding the kiln and creating a small rectangular chamber (1.2×1 m) west of it. to these remains one can apparently attribute a layer of small fieldstones associated with many mediumsized white tesserae (loci 120*, 124*), identified by the iaa excavators as possible foundations of a mosaic floor above part of the older ashlar paving of Courtyard 2. the pottery sherds found in this layer are dated to the late Byzantine/Umayyad period. Our excavations revealed that the entire area of Room 4 was covered with ashy earth and the stones of the lime kiln and the related walls were blackened by fire. the pottery found in relation to these remains indicates that the use of the lime kiln was only during the later part of Phase iii. the pottery from the fill that covered this area (loci 265, 270, 275-277) included types which dated mostly to the 7th-8th century, although some belong to the >abbasid period (Fig. 3.17:13, 14). the ashy fill inside the lime kiln (locus 274) also contained a few pottery sherds. Most of them can be dated to the late Byzantine/ Umayyad period, although one bowl fragment probably belongs to the >abbasid period. Some of these sherds were partly or completely covered with lime, which can reinforce the identification of the installation as a lime kiln. the charcoal from inside the kiln was examined and found to belonging to aleppo pine (Pinus halepensis). this tree was not common in Judaea in antiquity. Most of the examples of allepo pine found in archaeological sites in this region (including Jerusalem) were probably specially imported from neighbouring countries for use in buildings (liphschitz, Biger and Mandel 1990:147, 149). therefore, the origin of the wood found at Khirbet es-Suyyagh could have been either from native pines (such as those grown until today in the Masrek National Park 6 km northeast of the site), or from imported logs. if the latter possibility is true, we can assume that the imported pine logs were used in the construction of the church in the late Byzantine period, and were taken from its ruins in order to use as fuel for the lime kiln in the >abbasid period. Major changes occurred also in the southeastern corner of the central unit – in the large, rectangular hall and its smaller back room (Rooms 15 and 16). the common wall of these rooms and the rooms and small courtyard to the west (W131) was probably demolished down to the foundation course, and a new wall (W147) was built parallel to it from its east. a 0.9 m wide gap remained between the southern end of this wall and the southern wall of the complex, probably to use as a doorway to this part of the building. the foundation course of the new wall was built of ashlars (apparently robbed from the older wall), but its upper courses, or at least the second course which was preserved, were built of small and medium-sized fieldstones. the wall which divided between room 15 and room 16 was also demolished, and a new wall (W135) was built 3 m north of it. the southeastern corner of the complex was now divided into two almost equal rooms, 6.2 and 6.6 m length, respectively. the more southerly of these two rooms was paved with a new floor (locus 400), built of small pebbles, which covered the foundation course of the older latitudinal wall (Figs. 2.76, 2.83). Only a small section of the floor was preserved, and the few pottery sherds found above it (locus 279) can not provide a date more precise than the mid6th–mid-8th century. the northern room probably had a levelled bedrock floor. Remains of a circular Fig. 2.81. Section e-e, from the large courtyard in the east to W12 in the west. 77 Itamar taxel Fig. 2.82: lime kiln (l285) in Room 4, looking north. installation built of small fieldstones (locus 273; 1 m in diameter), probably the foundations of a tabun, were found in the centre of the room. However, on the basis of the few pottery sherds found here, we cannot determine for sure if this tabun belonged to Phase ii or iii. another alteration identified in this area is the rebuilding of the last 10 m of the southern end of W33. this section of the wall, excluding maybe the foundation course, was completely destroyed and rebuilt in a different manner than was usually common at the site. the rebuilt part, which was 0.4 m wider than the original wall, was made of small and medium-sized fieldstones bound with mortar. theoretically, it seems that the southern end of W33 was destroyed during the earthquake that hit other parts of the monastery at the end of Phase iia. However, we cannot say if the renovation of the damaged section occurred during Phase iiB or only in Phase iii. 78 the southern part of the gap between the new western wall of this area (W147) and the remains of the older western wall was now used as a storage chamber. that is apparent from a crushed storage jar and its matching lid (Figs. 2.84, 3.17:17, 18), found in situ above a small stone-paved platform between the two walls (loci 271, 286). this type of jar is usually dated to the late 6th/7th to 8th century. a mid/late 7th or 8th century date to this assemblage seems reasonable also on the basis of the rest of the pottery sherds found here (Fig. 3.17:15, 16). a highly important object found beside the crushed jar, is an iron bident hoe (Fig. 5.5), which points to the agricultural nature of the new population that resettled the site (see Chapters 5 and 10). architectural remains which belong to Phase iii were found in several other rooms of the central unit. Rooms 12 and 13 were re-divided by the demolishing of the wall which originally separated between the rooms (W126), down to its foundation course. a new thinner wall (W148) was built above the older one. a line of three small chambers was built along the common western wall of the rooms (Fig. 2.75). the chambers were built of thin, crude walls made of small and medium-sized fieldstones and ashlars in secondary use (Fig. 2.85). the northern chamber (locus 131*; 1.2×1.2 m) was paved with small fieldstones, and the two Fig. 2.83. Phase iii floor (l400) in Room 16, looking west. Chapter 2: StratIgraphy and arChIteCture other chambers (loci 246, 401; 1.3×1 m each) probably had a beaten earth f loor. another feature which can be attributed to Phase iii was found in the northwest corner of room 12. this was a semicircular installation built of small and medium-sized fieldstones (locus 155*; 0.7 m in diameter). this installation and its surroundings were found covered with ash, so it was probably a tabun. the two rooms were found covered down to bedrock with fill which contained pottery sherds. the pottery from the fill inside room 12 (loci 128*, 130*, 132*, 137*, 154* 231, 239) included types dated to the late Byzantine-Umayyad and >abbasid periods (Fig. 3.28:5), the latest of which can date the last phase of use in this room and maybe even the construction of the later features. in addition, a coin of Constantine ii (337-340 Ce) was also found here (Chapter 7: No. 9). two fragments of a marble chancel screen and a marble altar table, which were found below the floor of the northern chamber (locus 190), are another proof that it was built after the abandonment of the monastery and the demolishing of the church and its furniture. the pottery from the fill in room 13 (loci 248, 252) include types dated generally to the late Byzantine/Umayyad period. Changes also occurred in the large rooms north of Room 5. in Room 19, a new wall (W106) was built in order to block up the opening between it and the small entrance room 18 in the northeast. the new entrance to the room was now through a breach in the eastern wall (W140) (Fig. 2.75). in addition, most of the paving stones of this room were uprooted and only small patches remained in situ. the fill found inside the room (loci 208, 219, 220) was accumulated, therefore, during Phase iii. the finds from the fill included pottery sherds dating from the mid-7th–early 8th century (Fig. 3.18:1-9), and two coins of Constantine i (307-337 Ce; Chapter 7: No. 7) and tiberius i (580-581 Ce; Chapter 7: No. 29). apparently during Phase iiB, Room 20 was divided into two smaller rooms. in the middle of the eastern room, a circular pit (locus 292; 1.5 m in diameter, 0.8 m depth) was later dug in the soft bedrock (Fig. 2.75). the pit was found full of dark ash, which covered also the area to its north, east and south, including parts of the demolished walls of the room. the latter detail indicated that the hewing of the pit and the activity connected to it occurred when at least some of the surrounding walls were already robbed of their stones. this assumption is strengthened by the pottery that was found in the fill inside the pit (locus 255). it included a complete Fine Byzantine Were bowl/cup of a type dated to the 8th-9th centuries (Fig. 3.28:4) and a few other sherds, at least one of which also characterized the late Umayyad and >abbasid periods. therefore, it seems that the activity connected to the pit (burning debris?) took place at the later part of Phase iii. THE INdUSTRIAl UNIT Finds and remains which can be related to Phase iii were retrieved from Cistern 25. this contained a deep layer of collapsed material composed of ashlars and fieldstones (locus 179a), which belonged to the upper courses of the walls and to the ceiling. However, here the collapse lay above at least 1 m thick layer of earth (locus 179B). the plastering method of this cistern differed from that of Cistern 24. at least four layers of plaster, some of which were pale or dark grey, could be distinguished on Fig. 2.84. Crushed storage jar and a lid between W131 and its walls. the plaster was mixed with small pieces W147. 79 Itamar taxel Fig. 2.85: Phase iii walls (W115, W116) in Room 13, looking north. of charcoal and stone and pottery grits. One of the upper layers was made of pinkish plaster, mixed with pottery grits and tiny pieces of straw – similar to the plaster layer which covered Cistern 24. the latest layer of plaster that coated the lowermost level of the cistern, up to a height of ca. 1.8 m, was clearly different from the other layers. it was made of brittle grey plaster, in which pottery sherds were embedded and is thicker than the others, sloping down towards the bottom of the cistern. the finds from inside the cistern, i.e., from the collapse, the lower fill and the latest layer of plaster, are important for the understanding of the last phases of its use. the stone collapse (locus 179a) contained a small amount of pottery sherds, the latest types being typical of the late Umayyad and early >abbasid period (Fig. 3.18:10-14), and an Umayyad post-reform coin (i.e., post-dated 696 Ce; Chapter 7: No. 40). therefore, the first half/ middle of the 8th century is the terminus post quem for the collapse of the cistern. the lower level of earth which covered the bottom of the cistern (locus 179B) contained no stones at all, but many pottery fragments and some partially restorable vessels (Figs. 3.19, 3.20:1, 2), in addition to some glass fragments, many chared wood pieces (identified as Olea europaea), and charred olive 80 pits. the ceramic assemblage can be dated not earlier than the first half of the 8th century. the pottery types which were embedded in the plaster that coated the bottom of the cistern (locus 179C) included not only body sherds of bag-shaped jars (as usually known from plastered installations of the late Roman to early islamic periods), but also several more diagnostic sherds of other vessels. the latter belong to a coarse basin, Fine Byzantine Ware bowls and jug, a Cypriote Red Slip bowl and a roof tile (Fig. 3.20:3-6). the latest types can be dated to not before the early/mid-7th century, but on the other hand the assemblage does not include types which appeared only in the mid/late 8th to early 9th century. During the demolition of the cistern as part of the construction activities which took place at the site in 2005, two important architectural elements made of limestone were found by inspectors of the iaa. However, it cannot be said whether these elements were found in the collapse which filled the upper part of the cistern, or in the fill layer in its bottom. the first element was a partly broken limestone column base, which probably belonged to the church of Phase ii (see above). the second element is a doorjamb capital. this is a rectangular ashlar with a slightly trapezoidal cross-section Chapter 2: StratIgraphy and arChIteCture (0.47 m length, 0.3 m width, 0.32 m front height, 0.25 m rear height). the middle of the stone is carved in the form of an inverted, rectangular pyramid which only two of its faces projects out of the jamb’s line (Fig. 2.86). Similar elements, mostly relief-decorated, are known from quite many Byzantine-period buildings in Palestine, such as some discovered in the Negev large villages (e.g. Negev 1988: Figs. 3:86, 6:55; 1997: Phs. 235, 288). this element was most probably taken from one of the doorways of the abandoned monastery, maybe even from the church, and reused in the settlement of Phase iii. the history of Cistern 25 can be summed up as follows. the cistern itself was hewn not before the mid-6th century and continued to be used at least until the late Umayyad period. it probably did not collapse at once, as was the case with Cistern 24. it is reasonable to assume that it was re-plastered at least once during Phase iiB or iii, around the late 7th century. However, sometime in the second half of the 8th century, domestic garbage started to be dumped into the cistern and to accumulate on its bottom, together with silt. Use of the cistern finally ceased when it collapsed in the late 8th or early 9th century. Other remains which can be safely attributed to Phase iii were found within the tower. in the collapse which covered the area west of the southern doorway (loci 103, 133), in addition to a few pottery sherds that can be generally dated to the late Byzantine-Umayyad period, a large fragment of a marble chancel post and fragment of a glass window pane were found. these two items originally belonged to the church, and were apparently brought here as building materials after its abandonment, during Phase iii. additionally, immediately south of the doorway, on the original stone paving of the tower and adjacent to its southern wall, a massive bulk of stones (W1; 2.6 m length, 1.1-1.5 m width) arranged more or less in a straight line was unearthed. it may have served as a supporting wall for the southeastern corner of the tower. this wall (?) was built of ashlars and architectural elements in secondary use, including a small voussoir and an oil press beam weight with a reversed t-shaped bore (see Fig. 2.37). During the removal of this feature (locus 500), a few pottery sherds which can be generally dated to the late Byzantine/Umayyad period were found between the stones. However, the find which best dates this remain is an >abbasid coin dated to 768-800 Ce (Chapter 7: No. 43), found in its foundations, above the older Byzantine floor. another point related to the tower in which finds from the later stage of Phase iii were found is, surprisingly, beside the foundations of its southwestern corner. While excavating along the outer face of the tower’s south wall (loci 116, 125), a small quantity of storage jar fragments was found beside the foundation course of the southwestern corner. these jars are of a type which is typical of the late Umayyad but mainly the >abbasid periods (Fig. 3.28:7). together with them was found also a coin generally dated to the 5th century (Chapter 7: No. 16). the most reasonable explanation for the presence of these pottery sherds beside the foundation of a structure at least two centuries older is that during the later stage of Phase iii a refuse pit was dug outside the tower. Probably modern bulldozing of the area removed this feature almost entirely, excluding its base, which was unearthed in the excavations. as already noted, there was quite convincing evidence that at least some of the features in the oil press complex were built or rebuilt during the Fig. 2.86. Doorpost capital. 81 Itamar taxel Umayyad period (Phase iiB) and furthermore that it ceased to be used during Phase iii. this is indicated by the latest pottery sherds found in the topsoil layer which covered the oil press (loci 162, 256), which, excluding a few finds from the late Ottoman period, are dated to the >abbasid period. However, more solid evidence comes from the fill (locus 186) found below the broken crushing basin and above the staircase which entered the complex in the form of pottery sherds dated to the late Byzantine/Umayyad and the >abbasid periods (Fig. 3.22:1). the latest of these clearly shows that the dismantling of the crushing system could not have occurred before the >abbasid period. the excavation of the lower pressing system indicates that it was probably turned out of use sometime during the Umayyad period. although this event is tentatively attributed to Phase iii, we cannot rule out the possibility that it happened during Phase iiB, when maybe the volume of oil production was restricted only to the upper pressing system. the pressing base, which was originally placed over the collecting vat, was removed from the vat and put, turned on its face between the vat and W29. the pottery found in the fill which covered the collecting vat and the fallen pressing base (loci 172, 174, 176) and in the fill which sealed the screw weight and W29 (locus 136*), contained relatively many ceramic finds. Most of them were single sherds, but three complete and nearly complete lamps and one complete basin were also found. One of the most important finds is a fragmentary ceramic mould-made pilgrims’ ampulla. the latest lamp type found here is dated to the Umayyad period, and the latest sherd (of a Fine Byzantine Ware bowl) is dated to the 8th9th century (Figs. 3.21, 3.22:2-4). it seems that at least the complete lamps and the basin belong to the post-oil press stage, when the area of the lower pressing system was probably converted into a living room. During the time which passed between the final abandonment of this room and the collapse of its walls, the other pottery sherds maybe washed down from the upper part of the site. the collapse itself occurred only during the >abbasid period or later, as indicated by the latest 82 pottery sherds found in the debris that covered the area between the collecting vat and the screw weight (loci 103*, 126*, 173). in addition, this debris contained an Umayyad coin dated to 737 Ce (Chapter 7: No. 38). THE CHURCH COMPlEX this area, which was the social and religious heart of the monastery of Phase ii, probably symbolised the total opposite in the eyes of the new inhabitants of Phase iii. No clear architectural changes related to this phase can be identified within the church itself. although we believe it was transformed into a domestic complex during that time, no solid evidence for this was found due to the drastic damage caused in modern times. the only remains from Phase iii found in the church complex are from Hall 31. Here, the two internal walls and the round base can be attributed, with caution, to the time of Phase ii. the most prominent remnant, attributed to Phase iii, is W209 which connected W138 and W201. this wall was very crudely built from ashlars and fieldstones of various sizes, in addition to a large fragment of a marble ciborium column in secondary use (Figs. 2.75, 2.87; see Chapter 6). the wall was preserved to a height of one course only, but it does not seem to have been much higher and probably served as a dividing wall between two internal ‘rooms’. another feature from this phase is a clay tabun (locus 394; 0.5 m in diameter; 0.4 m preserved height), built near the southeastern corner of the hall (Fig. 2.88). it was not possible to distinguish between the different layers of the fill which had accumulated in the eastern part of Hall 31. this fill was covered by a collapse of building stones (loci 307, 312, 317), which contained pottery sherds (the latest are dated to the 9th century; Fig. 3.23:1, 2), glass fragments (including one belonging to a window pane), an Umayyad coin of a post-reform date (Chapter 7: No. 41), and a fragment of a marble chancel screen. Some >abbasid pottery sherds and two additional fragments of marble chancel screens were found also in the upper layer of the fill, under the collapse (locus 322). the reuse of Chapter 2: StratIgraphy and arChIteCture the marble ciborium column in the building of the later internal wall and the finding of additional chancel screen fragments in the fill and collapse related to this wall all indicate that these remains belong to a time when the church was no longer in use. the ceramic and numismatic evidence points to a late Umayyad->abbasid date to this phase. Fig. 2.88. Phase iii tabun (l349) in Hall 31, looking south. PHaSe iv. tHe MaMlUK aND late OttOMaN PeRiODS the last phase of habitation of the site prior to the 20th century took place during two periods – the Mamluk (13th-15th centuries) and the late Ottoman (probably the 19th century). Some pottery sherds of the Mamluk period (Fig. 3.29:7-9) were found in the topsoil at the north and south of the site. Fewer sherds, in addition to one fragmentary smoking pipe (Fig. 3.29:5) and few metal objects of the late Ottoman period were found only at the north of the site. No architectural remains can be attributed to either of these periods. it seems, therefore, that during the Mamluk and late Ottoman periods the site was only randomly visited by peasants of the neighbouring villages and/or by passers-by. Fig. 2.87: Phase iii wall (W209) in Hall 31, looking north. Note the reused marble column fragment at the middle of the wall. 83 CHaPteR 3 POtteRY itamar taxel this chapter presents the ceramic finds retrieved from the site. they include pottery artefacts dated to the late Hellenistic/early Roman, late Roman/early Byzantine, late Byzantine, Umayyad, >abbasid, Mamluk and late Ottoman periods. Most of the pottery plates are arranged according to assemblages represented by either one or more loci. type 2: a local, rounded bowl, almost metallically fired, with incurved rim and thin walls (Fig. 3.1:2). this bowl was very common in Judaea from the 1st century BCe to the 2nd century Ce (Bar-Nathan 2002: Pls. 14:203-204, 207-211, 15:222-223; 2006: Pl. 25:8-9, 11-12, 14; Fischer and tal 2000: Fig. 2.2:2-10; Geva and RosenthalHeginbottom 2003: Pls. 6.2:37-45, 6.6:15). late HelleNiStiC/eaRlY ROMaN PeRiOD KRATERS the bulk of the pottery dated to the early Roman period was found in a fill at the foundations of the northern wing of the tower. this group of sherds and partly restorable vessels represents a typical household assemblage of Judaean sites in the 1st century BCe - 1st century Ce. Unless otherwise mentioned, all of the pottery types are made of pale brown, pinkish-brown or orange-brown ware, usually fired at high temperatures, which is typical of the common pottery of the Second temple period in Judaea. BOWlS Only two types of bowls are represented in the assemblage. type 1: a local imitation of the Hellenistic ‘fish plates’, made of buff ware, with rounded body and sharply down-curved rim (Fig. 3.1:1). according to Bar-Nathan (2002:94, 97), this bowl is typical only of the late Hasmonean period (end of 2nd/end of 1st century BCe), and is absent from Herodian assemblages (c.f. ibid.: Pl. 16:268; Crowfoot, Crowfoot and Kenyon 1957: Fig. 56:5; Geva and Hershkovitz 2006:Pl. 4.5:1). 84 a krater with a wide, short neck and triangular rim (Fig. 3.1:3). it has parallels, though with a somewhat different rim, at other assemblages dated to the 1st century Ce (Bar-Nathan 2006: Pl. 23:1-2; Killebrew 1999: Fig. iii.57:1; loffreda 1996: Fig. 19:24-25). COOKING WARE the cooking vessels are represented by two types of casseroles and two types of cooking-pots, all of which are made of a reddish-brown, brittle cooking-pot ware. CaSSeroleS type 1: a casserole which has a high, out-turned rim, carinated body and small loop handles from rim to shoulder (Fig. 3.1:4). it is a typical form of the Hasmonean and mainly the Herodian period (Bar-Nathan 2002: Pls. 13:164-165, 27:495-499; 2006: Pl. 30:53-56; Geva and Hershkovitz 2006: Pls. 4.5:22, 4.10:21, 22, 4.12:16, 4.13:17). type 2: Casserole with a somewhat concave ledge rim (Fig. 3.1:5) and is dated to the Herodian period (Bar-Nathan 2002: Pl. 27:494; 2006: Pl. 30:62-64; Geva and Hershkovitz 2006: Pls. 4.5:21, 4.8:13). Chapter 3: pottery Fig. 3.1: late Hellenistic/early Roman pottery. 85 Itamar taxel FiGURe 3.1: late HelleNiStiC/eaRlY ROMaN POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Bowl Bowl Krater Casserole Casserole Cooking-pot Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar Storage jar 141 142 138 135 142 138 142 141 143 141 142 142 156 138 135 138 142 138 143 1180/2 1155/3 1129/6 1118/2 1147/1 1129/4 1155/1 1180/1 1141/1 1149/1 1140/1 1140/2 1206/1 1129/1 1118/1 1129/2 1147/2 1129/3 1141/2 256.42 256.30 256.60 257.03 256.45 256.60 256.30 256.42 256.60 256.51 256.50 256.50 255.45 256.60 257.03 256.60 256.45 256.60 256.60 FiGURe 3.2: late HelleNiStiC/eaRlY ROMaN POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Juglet Jug Jug Jug Flask Jug lamp Cooking-pot Storage jar Storage jar Storage jar Storage jar Storage jar Jug 143 156 141 135 135 142 142 518 518 600 600 600 600 253 1141/4 1206/2 1180/5 1118/6-7 1118/6-7 1155/2 1147/3 3084/1 3084/2 6000/1 6000/2 6000/3 6000/4 2241/1 256.60 255.45 256.42 257.03 257.03 256.30 256.45 252.05 252.05 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 256.35 86 Chapter 3: pottery Fig. 3.2. late Hellenistic/early Roman pottery. 87 Itamar taxel CooKIng-potS type 1: a cooking-pot with a short, slightly outturned neck, triangular rim and small loop handles (Fig. 3.1:6). this is the so called ‘Herodian’ cooking-pot, which continued to exist until the first half of the 2nd century Ce (Bar-Nathan 2002: Pls. 12:148-149, 26:477-479; 2006: Pls. 27:4, 6, 28:28; Fischer and tal 2000:34, Figs. 2.4:5, 14, 15, 2.5:1, 3, 5, 7, 9). type 2: a cooking-pot which has rather thin walls, short, flaring neck with a slight carination at the joining point with the shoulder, and flattened rim. the beginning of a flat strap handle is visible on the rim (Fig. 3.2:8). this form seems to be a variant of a relatively well-known type of Judaean cooking-pot with a broader rim and a sharper carination, dated to the 1st to mid 2nd centuries (Bar-Nathan 2006: Pl. 29:46; Fischer and tal 2000: 34-35, Fig. 2.5:15). STORAGE JARS Five types of storage jar belong to the period under discussion. type 1: this type is represented by different variants of collar-rim jars (Fig. 3.1:7-9), which characterized the Hasmonean and Herodian periods (1st century BCe-1st century Ce), mainly in Judaea (Bar-Nathan 2002: Pls. 3:19-21, 4, 5:26-27; 2006: Pl. 4:14-17; loffreda 1996: Figs. 1-3). type 2a: With a vertical neck with a ridge at its base and slightly everted rim (Fig. 3.1:18). this is an early form of the Palestinian bagshaped jar, which is dated mainly to the 1st century BCe-1st century Ce (Bar-Nathan 1981: Pls. 3:6-9, 15, 9:3; 2002: Pls. 6:40-41, 24:397-405; 2006: Pls. 4:19-20, 5:21, 25). another jar which can be related to this type lacks the ridge at the neck’s base (Fig. 3.1:10; Magen 2004: Pls. 1:18, 4:18). type 2B: Similar to the latter but differs from it by its more convex neck and thickened rim (Fig. 3.1:11, 12), but dates to the same period (BarNathan 2002: Pl. 24:415; 2006: Pl. 8:39-40; Fischer and tal 2000: Fig. 2.6:28). 88 type 3: With higher neck and wide, out-folded rim (Fig. 3.1:13, 14, 17, 19) which is dated to the 1st century BCe-1st century Ce (Bar-Nathan 2002: Pls. 5:28-30, 6:32-33; loffreda 1996: Fig. 4). type 4: With a high, concave and slightly ridged neck and flaring rim (Figs. 3.1:15, 3.2:9). these characters resemble those of the bell-shaped jars of the 1st-mid 2nd century Ce (Fischer and tal 2000: Fig. 2.6:31, 34), although our example can also be another type of Herodian bag-shaped jar (Bar-Nathan 1981: Pl. 9:2). type 5: With a concave neck as the previous type, but it lacks the ridges and its rim is triangular (Fig. 3.1:16). it also has parallels in Herodian contexts (Bar-Nathan 1981: Pl. 3:18; Hershkovitz 1992: Fig. 5:3). JUGS Jugs and juglets, mostly locally-produced but also imported, are also represented in the early Roman pottery from the site. type 1: a jug with cylindrical neck and narrow, folded ledge rim (Fig. 3.2:2) which is dated to the Herodian period (Bar-Nathan 2002: Pl. 25:430-431; loffreda 1996: Fig. 26:36). type 2: a jug with a wide, flaring and pointed rim (Fig. 3.2:3). this is another typical Judaean form of the 1st century Ce (Bar-Nathan 2002: Pl. 8:52; 2006: Pl. 19:23; Saller 1957: Fig. 61:1-2). type 3: this jug, also typical to the region in the 1st century BCe-1st century Ce, differs from the previous type mainly by its triangular rim (Fig. 3.2:4; Bar-Nathan 2002: Pls. 8:56 25:423; 2006: Pl. 18:15; Fischer and tal 2000: Fig. 2.8:10-18). type 4: Represented only by a body sherd, probably from the shoulder. it is decorated with horizontal combing and plastic thumbed application (Fig. 3.2:6). these features, in addition to the buff colour firing of the vessel, indicate that this is a Nabataean Cream Ware jug. this unique group of vessels, which is typical of sites in southern Palestine and transjordan, first appeared in the mid 1st century BCe and flourished mainly in the 1st2nd century Ce (Bar-Nathan 2006:282-283, Chapter 3: pottery Pl. 52:22-23; Clamer 1997:73-79, Pls. 9:1, 14-15, 19; Fischer and tal 2000:39-40, Fig. 2.10, top left). Until now, only one example of Nabataean Cream Ware jug was published or reported from north of the Negev and the Judaean desert. this jug was found in an early Roman assemblage at Yoqne>am (avissar 2005: Fig. 2.7:4). However, it seems that the jug from Yoqne>am and the jug fragment from Khirbet es-Suyyagh are no more than coincidental finds which do not necessarily indicate trade in these vessels between Judaea and the areas under Nabataean cultural influence. type 5: this is the only true early Roman imported jug found at the site, represented by a low ring base. it is made of a yellowish, slightly micaceous ware and has a burnished red slip on the exterior (Fig. 3.2:14). this fragment belongs to a type of jug of the eastern terra Sigillata ware, with a cylindrical body, carinated shoulder and trefoil rim, which dates to the 1st century BCe-1st century Ce (Hayes 1985:44, Form 105, tav. 9:6), and is known already from other Palestinian sites (elgavish 1977: Pl. 13:108; Silberstein 2000: Pl. 33:16). JUGlETS this is a piriform juglet with narrow neck, cupshaped rim and handle from rim to shoulder (Fig. 3.2:1). this type, probably used as a container for perfumes or the like, is one of the most characteristic Judaean vessels in the 1st century BCe-1st century Ce (Bar-Nathan 2002: Pls. 10:85-87, 25:443-444; 2006: Pl. 33:1-14; Fischer and tal 2000: Fig. 2.11). FlASKS the only type of flask from the period under discussion found at the site has an elongated cylindrical neck, thickened rim, two twisted handles and an asymmetrical body (Fig. 3.2:5), and is typical of Judaean sites in the 1st century BCe-1st century Ce (Bar-Nathan 2002: Pls. 10:120-122, 26:468-475; 2006: Pl. 22:70-73; Geva and Hershkovitz 2006: Pls.4.4:15, 4.7:20, 21, 4.9:11, 4.10:13, 4.11:11, 12, 4.13:9). lAMPS these are also represented by a single type which is a knife-pared lamp with a bow-shaped nozzle. Our example has a reddish slip and a decoration of two circles with a central dot on its nozzle (Fig. 3.2:7). according to Barag and Hershkovitz (1994:46), this lamp represents the most prominent class in Palestine in the 1st century Ce, probably particularly in areas inhabited by Jews (ibid.: Figs. 8:58, 60, 10:67-69). SUMMARy the late Hellenistic and early Roman pottery found at the site represents a homogeneous and relatively varied assemblage. the majority of the types are locally produced, with the two exceptions of the Nabataean Cream Ware jug and eastern terra Sigillata jug. Most of the types first appeared in the 1st century BCe, although others were common mainly in the 1st century Ce. Since we assume that the settlement (a small farm?) was inhabited by Jews, it seems reasonable to date the pottery discussed above to the mid/late 1st century BCe and no later than 70 Ce. this characteristically Judaean assemblage, which contained mainly storage jars and other closed vessels, is typical of small agricultural settlements such the one that existed at Khirbet es-Suyyagh in the early Roman period. late ROMaN/eaRlY BYZaNtiNe PeRiODS two locations in the site yielded most of the pottery dated to the late Roman/early Byzantine period (4th-5th centuries Ce). the first is the fill which sealed the plastered installation found in the western foundations of the oil press. the second, in which the bulk of the pottery was found, is the fill from inside the treading surface of the late Roman/early Byzantine wine press. Other sherds from this period were also found sporadically in other parts of the site. as in the case of the pottery from the previous period, the present assemblage is also characteristic mostly of the area of Judaea. 89 Itamar taxel BOWlS Common bowlS Five types of common bowl can be dated to the 3rd5th centuries. type 1: a rounded bowl, made of yellowish ware, has an external burnishing and a triangular, vertical rim (Fig. 3.3:1). the only parallel to it was found in a late 3rd/4th-5th century assemblage from Jerusalem (Hamilton 1944: Fig. 22:13). type 2: Bowls with thick, slightly carinated walls, flat rim with a ridge beneath, rouletted decoration from the outside and sometimes reddish slip over the yellowish-brown ware (Figs. 3.3:2, 3.4:11, 12, 16). Until recently, it was thought that these bowls appeared in the late 3rd/early 4th and continued to exist at least until the late 5th century (Magness 1993:154, 185-187, Rouletted Bowls Form 1). However, the recent excavations at the site of the Jerusalem Convention Centre (Binyanei Ha’uma) show that this type of rouletted bowl probably appeared a century earlier, in ca. 200 Ce (Magness 2005:105). type 3: Small, thin-walled and slightly carinated bowls with low ring bases and gentle ridges below the rim, which are made of pale brown, well-fired clay. they have a dark grey slip, sometimes over an orange slip, on the upper part of the exterior and the interior wall. One bowl has also a shallow exterior rouletting on the carinated part (Fig. 3.3:12, 13). Some of the characters of these bowls – the carinated body, the low ring base and the hard-fired clay – might indicate that the bowls should be related to the Rouletted Bowls of the previous type. a few identical parallels were found in 4th-early 5th century assemblages from Jerusalem (Mazar and Gordon 2007: Figs. 15.2:2, 15.3:1, 15.9:1). type 4: Small, delicate bowl, with rounded body and narrow ledge rim. it is made of yellowish ware and decorated with a reddish slip from the interior, brown slip on the upper part of the exterior and single dark grey line on the rim (Fig. 3.3:15). it has a parallel in a 4th90 early 5th century assemblage from Jerusalem (Mazar and Gordon 2007: Fig. 15.6:3). type 5: large bowl/basin made of yellowishorange ware, with an everted rim, rounded from the exterior and stepped from the interior (Figs. 3.3:16, 3.4:13). it is dated by Magness to the late 3rd/early 4th-6th century (1993:203-204, Rilled-Rim Basins). afrICan red SlIp ware (arSw) bowl One of the few early Byzantine imported late Roman Red Ware (lRRW) bowls found at the site is a shallow african Red Slip Ware (aRSW) bowl made of orange-brown ware with reddish slip, with thick walls and triangular rim (Fig. 3.4:14). Hayes dated this type to 325-400/425 Ce (1972:100-107, Form 61a). late roman C/phoCaean red SlIp ware (prSw) bowl this group of lRRW bowls, imported from western asia Minor, is also represented by a single early Byzantine example. this is a rounded bowl made of orange-brown ware with reddish slip, with a slightly in-curved rim (Fig. 3.4:15), which was dated by Hayes to the late 4th-early 5th century (1972:325-327, Form 1a). COOKING WARE three types of cooking vessels belong to the late Roman/early Byzantine periods, all of which are made of a reddish-brown, brittle cooking-pot ware. CaSSeroleS this casserole has thin plain walls and thin horizontal handles (Figs. 3.3:3, 3.4:17). this type of casserole is clearly different from those of the late Byzantine and Umayyad periods (see below). On the basis of parallels, it appeared sometime between the 3rd and 4th centuries, and continued into the 5th (Bar-Nathan and adato 1986: Fig. 2:15; Magness 1993:211-213, Casseroles Form 1; tomber 1999: Fig. 3:157). CooKIng-potS type 1: a cooking-pot with vertical neck, grooved rim and small strap handles (Figs. 3.3:4, 3.4:1, 2). this is a long-lived type, which appeared Chapter 3: pottery already in the 1st century Ce and continued into the late Roman or early Byzantine period (Bar-Nathan and adato 1986: Fig. 2:12; Fischer and tal 2000: Fig. 2.5:11; tomber 1999: Fig. 2:41-42). it is possible indeed that some of the specimens found at the site are from the Herodian period, but those who were found, for instance, in the fill of the lower treading surface should be dated to the 4th century. type 2: a cooking-pot with short, slightly everted neck, flattened or grooved rim, sharply sloped shoulder and small strap handles (Fig. 3.3:5). this type is dated to the 3rd-4th centuries (elgavish 1977: Pl. 15:122; Magness 1993:217, Cooking-pots Form 2; Riley 1975:41, No. 56). type 1: a jug with swollen neck which widened towards the top, thickened rim and strap handle from rim to shoulder (Figs. 3.3:8, 3.4:6). identical vessels were found in late Roman/ early Byzantine contexts in Jerusalem and its vicinity (Baramki 1932: Pls. 12:5, 16:9; Saller 1957: Fig. 60:7255). type 2: a jug with a narrow cylindrical neck with a ridge on its upper third, plain rim and handle from the ridge to the shoulder (Fig. 3.4:7, 8). it was dated by Magness to the 3rd-5th centuries (1993:244, Jugs and Juglets Form 1:1). type 3: a jug with a wide, short neck, flattened rim and a ridge beneath (Fig. 3.4:9), which has parallels in a 3rd-4th century assemblage from Jerusalem (Hamilton 1944: Fig. 23:10). STORAGE JARS three types of storage jars can be attributed to the period under discussion. all jars are made of a pale brown or yellowish-brown, high-temperature fired ware. type 1: a bag-shaped jar which has a high, slightly everted neck with a sharp ridge at its base, thickened or infolded rim and ribbed body (Figs. 3.3:6, 7, 3.4:3, 4). Magness dated this type to the 3rd-4th centuries (1993:223-224, Storage Jars Form 4a). type 2: a holemouth jar with a very short, wide neck with a ridge at its base and a flattened rim (Fig. 3.4:18). Magness dated this type, which is typical to Judaea, to the 2nd-5th centuries (1993:231-233, Holemouth Jars Form 1a). type 3: another neckless jar, with a broad ledge rim and sloping shoulder with a band of horizontal incisions beneath (Fig. 3.4:5). this type, which is also characteristic to the region, was dated by Magness to the 3rd-5th centuries (1993:235, Wide-Necked Jars Form 1). JUGS Seven types of late Roman/early Byzantine jugs and juglets were found in the site. Unless otherwise mentioned, all jugs and juglets are made of a yellowish, yellowish-orange or orange-brown, high-temperature fired ware. JUGlETS type 1: a small jug/juglet with very thin walls, conical body, rounded shoulder, short neck, thickened rim and handle from rim to shoulder (Fig. 3.3:9). More complete examples of this type, with pointed bases, were dated by Magness to the 2nd-4th centuries (1993:242, Jugs and Juglets Form 1:1). type 2: a small jug/juglet, made of cooking-pot ware, with a low ring base, conical (?) body and red slip from the exterior (Fig. 3.3:10). this type was dated to the 2nd-5th centuries (Magness 1993:242-243, Jugs and Juglets Form 2a). type 3: a juglet with a flat, thickened base and cylindrical, ribbed body (Fig. 3.3:11). these juglets are very common in the central hill country, but although dated by Magness to the 3rd-8th centuries (1993:246-247, Jugs and Juglets Form 6a) the few examples from Khirbet es-Suyyagh were found only in late Roman/early Byzantine contexts. type 4: a small jug/juglet made of dark brown ware, with a very short, relatively wide and ridged neck, and sloping shoulder (Fig. 3.4:10). it has a parallel dated to the 3rd century from Shiqmona in northern Palestine (elgavish 1977: Pl. 12:104). 91 Itamar taxel lAMPS late BYZaNtiNe/UMaYYaD PeRiODS the nearly complete lamp and a few other lamp fragments dated to this period all belong to one type. they are made of a yellowish ware, redslipped, and have an ovoid squat body, large filling hole, concave base, rounded nozzle and moulded decoration (Fig. 3.3:14). according to Magness, this type is dated to the 3rd-5th centuries (1993:249-250, Oil lamps Form 1). the pottery dated to the late Byzantine and Umayyad periods represents the great majority of finds unearthed at the site. Most of the ceramic finds from these periods were found as sherds, although there were also some assemblages with restorable or intact vessels, either in fills or in situ. Most of the types (excluding the imported and some of the local types) are typical of Judaea and the south of the country. the highly varied array which will be discussed below contained types that dated mostly within the mid-6th to mid-8th centuries and seem to represent Phase ii at the site, although some appeared somewhat earlier or continued after this time. therefore, it is not impossible that some of the vessels described below actually represent the late Umayyad or even the early >abbasid period (Phase iii). GENERAl ROMAN TyPES this title was given to two types of storage jars, which have parallels dated to the 1st-3rd centuries, and therefore they cannot be precisely dated to a specific phase within the Roman period. type 1: With a relatively high and slightly convex neck with a ridge at its base and thickened, outfolded rim (Fig. 3.2:10, 11). Similar jars were published from early Roman (Fischer and tal 2000: Fig. 2.6:28; Silberstein 2000: Pl. ii:12) as well as from later Roman (BarNathan and adato 1986: Fig. 2:8-9; elgavish 1977: Pl. 19:152-153) assemblages. type 2: With a similar neck as the previous type (sometimes a little bit higher) and a stepped rim (Fig. 3.2:12, 13). its parallels were found in early Roman (loffreda 1996: Fig. 14) as well as in later Roman (Bar-Nathan and adato 1986: Fig. 2:1-3) assemblages. SUMMARy the relatively limited but homogeneous ceramic array described above represents a typical assemblage which is characteristic of Judaea in the 3rd to 5th centuries. excluding the aRSW and PRSW bowls, all the pottery types are local products. the dating of this assemblage is not so easy, but although some of the types continue to exist until the 5th century and sometimes even later, there are many others that probably ceased to exist in the 4th century. therefore, i intend to put the end of the late Roman/early Byzantine phase in the site at around 400 Ce. 92 COMMON BOWlS ten types and sub-types of common bowls are included in the assemblage of these periods. Most of these bowls are made of brown, orange-brown or pinkish-orange, usually high-temperature fired, ware. type 1a: a deep, rounded bowl with out-folded rim (Figs. 3.17:15, 3.18:1, 3.23:3), which is the most common type of local bowl in the site. Most of the specimens are made of brown or pinkish-orange ware, although some – probably those from the Umayyad period or even later – are made of buff ware. there are also bowls with wavy combed decoration on the body and rim (Figs. 3.14:1, 3.23:4). these bowls can be found in all of central and southern Palestine, but are especially widespread in Judaea, between the 6th and the mid 8th centuries (aharoni 1962: Fig. 3:10; Cohen Finkelstein 1997: Fig. 1:3; Hamilton 1944: Fig. 9:5; Magness 2003a: Pl. 18.1:11; Wightman 1989: Pls. 9:5, 20:3, 38:13). type 1B: Similar to the latter but differs from it by its wide, triangular rim and straighter walls (Fig. 3.23:5). this bowl perhaps represents a local Chapter 3: pottery Fig. 3.3. late Roman/early Byzantine pottery. 93 Itamar taxel FiGURe 3.3. late ROMaN/eaRlY BYZaNtiNe POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Bowl Bowl Casserole Cooking-pot Cooking-pot Storage jar Storage jar Jug Juglet Juglet Juglet Bowl Bowl lamp Bowl Basin 164 164 164 164 144 164 164 164 164 164 164 511 514 164 512 512 1228/1 1228/2 1223/1 1223/2 1149/1 1225/1 1223/4 1228/3 1228/4 1225/2 1225/3 3062/1 3079/1 1225/4 3069/1 3070/1 256.30 256.30 256.55 256.55 256.05 256.40 256.55 256.30 256.30 256.40 256.40 255.40 254.80 256.40 255.05 254.90 FiGURe 3.4. late ROMaN/eaRlY BYZaNtiNe POtteRY 94 No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Cooking-pot Cooking-pot Storage jar Storage jar Storage jar Jug Juglet Juglet Jug Jug Bowl Bowl Basin Bowl Bowl Bowl Casserole Storage jar 512 512 512 512 512 512 512 512 512 512 252 142 142 240 341 141 142 122 3063/1 3070/2 3073/1 3077/1 3073/2 3063/2 3074/1 3066/1 3071/1 3078/1 2240/1 1190/1 1155/4 2196/1 3150/1 1180/4 1155/6 1057/1 255.30 254.90 254.70 254.65 254.70 255.30 254.70 255.20 254.80 254.65 255.80 256.20 256.30 256.15 256.03 256.42 256.30 257.70 Chapter 3: pottery Fig. 3.4. late Roman/early Byzantine pottery. 95 Itamar taxel imitation of a very common type of imported late Roman C/Phocaean Red Slip Ware (PRSW) bowls (Hayes’ Form 3, and see below), as noticed also in other sites in Judaea (Harper 1995: Fig. 14:100; Shurkin 2004: Fig. 16:1) and transjordan (Harrison 1994: Fig. 1:17). type 2: Rounded bowls and/or small basins with thickened rims (Fig. 3.14:3), which have parallels from the 6th-early 8th centuries in Judaea (Cohen Finkelstein 1997: Fig. 1:7; Rahmani 1991: Fig. 14:6) and the northern Negev (Ustinova and Nahshoni 1994: Fig. 3:4). type 3a: Rounded bowls with in-turned rim and sometimes horizontal combing beneath it (Fig. 3.17:10). these bowls have a clear southern orientation (i.e., the northern Negev and the southern coastal plain: Nahshoni 1999: Fig. 4:3; Nikolsky and Figueras 2004: Fig. 32:24, 27; Rosenthal-Heginbottom 1988: Pl. 4:173).However parallels were found also in Jerusalem (tushingham 1985: Fig. 29:10) and in nearby Khirbet el-Jiljil (Mlynarczyk 2005: Fig. 5:15). all these parallels are from the 6th7th centuries. type 3B: Similar to the latter, but with a slightly carinated body and a wavy combing below the rim (Fig. 3.23:7). the parallels to this bowl, albeit are not identical (Magness 1992a: Fig. 12:6; Nikolsky and Figueras 2004: Fig. 32:3), are also from the 6th-7th centuries. type 4: Deep bowl made of yellowish-greenish sandy ware, with everted, infolded rim and (originally) a carinated body (Fig. 3.14:2). these bowls were produced in the western Negev (Fabian and Goldfus 2004:14*, n. 3), and were typical mainly of the northern and western Negev and the southern coast in the 6th-7th centuries (Fischer and tal 1999a: Fig. 6.129:13; Nahshoni 1999: Figs. 4:8, 5:8; tubb 1986: Fig. 6:8). However, some reached also more northern sites – coastal as well as inland, like YavnehYam, Mazor and Khirbet es-Suyyagh. type 5: a bowl with carinated body and flattened rim (Fig. 3.23:6), with parallels in 6th-7th century assemblages in the central hill country (taxel 2005: Fig. 35:10; Wightman 96 1989: Pl. 40:9) and transjordan (Harrison 1994: Fig. 1:15). type 6: a bowl or basin with wide, up-turned ledge rim (Fig. 3.23:9), which has no close parallels. it could be a variant of the RilledRim Basins (Magness 1993:203-204, and see above, late Roman/early Byzantine bowls, type 5), although it lacks the characteristic ridges on the rim. type 7: a deep bowl with triangular rim, which is sometimes decorated with wavy combing (Fig. 3.23:8). it has a single 7th century parallel from the northern Negev (Nikolsky and Figueras 2004: Fig. 32:8). type 8: a rounded bowl with outfolded rim and rouletted decoration on the exterior (Figs. 3.14:4, 3.15:6). according to Magness, this is a later development of the Rouletted Bowl of the 3rd-5th centuries (see above, late Roman/early Byzantine bowls, type 3), which she dated to the 6th century (1993:187-189, Rouletted Bowls Form 2). Only a few examples of this bowl were found at the site. BASINS Nine types and sub-types of basins were found. Most of these basins are made of brown, orangebrown or pinkish-orange, usually high-temperature fired ware. type 1: a basin with narrow ledge rim, including a variant with a more vertical rim and a ridge beneath (Figs. 3.5:1, 2, 3.17:2, 3.21:1, 3.23:11). this is probably the second most widespread type of basin at the site. according to Magness, it is dated to the 3rd/early 4th-6th centuries (1993:204-206, arched-Rim Basins Form 1), although in Khirbet es-Suyyagh it appeared in contexts dated not before the 6th century. No parallels were found to the variant with the ridge below the vertical rim. type 2: a similar basin, with horizontal and wavy combed decoration on the walls and sometimes on rim (Figs. 3.5:3-6, 3.9:1, 3.13:2), which is the most common type of basin in the site. Magness dated it to the 6th-7th/early 8th centuries (1993:206-207, arched-Rim Basins Form 2a). Chapter 3: pottery type 3: a deep bowl/basin with almost vertical walls, thickened, everted rim and horizontal combing beneath (Fig. 3.15:4). it seems to be a variant of a type dated by Magness to the 6th-late 7th/early 8th centuries (1993:209, arched-Rim Basins Form 3; for an identical parallel, also identified as Magness’s type, see: Rapuano 1999: Fig. 5:79). type 4a: a basin made of yellowish-brown, sandy ware, sometimes with a yellowish-greenish slip. it has a thickened rim, large loop handles and horizontal and wavy combing (Figs. 3.5:7, 3.9:2, 3.19:1, 3.23:1, 12, 13). this type is most common in the southern regions of the country in the 6th-7th and maybe until the early 8th centuries (Fischer and tal 1999a: Fig. 6.129:9; levy 1960: Fig. 5:1-3; Nahshoni 1999: Fig. 5:9; Nikolsky and Figueras 2004: Fig. 33:4). type 4B: With the same characteristics as the previous type, it lacks the yellowish-greenish slip and its thickened rim is turned inward (Fig. 3.23:10). this variant seems to have developed from type 4a. it was probably developed during the Umayyad period, and continued – according to parallels – into the >abbasid period (arnon 2008:127-128, types 425a-b; Cohen Finkelstein 1997: Fig. 2:1; eisenberg and Ovadiah 1998: Fig. 14:7; Greenhut 1998: Fig. 24:1). it is possible, therefore, that some basins of this type found atthe site are indeed from the >abbasid period. type 4C: a few specimens, which differ from the previous variants by their more protruding rim (Fig. 3.20:3). according to its shape, parallels (Feig 1985: Fig. 1:15; Fischer and tal 1999a: Fig. 6.134:20; levy 1960: Fig. 5:1) and context in the site, it seems to be closer to type 10a than to type 10B. type 5: a basin with thickened, sharply incurved rim and horizontal and wavy combed decoration (Fig. 3.8:5). according to Magness, this type is dated to the 8th-10th centuries (1993:210-211, incurved-Rim Basins), but other parallels show that it appeared already in the late (?) 7th century (Nikolsky and Figueras 2004: Fig. 33:5, 7; taxel 2005: Fig. 37:5-6). type 6: a basin with vertical walls, triangular rim and loop handles (Fig. 3.23:15) that includes only one specimen. a possible parallel (from the 6th-7th century?) was found at nearby Khirbet Fattir (leszczyc 2003: Pl. 5:75). type 7: a deep bowl/basin with thickened rim and horizontal combing, but with thinner walls than those of the heavy basins mentioned above (Fig. 3.24:1). Parallels point on a 6th-8th centuries date to this type (Nikolsky and Figueras 2004: Fig. 32:12-13; taxel 2005: Fig. 37:2). FINE ByZANTINE WARE (FBW) ANd RElATEd BOWlS the bowls belonging to the ceramic group called Fine Byzantine Ware (FBW) include, according to Magness’ classification, ten types and subtypes (1993:193-201). all are represented at the site. these bowls are characterized by a well fired, grey-brown ware, burnished on the interior and exterior. Some of the variants have incised or painted decoration. according to Magness, this ware was produced in the vicinity of Jerusalem (ibid.: 166), as indicated by their high distribution in the city itself and in Judaea in general. another type of bowl, which seems as derived from the FBW bowls, is also included here. type 1: a small, rounded bowl/cup with low ring base and wavy incised decoration below the rim (Figs. 3.5:8, 3.8:1, 3.14:8). Magness dated it to the mid6th-late 7th/early 8th centuries (1993:193-194, FBW Bowls Form 1a). type 2: identical to the previous type and dates the same (Magness 1993:193-195, FBW Bowls Form 1B), but lacks the wavy incision (Fig. 3.19:2). type 3: a dipper bowl with carinated body and wavy incision below the rim (Figs. 3.5:9, 3.15:1), which is dated to the 7th-8th century (Magness 1993:193-198, FBW Bowls Form 1C). type 4: very similar to type 2, excluding its rounded and grooved base (Fig. 3.18:2, 3.24:2). Magness dated it to the late 7th/early 8th-9th/10th centuries (Magness 1993:194-196, FBW Bowls Form 1D). it must be noted that in cases when the base of the bowl was not found, it was sometimes quit difficult to distinguish between types 2 and 4. type 5: a deep bowl/cup with incurved walls and flat base. Most of the vessels have “strips” burnishing 97 Itamar taxel on the outer wall (Figs. 3.18:3, 3.23:2). these bowls were dated by Magness to the 8th-9th century (1993:194-198, FBW Bowls Form 1e), although in Khirbet es-Suyyagh it seems that most of the bowls came from Umayyad contexts. this is the third most widespread type of FBW bowl at the site (table 3.1). type 6: a deep, conical bowl with straight walls (Figs. 3.6:1, 3.7:16), which was dated to the 7th-8th centuries (Magness 1993:194-197, FBW Bowls Form 1F:3). type 7: Represented by small, deep bowls with a sharp ridge below the incurved rim (Fig. 3.18:4), which was dated to the mid 7th-9th/10th centuries (Magness 1993:198-199, FBW Bowls Form 2a). However, the two examples of this bowl found in the site cannot be dated later than the Umayyad period. type 8: a wide, shallow bowl with rim either incurved, triangular or narrow horizontal ledge (Figs. 3.6:2, 3.17:1, 3.18:5, 3.19:3, 3.20:4, 3.24:3), which was dated to the mid 7th9th/10th century (Magness 1993:198-201, FBW Bowls Form 2B). this is the most common type of FBW bowl in the site (table 3.1). in cases where it was found in undisturbed assemblages, it seems to be dated to no later than the 8th century. type 9: a similar bowl with narrow, down-turned ledge rim (Figs. 3.20:5, 3.23:14), which was dated by Magness as the former type (1993:198-201, FBW Bowls Form 2C). However, its presence in early 7th century assemblages (Magen Type 10 19% 1993:192, top; Watson 1992: Fig. 12:100) indicates that it appeared somewhat earlier than thought by Magness. like type 8, this bowl cannot be dated in Khirbet es-Suyyagh later than the 8th century. type 10: With thicker walls, wide, horizontal ledge rim and high ring base (Figs. 3.9:3, 3.14:5, 9, 3.15:7). this type, which was dated as the three preceding ones (Magness 1993:198-201, FBW Bowls Form 2D), represents the second most widespread type of FBW bowl at the site (table 3.1), and, like them, also seems to be attributed to the early/mid-7th–mid-8th century at the latest. type 11 (related FBW): Close in form and size to the deep FBW bowl/cup of type 5 (above), but it is made of coarser, brownorange ware and is not burnished. On the other hand, it is decorated with reddishbrown painting of a thin horizontal line on its rim and a thicker wavy line on the body (Fig. 3.19:4). this bowl/cup belongs to a wider ceramic group known as RedPainted Ware (RPW), which originated in transjordan and was mostly common there and in northern Palestine (see also below, Jug type 8). the bowls/cups belonging to this family probably appeared around the mid 8th century, and continued into the 9th century (Walmsley 1995:661, Fig. 6:3). the single example of this vessel found at the site originated indeed in a late Umayyad or early >abbasid context. Type 1 Type 2 11% 6% Type 9 6% Type 4 8% Type 8 26% Type 7 Type 6 4% 2% taBle 3.1: DiStRiBUtiON OF FBW BOWlS. 98 Type 3 2% Type 5 16% Chapter 3: pottery AFRICAN REd SlIP WARE (ARSW) BOWlS Only one type of aRSW bowl was found in relation to late Byzantine-early Umayyad assemblages at the site. this bowl is characterised by a thickened, concave ledge rim (Figs. 3.9:4, 3.15:3) and was dated by Hayes to 600-650 Ce (1972:171, Form 107). Only a few examples of it were found at the site. lATE ROMAN C/PHOCAEAN REd SlIP WARE (PRSW) BOWlS the most widespread group of lRRW bowls found in Khirbet es-Suyyagh (table 3.2), as in most of the contemporary levantine sites, is the lRC/PRSW. this ware, which was produced in western asia Minor between the late 4th and the mid 7th centuries, is represented by four types and subtypes. type 1: With a vertical rim, rectangular or triangular in section, sometimes decorated with rouletting (Figs. 3.6:3, 4, 3.9:5, 3.14:10, 3.24:4). this type includes many variants, which were dated by Hayes to the mid-5th until the mid-6th century (1972:329-338, Form 3). However, recent comparative studies show, in a high degree of certainty that the later variants of Hayes’ Form 3 continued to exist, at least in Palestine, until the late 6th or even the early 7th century (Magness 1999:193-194; tsuf 2003:138-140). this type is the most common type of lRC/ PRSW bowls in the site (table 3.3). type 2a: Seems to be a link between the former type and the next. it has a vertical, but knobbed square rim and it also lacks the rouletted decoration (Fig. 3.6:5, 3.9:6, 7, 3.15:2). Based on the dating of type 1 and that of type 2B, we can date this type to the late 6th century. type 2B: With a vertical rim, but more flattened and rectangular in section (Fig. 3.6:6, 7, 3.9:8, 3.14:12, 3.16:4). Hayes dated it to the late 6thearly 7th century (1972:343-346, Form 10a). this is the second most common type of lRC/PRSW bowls in the site (table 3.3). type 3: the latest development of lRC/PRSW bowls. it has a wide, flattened rim which continued the line of the wall (Figs. 3.6:8, 3.7:14, 17, 3.13:1, 103:3, 107:2), and was dated by Hayes to the first half of 7th century (1972:343-346, Form 10C). One of the low ring bases of the lRC/PRSW bowls found is decorated with a cross-monogram stamp on its bottom. this is a large Greek cross with double ribs, two circular 'pendants' below the side arms and the Greek letter rho (ρ) attached to the upper arm (Fig. 3.9:10). according to Hayes, this motif should be dated to the late 5th-early 6th century (1972:363-365, Motif 67). However, similar finds from various Palestinian sites indicate that lRC/PRSW bowls with this kind of cross-monogram appeared also in the late 6th century and maybe even later (aharoni 1964: Fig. 22:26; Gichon 1993: taf. 46:1-5; Mazar and Peleg 2003: Pl. i.15:2). CyPRIOTE REd SlIP WARE (CRSW) BOWlS the second most common group of lRRW bowls in the site (table 3.2) are those imported from Cyprus. two subtypes of CRSW bowls were found, although it is not always easy to distinguish between them because their base, which is the major feature that differs between the two, is always found separated from the rim. the two variants have rounded rims, concave from the interior and shallow rouletted decoration on the wall (Figs. 3.6:9, 3.8:2, 3.9:9, 3.14:11, 3.24:6). they were dated by Hayes to 550-600 Ce and to 580/600 Ce-end of 7th century (1972:379-383, Forms 9a and 9B). One of the base fragments of these bowls bears an impressed decoration of a small Greek cross (Fig. 3.15:8). according to Hayes, this type of cross impression is related to the bowls of Form 9a (1972:382, Fig. 84:k-o). EGyPTIAN REd SlIP WARE (ERSW) BOWlS the egyptian Red Slip Ware bowls are the third most common group of lRRW found in the site (table 3.2), with three representative types. type 1: a rounded bowl with vertical, slightly thickened rim (Fig. 3.16:1). in egypt, this form was dated to the 5th-8th centuries (Bailey 1998: Pl. 7:86; Rodziewicz 1976:56, 99 Itamar taxel Form O3, Pls. 23, 61). it was probably an imitation of aRSW bowls dated by Hayes to 230/240-300 and to the mid 2nd-mid 3rd century (1972:67-69, 200-201, Form 49 or 181), or an imitation of lRC/PRSW bowls dated to the late 4th-mid 5th century (ibid.:325-327, Form 1a or 1B). However, the egyptian bowls were also imitated, as indicated by some Palestinian specimens (for instance: tzaferis 1975: Pl. 15:1-3, 5-6). the lack of mica in the clay of the present example may also point to a local origin, although its pale brown, porous clay is identical to that of the egyptian bowls. although most of the pottery sherds found together with this bowl are dated to the 6th-8th centuries, its exact date is unclear, since among these sherds was also a local Rouletted Bowl of the 3rd-5th centuries (see above, late Roman/early Byzantine bowls, type 2). type 2: With a vertical, rectangular rim, and low ring base (Fig. 3.14:13). Hayes included this type among his eRSW ‘a’, and dated it to the early 6th-late 7th century (1972:389-392, Form J). Watson, however, believes that these bowls were imported to the southern levant only since the 7th century and until the first half of the 8th century (1995:305). indeed, at least in Palestine it is found in assemblages dated also to the late Umayyad period (Stacey 1988-89: Fig. 2:1-4; Watson 1995: Fig. 1:1-5). type 3: With more everted walls and thickened rim, grooved from the interior (Figs. 3.10:1, 3.17:4, 3.18:6, 3.24:5). it is included within Hayes’ eRSW ‘C’, and was dated to 620-700 Ce or later (1972:399-401). according to Watson, the finds from Pella in transjordan show that these bowls were imported into the region already in the second half of the 6th century (1995:310-311, Fig. 1:6). in Khirbet es-Suyyagh, this type is the most common among the eRSW bowls (table 3.4). 100 COOKING WARE thirteen types and subtypes of cooking vessels were identified in the ceramic assemblage of the late Byzantine and Umayyad periods; five are open cooking vessels and eight are closed cooking-pots. all the types are made of a reddishbrown, brittle cooking-pot ware. CaSSeroleS, fryIng panS and CooKIng ware lIdS type 1: a deep casserole with thin, ribbed walls and two horizontal handles (Figs. 3.15:5, 3.17:5, 3.19:6). it is included within the type dated by Magness to the late 3rd/early 4th8th/9th centuries (1993:211-213, Casseroles Form 1), although this certain variant is most widespread in assemblages dated to the 6th7th century (ibid.:211). type 2: a casserole which differs from the latter in its thicker and plain walls and larger and coarser handles (Figs. 3.6:10, 3.19:5). it seems that this variant was developed during the 7th century (Magness 1993:211) and continue into the Umayyad/>abbasid period, but at the site it can be attributed to no later than the 8th century. type 3: a deep casserole with thin, in-turned and plain walls and thin horizontal handles (Fig. 3.16:7). Magness dated this type to the late 7th/early 8th-9th/10th century (1993:214, Casseroles Form 3). type 4: a shallow frying pan with horizontal, hollow wishbone handles (Figs. 3.7:18, 3.10:2). this type is most typical to the central and southern regions of Palestine, and was dated by Magness to the 6th-7th century (1993:212, Casseroles Form 2). type 5: a casserole lid with knob handle (Figs. 3.8:3, 3.17:16, 3.19:7), which is usually perforated to allow the releasing of steams during cooking. these lids appeared in the late Roman period and continued until the >abbasid, almost without any change (Magness 1993:215). Chapter 3: pottery ERSW 8% Type 3 16% CRSW General 4% Type 1 38% 14% ARSW 3% Type 2B 26% LRC/PRSW 75% table 3.2. Distribution of lRRW bowls. Type 1 6% Type 2A 16% table 3.3. Distribution of lRC/PRSW bowls. Type 2 18% Type 3 76% table 3.4. Distribution of eRSW bowls. CooKIng-potS type 1: Cooking-pot with a short, vertical neck and plain rim (Fig. 3.6:11, 13). it has parallels in assemblages dated to the 6th-8th centuries (Magness 1995: Fig. 1:15; McNicoll, Smith and Hennessy 1982: Pl. 145:5). type 2a: Cooking-pot with a short – straight or slightly convex – neck, and outfolded rim (Figs. 3.10:3, 3.14:6). according to its parallels, it was common during the 6th7th centuries (Gichon 1993: taf. 39:7, 9-17, 20, 23-25, 28-33, 36-38; Magness 1995: Fig. 1:10-11; Watson 1992: Fig. 3:23). type 2B: Cooking-pot differing from type 2a in its higher and inward concave neck (Fig. 3.24:8). its parallels are contemporary to those of the former type (Gichon 1993: taf. 39:26, 27; Saller 1957: Fig. 48:515). type 3: Cooking-pot with a short, everted neck and thickened, triangular rim (Fig. 3.24:7). Magness dated it to the 5th/6th-late 7th/early 8th century (1993:219-221, Cooking-pots Form 4C), although most of its parallels are from 6th-7th century assemblages (Fischer and tal 1999b: Fig. 6.152:1; Ustinova and Nahshoni 1994: Fig. 6:22). type 4: Cooking-pot with a short, vertical neck and thickened, everted rim (Figs. 3.17:6, 3.24:9), which also has 6th-7th centuries parallels (Hamilton 1944: Fig. 7:11; Rosenthal-Heginbottom 1988: Pl. 4:195; Whitcomb 1989: Fig. 4:i). type 5: Cooking-pot with a short, slightly everted neck and hooked rim (Figs. 3.6:12, 3.10:4). according to Magness, it is dated to the 5th/6th-late 7th/early 8th century (1993:219-220, Cooking-pots Form 4B). type 6: Cooking-pot with a short neck and a flattened, everted rim (Figs. 3.10:5, 3.24:10). according to its parallels, this type was common mainly in the south of Palestine during the 6th7th centuries (Fabian and Goldfus 2004: Fig. 9:4; Gichon 1993: taf. 40:3-11, 15; RosenthalHeginbottom 1988: Pl. 4:197). 101 Itamar taxel type 7: Neckless cooking-pot with triangular rim (Fig. 3.18:10). Such cooking-pots are characteristic mainly of the north of Palestine. this particular type has 6th-7th century parallels (Calderon 2000: Pl. 22:45; Watson 1992: Fig. 2:11). type 8: Cooking-pot with a vertical, ridged neck and slightly everted, thickened rim (Fig. 3.19:8). according to parallels, this type, that characterized the central and southern parts of the country, appeared in the late 7th or early 8th centuries, and continued maybe into the 9th century (arnon 2008:73, types 711a-c; avner 1998: Fig. 13:8-10; Nikolsky and Figueras 2004: Fig. 46:9-11). the single example of this type found in the site originated indeed in a late Umayyad or early >abbasid context. STORAGE JARS ANd AMPHORAE ten types and sub-types of local and imported storage jars and amphorae from the late Byzantine and Umayyad periods were identified at the site. Most of the types are typical of Judaea and the central hill country, and only some of the types originated in other parts of Palestine. type 1: like the three following types, this is a bag-shaped jar of the group typical mainly of Judaea and the southern Samaria Hills in the late Byzantine and Umayyad periods. these jars are characterised by a hard, yellowish ware, relatively thick walls and a ridge at the base of the neck. this particular type has a ribbed body and shoulder, relatively short (vertical or slightly inverted) neck, and a rim thickened from the inside (Figs. 3.7:1, 2, 3.13:3, 3.14:16, 3.17:17). this type, which is the second most widespread in the site (table 3.5), was dated by Magness to the late 6th-7th century (1993:223-226, Storage Jars Form 4C). type 2: Differs from type 1 in its thicker walls, horizontally-combed shoulder and (usually) shorter neck (Figs. 3.7:3, 3.13:5, 6, 3.14:17, 18, 3.15:9, 3.17:7). this is the commonest type of storage jar at the site (table 3.5), and it was dated by Magness to the late 6th-early 8th centuries (1993:226-227, Storage Jars Form 5a). 102 type 3: larger and heavier jar with ribbed body, horizontal and sometimes wavy combing on shoulder and high, vertical neck (Fig. 3.13:7, 3.17:8). it was dated by Magness to the late 6th/7th-8th centuries (1993:227-230, Storage Jars Form 6a). this is the third most common type of jar in the site (table 3.5). type 4: identical to the latter, excluding its inturned neck (Fig. 3.20:1), and it is dated the same (Magness 1993:227-230, Storage Jars Form 6B). type 5: a holemouth jar with a stepped, flattened rim and a sharp ridge at its base, which was made from the same ware as the four previous types (Fig. 3.11:10). this jar, which is also typical to Judaea and dated by Magness to the 5th-6th centuries (1993:231-233, Holemouth Jars Form 1B), is represented in the site only by a single rim. type 6: a southern Palestinian bag-shaped jar, which differs from the first four types by its sandy, orange-brown ware, thinner walls and short, convex neck which lack the ridge at its base (Fig. 3.24:11, 12). it is mostly common in the southern and central coastal plain (adanBayewitz 1986: Fig. 1:6-7; tubb 1986: Figs. 3, 4:2), but appeared in large quantities also in more inland regions (Rosenthal-Heginbottom 1988: Pl. 2:26-27, 59, 90). For instance, at îorvat Zikhrin, in the western slopes of the Samaria Hills, it represents the most widespread type of jar (taxel 2005:68, Fig. 41:12-15). therefore, it is quite interesting that only few examples of this type were found in the site (table 3.5, and see below). according to Kingsley, this jar appeared only in the late 6th or early 7th century, and continued to exist until the late 7th century (1994-95:42; 2002:36), as indicated by most of its parallels. the jar in Fig. 3.20:2 seems to be a late variant of this type. it was found in a locus dated to around the mid 8th century, and differs from the earlier jars by its hard reddish-brown ware, buff slip, relatively sloping shoulders, and plain body. Possible parallels to this jar, some with ribbing on part/s of the body, were found Chapter 3: pottery in late Umayyad and >abbasid assemblages in Caesarea (arnon 2008: 223, type 931e) and Sde Boqer in the Negev (Nevo 1991: Pl. 5:1-3). in a petrographic analysis of this jar it was found that the vessel was made of a terra rossa clay with a ferugonous matrix. the clay contains medium sorted quartz (ca. 25%), rounded, worn lime grits (ca. 3%), and nari grits (ca. 1%). Terra rossa soil characterizes the main part of the central hill country. these facts indicate that the discussed jar most probably originated in the Judaean Hills, but its morphology was influenced more by the coastal bag-shaped jars rather than by the hill country jars. type 7: a northern Palestinian bag-shaped jar, characterised by its hard, metallic grey ware and white painting of vertical lines on the body. these jars were manufactured and were most common in the northern regions of the country (see, for instance, landgraf 1980:67-80), although it is found in small quantities also in southern sites, such as Jerusalem (Mazar and Peleg 2003: Pls. i.14:6, i.16:22-23). the single body sherd of such a jar which was found in the site (not illustrated) fits this scheme. type 8a: Represented by one of the variants of the so-called ‘Gaza/ashkelon amphora’, which is typical to the central and southern coastal plain of Palestine. this variant, with the vertical, concave rim and the trimmed base (Fig. 3.24:13), is dated to the 5th-6th centuries (Majcherek 1995:168-169, Form 3, Pl. 6) and it is represented in the site only by a few sherds. type 8B: the latest variant of the 'Gaza/ashkelon amphorae', with a plain rim that continues the line of the shoulder (Figs. 3.7:4, 15, 3.16:6). this variant is dated to the 6th-7th centuries (Majcherek 1995:169, Form 4, Pls. 7, 8) and appeared at the site in a much higher number than the previous type. type 9: a small bag-shaped jar, made of reddishbrown, highly micaceous ware, which has a low, convex neck (Fig. 3.18:11). this is an egyptian import, known as Red-Brown Ovoid amphora, which is relatively well known in Palestinian assemblages. in egypt they were produced from to the 7th to the 10th century (Gayraud 2003:559; Watson 1995:319), but common opinion maintains that in the southern levant their presence ceased at the end of the Umayyad period (Walmsley 1995:66; Watson 1995:319). However, some as yet unpublished examples of these jars found at Jaffa and other sites indicate that their import to Palestine continued well into the 9th or 10th century. Only a few sherds of these jars were found in Khirbet es-Suyyagh, in loci which cannot be dated prior to the 8th century. type 10: another imported type is an amphora made of yellowish ware, which has a narrow neck, thickened rim with a ridge beneath and two handles from neck to shoulder (Figs. 3.7:5, 3.24:14). this amphora was produced in the northeastern Mediterranean basin (Cyprus, coastal Syria and southern asia Minor) between the 5th and 7th centuries (Peacock and Williams 1986:185-187, Class 44a), and it is well known throughout Palestine (Calderon 2000: Pl. 19; Crowfoot and Fitzgerald 1929: Pl. 14:29; RosenthalHeginbottom 1988: Pl. 3:126-130). at Khirbet es-Suyyagh it is represented only by few sherds. the original content of these amphorae is uncertain. Peacock and Williams (1986:187) and Kingsley (1999:168) suggested that it was oil, while Calderon preferred to identify it as wine (2000a:133). according to Bass (1982:164-165) and Decker (2005:57, n. 32), the amphorae could have contained either oil or wine, although wine was the primary commodity exported in them. JUGS ANd JUGlETS twelve types and sub-types of coarse or semi-fine ware jugs and juglets were identified in the late Byzantine/Umayyad ceramic assemblage. Unless otherwise noted, these vessels are made of orangebrown, yellowish-brown or buff ware, usually high-temperature fired. 103 Itamar taxel Type 7 0% Type 4 3% Type 8B 8% Type 9 Type 10 2% 2% Type 8A 2% Type 1 30% Type 6 5% Type 5 0% Type 3 10% Type 2 38% table 3.5: Distribution of late Byzantine/Umayyad storage jars and amphorae. JugS type 1: a large jug with a thickened, everted or vertical rim, ridges on the upper part of the neck and thick strap handle from rim to shoulder (Fig. 3.24:16). this is the commonest type of jug in the site, and it has 6th-7th century parallels in Judaean and southern sites (Fischer and tal 1999a: Fig. 6.147:4; Harper 1995: Fig. 12:57; Mazar 2003a: Pl. iii.5:12). Some of the jugs of this type had hollow handles with a spout at their top (Figs. 3.7:6, 3.14:14), which were probably used for pouring the liquids storage in the jugs. this variant also has contemporary parallels in the same geographic area (aharoni 1962:Fig. 3:14; Crowfoot and Fitzgerald 1929: Pl. 13:23; Fischer and tal 1999a: Fig. 6.148:1; de vincenz 2005: Fig. 9:5-7). Few other handles have a crescent-like impressed decoration (Figs. 3.10:6, 3.17:12; leszczyc 2003: Fig. 126; Mlynarczyk 2005: Fig. 6:10; tushingham 1985: Fig. 30:12; de vincenz 2005: Fig. 9:3). it was sometimes difficult to distinguish between type 1 jugs to those of type 8 of the FBW jugs (see below), and it seems justified enough to entitle the former as a ‘related ware’ of the FBW jugs. type 2: a jug with a low, slightly convex neck, thickened rim and sloping, ribbed shoulder (Fig. 3.24:15). the brown, relatively hightemperature fired ware of this jug is close to that of the FBW jugs (see below), and it has 6th-7th century parallels from >ein Boqeq in the south (Gichon 1993: taf. 28:26-28). 104 type 3a: a jug with a narrow, cylindrical neck, everted rim and a handle projecting from it (Fig. 3.7:7). it has parallels in 6th-8th century assemblages from southern Palestine (Harper 1995: Fig. 12:53; Oked 1993: Pl. 5:3, 17:2). type 3B: Differs from the latter by the ridge approximately in the middle of its neck (Fig. 3.7:8). it has no parallels. type 4: a jug with a narrow, long cylindrical neck and plain rim (Fig. 3.25:1), that also lacks parallels. type 5: Jugs with low funnel-shaped, ridged neck, globular body and handle from rim to shoulder. One of the jugs has an X-like incision on its handle (Figs. 3.10:7, 3.11:1). these jugs were produced also from finer, hard-fired ware, and this variant, which was also found in the site, is described below within the FBW jugs and juglets (type 6). these jugs are typical to the central hill country and mainly to Judaea in the 6th-7th/early 8th centuries, although from the published parallels it cannot always be known if the vessel belongs to the coarse or the fine variant of the type (for coarse examples, see: Fischer and tal 1999a: Figs. 6.135:5, 6.145:10; taxel 2005: Fig. 43:10; de vincenz 2007: Pl. 25:33). type 6: a jug which has a low cylindrical neck with a ridge at its base, trefoil rim with a ridge beneath, strap handle projecting from the rim, and wide shoulder decorated with horizontal and wavy combing (Fig. 3.10:9). No exact parallels to such jugs were found. large closed Chapter 3: pottery vessels from the 6th-7th century, with wide shoulders decorated with combing, ridged at the base of the neck and single handled were published from Karkur >illit (Nikolsky and Figueras 2004: Fig. 44:11-13). However, their necks are missing so we cannot know if these vessels were like ours. type 7: a well-known form of jug made of cookingpot ware, which has a funnel-shaped neck with a ridge at its base, trefoil rim and handle from rim to shoulder (Figs. 3.10:8, 3.11:2). Magness dated it to the 6th-7th centuries and included it within the pottery which characterises the area of Jerusalem (1993:245, Jugs and Juglets Form 5), although it is very common also in southern Palestinian sites (e.g. Fischer and tal 1999a: Figs. 6.134:25, 6.135:4; Gichon 1993: tafn. 25:10-15, 26:1-5; de vincenz 2007: Pl. 25:18-20). type 8: Represented by body sherds of large jugs/ table amphorae decorated with crescent-like impressions (Fig. 3.25:2). Jugs and jars with a similar decoration were found mainly in the central hill country and the northern Negev, where they were dated to the 6th-8th century (leszczyc 2003: Fig. 127; taxel 2005: Fig. 42:9-12; de vincenz 2003: Fig. 23:200; 2003b: Fig. 2:12; 2005: Figs. 1:10, 13, 14, 8:1-11, 9:1, 2). according to de vincenz, the high frequency of these vessels in sites in the area of Beth Shemesh (i.e., Khirbet Fattir and Khirbet elJiljil) may indicate that they were produced somewhere in this region (2005:137). the jug neck with the strainer at its base, which is decorated with an impressed zigzag pattern (Fig. 97:4), cannot be related for sure to the previous body sherds. a similar jug neck, although undecorated, was published from ‘ein Boqeq (Gichon 1993: taf. 23:6). type 9: Jugs with a wide cylindrical or convex neck with a slight ridge at its base, thickened rim, carinated shoulder, strap handles and painted decoration of horizontal, wavy and diagonal red strips, sometimes over a white slip (Figs. 3.7:9, 3.16:9, 3.18:12, 13, 3.19:10). these jugs belong to the same ceramic group known as Red-Painted Ware (RPW), originated in transjordan (see above, related FBW bowl type 11). this certain type of jug probably appeared in the first half of the 8th century, but flourished mainly during its later part and in the 9th century (Walmsley 1995:661, Fig. 6:2). the few examples revealed in the site were indeed found together with pottery dated to the late Umayyad and >abbasid periods. JugletS type 1: a juglet with a narrow, thickened base and cylindrical body (Fig. 3.25:3), which is probably typical to the south of Palestine in the 6th-7th century, as indicated by its parallels (Fantalkin 2000: Fig. 9:11; RosenthalHeginbottom 1988: Pl. 4:162; Ustinova and Nahshoni 1994: Fig. 5:5). type 2: a juglet with a flat, string-cut base and cylindrical ribbed body with a handle emerging from its upper part towards the (missing) neck and rim (Fig. 3.11:3). this juglet belongs to a type already discussed above (late Roman/ early Byzantine Juglet type 4), which was dated by Magness to the 3rd-8th centuries (1993:246-247, Jugs and Juglets Form 6a). the relatively course ware of the present juglets and the fact that it was found together with late Byzantine/Umayyad pottery, indicate that this vessel represents the later form of the type. Such juglets were published also from other southern Palestinian sites of the 6th7th/8th centuries (Fantalkin 2000: Fig. 9:11; Gichon 1993: taf. 28:5). FINE ByZANTINE WARE (FBW) JUGS ANd JUGlETS Magness’ typology of FBW jugs and juglets includes four types dated by her to the mid-6th– early 8th centuries. Some are divided into subtypes (1993:237-241), all of which were identified at Khirbet es-Suyyagh. in addition to these types, we included within this ceramic group six other types of jugs and juglets, which were made of the same hard, brown-orange fine ware. 105 Itamar taxel JugS type 1: a jug with wide, cylindrical neck, triangular, everted rim, a loop handle from rim to shoulder and sometimes a spout (Figs. 3.20:6, 3.21:3, 3.25:5; Magness 1993:237-239, FBW Jars, Jugs and Juglets Form 1B). this is the most widespread type of FBW jug at the site. type 2: a jug with wide, short neck and everted (thickened or triangular) rim (Figs. 3.7:10, 3.11:6, 7; Magness 1993:237-239, FBW Jars, Jugs and Juglets Form 1C). Some of the vessels have incised decoration on their shoulder (Fig. 3.11:5; Mazar 2003b: Ph. i.92; 2003c: Pl. ii.1:3; Mazar and Peleg 2003: Pl. i.16:29). type 3: a jug which can be performed in different forms of body, but has a similar, although larger, neck as the former type (Figs. 3.7:11, 3.17:13; Magness 1993:239-241, FBW Jars, Jugs and Juglets Form 2B). type 4: a jug with funnel-shaped, ridged neck and thickened rim (Fig. 3.7:12). its parallels were found in 6th-8th century sites in Jerusalem and its vicinity (Mazar 2003a: Pl. iii.5:8-11; Rapuano 1999: Fig. 8:119; Saller 1957: Fig. 60:4660). as said before, these jugs are the finer variant of a course ware contemporary jugs characteristic to the region (above, Jugs and Juglets type 5). type 5: a jug with wide, concave neck and plain rim (Fig. 3.19:9). No exact parallels were found to it, but the context of the single example revealed in the excavations points on a mid7th-8th century date to this type. type 6: a large jug/table amphora with a ridge at the base of the neck, triangular, everted rim, two thick strap handles from rim to shoulder and incised (and sometimes a horizontal combing) decoration in different patterns on the shoulder (Figs. 3.18:7, 3.25:7, 8). this type seems to be a large variant of FBW Jugs type 1, and probably sometimes difficult to distinguish from the coarser Jugs type 1. the only parallels to such vessels were found at Jerusalem (tushingham 1985: Fig. 28:11, 26), from the second half of the 6th century. type 7: a jug which has a high, vertical neck with a ridge at its upper third, thickened rim, rounded 106 shoulder and a single handle from the neck’s ridge to the shoulder (Fig. 3.11:8). Such jugs (sometimes labeled table amphorae) – with incised-decorated body – were published from Jerusalem (Feig 2003: Fig. 25:2-3; Rapuano 1999: Fig. 8:114). type 8: a large jug/table amphora with a wide, everted neck and simple rim (Fig. 3.11:9). Only one possible parallel was found to this vessel, from a site in northwestern Judaean desert, where it was identified as Magness’ FBW jug Form 1C (above, our FBW Jug type 2; Sion 1997b: Fig. 6:13). However, the relatively high neck and the different rim form of the discussed vessel do not allow identifying it with the other jug type, and therefore it has to be defined as a separate type. JugletS type 1: a juglet with flat base, rounded body narrow neck, stepped rim, handle from rim to shoulder. the shoulder is incised-decorated with groups of three short lines (Figs. 3.8:6, 7, 3.17:9, 3.25:6; Magness 1993:239-241, FBW Jars, Jugs and Juglets Form 2a). type 2: a juglet with flat knobbed base and rounded or cylindrical body (Fig. 3.25:4). Parallels to such juglets, which the shape of their upper body is unknown, were found in southern Palestinian sites of the 6th-8th centuries (Fabian and Goldfus 2004: Fig. 9:18; Fischer and tal 1999a: Fig. 6.135:9-10, 6.148:8; Nikolsky and Figueras 2004: Fig. 35:18). FlASKS two types of late Byzantine/Umayyad flasks were found in the excavations. type 1: Made of orange-brown, hard-fired ware, which resembles that of the FBW Jugs and Juglets. it has a wide, ridged neck, which is narrow in its lower part (Fig. 3.14:7). this form is almost identical to that of the buff ware flasks of the late Umayyad and >abbasid periods (see below), and it was dated earlier only due to its different ware which apparently characterized the late Byzantine/Umayyad period. Chapter 3: pottery type 2: Made of cooking-pot ware, it has a short, concave neck and stepped rim (Fig. 3.7:13). it has a 6th–mid-7th century parallel from Caesarea (Magness 1995: Fig. 3:18). CylINdRICAl VESSEl a single, small cylindrical vessel with flat base was found in the area of the oil press (not illustrated). this type of vessel, which is sometimes called “ampoule”, is common in southern Palestine in assemblages dated to the 6th-8th centuries (Baly 1960: Pl. 53:88; Fabian and Goldfus 2004: Fig. 11:1; Nikolsky and Figueras 2004: Fig. 46:18-19; Ustinova and Nahshoni 1994: Fig. 6:33-34). the exact use of it is unknown, and according to Fabian and Goldfus it may have been used as a measuring vessel or as an oil lamps filler (2004:11*). PIlGRIMS’ AMPUllA Neck and body sherds of a single mould-made pilgrims’ ampulla were recovered. it has a narrow, vertical neck, plain rim, globular body and two handles from rim to shoulder. the moulded decoration consists of one horizontal band on the neck, and at least two arched aediculae (gates) with unidentified objects hanging from them and palm branches beneath on the body (Fig. 3.21:4). Petrographic analysis of this ampulla revealed that it was made of marl of the Moza formation, which originated in the area west of Jerusalem. this type of clay was used for pottery production in the Judaean hill country during many periods (Goren 2005:194). therefore, it may safely be assumed that this ampulla was produced in Jerusalem or its vicinity. Ceramic pilgrims’ ampullae are a relatively well-known find in Christian sites of the Byzantine and Umayyad periods (not before the 6th century) in Palestine, mainly from Jerusalem and Judaea. these ampullae and their contents (‘holy’ water, oil or earth) were bought by the pilgrims as blessing souvenirs (  ) from the Holy land (anderson 2004:80-82; israeli 2002:201, and references therein). the ampulla from Khirbet esSuyyagh is the first to be published from a clearly monastic site. the arched aedicula is a well-known artistic motif of eastern origin, which probably ought to resemble an arched shrine (avi-Yonah 1944:139-142). ampullae with the motif of arched aediculae are not very common. Only three parallels to such ampullae were found. the first is probably a Palestinian one, but its exact origin is unknown. its body is decorated with three aediculae; the two extreme ones have round dots along their inner side, and the central one has three objects, similar to those depicted on the ampulla from Khirbet es-Suyyagh, hanging from it (israeli 2002:202, bottom left). the origin of the second ampulla is also unknown (Palestine? egypt?). On both sides aediculae are depicted. in the middle of the first is a cross on a column and on the second there is an open double door with a figure standing (vikan 1982: Fig. 20). a third ampulla was found in Jerusalem. it is represented only by a small fragment which bears part of a Greek inscription below a motif that seems to be an aedicula (Di Segni 2007). the aedicula motif appeared also on glass pilgrims' bottles, probably originating in Jerusalem, which were dated to 578-614 Ce or until the end of the Byzantine period. However, within the aediculae these bear crosses on stepped bases, depicting the Holy Cross on the hill of Golgotha (Barag 1970a:45, Class 7). the objects hanging from the aediculae on the ampulla from Khirbet es-Suyyagh must also have a symbolic meaning. a similar arrangement of an aedicula with a round object hanging from it appears on a marble chancel screen from a church at Sussita. this object was first identified by the excavator as a lamp (anati 1957:32, Pl. 4:3) and later as a basket which symbolizes the Miracle of loaves and Fishes (epstein 1993:635). the latter interpretation may be applicable also in the case of our ampulla since the flat/horizontal upper part and rounded lower part is similar to that of a basket. lIdS ANd STOPPERS Five types of lids and stoppers were found. type 1: a bowl-shaped jar lid with flat base and everted, horizontal rim (Figs. 3.8:4, 3.13:4, 3.14:15, 3.25:10). it is made of thin, well-fired 107 Itamar taxel brown clay, identical to that of the FBW vessels. these lids were the most common in the site, and they were found only in assemblages dated to the mid-6th-7th century Ce. Magness identified these vessels as bowls, and included them within a type dated by her to the 7th-8th century or later (1993:194-198, FBW Bowls Form 1F:2). However, the excavations of Khirbet es-Suyyagh, in which these lids were found in storerooms beside jars, proved beyond any doubt that they are not bowls. the finding of such lids in 8th century assemblages (Baramki 1944: Fig. 8:7) indicates that their use continued at least until the middle of that century. type 2: identical to the previous one, except its courser ware (Figs. 3.11:11, 3.13:8, 3.17:18, 3.25:9). Coarse bowl-shaped jar lids, sometimes with different rims, are more common in Palestine, including in Judaea, where they were dated by Magness to the 6th-mid-8th century (1993:247-248, lids and Stoppers Form 1). this is the second most common type of lid at the site. type 3: a convex jar lid with a thickened, grooved knob handle (Fig. 3.25:11). this type is also made of a ware which resembles that of the FBW vessels. Only few parallels to such lids were found so far, in late Byzantine-Umayyad contexts (Calderon 1999: Pl. 4:9; taxel 2005: Fig. 44:22). type 4: a concave jar lid with a cone, concave handle (Fig. 3.25:12). it was dated by Magness to the 6th-mid-8th century (1993:248, lids and Stoppers Form 2). type 5: a stopper made of a body sherd of a jar, which was broken to form a roughly rounded shape (Fig. 3.17:11). although these stoppers are relatively common in Byzantine assemblages (Fischer and tal 1999a: Fig. 6.139:1; Magness 1992b: Figs. 59:10, 65:8), only two examples were found at Khirbet es-Suyyagh. lAMPS eight types and subtypes of late Byzantine and Umayyad lamps were found. type 1a: Represented by the well-known large 108 candlestick lamp, with the linear pattern (Figs. 3.12:1, 3.22:2, 3.27:1), that represents the most widespread type of lamp in the site. it was dated by Magness to the mid 6th-late 7th/early 8th centuries (1993:251-255, large Candlestick lamps Form 3a), but Hadad’s study of the lamps from Beth-Shean showed that these lamps were in use until the end of the Umayyad period (2002:66-68, type 28). type 1B: Differs from the latter by its wavy decoration pattern, but dates the same (Magness 1993:251-255, large Candlestick lamps Form 3B). Only one fragmentary example of it was found in the site (not illustrated). type 1C: another variant of the large candlestick lamp, which decorated with a Greek Christian inscription around the filling hole (Magness 1993:251-255, large Candlestick lamps Form 3C). Most of the inscriptions which appeared on these lamps are various versions of the formula “the light of Christ shines for all”, but other expressions, usually connected to Christ and/or light, also exist (loffreda 1990). according to Magness (1996) the lamps were manufactured in Jerusalem especially for the Christian pilgrims who visited the city. the nearly complete lamp (Fig. 3.22:3) has a bi-directional abbreviated inscription which starts from a common point (loffreda 1990:476, type D, Script Direction 9). it reads: . the central kappa (for  , lord) is flanked on both sides by a capital alpha and cursive upside down omegas (c.f. loffreda 1990:494, type D 4.6; Rosenthal and Sivan 1978:120, Nos. 498-499). the alpha-omega recalls the passage of Revelation 21:6: “i am the alpha and the Omega, the Beginning and the end”. the great resemblance between the present lamp and some of its parallels maybe indicates that they were made in the same mould. One of the fragmentary lamps (Fig. 3.27:2) has a counter-clockwise inscription, with some of the letters facing the centre of the lamp and some facing its periphery (loffreda 1990:476, type a or B, Script Direction 5 or 7). the inscription reads: Chapter 3: pottery  (   []   [], “the light of Christ shines for all”) (cf. loffreda 1990:483-486, type B 1.1-4.3). the nu and alpha are cursive, the pi is upside down, and the sigma is written as a capital letter. the inscription on another small lamp fragment (Fig. 3.18:9) is even less clear than the previous one; the remaining letters maybe include a phi and a sigma (abbreviation of   , light). type 1D: large candlestick lamps with high, stamped-decorated handle, which were dated by Magness as the three previous types (1993:251-255, large Candlestick lamps Form 3D). two fragments of such handles were found at the site. the first is in the form of a cross decorated with a spiral line and a pitchforklike pattern above it (Fig. 3.27:3). lamps with similar handles, although with different decorations on them, were found, for instance, in Jerusalem (Shapira and Peleg 2003a: Pl. i.17:16; 2003b: Pl. ii.4:7; tushingham 1985: Fig. 32:43). the second handle has a rounded finial, decorated with triangles (Fig. 3.27:4). this handle has some almost identical parallels from Jerusalem (Shapira and Peleg 2003a: Pl. i.17:13; tushingham 1985: Fig. 32:40) and the south (Fischer and tal 1999a: Fig. 6.135:5). another fragment, of a small, flattened loop handle decorated with two rows of square impressions (Fig. 3.12:2), has no parallels but it was tentatively attributed to the present type. type 2a: the earliest variant of the type of lamp which developed from the large candlestick lamp during the Umayyad period. it still has the narrow, pointed form of the late ByzantineUmayyad lamps, but also a channel-nozzle, small conical handle and geometric pattern (Fig. 3.22:4), which characterized the early islamic lamps. it was probably developed toward the end of the 7th century (Magness 1993:255-258, early Channel-Nozzle Oil lamps Form 4a). Only one specimen of this type was found in the site. type 2B: the later – and much more widespread – development of the previous type. it has a less pointed shape, higher conical handle and geometric decoration of short lines and dots (Figs. 3.12:3, 3.22:1, 3.27:5, 6). Magness dated its appearance and time of use to the second half of the 7th century until the early 8th century (1993:255-258, early Channel-Nozzle Oil lamps Forms 4B and 4C), but according to the finds from more northern sites it seems that it appeared only toward the early 8th century and continued to be used also in the early >abbasid period (Hadad 2002:82-95, type 36; Stacey 2004:149-150, Form 1a). type 2C: Seems to be a variant of the former type. it is decorated with two seven-branch menorahs from each side of the filling hole. the menorahs are baseless and have round arms (Fig. 3.26:1). this form of decoration on early islamic lamps is extremely rare. Only two more Umayyad period lamps decorated with two seven-branch menorahs have been published so far. these menorahs are also baseless and have round arms, but they are slightly different from those depicted on the lamp from Khirbet es-Suyyagh. the first lamp was found at îorvat Bireh in the southwestern slopes of the Samaria Hills (avissar 1997: Fig. 3:1), and the second lamp was found at Jaffa (erroneously dated to the 6th-7th century; Hachlili 2001: Pl. 90:l10.2). another, yet unpublished fragment of an Umayyad lamp decorated with a menorah was found at Khirbet Umm Deimnah, in relation to the postmonastic phase at this site (S. Batz, personal communication; for the preliminary report, see Magen and Batz 2008). the appearance of this symbol, which is mostly characteristic of the Jewish religion in the discussed period, is highly interesting. its probable meaning, with emphasis on the relationship to the certain context in which this lamp was found, will be discussed in Chapter 10. type 3: a wheel-made lamp with a rounded ribbed body and long nozzle (Fig. 3.27:7). this is a typical lamp of the southern coastal plain, southern Judaea and the Negev in the 6th7th/8th centuries (Gadot and tepper 2003: Fig. 18; Harper 1995: Fig. 19:3-7; Nikolsky 109 Itamar taxel and Figueras 2004: Fig. 47:6-8), and the single example found in Khirbet es-Suyyagh indicates that in the Judaean Shephelah this type was much less common. lANTERN a single lantern fragment was found in the site. it includes the hollow cylindrical handle which projected from the lantern’s body, and the beginning of the loop-like finial attached to this handle (Fig. 3.26:2). Such lanterns were relatively common in Judaea and southern Palestine in the late Byzantine and Umayyad periods (Corbo 1955: tav. 25.73:4; Fischer and tal 1999a: Fig. 6.135:6; Mazar 2003a: Pl. iii.7:5). WATER PIPES One complete section of ceramic water pipe was found in situ on the northeastern edge of the large courtyard, and fragments of another were found in the fill west of Cistern 24. they have a narrow opening with a ridge below, and a wider opening in the opposite end (Fig. 3.8:8). Such pipes were in use from the Roman period until modern times, but the context of those discussed here point on a late Byzantine or Umayyad date. ROOF TIlES Hundreds of roof tile fragments and a few nearly complete examples were found all over the site, but mainly in the church area. they include large flat lower tiles (tegulae) with square-sectioned fringes (Fig. 3.16:2, 3), and narrow convex upper tiles (imbrex; Fig. 3.16:4). Some of the tiles bear rounded impressions – mostly one but sometimes two – on their lower face. One lower tile has a unique impression, which resembles a hammer-like tool. No parallels were found for the latter impression, but the rounded impressions have many parallels from Jerusalem and sites throughout Judaea (avner 2000: Fig. 24; Mazar 2003b: Pls. i.3:3-4, i.5:1-6; Pele 2003: Pl. i.20:2, 4-5; Saller 1946: Pl. 36.1:8, 10-11, 14-22; tsafrir and Hirschfeld 1979: Fig. 27). according to Pele, the rounded impressions, which he dates to the 5th-7th centuries, are the second most common type of roof tile impressions 110 in the temple Mount excavations. in his opinion, these impressions were not potter’s marks but were designed to aid in counting quantities and output (2003:133). However, i prefer to identify the impressions as potter's marks representing certain workshops, as indicated by the variety of such impressions in the area of Jerusalem (for instance: lombardi 1956-1957: Figs. 8-9). the unique hammer-like impression probably symbolized a building tool, maybe of a kind used in roof construction. COARSE HANdMAdE VATS this title was given to two fragments of large, round vats with thick walls, which are made of a highly coarse ware. these features are typical also of ordinary tabuns, although only one of the discussed fragments can be identified as such. the first fragment, which was found in mixed late Byzantine->abbasid context, has at least three strips of shallow wavy combing below its rim (Fig. 3.16:8). Many similar fragments, dated to the >abbasid period, were found in an early islamic industrial site south of Ramla, where they were belonged to large, thick-walled storage vats (tal and taxel 2008: Figs. 6.67, 6.99). it is possible, therefore, that the fragment from Khirbet es-Suyyagh belonged to a similar, and maybe contemporary, vat. the second fragment had a wide, circular (?) opening (ca. 7 cm in diameter) below its rim, and it bears remains of soot (Fig. 3.17:14). in this case, we are probably dealing with part of a tabun. MIRROR PlAQUE One small fragment of ceramic mirror plaque was found in the site. this plaque (6 mm average thickness) belongs to the round type with a central sunken circle in which the glass mirror itself was placed. the central circle is surrounded by a sharp ridge, and the outer border of the plaque is reliefdecorated in a pattern of concentric circles around a central raised dot. around the outer circle there are remains of three (originally there were probably four) red-painted dots (Fig. 3.26:5). Such ceramic mirror plaques, with a similar or herringbone relief decoration on the outer border, were found Chapter 3: pottery in relatively many Byzantine (5th-7th century) assemblages throughout Palestine: in burial caves (avni and Dahari 1990: Fig. 7; Macalister 1911: Fig. 201; Rahmani 1964: Pl. 16a) as well as in domestic or other contexts (Colt 1962: Pl. 26:9-11; Fischer and tal 1999a: Fig. 6.142:13; Saller 1957: Pl. 130b:4; see also Saar 2003:79-86, 94). the most acceptable explanation for the use of these mirrors is as apotropaic objects, or, in Rahmani’s words: “the idea behind this practice was simply that any possible evil, inherent in the evil eye, would thus fall back upon itself in some sort of ‘auto fascination’” (1964:59-60). 3.26:7), and the second has a H-like incision (Fig. 3.18:14). the first marks may represent the number 9 or 11, and the latter mark maybe represents the number 8. However, we cannot rule out the possibility that these incisions are only mere potter’s marks and not registration marks relating to the jars’ contents. incisions on or beneath jar handles, mainly of X but also of other letters and marks, are quite common in Judaea (aharoni 1964: Fig. 8:12-13; Figueras 2004b: Fig. 49:11-14; Fischer and tal 1999a: Fig. 6.143:19; Saller 1957: Fig. 39:248, 250-252, 2641, 7318). POTTERS' MARKS ANd INSCRIBEd SHERdS SUMMARy Some body and handle fragments of local bagshaped storage jars of the late Byzantine-Umayyad period found in the site bear painted or incised marks and letters which have a commercial and religious connotation. One body sherd includes part of a crossmonogram (approximated measurements 8×8 cm), inscribed with red ink (Fig. 3.26:4). No parallels were found to this find. its clear Christian meaning is not surprising considering the monastic context in which it was found. it is possible that the jar which was marked with the cross-monogram contained a certain commodity that was especially sent to the monastery or sent from it. On another body sherd one can see part of an inscription – probably Greek – written in red ink. the small measurements of the inscribed fragment and the poor state of preservation of the ink do not allow identification of specific letters (Fig. 3.26:6). inscriptions on late Byzantine jars and amphorae, known as dipinti, are a common phenomenon, and they ought to denote the nature or quantity of the vessel’s contents (Fischer and tal 1999a:312). a third body sherd bears an incision which resembles the Greek letter a (Fig. 3.26:3), which could be part of an inscription. two of the jar handles found have incisions, probably made before firing. the first has an X-like incision and a vertical line beside it (Fig. the rich ceramic array reflected in the late Byzantine and Umayyad assemblages in Khirbet es-Suyyagh provides much information about the material culture, economic conditions and commercial connections of the inhabitants of the site during the 6th-8th centuries. as a whole, the ceramic array has a strong orientation to Judaea and the northern Negev. the preponderance of certain types of bowls and basins and all the known types of FBW characteristic of these regions, indicates the dependence of the site’s inhabitants on the ceramic industry of Jerusalem and its hinterland. this can be clearly seen by the distribution of local (i.e., Judaean) jar types at the site. Only 15% of the jars originated in the coastal region (table 3.6), indicating that most of the site's commercial connections took place within the limits of Judaea. the relatively high number of ‘Gaza amphorae’ points to the limited import of wine produced in the central and/or southern coast to the site. On the contrary, products that were transported in southern Palestinian bagshaped jars barely reached the site. the import of pottery from countries and regions outside Palestine including mainly lRRW bowls, most of them from asia Minor (lRC/PRSW). the distribution of lRRW bowls within the site (see table 3.2) is not unusual (see also aviam and Getzov 1998: table 1; avshalom-Gorni 2002: table 111 Itamar taxel 2), although the relatively high number of eRSW bowls is worth attention since these bowls usually appeared in much lower percentages in Palestinian sites (see Chapter 10). it is also interesting to note that the number of imported fine ware bowls at the site is higher by more than 10% than that of the local fine ware bowls (i.e., FBW; table 3.7; see also Gichon 1993:175, 207). imported jars and amphorae are also quite rare in the site, with a total of only 5% of the storage vessels (table 3.5). this fact is associated with the dominance of Judaean storage jars at the site, and is additional proof of the relatively limited economic connections of the inhabitants with regions outside Judaea (see Chapter 10). it is interesting to note that the local pottery at the site, made from fine, pale brown-grey clay, later became a well-fired and high-quality ware. this phenomenon is not restricted only to the bowls, jugs and juglets of the FBW group, but also to some basins of types 1, 3B, 8 and 9, jugs of types 1 and 2, flask of type 1 and jar lids of types 1 and 3. the main difference between most of these vessels and the ‘true’ FBW vessels is the lack of burnishing of the outer (and in cases of bowls, also the inner) walls. Only the jar lids have a finish identical to that of the FBW vessels. it is possible therefore, that some of the Judaean pottery workshops specialized in the production of well-fired small and medium-sized household vessels, which were utilized mainly by the local population. Greatest attention was given to the production of certain fine bowls, jugs and juglets, known as FBW, which were sought after also in regions outside Judaea, while the distribution of the less fine vessels was restricted mainly to Judaea. Coastal Types 15% FBW 44% LRRW 56% Inland Types 85% table 3.6. Distribution of local types of late Byzantine/ Umayyad storage jars. table 3.7. Distribution of late Byzantine/Umayyad fine ware bowls. FiGURe 3.5: late BYZaNtiNe/UMaYYaD POtteRY No. 1 2 3 4 5 6 7 8 9 112 Type locus Basket/Reg. No. Elevation Basin Basin Basin Basin Basin Basin Basin Bowl Bowl 240 240 240 330 240 358 240 368 240 2196/3 2142/1 2180/2 3159/7 2180/5 3228/1 2180/1 3214/3 2131/1 256.15 256.85 256.54 256.03 256.54 255.86 256.54 255.92 257.01 Chapter 3: pottery Fig. 3.5: late Byzantine/Umayyad pottery. 113 Itamar taxel FiGURe 3.6: late BYZaNtiNe/UMaYYaD POtteRY No. 1 2 3 4 5 6 7 8 9 10 11 12 13 114 Type locus Basket/Reg. No. Elevation Bowl Bowl Bowl Bowl Bowl Bowl Bowl Bowl Bowl Casserole Cooking-pot Cooking-pot Cooking-pot 368 240 358 240 240 240 368 368 330 358 240 368 368 3214/1 2142/5 3228/4 2180/10 2198/1 2180/9 3214/4 3214/5 3159/9 3228/2 2196/2 3214/7 3214/8 255.92 256.85 255.86 256.54 256.10 256.54 255.92 255.92 256.03 255.86 256.15 255.92 255.92 Chapter 3: pottery Fig. 3.7: late Byzantine/Umayyad pottery. 115 Itamar taxel FiGURe 3.7: late BYZaNtiNe/UMaYYaD POtteRY 116 No. Type 1 23 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Storage jar Storage jar Storage jar Storage jar amphora Jug Jug Jug Jug Jug Jug Jug Flask Bowl Storage jar Bowl Bowl Casserole locus 240 368 368 368 368 368 240 330 240 368 240 368 240 386 386 356 356 356 Basket/Reg. No. 2198/2 3214/9 3214/10 3214/11 3214/12 3214/13 2140/1 3159/1 2158/ 3214/14 2131/2 3214/15 2140/2 3260/1 3260/2 3183/1 3183/2 3178/1 Elevation 256.10 255.92 255.92 255.92 255.92 255.92 256.90 256.03 256.71 255.92 257.01 255.92 256.90 255.99 255.99 255.45 255.45 255.93 Chapter 3: pottery FiGURe 3.8: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 Bowl Bowl lid lid Basin Juglet Juglet Pipe 260 260 260 260 259 259 259 205 2247/1 2247/2 2247/3 2247/4 2207/1 2199/1 2208/1 2046/1 256.65 256.65 256.65 256.65 256.25 256.37 256.19 257.85 117 FiGURe 3.9: late BYZaNtiNe/UMaYYaD POtteRY No. 1 2 3 4 5 6 7 8 9 10 118 Type Basin Basin Bowl Bowl Bowl Bowl Bowl Bowl Bowl Bowl locus 600 287 293 293 287 182 287 182 182 600 Basket/Reg. No. 6000/5 2311/4 2318/1 2318/2 2312/7 1294/1 2321/5 1294/2 1294/3 6000/6 Elevation 257.40-258.00 257.87-258.10 257.70 257.70 257.87-258.10 257.83 257.87-258.10 257.83 257.83 257.40-258.00 Chapter 3: pottery FiGURe 3.10: late BYZaNtiNe/UMaYYaD POtteRY No. 1 2 3 4 5 6 7 8 9 Type Bowl Frying pan Cooking-pot Cooking-pot Cooking-pot Jug Jug Jug Jug locus 287 287 600 600 600 600 600 600 600 Basket/Reg. No. 2321/3 2312/1 6000/7 6000/8 6000/9 6000/10 6000/11 6000/12 6000/13 Elevation 257.87-258.10 257.87-258.10 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 119 Itamar taxel FiGURe 3.11: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 Jug Jug Juglet Jug Jug Jug Jug Jug Jug Storage jar lid 600 600 600 600 600 600 600 600 600 600 600 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 6000/8-18 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 257.40-258.00 120 Chapter 3: pottery FiGURe 3.12: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 lamp lamp lamp 287 182 296 2310/1 1294/4 2330/2 257.95 257.83 257.63-257.87 121 Itamar taxel FiGURe 3.13: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 Bowl Basin Storage jar lid Storage jar Storage jar Storage jar lid 114* 242 242 242 148 148 148 148 1061/1* 2159/1 2159/2 2153/1 1173/1 1167/1 1167/2 1182/1 257.74 257.01 257.01 257.10 255.75 256.05 256.05 255.58 122 Chapter 3: pottery Fig. 3.14: late Byzantine/Umayyad pottery. 123 Itamar taxel FiGURe 3.14: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Bowl Bowl Bowl Bowl Bowl Cooking-pot Flask Bowl Bowl Bowl Bowl Bowl Bowl Jug lid Storage jar Storage jar Storage jar 134 148 134 148 148 148 175 107 120 107 107 107 131 107 128 128 107 128 1117/5 1178/1 1127/1 1167/3 1178/4-5 1178/4-5 1276/1 1068/1 1047/2 1017/1-2 1017/1-2 1068/2 1105/1 1068/7 1085/4 1085/1 1035/1 1085/3 256.81 255.65 256.35 256.05 255.65 255.65 253.10 258.22 258.41 258.30 258.30 258.22 257.73 258.22 257.69 257.69 258.25 257.69 FiGURe 3.15: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 Bowl Bowl Bowl Bowl Casserole Bowl Bowl Bowl Storage jar 181 181 183 510 183 316 320 306 318 1293/1 1293/2 1291/6 3060/1 1291/1 3075/5 3057/2 3035/1 3078/1 258.05 258.05 257.82 255.70 257.82 256.85 256.71 257.05 256.68 124 Chapter 3: pottery Fig. 3.15: late Byzantine/Umayyad pottery. 125 FiGURe 3.16: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 Bowl Roof tile Roof tile Roof tile Bowl Storage jar Casserole vat Jug 320 318 316 316 349 349 322 321 322 3057/1 3067/3 3075/2-3 3075/2-3 3165/1 3165/2 3068/6 3064/4 3068/1 256.71 256.85 256.85 256.85 255.54 255.54 256.40 256.32 256.40 126 Chapter 3: pottery Fig. 3.17: late Byzantine/Umayyad pottery. 127 FiGURe 3.17: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Bowl Basin Bowl Bowl Casserole Cooking-pot Storage jar Storage jar Juglet Bowl Stopper Jug Jug vat/tabun Bowl lid Storage jar lid 390 343 361 343 343 361 343 361 361 210 210 506 277 265 271 271 271 271 3278/5 3145/2 3198/4 3145/8 3145/9 3197/1 3145/6 3198/1 3198/2 2080/6 2080/7 1332/4 2273/7 2220/5 2305/1 2302/3 2307/2 2306/1 256.15 255.72 256.02 255.72 255.72 256.20 255.72 256.02 256.02 258.15 258.15 257.74 256.10 256.41 255.90 256.15 255.37 255.69 FiGURe 3.18: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Bowl Bowl Bowl Bowl Bowl Bowl Jug Flask lamp Cooking-pot Storage jar Jug Jug inscribed storage jar handle 220 220 220 220 220 220 220 220 220 179a 179a 179a 179a 179a 2069/1-9 2069/2 2069/3 2069/4 2069/5 2069/6 2069/7 2069/8 2069/9 1286/1 1286/2 1286/3 1286/4 1286/5 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 257.38-257.62 253.30-255.82 253.30-255.82 253.30-255.82 253.30-255.82 253.30-255.82 FiGURe 3.19: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 Basin Bowl Bowl Bowl Casserole Casserole lid Cooking-pot Jug Jug 179B 179B 179B 179B 179B 179B 179B 179B 179B 179B 3088/2 1296/3 1296/4 3088/6 1300/1 3088/5 1300/2 3088/3 1300/7 3088/1 251.90 253.05 253.05 251.90 252.60 251.90 252.60 251.90 252.60 251.90 128 Chapter 3: pottery Fig. 3.18: late Byzantine/Umayyad pottery. 129 Itamar taxel Fig. 3.19: late Byzantine/Umayyad pottery. 130 FiGURe 3.20: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 Storage jar Storage jar Basin Bowl Bowl Jug 179B 179B 179C 179C 179C 179C 1300/3 3088/11 1302/1 1302/2 1302/3 1302/4 252.60 251.90 251.90-253.10 251.90-253.10 251.90-253.10 251.90-253.10 FiGURe 3.21: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 Basin Bowl Jug Pilgrims' ampulla 136* 174 174 136* 1121/4 1253/1 1253/2 1122/1 256.92 255.72 255.72 256.80 131 Fig. 3.21: late Byzantine/Umayyad pottery. Fig. 3.22: late Byzantine/Umayyad pottery. 132 Chapter 3: pottery FiGURe 3.22: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 lamp lamp lamp lamp 186 172 174 174 1308/1 1251/1 1252/1 1262/1 256.65 255.90 255.40 255.40 FiGURe 3.23: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Basin Bowl Bowl Bowl Bowl Bowl Bowl Bowl Bowl Basin Basin Basin Basin Bowl Basin 312 317 207 247 201 247 360 116 113 173 107* 346 201 360 123 3050/4 3063/1 2028/2 2138/3 2001/2 2164/6 3207/5 1040/1 1028/2 1247/6 1049/1 3160/1 2001/7 3233/1 1064/4 256.80 256.50 258.30 256.70 257.99 256.40 256.80 257.40 258.01 255.10 255.54 255.91 257.99 256.35 258.15 FiGURe 3.24: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Basin Bowl Bowl Bowl Bowl Bowl Cooking-pot Cooking-pot Cooking-pot Cooking-pot Storage jar Storage jar Storage jar amphora Jug Jug 201 360 165 360 217 336 329 275 375 202 155 214 247 329 252 375 2001/1 3233/2 1233/8 3233/4 2059/7 3120/1 3103/5 2252/3 3240/1 2006/4 1193/1 2064/3 2164/1 3103/2 2170/2 3240/1 257.99 256.35 256.80 256.35 258.30 256.45 256.29 256.26 255.75 258.91 256.35 257.61 256.40 256.29 255.30 255.75 133 Itamar taxel Fig. 3.23: late Byzantine/Umayyad pottery. 134 Chapter 3: pottery Fig. 3.24: late Byzantine/Umayyad pottery. 135 FiGURe 3.25: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 Jug Jug Juglet Juglet Jug Juglet Jug Jug lid lid lid lid 355 252 144 123 215 251 253 246 247 142 101 221 3187/1 2191/3 1209/1 1058/4 2066/2 2276/6 2266/5 2160/2 2152/1 1147/5 1006/1 2086/8 256.45 255.62 255.85 258.12 258.12 255.30 256.05 256.61 256.45 256.45 258.55 258.08 136 FiGURe 3.26: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 lamp lantern inscribed storage jar sherd inscribed storage jar sherd Mirror plaque inscribed storage jar sherd inscribed storage jar handle 251 226 334 216 271 214 100 2165/2 2106/1 3130/1 2084/1 2305/4 2060/7 1000/6 255.45 257.20 256.21 258.08 255.90 257.70 259.01 137 Itamar taxel FiGURe 3.27: late BYZaNtiNe/UMaYYaD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 lamp lamp lamp lamp lamp lamp lamp 249 212 136 236 150 163 236 2144/3 2035/1 1126/8 2179/5 1242/3 1222/6 2195/1 256.77 258.18 257.51 256.65 255.15 256.12-256.81 256.20 138 Chapter 3: pottery tHe >aBBaSiD PeRiOD Some of the ceramic types mentioned above are dated to the 8th-9th century (or even later), and it is possible that at least part of them are actually belong to the >abbasid phase at the site. the following paragraphs deals with those types which first appeared in the 8th century (i.e., in the late Umayyad or early >abbasid period), and therefore best represent this time. BOWlS Seven types of bowl were identified in the ceramic assemblage of the period. type 1: a rounded bowl with incurved rim, made of buff ware (Fig. 3.28:1). this is one of the most common types of early islamic bowls, mainly between the mid 8th-10th centuries (Kletter 2005: Fig. 14:5; Sion 2004: Fig. 10:17). type 2: almost identical to type 1 but has an outfolded rim (Fig. 3.28:2). this type, similar to bowls found in >abbasid assemblages at Ramla (Kletter 2005: Fig. 14:8; Sion 2004: Fig. 9:7-8), continues the late Byzantine and Umayyad tradition of rounded bowls, differing mainly by its buff slip or clay. type 3: a bowl with a down-turned, narrow and slightly concave ledge rim with an interior ridge and horizontal combing (Fig. 3.28:3). it has some parallels dated to the >abbasid period (Gophna and taxel 2007a: Fig. 3.6:3; de vincenz 2005: Fig. 2:3). type 4: a variant of a FBW bowl type with incurved walls and pointed rim (see above, FBW bowls type 5; Magness 1993:194-198, FBW Bowls Form 1e), which is decorated with white painting of diagonal lines and dots over black slip (Fig. 3.28:4). it has parallels in 8th-9th century assemblages; some are similar and even identical and some are decorated with black-over-white painting (arnon 2008:122-123, types 322e-f; Haiman 1995: Fig. 8:4-5, 7; Stacey 2004: Fig. 5.2:2-3, 6-7). type 5: a variant of another FBW bowl type, with sharp ridge below the incurved rim (see above, FBW bowls type 7; Magness 1993:198-199, FBW Bowls Form 2a). it is rather large than its prototype, and is decorated with a red slip from the exterior (Fig. 3.28:5). this variant seems to be developed in the 8th century, and according to parallels it continued into the 9th century (avissar 2003: Pl. 19.1:7; Baramki 1944: Fig. 7:10; Kelso and Baramki 1955: Pl. 30:a121). type 6: a black-burnished bowl with thick, vertical wall, pointed rim and incised decoration (not illustrated). these bowls are imitations of steatite incised-decorated bowls imported from the arabian Peninsula in the 8th-9th century (Magness 1994). type 7: Open buff ware bowls with everted rim and polychrome splash glaze from the interior and behind rim (Fig. 3.28:6). these highly common glazed bowls are dated to the 9th-11th centuries (arnon 2008:115-116, types 223a-e; avissar 1996: Fig. Xiii.2). STORAGE JARS three types of >abbasid period jars were found in the site. type 1: a large bag-shaped jar made of reddishbrown, high-temperature fired ware, which has a high, slightly convex neck with a ridge at its base and horizontal and wavy combed decoration on shoulder (Fig. 3.28:7, 8). this is the most common type of >abbasid jar in the site. Most parallels date it to the mid/late 8th-10th centuries (ayash 2000: Fig. 126:13; Rosen-ayalon and eitan 1969; Sion 2004: Fig. 11:41-42). type 2: a later variant of the large Judaean bagshaped jars (see above, late Byzantine and Umayyad jars types 1-4). it has a vertical neck with a ridge at its base and a wavy incised or combed decoration (Fig. 3.28:9). these jars are relatively well-known, but only few were published so far (for instance: arnon 2008:159, 222, 252, types 822b, 831g, 841; Kletter 2005: Fig. 19:11). according to parallels, this type is dated between the 8th-10th centuries. type 3: another variant of these jars, which differs from the latter type by its short, 139 Itamar taxel everted neck and thumbed decoration on rim (Fig. 3.28:10). its parallels date it to the mid/ late 8th-10th centuries (Baramki 1944: Fig. 4:4; Cohen Finkelstein 1997: Fig. 6:4; Sion 2004: Fig. 11:31). 1960:Pl. 56:20), amman (Harding 1951: Fig. 2:53) and Hammath Gader (Ben-arieh 1997: Pl. 15:14-19), and >abbasid ones were published from Caesarea (arnon 2008:72, 160, types 516a, 824a), Ramla (Kletter 2005: Fig. 17:2) and tiberias (Stacey 2004: Fig. 5.52). JUGS ANd JUGlETS Four types of jugs and juglets are included within the discussed pottery. JugS type 1: a typical Fine Buff Ware jug, with a funnelshaped neck and handle from rim to shoulder (Fig. 3.28:11). these jugs are usually dated to the 9th-11th centuries (arnon 2008:202-205, types 531a, b, d, f; Brosh 1986: Fig. 1:8; de vaux and Steve 1950: Pl. C:20, 24). this is the most common type of >abbasid jug in the site, although itself represented by only 6 examples. type 2: a larger and coarser variant of the previous type, and it sometimes has also a thickened rim (Fig. 3.28:12; arnon 2008:130, type 521f; Sion 2004: Fig. 12:52). JugletS type 1: a juglet with conical/globular body, short everted neck and basket handle projecting from the rim (Fig. 3.28:13). Only four published parallels were found to this uncommon form, all are probably from the early >abbasid period (Baramki 1944: Fig. 15:32; Fowler 1990:51, No. 8; Saller 1946: Pls. 34.1:39, 35.1:5; Sion 1998: Pl. 2:20). type 2: a small spheroid vessel with rounded base, short neck and triangular rim, made of metallic-fired reddish-brown ware and whitishslipped from the exterior (Fig. 3.28:14). these vessels, known also as 'grenades' or 'Greek Fire', were relatively common throughout the early islamic period. the exact use of these vessels is unknown (see Stacey 2004:138). the discussed example can be either Umayyad or >abbasid in date, since it was found in a very mixed locus. Umayyad parallels are known from Khirbet al-Karak (Delougaz and Haines 140 FlASKS One type of flask can be attributed to the >abbasid period. it is made of typically early islamic buff ware, but sometimes also of pale brown-orange ware. its neck is narrow at its base and wide and has sharp ridges at its upper part (Figs. 3.18:8, 3.29:6). these flasks appeared in the late Umayyad period (first half of 8th century), and continue to exist at least until the end of the >abbasid period (Baramki 1944: Fig. 5:10, 15:6; Rosenthal-Heginbottom 1988: Pl. 3:154). lAMPS as in the case of flasks, the >abbasid lamps are also represented by a single type. this is the almondshaped, mould-made lamp with the tongue handle and a channel-nozzle. these lamps, made of buff or brown ware, are dated to the late 8th-11th centuries (Hadad 2002:95-106, type 37; Magness 1993:258-259, Channel-Nozzle Oil lamps [Form 5]). ten examples of such lamps, represented mostly by fragments, were found in the site; four of which are presented here. One complete lamp and one fragment are decorated with stylized grape branches on body (Fig. 3.29:1, 2). this motif can be found on many >abbasid period lamps (Baramki 1944: Pl. 17:4, 8; tushingham 1985: Fig. 34:5). the third fragment is decorated on one side of the channel-nozzle with a stylized grape branch and a small bird standing on it (Fig. 3.29:3). this motif, which appeared most probably also on the other side of the channelnozzle, is also well-known from other >abbasid lamps (Hadad 2002: No. 453; Khalil and Kareem 2002: Fig. 21:1). the fourth lamp is almost complete. Both sides of its channel-nozzle are decorated with a four-legged animal (horse? lion?). two more Chapter 3: pottery Fig. 3.28: >abbasid pottery. 141 Itamar taxel FiGURe 3.28: >aBBaSiD POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Bowl Bowl Bowl Bowl Bowl Bowl Storage jar Storage jar Storage jar Storage jar Jug Jug Juglet Juglet/spheroid vessel 150 100 214 255 239 103 125 231 256 169 322 256 265 100 1186/7 1044/5 2060/10 2187/1 2140/5 1009/7 1079/1 2129/2 2214/6 1248/2 2080/1 2189/2 2220/8 1023/17 255.90 258.25 257.70 255.76 256.74 258.20 257.01 256.99 256.30 255.65 255.63 256.84 256.41 258.90 FiGURe 3.29: >aBBaSiD, MaMlUK aND late OttOMaN POtteRY No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 lamp lamp lamp lamp Smoking pipe Flask Bowl Bowl Jug 175 231 202 256 256 231 100 135 245 3081/1 2127/5 2002/2 2189/2 2189/6 2127/1 1074/8 1118/11 2137/9 252.60 257.17 258.80 256.84 256.84 257.17 257.30 257.03 255.82 142 Chapter 3: pottery Fig. 3.29: >abbasid, Mamluk and late Ottoman pottery. 143 Itamar taxel unidentified animals probably appeared also on the lamp’s body, but its worn condition precludes more accurate description (Fig. 3.29:4). Parallels for lamps decorated with four-legged animals exist, most of them probably from the >abbasid period (Brosh 1986: Fig. 5:12; Hadad 2002: Nos. 449-450; Macalister 1912: Pl. 101:1; Sion 2004: Fig. 14:89). SUMMARy the post-Umayyad pottery from Khirbet esSuyyagh belongs mostly to the >abbasid period. although some of the ceramic types continue to exist until the 11th century, no typical 11th century pottery was found at the site. Most of the types point to a late 8th-9th/10th century date, which also corresponds to the date given by the >abbasid coins found. type 1: a handmade, open bowl with thickened rim and a pinkish self slip over the very crude yellowish-grey clay (Fig. 3.29:7). this is one of the most common types of medieval (mainly Mamluk) bowls in Palestine, including Jerusalem and its vicinity (avissar and Stern 2005: Fig. 38:2-4). type 2: a rounded bowl, decorated with a gouged green glaze over yellowish slip from the interior to below rim (Fig. 3.29:8). this is a typical Mamluk period form (avissar and Stern 2005: Fig. 4). STORAGE JARS One type of Mamluk storage jar was found. it has a wide, cylindrical neck with a ridge at its middle, and flattened rim (Fig. 3.29:9; avissar and Stern 2005: Fig. 42:7-10). MaMlUK aND late OttOMaN PeRiODS Only a few pottery sherds dated to the Mamluk period (13th-15th century) were found at the site, mainly in topsoil loci. the amount of ceramic finds from the Ottoman period is even lesser than that of the Mamluk period. Only two fragments of the so-called dark grey ‘Gaza Ware’, characteristic to southern and central Palestine mainly in the 18th/19th-20th century, were found at the site (not illustrated). excluding these, the late Ottoman period is represented by a single fragment of a smoking pipe. this poor ceramic find points on a coincidental visits of passers-by and local peasants in the site during the 13th18th/19th centuries. BOWlS two types of bowl belong to the limited assemblage from the Mamluk period. 144 JUG a single type of Mamluk jug was found. it has a narrow, cylindrical neck, ridged upper part, and a single strap handle from the neck’s bottom to shoulder (not illustrated). this common type has many parallels from Mamluk assemblages, including from the area of Jerusalem (avissar and Stern 2005: Fig. 45:9). SMOKING PIPE the pipe is made of pale grey ware and covered from the exterior with a burnished purplish slip. it has a rounded bowl, which narrows toward the top and thickened rim. the bowl is decorated with notch rouletting and floral and geometric stamping (Fig. 3.29:5). this type of pipe, probably locallyproduced, is dated to the 18th century (Simpson 2000: Figs. 13.1:16-21, 13.2:25-31). CHaPteR 4 GlaSS itamar taxel the relatively small quantity of glass vessels and other objects found at the site is quite surprising and probably stems from the special needs and/ or economic abilities of the different people who inhabited the place throughout its history. the great majority of the glass is dated to the late Byzantine and Umayyad periods, although some of the finds can be dated to the >abbasid period. Only a few can be clearly dated prior to the late Byzantine period. BOWlS Four types of bowl were identified. the first is dated to the late Roman/early Byzantine period and the others belong to the early islamic period. type 1: an uncommon bowl, which has a wide, pinched scalloping rim with a ridge beneath (Fig. 4.1:1). a similar bowl fragment was found in Karanis, egypt, where it was dated to the 4th-5th centuries (Harden 1936: Pl. 14:259). Other, complete bowls of this type are kept in museums’ collections, and although two of them were found in Syria, in Harden’s opinion they all originated in egypt. it was also suggested that these bowls are copies of metal bowls, since a similar silver bowl was found in Karnak (ibid.:97). Palestinian parallels to this type were found at Khirbet Badd >isa, north of Jerusalem (Magen, tzionit and Sirkis 2004: Pl. 9:10), and at en-Gedi (Jackson-tal 2007: Pl. 6:1), where they were dated to the late Roman period. Other parallels were reported from Yavneh and lod (ibid.:483). it must be noted, however, that the present fragment was found in a late Byzantine/early Umayyad assemblage. type 2: with flaring walls and outfolded rim. this type is one of the commonest types of bowls in our region between the late Roman and >abbasid periods. the one shown here has a relatively wide rim (Fig. 4.1:2), and has parallels in Umayyad contexts (Hadad 2005: Pl. 3:58; Meyer 1987: Fig. 12:t). type 3: a deep bowl/cup with rounded walls and thickened, slightly incurved rim (Fig. 4.1:3). this type firstly appeared probably in the end of the Byzantine period, but flourished mainly in the Umayyad and >abbasid periods (Delougaz and Haines 1960: Pl. 59:14; GorinRosen 2000: Fig. 2:10; Hadad 2005: Pls. 1:1-3, 24-26, 25:500-502). these bowls are the commonest type of glass bowls in the site, and none was found in a clear post-Umayyad context. type 4: another deep bowl/cup, which differs from the previous one not only by its vertical walls but mainly by its decoration. the decoration consists of a pinched oval shape with a raised central boss (Fig. 4.1:4). Bowls/cups with similar or somewhat different pinched patterns were dated in our region to the >abbasid and Fatimid periods (Hadad 2005: Pls. 29:578, 32:637; lester 1996: Fig. Xvii.9:6; 2004: Fig. 7.14:174-175). GOBletS the single goblet type is represented by the wellknown ‘wine glass’ goblets of the late Byzantine and Umayyad periods, characterized by a concave or flat ring base, solid or hollow foot and slightly thickened rim decorated with a fused-in trail (Fig. 4.1:5, 6; Gorin-Rosen 1999: Fig. 2:24-26; Katsnelson 1999a: Fig. 2:9-14; 2004: Fig. 61:11-14; Meyer 1987: Fig. 8:aa-cc). this is one of the most widespread glass vessels in the site. BeaKeRS three types of beaker were found in the site, dating to the late Roman/early Byzantine and the late Byzantine-Umayyad periods. 145 Itamar taxel Fig. 4.1. Glass vessels. 146 Chapter 4: glaSS FiGURe 4.1: GlaSS veSSelS No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Bowl Bowl Bowl Bowl Goblet Goblet Beaker Bottle Bottle Bottle Bottle Bottle Bottle Bottle Bottle Bottle 600 319 142 100 251 346 227 179B 600 227 600 227 126 293 600 600 6001/3 3061/1 1162/2 1065/4 2298/1 3162/2 2095/5 1299/3 6001/1 2095/2 6001/7 2115/2 1087/3 2319/6 6001/4 6001/10 257.40-258.00 256.70 256.30 257.30 255.80 255.91 257.73 251.90 257.40-258.00 257.73 257.40-258.00 257.73 257.80 257.70 257.40-258.00 257.40-258.00 FiGURe 4.2: GlaSS veSSelS No. Type locus Basket/Reg. No. Elevation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Bottle lamp Window pane Bottle Bottle Bottle Bottle Jug Jug Jug Jar Jar Dropper Mould-made vessel Mould-made vessel lamp Double kohl tube Bracelet Bead 600 128 319 270 325 248 225 600 600 270 175 227 226 271 257 277 600 277 183 6001/12 1100/1 3061/4 2237/2 3081/1 2154/5 2117/1 6001/8 6001/11 2237/1 1254/3 2115/4 2112/1 2290/6 2193/4 2279/2 6001/6 2279/1 1295/1 257.40-258.00 257.60 256.70 256.70 256.50 256.30 257.90 257.40-258.00 257.40-258.00 256.70 253.10 257.73 257.20 255.90 256.79 256.10 257.40-258.00 256.10 257.82 147 Itamar taxel Fig. 4.2: Glass vessels. 148 Chapter 4: glaSS type 1: With a thick disc base (not illustrated), originally belonged to a deep beaker with almost vertical walls. this type is usually dated to the mid 4th-early 5th century (GorinRosen 1999:207, Fig. 1:5-6; Katsnelson 1999a: Fig. 2:1-2). type 2: With a tubular ring base and cylindrical body (not illustrated). this type is also usually dated to the early Byzantine period (Meyer 1987: Fig. 5:v-Y; Winter 1998: Fig. 2:8), but some vessels found in Karkur >illit, made of an “unusually thick glass”, were dated to the late Byzantine period (Katsnelson 2004: Fig. 61:16-17). type 3: With a concave base and slanting, slightly concave walls (Fig. 4.1:7). this type is dated, according to its parallels, to the late Byzantine-Umayyad period (Corbo 1955: Fig. 25:14; Gorin-Rosen 2000: Fig. 2:16-17; Katsnelson 2004: Fig. 62:27-28). BOttleS ten types and sub-types of glass bottle were identified at the site. Some of them appeared already in the early Byzantine period and continued into the late Byzantine or Umayyad periods, and some appeared only in the Umayyad period. However, none of the examples was found in a context earlier than the late Byzantine period. type 1a: With a cylindrical, plain neck and simple rim (Fig. 4.1:8; the illustrated example has a rather wide neck). it has parallels from the 5th century to the Umayyad period (Hadad 2005: Pl. 8:144-145; Katsnelson 1999a: Fig. 3:9-12; 2004: Fig. 62:1-6). this type, together with type 3 (below), is the most widespread type of glass bottle at the site. type 1B: Differs from the latter by its infolded rim (Fig. 4.1:9), but it has contemporary parallels (Delougaz and Haines 1960: Pl. 59:7-8, 12-13; Hadad 2005: Pls. 8:155-159, 9:166-168, 170-175; Katsnelson 2004: Fig. 62:7). type 2: Represented by bottles with a funnelshaped neck and infolded rim (Fig. 4.1:10), which have late Byzantine and Umayyad parallels (Katsnelson 1999a: Fig. 3:1-2, 4-6; 2004: Fig. 62:12-16; Meyer 1987: Fig. 9:O-S). type 3: With a cylindrical or slightly funnel-shaped narrow neck, simple rim and decoration of one or more trails on the rim and neck. the colour of the trail/s is either darker or identical to that of the bottles themselves (Fig. 4.1:11-14). this type has parallels from the early Byzantine to the Umayyad period (Delougaz and Haines 1960: Pl. 59:1-6; Hadad 2005: Pls. 12:223-234, 13:235-247; Katsnelson 2004: Fig. 63:3-7; Katsnelson and Jackson-tal 2004: Fig. 2:11; Meyer 1987: Fig. 11:G-H). type 4: large bottles with wide, almost vertical necks, plain rims and thick walls. the neck is decorated with groups of thick trails, ca. 2.5 cm below the rim (Fig. 4.1:15-16). they have parallels in 5th-7th century assemblages (Katsnelson 1999b: Fig. 9:7; Katsnelson and Jackson-tal 2004: Fig. 2:5). type 5: With a cylindrical neck, decorated with a mould-blown design of spiral ribbing, and simple rim (Fig. 4.2:4). its parallels are dated generally to the Byzantine period (Katsnelson 2004: Fig. 63:8; Katsnelson and Jackson-tal 2004: Fig. 2:13-14; Peleg and Reich 1992: Fig. 18:6). type 6: a uncommon bottle, which is also the only complete glass vessel found in the site. it has a concave base, body which is globular in its lower part and narrow and cylindrical in its upper part, sharply carinated shoulder and funnel-shaped plain neck with a narrow lower part. the narrower part of the body is decorated with a very fine vertical ribbing (Fig. 4.2:1). the Only exact published parallel to this bottle was found in a tomb dated to the 5th/6th-7th century near Netiv Ha-lamed He, 7 km south to Khirbet es-Suyyagh. the bottle, which lacks the fine ribbing of the one discussed here, was described by the excavator as an unusual type that has no parallels (Barag 1974:84, Fig. 2:7). a somewhat different bottle from the Umayyad period, with a less carinated shoulder, trail-decorated neck and 149 Itamar taxel vertical ribbing, was found in Beth Shean (Hadad 2005: Pl. 16:321). type 7: a small bottle with short cylindrical neck and thickened, infolded rim. Sometimes there are two wavy ridges on the neck (Fig. 4.2:5, 6). this form is typical to the late Byzantine and mainly the early islamic period (Hadad 2005: Pl. 9:182-184, 10, 37:738-742; lester 2004: Fig. 7.7). type 8: Represented by a single fragment (not illustrated) which has a long, narrow neck decorated with a single thick, wavy trail. this type appeared in the late Byzantine period but flourished mainly in the Umayyad period (Hadad 2005: Pls. 13:267, 14:270-277). type 9: another small bottle, characterized by its narrow cylindrical, ridged neck (Fig. 4.2:7). it is typical to the Umayyad and >abbasid periods (Hadad 2005: Pls. 11:204, 38:762-779; Gorin-Rosen 1999: Fig. 1:21; lester 2004: Fig. 7.8:95). JUGS two types of glass jug were found, both probably late Byzantine in date. type 1: With a narrow cylindrical rim which terminates in a funnel-shaped upper part, infolded rim and loop handle from rim to shoulder. the upper end of the handle, below the rim, is folded downwards (Fig. 4.2:8, 9). these jugs first appeared in the 4th century and continued until the 7th century (Katsnelson 1999a:73*). in Umayyad assemblages, however, like in Beth Shean, they are very rare and their revival is seen only in the >abbasid period (Hadad 2005:28). all the fragments of these jugs found in the site originated in late Byzantine-early Umayyad contexts (for contemporary parallels, see: Hadad 2005: Pl. 20:382; Katsnelson 1999a: Fig. 3:4-7; 2004: Fig. 62:12-15, 17; Sussman 1976: Fig. 4:1-2). type 2: Represented by a concave foot base, from which starts a globular (?) body. the base is decorated with a mould-blown design of vertical ribbing (Fig. 4.2:10). No published 150 parallels were found to this jug base, but an almost identical base with a spiral ribbing was found at Kafr >ana (taxel 2007a: Fig. 4.1:9). in both cases the jugs were found together with late Byzantine-Umayyad pottery. JaRS two types of glass jar were found in the site. according to their context and parallels both are probably dated to the Umayyad period. type 1: With a short funnel-shaped neck and thickened infolded rim (Fig. 4.2:11). it has parallels in Umayyad assemblages (GorinRosen 2000: Fig. 2:21; Hadad 2005: Pl. 20:370-377; Meyer 1987: Fig. 12:H). type 2: With a wide cylindrical neck and thickened infolded rim (Fig. 4.2:12). it has a single published parallel, dated generally to the early islamic period (lester 1996: Fig. Xvii.1:2). DROPPeR this unique vessel has a relatively wide, slightly bulging neck and a very small opening (1 cm in diameter) between neck and body (Fig. 4.2:13). these vessels, which appeared in the late Byzantine period and continued into the Umayyad period, have only a few published parallels (Barag 1985: Fig. 8:6-7; Hadad 2005: Pl. 20:379-381; tushingham 1972: Fig. 28:41). MOUlD-MaDe veSSelS two base fragments of mould-made vessels (bowls?) were found in the site. these are flat bases, thickened in their centre and decorated in a honeycomb concentric pattern (Fig. 4.2:14, 15). this unique mould decoration pattern is known from all over the Muslim world, between the 8th-11th centuries (Hadad 2005:36-37), and contemporary parallels were found also in Palestine (ibid.: Pls. 4:86, 30:599-600, 31:601; lester 1996: Fig. Xvii.9:3; 2004: Fig. 7.13:162-163). One of the discussed fragments was found in an assemblage dated to the Umayyad period, and the other was found in a mixed context which contained also >abbasid pottery. Chapter 4: glaSS laMPS two types of glass lamp were retrieved. Both are typical of the late Byzantine and Umayyad periods. type 1: the more common of the two types is bowl shaped with an outfolded rim, pushed-in base, vertical inner tube and three suspension handles projecting from the rim (Fig. 4.2:2). these lamps appeared in the 4th or 5th century, and continue to exist until the medieval period (Barag 1970b:184-185; Hadad 1998:64, 68, type 1). Most, if not all, the examples of this lamp from the site were found in late ByzantineUmayyad contexts (Hadad 1998: Fig. 1:1-7; 2003: Pl. ii.5:1-5; Patrich 1988a: Pl. 12). type 2: this is a stemmed lamp characterised by a conical hollow stem, sometimes with a solid base (Fig. 4.2:16). these lamps have simple rims, similar to some found in the excavations; however none can be safely attributed to oil lamps. Stemmed lamps were usually put into a bronze chandelier (polycandela) or a short hooked rod which was attached to the wall. this type of lamp probably appeared in the late Byzantine period, and continued into the mediaeval period (Barag 1970b:182-183). according to Hadad, the variant represented here did not continue to exist after the Umayyad period (1998:69, 72, type 4, Fig. 4), as can be seen indeed through other parallels (Hadad 2003: Pl. ii.5:6-15; Patrich 1988a: Pl. 13:14-18; Peleg and Reich 1992: Fig. 20). according to O’Hea (2007:239-244), glass oil lamps appeared in the 4th century as Christian liturgical vessels, although this lighting method soon spread to bathhouses as well as private homes. indeed, at Khirbet es-Suyyagh glass lamps were found in various architectural contexts, and not only in relation to the church. DOUBle KOHl tUBe One fragment (base) of a double kohl tube was found (Fig. 4.2:17). these vessels, which were used for mixing cosmetics, were dated to the 4th7th centuries (Barag 1970b:175). the example from Khirbet es-Suyyagh was found in a late Byzantine-early Umayyad context (for 5th-6th/7th centuries parallels, see: Delougaz and Haines 1960: Pl. 50:11-12; Sussman 1976: Pl. 28:15). WiNDOW PaNeS One of the most interesting and important types of glass artefact found at the site are window panes. these panes, represented only by nine secure fragments, belong to the round type with a folded rim (the so-called 'bulls-eye' pane). the pane is usually rather thick and its rim is wide (Fig. 4.2:3), but in one case the body is thin and the rim is narrower and more delicate. Window panes are usually found in relation to Christian religious buildings, but also in private and secular public structures, such as bathhouses (O’Hea 2007:236-239). Most of the pane fragments from Khirbet es-Suyyagh were found in the vicinity of the church, and probably belonged to this building. the great majority of parallels to these panes are from the late Byzantine and Umayyad periods (Gorin-Rosen 2000: Fig. 3:40; Hadad 2005: Pl. 24:476-478; Meyer 1987: Figs. 7:bb, 11:O-U; taxel 2005: Fig. 54:2-8). BRaCelet aND BeaD Glass jewellery is rare at the site; only one fragment of a glass bracelet and one glass bead were found. the bracelet is made of black opaque glass, and has a semi-circular cross-section (Fig. 4.2:18). according to Spaer, this type of bracelet did not precede the Byzantine period, and continued also later (1988:54, type a2b; for contemporary parallels, see: agady et al. 2002: Fig. 24:44; Hadad 2005: Pl. 23:458-459; Katsnelson 2002: Fig. 1:2). the bead belongs to the 'segmented silver-glass' type. it is made of two layers of glass with silver leaf between them, which were shaped as three globular sections attached together (Fig. 4.2:19). according to Spaer, this technique of bead making originated in egypt, although we cannot rule out the possibility of their production also out of egypt. these beads were common in the Hellenistic through Byzantine periods (2001:130-134), although most of the published parallels from Palestine are from the late Roman and Byzantine periods (Hizmi 1997: Fig. 5; Katsnelson 2002: Fig. 1:8; Spaer 2001: Fig. 58). 151 CHaPteR 5 StONe, BONe aND Metal OBJeCtS itamar taxel StONe a small but varied number of stone objects (mostly in fragmentary condition) was found. the majority can be dated to the Byzantine and early islamic periods, although at least one is dated to the early Roman period. CRUSHING ANd GRINdING IMPlEMENTS Most of the stone artefacts found at the site were associated with grain crushing and grinding, which is not surprising considering its agricultural nature throughout its history. the simplest implements are three oval limestone pestles or grinding stones (Fig. 5.1:1-3), which originated in late Byzantine-Umayyad contexts (cf. Greenhut 2004: Fig. 15:1-2). Other crushing vessels are represented by fragments of two three-footed round basalt mortars (Fig. 5.1:4). according to ayalon, these highly common mortars were used more for the grinding of spices, medicinal and make-up materials, dyes etc., than for crushing grains (2004:269, and cf. ibid.: Figs. 1:9, 2:1; Shalem 2002: Fig. 17:1). another crushing device is a roughly rectangular mortar (0.23×0.19×0.16 m), carved from a hard limestone, with a round depression (0.11 m in diameter, 6 cm depth) in its middle (Fig. 5.1:5). Four types of mill were found in the site. the simplest is the saddle quern, represented by two fragments of flat and slightly concave lower stones made of basalt and a complete convex upper stone made of hard limestone (Fig. 5.1:6, 7). these mills, which were the only type used in our region until the Persian period (Frankel 2003a:44), continued in use also during the Roman and Byzantine periods, as indicated by parallels (Kol-Yaakov 2000: Pls. 2:4-5, 5:9-11; Meyers, Strange and Meyeres 1981: Pl. 9.24:10) and by the present finds. 152 the second type of mill is the so-called 'Olynthus mill' which is characterized by a large, flat lower stone and upper frame-like stone made of basalt, with a central depression and grooved lower face (Fig. 5.1:8). these mills, which originated in Greece or anatolia of the Classical period, were the commonest type in Roman Palestine, although they continued to be in use during the Byzantine period (Frankel 2003a:45-46; 2003b). the fragmentary examples from Khirbet es-Suyyagh were found in assemblages dated to the early Umayyad period, although they are probably earlier (for contemporary parallels, see ayalon 2004: Fig. 4:6, 7; Delougaz and Haines 1960: Pl. 49:5). the third type is represented by a single fragment of a basalt rotary hand quern, probably the upper stone (estimated diameter 0.3 m, 6 cm thickness; Fig. 5.2:1). according to Frankel, this implement was brought to Palestine by the Roman legions, probably around the 1st century Ce, but became the most widespread type of mill only in the Byzantine period (2003a:46; for contemporary parallels, see: Fischer and tal 1999c: Fig. 10.1:16; Gichon 1993: taf. 59:4; Harper 1995: Fig. 26:1). to the fourth type belongs a fragment of a lower stone of a Pompeian or hour-glass mill, shaped like a short cone (0.14 m height; Fig. 5.2:2). this type was made up of two parts, the upper stone being in the form of two hollow cones. it originated in the western Mediterranean and probably appeared in Palestine already in the Hellenistic period (Frankel 2003a:47). Nevertheless, the archaeological parallels show that it was most frequent during the late Roman and Byzantine periods (ayalon 2004: Fig. 3:17; Delougaz and Heines 1960: Pl. 49:6; Neidinger, Matthews and ayalon 1994: Fig. 11). the discussed fragment was found in topsoil containing pottery dated to the 6th-8th centuries. Chapter 5: Stone, bone and metal obJeCtS Fig. 5.1. Stone objects: 1) l128*; 2) l166; 3) l101; 4) l241; 5) l262; 6-7) l175; 8) l208. 153 Fig. 5.2. Stone objects: 1) excavated by the iaa; 2) l226; 3) l200; 4) l121*; 5) l518; 6) l153; 7) l248. 154 Chapter 5: Stone, bone and metal obJeCtS ROllERS two stone rollers were excavated. a fragmentary roller (0.21 m preserved length, 0.25 m in diameter) is made of hard dolomite limestone, and has a cylindrical depression at one end (4.5 cm depth and diameter) which was used for attaching a wooden handle (Fig. 5.2:3). thick, heavy stone rollers (latin: cylindrus; Greek:   ) were used for levelling beaten earth roofs and floors, and were in use for many periods, up to the present (Hirschfeld 1995:123, 244, Fig. 178; White 1975:3-5). this roller was most probably used in the monastery, although its context (the area of the large paved courtyard) indicates that it was reused in some way during the post-monastery phase. Contemporary (i.e., late Byzantine and Umayyad) parallels for limestone rollers have been found mainly in private domestic contexts (Greenhut 1998: Fig. 6; Hirschfeld 2000a: Fig. 73; Strus 2003: Fig. 5.30), but also in the monastery of Martyrius in the Judaean desert (Magen 1993:175). the second roller was made of a round narrow marble column-like object, probably a leg of an altar/offering table taken from the monastery’s church in secondary use and thus dating to the post-monastery phase (Chapter 6). this roller is much smaller than the limestone roller (0.14 m length, 9 cm in diameter), and has two unequal depressions (1.5 cm depth, 6.5 cm in diameter; 1 cm depth, 2.1×1.3 cm in diameter) at either end (Fig. 5.2:4). the small size and relative lightness of this roller eliminate the possibility that it was used as roof/floor roller. Rather, it is much more likely that this small roller was used as a crushing implement, probably for crushing olives. the use of stone rollers, either small or large, for crushing olives was common in Palestine at least since the Bronze age and until the early 20th century, as indicated by archaeological evidence (mainly from the iron age ii) and ethnographic observations (Warnock 2007:34-36, Figs. 3.27-3.30). Olive crushing with a roller could have been done either on a flat surface or in a stone basin (ibid.:75-78, Figs. 4.6-4.11). No parallels for ancient small rollers made of re-used architectural elements have been published so far, though large rollers made of re-used column drums are documented ethnographically (ibid.:34, Fig. 3.27). BOWlS three different types of stone bowl were recognised. the first type, dated to the early Roman period, is a chalk lathe-turned bowl. it is hemispherical in shape and highly smoothed, and has an incised line below the pointed rim (Fig. 5.2:5). Such bowls were found in chalk-vessel workshops at îizma (Magen 2002a: Figs. 2.11-2.13), Mount Scopus (amit, Seligman and Zilberbod 2008: Fig. 20.13:5-8), northeast of Jerusalem and in various places in and around Jerusalem such as the temple Mount (ibid.: Figs. 3.13:1-3, 3.14) and the City of David (Cahill 1992: Fig. 16:2-6, Ph. 132). according to Jewish Halakha, stone vessels do not acquire impurity, compared to vessels made of other materials (Magen 2002a:144). thus, stone vessels became an important characteristic of the Jewish culture of Jerusalem and Judaea in the early Roman period. the second bowl is a square or rectangular limestone vessel with vertical walls, flat rim and a long wide horizontal handle with a triangular end. traces of an incised decoration can be seen on one of the walls, and in one corner there is a round hole in the rim (Fig. 5.2:6). No exact parallel was found for this bowl. Fragments of two bowls made of soft, greenishblack steatite belong to the third type. they have a flat base, vertical walls and a pointed rim (missing from our examples). they are decorated with incisions of diagonal lines and concentric circles on the outside (Fig. 5.2:7). these bowls were imported from the arabian Peninsula in the 8th9th centuries (Magness 1994:204; see also Harrell and Brown 2008:41-42, 63), and are quite common in late Umayyad and early >abbasid assemblages in Palestine (e.g. alliata 1990: Nos. 42-43; arnon 2008:62-63, types 116a-e; Saller 1941: Pl. 133.1:1-2). Both of the bowls found came from loci dated, on the basis of finds or stratigraphy, to the early >abbasid period. 155 Itamar taxel BASIN among the finds unearthed below the ashlar floor of the large courtyard were also two large fragments of a limestone basin. this is round and rather shallow (1.2 m in diameter, 0.4 m height, 0.18 m depth), and has vertical walls and a central round depression (Fig. 2.27). the only possible parallel to this basin was published from îorvat îermeshit, where a similar basin was found in an oil press (Greenhut 1998: Fig. 19). according to Greenhut, this basin was probably part of an oil pressing installation. He suggested that the stone basin carried a perforated wooden plate on which crushed olives were placed. the olives were pressed by stones and the oil drained through the perforated plate into the depression in the basin’s base (ibid.:133). STOPPERS two limestone stoppers were found. the first was made of hard limestone, roughly carved to a round shape with a flat upper face and a conical lower face (10 cm in diameter, 6 cm thickness; Fig. 5.3:1). the second stopper was made of soft limestone, carved into an oval shape with two flat faces (7.2×6.5 cm in diameter, 3 cm thickness; Fig. 5.3:2). the first stopper was probably used for sealing a storage jar, as indicated by its diameter which fits the average diameter of the local bag-shaped jars and by the fact that it was found in one of the storerooms, together with jars and ceramic lids. the second and smaller stopper was perhaps used to seal a jug. Not many parallels for stone stoppers have been published (Greenhut 2004: Fig. 14:1; Patrich 1988b: Pl. 11:67; taxel 2005: Figs. 56:4-5, 57:4), perhaps suggesting that they were not commonly used. WEIGHTS Four weights were made of small pieces of unworked limestone with a round hole drilled into one of the edges (Fig. 5.3:3-6). these simple weights were found mainly in late Byzantine-Umayyad contexts, and have no specific parallels. two other weights are smaller and more elaborate. One (Fig. 5.3:10) is disc-shaped and made 156 of limestone (5 cm in diameter, 1.5 cm thickness, 1 cm diameter hole). there are Byzantine and early islamic parallels (Birger 1981: Pl. 11:15; Shalem 2002: Fig. 17:8-9). the other (Fig. 5.3:11) is a convex spindle whorl made of polished basalt (2.3 cm in diameter, 0.8 cm thickness, 1 cm diameter hole). Such spindle whorls were in use since at least the iron age ii, and this example was found in a late Byzantine context (Coen Uzzielli 1997: Pl. 1:12; Mazar 2003b: Pl. i.10:8; taxel 2005: Fig. 56:3). WHETSTONE an object that can be identified as a whetstone, made of hard greenish slate, was found. it has an irregular shape, worn and scratched from use (Fig. 5.3:12; for late Byzantine/early islamic parallels see tal and taxel 2008: Fig. 6.130:2-4 with additional references). MISCEllANEOUS the exact nature of three stone objects could not be determined. the first is a roughly carved cylindrical limestone object (0.13 m height, 10.5 cm wide diameter, 7.5 cm narrow diameter), with a shallow round depression at each end (Fig. 5.3:7). a similar object, although twice the size, was found at another site, where it was dated to the 2nd century Ce and identified as a possible door socket (Glick 2006: Fig. 13:1). the second object (Fig. 5.3:8), carved from soft limestone, is rounded and flat (0.28 m in diameter, 0.12 m thickness) with a circular hole (0.11 m in diameter) in the centre. Similar object, also identified as part of a door socket, was found at Raqit (ayalon 2004: Fig. 3:11). the third object (Fig. 5.3:9) is a small irregular lump of reddish-brown scoria (11.5 cm length, 6.5 cm maximal width, 4 cm maximal thickness). the use of this material, which was brought to the site from afar, is unknown. it may have been used for abrading other objects, maybe even in body hygiene (for parallels of similar early islamic scoria objects, see tal and taxel 2008: Fig. 6.130:1; taxel 2007b: Fig. 6.1:7). Fig. 5.3: Stone objects: 1) l148; 2) l329; 3) l185; 4) l150; 5) l271; 6) l512; 7); l287; 8) l241; 9) l116; 10: l101; 11) l114*; 12) l128*. BONe Metal Only one artefact made of bone was found in the site (Fig. 5.4:1). this is a rectangular plaque (0.1 m preserved length, 3 cm preserved width, 0.7 cm maximal thickness), probably cut from a bovine tibia. it is decorated with curved longitudinal stripes of unequal width. Similar plaques were found at Caesarea, where they were dated to the early Roman to late Byzantine periods (ayalon 2008: Figs. 37:351-352, 41:385, 42:390). the plaques from Caesarea were identified as inlays or carved pieces. Some of the grooved plaques, which originally had a semi-circular section, were probably intended to imitate columns of the Doric order (ibid.:93). Metal objects, made of iron, copper/bronze and lead were found in almost every part of the site, mostly in fragmentary and/or corroded condition. Most of them were found in relation to finds from the late Byzantine and Umayyad periods, a few can be dated to the late Ottoman period and others are undatable. IRON naIlS the forty five fragmentary or complete iron nails found all over the excavated area are of two types. this is true also of an additional unknown number of nails found in the iaa excavations. 157 Itamar taxel type 1 (Fig. 5.4:2-7), which represents the great majority of the nails, includes medium-sized and large nails with a round flat or slightly convex head and square cross-section shaft (5-10 cm length, 0.5-1 cm thickness, 1.4-3 cm diameter of head). these nails were most probably used in various construction activities, such as nailing wooden roof and door beams. type 2 (Fig. 5.4:8, 9) includes a few much smaller and thinner nails with rounded or oval convex head and square cross-section shaft (2-2.5 cm length, 0.3-0.4 cm thickness, 1.3-2 cm diameter of head). these nails were probably used for nailing wooden boxes or other small objects. rIng Half of an iron ring (5 cm in diameter, 0.7 cm thick) was found in the site (Fig. 5.4:10). Such rings were probably used for holding ropes, bells or chandeliers (cf. Goldsmith, Ben-Dov and Kertesz 1999: Fig. 14.15:6; Harper 1995: Fig. 24:5; Nikolsky et al. 2004: Fig. 55:6). hooKed obJeCt this surface find is a narrow rod with a circular cross-section (0.8 cm thickness) terminating in a hook (Fig. 5.4:11). tool handle (?) the same mixed locus is which the hooked rod was found yielded another un-clear iron object. this is a straight rod with a circular cross-section (0.6 cm thickness) terminating in a flaring flat end (1 cm thickness) with a semi-circular depression (Fig. 5.4:12) came from the same surface locus as the hooked object described above. its function is unknown but it could be the handle of some tool. a similar, although undated parallel was found in Jerusalem (tushingham 1985: Fig. 70:2). VeSSel handle an interesting iron object is a large flat ribbon handle (12 cm length; 1.2 cm width, 0.5 cm thickness), which was probably rectangular (Fig. 5.4:13). its turnedout ends were somewhat wider than the body. the ends were pierced and attached to the vessel’s body 158 with rivets, one of which was partly preserved. an identical lid handle, although of bronze, was found in a Byzantine context at Sardis (Waldbaum 1983: Pl. 36:548). two other handles, one of bronze and one of iron, which are more rounded than ours, were found in Yassi ada (late Byzantine period; Womer Katzev 1982: Fig. 12-4:10) and Meiron (late Roman period; Meyers, Strange and Meyeres 1981: Pl. 9.4:11). the shape of the handle indicates that it was part of a lid or bucket. bIdent hoe Worthy of note is a two-pronged hoe (Fig. 5.5, left) which was found in a highly corroded condition. it was identified and restored for photographing thanks to its parallels. the hoe (approximate length and width 0.20 and 0.12 cm, respectively) has two long narrow prongs and a round socket between them into which the wooden haft was affixed. this kind of agricultural tool (latin: bidens; Greek: ) was amongst the most common manual implements in the Roman and Byzantine world. it is known both from contemporary written sources (White 1967:47-52) and archaeology. the latter can be divided into iconographic evidences and physical finds. in at least two Byzantine mosaic floors – one from Constantinople (ibid.: Pl. 3; Fig. 5.5, right) and another from a monastery at Beth Shean (Fitzgerald 1939: Pl. 8:1) – working men holding or working with a bident hoe are depicted. Similar objects from Palestine were published so far only from three northern sites – even Menaúem (late Roman period: Katz, Kahane and Broshi 1968: Pl. 113), Shiqmona (late Byzantine period: elgavish 1994: Fig. 17) and îorvat Ovesh (late Byzantine period: aviam and Getzov 1998: Fig. 16:5). the implement from Khirbet es-Suyyagh was found in an Umayyad period context and provides evidence for the agricultural nature of the site’s population in Phase iii. horSeShoe One of the few finds dated to the late Ottoman period and found in the topsoil layer is an iron horseshoe (Fig. 5.4:14). this is a thin oval horseshoe of the levantine type (9.5×6.5 cm in diameter; 2 Chapter 5: Stone, bone and metal obJeCtS Fig. 5.4: Bone and metal objects: 1) l100; 2) l325; 3) l253; 4) l210; 5) l318; 6) l375;7) l236; 8) l157; 9) l245; 10) l354; 11-12) l100; 13) l260; 14) Surface find; 15) l303; 16) l213; 17) l240; 18) l262. 159 Itamar taxel Fig. 5.5: iron bident hoe (l286) (left) and a Byzantine mosaic from Constantinople depicting a man use a bident hoe (after White 1967: Pl. 3). mm thickness). it designed to cover the whole foot, except for a round hole in its centre, on either side of which are two small rectangular holes for nails. this type of horseshoe, which is clearly different from the western crescent-like horseshoe, is known from Palestinian ethnography of the 19th-20th century (avitsur 1976: Fig. 397) and archaeological contexts (Boas 2000: Pl. 5). COPPER/BRONZE weIghtS two small bronze weights were found in the excavations. the first weight (Fig. 5.4:17) has a flat oval shape (0.9×1.82 cm in diameter, 1-1.5 mm thickness), and weighs 1.43 gr. the Greek letter H is incised on one face. this weight originates in a context dated to the 7th century. Parallels of small, usually square, bronze weights marked with the letter H, weighing more or less the same as the one discussed here, were found in other late Roman to Umayyad sites such as Beth Shean (1.36 gr; Khamis 1998: Pl. 2:11), îorvat Raqit (2 gr; ayalon 2004: Fig. 6:14), legio 160 (1.9 gr and 1.85 gr; tepper 2003: Fig. 19:101-102) and Sardis (1.4 gr; Waldbaum 1983: Pl. 29:466). the letter H probably symbolises the value of 8 siliquae (1 siliqua=0.189 gr), or ⅓ nomisma (Khamis 1998:21; Kisch 1965:153, table 7). according to Waldbaum, the weights from Sardis do not fit the Roman weight system although their marks continue the Roman tradition (1983:84). Sometimes, however, as can be seen in the case of the present weight and its parallels, Greek letters on Byzantine weights do represent their true weight. the second weight (Fig. 5.4:18) is square and flat with a trapezoidal cross-section (1.3×1.3 cm, 2 mm thickness). it is unmarked and weighs 5.6 gr. this weight corresponded to almost 5 scripulum /  or almost 10 obolus (Kisch 1965:220, table 11). the parallels to this weight are either of lower or higher values. a late Byzantine weight from Shiqmona, made of lead and coated with bronze, weighs 4.53 gr, which corresponds to 1 nomisma (elgavish 1994: Fig. 108, bottom left). an Umayyad bronze weight from Beth Shean weighs 7.62 gr and is marked with the letters NB which symbolise 2 nomisma (Khamis 1998: Pl. 2: 6). Chapter 5: Stone, bone and metal obJeCtS Bronze weights of small denominations (1 nomisma or less) were very common in the Byzantine period, mainly for weighing coins. they were used for the equal-arms balance, and many times marked with Greek letters or other signs, like crosses (agady et al. 2002:495-496; Qedar 2001:24*). the Byzantine weighing system continued with minor changes after the Muslim conquest until the reform of the caliph >abd alMalik in the late 7th century (Khamis 1998:31). the presence of these bronze weights, and another balance lead weight (see below) at Khirbet esSuyyagh indicate that some kind of commercial or industrial activity, which demanded weights, took place at the site during the late Byzantine-early Umayyad period. r Ing a complete copper/bronze ring (Fig. 5.4:15) was found (3.3 cm in diameter, 2.5 mm thickness). the use of this ring was probably similar to that of the iron ring discussed above (cf. Gichon 1993: taf. 60:1; Nikolsky et al. 2004: Fig. 53:9-10; Waldbaum 1983: Pl. 49:857-866). bell Part of a copper/bronze bell (handle and body) (Fig. 5.4:16) was also found. inside the body, remains of the iron tongue were preserved. this bell belongs to the large conical-flattened type which was probably tied to the necks of cattle or sheep (cf. ayalon 2004: Fig. 13:4; Dar 1999: Fig. 27:5). bowl the only copper/bronze vessel belongs to a bowl with straight everted walls and a plain rim (Fig. 5.6:1). ChaIn a short section and two separate links of a bronze chain (total length 14.5 cm; length and diameter of each link 2.8 cm and 1 cm, respectively) were unearthed. the links are of the figure-eight type, and were made by bending short bars with a flat section (Fig. 5.6:2). Chains with similar or slightly different links are known from other Byzantine sites (aviam 2002: Fig. 78; ayalon 2004: Fig. 10:10; Mazar 2003b: Pl. i.6:12), where they were used for hanging chandeliers or incense bowls. therefore, this chain was possibly used for hanging glass chandeliers (polycandela) in the church building. Spatulae the six bronze spatulae found can be divided into three types. type 1 includes three different spatulae, which are characterized by a plain shaft and leaf-shaped or rounded spoon. the shaft’s base is decorated with a simple incised geometric design. the first spatula (Fig. 5.6:3) has a shaft with a square cross-section and flat leaf-shaped broken spoon (preserved length 8 cm, thickness of shaft 3 mm, width of spoon 1 cm). the shaft’s base is decorated with an incised zigzag line. Similar spatulae were published from late Roman, Byzantine (andersen 1985: Pl. 15:288; taha 2003: Fig. 11:10; tushingham 1985: Fig. 71:13) and even late Mamluk/Ottoman (Khamis 199: Fig. Xviii.6:1) contexts, although the present example was found in a clear late Byzantine context. the second spatula (Fig. 5.6:4) has a shaft with square cross-section and a concave rounded spoon (preserved length 4 cm, thickness of shaft 3 mm, diameter of spoon 7 mm). the shaft’s base is decorated with incised diagonal lines and the spoon is decorated on the inside with an impressed floral pattern. Somewhat different spatulae, with undecorated spoons, were published from Byzantine (Goldsmith, Ben-Dov and Kertesz 1999: Fig. 14.6:4) and early islamic (Shalem 2002: Fig. 20:1) contexts, and others, with an impressed shell pattern, were published from Byzantine Sardis (Waldbaum 1983: Pl. 41:631-634). the third spatula (Fig. 5.6:5) is represented only by its circular cross-sectioned shaft (preserved length 9 cm, diameter of shaft 3-4 mm). type 2, represented by two spatulae (Fig. 5.6:6, 7) is characterized by a plain shaft with a twisted end, circular or oval cross-section and rounded, concave spoon (length 12.3 cm, diameter of shaft 2-3 mm, diameter of spoon 5 mm and preserved length 12 cm, diameter of shaft 3×4 mm, respectively). Parallels for such spatulae were 161 Fig. 5.6: Metal objects: 1) l214; 2-3) l240; 4) l302; 5) l301; 6) l256; 7) l204; 8) l240; 9) l368; 10) l200; 11) l162; 12) l240; 13) l322; 14) l240; 15) l317; 16) l214; 17) l205; 18) l236; 19) l348; 20) l355; 21) l100. 162 Chapter 5: Stone, bone and metal obJeCtS found in late Roman and Byzantine contexts (Goldsmith, Ben-Dov and Kertesz 1999: Fig. 14.6:1; tacher, Nagar and avshalom-Gorni 2002: Fig. 10:12; tushingham 1985: Fig. 71:14). type 3 is represented by one spatula (Fig. 5.6:8) with a mould-decorated shaft in the form of alternately square and biconical links. the square links are decorated with straight incised lines (preserved length 8 cm, thickness 3-6 mm). Similar spatulae were published from late Byzantine assemblages (Nikolsky et al. 2004: Fig. 53:16; tushingham 1972: Fig. 26:42). u nIdentIfIed fragmentS another object difficult to identify is a narrow sheet with narrower folded and stapled ends (Fig. 5.6:12). this was maybe a tack, which was used to attach two separate objects to each other. it is also possible that it was a simple button. No exact parallels were found for this object. the function of two small copper/bronze fragments, one a thin folded sheet with a circular hole in its middle (Fig. 5.6:13) and the second a narrow straight sheet with broken ends (Fig. 5.6:14), could not be determined. deCorated plaque a thin square copper/bronze plaque (3.2×3.2 cm) was found in a fill dated to the 7th century (Fig. 5.6:9). it is decorated along two of its borders with punctured dots. a similar copper plaque, decorated with a different pattern, was published from a 7th9th century site in the Negev (israel, Nahlieli and Ben Michael 1995: Fig. 7:18). lEAd buCKle One fragment of a copper/bronze object was identified as belonging to a buckle (Fig. 5.6:10). it includes parts of the straight bar and the U-shaped bar, which join together in a loop, and the buckle’s tongue that looped around the straight bar. Parallels of similar buckles were found in late Roman and Byzantine contexts from Jerusalem (Mazar 2007: Fig. 1.17:1) and asia Minor (Russell 1982: Fig. 6:7-8; Waldbaum 1983: Pl. 44:703-704). braCelet/handle One of the unidentified copper/bronze objects (Fig. 5.6:11) is almost semi-circular (5 cm in diameter) with an oval cross-section (7×8 mm) and narrower ends (4×8 mm). the ends are broken but it is possible that this object was originally a complete circle. if so, it might be a bracelet. Bracelets with a similar cross-section are known from late Roman and late Byzantine assemblages, although all of them have a uniform thickness (Stern 1997: Fig. 12:49; tacher, Nagar and avshalom-Gorni 2002: Fig. 10:30; tushingham 1972: Fig. 27:20). another possibility is that this object was part of a handle of a vessel. weIghtS two lead weights were found at the site, both in loci dated to the Umayyad period. the first weight comprised two parts (Fig. 5.6:15). the main part is a roughly trapezoid piece of lead (6.2×2.3×2.2 cm). an iron suspension loop with a square cross-section (external diameter 3.5 cm, thickness 0.8 cm) was inserted into its wide flat face. the total height of the weight, therefore, is 5.7 cm, and its weight is 266 gr, which corresponds to 9.8 ounce (uncia/; 1=27.2 gr) or 58.7 nomismae. Parallels to heavy lead weights with iron suspension loops have only been found in asia Minor. the first was found in the 7th century Byzantine shipwreck of Yassi ada. it is almost identical to the present weight, although almost 3 times heavier (933 gr; Kenneth Sams 1982: Figs. 10-4:1, 10-6). the second parallel is an undated weight from Sardis. it is more rounded and 5 times heavier (1340 gr) than the weight from Khirbet es-Suyyagh (Waldbaum 1983: Pl. 29:447). the second weight (Fig. 5.6:16) has a flat rectangular shape with rounded corners (2.5×2×0.5 cm, weight 29 gr). One of its wide faces is smoother than the other. its weight corresponds to slightly more than 1 ounce, or 6.4 nomismae. Parallels of similar, although heavier, weights were found at Beth Shean (58.36 gr and 38.55 gr; dated ByzantineUmayyad; Khamis 1998: Pl. 12:38-39) and Sardis (108.1 gr=⅓ libra; dated Roman-Byzantine; Waldbaum 1983: Pl. 30:482). this weight, like the 163 Itamar taxel two bronze weights described above, was probably also a balance weight. as can be seen, both the lead weights are much larger and heavier than the bronze weights described above. their higher basic weight made them the commonest commercial weights (for various commodities) in Byzantine Palestine (agady et al. 2002:494-496; Qedar 2001:24*), and most probably also in the early Umayyad period. of the neck (Fig. 3.15:9). the second jar (Fig. 5.6:19) is represented only by a body sherd, with a small, folded piece of lead (3.5 cm length, 4 mm width, 2 mm thickness) inserted through a hole in it. Parallels for late Byzantine and early islamic pottery vessels repaired with different kinds of lead clamps were found at other sites (Fischer and tal 1999a: Figs. 6.116, 6.129:12; Gudovitch 1999: Fig. 182:1). ClampS two lead clamps, composed of two convex strips attached to each other by two narrow, rounded pieces were found at the site (Fig. 5.6:17, 18). these clamps were used for repairing cracked pottery vessels and in building, and are quite common in late Roman and Byzantine assemblages in Palestine (Dar 1999: Fig. 18:20; Nikolsky et al. 2004: Fig. 53:4; Rahmani 1960: Pl. ii:2, upper right). the excavations also yielded fragments of two pottery jars of the late Byzantine/Umayyad period, which were repaired with somewhat different lead clamps. the first jar is represented by a neck, into which were drilled seven holes. Short lead pins were inserted into these holes and attached flat lead strips (5 cm length, 1 cm width, 3 mm thickness) to the inside m uSKet bullet One round lead bullet of a musket rifle (1.9 cm in diameter) was found in the topsoil (Fig. 5.6:21). Similar bullets have been published from Ottoman assemblages in acre and Ramat Hanadiv (Berman 1997: Fig. 12; Boas 2000: Fig. 8). 164 u nIdentIfIed fragmentS Five unidentified pieces of lead were found in loci dated to the late Byzantine-Umayyad periods. Four of these have no definable shape (Fig. 5.6:20), and they might be the raw material for making lead objects, such as the common clamps described above. the fifth piece is flat and round (12 mm in diameter, 2 mm thickness), so it may be either waste, a raw piece or a broken object. CHaPteR 6 MaRBle FURNiSHiNGS itamar taxel Numerous fragments of various marble elements were found during the excavations. the reuse of the church for domestic purposes in the post-monastery phase led to the marble furniture that once adorned it being spread in fragmentary condition all over the site. Some of the elements were reused in the new building of Phase iii, some were most probably burnt for lime making, but many fragments were found without any defined context. Most of the marble elements identified belonged to the bema of the church – chancel screens and posts, altar and/or offering tables. More portable objects are marble bowls and lids of reliquary chests. CHaNCel SCReeNS three fragments of marble chancel screens were found, all belonging to the screens’ central parts. two fragments (2.7 and 3.4 cm thickness) are decorated in relief with a wreath pattern (Fig. 6.1:1, 2). Screens decorated with a central wreath are the most common in Byzantine Palestine (e.g. Patrich 1988b: Pl. 10:46, ill. 187; Peleg 2003: Pl. i.21; Saller 1941: Pls. 124:3, 125:24, 26-28, 30). Most, if not all, chancel screens of this type have an additional motif within the wreath’s borders, and the wreath itself has two tendrils emerging from its base towards the screen’s sides. these parts, however, were not preserved in the fragments retrieved. a small part of a right-angled central motif within a wreath can be seen in the third fragment (2.3 cm thickness). this may be a cross (Fig. 6.1:3). a parallel decorated with a straight-angled cross within a simple circle was found in aliki, Greece (Sodini and Kolokotsas 1984: Fig. 77). CHaNCel POStS Six fragments of chancel posts all of the same type were found in the excavations. these posts have a square section and a moulded decoration of a rectangle surrounded by a plain frame on their front. the side slot for affixing the chancel screen can be seen only in two of the fragments (Fig. 6.1:4, 9). this is the most widespread type of chancel post in religious buildings of the Byzantine period in the levant (acconci 1994: Nos. 27, 29, 31; Saller 1941: Pl. 132:4-10; Patrich 1988b: Pl. 10:48-49, ills. 190-191; Peleg 2003: Pl. i.26:1-7). altaR/OFFeRiNG taBleS Nine or ten fragments of marble table plates and two fragment of table legs were found. the total number of tables represented is seven to ten, since two pairs of plate fragments are of identical thickness. they are divided into four types. type 1 (Fig. 6.2:1-4) is a square/rectangular table characterized by a sunken surface (2.2-3.3 cm thickness) and wide frame (4.2-5 cm thickness, 7-7.5 cm width), either plain or with a groove along its inner edges. One of the sides of the grooved fragment has a concave profile, which probably appeared also in the opposite side (Fig. 6.2:1). this was the most common type of altar/offering table in Byzantine churches in the levant (acconci 1998: Nos. 55-58; Habas 1999: Pl. 1:1; Margalit 1987: Figs. 11-12; Patrich 1988b: Pl. 10:59, ill. 200), and throughout the Byzantine world (Habas 1999:122-123, and references therein). two other fragments, one belonging to a table’s corner, have plain frames (Fig. 6.2:3, 4). this sub-type is less frequent, but parallels were found at one site at least (acconci 1994: Nos. 44-45). type 2 (Figs. 6.2:5, 6, 6.3:1, 2) is a polylobed table. One of the fragments has straight edges (Fig. 6.2:5) while those of two others are slightly rounded (Figs. 6.2:6; 6.3:1). the fourth fragment belongs to the inner side of a corner (Fig. 6.3:2). these tables have flat frames with plain profiles, and horseshoe-shaped lobes 165 Itamar taxel arranged along their inner edges. three subtypes of polylobed tables are known: square/ rectangular, round and sigma-shaped. the fragment with the straight edges seems to have belonged to a table of the first sub-type, and the two round fragments were probably parts of sigma-shaped tables. the fragment of the inner corner could theoretically belong to either of the two sub-types, but its thickness, which is identical to that of one of the round fragments, maybe indicate that it belongs to the same sigma-shaped table. Parallels to such tables, which were rather common in Byzantine churches, were found at sites in the levant (Habas 1999: Pl. 2:1; Saller 1941: Pl. 126:1-2, 4-8, 20; 1957: Pl. 79:1-2) and beyond (Sodini and Kolokotsas 1984: Fig. 178; Rodziewitz 1984: Pl. 69:342; Roux 1973: Figs. 84-87). in liturgical contexts, these tables were used as bread tables or offertory tables, but they are known also from secular and funerary contexts (Habas 1999:125, and see below). type 3 is represented by a single fragment. it has straight edges and a thick round frame (Fig. 6.3:3). it must be noted also that the fragment bears traces of fire that may indicate burning for lime production. a parallel was found in Kourion, Cyprus (loverance 1990:229, Fig. 17). the tenth fragment has a stepped profile and wide frame with two parallel grooves. in contrast to its upper face, which is well-smoothed, its lower face is roughly dressed. in addition, traces of several drilled holes can be seen in its sunken part (Fig. 6.3:4). it cannot be said whether these holes were part of the original plate, or if they were drilled later. the exact identification of this plate is unclear. its stepped profile and roughly dressed lower face can indicate that it was affixed inside another installation. For instance, it could have been used as a base for altar table, like slightly different bases found in other ecclesiastical sites in Palestine (Batz 2002: Fig. 10; Habas 1999: Pl. 1:8). However, an almost identical piece was found in Reúovot-in-the-Negev, where it was identified as an altar table’s plate (Patrich 1988b: Pl. 10:57). 166 Only two marble fragments which can securely identified as leg of altar/offerings table were found. the first is the large fragment which was reused as a roller (see Fig. 5.2:4). the diameter of this fragment (9 cm) makes it feasible to suggest that it was part of a table leg. the second fragment (Fig. 6.3:5) is a well-polished cylindrical leg (0.12 m estimated diameter) with a protruding horizontal strip which indicates that the fragment was part of the column’s base or capital (acconci 1994: Nos. 50-55; 1998: Nos. 26-29; Habas 1999: Pl. 1:6-7). it must be remembered that marble and other stone tables (usually sigma-shaped) were in use not only as liturgical objects in churches but also in monastic dining rooms (Popović 1998:299-301, Fig. 28). the best-known Palestinian examples of such tables are the sigma-shaped marble tables which were found in the refectory and kitchen of the monastery of Martyrius in the Judaean desert (Magen and talgam 1990:108-109, Fig. 22). therefore, it is possible that some of the marble tables found at Khirbet es-Suyyagh also originated in the monastery’s refectory, and not necessarily in its church. COlUMN two fragments of a round marble column (0.14-0.17 m in diameter, 0.45 m total length) were found, one of which was reused in a Phase iii wall (W209) north of the church (Fig. 2.87). the two fragments have a side slot, maybe for a screen (Fig. 6.3:6). Nevertheless, this column was probably not part of the chancel, but rather of the ciborium (the altar’s canopy), as indicated by round columns of similar diameter and sometimes also a side slot found in other levantine churches (acconci 1998: Nos. 16-19; alliata 1994: No. 5; Patrich 1988b: ill. 194-195). BOWlS two fragments of marble bowls were found. One (Fig. 6.4:1) is a small rounded bowl with a flat rim and four semi-circular handles. it is a small variant of a well-known type of marble bowl used for grinding spices and incense found in secular as Chapter 6: marble furnIShIngS Fig. 6.1: Marble objects: 1) l322; 2) l175; 3) l319; 4) l103; 5) l225; 6) l210; 7) l150; 8) surface; 9) l100. 167 Fig. 6.2: Marble objects: 1) l256; 2) Surface find; 3) l322; 4) l190; 5) l326; 6) l190. 168 Fig. 6.3: Marble objects: 1) l100; 2) l325; 3) l227; 4) l225; 5) l175; 6) left: taken from W209; Right: surface find. 169 Itamar taxel Fig. 6.4: Marble objects: 1) l293; 2) l330; 3) l235; 4) l247. 170 Chapter 6: marble furnIShIngS well as religious contexts (agady et al. 2002:505). Parallels for bowls of more or less the same size come from Byzantine assemblages in Palestine (ibid.: Fig. 26:7; Corbo 1955: tav. 28.80:3) and Sardis (Crawford 1990: Fig. 266). the second bowl (Fig. 6.4:2) is represented by a fragment of a high ring base. Parallels for bowls with similar bases were found at Jerusalem (Bagatti 1969: Fig. 7:6; Mazar 2007: Figs. 1.7:1, 1.23:2) and îorvat Karkur >illit (Monelli 2004: Fig. 17:6). ReliQUaRY liDS two marble items can be identified as flat lids of reliquary chests. Fig. 6.4:3 (12.5 cm preserved length [25 cm estimated total length], 12.5 cm width, 1.7 cm thickness) is made of white marble, and has a round hole (2 cm in diameter) in its centre. the lower face is crudely dressed and its margins are damaged so that whether they have a diagonal or vertical profile cannot be determined. Parallels for flat reliquary lids with a central hole were found in two other Palestinian churches (aviam 2002: Fig. 52; Negev 1988: Ph. 50). in both cases the reliquary chests were fixed in the bema’s floor. the other lid (Fig. 6.4:4) was made of a distinctive variegated white, red and greyish-bluish marble. this marble, known as Pavonazzetto, originates in asia Minor (Phrygia), and is known to have been used for making columns and decorative objects from the 2nd century Ce until the Byzantine period. in ancient times it was known as marmor Phrygium, marmor Synadicum (after Synada, place of origin in Phrigia) or marmor docimium (Mielsch 1985:59, taf. 18:606). the lid is rectangular, although it is somewhat narrower in one side (10 cm length, 6.3-7 cm width, 1.5-1.7 cm thickness), and a diagonal profile. No parallels for reliquary lids made of this variety of marble have been found. a flat square lid made of reddish limestone was found in church from southern Judaea (Batz 2002: Fig. 11). the latter was one of two lids of a reliquary with two separate compartments. another reliquary with a central round compartment and two lateral rectangular compartments, similar in measurement to that discussed here, was found in a church at Sussita. the lateral compartments were covered with flat lids made of grey marble (Segal et al. 2005: Fig. 101). it is possible, therefore, that the small lid from Khirbet es-Suyyagh also belonged to such a reliquary. it should be noted that similar objects, made of marble or stone, which were found at Shiqmona, were identified as writing plaques (elgavish 1994: Figs. 48, 106). However, the uncommon type of marble from which the present object was made indicates that it had special significance. 171 CHaPteR 7 COiNS a total of 50 coins was retrieved from the excavations at Khirbet es-Suyyagh. these coins represent the major periods during which the site was inhabited. Most of the coins were found in well- or relatively well-dated assemblages, while others originated in mixed or topsoil loci. the first section is dedicated to the description of the 36 coins dated to the early Roman, late Roman and Byzantine periods including also some Byzantine coins post-dated to the Muslim conquest. in addition, four more unidentifiable coins are presented. the second section describes the ten early islamic coins of the Umayyad and >abbasid periods discovered at the site. all appear on a 1:1 scale with the obverse view on the left. ROMaN aND BYZaNtiNe COiNS arieh Kindler and Zvi Gur the two earliest coins are quadrans. the first (No. 1) was issued by the Roman procurator for Iudaea under augustus, and the second (No. 2) was issued in the 2nd year of the Great Jewish Revolt against Rome (66-70 Ce), namely in 67/68 Ce. No. 1. locus 214, Basket No. 2062. date: Roman procurators under augustus; 6-14 Ce Obverse.: ear of barley upright. Border of dots. around from left below: KaiCa POC. Reverse.: Palm tree with eight leaves and two bunches of dates. Border of dots. On left below: l. Æ; 2.25 gr; 13.64×16.26 mm; axis 12 No. 2. locus 206, Basket No. 2063. date: Great Jewish Revolt; 67/68 Ce Obverse: vine leaf. around from left: [‫ל[חרת ציון‬. Reverse: large amphora. around from left: ‫ש[נת ש]תים‬. Æ; 2.66 gr; 16.16×17.25 mm; axis 3 the 3rd century Ce yielded only two coins, namely No. 3, which was struck at Caesarea Maritima during the reign of elagabalus (218-222 Ce) and the other, No. 4 a Roman imperial sesterce from Maximinus trax (235-238 Ce). No. 3. locus 226, Basket No. 2104. date: elagabalus; 218-222 Ce; Caesarea Maritima Obverse: Bust of emperor to right, laureate and draped. Reverse: Head of Serapis to right, wearing ornamented kalathos. around from left below: COl[iaFvFCCaeSa]. Æ; 8.22 gr; 21.65×22.4 mm; axis 12. 172 Chapter 7: CoInS No. 4. locus 210, Basket No. 2081. date: Maximinus trax; 235-238 Ce; Sesterce. Obverse: Bust of emperor to right, laureate and draped. Cable border. around from left below: iMP MaXiMiNvS PF avG. Reverse: Female figure standing to left. Æ; 21.4 gr; 28.53×30.02 mm; axis 12. the 4th century Ce is represented by the following emperors: the house of Constantine: a. Constantine i (307-337 Ce) is represented by three coins (Nos. 5-7). 5. locus 142, Basket No. 1159. date: Constantine i; 307-337 Ce; Constantinopole. Obverse: Bust of Constantinopole to right, helmeted and draped. around from left below: CONStaNtiNOPOli. Reverse: victory standing with spread wings, holding shield in left hand and spear in right hand. Right foot on prow of galley. in exerge: CONS. Æ; 2.35 gr; 17.4×18.34 mm; axis 12. 6. locus 250, Basket No. 2146. date: Constantine i; 307-337 Ce; Obverse: Bust of emperor to right, diamed and draped. Border of dots. around from left below: DN CONStaNtiNvS PF avG. Reverse: two legionaries standing, between them legionary standard. around from left below: GlO[Ria eXeCi]tvS. Æ; 1.48 gr; 14.39×15.17 mm; axis 12. 7. locus 220, Basket No. 2082. date: Constantine i; 307-337 Ce; arelatum. Obverse: Bust of emperor to right. around from left below: iMP CONStaNtiNvS [avG]. Reverse: two victories standing opposite, between them altar above which shield. around from left below: viCtORiae laetae PRiNC PeRP. On shield: vOt/ PR. in exergue: aRl. Æ; 2.53 gr; 17.38×17.78 mm; axis 6. b. Crispus (324-326 Ce) is represented by one coin (No. 8). 8. locus 330, Basket No. 3096 date: Crispus; 324-326 Ce; Obverse: Bust of emperor to right, diademed and draped. Border of dots. around from left below: [DN Flivl CRiSPvS NOB] CaeS. Reverse: Jupiter standing to right, holding long sceptre in right hand. Æ; 0.84 gr; 19.2 mm; axis 6. 173 arIeh KIndler and ZVI gur Chapter 7: roman and byZantIne CoInS c. Constantine ii (337-340 Ce) is represented by one coin (No. 9) of a similar type as coin No. 6 of Constantine i. 9. locus 231, Basket No. 2127. date: Constantine ii; 337-340 Ce; Obverse: Bust of emperor to right, diademed and draped. Border of dots. around from left below: CONStaNtiNvS ivN NOBC. Reverse: two legionaries standing, between them legionary standard. Border of dots. around from left below: GlORia eXeRCitvS. in exerge: SMaNt. Æ; 2.44 gr; 15.92×16.08 mm; axis 6. d. Constantius ii (337-361 Ce) appears on one coin (No. 10). 10. locus 282, Basket No. 2297 date: Constantius ii; 337-361 Ce; Pierced in centre. Obverse: Bust of emperor to right, diademed and draped. traces of letters. Reverse: legionary killing a foe on horseback. around from left below: Fel teMP [RePaRatiO]. Æ; 2.20 gr; 13.89×14.96 mm; axis 6. e. Further from the house of Constantine: this is a coin (No. 11) of the type of the two legionaries flanking a legionary standard. No specific identification was possible because of the bad state of preservation of the coin. 11. locus 100, Basket No. 1027 date: Constantinian family. Obverse: Bust of emperor to right, diademed and draped. Nearly effaced. Reverse: two legionaries standing, between them legionary standard Border of dots. [GlORia eXeRCitvS]. Æ; 0.95 gr; 13.60×13.85 mm; axis 12. two further coins of the 4th century Ce (Nos. 12, 13) on both of which a standing victory is depicted. 12. locus 148, Basket No. 1179 date: 4th century Ce. Obverse: Bust of emperor to right, diademed and draped. around from left below: …aNvS…. Reverse: victory standing to front, holding wreath aloft in right hand. around from left below: [SeCvRitaS] Rei PvB[liCae]. Æ; 1.61 gr; 15.1×16.1 mm; axis 6. 13. locus 244, Basket No. 2172 date: 4th century Ce; Æ; 1.2 gr; 1414.85 mm. Obverse: Completely effaced. Reverse: victory standing to front, holding wreath aloft in right hand. [SeCvRitaS Rei PvBliCae]. Æ; 1.2 gr; 14.85 mm. 174 Chapter 7: CoInS ten coins from the 5th century Ce were found. However, only one coin (No. 14) could be identified as being of theodosius ii (408-450 Ce). all the others, namely Nos. 15-23, were classified as belonging to the 5th century only on the basis of their shape and size, because of their bad state of preservation. 14. locus 135, Basket No. 1119 date: theodosius ii; 408-450 Ce. Obverse: Bust of emperor to right. around from left below: [DN tHeOD]OSivS PF [avG]. Reverse: Figure standing. Cable border. traces of letters. Æ; 1.05 gr; 10.96 mm; axis 12. 15. locus 142, Basket No. 1142 date: 5th century Ce. Obverse: Bust of emperor to right. Reverse: Figure standing. Æ; 1.87 gr; 13.93×13.99 mm. 16. locus 116, Basket No. 1082 date: 5th century Ce. Obverse: Bust of emperor to right, diademed and draped. Reverse: Figure standing (?). Æ; 1.72 gr; 14.94×15.91 mm. 17. locus 261, Basket No. 2203 date: 5th century Ce. Æ; 1.52 gr; 14.5×14.62 mm; completely effaced. 18. locus 249, Basket No. 2252 date: 5th century Ce. Æ; 1.28 gr; 11.52×12.51 mm; completely effaced. 19. locus 240, Basket No. 2145 date: 5th century Ce. Obverse: Bust of emperor to right. Reverse: Figure standing. Æ; 1.09 gr; 13.55×14.68 mm; axis 12. 20. locus 322, Basket No. 3115 date: 5th century Ce. Obverse: Bust of emperor to right, diademed and draped. Border of dots. Reverse: vOt XX Mvlt XXX Æ; 0.94 gr; 12.68×15.11 mm; axis 6. 21. locus 134, Basket No. 1143 date: 5th century Ce. Æ; 0.93 gr; 9.92×10.04 mm; completely effaced. 175 arIeh KIndler and ZVI gur Chapter 7: roman and byZantIne CoInS 22. locus 227, Basket No. 2099 date: 5th century Ce. Obverse: Bust of emperor to right. Reverse: Completely effaced. Æ; 0.92 gr; 11.77×11.86 mm. 23. locus 206, Basket No. 2019 date: 5th century Ce. Æ; 0.52 gr; 12.41 mm; completely effaced. in addition 15 Byzantine coins were discovered. they range from 491 Ce with the accession to the throne of anastasius i and represent all the Byzantine emperors of the 6th and up to the second half of the 7th century Ce ending with Constans ii in 668 Ce. all these coins are of either follis (M) or half-follis (K) denomination. a. anastasius i (491-518 Ce) is represented by one coin (No. 24). No. 25 also seems to belong to this emperor but this is uncertain. 24. locus 148, Basket No. 1160 date: anastasius i; 491-518 Ce; Follis. Obverse: Bust of emperor to right. Cable border. around from left below: [DN aNaStaSivS P]. Reverse: large M, above which is small cross. in field left and right – star. Cable border. Between legs of M: Δ. Æ; 10.81 gr; 26.29×28.5 mm; axis 6. Comparisons: Wroth 1908:4, No. 27, Pl. 1:9. 25. locus 201, Basket No. 2002 date: anastasius i (?); 491-518 Ce (?). Obverse: Bust of emperor to right. Reverse: large M, above which is small cross. On right: long cross. Æ; 13.69 gr; 24.5×25.53 mm; axis 6. b. Justin i (518-527 Ce) is represented by two specimens, namely Nos. 26, 27. 26. locus 214, Basket No. 2056 date: Justin i; 518-527 Ce; Constantinopole; 20 Nummia. Obverse: Bust of emperor to right, diademed, wearing cuirass and paludamentum. around from left below: DN ivSti[NvS PP avG]. Reverse: large K, above which is a star. On left: large cross. Cable border. On field right: Γ. Æ; 6.68 gr; 23.8×27.6 mm; axis 12. Comparisons: Wroth 1908:15, Nos. 33-35. 176 Chapter 7: CoInS 27. locus 260, Basket No. 2205 date: Justin i (?); 518-527 Ce (?); Follis. Obverse: Bust of emperor to right. traces of letters. Reverse: large M, above which is a small cross. in field left: star. Cable border. in field right: illegible letter. Æ; 11.96 gr; 29.2×30.59 mm; axis 6. c. Justinian i (527-565 Ce). Here we have only one coin (No. 28). 28. locus 330, Basket No. 3146 Date: Justinian i; 536/537 Ce; thessalonica; 20 Nummia. Obverse: Bust of emperor to front, crowned and draped. Reverse: large K, above which a small cross. Cable border. in field left downwards: aNNO. in field right: X. in exergue: teS. Æ; 5.74 gr; 19.6 mm; axis 6. Comparisons: Wroth 1908:41, Nos. 160-162, Pl. 6:10. d. tiberius i (578-582 Ce) is represented by one coin (No. 29). 29. locus 219, Basket No. 2071 date: tiberius i; 580/581 Ce; Constantinopole; Follis. Obverse: Bust of emperor to front, crowned and draped. Cable border. Around from left below: σMTIb]CONSTAN TPPAVI[. Reverse: large cursive M, above which is a small cross. Cable border. In field left downwards: ANNO. In field right: ЧI. in exergue: CONS. Æ; 10.27 gr; 29.39×30.25 mm; axis 6. Comparisons: Wroth 1908:109, No. 33, Pl. 14:5. e. Maurice (582-602 Ce). Here we have two specimens, Nos 30, 31. 30. locus 259, Basket No. 2219 date: Maurice; 582-602 Ce; antioch; 20 Nummia. Obverse: Bust of emperor to front, crowned and draped, holding globe surmounted by cross in right hand. Cable border. Around from left below: ]σNMAɥCNPaɥ]. Reverse: large K, above which is a small cross. Cable border. in field left downwards: aNNO. in field right: ii. Between legs of K: B (=second oficina). Æ; 4.77 gr; 22.16×24.02 mm; axis 12. Comparisons: Wroth 1908:146, Nos. 199-207, Pl. 18:8. 177 arIeh KIndler and ZVI gur Chapter 7: roman and byZantIne CoInS 31. locus 213, Basket No. 2083 date: Maurice; 589/590 Ce; 20 Nummia. Obverse: Bust of emperor to front, crowned and draped. Cable border. around from left below: [ v] PPavG. Reverse: large K, above which is a cross. Cable border. in field right: ЧII. Æ; 3.52 gr; 22.24×23.25 mm; axis 6. Comparisons: Wroth 1908:135, No. 83. f. Heraclius (610-641 Ce). three coins of this emperor were discovered (Nos. 32, 33, 34). 32. locus 359, Basket No. 3190 date: Heraclius; 610-641 Ce; Constantinopole; 40 Nummia. Obverse: in field left: cross and globe. illegible legend. Reverse: large M. illegible legend. Æ; 3.17 gr; 25.18×27 mm; axis 12. Comparisons: Wroth 1908:208, No. 18. 33. locus 278, Basket No. 2282; date: Heraclius; 610-641 Ce; 20 Nummia. Obverse: emperor standing to front, wearing long robe. traces of letters. Reverse: large K, above which is a small cross. On left downwards: aNN[O]. On right: …X…. Æ; 5.38 gr; 23.36×26 mm; axis 6. 34. locus 387, Basket No. 3269 date: Heraclius; 629/630 Ce; Half Follis. Obverse: Heraclius and Heraclonas standing to front. Heraclonas holding globe surmounted by cross in left hand. Cable border. Reverse: large K. in field left: aNN. in field right: XX. Æ; 3.86 gr; 21.52×22.16 mm; axis 12. g. Constans ii (641-668 Ce) is represented by two secure coins (Nos. 35, 36). 35. locus 359, Basket No. 3191 date: Constans ii; 641-668 Ce. Obverse: emperor standing to front, wearing long robe and holding a globe surmounted with cross in left hand. in field right: i. Reverse: large cursive M surmounted by small cross. in field left and below: [a]NNO. Æ; 3.28 gr; 19.26×21.91 mm; axis 6. 178 Chapter 7: CoInS 36. locus 281, Basket No. 2291 date: Constans ii; 641-668 Ce. Obverse: emperor standing to front, wearing crown and long robe, and holding long cross in right hand. in field left: ɥ. Reverse: large cursive M, above which is small cross. in field left: legend (?). in exergue: Ri. Æ; 3.11 gr; 18.92×21.13 mm; axis 12. Four coins (Nos. 47-50) could not be identified, as they are completely blank on both sides. 47. locus 241, Basket No. 2150 Æ; 3.58 gr; 15.83×18.05 mm; completely effaced. 48. locus 265, Basket No. 2238 Æ; 3.33 gr; 14.15×14.6 mm; completely effaced. 49. locus 137, Basket No. 1130 Æ; 2.45 gr; 21.26×21.76 mm; completely effaced. 50. locus 200, Basket No. 2013 (Fig. 7.2:7) Æ; 0.56 gr; 13.8×15.42 mm; pierced; completely effaced. 179 nItZan amItaI-preISS Chapter 7: early ISlamIC CoInS eaRlY iSlaMiC COiNS Nitzan amitai-Preiss ten coins range from the early decades of the early islamic period to the >abbasid period, ending with a coin from the days of al-Wāthiq (227 AH / 841-842 CE). Two of the coins are silver dirhams (Nos. 45, 46) while the remainder are bronze coins of various denominations (Nos. 37-44). the earliest coin is a Byzantine or an arab-Byzantine coin of the type called Pseudo-Byzantine, which is dated until ca. 668 Ce.12 37. l. 266, Basket No. 2225 date: Constans ii or arab-Byzantine imitation of its coins; 641-668 Ce. Obverse: emperor standing to front, wearing crown and long robe, holding long cross in right hand and globe surmounted by cross in left hand. Reverse: large cursive M. above, to left: aN[a]. Æ; 1.14 gr; 16.7×19.61 mm; axis 12. Comparisons: Grierson 1968:442-449, Nos. 59a-68.2. Possible dates are 641/2-647/8 Ce. However, due to this fact that the M on our coin is cursive, we assume that this is an arab-Byzantine imitation of a coin of Constans ii (cf. Goodwin 2002: Pl. 36:508). Five Umayyad post-reform coins are represented in the excavations, one of which (No. 38) was minted at al-Ramla, another one (No. 39) was minted at egypt, and another one (No. 40) was minted at Damascus. 38. l. 173, Basket No. 1255 date: Umayyad post-reform coin; 120-130 a.H. / 737-747 Ce; al-Ramla. Obverse: in the centre within a dotted circle a small tree or branch with four leaves. Outside marginal legend: ‫ ا اله اا اه‬. Reverse: Within a dotted circle a crescent, outside marginal legend: ‫ضرب هذا الفلس بالرملة‬ Æ; 2.57 gr; 15 mm; axis 12. Comparisons: ilisch 1993:14, Nos. 64-78. 12. “the production of pseudo-Byzantine coins largely ceased soon after the accession of Constantine iv in 668” (Goodwin 2002:106). 180 Chapter 7: CoInS 39. locus 100, Basket No. 1078 date: (Governor) al-Qāsim b. Ubayd Allah; 116-124 AH / 734-742 CE; Misr. Obverse: Within double beaded circles: ‫ اه وحده‬/ ‫ ا اله اا‬/ ‫بسم اه‬ at bottom a fleur-de-lis flanked by two dots. Reverse: in centre, within a small circle: ‫ اه‬/ ‫ رسول‬/ ‫محمد‬ Marginal inscription: only ‫ القاسم بن‬is seen from the full inscription that should be: ‫ امر القاسم بن عبيد اه‬. Beaded border. Æ; 4.78 gr; 17.19×19.28 mm; axis 6. Comparisons: Miles 1958:477, No. 3. 40. locus 179, Basket No. 1317 date: Umayyad post-reform coin; Dimashq. Obverse: Within three circles: ‫ا اله اا اه وحده‬ Reverse: in the centre within a circle: ‫ضرب هذا الفلس بد مشق‬. Marginal inscription: [ ‫]بسم اه محمد رسول اه‬. Æ; 1.92 gr; 15.74×16.28 mm; axis 6. Comparisons: Walker 1956:251, No. 831. 41. locus 307, Basket No. 3046 date: Umayyad post-reform coin. Obverse: ‫ [ الل] ه وحده‬/ ‫ ا اله اا‬/ ‫بسم اه‬. Reverse: in the centre there is a circle inside it a six-pointed star. Marginal inscription: ‫محمد رسول اه‬. Æ; 3.9 gr; 13.83×14.09 mm; axis 12. Comparisons: Walker 1956:222, No. 722. but here the obverse legend is around the circle with the six pointed star. 42. locus 162, Basket No. 1221 date: Umayyad post-reform coin. Obverse: traces of a circle inside which is a corrupted legend: traces of the Shahada, of which can be seen only: ‫ اا ا‬/ ‫اا‬ Reverse: a corrupted unclear legend: [ ‫ ] ر]سو[ل‬/ ‫اا‬ Æ; 2.14 gr; 10.56×10.93 mm; axis 6. Comparisons: Miles 1958:473, No.1 but here minted on a smaller flan. 181 nItZan amItaI-preISS Chapter 7: early ISlamIC CoInS Four >abbasid coins were found in the excavations. each represents a different mint. 43. locus 500, Basket No. 1326 date: Kufa type imitation; 170’s to early 180’s aH / ca. 768-800 Ce. Obverse: Within a circle: .. / ‫ ا شريك له‬/ ‫ اه وحده‬/ ‫ا اله اا‬ two circles, five ringlets, two dots below the central inscription. Reverse: Within a circle: ‫ عد ل‬/ ‫ اه‬/ ‫ رسول‬/ ‫ محمد‬/ *.* above and >adl below the central inscription. (…) ‫)…( مؤمنين سنة سبع ومئة‬. Æ; 0.84 gr; 17×19 mm; axis 4. Comparisons: Heideman 2003:148-154.13 44. locus 177, Basket No. 1280 date: al-Ma’mūn; 217 or 218 AH / 832-833 or 833-834 CE; al-Ramla, or alQuds or Gaza. Obverse: Within a double twisted cable: ‫ ا شريك له‬/ ‫ اه وحده‬/ ‫ا اله اا‬. Reverse: Should be: ‫ اه‬/ ‫ رسول‬/ ‫ محمد‬but completely effaced. Æ; 3.12 gr; 19.45×22.13 mm. Comparisons: For a coin minted at al-Ramla in 218 aH see Berman 1976:39, No. 91. 45. locus 227, Basket No. 2098 date: al-Mu>tasim; 223 aH / 837-838 Ce; Dimashq. Obverse: in centre: ‫ا اله اا اه وحده ا شريك له‬. Marginal inscription: inner circle: ‫بسم اه ضرب هذا الد رهم بد مشق سنة ثلث عشرين ومئتين‬ Outer circle: ‫ه اامر من قيل ومن بعد ويومئذ يفرح المؤمنون‬ Reverse: in centre:‫ اه‬/ ‫ رسول‬/ ‫ محمد‬/ ‫ له‬. Marginal inscription: ‫بسم اه ارسله بالهدى ودين الحق ليظهره على الد ين‬ ‫كله ولو كره المشركون‬ aR; 2.42 gr; 25.39×25.7 mm; axis 2. Comparisons: tiesenhausen 1873:203, No. 1827. 46. locus 210, Basket No. 2047 date: al-Mu>tasim; 227 aH / 841-842 Ce; Madinat al-Salam (Baghdad). Obverse: in centre: ‫ ا شريك له‬/ ‫ اه وحده‬/ ‫ا اله اا‬. Margins: ‫ بسم اه ضرب هذا الدرهم بمد ينة السام سنة اثنين وسبع ومئتين‬. Reverse: in centre: ‫ اه‬/ ‫ رسول‬/ ‫ محمد‬. below: ‫ له‬instead of ‫ه‬. Margins: ‫بسم اه ارسله بالهدى ود ين الحق ليظهره اعلى لد ين كله ولو كره المشركون‬ aR; 2.85 gr; 21.1×21.39 mm; axis 9. Comparisons: lane-Poole 1889:61, No. 315. 13. the author wishes to thank Dr. S. Heideman for his help in identifying coin No. 43. 182 CHaPteR 8 FaUNal ReMaiNS aharon Sasson the excavation at Khirbet es-Suyyagh revealed 268 animal bones, derived from eight loci, nearly all from the late Byzantine and Umayyad periods. One hundred and sixty bones (60%) were classified to body part and species, while the rest were partly classified to a level of group size (table 8.1). the latter bones were divided into three group sizes: bones that were identified as large mammal are probably of cattle or equids; bones that were identified as medium mammal are probably of caprines, pigs or canids; bones that were identified as small animal are probably of rodents or birds. ReSeaRCH MetHODS various modifications of the identified bone were recorded in order to gather as much information as possible: (1) Preservation was defined as a percentage. For instance, half of a distal femur fragment received a rate of 50% preservation. (2) Patina was recorded as such. (3) Fresh breaks caused during excavation were recorded. (4) Bones bearing marks of cutting, chopping or burning were recorded. the bone assemblage was identified based on the reference collection of the institute of archaeology in tel aviv University and atlas of animal bones (Schmid 1972). Sheep and goat bones were differentiated on the basis of criteria outlined by Boessneck (Boessneck 1969) and by Prummel and Frisch (Prummel and Frisch 1986). Caprid bones that could not be separated to sheep or goat were referred as sheep/goat (table 8.1). age at death determination of caprids was based on teeth eruption and attrition (Payne 1973) and fusion condition of bones (Silver 1969). Species representation is presented according to two counting methods, NiSP (number of identified specimens) and MNi (minimum number of individuals, and see Reitz and Wing 1999:121-126). PRiMaRY taXa RePReSeNteD SHEEP/GOAT Sheep and goat bones are the most abundant in the assemblage (61 bones; table 8.1). Of the 61 bones, seven were identified as domestic sheep (Ovis aries) and ten as domestic goat (Capra hircus). the small sample prevents us from concluding that goats were bred in larger numbers than sheep at the site. the MNi values that were derived from the bone assemblage were: three for the sheep/goat category based on three left distal tibias and two for the goat category, based on two right proximal radii (table 8.1). Fourteen unfused bones were identified (table 8.2). Nearly all of these are bones that fuse in prematurity and maturity (18-28 months and 30-42 months). the small sample was not adequate for producing a reliable mortality profile. the three cut marks identified were insufficient for indicating slaughtering patterns. CATTlE Forty two bones of domestic cattle (Bos taurus) were identified, utilizing MNi value of 2, based on two left mandibular premolars (table 8.1). Only one unfused bone and one cut mark were identified (table 8.2). the small sample did not allow further analysis. PIG thirty six pig bones (Sus scrofa) were identified (23% of the total assemblage). the MNi value is 3, based on three left distal humeri (table 8.1). two cut marks were identified (table 8.2). Unfused bones were not identified. Due to the small sample, it would be hasty to conclude that pigs in Khirbet es-Suyyagh were slaughtered solely at maturity. 183 aharon SaSSon aNalYSiS OF SeleCteD lOCi lOCUS 175 this locus represents the fill and collapse inside the northern cistern. the finds from the fill are dated to the late Byzantine to >abbasid periods. the locus is the source for almost half of the total bones assemblage, 113 partly identified and identified bones. the identified bones are comprised of 32 sheep/goat, seven goat, seven sheep (total of 46 sheep or goats) and 18 cattle bones, as well as six chicken (Gallus gallus) and three small cat bones (Felinae). the absence of pig bones in this locus may relate it to the post-monastery phase (Phase iii) rather than to the time of the monastery’s existence (Phase ii). Nearly all of the cattle bones are part of the lower hind limbs; the metatarsals, tarsals and the three phalanges of both sides (left and right). evidently, the lower hind limbs were discarded articulated at time the site was active. Discarding of feet is related to butchery refuse (Hellwing and Gophna 1984; Wapnish and Hesse 2000) however, the entire zoo-archaeological context of this locus comprise also of various sheep/goat body parts that are related to food refuse such as femur, tibia and humerus (three fragments for each body part). Wapnish and Hesse (2000) discuss a similar phenomena of discarded articulated feet bones in the early Bronze age tel Megiddo. they also point out that articulated bones were found as part of dense bone concentrations, similar to locus 175 in our case. two pairs of sheep/goat mandibles (left and right) with teeth attached were identified. Both pairs were most likely articulated. the first pair had three morals that, based of the stage of attrition, were aged to over four years at death. the second pair had three permanent premolars that were aged to 6-12 months at death (Payne 1973). three body parts (humerus, metacarpal and calcaneus) of an infant caprine are possibly related to natural death 184 of newborns. in conclusion, the rich and diversified content of body parts indicates that locus 175 was a dumping spot for slaughter and food refuse. lOCI 330 ANd 358 locus 330 is a fill, rich in finds, related to a construction phase which was undertaken probably around the mid-7th century. locus 358 is a fill with the same nature and probably the same date, not far from the former. a hundred animal bones were collected from these two loci. Of the fragments identified, 28 are pig, 16 are cattle and only nine sheep/goat. the relative abundance of pig bones in these loci in contrast to locus 175 may relate them to Phase ii rather than to Phase iii. two pig mandibles from locus 330 and one maxilla from locus 358 that contained teeth were aged. the mandibles came from two animals and both were aged to 19-23 months at death (Bull and Payne 1982). the maxilla that contained three molar teeth was aged to 7-11 months at death. these age groups may point to slaughter of immature pig (Rappaport 1984:149), however the sample is not sufficient to establish a mortality profile. CONClUSiONS the relative frequency of species in Khirbet esSuyyagh points to a mixed animal husbandry of sheep, goats, cattle and pigs. Due to the small sample of bones, mortality profiles by species could not be obtained. three loci that were characterized as fills contained a relatively large number of bones. Several groups of bones in locus 175 were probably articulated, providing additional indication that this locus was a refuse area. the absence of pig bones in locus 175 and their abundance in loci 330 and 358 strengthens the hypothesis that the first assemblage should be related to Phase iii and the latter assemblage to Phase ii. Chapter 8: faunal remaInS taBle 8.1: Relative FReQUeNCY OF SPeCieS Species NISP # 44 7 10 61 42 36 6 6 3 2 2 2 160 37 65 6 108 Sheep/Goat Sheep Goat Sheep/Goat total Cattle Pig Ursus Gallus Felis Canis Anser Donkey Total of identiied large Mammal Medium Mammal Small animal Total of partly identiied MNI % 3 1 2 38 26 23 4 4 2 1 1 1 100 2 3 1 1 1 1 1 taBle 8.2: Relative FReQUeNCY OF BODY PaRtS, UNFUSeD BONeS, CUt MaRKS aND FReSH BReaKS Body Parts astragalus atlas axis Calcaneum Carpal Cranium Femur distal Femur proximal Horn Humerus distal Humerus proximal Mandibula Metacarpal proximal Metatarsal distal Metatarsal proximal Metacarpal distal Metapod Pelvis Phalange i Phalange ii Phalange iii Radius distal Radius proximal Scapula tarsal tibia distal tibia proximal Ulna teeth Total Species NISP Sheep/Goat Cattle Pig 2 1 0 0 1 0 1 1 4 0 2 0 2 2 4 2 1 1 0 1 0 6 1 3 0 0 8 2 5 2 1 2 2 4 2 1 2 1 1 2 1 0 3 3 5 0 4 1 3 1 1 3 0 1 1 2 0 1 5 0 1 1 1 1 1 7 10 14 54 32 22 Total 3 1 2 4 2 8 3 1 1 10 0 15 3 4 7 3 3 4 8 4 4 2 4 3 1 5 2 3 31 108 Unfused S/G 0 0 0 4 0 0 1 0 0 1 0 0 0 2 0 2 0 0 0 0 0 1 0 0 0 2 1 0 C 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 14 1 Cut Marks S/G C P Fresh Break S/G C P 1 1 1 1 1 2 1 1 1 1 5 2 2 2 1 1 1 2 1 2 1 1 1 1 1 1 1 1 1 1 2 1 1 4 1 1 3 1 2 26 14 4 S/G = sheep/goat, C = cattle, P = pig 185 CHaPteR 9 eaRtHQUaKe-RelateD DaMaGe Shmuel Marco Damage to man-made features and natural bedrock is recognized at some points at the site of Khirbet esSuyyagh. there are three types of plausible causes for this damage: deliberate man-made destruction, slow protracted wear of masonry and erosion of underlying rock, and catastrophic earthquakes. Distinguishing the three types is important for understanding the human and natural history of the region in general and of Khirbet es-Suyyagh in particular. Damaged architectural remains can be recognised throughout the site. Signs of destruction and nearly immediate rebuilding combined with absence of signs of man-made violent actives are typical earthquake-related features. the area of the large courtyard (Fig. 2.1:8-10) had been completely rebuilt after a destructive event. an earlier construction phase, which is observed south of the centre of the courtyard (Fig. 2.1:9), is covered by a later floor. Fallen masonry and subsequent repairs were observed in the southern part of the apse of the church, with its inner face remaining asymmetric. Since the damage is observed close to the foundations of the church it seems that the damage had a pervasive affect on the entire structure. a section of about 10 m in the southern end of W33 seems also to have been rebuilt. Similarly, in W100 there is a warped contact in room 19, where two different styles of masonry meet but are misaligned. another type of damage appears in two broken door thresholds, that of the main gate and that of the small courtyard in the south of the monastery. the large, monolithic and nicely carved stones are placed in-situ but broken by a width wise crack into two pieces. assuming the thresholds were carved from intact rocks without significant fractures, we can envision strong vertical acceleration, perhaps 186 of the order of 1g, which caused the fracturing. Such strong shaking is known based on modern earthquakes to occur either near the epicentre of strong earthquakes (of the order of magnitude 7 and above) or in places with strong local amplification of seismic waves. each of the damaged elements alone would not suffice to indicate an earthquake as the damaging agent. However, the occurrence of many such elements, the extensive repair and reconstruction of features without any sign of human violence and in short time, together with the frequent occurrence of earthquakes in the region supports the association of the damage to earthquake/s. ideally, the proposed evidence for earthquakerelated damage would be corroborated by historical accounts and/or by additional independent archaeological observations in contemporary sites and/or geological evidence. Geological evidence for strong earthquake shaking in the vicinity of the site is found in the form of fallen cave deposits (speleothems), in particular stalactites and stalagmites. the damage events to speleothems were dated using the Uranium-series method (Kagan et al. 2005). Several breakage events of speleothems were identified in the Soreq and Hartuv caves, the age of the youngest of which is estimated by five millennia. a doctorate in progress by e. Kagan (the Hebrew University of Jerusalem) indicates that the analytical methods that were used in the previous stage (i.e., alpha counting) are not sensitive enough and by using MC-iCP-MS system younger historical events are emerging (personal communication). Hence, the regional seismic activity is capable of triggering severe damage in the vicinity of Beth Shemesh. the time of the damaging episode in Chapter 9: earthquaKe-related damage Khirbet es-Suyyagh is determined by the ceramic and numismatic finds to around the middle of the 7th century. this is supported by historical accounts on 7th century earthquakes, which report considerable damage in Judaea by earthquakes that are dated to 631/2 Ce and 659 Ce (amiran, arieh and turcotte 1994; Guidoboni 1994:355-358; and see Chapter 10). these reports need to be examined in detail and cross-checked in further studies. the most likely source of earthquakes in this region is the Dead Sea fault, an active boundary between the Sinai and the arabia tectonic plates. the plates’ relative movements of average 4-5 mm/yr trigger occasional strong earthquakes. the earthquakes are recognized in numerous damaged archaeological sites throughout the Middle east as well as in deformed rock units (e.g., amit et al. 2002; Kagan et al. 2005; Marco and agnon 2005; Marco et al. 2005) the nearest segment of this fault is in the Jordan valley, some 50 km east of the site, which ruptured in the earthquakes of 31 BCe, 363 Ce, 749 Ce, and 1033 Ce (Marco et al. 2003). in conclusion, it is highly likely that the observed damage and subsequent repairs in Khirbet esSuyyagh were caused by one or more earthquakes. 187 CHaPteR 10 KHiRBet eS-SUYYaGH iN CONteXt itamar taxel the relatively sparse finds, mainly pottery, dated to the late Hellenistic/early Roman period bear witness to the small scale of the settlement which existed here during that time. Unfortunately, the re-inhabitation of the site in the late Roman period, and above all the massive building activity that took place here in the late Byzantine period, completely demolished almost every remnant of the oldest settlement. structures. Such hip-baths, oval or rectangular in shape, usually rock-cut but also built, were found in early Roman period sites throughout the country (Reich 1990:129-130; Hirschfeld 2000a: Figs. 20, 21; 2000b: Figs. 31-33; Magen 2004: Fig. 36; Netzer 2001: ills. 47, 225, 332). the present feature lacks the seat that is usually placed in one of the bath’s edges, but this could have been in its western part which was demolished during the building of the oil press’s western wall. the dating of this feature is not clear, since no dateable finds were found associated with its foundations or building matrix. We know only that it was fell from use in the late Roman/ early Byzantine period. although similar features are known also from late Roman contexts (e.g. Seligman 1995:62-63), on the basis of most of the parallels it can be dated to the early Roman period. THE ARCHITECTURAl CONTEXT THE REGIONAl CONTEXT the only built feature which can be securely dated to the early Roman period is Cistern 24. Rockcut or built cisterns are a very common feature in settlements of all types from this period. However, the great majority of cisterns of that time are round, cylindrical or bell-shaped. Only a few contemporary square or rectangular cisterns have been published to date (e.g. tsuk 1994:136, No. 2). another feature which might be dated to the early Roman period is the oval plastered installation unearthed beside the western foundations of the late oil press. the use of this small feature is not clear. although theoretically it could have been used as the collecting vat of an industrial installation, it seems that this was not the case here, primarily because of its oval shape. the more plausible identification of this feature, therefore, is within a domestic context. if it were indeed part of a residential structure, it might have been used as a hip-bath, a well-known feature mainly in late Hellenistic/early Roman Khirbet es-Suyyagh lies in a region which was densely settled during the late Second temple period. Surveys and excavations conducted in its vicinity discovered evidence for the existence of settlements in the Hellenistic and early Roman periods at relatively many sites. an area of 10 km², in the northwestern corner of which Khirbet es-Suyyagh is located, was surveyed for the map of Nes Harim (Weiss, Zissu and Solimany 2004). Within the mapped area 33 sites were dated to the Hellenistic period, although the surveyors had difficulties to distinguish between early and late Hellenistic (i.e., Hasmonean) sites based on the ceramic finds only. they identified the nature of the Hellenistic sites as small villages, agricultural estates and farmsteads (ibid.:14*). More than 30 sites were generally labelled as Roman, among which 24 were dated to the early Roman period. Of these 14 were already inhabited in the Hellenistic period. the Roman period sites were identified as this chapter is dedicated to a synthesis of the site as a whole and of its various units from the archaeological, historical, economic and social points of view. it deals with the main historical phases of the site's existence, from the Second temple period to the early islamic period. late HelleNiStiC/eaRlY ROMaN PeRiODS 188 Chapter 10: KhIrbet-eS-Suyyagh In Context small villages and farmsteads. at least two of these sites were probably inhabited by Jews, since they include hewn installations which may be ritual baths (miqva’ot) (ibid.:14*-15*). Unlike the area mentioned above, that which surrounds Khirbet es-Suyyagh on the north, west and southwest was not thoroughly investigated. Parts of the map of Beth Shemesh, which spread over an area of 10 km² west and southwest of Khirbet esSuyyagh were surveyed, but no final report has yet been published. the few surveyed sites are located 2-4 km southwest of Khirbet es-Suyyagh, and all yielded pottery which was generally defined as Roman (Dagan 1991). the closest most extensively explored site in the vicinity of Khirbet es-Suyyagh is îorvat >illin, located 0.5-0.6 km to its southwest. excavations conducted at the site revealed, inter alia, burial caves and cisterns dating to the late Hellenistic and early Roman periods (Seligman and May 1993:78; Seligman, Zias and Stark 1996:43-44; Weksler-Bdolah 1996). another nearby site is Khirbet Fattir, located 2.8 km southwest of Khirbet es-Suyyagh. excavations conducted at this site revealed remains of a Jewish village from the late Hellenistic/early Roman period, which includes cisterns, a ritual bath (mikveh), a wine press, storage basements and burial caves (Strus 2003:73-97). these rock-cut features are the only remains of the village preserved, due to intensive building activity took place at the site in later periods. a subterranean hewn complex and several rock-cut and built features dated to the early Roman period were unearthed also at Khirbet el-Jiljil, 3.2 km southwest of Khirbet es-Suyyagh. Here too, the remains were attributed to a small Jewish settlement which was almost completely demolished by later building activity (Strus and Gibson 2005:47-55). Other Jewish miqva’ot and burial caves were discovered within the territory of the modern monastery of Beit Jimal, 1 km southwest of Khirbet Fattir (ibid.: 455-464; Zissu 2001:149-150). Remains of a rural settlement which included oil and wine presses, terraces and various hewn features (eisenberg 2000) were unearthed at Naúal Yarmut, 4 km south of Khirbet es-Suyyagh. Within a radius of ca. 7 km to the west, northwest and northeast of Khirbet es-Suyyagh, early Roman period finds and remains (burial caves and miqva’ot) were reported from Deir Rafat (Bagatti 1990), Khirbet el-Marmita (Gershuny 2006), Naúal Yo’el (Weiss 1994) and îorvat îushsham (Zissu 1999). dISCUSSION During the early Roman period the northern Judaean Shephelah was an integral part of the province of Iudaea (avi-Yonah 1966:94-96). However, it cannot be said for sure whether Khirbet es-Suyyagh itself was included within the territory (toparchy) of Jerusalem, or within the territory of nearby Beth Nattif (Pella, according to Josephus, The Jewish War iii 3, 5; Betholeptephe, according to Pliny, Historia Naturalis 5, 70). at any rate, the northern Shephelah was a typically Jewish region which was dominated by rural settlements of various types. the ceramic assemblage of the 1st century BCe-1st century Ce from Khirbet es-Suyyagh is quite varied, and including a few fine open forms, cooking vessels, jugs, juglets, lamps and many storage jars. this variety of vessel types indicates that the site was not used temporarily, but as a permanent agricultural settlement, albeit a small one. the location of the site, the limited area within it which yielded finds from this period and the relatively meagre finds themselves, indicate that this was a farmhouse, probably inhabited by no more than one family. as pointed out by Hirschfeld, the fact that such small settlements were founded in open agricultural land, frequently on elevated prominent points in the landscape, is an indication of a high degree of security (1997:74). two types of farmhouse are known in late Hellenistic and early Roman Palestine: a farmhouse with a defensive corner tower, and a simple farmhouse without a corner tower. the two types of farmhouse were situated in similar topographic locations, namely on high points which dominated the surroundings, although unfortified farmhouses were also built in low areas. Both types of farmhouse were very common in early Roman Palestine, mainly in the central hill country. Many 189 Itamar taxel of them were inhabited by Jews, as indicated by the finding of miqva’ot within their area (Hirschfeld 2000c:709-720; Zissu 2001:253-257). intensive construction at the site during the Byzantine period, which removed most earlier remnants, precludes identification of the exact nature of this complex. the farmhouse was situated on a rise which overlooked its surroundings, but there is no way to know whether it was fortified with a tower or not. We should not rule out the possibility that not only the Cistern 24 but also other late Byzantine walls and features were built more or less in the same place as those of the early Roman period. the coin issued during the second year of the Great Jewish Revolt (67/8 Ce) provides the terminus post quem for the abandonment of the early Roman settlement. thus it is tempting to relate this abandonment to the Roman Fifth legion’s campaign in Judaea and idumaea in spring-summer 68 Ce (Josephus, The Jewish War iv, 419-439, 443-449, esp. 445-448; Rappaport 1983:54; Schürer 1973:498-499). evidence of resettlement by Jews after the First Jewish Revolt (Zissu 2001:308) was found in relatively many Jewish villages in the Shephelah and other regions. However, this was not true of farmhouses and other isolated rural settlements. it has been explained as stemming from security and Halakhic considerations (Hirschfeld 1997:81-84). this is reinforced by the situation reflected in the excavation of Khirbet es-Suyyagh where no pottery types or coins of the late 1st and 2nd centuries were found. late ROMaN/eaRlY BYZaNtiNe PeRiODS after the abandonment of the site, probably during or shortly after the Great Jewish Revolt, the site was unoccupied for at least 200 years. apparently, the terminus post quem for the reinhabitation of the site can be determined by the two 3rd century coins dated to 218-222 Ce and to 235-238 Ce which were found in the excavations. However, since two complete bowls dated to the 4th century were embedded in the foundations of the wine 190 press treading floor, this date seems to be a much more reasonable terminus post quem than the early 3rd century. the end of this phase occurred, again according to the ceramic evidence, around the late 4th/early 5th century. THE ARCHITECTURAl CONTEXT the relatively small quantity of finds, mainly pottery, dated to the late Roman/early Byzantine period shows that the degree of activity at the site during that time was similar to that of the preceding period. the only architectural remains which can be securely dated to this time are the treading floor of a wine press and the later phase of the northern cistern in the north of the site. there is no other information about the 4th-5th centuries, but the nature of the finds and remains indicates that it was no more than a small agricultural settlement, probably a farmhouse. the estimated size of the treading floor (ca. 7×7 m) indicates that the wine press to which it belonged could produce a large quantity of wine. However, since neither the central part of the treading floor nor any other part of the wine press was preserved, neither its complexity or sophistication can be assessed. From the Roman period onwards, both simple and complex wine presses of various sizes were in use in Palestine. However, since only rarely does the size of treading floors in simple wine presses extending over 5×5 m (Frankel 1999:52-53), it is more reasonable to assume that this particular example belonged to one of the complex (or improved) types. the most common type of complex wine press is the one known as the 'four-rectangle plan' which flourished mainly from the late Roman period onwards. the basic units of such a wine press included a large treading floor, a settling pit and a collecting vat. Sometimes there was more than one collecting vat. Many wine presses also had a square or round stone for anchoring a crushing screw in the middle of the treading floor (ibid.:149-150). Wine presses dated with certainty to the late Roman/early Byzantine period are not very frequent in Palestine. the examples published show that these wine presses were paved either with mosaic or with Chapter 10: KhIrbet-eS-Suyyagh In Context plaster. in the region of Jerusalem and the Shephelah there are simple wine presses at >ein Karim (dated to the 4th century; Saller 1946:92-95) and Khirbet Badd >isa (same date; Magen, tzionit and Sirkis 2004:181-182) and a complex wine press at nearby Khirbet Fattir (dated to the 3rd or 4th century; Strus 2003:105, 419). More distant examples are a complex wine press with several treading floors and collecting vats at akhziv, on the coast of western Galilee (dated to the 4th century; Syon 1998), and two complex although smaller wine presses near tel Qasile, in the central coastal plain (dated to the 2nd century; ayalon 1984). an important detail concerning the wine press is the fact that the tesserae for its mosaic pavement were manufactured at the site itself. this is demonstrated by the large amount of tesseraemanufacturing waste that was found below the floor and in the fill which covered it. the tesserae were made of hard white dolomitic limestone, which does not exist at the site itself but is very common in the Judaean Hills to the east. Waste from tesserae-manufacture was found below mosaic floors at other rural sites too, although they are from the Byzantine period, e.g. the monastery of Martyrius (Magen and talgam 1990:149), the church at îorvat Berachot (tsafrir and Hirschfeld 1979:297, Fig. C), the synagogue at Meroth (ilan and Damati 1987:100) and wine presses at emmaus (Hirschfeld 1981:385) and near Jerusalem (Rahmani 1991:101) (see also talgam 2002:9). the two large rectangular beam weights of the reversed-t type found in secondary use in later parts of the site suggest the presence of an older oil press here prior to the late Byzantine period. according to the archaeological evidence, the reversed-t weight was the main type of beam weight in the levant from the Hellenistic until the Byzantine period (Frankel 1999:101), but their most widespread use was during the Hellenistic and Roman periods. examples for contemporary oil presses with reversed-t weights, some very similar to those from Khirbet es-Suyyagh, are known from other sites in the central hill country, such as Pisgat Ze<ev (Shukrun and Sauariego 1993: Fig. 67), Shoham Bypass Road (Dahari and >ad 2000: Fig. 110), Naúal Yarmut (eisenberg 2000: Fig. 171), Beth Neúemia (Yekutieli et al. 2001: Ph. 5), Khirbet Badd >isa (Magen, tzionit and Sirkis 2004: Fig. 22) and îorvat Burnat (amit, torge and Gendelman 2008:101). all these are dated to the Second temple period. Weights very similar to ours were found in an oil press from Sumaqa at the Carmel. in this case they were dated to the 3rd-5th/6th centuries (Dar 1999: Figs. 58-59), and thus closer in time to the late Roman/early Byzantine phase at Khirbet es-Suyyagh. THE REGIONAl CONTEXT in contrast to the relatively dense settlement which existed in the region surrounding Khirbet esSuyyagh in the early Roman period, the settlement pattern of the late Roman/early Byzantine period seems much less intensive. the surveyors of the Nes Harim map identified only three sites as late Roman, but they also admitted difficulty in distinguishing between the Roman and the Byzantine periods at many other sites (Weiss, Zissu and Solimany 2004:14*). these sites are: Deir el-Hawa (3.5 km east of Khirbet es-Suyyagh; ibid.:32*), Naúal azan (5.5 km to the southeast; ibid.:47*) and Har Kitron (9 km to the southeast; ibid.: 49*). Secure late Roman and/or early Byzantine remains, mainly burial caves but also other features, were excavated in a small number of sites south and southwest of Khirbet es-Suyyagh, namely at Beth Nattif (Baramki 1936), îorvat >illin (Seligman, Zias and Stark 1996:44-60), Khirbet >ein Shams (Weiss 1995), Khirbet en-Nabi Bulus (Rapuano and Yas 1996:89), Khirbet Fattir (Strus 2003:418-424), Beit Jimal (ibid.:465-479) and Khirbet el-Jiljil (Strus and Gibson 2005:55). dISCUSSION During the period under discussion the northern Judaean Shephelah was part of the province of Palaestina, and after the re-division of the latter in the early 5th century it became part of the province of Palaestina Prima. the municipal affiliation of the region was also changed, since in 200 Ce the city of Beth Guvrin was refounded as the polis of eleutheropolis. the city territory of Beth Guvrin 191 Itamar taxel was the largest in Palestine, absorbing inter alia the territory of Beth Nattif and parts of the territory of Jerusalem. its northern border passed to the north of Naúal Soreq (avi-Yonah 1966:115, 159, Maps 15, 16). Khirbet es-Suyyagh seems to be included within this area, but very close to the eastern territory of Jerusalem and the southern territory of emmaus. the identification of the site's inhabitants during the late Roman/early Byzantine period is not clear. Historical sources and the archaeological record point to a mixed population in the northern Shephelah in the late Roman period. From the end of the Bar-Kokhba Revolt until the 4th century, most of the evidence relates to the Jewish and pagan components in the population, and from the 4th century on these groups are joined by Christians. However, during the late Roman period the pagans were the most prominent component in the population of the northern Shephelah and later the Christian community of the region originated from this ethnic group (Schwartz 1986:94-95). these two groups continued to settle in small farmhouses and frequently in abandoned Jewish settlements (Hirschfeld 1997:83; Zissu 2001:268). this was in contrast to the Jews who preferred to live in larger villages and in urban centres. For instance, at nearby Beth Nattif, which was probably a village in antiquity, late Roman oil lamps decorated with Jewish motifs were discovered (Baramki 1936), and excavation of a cemetery in the city of Beth Guvrin yielded evidence for a Jewish presence there in the late Roman period (avni, Dahari and Kloner 2008). therefore, it seems less likely that the inhabitants of the small farmhouse which existed at Khirbet es-Suyyagh in the late Roman/early Byzantine period were Jews. Furthermore, although the city of Beth Guvrin adopted Christianity in the early 4th century, as indicated by the participation of its bishop in the Council of Nicaea in 325 Ce (Geiger 1982:225), it does not seem this was the case for its rural hinterland also. Christianity in 4th century Palestine was concentrated almost exclusively in the Hellenized cities, and even there Christians were a minority compared to the pagan majority (ibid.:225-226). Hence it is very feasible that the 192 small rural community which settled at Khirbet es-Suyyagh in the late Roman/early Byzantine period was pagan. the reoccupation of Khirbet es-Suyyagh in the late Roman/early Byzantine period can perhaps be identified with the phenomenon of the establishment of many new farmhouses and villages throughout the country in the 4th and 5th centuries (Hirschfeld 2005:523-532). Some of these agricultural settlements were founded over the remains of older and abandoned Roman structures. among the factors that brought to this settlement expansion Hirschfeld mentions the late Roman and early Byzantine imperial legislation of laws known as agri deserti. according to these laws, legal ownership of previously uncultivated land was granted to any person who could make it fertile (ibid.:533-534). it is possible that sites which were cultivated in the past but stood deserted until the late Roman/early Byzantine period – such as Khirbet es-Suyyagh and other sites mentioned by Hirschfeld – were included within the agri deserti. the reason for the abandonment of this settlement within a century or so after its establishment is also vague. One reason could have been Christian religious pressure on the region’s pagan population. anti-pagan policy, such as imperial anti-pagan legislation, riots and forced conversions to Christianity, is documented in the history of late 4th and 5th century Palestine. Such pressure on the pagans of the Judaean Shephelah could have been initiated by the civil and/or religious Christian authorities of Beth Guvrin or by individuals who enjoyed the latter’s support (such as the fanatic monk Barsauma of Nisibis) (Rubin 1982:246-249). it is not unrreasonable to assume that such Christian enthusiasm was motivated, inter alia, by the discovery of the relics of some biblical prophets in and near villages in the Judaean Shephelah during the 4th and 5th centuries (Perone 2006:151; taxel 2008:66). another possible cause for the abandonment of the late Roman/early Byzantine settlement at Khirbet es-Suyyagh is the earthquake of 419 Ce. this earthquake is mentioned in literary sources as severely affecting Jerusalem and other settlements in Palestine. archaeologically, the Chapter 10: KhIrbet-eS-Suyyagh In Context destruction of some Galileen synagogues has been related to this earthquake (Russell 1985:42-43, with references), though no clear evidence for such destruction was identified in the scant remains attributed to this period at Khirbet es-Suyyagh. an economic crisis, such as that supposed to have occurred in Palestine in the 5th century (e.g. Fiema 2006:82; Safrai 1998:129-130) is another factor which should be take into account when dealing with the abandonment of a small rural settlement. Yet, the archaeological evidence for such a crisis, especially in regard to the early 5th century (the latest approximate time in which the settlement at Khirbet es-Suyyagh was abandoned), is still relatively scanty and ambiguous (see e.g. Bijovsky 2000-02:204-205, 209), and in any case could not be identified at the site. late BYZaNtiNe/eaRlY UMaYYaD PeRiODS the heyday of the site occurred in the late Byzantine and early Umayyad periods, in the 6th and 7th centuries Ce. During this time, massive construction work was carried out in order to build the well-designed complex of a rural monastery. in the course of its existence the monastery underwent architectural changes due to unexpected natural and/or man-made events. the following is a synthesis of the monastic phase in its historical, archaeological and socio-economic context. IdENTIFICATION OF THE MONASTERy Since no inscription or any other finds which might provide information about the ancient name of the monastery was found in the excavations, this detail may be revealed only by literary sources. Of the dozens of Palestinian monasteries mentioned in the literary sources of the Byzantine period, the one which seems to be the most suitable to be identified with the monastery of Khirbet es-Suyyagh is the Monastery of Samson (Sampso) mentioned by the monk John Moschos in late 6th-early 7th century (Pratum Sprituale 170). the latin translator of John Moschos’ account notes that the monastery of Samson is located 20 miles from Jerusalem, but without giving the direction from the city: distabat ab Jerosolymis fere viginti passuum millibus monasterium quoddam, quod Sampsonis dicitur. Ex hoc monasterio duo Patres abierunt ad monastem Sina orationis gratia… (“about twenty miles from Jerusalem there is a monastery called Sampson, from which two fathers went up to Sinai to pray”). the distance of 20 miles from Jerusalem, and mainly the strong relations between the figure of Samson and the northern Judaean Shephelah (the birthplace of this biblical judge and the region where many of his acts took place), have already led some scholars to identify the Monastery of Samson with sites located in the northern Judaean Shephelah. the first identification was made by abel (1936:539-540), who suggested that the Monastery of Samson should be identified with the so-called Byzantine convent excavated by Mackenzie at tel Beth Shemesh (Mackenzie 1911:75-84). However, as rightly mentioned by Gass and Zissu (2005), that structure lacks any clear evidence of being a monastery, and provides no evidence for the existence of a chapel/church. alternatively, they suggest seeing it as a fortified building, either a road station, a fortified estate or a guard-post (2005:173-175). Gass and Zissu suggest identifying the Monastery of Samson with the site of >iraq isma>in, situated on a cliff on the northern bank of Naúal Soreq (3 km northeast of Khirbet es-Suyyagh). this site, which has been surveyed but as yet not excavated, consists of a large natural cave with a long narrow terrace in front of it. On parts of the terrace and at its edges rock-cut and stone-built remains have been found, including paths, two cisterns and rooms. Near the entrance to the cave, remains identified as a chapel, which was partly built and partly rock-cut, were found. inside the cave, some other built remains were found, some of which are later walls built of re-used ashlars (ibid.:176-180). according to Gass and Zissu (2005), >iraq isma>in is the site of the Rock of etham, Samson’s hiding cave (according to Judges 15:8), a place mentioned by eusebius in the early 4th century (Onomasticon 96:5). Gass and Zissu suggest that this apparently sacred cave site was the place in 193 Itamar taxel which, during the Byzantine period, a monastery (of the cliff coenobium type, in Hirschfeld’s typology of Judaean desert monasteries [1992:34-42, 55-58]) with a memorial church was founded for the veneration of Samson. However, in my opinion, based on personal observation of the site of >iraq isma>in, it is hard to accept both Gass and Zissu’s interpretations of the site’s nature and its identification with the Monastery of Samson. Firstly, this modest site lacks any features which can be identified with a typical memorial pilgrimage-centred monastery, such as a basilical church and a hostel. Furthermore, the site is situated, as Gass and Zissu themselves noted (2005:173), far from any major route. Secondly, the few known monasteries of the cliff coenobium type in Palestine are all located in the Judaean desert (Hirschfeld 1992:33-42), and seem to be an endemic phenomenon of that region. alternatively, i suggest seeing >iraq isma>in as a hermitage of a small group of monks, which were most probably related to a nearby mother monastery (for similar complexes in other regions, see e.g. aviam 2004:201; Hirschfeld 1992:213-222). thirdly, it cannot be proven from John Moschos’ description and its latin translation that the Monastery of Samson was indeed located in the northern Judaean Shephelah, nor that it was built in a cave. Furthermore, we cannot prove that the Monastery of Samson was called after biblical Samson. alternatively, it could have been named for a certain contemporaneous Christian figure by the name of Samson, either a monk or a layman, who founded the monastery and/or financed its construction. if this were the case, the monastery of that name could theoretically be located somewhere else, 20 miles from Jerusalem. Nevertheless, if indeed the Monastery of Samson was located in the northern Judaean Shephelah, its identification with Khirbet esSuyyagh seems much more plausible than that suggested by Gass and Zissu. the monastery of Khirbet es-Suyyagh is also located 20 miles from Jerusalem, and, unlike >iraq isma>in, it is a typical coenobium standing beside a local road, with clear evidence for being engaged in pilgrim hospitality, 194 including a basilical church, a probable hostel and certain small finds (reliquary chest/s, ceramic pilgrims’ ampulla, etc.). My identification of the Monastery of Samson with Khirbet es-Suyyagh is of course tentative pending a more reliable alternative identification. RURAl MONASTERIES the monastery of Khirbet es-Suyyagh belongs to the coenobium (  ) type or, more specifically, to the rural coenobium type. the coenobium in general is “…a monastery in which monks live a communal life, with a daily routine of communal prayer, work, and meals” (Hirschfeld 1992:33). the typical coenobium includes the following components: surrounding wall, gate, tower, central courtyard, one or more churches/ chapels, dining room, kitchen and bakery, storerooms, stables, cisterns, and sometimes even workshops, hospice/s for pilgrims and local visitors and hospital/s. as can be learned from both the archaeological and literary evidence, occupation in agriculture in varying volume of intensity characterized most of the Palestinian monasteries in the Byzantine period, both the coenobia and the laura types, whether they were located in the desert or in the sown regions (Brenk 2004; Hirschfeld 1992:33; Patrich 1993:1066). the present study will focus mainly on the true solitary rural monasteries which are closest in form, architectural components and nature to the monastery of Khirbet es-Suyyagh. in my opinion the title 'rural monastery' should be applied only to those coenobia which reflect most or all of the following characteristics: 1. Geographically, these monasteries are located in the sown and desert fringe regions of the country, in close proximity to rural settlements. Rural monasteries can thus be found in Galilee (mainly western Upper Galilee), in the Beth Shean valley, in the western Samaria Hills, in the Judaean Shephelah and the Judaean Hills (including the western fringes of the Judaean desert), in the coastal plain and in the northwestern Negev (see also Hirschfeld 2006:408). Chapter 10: KhIrbet-eS-Suyyagh In Context 2. topographically, these monasteries are located on landscape formations with moderate slopes (on a plain or summit/moderate slope of a hill or a spur), i.e., they belong to the type of monastery determined by Hirschfeld as the level type coenobia (1992:34-42). 3. all these monasteries are modest in size (up to 3 dunam) and usually also in their internal and external architecture. 4. the great majority of these monasteries have remains associated with them which indicate the cultivation of agricultural lands in the environs of the monastery and/or processing of agricultural crops within the monastery complex or in its immediate vicinity. in some of the monasteries evidence for the raising of farm animals (remains of stables/pens or faunal remains) were also found. these monasteries functioned as farmsteads to all intents and purposes. a few complexes, indeed, have been labelled as monastic farms and not as real monasteries by scholars, but they will be treated below as the rest of the monasteries. it is true that there is evidence for agricultural activity, such as terraced or walled plots and installations used for storing and conducting water (but not for processing agricultural crops), at some of the coenobia situated in the Judaean desert. these were mainly in its central and eastern sectors, such as the monasteries of Martyrius (Damati 2002), Chariton (Hirschfeld 2002c), euthymius (Hirschfeld 1993:359, Fig. 5) and Khirbet ed-Deir (Hirschfeld 1999:91-94). However, these monasteries cannot be considered as true rural monasteries due to their relative remoteness from the heartland of rural settlement and also due to the nature of their agricultural activity. it is most probable that the extent of agricultural activity undertaken in these monasteries was more limited than that of the true rural monasteries, where agriculture was their main (and sometimes only) source of livelihood. One of the main reasons for this was the drier climate of the desert regions compared to that of the sown and even the desert fringes, which “… permits cultivation of crops without irrigation, including winter cereals, fruit trees such as olives and deciduous trees” (Hirschfeld 1992:8). another factor was the considerable financial support given to many of the Judaean desert monasteries, reflected in the size and relative material wealth of some of them (Hirschfeld 1992:102-104). this sponsorship allowed them to engage in low-scale agriculture (mainly irrigated plots and perhaps animal husbandry) for their own consumption. this was most probably also the nature of the coenobia in the eastern part of the Beersheba valley (Hirschfeld 2004a:83) – at tell Masos (Figueras 1995:443-445; Fritz 1983), tell >ira (Cresson 1999; Figueras 1995:442-445) and tel Yeshua> (Figueras 1995:445-447; Govrin 1991:61*-62*). in this context it is worth noting that several urban monasteries which are quite similar to the more remote rural monasteries were established close to Jerusalem (up to 2 km from the city). they included agricultural plots and/or installations used for processing crops. examples are the monasteries at Dominus Flevit on the Mount of Olives [Bagatti 1955-1956; 1969] and at Mount Scopus [amit, Seligman and Zilberbod 2003]). However, since these monasteries were no doubt part of the religious-pilgrimage fabric of Christian Jerusalem, despite being located outside the city walls and depending also on agriculture for their living, they are not considered here as true rural monasteries. On the other hand, there were monasteries situated in the countryside whose livelihood was based mainly on pilgrimage (and donations), such as Kursi near the Sea of Galilee (tzaferis 1983) and Khirbet ed-Deir in the southwestern Judaean Hills (Kopp and Stève 1946; tal 1997:274-275). Besides the solitary rural monasteries built some distance from settlements, there are monasteries (or complexes thought to be such) which were built as an integral part of villages, usually on their fringes. these may be called village-annexed monasteries. Some of these complexes were similar in plan and architectural components to the solitary rural monasteries and others were different (e.g. Ramat Raúel [testini 1962; 1964], Shoham Bypass Road [Dahari and >ad 2000:57*-58*], îorvat Zikhrin [Fischer 1989:1793-1796], îorvat Beth loya [Patrich and tsafrir 1993] and Khirbet Yattir [eshel, 195 Itamar taxel taBle 10.1. RevieW OF PUBliSHeD SOlitaRY RURal MONaSteRieS iN PaleStiNe. Site Area* Church Tower Oil press Wine press Water source >ain el-Jedide Beth ha-Shittah Bir el-Qutt Deir el->asfura Deir Ghazali Deir Qal>a ein el-Sachaniah îorvat >amude Qerayot îorvat Hani îorvat Migdal Khirbet abu Ghunneim Khirbet abu Rish Khirbet ed-Dawwara Khirbet el-Bureikut Khirbet Deir >arab Khirbet Deir Sam>an Khirbet ed-Duweir Khirbet Jemameh Khirbet Ma>ar Khirbet el-Qa§r Khirbet el-Quneitra Khirbet el-Qu§eir Khirbet es-Suyyagh Khirbet et-tina Khirbet Umm Deimnah Khirbet Umm îalasa Mevo Modi>im Naúal Kidron el-Qa§r ca. 530 ca. 240 ca. 1400 ca. 1500 ca. 700 ca. 2100 ca. 1.3 ca. 500 ca. 1130 Known 1000 ca. 450 ca. 350 ca. 1500 ca. 1500 ca. 2600 ca. 1600 ca. 900 Known 750 ca. 2250 ca. 1500 ca. 1500 ca. 1800 ca. 1900 ca. 560 ca. 480 ca. 2000 ca. 500 ca. 1750 ca. 500 (including farmhouse) Known 170 ca. 350 ca. 700 ca. 3000 C C C ? C C C C? C UD C C ? ? C C C UD B C C ? B UD C C C C C ? + + ? + + ? + ? ? ? ? ? + + ? + + +? + ? ? 1i 1i 1e? 1i 2e 2i 1i ? ? ? ? 5i 1i 1e 1i 1i 1i 1i 1i 1i 1i 1i and 1e ? 2i and 2e 1i 2e 1e 1e 1 i? 2e 1e ? 1e ? ? 2e 1e 2e 1e 1e 2e 1 CN, 1 R ? 3 CN 1 CN 1 CN 2 CN 4 CN 2 CN 2 CN C? C C? UD ? + - UNe 1i 2i 1e 1e 1e 1e 2 CN 1 CN 1 CN 3 CN; 1 P Ramot Ras et-tawil Rujm Jerida Siyar al-Ghanam Spring; 1 CN 1 CN 1 CN 1 CN Spring; 1 CN; 5 P Spring; 1 CN 1P 3 CN 1 CN 1 CN; 1 P 2 CN 3 CN; 1 R ? 2 CN; 2 P 2 CN; 4 P 1 CN 3 CN 1 CN? legend: B=Basilica; C=Chapel; CN=Cistern; e=external; i=internal; P=Pool; R=Reservoir; UD=Undefined; UN=Unknown Number. * the area (in m²) within the enclosing walls of the complex, not including external separated features/structures. in cases when only part of the monastery was preserved/excavated, only the known area will be mentioned. 196 Chapter 10: KhIrbet-eS-Suyyagh In Context Magness and Shenhav 2000]; there is no certainty about the identification of all of these complexes as monasteries). the great majority of rural coenobia was discovered in the central hill country and in the neighbouring regions. Clusters of monasteries are known from the vicinity of Jerusalem, the western fringes of Samaria and Judaean Hills and the Judaean Shephelah. Other clusters of rural monasteries are known from the vicinity of Gaza, Beth Shean and other regions in Palestine. Most of these monasteries are known only from surveys, although relatively many were excavated fully or partially (see e.g. aviam 2004:197-200; Hirschfeld 2002a:185-189, 2004b). in some cases, when the remains were fairly well preserved, a complete or nearly complete plan of a monastery was made even if it was not fully excavated. the complexes identified with high certainty as rural monasteries (of the solitary, not village-annexed form) from the central hill country and the neighbouring regions, which their preliminary or final publication includes their complete or nearly complete plan are: 1. In the Judaean Hills: Deir Ghazali (avner 2000), Khirbet abu Rish (Baruch 1998; Magen and Baruch 1997), Khirbet Siyar el-Ghanam (Corbo 1955:11-88), Bir el-Qutt (ibid.:112-139), Khirbet abu Ghunneim (ibid.:141-145), Ras ettawil (Gibson 1985-6), >ain el-Jedide (Hamilton 1935), Naúal Kidron (Hervé and Zelinger 2006), ein el-Sachaniah (Sar-avi 1999), Khirbet Umm Deimnah (Magen and Batz 2008), Khirbet edDawwara (Batz and Sharukh 2008) and Rujm Jerida (Magen, Peleg and Sharukh 2008). among the lesser-preserved published monasteries in this region one can mention Ramot (arav, Di Segni and Kloner 1990). 2. On the western fringes of the Samaria Hills: îorvat îani (Dahari 2003), Khirbet Deir >arab (Conder and Kitchener 1882:311-313; Yitach 2001), Khirbet Deir Sam>an (Conder and Kitchener 1882:319-320; Dar 1986:26-35; Hirschfeld 2002a:185), Khirbet ed-Duweir (Conder and Kitchener 1882:311, 333; Hirschfeld 2002a:187), Deir Qal>a (Hirschfeld 2002a:155-181; Magen and aizik 2008) and îorvat Migdal (ayalon 2002; Matthews, Neidinger and ayalon 1990; Neidinger, Matthews and ayalon 1994). 3. In the Judaean Shephelah: Mevo Modi>im (eisenberg and Ovadiah 1998) and Deir el>asfura (Strus 2003:44-48; Waliszewski 1994). 4. On the western fringes of the Judaean desert: Khirbet Umm îalasa (amit 1997), Khirbet ettina (Hirschfeld 1985a:51*-52*; 2002:139-142), Khirbet el-Quneitra (Hirschfeld 1985b; 2002a:139), îorvat >amude Qerayot (Hirschfeld 1988-1989), Khirbet el-Bureikut (Hirschfeld 1990:46-48), el-Qa§r (Hirschfeld 1990:56; 2002b:136-138) and Khirbet el-Qa§r (Magen, Har-even and Sharukh 2008). 5. In the vicinity of Gaza: Khirbet Jemameh (Gophna and Feig 1993). among the rural monasteries (both solitary and village-annexed) or complexes thought to be such from northern Palestine, it is worth mentioning Sede Naúum (Goldfus 1998:230-233; tsori 1962), Beth Hashita (aharoni 1954), Khirbet Ma>ar (aviam 2004:198) and Khirbet el-Qu§eir (western complex?; aviam 2004:198; Frankel 1992:49-59). the classical plan of a level type coenobium is a square or rectangular centralized complex, which included within its surrounding walls all the most important units of the monastery, including the church. almost invariably the church was situated within the coenobium’s outer walls, with at most only the church’s apse projecting outwards (e.g. Corbo 1955:19-23, tav. 63; Cresson 1999:93, Fig. 3.83; Hamilton 1935:112-113; Ma>oz 1993). Sometimes the monastery church was freestanding (Conder and Kitchener 1882:311; Marti 1880:36, taf. 2), or even exactly in the centre of the monastery (tzaferis 1983:5, Plan 2). Only one complex which was fully excavated – that of Shoham Bypass Road – reflects a situation similar to that at Khirbet es-Suyyagh. However, this site was not securely identified as a monastery, and in any case it was probably annexed to a village (Dahari and >ad 2000:57*-58*). this complex is built as a large rectangle, composed mainly of a numerous rooms and halls of various sizes. the church belonging 197 Itamar taxel to this complex is of the basilica type and projects from its southeastern corner, toward the east. Only the atrium and the western half of the narthex are included within the borders of the rectangular main unit (ibid.: Fig. 112). the entry into this church was, therefore, from the complex itself, differing from the situation at Khirbet es-Suyyagh. two additional examples of monasteries in which the church was built as an independent unit outside and adjacent to the living quarters, were found in northern Syria. at Umm al-Kutten the church (basilica) is built beside the living quarters on its northwest (Butler 1969: ill. 89), and in the monastery of St. George at Sameh the church (also a basilica) is built adjacent and outside the southern wall of the complex (ibid.: ill. 92). PIlGRIMAGE IN THE HOly lANd the monastic movement in Palestine, from its beginning, was closely connected to pilgrimage, spiritually as well as practically. Hospitality was one of the famous characters of the monks, and organized hostels () were built in many Palestinian monasteries. Most visitors to the monasteries, and those who enjoyed the hostel services, were pilgrims who donated money for the maintenance of the hostels (Bitton-ashkelony 2005:148-158; Hirschfeld 1992:102, 196; 2002b:268-269; limor 2006). Monks, however, also visited holy places and holy men in different parts of the country and abroad (Bittonashkelony 2005:142-143). Most hostels were built in urban monasteries, and very few are known from desert or rural monasteries. the most impressive example among the latter is the completely-unearthed hostel at the monastery of Martyrius (Magen and talgam 1990:106-107). Other hostels or pilgrims’ dormitories were claimed to be found or existed at the monasteries of Kursi (tzaferis 1983:20), Siyar el-Ghanam (Corbo 1955:42), Mount Scopus (amit, Seligman and Zilberbod 2003:142), îorvat îani (Dahari 2003:106), Deir en-Nuserat (Hirschfeld 2004a:76) and Mount aaron (Fiema 2003:354-355). as remembered, it was suggested that the large hall and its small backroom at the southeast corner of the monastery of Khirbet es-Suyyagh (Rooms 15 and 16) served as a hostel for pilgrims and guests. 198 a church or chapel was an essential component of every monastery. the routine of the monastery’s community was dictated by the schedule of prayers (Hirschfeld 1992:80, 112). Rather than a large church, monasteries usually included a only a more modest chapel for the monks' devotions. the chapels discovered in Palestinian monasteries have either a long narrow prayer hall (monastic type chapel; Hirschfeld 1992:114), or a smaller and more compact hall. these chapels ended in a single, usually rounded but sometimes square, apse. the monastic type of chapel represents most of the churches built in the Judaean desert monasteries (Hirschfeld 1992:114-117), as well as some outside this region. the latter include some rural monasteries in the central hill country: Bir el-Qutt (Corbo 1955:112-139, Fig. 30), Khirbet abu Ghunneim (ibid.:141-145, Fig. 42), Deir Qal>a (Hirschfeld 2002a:173, Fig. 21; Magen and aizik 2008: 1694), Khirbet Deir >arab (Conder and Kitchener 1882:311), Khirbet ed-Duweir (Conder and Kitchener 1882:311, 333), Khirbet Deir Sam>an (ibid.:320), îorvat îani (Dahari 2003:102), Ras et-tawil (?) (Gibson 1985-6:70-71, Fig. 1), Naúal Kidron (Hervé and Zelinger 2006:290, Fig. 1), Khirbet el-Quneitra (Hirschfeld 1985b:250, Fig. 3; 2002:Fig. 103), el-Qa§r (Hirschfeld 1990:56, Fig. 65) and Khirbet el-Qa§r (Magen, Har-even and Sharukh 2008). Rural coenobia with compact chapels are, for instance, those of >ain el-Jedide (Hamilton 1935:112-113), ein el-Sachaniah (Sar-avi 1999:188, Fig. 2), Khirbet abu Rish (Magen and Baruch 1997: 138, Fig. 2), Khirbet Umm Deimnah (Magen and Batz 2008), Khirbet Umm leisun (Seligman and abu Raya 2002: 129, Plan 1), Khirbet Umm îalasa (amit 1997:263, Fig. 1), Mevo Modi>im (eisenberg and Ovadiah 1998:1*, Plan 1), Sede Naúum (Goldfus 1998:231, Fig. 151), and Beth haShittah (aharoni 1954:211, Fig. 1, Pl. 7:2). Basilical churches are very rare in Palestinian monasteries. examples are known from some Judaean desert and village-annexed monasteries (such as Shoham Bypass Road mentioned above). in transjordan, one can mention the basilica in the large monastic complex at Mount Nebo Chapter 10: KhIrbet-eS-Suyyagh In Context (Saller 1941:45-85, Fig. 27, Pl. 161), and that in the monastic or pilgrimage centre at Mount aaron near Petra (Fiema 2003:347-349, Fig. 2). Both these are pilgrimage-based and not true rural monasteries. as to rural monasteries or thought to be such, basilicas are known from Khirbet Deir Daqla (Drake 1919), Sheikh el-Qatrawani (taha 2002:445-448, Fig. 1) and Khirbet Ma>ar (aviam 2004:198, Fig. 17.18). the main reason, in our opinion, for erecting a basilica at Khirbet es-Suyyagh was the connection between the monastery and the phenomenon of Christian pilgrimage to Palestine in the Byzantine period, especially during the 5th and 6th centuries. Pilgrims from all over the Byzantine empire and the west visited the Holy land’s loca sancta, particularly Jerusalem. they travelled throughout the country, using pilgrim guides and maps which informed them about the holy places and the roads leading to them (limor 2006; tsafrir 1986:129-130; Wilkinson 1977:33-43). the Judaean Shephelah was among the regions included in pilgrims’ itineraries in the Byzantine period. Pilgrims who traveled through this area used the roads which led from Jerusalem and its vicinity to Beth Guvrin (eleutheropolis) and the Shephelah, not only in order to reach Gaza and the southern coastal plain (and from there to Sinai), but also for visiting local sites of Christian interest, which commemorated traditions about figures of the Old testament. One such site was the 'Well of Samson', north of eleutheropolis, which was identified as the place where Samson killed a thousand men with a jawbone of an ass. this place was mentioned, inter alia, by the pilgrim antoninus Placentinus in ca. 570 Ce (Itinerarium 32; see also Hieronymus, Epistolae 108, 14). the battle between David and Goliath is another story which attracted pilgrims to the area. the battle site itself is identified in the itinerary of theodosius the archdeacon (early 6th century) in a place called Mount Buzana, situated exactly in the middle of the road between Jerusalem and eleutheropolis (de situ terrae sanctae 3; tsafrir 1986: 130, 138). this “mountain” is mentioned also by antoninus Placentinus, who erroneously identified it with Mount Gilbo>a. according to him, this was also Goliath’s burial place, marked by a huge mound of stones (Itinerarium 31). Shenhav suggested identifying this site at îorvat îanot, in the southeastern fringes of the elah valley, ca. 7 km south of Khirbet esSuyyagh (2003:169). this was also the region in which the burial sites of some biblical prophets – Zachariah (e.g. antoninus Placentinus, Itinerarium 32; theodosius, de situ terrae sanctae 3), Micah (e.g. Hieronymus, Epistolae 14) and Habakkuk (e.g. antoninus Placentinus, Itinerarium 32) – were believed to be located. On their way to these sites or others pilgrims needed a place in which they could rest, eat, pray and even spend a night or so. these services could be received in road stations and monasteries situated along the main and secondary roads connecting the Judaean Shephelah to Jerusalem and Bethlehem (Di Segni 2001:36). the monastery of Khirbet es-Suyyagh was apparently among the places which had a role in this system; inter alia due to its location beside the road at Naúal Zanoaú and not far from the Jerusalem-Beth Guvrin road. the presence of pilgrims at the site can be attested by the marble lids of the reliquary chest/s and the ceramic pilgrims’ ampulla. as can be seen in other monasteries which are known to have been visited by many pilgrims, the preferred type of church erected in them was the basilica. this was the case at St. Peter (Qa§r >ali) monastery, which served pilgrims travelled along the Jerusalem-Jericho road (Hirschfeld 1992:130, Fig. 30); at the monastery of euthymius in the Judaean desert, that attracted pilgrims who visited the grave of St. euthymius (idem 1993:362-365, 367, Fig. 7); at Kursi, the place identified with the occurrence of the Miracle of the Swine, east of the Sea of Galilee (tzaferis 1983:47-48, Plans 3, 5); and at the monastery of the Burning Bush (St. Catherine monastery) at Sinai (Forsyth and Weitzmann 1970:8, 10, Fig. B). in the monastic complex at Mount Nebo the existence of a basilica can be explained both by the sanctity of the place and by the fact that this church had to serve three different monasteries (Saller 1941). the identification of the ecclesiastical complex at Mount aaron as a monastery is not sure, but the site, identified as 199 Itamar taxel the burial place of aharon, brother of Moses, was visited by many pilgrims, and thus necessitated a basilical church (Fiema 2003:354-355). a possible reason for building a basilica instead of a chapel in the monastery of Khirbet es-Suyyagh, is that the church was used also by the local population from nearby settlements. the environs of Khirbet es-Suyyagh were densely settled during the Byzantine period by villages, farmhouses and other monasteries. Many of these settlements, such as the nearby sites of îorvat >illin and Khirbet Fattir, had churches. Nevertheless, people who lived in smaller farmhouses might have used the monastery’s church on Sundays and other religious occasions. this church could also have attracted people from villages which had their own church, if for example it contained some important sacred relics. this assumption is not unreasonable, taking into account the two reliquary chest lids found at the site. the chest/s and the sacred relics of a martyr or holy man stored inside were undoubtedly kept in the church and may have featured in a specific rite celebrated there. Peasants employed by the monastery in different agricultural and industrial tasks such as the production of oil and wine would also have used the church. THE MONASTERy OF KHIRBET ES-SUyyAGH as can be seen from the architectural description (Chapter 2), the plan of the late Byzantine monastery reflects a typical coenobium. the complex has a thick wall surrounding living quarters, inner courtyards, cisterns and an oil press. a main gate and a corridor lead to the central and western units of the complex from the south, and a subsidiary gate leads to the central unit from the west. a massive tower at its northwestern corner dominates its highest point. a large external courtyard and storeroom lies west of the surrounding wall, and is bounded on the west by a thick defense wall. a church and a large adjacent complex – probably a public dining room and kitchen – stand southwest of the central unit. additional agricultural installations (including two wine presses) were built outside the monastery. 200 in the nature and arrangement of most of its units, the monastery of Khirbet es-Suyyagh is no different from many other rural or desert coenobia. the most striking difference between it and the great majority of the known coenobia is the architectural separation between the main unit and the church complex. the monastery complex was built at the eastern end of a spur probably in order to make use of features originally belonging to the Roman farmhouse. Furthermore, the commanding view to the north and east seem to be a good reason for erecting of the monastery at this spot. the northern and eastern walls of the main complex were built quite close to the steep slopes of the spur and not much space remained outside them for additional building. the most convenient place for building the church was, therefore, west or southwest of the main complex. Of these positions, the southwestern corner was the most appropriate because access to the monastery in general was easiest from the relatively flat area on the west. in addition, since churches were usually oriented east-west, it was reasonable to design it so that passers-by could enter it from the west and/or north without disturbing the privacy and daily life of the monks who lived and worked in the main complex. another example of such monastery planning can be found in the monastery of alahan in anatolia, where a hospice and a large basilical church were built at the western edge of the complex. the church, which was designed to serve only foreign visitors (the monks had their own church, inside the monastery), was entered from outside the monastery (Bakker 1985:100-101, Fig. 71). the ChurCh this is a single-apse basilica, without pastophoria at the eastern ends of the aisles. the church’s gabled roof was probably supported by two rows of columns, founded on stylobate walls. Not much can be said on the upper structure of the church, due to the poor preservation of its walls and the almost complete demolishing of its floors. it is known, however, that the church had relatively rich marble furniture, which adorned its bema, and it Chapter 10: KhIrbet-eS-Suyyagh In Context is possible that at least part of its floor was paved with marble slabs. the most remarkable feature of the church is undoubtedly, its polygonal apse. Churches – whether basilicas or chapels – with a polygonal apse were not very common in the Byzantine period, especially not in the levant. they seem to be, however, slightly more common in asia Minor, Greece and the Balkans, where churches with polygonal apses were built as early as the 4th century (Poulter 1995:163; Winfield and Wainwright 1962:142). Some of the examples have two pastophoria rooms flanking the apse on both sides; a fact which made it sometimes an inscribed apse and not a projecting one, like in the church of Khirbet es-Suyyagh. examples from Byzantium and the neighbouring regions are the Studius Basilica (4th century: Mango 1985: Fig. 43), the St. irene church (6th century: Hamilton 1933: Fig. 14) and the SS. Sergius and Bacchus church (6th century: Krautheimer 1965: Fig. 63) at Constantinople. Others are found at Pergamon (5th or 6th century; Krautheimer 1965: Fig. 30), amorium (late 5th century; lightfoot and irison 2001: Fig. B), Sofia (City Cathedral, 6th-7th century; Krautheimer 1965: Fig. 73), Nicopolis ad istrum (northern Bulgaria, 5th century; Poulter 1995: Fig. 60, Pl. 22) and tsahdripch (Georgia, 6th century; Khroushkova 1998: Fig. 1). in the levant, churches with polygonal, mostly projecting apses, are known from Simdj (6th century; Butler 1969: ill. 118), Chúim-Marjiyat (late 5th century; Waliszewski and Ortali-tarazi 2002: Fig. 45), Jerusalem (St. Stephan church, mid 5th century: vincent and abel 1926: Pl. 77; the Kathisma church, 5th century: avner 2003: Fig. 5), the monastery of Siyar el-Ghanam (chapel, 6th century; Corbo 1955: tav. 6:11, 63), Ramat Raúel (mid-5th century; testini 1964: Fig. 39), emmaus (late 5th or early 6th century; Ovadiah 1970:63-64; vincent and abel 1932: Plan ii), îorvat >eirav (two churches, one is generally dated to the Byzantine period, and the second is dated to the 6th century; ilan 1986:510-511), Jerash (St. theodore church, late 5th century; Kraeling 1938: Plan 33), Madaba (late 6th century; Piccirillo 1982: Plans i, iii), Umm al-Rasas (aedicula church, 6th century; idem 1994: Fig. 27), Machairos (late 5th century; idem 1995: Figs. 1, 2), Khirbet es-Samra (construction date unknown, but functioned until the 8th century; Humbart 1986: Pl. 76:2; Schick 1995:378), and the church on the summit of Mount Sinai (6th century; Dahari 2000: Plan 5). Of all these churches, those closest to that from Khirbet es-Suyyagh are the Studius Basilica in Constantinople, St. Stephan church in Jerusalem and the church at Ramat Raúel. according to testini, churches with a projecting polygonal apse in our region are dated not earlier than the 5th century, and their relatively small number points to foreign influence in all that relates to the apse shape (1962:81). However, the possible polygonal apse of the eleona church in Jerusalem, which is dated to the 4th century (vincent and abel 1926: Fig. 154), perhaps indicates that this form of apse existed in our region as early as the time of emperor Constantine i. the tower a tower ( ς) was a common feature in coenobia, mainly those built in the Judaean desert, but seldom seen in other regions. the towers, characterized by their thick walls and simple plan, were built at the highest point of the monastery, usually overlooking the gate (or one of the gates). their main aim was to provide refuge for the monks in times of danger and to signify ownership of the land. Sometimes they contained not only living rooms but also a cistern and storeroom/s. Based on the surviving remains of towers in the monasteries of the Judaean desert, it seems that they were usually built to a height of no more than three stories (Hirschfeld 1992:171-175, Fig. 94). towers varied in size and shape, being square or rectangular, and usually projecting from the outer face of the monastery’s wall, although cases of ‘inner’ towers are also known. Rural monasteries in the Judaean desert which have towers are those of Khirbet et-tina (Hirschfeld 1985a:52*, Fig. 73), Khirbet elQuneitra (Hirschfeld 1985b:247, Fig. 3), îorvat >amude Qerayot (Hirschfeld 1988-1989:4, Figs.3, 4), Khirbet Umm Rukba (Hirschfeld 1990:15, Figs. 14, 15), Khirbet el-Qa§r (Magen, Har-even and Sharukh 2008) and maybe Khirbet Umm îalasa 201 Itamar taxel (amit 1997:264, Fig. 1). Rural monasteries with towers from other regions are, for instance, those of Deir Qal>a (Hirschfeld 2002a:179, Figs. 10, 33; Magen and aizik 2008: 1694), îorvat îani (Dahari 2003:102-103), Deir Ghazali (avner 2000:36*, Plan 1:16), Khirbet ed-Dawwara (Batz and Sharukh 2008) and Rujm Jerida (Magen, Peleg and Sharukh 2008: 2023). all of these towers could have been protruding, ‘inner’ (i.e., built in adjacent to the inner face of one of the enclosing walls) or even external ones. as can be seen, the tower at Khirbet es-Suyyagh is an important addition to the limited list of towers in rural monasteries. its plan and position fall into the definition of a monastic tower, and can be compared with towers in other monasteries. it is built at the highest point of the monastery, projecting from its northwestern corner, and overlooks the subsidiary gate of the complex. its size (15.8×10 m) is very similar to the northern tower of the monasteries of Chariton (15.4×11.4 m) and Khirbet et-tina (11.4×9.8 m) (Hirschfeld 2002b:245, 249). Similar internal planning was identified also in the southern tower of the monastery of Chariton (ibid.:247). lIVIng quarterS Seven rooms (Fig. 2.1:12-14, 17, 19-21) were identified as cells for the monks who lived in the monastery. two more rooms (Fig. 2.1:5-6) seem to have housed the head of the monastery. the guardroom (Fig. 2.1:3) could have also been occupied by the gatekeeper. the large hall and the small room to its south (Fig. 2.1:15-16) were very probably a hostel. the monks' cells are all rectangular in plan, but vary in area (Room 12: 10.8 m²; Room 13: 12.54 m²; Room 14: 11.48 m²; Room 17: 8.74 m²; Room 19: ca. 22.9 m²; Room 20: 30.1 m²; Room 21: 9.43 m²). therefore, the total area of the cells, is ca. 10.6 m². according to the Codex of Justinian, monks of coenobia monasteries had to sleep in communal halls (Novella 3.3, 5). Nevertheless, it seems that (at least in the Judaean desert coenobia) communal halls were used mainly for young monks and protégés, while senior monks and monks who had 202 specific duties lived in individual cells (Hirschfeld 2002b:274). in the case of Khirbet es-Suyyagh, it is probable that the relatively small Rooms 17 and 21 were private cells. the medium-sized cells, i.e., Rooms 12-14, were maybe used by two monks each, and the largest cells, Rooms 19 and 20, were probably occupied by no fewer than three monks each. all in all, together, maybe, with a few monks that lived in the tower, it seems that the community of the monastery numbered around 20 monks. in his study on the Judaean desert monasteries, Hirschfeld estimated the average number of monks lived in the small monasteries of that region (coenobia and lauras alike) in 20 people, based on general comments in the historical sources and the archaeological evidence (1992:78-79). another rural coenobium, that of Ras et-tawil, which was smaller than that of Khirbet es-Suyyagh, was inhabited, according to the excavator’s opinion, by 10-15 monks (Gibson 1985-6:72). therefore, the suggested number of ca. 20 monks as the size of the monastic community of Khirbet es-Suyyagh fits not only the archaeological reality in the site, but also the conclusions of Hirschfeld’s study. living cells or halls were found in many level type coenobia, although it seems that at least in the Judaean desert coenobia the cells were usually located on the second floor and thus were not preserved (Hirschfeld 1992:176). the information about the living quarters in rural monasteries is not always clear, although in the case of singlestoreyed complexes it is easier to identify the living rooms. However, the excavators/surveyors of monasteries did not always attempt to locate the living rooms. the following cases are exceptions. three long dormitory halls were identified on the ground f loor of the monastery of Deir Qal>a (Hirschfeld 2002a:180, Fig. 10). at îorvat îani, a large room near the entrance to the monastery was identified as the living room of the Mother Superior, and a cluster of rooms in the north of the complex was identified as the nuns’ cells. it was suggested that the monastery’s tower was also used for living (Dahari 2003:102-103). Some Chapter 10: KhIrbet-eS-Suyyagh In Context small rooms in the monastery of îorvat Migdal were identified as monks’ cells (Matthews, Neidinger and ayalon 1990:10, Fig. 7). at Bir el-Qutt, a large hall in the southeastern corner of the monastery was identified as the monks’ dormitory (Corbo 1955:120, Fig. 30:9). in my opinion, the senior monks in this community, or at least the head of the monastery, did not live in the large common hall but in the rooms at the northwestern corner of the complex, near the church. at Ras et-tawil some of the rooms at the west of the monastery were identified as living and storage rooms (Gibson 1985-6:71, Fig. 1). the living quarters of the monastery at Naúal Kidron are supposed to have been located on the second f loor, which was not preserved (Hervé and Zelinger 2006:289, Fig. 1). the monks’ cells in the monastery of Mevo Modi>im were located on the second storey, which was not preserved (eisenberg and Ovadiah 1998:17*). the rooms along the northern wall of Khirbet et-tina were identified as living quarters (Hirschfeld 1985a:52*, Fig. 73). Rooms built north of the chapel of Khirbet el-Quneitra were identified as living quarters (Hirschfeld 1985:250-251, Fig. 3). at Khirbet el-Qa§r the living quarters were identified in the northeastern part of the monastery (Magen, Har-even and Sharukh 2008) and in the north of the monastery at Rujm Jerida (Magen, Peleg and Sharukh 2008: 2023). three large rooms in the south and northeast of the monastery of îorvat >amude Qerayot were identified as the residential units (Hirschfeld 1988-1989:4, Fig. 5). the living rooms of the monastery of Khirbet Jemameh were probably placed partly on the ground f loor and partly on the second storey built above the refectory (Gophna and Feig 1993:100, Plan 2; Hirschfeld 2004a:79). Some of the smaller rooms in the monastery of Beth ha-Shittah were used for living (aharoni 1954:210, Fig. 1). the gatehouSe and the SubSIdIary gate all the features connected to the southern gate of the monastery, i.e., the gateway itself, the path which led to it, the guard room, the small cell and the entrance corridor, represent an independent unit – that of the main gatehouse. the gatehouse was an element associated in historical sources mainly with coenobia (Hirschfeld 1992:161, 163). the term door-keeper (), whose duty was to guard the doors of the church, appears in the register of church officials in the Byzantine period, (Meimaris 1986:253-254). Such an official would seem to be appropriate in a monastic community where there is a true need to guard the main gate of the monastery. indeed, the gatekeeper and the gatekeeper’s cell () are mentioned in the vitae of some of the Judaean desert monks (Hirschfeld 1992:163). the gateway itself was composed of two doors, 0.6 m and 0.75 m wide respectively. the narrower door was probably more frequently used since it was wide enough to easy passage. the wider door would have been used when transporting a large and heavy equipment. the division of the gate into two also made it easier to control at night or in times of hazard. examples of other main gates with two doors can be found also in the monasteries of Khirbet abu Rish (Magen and Baruch 1997:138, Fig. 2), Khirbet Deir >arab (Yitach 2001, Figs. 137, 139), îorvat îani (Dahari 2003:102) and îorvat Migdal (Neidinger, Matthews and ayalon 1994:11, Plan a). if our interpretation of the function of the room situated east of the main gate is correct, this is one of the few examples of guard rooms found in Palestinian monasteries and the only published example of an external guard room. Other such rooms, placed inside the gate, are known from the monasteries of Martyrius (Magen and talgam 1990:95, Fig. 4:room 212), Mount Scopus (amit, Seligman and Zilberbod 2003:142, Fig. 1:4), Bir elQutt (Corbo 1955:122-124, Fig. 30:14, 15), îorvat îani (Dahari 2003:102), the southern monastery at Mount Nebo (Saller 1941:166, Pl. 161:room 72) and maybe also from Khirbet Deir >arab (the room south of the main gate; Yitach 2001: Fig. 137). the corridor, which in this case leads from the main gate northwards towards the external courtyard, is a feature known in other monasteries too. a similar entrance corridor, which, as at 203 Itamar taxel Khirbet es-Suyyagh, also included a staircase, was found at the monastery of Khirbet ed-Deir in the Judaean desert (Hirschfeld 1999:19-21, Figs. 13, 19). Other, non-stepped entrance corridors were built, for instance, in the rural monasteries at >ain el-Jedide (Hamilton 1935: 113), Bir el-Qutt (Corbo 1955:122-124, Fig. 30:15), Deir Ghazali (avner 2000:28*, Plan 1), Khirbet Umm îalasa (amit 1997:260-261, Fig. 1), îorvat >amude Qerayot (Hirschfeld 1988-1989: Fig. 5), Khirbet Umm Deimnah (Magen and Batz 2008), Khirbet Deir >arab (Conder and Kitchener 1882:311; Yitach 2001: Fig. 137), Deir Qal>a (Hirschfeld 2002a:177, Fig. 10; Magen and aizik 2008:1694) and Khirbet Jemameh (Gophna and Feig 1993:97, Plan 2). the second entrance to the monastery, from the west, was a subsidiary gate. it seems that it was used mainly by the monks. Other monasteries with subsidiary gates are those at Deir Ghazali (avner 2000: Plan 1), Khirbet el-Quneitra (Hirschfeld 1985b:245, Fig. 3), îorvat >amude Qerayot (Hirschfeld 1988-1989:4, Fig. 5), Khirbet Deir >arab (Conder and Kitchener 1882:311; Yitach 2001: Fig. 137) and Deir Qal>a (Hirschfeld 2002a:173, Fig. 10; Magen and aizik 2008:1694). the refeCtory a communal dining room was one of the major components in the coenobia monasteries. in most of the coenobia the refectory (refectorium,  ) was built adjacent or close to the church due to the daily practice of the monks to eat together after prayers. Therefore, Popović sees the refectory as a sacred rather than a secular unit of the monastery (Hirschfeld 1992:190-191; Popović 1998:297, 299-303). the relatively few refectories identified so far in coenobia of the level type in Palestine were indeed placed in proximity to the church, usually on its north or west (Popović 1998:287). In almost all cases the refectory was a rectangular single-aisled hall which was built in the same quality as the church (ibid.:297-298). the usual internal arrangement was one or two rows of tables flanked by benches (ibid.:299). the estimated size of the community in the monastery of Khirbet es-Suyyagh (ca. 20 204 monks) makes it probable that there was only one longitudinal table in the centre of the hall. the following is a review of the known examples of refectories in rural monasteries in Palestine. at Siyar el-Ghanam, the refectory (9×5-7 m) was located at the northern end of the central courtyard (Corbo 1955:32, Pl. 63). at Bir al-Qutt, the refectory (105.4 m) and the church, which were built on the western and northern sides of the courtyard respectively, had entrances which faced each other (Corbo 1955:116, Fig. 30:4). at Mevo Modi>im, the refectory (10.5×4.3 m) was built adjacent to the church’s western wall, and the two units had a common doorway (eisenberg and Ovadiah 1998:5*, 17*, Plan 1, Fig. 5). at Khirbet ed-Dawwara, the kitchen (ca. 12×12 m) and refectory (apparently on a second storey) were identified in the northwestern corner of the monastery (Batz and Sharukh 2008). at Khirbet el-Qa§r, the kitchen (18×5 m) and refectory (12×4 m) were located at the north of the monastery (Magen, Har-even and Sharukh 2008). at îorvat îani, the refectory (ca. 132.5 m) was part of a separate building built south of the main complex with the church (Dahari 2003:102, 106). at Deir Qal>a, the hall identified as the refectory (9.2×7.8 m) is situated at the southern wing of the monastery, not far from the church (Hirschfeld 2002a:174, Fig. 10). at Khirbet Jamameh, the supposed refectory (8×3.5 m) was placed in the north side of the courtyard whose western side was occupied by the church (Gophna and Feig 1993:100, Plan 2:iv, Fig. 5; Hirschfeld 2004a:79). among these examples, the refectory at Khirbet Jamameh is the closest in size to Hall 31 at Khirbet es-Suyyagh, and the proximity of the latter to the monastery’s church found its closest parallel in the monastery of Mevo Modi>im. the CourtyardS Courtyards were discovered in most of the Judaean desert monasteries. they are an important feature which “… provided a focal point and served as a gathering place before various ceremonies or during the monks’ leisure time”, and “… also had an important architectural function, as a source of light and air for the surrounding buildings Chapter 10: KhIrbet-eS-Suyyagh In Context of the monastery” (Hirschfeld 1992:166). there was usually one courtyard in each monastery, although cases of two courtyards, either linked to each other or placed in two points of the complex, are also known. the courtyards varied in shape and area and in their inner arrangement, if there was any. Many times a cistern was situated below the courtyard. their floor was made of levelled bedrock, stone slabs or simple mosaic. Courtyards – either rectangular or irregular – were an essential feature also in most of the rural monasteries: Siyar el-Ghanam (three courtyards: ca. 280 m², ca. 64 m² and ca. 66 m²; Corbo 1955:30, tav. 63:14, 15), Bir el-Qutt (ca. 135 m²; ibid.:120-122, Fig. 30:13, Fot. 117), Khirbet abu Ghunneim (ca. 150 m²; ibid.:142, Fig. 42), Khirbet Umm Deimnah (66 m; Magen and Batz 2008), Ras et-tawil (two linked rectangular courtyards: estimated total territory ca. 100 m²; Gibson 1985-6:69, Fig. 1), >ain el-Jedide ca. 110 m²; Hamilton 1935:111, 113), Naúal Kidron (two courtyards: ca. 400 m² each?; Hervé and Zelinger 2006: Fig. 1), Khirbet abu Rish (ca. 35 m²; Magen and Baruch 1997: Fig. 2), îorvat Migdal (96 m²; Neidinger, Matthews and ayalon 1994:11, Plan a), îorvat îani (two courtyards: ca. 240 m² and ca. 30 m²; Dahari 2003:102-103, 106), Khirbet Deir >arab (ca. 500 m²; Conder and Kitchener 1882:311), Khirbet ed-Duweir (ca. 500 m²; ibid.:311, 333), Khirbet Deir Sam>an (ca. 285 m²; ibid.:320), Deir Qal‘a (ca. 750 m²; Hirschfeld 2002a: Fig. 10; Magen and aizek 2008), Mevo Modi>im (100 m² known area; eisenberg and Ovadiah 1998:8*, Plan 1, Figs. 10, 11), Khirbet Umm îalasa ca. 85 m²; amit 1997: Fig. 1), Khirbet et-tina (ca. 90 m²; Hirschfeld 1985a:52*, Fig. 73), Khirbet el-Quneitra (ca. 160 m²; Hirschfeld 1985b:251, Fig. 3), îorvat >amude Qerayot (110 m²; Hirschfeld 1988-1989: Fig. 5), Rujm Jerida (two courtyards: 108 m² and 60 m²; Magen, Peleg and Sharukh 2008), el-Qa§r (32 m²; Hirschfeld 1990:2002b:138, Fig. 66), Khirbet el-Qa§r (85 m²; Magen, Har-even and Sharukh 2008) and Khirbet Jemameh (ca. 180 m² known area; Gophna and Feig 1993:99, Plan 2, Fig. 3). the example of îorvat îani is the closest to the arrangement of the two courtyards at the monastery of Khirbet es-Suyyagh, and the size of the small courtyard at the latter (22.5 m²) is not much smaller than that of the small courtyard at îorvat îani. the size of the main courtyard at Khirbet es-Suyyagh (ca. 157 m²), however, is closest to the courtyard of the monastery of Khirbet Jemameh. the lack of cistern below its floor makes the main courtyard of the monastery of Khirbet es-Suyyagh somewhat unusual, although it bordered the two cisterns of the complex in its northeast. the CISternS Water supply was a vital issue in every settlement which lacks a steady source of water, like a spring. it was, of course, especially critical among the monasteries of the Judaean desert and the Negev (Hirschfeld 1992:148), although the same methods of collecting and hoarding water were in use also in the rural monasteries of the sown regions. the main source of water for these monasteries were the cisterns, which were hewn or built below the courtyard/s and/or near them. Other means of hoarding water were large hewn/built pools, usually located outside the monastery’s walls. the cisterns (and pools) were fed by rainwater collected by a network of gutters and underground channels. Cisterns have been reported from almost every known rural monastery (table 10.1). the commonest type was bell-shaped, although rectangular cisterns are also known. the eastern cistern inside the monastery of Khirbet es-Suyyagh is rectangular, and the one hewn in the agricultural area south of the monastery is bell-shaped. Remains of only one water channel were found below the floor of the main courtyard, in the direction of the two linked cisterns, but it is quite certain that other channels or gutters drained water into the cistern from other directions too. the size of the rectangular cistern of Khirbet es-Suyyagh (4.1×2.6 m) is close to those found at Deir Ghazali (5×2.75 m; avner 2000:36*, Plan 1) and Khirbet Jemameh (4.6×3 m; Gophna and Feig 1993:97, 102, Plans 1, 2). the water conserved in the two cisterns at the monastery of Khirbet esSuyyagh was used for drinking by both humans and animals and for other daily needs (cooking, cleaning, laundry, etc.). When the oil and wine presses were 205 Itamar taxel in use, in the late summer and autumn, the cisterns also provided water for the special needs of these installations (mainly for the washing of the picked olives and grapes and for cleaning the working surfaces and vats). it is very possible that there was another cistern outside the complex, maybe near the wine press (as at Dominus Flevit), but no remains of such were discovered. the cistern found in the agricultural area south of the monastery was used for irrigating the crops cultivated in this area. the total capacity of the two cisterns built inside the monastery was at least 91 m³ (northern cistern [later stage] – at least 43 m³; eastern cistern – at least 48 m³). the annual amount of water which was available to each of the ca. 20 inhabitants of the monastery was, therefore, at least 4.5 m³. indeed, the irrigation of vegetable gardens/plots (which were most probably existed near the monastery; see talbot 2002) and watering the monastery’s animals also required water. Some of the water could, of course, be taken from the bell-shaped cistern in the agricultural area south to the monastery. However, i intend to believe that at least one more cistern existed closer to the monastery (maybe in the flat area to its west). if even one additional cistern had the average capacity of 46 m³ (see also tsuk 1994:147), the annual water amount for each person is grown to at least 6.9 m³ = a daily amount of almost 19 litres. and, since the real capacity of the two known cisterns was larger than that mentioned above, the annual and daily water amount was even larger, and enough to fulfill all the daily needs. the oIl preSS the processing of agricultural crops was characteristic of non-urban monasteries, mainly rural monasteries of the sown regions but also some desert monasteries. the main agricultural/industrial branches in these monasteries were the cultivation of olives and vines and the production of oil and wine. these activities played a major role in the daily routine, at least during part of the year, and were one of the monks’ major sources of livelihood (Brenk 2004; Hirschfeld 1992:104, 106-107). Oil and/or wine presses, therefore, are features which can be found in almost every known rural 206 monastery in Palestine. the monastery of Khirbet es-Suyyagh included an oil press within its main built complex, a wine press just outside its walls, and at least one more wine press in its agricultural area. the oil press comprised one crushing system and two pressing systems of the lever and screw type. this pressing technique, which probably originated in italy, appeared in the levant in the Roman period, flourished during the Byzantine period (Frankel 1994:50) and continued until the late Umayyad or early >abbasid period. the main components of this system, which help to place them within the cultural and regional context of the discussed periods, are the screw weights. those found in the oil press of Khirbet esSuyyagh belong to two different types. the weight of the upper pressing system belongs to a subtype of Frankel’s 'Samaria' screw weight which has a central round socket and two dovetail mortises. the 'Samaria' weight is the most common type of screw weight in Palestine and abroad (mainly in italy), and is concentrated mainly in the central hill country and the central coastal plain (Frankel 1994:61; 1999:111-113, Map 19). the weight of the lower pressing system is more unusual. it has an internal dovetail mortise surrounded by a rectangular frame, in addition to four external dovetail mortises. it seems to represent a combination of two types that are included in Frankel’s typology – a subtype of the 'Samaria' weight with a central round socket and four external mortises (ibid.:112, Map 19), and a subtype of the 'Kasfa' screw weight with an internal dovetail mortise surrounded by a rectangular frame and two external mortises. Only a few examples of the 'Kasfa' weight were found so far, all from the vicinity of Jerusalem and the Judaean Shephelah (ibid.:114, Map 21). While the upper weight, of the 'Samaria' type, has relatively many parallels in rural monasteries as well as in civilian settlements of the late Roman period and later, the lower weight of the combined type is so far unique. a few parallels were found, however, for some features which existed in this weight, such as the rectangular frame that surrounding the central socket. all these parallels were found in southern Judaea. Chapter 10: KhIrbet-eS-Suyyagh In Context various remains which the excavators suggested belonged to a rural monastery were unearthed at îorvat Beth loya, southeast of Beth Guvrin. these included an oil press which was made up of two screw weights, one of which had an internal dovetail mortise surrounded by a rectangular frame, and two external dovetail mortises (Frankel, Patrich and tsafrir 1990: Figs. 9, 11). the surrounding rectangular frame has parallels in screw weights from Syria, therefore it was suggested that the weight from îorvat Beth loya was influenced by the Syrian weights. Based on the presence of this weight in a complex suspected as a rural monastery, and on the presence of another uncommon screw weight with parallels in Syria and asia Minor in the rural monastery of >ain el-Jedide (see Hamilton 1935), it was suggested that the design of these unusual weights was influenced by foreign monks who brought with them to Palestine the techniques they knew from their homelands (Frankel 1999:169, 178; Frankel, Patrich and tsafrir 1990:294-297). a screw weight identical to that from îorvat Beth loya was found also in another Christian site in Judaea (from the vicinity of Hebron) – Ramet el>amle (Mader 1918: Fig. 4:a). the concentration of these four weights in ecclesiastical/monastic sites in the vicinity of Jerusalem can maybe point to the presence of foreign Christians – probably monks – in this area during the late Byzantine period. another element attributed to the monastery’s oil press is the third, relatively small beam weight found embedded in a later >abbasid wall. it was suggested that this weight used in one of the lever and screw pressing systems in order to increase the pressure on the pressing base by hanging it on the wooden beam in front of the screw. at least five other published examples of Byzantine oil presses from Palestine which reflect a combination of screw and beam weights are known: îorvat Karkara in Upper Galilee (Frankel 1992:46-47, Figs. 9-13); Kafr Samir in the Carmel coast (Zemer 1999:65; J. Finkielsztejn, oral communication); the Byzantine settlement (monastery?) at lower Herodium (Kalman 2001:146); the monastery of Siyar el-Ghanam (Corbo 1955: Fot. 42); and the monastery of Deir Ghazali (avner 2000:29*:Fig. 3). another possible example is the monastery of Mevo Modi>im, were the relatively small measurements of the screw weight brought the excavators to assume that additional beam weights were needed during the pressing operation (eisenberg and Ovadiah 1998:8*). the conversion of one of the large beam weights of the Roman oil press into a pressing base of the lower pressing system in the monastery’s oil press has a parallel in Kafr Samir which reflects the reverse process. in one of the oil presses unearthed in this large village/town was a screw weight which represented three stages of use. Originally, this element was a pressing base. later, it was broken into two pieces which were used as beam weights. in the third stage, the two parts were bound together and functioned as a single screw weight (Yeivin and Finkielsztejn 1999:25*, Fig. 47; J. Finkielsztejn, oral communication). it has already been remarked that oil presses are a relatively common feature in many rural monasteries (see also table 10.1). Oil presses of the lever and screw pressing system (usually with 'Samaria' type screw weights) were found at the monasteries of Siyar el-Ghanam (Corbo 1955:36-38, tav. 63:31-35, Fig. 9, Fots. 37, 46), Bir el-Qutt (ibid.:118-119, Figs. 30:7, 32, Fot. 108), Deir Ghazali (in addition to elements belong to an older a direct pressing system; avner 2000:29*, Plan 2, Fig. 3), >ain el-Jedide (Hamilton 1935:113-115, Pl. 45:3, and see above), Khirbet edQawwara (Batz and Sharukh 2008:1691), Mevo Modi>im (eisenberg and Ovadiah 1998:6*-8*, Plan 1, Figs. 4, 6-9), îorvat Migdal (Dray 1994; Matthews, Neidinger and ayalon 1990:13-17, Figs. 15-17), Khirbet Umm îalasa (amit 1997:261, 263, Fig. 6) and el-Qa§r (Hirschfeld 1990:56, Fig. 66). Some oil presses (number and type not mentioned) were found scattered around the monastery of Ramot (arav, Di Segni and Kloner 1990:315). Poor remains of an oil press were found in the monastery near Naúal Kidron (Hervé and Zelinger 2006). two oil presses were found near the monastery of îorvat îani (type not mentioned; Dahari 2003:102). an oil press of the direct pressing technique was identified at Khirbet el-Quneitra (Hirschfeld 1985b:251, Fig. 3, Pl. 50:3), Khirbet el-Qa§r (Magen, Har-even 207 Itamar taxel and Sharukh 2008:1997) and Rujm Jerida (Magen, Peleg and Sharukh 2008:2024). Oil presses were found also in almost all of the complexes identified as rural monasteries in western Galilee – îorvat Bata (one oil press, type not mentioned; aviam 2004:198), Khirbet el-Shubeika (one oil press of a north-Palestinian lever and screw pressing type; avshalom-Gorni 2002: Figs. 3, 6), Khirbet el-Qu§eir (one oil press of a north-Palestinian lever and screw pressing type; Frankel 1992:58, Figs. 28-32), Khirbet Ma>ar (five oil presses, at least one is of the direct pressing technique; aviam 2004:198, Fig. 17.19), Khirbet Mujeidal (two or three oil presses, type unknown; ibid.:199) and îorvat Gov (one oil press, type unknown; ibid.:199). the wIne preSSeS Of the wine press built outside the northwestern corner of the complex, only the lower part of the collecting vat was preserved and that situated in the agricultural area south of the monastery was not excavated. in the latter, however, the treading floor and collecting vat could be identified. Both of these were probably of the simple type which was the most common in Palestine for thousands of years, including the Byzantine period (Frankel 1999:51). the (almost) square shape of the treading floor of the southern wine press is also the most common among wine presses in Palestine as is the square shape of this wine press’s collecting vat, and the rectangular shape of the collecting vat of the wine press built outside the monastery (ibid.:52-53). the two square and circular compartments identified beside the treading floor of the southern wine press are features known also from other wine presses, where usually a larger number of semicircular compartments was built/hewn along one or more wall of the treading floor (e.g. avshalom-Gorni, Frankel and Getzov 2008; Sidi, amit and >ad 2003). the function of the small side compartments has been dealt with by many scholars. Some of the suggestions were that the compartments were collecting vats for must extracted by natural pressure of the grapes put in them, storage vats 208 for different components added to the must or fermenting vats for the residue of the grapes in order to create an alcohol-rich beverage (avshalomGorni, Frankel and Getzov 2008:62-64; Sidi, amit and >ad 2003:261, 263, with references). the northern wine press was built beside (and maybe even partially above) a wine press from the late Roman/early Byzantine period. a case of continuous use of an early Roman wine press during the late Byzantine period was identified at the farmhouse site of Ras abu Ma>aruf, north of Jerusalem (Seligman 1999:144-148). another example comes from the nearby site of Khirbet Fattir where, it was suggested, the Byzantine wine press was built over the remains of an earlier wine press from the early Roman period (Strus 2003:81-83). the re-paving of the collecting vat of the northern wine press at Khirbet es-Suyyagh has at least two parallels from the vicinity of Jerusalem. at Bethphage, a late Byzantine wine press was paved three times, always with a coarse white mosaic (Saller and testa 1961:30). three paving stages were identified also in a late Byzantine wine press from the valley of the Cross. the three floors of the treading surface were all made of white mosaic. the collecting vat, however, first had a levelled bedrock floor, then a plaster floor and finally a coarse white mosaic floor (Rahmani 1991:100-101). Wine presses, like oil presses, are a frequent feature in Palestinian rural monasteries (see also table 10.1). Complex and/or simple wine presses were found at various distances outside many of the rural monasteries, but seldom inside them. this is seen at Siyar el-Ghanam (Corbo 1955:47-49, tav. 62:63, 64, Fig. 11), Bir el-Qutt (ibid.:118, 128, Figs. 30:6, 31, 38, Fot. 106), Khirbet abu Ghunneim (Kloner 2000:98*), Deir Ghazali (avner 2000:32*-34*, 36*, Plan 2, Figs. 11-13, 17), Ras ettawil (Gibson 1985-6:72, Fig. 1:4), Ramot (arav, Di Segni and Kloner 1990:315), Naúal Kidron (Kloner 2000:88*), ein el-Sachaniah (Sar-avi 1999:188), Khirbet abu Rish (Magen and Baruch 1997:140-141, Figs. 7, 8), Khirbet Umm Deimnah (Magen and Batz 2008:2058), Mevo Modi>im (eisenberg and Chapter 10: KhIrbet-eS-Suyyagh In Context Ovadiah 1998:9*-10*, Plans 2-3, Fig. 13), Deir Qal>a (Magen and aizik 2008:1694), Khirbet Deir Sam>an (Dar 1986:29), Deir Daqla (Dar 1986:149, Fig. 91, Pl. 71), îorvat Hani (Dahari 2003:102), Khirbet Umm îalasa (amit 1997:265-267, Figs. 8-10), Rujm Jerida (Magen, Peleg and Sharukh 2008), el-Qa§r (Hirschfeld 1990:56, Fig. 66), Khirbet el-Bureikut, (Hirschfeld 1992:107, Fig. 47), Khirbet Umm Rukba (Hirschfeld 1990:15, Fig. 15; 1992:108-109, Fig. 49), Beth ha-Shittah (aharoni 1954:210-211, Fig. 1, Pl. 7:1) and îorvat Medav (in the western Galilee; aviam 2004:200). ECONOMIC ASPECTS the data regarding to the economy of the rural monastery of Khirbet es-Suyyagh can find parallels in other monasteries as well as in contemporary civil rural settlements. the livelihood of monasteries in Byzantine Palestine leaned on two main factors – donations and sale of various household objects and agricultural surpluses which were produced in the monasteries. as noted by Hirschfeld, donations were the most prominent source of livelihood among the Judaean desert monasteries. these donations, mostly in cash, could have been given by pilgrims and visitors, monks who joined the monastery, the local ecclesiastical establishment or the imperial court. this money was used mainly for the initial construction of the monastery and for renovation and building activities throughout its existence (Hirschfeld 1992:102-104). Donations of lands and other agricultural property were especially important in the case of rural monasteries. the sources tell us about Gelasios, a 5th-century monk who inherited agricultural lands from an old monk in the vicinity of emmaus/Nicopolis and established there a monastery (Apophthegmata Pratum 177). according to Brenk, wealthy patrons played a central role in the financing of monasticism, mainly by giving steady or disposable money donations. in his opinion, that was true mainly in regarding to monasteries which were situated in arid regions (such as the Judaean desert); in especially large monasteries; and in those which reflect a lavish monumental architecture. these monasteries could also have owned agricultural lands, but the income from this source was not sufficient to achieve their desired life-style (2004:449-454, 472-473). Nevertheless, in true rural monasteries the situation was probably different, as explained by Brenk in regarding to the example of Bir elQutt: "i am presuming that only monasteries that were financially poorly-endowed engaged in agricultural production and kept animals actually inside the monastery walls. Bleating goats and spiritual exercises side by side cannot have been inspiring. But at Bir el-Qutt it is certain that the monastery owned pastures, along with olive trees and vineyards and the upkeep of the site was financed in large part by the sale of the produce" (2004:469). it seems that the case of Khirbet es-Suyyagh was not much different from that described above. this small rural monastery reflects a very modest, provincial building. it is indeed a planned complex which was without a doubt built by a skilled labor force. Yet the almost completely lack of splendour and adornments, and the almost exclusive use of local building materials suggest that this monastery did not enjoy the financing of a rich patron. it is possible of course that the money for the foundation of the monastery was given by a private donor or by the local ecclesiastical authorities, such as the patriarchate of Jerusalem or the bishopric of Beth Guvrin. Some of the money could have been brought by the founder monks themselves, and/ or by monks who joined the monastery later. it must be also remembered that the monastery was situated beside a secondary though still important road and was certainly visited by pilgrims from whom the monastery may have received donations from time to time. initial funding enabled the monks to build their monastery and to acquire all they needed to maintain a religious agricultural community. apart from various household vessels and implements, agricultural tools, animals, etc., the monastery’s church had relatively rich marble bema furniture and related liturgical artefacts, such as altar/offering tables and reliquary chest/s. these heavy marble pieces, which must have been transported from one of harbour cities situated along the central or southern coast, were maybe the 209 Itamar taxel most expensive items required by or donated to the monastery. However, donations were probably of secondary importance in the monastery’s economy compared to the income from agricultural products. We have no evidence that the local monks manufactured and sold baskets, mats, ropes or any other household artefacts, as did monks in the Judaean desert monasteries (Hirschfeld 1992:104), although the absence of such activity from the archaeological record does not mean that it did not take place. On the other hand, the remains of an oil press and two wine presses within its precincts and the relatively varied finds of domestic animals bones clearly indicate that this monastic community leaned on agriculture as its main source of livelihood. the oil press at Khirbet es-Suyyagh had a major role in the monastery’s economy, as a producer of what seems to be a considerable amount of oil (due to the use of the screw pressing technique in the monastery’s oil press; see Decker 2007:85-86). Calculation of the area of olive groves whose crop this oil press processed, is based on Ben David’s conclusions about oil production in the Golan in the late Roman and Byzantine periods. according to Ben David, the average ratio between the area of olive groves and one oil press (apparently with one pressing system) was 345 dunams, and the minimal ratio was 260 dunams (1998:50). thus, an oil press with two pressing systems, like that at Khirbet es-Suyyagh, could process the crop from 420-700 dunams of olive groves, or even more. Calculation of the amount of crop per dunam of these olive groves, relies on Zinger’s data about modern olive groves in the Galilee. according to Zinger, the olive crop in eastern Galilee, where the annual amount of precipitation is 400 mm, is 100 kg per dunam, and the number of trees per dunam is 10 (1985:13). the annual amount of precipitation in the northern Judaean Shephelah is also 400 mm. thus, it can be assumed that similar quantities of olives were produced also in this region. if the number of trees per dunam was larger – 10-15, as suggested by Frankel (1994:124), or 15-25, as suggested by Dar (1986:186) – the crop would also have been larger. even taking the minimal number of ten trees 210 per dunam, we find that 42,000-70,000 kg of olives could have been picked in the olive groves of Khirbet es-Suyyagh. it is accepted today that traditional oil presses can produce only 20% of oil from 1 kg of olives (Ben David 1998:51). thus, the amount of oil that could have been produced at Khirbet es-Suyyagh was 8400-14,000 kg, or 9240-15,400 litres. this amount is more than eight to 13 times larger than the maximal total capacity of the two collecting vats of the monastery’s oil press, which is 1.13 m³ (1130 litres). the olive-picking season lasted one or two months between October and December, depending in the size of the oil press, the amount of crop and the number of available workers. Due to the short picking season, some scholars believes that oil presses were often worked day and night in order to finish the initial pressing when the olives were still fresh (avitsur 1994:103-104; Frankel 1994:23). On the contrary, Safrai does not thinks that ancient oil presses worked also at nights, but only about 14 hours a day or ca. 80 hours in a week of ca. 5.5 working days (1994:124). Dar calculated that the weekly oil amount which was produced in an oil press (apparently with one pressing system) that worked 136 hours a week (ca. 5.5 days) is 1360 kg (1496 litres, or 272 litres per day) (1986:184). according to avitsur, in a working day of 20-24 hours, 300-400 kg (330-440 litres) of oil could have been produced by a given pressing system. thus, during the picking season 10-15 tons of oil could have been produced, depending in the region, the amount of crop and its quality (1994:110, 124). Safrai calculated that the weekly oil production of a given oil press (with one pressing system) was 400-1000 litres, and that the total production during a ten-week season was 4000-10,000 litres. He also claimed that the average capacity of collecting vats in ancient oil presses is 40-50 litres, of which only 25% was oil (1994:124). the numbers presented by Safrai seem to be too low to fit the potential oil production of Khirbet es-Suyyagh, and, of course, the capacity of collecting vats mentioned by him is much lower than that reflected from the vats in Khirbet es-Suyyagh as well as from other sites! as to the daily number of working hours of a given Chapter 10: KhIrbet-eS-Suyyagh In Context oil press, here we must admit that we cannot say for sure whether indeed oil presses worked around the clock, and if so – for how many days. We can assume, however, that during at least part of the season the oil press of Khirbet es-Suyyagh worked also at night, as can be presumed from the large number of lamps found in it. the certain details of the oil press from Khirbet es-Suyyagh, thus, indicated that the weekly and total amount of oil produced in it was larger than that suggested by Safrai, but not necessarily as high as suggested by Frankel and Dar. the data given by Dar and avitsur about the daily output of oil of one pressing system (i.e., between 270 and 440 litres per day) are closest to those reflected from Khirbet es-Suyyagh. if we take the average amount of 355 litres per day their data, we can calculate that none of the collective vats in the oil press at Khirbet es-Suyyagh was completely filled in the end of a given day. the upper one with a volume of 570 litres, and the lower one with a volume of 660 litres could have been filled within 1.6 and 1.8 days, respectively. We cannot know, however, if the vats were indeed filled up to their rim before emptying, or whether they were filled only for ¾ or so of their volume. if we still take a total amount of 1000 litres (little bit less than the maximum capacity of 1130 litres of the two vats), we can calculate that in a week of 5.5 full working days (according to Dar) this amount could have been obtained ca. 3 times. thus, the weekly amount of oil produced in the oil press was ca. 3000 litres. if we take the minimal potential amount of oil (9240 litres) which could have produced from the olive groves of Khirbet es-Suyyagh, we can find that the monastery’s oil press worked for 3 weeks during the picking season, and if we take the maximal potential amount (15,400 litres) we can find that the oil press worked for five weeks. if the monastery’s olive groves produced indeed the maximal potential amount, but the oil press worked less than ca. 136 hours a week, it was maybe active during more than five weeks. On the other hand, in years blessed with high amount of crop, part of it could have been sold. according to Dar, the annual consumption of olive oil among the arab rural population of the Samaria hill country in the 1970’s was 15-25 litres per capita, which were used only (or almost only) for food. in his opinion, an annual quantity of 17-20 litres per capita is not unrealistic, and was probably common also in the late Roman and Byzantine periods (1982:309-310). However, while in modern times the almost only use of olive oil is for food, in antiquity oil was also widely used for lighting (Frankel 1994:93). Dar found that a typical oil lamp of the Byzantine period consume 0.025 litre of oil in two hours of lighting. if a lamp worked only two hours every night, it consumed an annual oil amount of nine litres. However, since it is reasonable that every household contained at least two lamps which worked more than two hours every night, we can assume that every household consumed an annual amount of ca. 20 litres of oil for lighting. therefore, a family of seven persons consumed an annual amount of 140-160 litres of oil for food and lighting (Dar 1982:310, n. 78). if we take the approximated number of 20 monks as the monastery’s community and the estimated annual oil consumption for food of 17-20 litres per capita, we can calculate that the monastery’s annual oil consumption for food was 340-400 litres. the annual amount of oil for lighting in the monastery, however, is somewhat harder to calculate. We can assume that the number of oil lamps used every evening/night was equal to the number of the inhabitants, and probably even larger, i.e., at least 20. in addition, the church also needed to be lit in the evening and sometimes all night, mainly by glass lamps which seem to consume a larger amount of oil than the ceramic lamps. the monastery’s annual oil consumption for lighting can be roughly estimated at 300 litres, and together with consumption as food it reached ca. 700 litres. to this we have to add the quantity of oil consumed by the variable number of pilgrims, guests and daily workers who stayed in the monastery during the year. However, even if a larger quantity (up to 1000 litres) was kept in the monastery for times of shortage and for sale, a possible surplus of ca. 8000-14,000 litres still remains! 211 Itamar taxel Other uses of the olives processed in the monastery’s oil press regard to the pressing wastes. these wastes, and mainly the solid ones (i.e., olive flash and pits), could have been used as fuel in domestic contexts (cooking, heating), as animal food, as fertilizer, and as construction material (when mixed in mortar/plaster) (Warnock 2007:45-52, 59-62). the production of wine in the monastery was probably on an even larger scale than that of oil. the data regarding the capacity of the collecting vats of the two wine presses which were attributed to the monastery is only partial (the vat of the northerwestern wine press was not fully preserved, and the southern wine press was documented only in survey). Nevertheless, the estimated capacity of the first collecting vat (ca. 5 m³) and the minimal capacity of the latter collecting vat (3 m³) give us at least a partial idea about the production ability of the two wine presses (at least 8000 litres). according to Gal, the amount of juice produced in antiquity from one ton of grapes was ca. 500 litres. He also estimated that 300-400 kg of grapes can be trodden at one time in an area of 12 m² (1986:137). if so, ca. 700-1000 kg of grapes could have been trodden on the treading floor of the southern wine press of Khirbet es-Suyyagh, and at least six treading cycles were needed in order to fill its collecting vat. if we assume that the treading floor of the northwestern wine press of Khirbet es-Suyyagh had a similar area to that of the southern wine press, it can be calculated that at least 10 treading cycles were needed in order to fill its collecting vat. Based on the output of recent arab vineyards, Gal assumed that vineyards in the Byzantine period yielded ca. 200 kg of grapes (or 100 litres of juice) for one dunam (1986:138). the data given by Dar regarding the arab vineyards during the British Mandate period are totally different. according to him, the local species of vines produced at that time 750-1000 kg grapes for one dunam, and in the Hebron Mountain there were vineyards which produced even 3000-5000 kg of grapes for one dunam (1986:154). Dar, however, did not compare the modern arab vineyards and those of antiquity, so we will not use his data here. if we take Gal’s 212 data about vineyard production, we can calculate that the filling of the two collecting vats of the wine presses at Khirbet es-Suyyagh required at least 16 tons of grapes, which were grown in 80 dunams of vineyards. and if these collecting vats were filled more than once during the harvest season, the required amount of grapes and the area of vineyards were at least twice larger. the harvest season lasted for four to six weeks between June and September, depending in the species of vine, the geographic region and the nature of treatment of the vines (Frankel and ayalon 1988:12). in Dar’s opinion, wine presses in antiquity were active for 50 days during the harvest season. the first fermentation of juice, which lasted a few days, occurred in the collecting vats. afterwards, the juice was sieved and removed to jars which were stored in cool and dark places (such as basements), for the second fermentation. after a period of six months to one year the wine was ready, and could have been used and sold (Dar 1986:154-157). Some scholars, such as Dar (1986:161) and Safrai (1994:129, 131-132) tried to calculate the annual wine consumption in late RomanByzantine Palestine. their calculations were based on a rabbinical law which says that a potter may sell 15 wine jars to one person (Mishnah, Shevi>it 5, 7), and on the assumption that the average volume of these jars is 20-25 litres. according to this data, they suggested that the wine consumption for a family was 330-375 litres. However, as noted by Kingsley, it does not seems that the rabbinical source refers to real consumption levels, and the number of people who bought the 15 wine jars is also unknown. in addition, different types of wine had different levels of strength and therefore they were diluted with water to varying extents. Moreover, different levels in the population and different ethnic or religious groups (Jews, Christians and Samaritans) also consumed different quantities of wine (2002:70-71). therefore, the quantity of wine consumed by the inhabitants of the monastery of Khirbet esSuyyagh cannot be estimated, especially when dealing with monks. Chapter 10: KhIrbet-eS-Suyyagh In Context Wine was forbidden in Syrian monasteries, although it was consumed by some, and not only in liturgy (abouzayd 2005:136-141; vööbus 1960:264). in egyptian monasteries the drinking of undiluted wine by monks was absolutely forbidden. the drinking of diluted wine was common in these monasteries, though probably in small quantities, since it was served only to sick and weak monks and in the liturgy. Based on this data, Dembińska assumed that the vineyards in egyptian monasteries were very small and that most of the produced wine was sold or exchanged in markets (1985:442). that the area of vineyards in the monastery of Khirbet es-Suyyagh was quite large is indicated by the existence of the two wine presses and by the volume of their vats, and it is certain that the lion’s share of the wine produced in them was sold. Nevertheless, we still cannot know if the local monks abstained from drinking wine, as did egyptian monks, or if they used wine more frequently in their daily life. the importance of wine production in Palestinian monasteries is highlighted by a Greek inscription dated to 529 Ce found near tel ashdod. the inscription mentions the construction of a wine press and monastery by a certain abbot, and according to Di Segni, it most probably adorned the entrance of that monastery’s wine press. Di Segni rightly assumed that the sale of surplus wine was probably an important source of income for that monastery (2008:32*-33*). Some of the monastery’s production of wine and oil was most probably collected as tax. ecclesiastical property was taxed (Dauphin and Kingsley 2003:66), although as noted by Kingsley there is no historical or archaeological source which can provide us details about the level of taxation in the Byzantine period. Still, he suggested that the taxes were not particularly high (1999:182). the manpower needed for the production of oil and wine in the monastery of Khirbet es-Suyyagh is another important issue. according to avitsur, every shift of 10-12 hours in an oil press with one pressing system required two-three workers (1994:105). thus, it seems that ca. five people worked in every shift in the oil press of Khirbet esSuyyagh, and ca. ten different people worked there in the peak season when the oil press functioned around the clock. theoretically, we could assume that about half of the monastery’s community worked in the oil press itself during the four weeks or more of the olive-picking season. However, before and during the pressing activity the olives themselves should have been picked, and this work required much more manpower that does not seem to be fulfilled only by the monastery’s inhabitants. the case of the wine presses does not seems to be different. the two wine presses of the monastery seem to required three-four workers each, and if they functioned simultaneously so up to eight people had to work in wine production in 50 days or so during and after the harvest season. as in the case of olive picking, here too the grapes had to be harvested, partially before the working time of the wine presses and partially during it. therefore, additional workers were needed to carry out both operations. Fortunately, the grape harvesting and the olive-picking seasons did not overlap but roughly followed each other. Nevertheless, the fulfillment of these two hard tasks, which occupies about one quarter of the year, must have been assisted by additional manpower apart from the monastery’s inhabitants. Certain legislations in the Novellae of Justinian, such as Novella 7 of 535 Ce, mention that ecclesiastical property included, inter alia, also agricultural slaves. However, ecclesiastical (in this case – monastic) agricultural lands could also be rented in a long-term contract called Emphytheosis. in such a contract the ownership of the land was transferred to tenants, with a time limitation of three generations, and in any case the contract had to be renewed every 25-29 years. after three generations the land was returned to its ecclesiastical owner, and could be rented again if the owner wished to do so (Dauphin and Kingsley 2003:66-67, and references therein). it was suggested that the ecclesiastical complex excavated at Shelomi in western Galilee was a monastic farm which was cultivated by tenants through such a contract (ibid.:67). already before the excavations at Shelomi, Dan suggested that some local monasteries had considerable land holdings which were cultivated (at least partially) 213 Itamar taxel by tenants or lessees, although we have no certain data as such in the case of Palestinian monasteries (1977:186-188, and references therein; see also Di Segni 2001:34). i suggest, therefore, that farmers from the neighbouring villages and/or tenants were employed by the monastery of Khirbet es-Suyyagh at least during the seasons of the grape harvest and olive-picking. it it also possible that some of the monastery’s lands were leased by farmers, who also used the monastery’s agricultural installations. For instance, we could assume that the northwestern wine press was managed by the monks, while the more distant wine press in the south was used by lessees. Similarly, it is possible that the work in the oil press was divided between the monks and the lessees, each group processing its own crop. Similar conclusions were recently reached also by Hull (2008:102, 105) in regard to rural monasteries in northern Syria, whose main source of livelihood was olive oil production. according to Hull, “a portion of the olive harvest was awarded as tribute to monasteries, or alternatively that the olives were grown within the monastic domain and that harvesting and processing labour was provided in lieu of rent” (ibid.:105). Hull’s second alternative seems more acceptable, at least in the case of the solitary Palestinian rural monasteries, such as Khirbet es-Suyyagh. We cannot know if salaried or lessee farmers worked in the monastery’s lands also outside the season of wine and oil production, because the extent of these lands and the agricultural crops grown on them are unknown. if, apart from vineyards and olive groves, the monks had only small plots of vegetables and fruit trees, it does not seem necessary to cultivate them together with laymen workers. However, if the monastic lands included also some grain plots, additional manpower would have been needed for their cultivation. We should not forget that the monastery had also an unknown (but probably not large) amount of domestic animals which had to be taken care of, and – of course – that the manual labor occupied only part of the monks’ daily routine. Hard work was indeed an integral and welcome part of monastic life, but 214 prayers (up to seven every day), meals and rest occupied also parts of the day (Hirschfeld 1992:7981). thus, it seems reasonable that a small group of laymen farmers (tenants?) worked regularly in the monastery’s lands, and maybe even lived in the area around the monastery. according to Hull (2008:105-106), such economic relationships must have created stronger social bonds between the neighbouring rural secular population and the monastery. this may also explain the existence of a basilical church in the monastery of Khirbet es-suyyagh, whose community numbered only 20 monks or so. the faunal remains from the site and mainly from loci 330 and 358, which are attributed to Phase ii (Chapter 8), indicate that sheep/goats, cattle and pigs were raised in the monastery. this distribution of species, especially the presence of pigs, is similar to other Christian Byzantine sites (Pella: Rielly 1993: table 1; Hesban: von den Dreisch and Boessneck 1995:71-73, table 5.10; Caesarea: Cope 1999:407, table 1; tel >ira: Dayan 1999; Kafr Samir and Dor: Kolska Horwitz 2006:840). the total number of identified individual animals is rather low, a fact that can derived from the relatively small quantity of faunal remains in the excavation but also from the nature of the site’s economy. animals were kept in monasteries mainly for their by-products (milk, wool and manure), as well as for draught and ploughing (talbot 2002:39, 52-53). Meat was consumed only rarely (Dembińska 1985:442; talbot 2002:52), and in Syrian monasteries it was absolutely forbidden and consumed only by few (abouzayd 2005:136-141). Rules regarding the latter permitted only one donkey and one ox to be kept in every monastery (vööbus 1960:260, 262), although the archaeological evidence show that in reality some Syrian monasteries had stables in which quite a large number of animals could have been housed (Brenk 2004:463-464, 467). the property of Gelasios’s monastery from the vicinity of emmaus/Nicopolis included not only many lands with olive groves, but also cattle and draught animals (Apophthegmata Pratum 177, 180; see also Dan 1977:185; vailhé 1899:537). Chapter 10: KhIrbet-eS-Suyyagh In Context the sheep/goats and cattle whose remains were found in the monastery of Khirbet es-Suyyagh were apparently kept mostly for their various by-products and for labour, as indicated by the absence of cut marks on their bones (contrary to the situation in locus 175 which is attributed to Phase iii, and see Chapter 8). On the other hand, three of the pig bones from the same assemblage bear cut marks, clearly proving that pigs were eaten. the sources tell us that the monastery of theodosius in the Judaean desert had a pig farm at Phasaelis in the Jordan valley (John Moschos, Pratum Spirituale 92), and it is interesting to note that pigs were raised at least until recently in the monastery of Deir îajla near Jericho (personal observation, 1994). the raising of cattle (or other animals) in the monastery/monastic farm of Shelomi is apparently indicated by the finding of fodder remains at that site (Dauphin and Kingsley 2003:65). after setting aside the quantities of oil, wine and other agricultural products needed for homeconsumption and for random selling, and after the deduction of unknown amounts of taxed quantities, considerable surpluses remained in the hands of the monks of the monastery of Khirbet es-Suyyagh. Part of the surpluses, mainly of oil and wine, was certainly given to the tenants or independent farmers who helped with the hard work. the majority of the surplus was most probably sold and provided the main source of income for the monastery’s inhabitants, and one of the main factors in the establishment of its status in the vicinity (Zeisel 1975:273-275, 281-284). the products could have been sold in the monastery itself to merchants, and/or in markets or fairs in the neighbouring villages and even in the city of Beth Guvrin. Despite a ruling given by Rabbula, Bishop of edessa (411-435 Ce), which prohibit greediness and making a profit from selling the monastery’s products, Brenk believes that this rule was taken only as a recommendation and that monks sold their surpluses at market prices and made profits when possible in order to maintain their lifestyle (2004:459). the surpluses of wine and oil were stored and transported in large bag-shaped jars of the types characteristic of Judaea which were found in large quantities at the site (see Chapter 3, late Byzantine and Umayyad pottery, jars of types 1-4). there is not enough evidence about the export of Palestinian oil in maritime trading. according to Ben David, the large surpluses of oil from the central and southern Golan were exported to inland regions in the north and east (1998:53). However, aviam believes that some of the surplus from the western Galilee was exported (1994b:35), and it is very possible that this was by sea since the coast was accessible. local wines were one of the most famous and important products exported from Palestine in the Byzantine period. this is apparent from the large numbers of Palestinian wine jars, i.e., bagshaped jars and 'Gaza/ashkelon amphorae' which were found in shipwrecks, harbour deposits and numerous coastal and inland sites throughout the Mediterranean and beyond (Decker 2007:69-72; Kingsley 2002:74-84; Mayerson 1992, and references therein). the great majority of Palestinian jars found abroad represented coastal types of bag-shaped jars and 'Gaza/ashkelon amphorae' (Kingsley 2002:74-80). inland types of bag-shaped jars, including the Judaean types common at Khirbet es-Suyyagh, are rarely found in Byzantine coastal sites, ship wrecks and harbour deposits in Palestine (e.g. Haddad 2009: Figs. 2:13, 14, 6:10, 11; Kingsley and Raveh 1996: Fig. 36:P41; Riley 1975:28, No. 3) or other Mediterranean countries (e.g. Bonifay and Piéri 1995: Fig. 10:69; Rautman 2003: Fig. 3.21:i-31-1; Riley 1981: Fig. 8:72; tomber 1988: Fig. 26:500). this fact perhaps indicates that wine from inland regions in Palestine, including the central hill country, was exported abroad in much smaller quantities than wine from the coast. the distribution of the wine and oil manufactured in the monastery of Khirbet es-Suyyagh was concentrated, therefore, in the relatively close vicinity. Jerusalem and Beth Guvrin seem to represents the main and maybe the most distant markets which these products reached. Nevertheless, we cannot rule out the possibility that some of the wine and/or oil from Khirbet es-Suyyagh did reach the coastal plain and was maybe even exported or taken abroad by pilgrims 215 Itamar taxel who puchased these commodities at the monastery. there was a great demand for ‘holy’ wine and oil, especially those produced by monks, and many monasteries in the levant took advantage of this demand in order to maintain themselves (Kingsley and Decker 2001:10; Majcherek 2004:235). the restricted economic contacts of the monastery with the coastal plain are evidenced by the very small number of coastal bag-shaped jars, 'Gaza/ashkelon amphorae' or imported northeastern Mediterranean amphorae found at the site (for a similar situation at nearby Khirbet el-Jiljil, see Mlynarczyk 2005:154). this further proves that the monastery was self-sufficient, and that the wine and oil consumption of its inhabitants was based primarily on local manufacture. the coastal and foreign commercial containers could have been brought to the site also as recycled vessels – either empty or filled with commodities other than their original ones.1 a similar picture of local versus imported amphorae/jars is reflected from rural and urban sites throughout Palestine (Kingsley 1999:180, table 10). in Kingsley’s opinion, the relative scarcity of 'Gaza/ashkelon amphorae' in northern and inland sites in Palestine can be explained either by the self-sufficiency of these regions in wine of similar quality, or that the wine from the southern coast was too expensive to be acquired in large quantities (1999:182). the main imported, and presumably the most expensive commodity which was bought by the monastery with the money earned from selling wine and oil was probably the fine lRRW table ware (see also Kingsley 1999:182-193). the imported bowls from North africa, Cyprus, egypt and mainly western asia Minor were most probably acquired in the large markets of Beth Guvrin or Jerusalem, by the monastery’s economist (;  ) or its agents (on the role of the economist, see Hirschfeld 1992:73-74; Meimaris 1986:256-259). 1. John Moschos tells a story from a certain coenobium, about one of the monks who stole the shawl of another monk and hide it inside a Gaza amphora () which was in his cell (Pratum Sprituale Supplementary tales 9). 216 the local markets were also the places were the locally-produced pottery – cooking vessels, storage jars, FBW bowls and jugs and all the rest of the varied ceramic vessels used in the monastery – was bought. the dominance of imported western asia Minor (lRC/PRSW) bowls in the site (see above, Chapter 3) matches the situation in the rest of the country, including the Jerusalem region. On the other hand, the small number of aRSW bowls in the site reflects an opposite picture to that presented by tsuf, according to which these bowls are quite common in Jerusalem and its vicinity (2003:xiii). the relatively high amount of eRSW bowls in the site is remarkable when looking on their general distribution in Palestine, where they were less common mostly in inland sites compared to the coastal ones. However, the situation from Khirbet es-Suyyagh seems to characterize also other sites in Judaea and southern Palestine, in which eRSW bowls are relatively common (ibid.:xiii, 87-88, Graphs 5, 6). there is no clear evidence that the wine presses at Khirbet es-Suyyagh ceased to function after the Muslim conquest, while the monastery continued to exist. it is now quite clear that during the Umayyad period wine production in Palestine and adjacent regions was gradually reduced, until its complete disappearance in the late Umayyad or early >abbasid period (ayalon 1997; Watson 2004:501). according to Kingsley, the wine industry of the southern coastal plain was the first to be ceased, and by the end of the 7th century it disappeared also from central and northern Palestine (1999:193). Nevertheless, the processes of islamisation and arabisation, which were among the factors which led to the cessation of wine production, accelerated and started to be felt only from the 8th century onwards. throughout the 7th century and even later the majority of the population in the central hill country, and especially in Judaea, remain nonMuslim (amar 1995:258; levy-Rubin 2000:271). the southern wine press was not excavated, and the few pottery sherds found in the earth fill which sealed the northwestern wine press cannot give a more accurate date than the late 6th-7th/8th centuries to its destruction/abandonment. Still, at Chapter 10: KhIrbet-eS-Suyyagh In Context least in regarding to the northwestern wine press it does not seem very reasonable that it ceased to function immediately after the Muslim conquest in the 630’s. there are relatively many examples from rural sites in Palestine for wine presses which continued to function into the second half of the 7th century or even later. examples of wine presses in rural monasteries which continued to exist after the Muslim conquest are those from îorvat Migdal (Matthews, Neidinger and ayalon 1990:21), Siyagh el-Ghanam (Corbo 1955:47-49), Bir el-Qutt (ibid.:118, 128), Ras et-tawil (Gibson 1985-6:72) and Ramot (arav, Di Segni and Kloner 1990:319-320). the presence of pigs in the monastery indicates that these animals probably continued to be raised after the Muslim conquest. in its early days, islam did not legislate against the continuation of pig raising by Christians (Fatal 1958:151-155). THE REGIONAl CONTEXT Palestine reached a peak in population and settlement density during the Byzantine period (Broshi 1980:5, 7; tsafrir 1996:270) although this was not true for all regions (Fiema 2006). the northern Shephelah was no exception, as demonstrated by surveys and excavations conducted in this region which revealed finds and remains from this period at numerous sites. in the survey of Ramat Beth Shemesh (an area extending from Naúal Yarmut in the south to the eastern fringes of modern Beth Shemesh in the north), 392 sites dated to the Byzantine period were identified. these included remains of settlements of various size and nature (a town, villages, farmhouses, scattered structures, single structures), burial sites, clusters of installations, agricultural terraces and spots of finds (mainly pottery sherds) (Dagan, forthcoming). in the survey of the Nes Harim map, extended southeast to Khirbet es-Suyyagh and overlapping part of the Ramat Beth Shemesh survey, 94 sites that yielded finds and/or remains from the Byzantine period were discovered. Most of these sites were identified as villages or farmsteads, and four to seven of them were excavated or surveyed churches or related remains (Weiss, Zissu and Solimany 2004:15*). two of these sites include the remains of what seem to be monasteries. Khirbet Nabhan, 5 km northeast to Khirbet es-Suyyagh, composed of several ashlars-built structures, including a large rectangular building (30×50 m) that was identified as a monastery, and another building with remains of an apse, which was identified as a probable church (ibid.:29*). Khirbet Deir abu >ali, which placed 4 km southeast to Khirbet es-Suyyagh, is a large square complex (100×100 m) surrounded by ashlars-built walls. Within its area were seen various architectural elements, such as thresholds, lintels, pilasters and columns and remains of an oil press, and outside it was identified a hewn wine press (ibid.:51*). although the surveyors did not suggest to identify Khirbet Deir abu >ali as a monastery, its nature and finds can indicate that this site was a large rural monastery. Four more distant sites in the map of Nes Harim (placed within 7-10 km east or south of Khirbet es-Suyyagh) – Matta>, Khirbet Jurish, Khirbet elîammam, Khirbet el-Jurfa/Rogelit and Khirbet Umm er-Rus, also yielded evidence for Christian presence in the Byzantine period. at Matta> was unearthed a chapel dated to the 6th century (Ovadiah, Ovadiah and Gudovitch 1976). Khirbet Jurish, which was quite a large village, includes some burial caves, one of which is decorated with a curved cross (Weiss, Zissu and Solimany 2004:55*). at Khirbet el-Jurfa/Rogelit, which was another village, was excavated a basilical church dated to the 6th century (Gophna and Zvilichovsky 1959; Weiss, Zissu and Solimany 2004:71*). at Khirbet el-îammam, which also seems to be a village, some marble fragments were found, including one that the surveyors identified as part of a chancel screen (Weiss, Zissu and Solimany 2004:73*, Fig. 236.2:6), although it looks more like an altar table. at any rate, the marble finds are clear evidence to the presence of a church at the site in the Byzantine period. at Khirbet Umm er-Rus was unearthed a small basilical church dedicated to St. John, which was part of a larger complex (a monastery?) (Macalister 1889; Ovadiah 1970:127-128, Pl. 54; Weiss, Zissu and Solimany 2004:67*). 217 Itamar taxel the close vicinity of Khirbet es-Suyyagh from its northwest and southwest was also densely settled during the Byzantine period. evidence that many of the sites identified in this area were inhabitated by Christians was found. at Deir Rafat architectural elements decorated with crosses were found as well as burial caves which contained finds with Christian relation (Bagatti 1990). at Deir et-taúuneh three carved lintels, two of them decorated with crosses, were found. the site was identified as a monastery (Conder and Kitchener 1883:92-93). at tel Beth Shemesh a large building and architectural elements, some decorated with crosses, were unearthed. this complex was also identified as a monastery (Mackenzie 1911:75-84), though this identification does not seems to be true. at tell Zakariye/tel >azeqa incisions of Greek inscriptions and crosses were discovered on some hewn features (Bliss and Macalister 1902:217-223). at Khirbet Fattir parts of a Byzantine and Umayyad village, including a church and related finds, were excavated (Strus 2003:151-215). at Beit Jimal a Byzantine church was unearthed (ibid.: 481-492) and at Khirbet el-Jiljil remains of Christian-occupied farmhouse were surveyed and partially excavated (ibid.:31-39; Di Segni and Gibson 2007; Strus and Gibson 2005). Remains of a possible monastery, including mosaics, were surveyed and partially excavated at Deir el->asfura (Strus 2003:44-48; Waliszewski 1994). at îorvat >illin fragments of marble elements, originally belonging to a church, were found in contexts from the early islamic period (Greenhut 2004). at Khirbet en-Nebi Bulus remains and finds which were attributed to a Byzantine monastery were discovered (Bagatti 2002:136-137; lehmann 1996; Rapuano and Yas 1996:89). Near îorvat Socho remains of a building, maybe a monastery which included a chapel, were unearthed (Gudovitch 1996). at Khirbet Jannaba et-taúta scattered remains were excavated, including those of a church (avner 1995; Dauphin 1991). at îorvat îanot remains of a church, which was probably part of a hostel maintained by a monastic community, were excavated (Di Segni 2003; Shenhav 2003). 218 Northeast of Khirbet es-Suyyagh, at least one Byzantine site of Christian nature is known – the cave site of >irak isma>in in the ravine of Naúal Soreq (3 km away). this site, which includes rock-cut and built remains (some belong to a chapel) inside and outside a large natural cave, was identified by Gass and Zissu (2005) as the Monastery of Samson, although in my opinion this identification is insecure. dISCUSSION in the late Byzantine period Khirbet es-Suyyagh and its vicinity continued to be part of the territory of Beth Guvrin/eleutheropolis. the many Byzantine sites in the Judaean Shephelah demonstrate that it was a typical Christian region (Fig. 10.1). the urban centres closest to Khirbet es-Suyyagh were emmaus in the north and Beth Guvrin in the southwest. Beth Nattif, 5.5 km south of Khirbet esSuyyagh, was, according to the surveyors of Ramat Beth Shemesh, the largest settlement in the region (a town), after Beth Guvrin (Dagan, forthcoming). these were also important Christian centres. Christianity spread throughout Beth Guvrin’s hinterland, established itself in villages as well as in rural monasteries (Beyer 1931; Schwartz 1986:95-97; taxel 2008:65). No clear evidence for the existence of a Jewish or other non-Christian population in Beth Guvrin’s countryside in the Byzantine period was found. Jewish presence in the region during the Byzantine period was archaeologically recorded so far only from Beth Guvrin itself, but the finds on which this evidence is based are themselves not clearly dated (ilan 1991:260-261; Magness 2003b:108). the involvement of the monastery of Khirbet es-Suyyagh in the phenomenon of pilgrimage to the vicinity of Jerusalem in general and to the Judaean Shephelah in particular was already mentioned. the increase in Christian pilgrimage to the Holy land, mainly in the 5th and 6th centuries, occurred side by side with the flourishing settlement and economy of Palestine. the peak of this development occurred in the time of Justinian (527-565 Ce). During the reign of this emperor, dozens of public buildings – mainly churches – were erected or renovated Chapter 10: KhIrbet-eS-Suyyagh In Context throughout Palestine and transjordan (Di Segni 1999:164, tables 4, 5). Justinian’s most special attention – in building and donations – seems to be given to Jerusalem, which became a highly rich metropolis that attracted pilgrims from all over the empire and beyond (Rubin 1999:231-232). Justinian financed also the renovation and enlargement of several monasteries in the vicinity of Jerusalem (Procopius Buildings: 5, 9; Hirschfeld 1992:103), as well as the construction of a monastery and city wall on Mount Berenice, in the eastern fringes of tiberias, which became another pilgrimage centre (Hirschfeld 2004b:220). the construction of several rural monasteries in the western slopes of the Samaria Hills was also attributed to the time of Justinian (Hirschfeld 2002a:183). the pottery found in the foundations of some of the walls of the monastery of Khirbet es-Suyyagh indicates that it was built in the 6th century, and maybe even in its second half. this date suggests that this complex was also founded in the time of Justinian or in the time of his successor, Justin ii (565-578 Ce). although the construction of the monastery of Khirbet es-Suyyagh does not seem to have been financed directly by the imperial court, it is reasonable to attribute its founding to the general phenomenon of strengthening Christianity's hold throughout Palestine. the monastery of Khirbet es-Suyyagh was founded over the remains of older settlements, which was not an unique situation. Many coenobia and laurae monasteries in the Judaean desert, e.g. Marda (Masada), Castellion (Hyrcania), Doq, Herodium, Cypros, Nuseib >Uweishira, Khirbet el-Qa§r and Khirbet el-Khilya, were built within the remains fortresses from the Hasmonean and Herodian periods or from the late Roman period (Hirschfeld 1992:46). the foundation of the monasteries of Marda and Castellion, and the use that the founder monks made of the ancient remains was described by Cyril of Scythopolis (life of St. Euthymius 11; life of St. Sabas 27). Other desert and rural (including village-annexed) monasteries which were built over remains from the iron age onwards are known from the Negev (tel Masos: Fritz 1983:138-146; tel >ira: Creson 1999), the central hill country (Deir Qal>a: Magen and aizik 2008; Khirbet Umm Deimnah: Magen and Batz 2008; Khirbet ed-Dawwara: Batz and Sharukh 2008; Khirbet el-Qa§r: Magen, Hareven and Sharukh 2008; Ramat Raúel: testini 1962; 1964; Shoham Bypass Road: Dahari and >ad 2000) and the Galilee (Yeúi>am: Foerster 1977). the phenomenon of resettlement of older sites by founding monasteries in them is known also from other countries, and mainly from Syria. the topographically-dominant location of some of these sites, the opportunity to use the walls, water installations and building materials which already existed there, and the absence of private ownership on these deserted sites can explain this phenomenon (Hirschfeld 1992:49; vööbus 1960:164-165). the monastery in its original form (Phase iia) continued to exist as a strong economically independent unit until the end of the Byzantine period or sometime after, when the complex was partially destroyed probably by one of the earthquakes which hit Palestine in the 7th century. the damaged parts of the monastery were then renovated in a relatively rough manner. the terminus post quem of these repairs is given most precisely by the coin dated to 629/630 Ce which was found below the mosaic floor in the northern aisle of the church (locus 387) attributed to Phase iiB. there are also other finds which can date the beginning of Phase iiB to the mid-7th century. the first is a coin of Constans ii (641-648 Ce), which was found in the fill that covered the corridor north of the main gate (locus 281). this coin, however, can also be related to the construction of the blocking wall of the corridor in Phase iii. the other finds are represented by the pottery found below the fieldstone paving which abutted on the new (southern) storeroom in the external courtyard and to the repaired doorway of the subsidiary gate (loci 181 and 183). theoretically, the Sassanid-Persian conquest of 614 Ce could have been responsible for the destruction at the end of Phase iiB. the literary sources which deal with the conquest mention the destruction of churches and monasteries and even massacre of Christian communities in western Galilee but mainly in Jerusalem and nearby (Schick 219 Itamar taxel Fig. 10.1: Map of the northern Judaean Hills and Shephelah, showing cities, main roads and main Christian sites in the vicinity of Khirbet es-Suyyagh: 1) Matta>; 2) Rogelit; 3) Deir Rafat; 4) tel Beth Shemesh; 5) tel >azeqa; 6) Khirbet Fattir; 7) Beit Jimal; 8) Khirbet el-Jiljil; 9) Deir el->asfura; 10) Khirbet en-Nabi Bulus; 11) îorvat Socho; 12) Khirbet Jannaba et-taúta; 13) îorvat îanot. 14) >irak isma>in. 1995:21, 31, 33-39). However, these events occurred only in the initial stage of the conquest, and were not reported from the period of the Persian occupation which lasted until 628 Ce. Some of the damaged buildings – mainly in Jerusalem – were repaired soon after the conquest, and the reoccupation of some of the abandoned monasteries also occurred at about that time (Schick 1995:41-43). Some renovations were probably carried out after the renewal of the Byzantine rule in Jerusalem in 630 Ce, but these were few and undocumented in the archaeological record (Schick 1995:65-67). therefore, it does not seem very plausible that the monastery of Khirbet es-Suyyagh was damaged in the conquest of 614 and remained partially destroyed (and abandoned?) until at least 629/630 Ce. Furthermore, the nature of the damage and renovations does not indicate 220 violent destruction by human agency. the almost total absence of traces of fire weaken the possibility that the monastery of Khirbet es-Suyyagh was damaged during the Persian conquest. Nor does it seem that this destruction should be attributed to the Muslim conquest of Palestine in the 630s. there is little evidence for deliberate violent destruction of Christian sites at the beginning of the early islamic period (Schick 1995:128). Nonetheless, the excavator of nearby Khirbet Fattir suggested that the destruction of that site in the 7th century was caused either by the Persian invasion or by the Muslim army who defeated the Byzantines at the battle of Ijnādayn in 634 Ce and “…plundered and destroyed the Christian villages with their sanctuaries in the neighborhood…” (Strus 2003:430). Similarly, the Chapter 10: KhIrbet-eS-Suyyagh In Context Persian or Muslim conquest was identified as the cause of the destruction of the farmhouse of Khirbet el-Jiljil, although the excavators did not rule out the possibility that an earthquake was responsible (Strus and Gibson 2005:72). One of two earthquakes documented by contemporary and later sources could have hit Khirbet es-Suyyagh. according to Russell, the first earthquake occurred in September 633 (1985:46), although Guidoboni predates this event to 631/2 Ce (1994:504). the intensity of this earthquake and the exact area affected by it are unknown (the only place mentioned in the sources as damaged by the earthquake is the church of the Holy Sepulchre in Jerusalem), but according to Russell it was less destructive than other recorded earthquakes and affected a relatively limited area (1985:46). Schick suggested that this earthquake could have been responsible for the destruction of ecclesiastical complexes in Palestine and transjordan, mainly along the Jordan valley (1995:24, 66, 125). a stronger earthquake occurred in June 659, followed by another in 660 Ce (Russell 1985:47) which caused destruction throughout Palestine and transjordan. the sources mention several specific settlements and ecclesiastical complexes in Jerusalem and the Jordan valley that were damaged or destroyed by this earthquake, such as the monastery of euthymius. indeed, excavation of this monastery revealed evidence of its almost complete renovation after the earthquake of 659 Ce (Hirschfeld 1993:357). this earthquake was identified archaeologically also at other sites in the Jordan valley, including Pella (Schick 1995:66, 125126; Walmsley 1992:254), and at Caesarea (Raban, Kenneth and Bukley 1993:59-61). it is hard to determine which of the two above-mentioned earthquakes hit Khirbet esSuyyagh. apparently, the finding of the 629/630 coin below the church mosaic make the option of the 631/2 or 633 earthquake more plausible. However, archaeological evidence from other sites in Palestine and the neighbouring regions clearly shows that coins of Heraclius remained in circulation well into the Umayyad period (Magness 2003b:149, 170; Walmsley 2000:332-333). On the other hand, support for the 659/660 earthquake can be given by the pottery found in loci 181 and 183, which can be dated to the mid-late 7th century. thus, the possibility that the site was hit by the 659/660 earthquake rather than that of 631/2 or 633 seems somewhat more reliable. Still, the possibility that another undocumented earthquake hit the site sometime around the mid-7th century, as maybe happened at Beth Shean, cannot be ruled out (tsafrir and Foerster 1997:143-144). according to Schick, “in a thriving population an earthquake has only a transitory impact and the damage is quickly repaired. But a community in decline does not has the resources to repair monumental structures built in a previously more prosperous time. thus, whether churches were repaired or not is an important indication of the state of the Christian community at the time of the earthquakes. a church substantially rebuilt after an earthquake indicates that the Christian community at the site was still thriving, while absence of repair suggest not so much that the population was decimated by the earthquakes as that the Christian community there had already declined. the Christians could still use a church building that had once been suited to their needs, but which was now more substantial than they required, without spending much money or effort on upkeep. a church damaged or destroyed in a major earthquake, however, would only be restored to a standard required by current needs and allowed by the available resources of the community” (1995:123-124). after being repaired, the monastery continued to exist for several decades at the beginning of the early islamic period although its economy was undoubtedly less robust than before. this is amply demonstrated by the poor quality of the building as seen particularly in the apse of the church. Restoration of the monastery was probably possible only due to money that the community earned by selling the surpluses of the wine and oil produced here. these resources, however, were not enough to bring the monastery, or at least its church, to their previous state. 221 Itamar taxel in addition to the physical damage and deterioration in the monks' standard of living, the decrease in Christian pilgrimage to the Holy land after the Muslim conquest (Schick 1995:109) reduced the economic viability of rural monasteries. this, together with the severance from Byzantium, also led to a gradual decline in the export of wine (Hirschfeld 2002a:189, n. 82). thus, these factors can explain the deterioration of the economic abilities of the monastic community of Khirbet es-Suyyagh around the mid 7th century. Furthermore, the domino effect was that, no longer having to serve a large number of worshipers, including pilgrims, a high standard of renovation of the basilical church became unnecessary. the continued existence of monasteries after the Muslim conquest is a well-known phenomenon. the relative tolerance of the caliphs and their openness to local cultures even permitted many monasteries to flourish into the Umayyad period and sometimes into the >abbasid period and later (Schick 1995:96-100). the following solitary rural monasteries yielded more or less solid evidence of their continuation after the Muslim conquest: Deir Qal>a (Magen and aizik 2008:1695), îorvat îani (Dahari 2003:104-105), îorvat Migdal (ayalon 2002:284; Matthews, Neidinger and ayalon 1990:7), Ramot (arav, Di Segni and Kloner 1990), Khirbet Umm Deimnah (Magen and Batz 2008:2059) and Khirbet el-Shubeika (Syon 2003). Other monasteries, such as those of Siyar elGhanam (Corbo 1955), Bir el-Qutt (ibid.), Khirbet abu Rish (Magen and Baruch 1997), Khirbet ed-Dawwara (Batz and Sharukh 2008), Khirbet el-Qa§r (Magen, Har-even and Sharukh 2008), Rujm Jerida (Magen, Peleg and Sharukh 2008), Deir Ghazali (avner 2000), Ras et-tawil (Gibson 1985-6) and Mevo Modi>im (eisenberg and Ovadiah 1998) had a post-monastic phase dated to the early islamic period, though it cannot be said how long they continued to function as monasteries after the Muslim conquest. in the Byzantine period Khirbet es-Suyyagh was included within the northern border of the toparchy of Beth Guvrin, and was most probably subordinate to its bishopric. in the course of the 222 early islamic period the status of this city was weakened, and so was the status of its Christian leadership. During the early islamic period (at least until the >abbasid period) the Christian settlements in the Judaean Shephelah fell either within the area of the patriarchate of Jerusalem or the archbishopric of Beth Guvrin. according to the Taktikon (Τακτικόν) lists, which describe the organization of the Palestinian patriarchates in the early islamic period, the northern border of the archbishopric of Beth Guvrin was the valley of elah. Khirbet es-Suyyagh, therefore, was not part of this archbishopric, but belonged to the patriarchate of Jerusalem, being subordinate to nearby emmaus (levy-Rubin 2003:211, 218, Map 2). tHe late UMaYYaD/>aBBaSiD PeRiODS after the abandonment of the monastery, probably around the middle of the Umayyad period, the site seems to have remained deserted for a short time. Not later than the first half of the 8th century, however, new inhabitants established a small rural settlement on the older remains. this settlement continued to exist well into the 9th century, but was probably abandoned not later than its second half. THE ARCHITECTURAl CONTEXT Characteristic of the settlement of Phase iii is the construction of walls and floors which redivided existing spaces into smaller rooms and chambers and blocked some of the original doorways. it seems that most of the building stones for the newly built walls were taken from the existing remains of the monastery. the nature of these alterations and the finds which were attributed to them indicate that most of the complex had a domestic nature throughout Phase iii. the church complex and the large courtyard were also used now for various domestic needs. it is possible that the oil press complex continued to function, at least partially, during Phase iii as indicated by the latest pottery sherds which were found in the fill that covered the complex. these finds date the latest use of the oil press to the >abbasid period, but not necessarily for its original purpose. Chapter 10: KhIrbet-eS-Suyyagh In Context the conversion of the eastern cistern into a refuse pit during Phase iii is a clear example of secondary refuse (Schiffer 1995:211) dumped by later inhabitants of a given site. the use of subterranean (either built or rock-cut) features as dump receptacles is a well known phenomenon with parallels in many contemporary sites (see, e.g. tal and taxel 2008:111, 160). the use of older building material, including the church’s marble furnishings, and the transformation of the church complex into a secular place clearly indicate that the new inhabitants had a different religious and social orientation from that of the previous Christian monastic community. the phenomenon of re-occupation of deserted Byzantine monasteries by new inhabitants in the late Umayyad or >abbasid period is known also from other sites in Palestine. it is more remarkable among the rural monasteries (both solitary and village-annexed) in the arable regions, although there are some examples also from the desert monasteries. in most cases this phase did not last for long, and ended not later than the 10th century. Some examples of rural monasteries with evidence of a post-monastic phase are Mevo Modi>im where the refectory and chapel’s narthex were used as dwelling spaces by dividing them up by means of new walls. the chapel was transformed into an oil press by moving some of the elements of the monastery’s oil press into the chapel. this phase was dated according to the pottery to the 8th-10th centuries (eisenberg and Ovadiah 1998:10*, 18*). the later phase in the monastery of Deir Ghazali is reflected by the thickening of existing walls by dismantled architectural elements of the original phase, the redivision of rooms and the blocking of doorways. this phase was dated, based on the pottery, to between the late Byzantine and the >abbasid period (avner 2000:28*, 50*). the monastery of îorvat îani ceased to function as such in the 9th century. Parts of its building materials were robbed, and its area became a Muslim cemetery which existed well into the medieval period (Dahari 2003:105). early >abbasid pottery (erroneously dated to the late Umayyad period) which was found in one of the oil presses of the complex at îorvat Migdal apparently dates the time of its turning out of use (Matthews, Neidinger and ayalon 1990:17, Fig. 18). at Khirbet Umm Deimnah, a new northern wing was added to the abandoned monastery, and some dividing walls, floors and a staircase were built within the old building. in addition, an oil press was built in one of the monastery’s rooms, with reused liturgical furniture of the chapel (Magen and Batz 2008:2059). the monastery of Shoham Bypass Road was abandoned in the 10th century. a new settlement was established in the 11th century on the ruins: doorways and passages were blocked, new walls were erected over collapses of older walls, tabuns were built over the remains of older walls and lime kilns were built on the edges of the site. Based on arabic inscriptions on lamps and other pottery vessels, the excavators identified the new inhabitants as Muslims. this settlement existed until the 12th century (Dahari and >ad 2000:58*). the monastic church at îorvat Beth loya also had a later phase, dated to the 9th-10th centuries, which is represented by new walls that were built in the bema area and by domestic pottery (Magness 2003b:109; Patrich and tsafrir 1985:106). a rectangular building was built in the centre of the western wing of the monastery of Martyrius in the Judaean desert after its abandonment. the builders reused stones, architectural elements and marble furniture taken from the abandoned monastery. an agricultural area with irrigation channels east of the building was also related to this phase. the later construction was identified as an agricultural farm, and was dated by a coin, an arabic inscription and pottery to the 8th century. Probably more or less at the same time some of the monastery’s architectural elements (such as columns and capitals), mosaics and roof tiles were systematically dismantled for use in the building of new structures, such as the Umayyad palace at Khirbet al-Mafjar (Magen and Hizmi 1985:86, 90; Magen and talgam 1990:106). the southern building in the monastery of Khirbet el-Khillia in the Judaean desert also reflects a post-monastic phase, which was dated by the pottery to the Umayyad and >abbasid periods (the 7th-9th centuries). this phase 223 Itamar taxel includes many architectural alterations – redivision of rooms, construction of small installations (such as troughs) and blocking of passages between pilasters (Sion 1998:200-201). Similar changes were made also in the monastery of tel >ira. Some of the doorways of the chapel and atrium were blocked, and the atrium was redivided by new walls. Small storage installations were built in two other rooms (Cresson 1999:95). this post-monastic phase was dated by Magness, based on the pottery, to the 8th to early 9th century (2003b:56-57). the phenomenon mentioned above is not restricted to monasteries. On the contrary, it is better known from villages where the local churches were reoccupied by a new population. Schick differentiated between churches which were reoccupied after a gap in the use of the church – for example, after destruction by an earthquake – and between churches that were reoccupied with connection to their abandonment, maybe as a result of a deliberate destruction. However, the latter scenario is barely attested in the archaeological record from Palestine and transjordan, and in the few possible cases of deliberate destruction the churches were either rebuilt as churches or left in ruins (1995:129-130). the commonest use of abandoned churches (and monasteries) was for domestic occupation. very few churches were converted into mosques. they could be used as animal pens and/or dwellings. Building stones and other architectural elements were robbed out of the abandoned churches, whether reoccupied or not. the marble furniture of the churches was also used in building or in lime production, as evidenced also by the finding of lime kilns inside or near some of the churches (Schick 1995:130-132). SOCIAl ANd ECONOMIC ASPECTS Who were the people that settled in Khirbet esSuyyagh after the abandonment of the monastery? the exact size of this rural community is impossible to assess. the number of the former monastery rooms reused as living rooms is unknown, nor how many people lived in each room. Nevertheless, it does not seem that this site was inhabited by more 224 than one clan or two nuclear families. according to Schick, it is not always possible to identify the religious identity of the new inhabitants of the deserted Christian complexes. He mentioned only three cases of rural churches in which there is quite clear evidence that the later occupants were Muslims (1995:131). Similarly, Magness preferred to be very cautious when, despite the clear evidence that the two churches of Khirbet Yattir went out of use sometime after the Muslim conquest, she said that “there is no archaeological basis for determining whether there were changes in the religious orientation of the village’s population after the Muslim conquest” (2003b:107). From Schick’s words it can be understood that, although the abandonment of certain churches does not necessarily indicate the complete disappearance of Christian communities in those settlements, the general picture points to at least a partial shift in population and/or religious identity in the late Umayyad and early >abbasid periods in Palestine and transjordan. However, such shifts are complex and connected to various economic, political and social changes (1995:134-135). the lack of numerical data with regard to the population of early islamic Palestine makes the estimation of the ratio between the different religious groups very difficult, although the Muslims themselves were a minority at least until the end of the period (Gil 1992:170). Based on the writings of local Muslim scholars from the late 10th century (al-Muqaddasī and altamimi), amar assumed that in the 10th and 11th centuries the Christians were a considerable part, and maybe even the majority of the population of Judaea (or at least of greater Jerusalem). the process of islamisation in Judaea started to be felt only from the second half of the 10th century (amar 1995:251). Similarly, ellenblum is of the opinion that true islamisation of rural Christian communities occurred only after the creation of established Muslim communities, i.e., in a relatively advanced stage of the early islamic period (1998:255-256; see also Gil 1992:221-222). in contrast to villages, rural monasteries were inhabited by a homogeneous religious community. Chapter 10: KhIrbet-eS-Suyyagh In Context thus it is much easier to identify the total abandonment of monasteries than the desertion of villages which was not always complete. Christian villages which were partially abandoned would have remained settled by a smaller community (taxel, forthcoming). this nucleus could subsequently have been augmented by a non-Christian population. the resilience theory, suggested by Walmsley as a way of interpreting social, economic and cultural changes in early islamic Palestine (2007:146-147), seems reasonable and helpful mainly when dealing with relatively complex, socially-diverse communities, such as towns and villages. the “ability of an individual, family, community or state to resist or recover from potentially destructive challenges through the construction of successful adaptive strategies” (Walmsley 2007:146) is much harder to identify in a type of settlement such as the rural monasteries. abandoned sites were reoccupied by a Muslim population who could have been immigrants to Palestine from the arabian Peninsula, Persia or Syria, under the encouragment of the Muslim caliphs. the other possible settlers were nomads who were in the process of sedentarisation (Safrai and Sion 2007; Whitcomb 2009:242-245). the properties themselves were transferred into the hands of the local Muslim authorities. Some of these were agricultural estates (arabic: èiyā>, sing. èay>a; villages and individual houses could also form part of the èay>a) which were allocated by the authorities to rich Muslim immigrants, who became part of the civilian élite. >amr b. al->Ā§, one of the major Muslim commanders who lead the conquest of Palestine from the Byzantines, owned such an estate, called >Ajlān, southwest of Beth Guvrin (lecker 1989:31-37). the lands of these estates were cultivated by tenants, and some of the lands could have been rented to the tenants in exchange for a certain percentage of the crop. the estates could increase their property by giving protection to villagers and small land-owners. Sometimes, the latter preferred to be protected by wealthy landlords in order to avoid extortion by tax collectors and to get a reduction in taxes. in practice, these private lands became part of the estate’s property (Gil 1992:137, 224-225; Kennedy 2004:24; Yusuf 1985:28, 32, 50). amar and Serri suggested that the agricultural class called by the local Muslim scholar of the late 10th century al-tamimi akara (sing. akkâr), refers to the landlords, while the term falaúin refers to the tenants (2004:93-94). However, regarding another estate from the vicinity of Beth Guvrin, which owned by the Umayyad caliph al-Walīd, it was said that the estate was granted to a Bedouin leader named Rawú b. Zinbā> together with the slaves and akara who lived in it (lecker 1989:32). therefore, the term akara should be referred not to the landlords but to the tenants. the term falaúin, if so, maybe refers to the free peasants who cultivated their own lands. into what category, therefore, does the humble agricultural settlement founded within the deserted monastery of Khirbet es-Suyyagh fall? Were its inhabitants akara or falaúin? On the one hand, a nomadic group may have decided to settle here, and perhaps were even encouraged by the authorities to do so. On the other hand, this settlement could have been part of a larger estate owned by a landlord who lived in the vicinity, maybe even in Beth Guvrin. it is also possible that the settlement was founded by free people ( falaúin?), who later became protected tenants. With regard to the religious affiliation of the new settlers, it is known that a Christian population continued to live in the Judaean Shephelah in the late Umayyad and >abbasid periods. it was probably concentrated in the close vicinity of Beth Guvrin (Constantelos 1972:343; Magness 2003b:108-109) and did not reinhabit deserted settlements/monasteries. the three other ethnic groups who lived in Palestine in this periods were Muslims, Jews and Samaritans. the possibility that the newcomers were Samaritans seems least likely since these communities were concentrated in the Samaria Hills, on the coastal plain and in the area of Beth Shean. the Judaean hill country remained almost unsettled by Samaritans (levy-Rubin 2002:564; Magen 2002b:267-269, Fig. 1). 225 Itamar taxel the extent of the Jewish settlement in Palestine at the beginning of the early islamic period is unknown. Jews apparently continued to live in many of the cities and villages that they previously inhabited (Gil 1992:173). However, neither the written sources nor the archaeological finds provide any secure evidence regarding the establishment of new rural settlements by Jews in the early islamic period. the excavators of îorvat Bireh, a rural settlement on the southwestern fringes of the Samaria Hills, identified the last phase of the site, which was dated to the Umayyad period, as a Jewish estate (Safrai and Dar 1997:57-58). the attribution of this phase to a Jewish community is based on a lamp of an Umayyad period type which was decorated with two seven-branched menorahs and a hewn installation identified as a Jewish ritual bath (miqveh) (avissar 1997:Fig. 3:1; Safrai and Dar 1997:93). However, such artefacts also characterized the Samaritans (taxel 2005:240-241). Similarly, the lamp fragment from Khirbet esSuyyagh, which was decorated with two sevenbranched menorahs (Fig. 3.26:1), may suggest that Jews inhabited the site in the early islamic period. Nevertheless, extreme caution must be exercised in basing such an identification on a single find. Furthermore, the menorah – and various stylized versions of it – was adopted not only by the Samaritans as a religious symbol, but also by the Christians as an artistic motif and religious symbol (Hachlili 2001:263-274). Seven- and fivebranched menorahs were also depicted on a group of Umayyad coins which bear the Muslim shahāda (“there is no god except allah alone, Muhammad is allah’s messenger”). However, here also the menorah is no more than an echo of its original Jewish meaning (Barag 1988-89). thus, despite the temptation to connect it to a Jewish community, the menorahs on the Umayyad lamp from Khirbet esSuyyagh cannot be used to determine the religious identity of the site’s inhabitants in Phase iii. the major economic centres of the region in the late Umayyad and >abbasid periods remained, as before, Jerusalem and Beth Guvrin (now called in Arabic Bayt Jibrīn). The late 10th century geographer al-Muqaddasī said that “Bayt Jibrīl .... 226 is a town in jund Filastīn” and described the fertile agricultural lands and the estates in its vicinity. He also said the the region sent its agricultural products to the district’s capital, Ramla (155, 174). at this time the settlement at Khirbet es-Suyyagh was already deserted, but there is no reason to believe that the economic situation in the 8th and 9th centuries was very different from that described by al-Muqaddasī. the economy of the settlement of Phase iii was based on agriculture. Olives continued to be a very important crop in Palestine also in the early islamic period although it cannot be said whether the new inhabitants continued to use the monastery’s oil press for producing oil. Oil production during Phase iii could also have been done by much simpler methods. the marble roller (Fig. 5.2:4), which is attributed to Phase iii, could have been used for crushing a very small amount of olives at a time. either using parts of the earlier oil press or stone rollers, the inhabitants of Khirbet esSuyyagh in Phase iii could produce only relatively small amount of olive oil for private domestic use, which stands in a sharp contrast to the industrial oil production of Phase ii. the iron bident hoe found in a context dated to Phase iii provides additional evidence for the agricultural nature of the settlement. this implement could have used in the cultivation of various kinds of crops, from olive trees to vegetables. the faunal remains which were attributed to Phase iii (mainly those from locus 175, and see Chapter 8) indicate that the settlement’s inhabitants raised mainly sheep and goats, but also cattle and chickens. Some of the sheep, goats and cattle bones bear cut marks, which indicate that these animals were raised not only for their milk and wool, but also for their flesh. these animals were among the most common domestic types in the early islamic period besides various beasts of burden (Yusuf 1985:48). it is interesting to compare the faunal finds from Phase iii at Khirbet es-Suyyagh to those from a number of farms and small villages (elatelot, Naúal la>ana, Naúal Mitnan, Naúal Shaúaq and Upper Naúal Besor) which were excavated in the Negev and the arava and were dated to the Chapter 10: KhIrbet-eS-Suyyagh In Context Umayyad and early >abbasid periods. all these settlements yielded bones of sheep/goat, which represent the lion's share of the faunal remains. Most of the sites yielded also bones of chickens and/ or of beasts of burden, usually camels, in addition to some wild animals and birds (Heker 1996; 2004; Horwitz 1995; 1998; israel, Nahlieli and Ben Michael 1995:10*). Sheep/goat and chicken bones were found also in the Umayyad and >abbasid village of Khirbet abu Suwwana in the northern Judaean desert (Sion 1997a:191). None of the sites, however (excluding one uncertain case from Naúal Mitnan; Horwitz 1995:15), yielded cattle bones. this difference between Khirbet es-Suyyagh and the settlements of the arid regions can be explained by the climatic conditions which did not allow the inhabitants of the latter sites to rise cattle. On the other hand, the >abbasid- and Fatimid-period phases at Khirbet ibreica in the Sharon plain and Kafr >ana in the lod valley, whose climate is much closer to that of the Judaean Shephelah, yielded bones of sheep/goat, cattle, camels and chickens (Sade 2006:62; 2007:106-107). Cat bones found in the collapse which sealed Cistern 24 (locus 175) may have interesting implications regarding sanitary conditions in the settlement during Phase iii. the existence of cat/s in such a small relatively isolated settlement can indicate the existence of mice and/or rats in quantities that justified the appearance or bringing of the small predators (for the association between cats and rats in human communities, see McCormick 2003:21). indeed, we cannot rule out the possibility that mice and/or rats existed in the site already in its former, monastic phase. However, when comparing the economic abilities and maybe even the skills of the site’s inhabitants during the monastic and post-monastic phases, we cannot ignore from the feeling that the preciseness on appropriate hygiene conditions during the latter phase was felt much less. THE REGIONAl CONTEXT in contrast to the situation in the Byzantine and early Umayyad periods, surveys and excavations show that the number of settlements in the region started to decline sharply in the late Umayyad or >abbasid period. Only 20 sites in the Nes Harim survey map were described as yielded “early islamic” finds. Most of these sites were identified as small villages, some situated beside main roads (Weiss, Zissu and Solimany 2004:16*). However, the surveyors did not describe their criteria for identifying early islamic sites. Presumably, as is the case in other surveys conducted in Palestine, such sites were identified mainly by ceramic types such as Fine Buff Ware, glazed bowls and channelnozzle lamps, which are usually dated to the 9th century onward. thus, theoretically sites which were inhabited during the 8th century were not identified as early islamic sites. the early islamic sites closest to Khirbet es-Suyyagh are: Khirbet Jinnā>ir, Khirbet Rabi>a, Khirbet îaraza, Khirbet >Umran, Khirbet Zanu> and Naúal Ha-Nativ, all were inhabited already in the Byzantine period (Weiss, Zissu and Solimany 2004:27*, 31*-33*, 44*, 56*). îorvat >illin, situated 0.5-0.6 km southwest of Khirbet es-Suyyagh, is the closest site which yielded remains and finds from the early islamic period. these remains includes parts of simple structures, which were identified as belong to a small village. two construction phases were identified in some of the remains; both were dated to the Umayyad and >abbasid periods. Some older architectural elements, including fragment of marble furniture of a Byzantine church, were reused in the early islamic building activity. Some older burial caves from the early Roman and Byzantine periods, were robbed in the >abbasid period. the site continued to exist also in the Fatimid and Crusader periods, but probably not as a permanent settlement (Greenhut 2004:15*-19*, 30*; Seligman and May 1993; Weksler-Bdolah 1996). at Khirbet Fattir, a little bit further away to the southwest, evidence of occupation in the late Umayyad and >abbasid periods was also found. Some installations and buildings dated to the Byzantine period (including the site’s church) were reoccupied as dwelling places (Strus 2003:103, 122, 147). at the same time stones were looted from nearby Khirbet el-Jiljil by squatters (Strus and Gibson 2005:76-77). 227 Itamar taxel SUMMaRY throughout the settlement history of Khirbet esSuyyagh the size, nature, and identity of the site’s population changed several times. For nearly 1000 years, including a few gaps in occupation, it remained a rural agricultural settlement. the small farmstead of the late Hellenistic/early Roman period was reoccupied in the late Roman/ early Byzantine period. it was replaced by a wellplanned monastery in the late Byzantine and early Umayyad periods. after being abandoned some time in the late 7th century, this was superseded by a poor hamlet in the late Umayyad and >abbasid periods. the latest coin found in the excavations is dated to 841-842 Ce. it was found in the foundations of a wall; a fact which indicates that the settlement continued to exist at least some years after the coin was minted. the reasons for the abandonment of the last permanent settlement at Khirbet es-Suyyagh are unknown. However, it is plausible to assume that it was connected to the political upheavals in Palestine during the >abbasid period. In the reign of the caliph al-Wāthiq (842-847 CE), a widespread rebellion led by Abū îarb Tamīm broke out in Palestine. Farmers and Arab tribes from southern Palestine participated and some major settlements in Judaea, including Jerusalem and Hebron, were affected (Gil 1992:295-296). it is tempting to connect the abandonment of Khirbet es-Suyyagh to this uprising, but there is no substantiating evidence for this. Other rebellions, which affected mainly 228 northern Palestine and the region of Ramla, occurred also in the 860’s (Gil 1992:299-600), but their effect on more southerly regions is unclear. in addition, Palestine was raided by arab tribes (including Bedouin nomads) (Kennedy 1991:110) from the late 9th century onwards. among these, were the Qarma‹īs’ raids in 900 and 906 Ce which led to the >abbasids reoccupation of Palestine after being under Tū lūnid reign since 878 Ce (Gil 1992:312-313). it cannot be said, however, if the settlement at Khirbet es-Suyyagh was directly affected by any of the events mentioned above, or whether the deterioration of security and economic conditions forced the inhabitants to move to a larger, and presumably safer, settlement. the excavation of Khirbet es-Suyyagh provides important information for the study of the history and material culture of the Judaean Shephelah in the first millennium Ce. First and foremost is its contribution to the study of rural monasticism in Palestine in the Byzantine period and the beginning of the early islamic period. the opportunity to fully excavate a monastery complex is not very common in modern-day israel. Most salvage excavations are limited to a small arbitrarily chosen area within the area of the site. very little is known about the monastic history of this region since only a relatively small number of rural monasteries have been fully or nearly fully excavated. Khirbet es-Suyyagh is an important addition to this list. referenCeS HiStORiCal SOURCeS antoninus Placentinus. Milani, C. 1977, ed. Itinerarium Antonini Placentini. Un viaggio in Terra Santa del 560-570 d.c. Milano. Apophthegmata Pratum. Miller, B. 1965, ed. Apophthegmata Pratum, auch Gerontikon oder Alphabeticum genannt. Freiburg im Breisgau. Codex of Justinian. Kruger, P. 1877, ed. Codex Iustinianus (Corpus iuris Civilis 2). Berlin. eusebius, Onomasticon. Klostermann, e. 1904, ed. das Onomasticon der biblischen Ortsnamen. leipzig. theodosius, de situ terrae sanctae. Geyer, P. 1965, ed. in: Itineraria et alia geographica (Corpus Christianorum, Series latina 175:111-125). John Moschos, Pratum Spirituale (Patrologia Graeca 87.3). John Moschos, Pratum Spirituale Supplementary Tales. Nissen, t. 1938. Unbekannte erzählungen aus dem Pratum spirituale. Byzantinische Zeitschrift 38:351-376. Josephus. The Jewish War. thackeray, H.St.J., 1956, ed. Cambridge, Ma Hieronymus, Epistolae. Hilberg, i. 1955, ed. Saint Jérôme lettres, Tome 5. Paris. Al-Muqaddasī. The Best divisions for Knowledge of the Regions (Aúsan al-Taqāsīm fī Ma>rifat al-Aqālīm). Collins, B., ed. 2001. Reading. Pliny, Historia Naturalis. in: Stern, M. 1976, ed. Greek and latin Authors on Jews and Judaism, Vol. I. Jerusalem. pp. 468-469. Procopius, Buildings. Dewing, H.B. 1954, ed. london. life of St. Euthymius. Price, R.M. 1991, ed. Cyril of Scythopolis: The lives of the Monks of Palestine. Kalamazoo, Mich. pp. 1-92. life of St. Sabas. Price, R.M. 1991, ed. Cyril of Scythopolis: The lives of the Monks of Palestine. Kalamazoo, Mich. pp. 93-219. ReFeReNCeS abel, F.M. 1936. le monastère de Beth-Shémesh. Revue Biblique 45:538-542. abouzayd, S. 2005. the prohibition and the use of alcohol in the Syrian ascetic tradition and its biblical and spiritual origins. Aram 17:135-156. acconci, a. 1994. l’arredo liturgico. in: Piccirillo, M. and alliata, e., eds. Umm al-Rasas – Mayfa>ah, I. Gli scavi del complesso di Santo Stefano. Jerusalem. pp. 290-313. acconci, a. 1998. elements of the liturgical furniture. in: Piccirillo, M. and alliata, e., eds. Mount Nebo: New Archaeological Excavations, 1967-1997. Jerusalem. pp. 468-542. adan-Bayewitz, D. 1986. the pottery from the late Byzantine building. in: levine, l.i. and Netzer, e. Excavations at Caesarea-Maritima 1975, 1976, 1979, Final Report. (Qedem 21) Jerusalem. pp. 90-129. agady, S., arazi, M., arubas, B., Hadad, S., Khamis, e. and tsafrir, Y. 2002. Byzantine shops in the Street of the Monuments at Bet Shean (Scythopolis). in: Rutgers, l.v., ed. What Athens Has to do with Jerusalem. Essays on Classical, Jewish, and Early Christian Art and Archaeology in Honor of Gideon Foerster. leuven. pp. 423-506. aharoni, Y. 1954. excavations at Beth-Hashitta. Bulletin of the Israel Exploration Society 18:209-215. (Hebrew) aharoni, Y. 1962. Excavations at Ramat Raḥel, Seasons 1959 and 1960. Rome. aharoni, Y. 1964. Excavations at Ramat Raḥel, Seasons 1961 and 1962. Rome. 229 alchermes, J. 1994. Spolia in the Roman cities of the late empire: legislative rationales and architectural reuse. dumbarton Oaks Papers 48:167-178. alliata, e. 1990. Nuove settore del monastero al Monte Nebo-Siyagha. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 427-466. alliata, e. 1994. i reliquiari ed altri elementi architettonici. in: Piccirillo, M. and alliata, e., eds. Umm alRasas – Mayfa>ah, I. Gli scavi del complesso di Santo Stefano. Jerusalem. pp. 313-317. amar, Z. 1995. Grape cultivation and wine production in Judaea and Samaria during the Middle-ages. Judea and Samaria Research Studies 4:247-261. (Hebrew) amar, Z. and Serri, Y. 2004. The land of Israel and Syria as described by al-Tamimi, Jerusalem Physician of the 10th Century. Jerusalem. (Hebrew) amiran, D.H.K., arieh, e., and turcotte, t. 1994. earthquakes in israel and adjacent areas: Macroseismic observations since 100 B.C.e. Israel Exploration Journal 44:260-305. amit, D. 1997. Kh. Umm îalasa: an additional monastery in the Wilderness of Ziph. Judea and Samaria Research Studies 6:259-270. (Hebrew) amit, D. 2003. The Synagogues of Hurbat Ma>on and Hurbat >Anim and the Jewish Settlement in the Southern Hebron Hills. (Ph.D. dissertation, the Hebrew University of Jerusalem) Jerusalem. (Hebrew) amit, D., Seligman, J. and Zilberbod, i. 2003. the “Monastery of theodorus and Cyriacus” on the eastern slope of Mount Scopus, Jerusalem. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 139-148. amit, D., Seligman, J. and Zilberbod, i. 2008. Stone vessel production caves on the eastern slope of Mount Scopus, Jerusalem. in: Rowan, Y.M. and ebeling, J.R., eds. New Approaches to Old Stones: Recent Studies on Ground Stone Artefacts. london and Oakville. pp. 320-342. amit, D. torge, H. and Gendelman, P. 2008. îorvat Burnat, a Jewish village in the lod Shephalah during the Hellenistic and Roman periods. Qadmoniot 136:96-107. (Hebrew) amit, R., Zilberman, e., enzel, Y. and Porat, N. 2002. Paleoseismic evidence for time dependency of seismic response on a fault system in the southern arava valley, Dead Sea Rift, israel. Geological Society of America Bulletin 114:192-206. anati, e. 1957. Sussita. Bulletin of the department of Antiquities of the State of Israel 5-6:31-33. (Hebrew) andersen, F.G. 1985. Shiloh. The Danish Excavation at Tall Sailūn, Palestine in 1926, 1929, 1932 and 1963, Vol. 2: The Remains from the Hellenistic to the Mamlūk Periods. Copenhagen. anderson, W. 2004. an archaeology of late antique pilgrim flasks. Anatolian Studies 54:79-93. arav, R., Di Segni, l. and Kloner, a. 1990. an eighth century monastery near Jerusalem. liber Annuus 40:313-320. arnon, Y.D. 2008. Caesarea Maritima, the late Periods (700-1291 CE). (British archaeological Reports international Series 1771) Oxford. avi-Yonah, M. 1944. Oriental elements in the art of Palestine in the Roman and Byzantine periods. Quarterly of the department of Antiquities in Palestine 10:105-151. avi-Yonah, M. 1966. The Holy land from the Persian to the Arab Conquests (536 B.C. to A.d. 640): A Historical Geography. Grand Rapids, Mich. aviam, M. 1994. large-scale production of olive oil in Galilee. Cathedra 73:26-35. (Hebrew) aviam, M. 2002. Five ecclesiastical sites in western upper Galilee. in: Gal, Z., ed. Eretz Zafon: Studies in Galilean Archaeology. Jerusalem. pp. 165-218. (Hebrew) aviam, M. 2004. Churches and monasteries from the Byzantine period in western Galilee. in: aviam, M., ed. Jews, Pagans and Christians in the Galilee. 25 years of Archaeological Excavations and Surveys, Hellenistic to Byzantine Periods (land of Galilee 1). Rochester. pp. 181-204. 230 referenCeS aviam, M. and Getzov, N. 1998. a Byzantine smithy at îorvat Ovesh, Upper Galilee. >Atiqot 34:63-83. (Hebrew) avissar, M. 1996. the Medieval pottery. in: Ben-tor, a., avissar, M. and Portugali, Y., eds. yoqne>am I: The late Periods. (Qedem Reports 3) Jerusalem. pp. 75-172. avissar, M. 1997. the pottery of Hurvat Bireh. in: Friedman, Y., Safrai, Z. and Schwartz, J., eds. Hikrei Eretz. Studies in the History of the land of Israel, dedicated to Prof. yehuda Feliks. Ramat-Gan. pp. 109-125. (Hebrew) avissar, M. 2003. early islamic through Mamluk pottery. in: Geva, H., ed. Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 1969-1982. Vol. 2: The Finds from Areas A, W and X-2, Final Report. Jerusalem. pp. 433-446. avissar, M. 2005. Tel yoqne>am – Excavations on the Acropolis. (iaa Reports 25) Jerusalem. avissar, M. and Stern, e.J. 2005. Pottery of the Crusader, Ayyubid, and Mamluk Periods in Israel. (iaa Reports 26) Jerusalem. avitsur, S. 1976. Man and His Work, Historical Atlas of Tools and Workshops in the Holy land. Jerusalem. (Hebrew) avitsur, S. 1994. Olive oil production in the land of israel. in: Frankel, R., avitsur, S. and ayalon, e. History and Technology of Olive Oil in the Holy land. tel aviv. pp. 91-158. avner, R. 1995. Khirbet Jannaba et-taúta. Excavations and Surveys in Israel 16:114-117. avner, R. 1998. elat-elot – an early islamic village. >Atiqot 36:21*-39*. (Hebrew) avner, R. 2000. Deir Ghazali: a Byzantine monastery northeast of Jerusalem. >Atiqot 40:25*-52*. (Hebrew) avner, R. 2003. the recovery of the Kathisma Church and its influence on octagonal buildings. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 173-186. avni, G. and Dahari, U. 1990. Christian burial caves from the Byzantine period at luzit. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 301-314. avni, G., Dahari, U. and Kloner, a. 2008. The Necropolis of Bet Guvrin-Eleutheropolis. (iaa Reports 36) Jerusalem. avshalom-Gorni, D. 2002. excavations at Khirber el-Shubeika, 1991, 1993: architecture and stratigraphy. in: Gal, Z., ed. Eretz Zafon: Studies in Galilean Archaeology. Jerusalem. pp. 220-226. (Hebrew) avshalom-Gorni, D., Frankel, R. and Getzov, N. 2008. a complex winepress from Mishmar Ha->emeq: evidence for the peak in the development of the wine industry in eretz israel in antiquity. >Atiqot 58:49-66. (Hebrew) ayalon, e. 1984. two wine presses from the Roman period at the eretz-israel Museum. Tel Aviv 11:173-182. ayalon, e. 1997. the end of the ancient wine industry in the central coastal plain. in: Friedman, Y., Safrai, Z. and Schwartz, J., eds. Hikrei Eretz. Studies in the History of the land of Israel, dedicated to Prof. yehuda Feliks. Ramat-Gan. pp. 149-166. (Hebrew) ayalon, e. 2002. îorbat Migdal (tsur Natan) – an ancient Samaritan village. in: Stern, e. and eshel, H., eds. The Samaritans. Jerusalem. pp. 272-288. (Hebrew) ayalon, e. 2004. the stone and metal implements from îorvat Raqit. in: Dar, S. Raqit, Marinus’ Estate on the Carmel, Israel. (British archaeological Reports international Series 1300) Oxford. pp. 268-296. ayalon, e. 2008. The Assemblage of Bone and Ivory Artefacts from Caesarea Maritima, Israel, 1st-13th Centuries CE. (British archaeological Reports international Series 1457) Oxford. ayash, e. 2000. Ramla (B). îadashot Arkheologiyot – Excavations and Surveys in Israel 112:69*. Bagatti, B. 1955-1956. Scavi di un monastero al “Dominus Flevit” (Monte Oliveto – Gerusalemme). liber Annuus 6:240-270. 231 Bagatti, B. 1969. Nouvi apporti archeologici al “Dominus Flevit” (Oliveto). liber Annuus 19: 194-236. Bagatti, B. 1990. le antichità di Rafat e dintorni. liber Annuus 40:263-286. Bagatti, B. 2002. Ancient Christian Villages of Judaea and the Negev. (trans. by Rotondi, P.) Jerusalem. Bailey, D.M. 1998. Excavations at el-Ashmunein, Vol. 5: Pottery, lamps and Glass of the late Roman and Early Arab Periods. london. Bakker, G. 1985. the buildings of alahan. in: Gough, M., ed. Alahan: An Early Christian Monastery in Southern Turkey. Based on the Work of Michael Gough. toronto. pp. 75-196. Baly, t.J.C. 1962. Pottery. in: Colt, H.D., ed. Excavations at Nessana, Vol. 1. london. pp. 270-303. Bar-Nathan, R. 1981. Pottery and stone vessels of the Herodian period. in: Netzer, e., ed. Greater Herodium. (Qedem 13) Jerusalem. pp. 54-70. Bar-Nathan, R. 2002. Hasmonean and Herodian Palaces at Jericho. Final Report of the 1973-1987 Excavations, Vol. III: The Pottery. Jerusalem. Bar-Nathan, R. 2006. Masada VII. The yigael yadin Excavations 1963-1965, Final Reports: The Pottery of Masada. Jerusalem. Bar-Nathan, R. and adato, M. 1986. Pottery (of the Promontory Palace). in: levine, l.i. and Netzer, e. Excavations at Caesarea-Maritima 1975, 1976, 1979, Final Report. (Qedem 21) Jerusalem. pp. 160-175. Barag, D. 1970a. Glass pilgrim vessels from Jerusalem. Journal of Glass Studies 12:35-63. Barag, D. 1970b. Glass Vessels of the Roman and Byzantine Period in Palestine. (Ph.D. dissertation, the Hebrew University of Jerusalem) Jerusalem. (Hebrew) Barag, D. 1974. a tomb of the Byzantine period near Netiv Ha-lamed He. >Atiqot 7:81-88. (Hebrew) Barag, D. 1985. Finds from a tomb of the Byzantine period at Ma>in. liber Annuus 35:365-374. Barag, D. 1988-89. the islamic candlestick coins of Jerusalem. Israel Numismatic Journal 10:40-48. Barag, D. and Hershkovitz, M. 1994. lamps from Masada. in: aviram, J., Foerster, G. and Netzer, e. Masada IV. The yigael yadin Excavations 1963-1965, Final Reports. Jerusalem. pp. 1-147. Baramki, D.C. 1932. Note on a cemetery at Karm al-Sheikh. Quarterly of the department of Antiquities in Palestine 1:2-9. Baramki, D.C. 1936. two Roman cisterns at Beth Nattif. Quarterly of the department of Antiquities in Palestine 5:3-10. Baramki, D.C. 1944. the pottery from Khirbet el-Mefjer. Quarterly of the department of Antiquities in Palestine 10:65-103. Baruch, Y. 1998. Kh. Bet-einun – a Christian holy place at Mt. Hebron during the Byzantine period. Judea and Samaria Research Studies 7:169-179 (Hebrew) Bass, G.F. 1982. the pottery. in: Bass, G.F. and van Doorninck, F.H., eds. yassi Ada: A Seventh Century Byzantine Shipwreck. College Station, tX. pp. 155-188. Batz, S. 2002. the church of St. theodore at Khirbet Bet Sila. Israel Musuem Studies in Archaeology 1:39-54. Batz, S. and Sharukh, i. 2008. Dawwara, Khirbet ed- (South). in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 5. Jerusalem. pp. 1690-1691. Ben-arieh, R. 1997. the Roman, Byzantine and Umayyad pottery. in: Hirschfeld, Y. The Roman Baths at Hammath Gader. Jerusalem. pp. 347-381. Ben David, H. 1998. Oil presses and oil production in the Golan in the Mishnaic and talmudic periods. >Atiqot 34:1-61. (Hebrew) Berman, a. 1976. Islamic Coins. Jerusalem. Berman, a. 1997. excavation of the Courthouse Site at akko. >Atiqot 31:91-103. Beyer, G. 1931. Die Stadtgebeite von eleutheropolis im 4. Jahr h. n. Chr. und seine Grenznachbarn. Zeitschrift des deutschen Palästina-Vereins 54:209-277. 232 referenCeS Beyer, G. 1933. Die Stadtgebeite von Diospolis und Nikopolis im 4. Jahr h. n. Chr. und ihre Grenznachbarn. Zeitschrift des deutschen Palästina-Vereins 56:218-253. Bijovsky, G. 2000-02. the currency of the fifth century C.e. in Palestine – some reflections in light of the numismatic evidence. Israel Numismatic Journal 14:196-210. Birger, R. 1981. Pottery and miscellaneous finds of the Byzantine period. in: Netzer, e. Greater Herodium. (Qedem 13) Jerusalem. pp. 75-77. Bitton-ashkelony, B. 2005. Encountering the Sacred. The debate on Christian Pilgrimage in late Antiquity. los angeles. Bliss, F.J. and Dickie, a.C. 1898. Excavations at Jerusalem 1894-1897. Jerusalem. Bliss, F.J. and Macalister, R.a.S. 1902. Excavations in Palestine during the years 1898-1900. london. Boas, J.a. 2000a. Pottery and small finds from the late Ottoman village and early Zionist settlement. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 547-580. Boas, J.a. 2000b. Medieval and Post-Medieval Finds. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 211-225. Boessneck, J. 1969. Osteological differences between sheep (Ovis aries linne) and goat (Capra hircus linne). in: Brothwell, D. and Higgs, H., eds. Science in Archaeology. london. pp. 331-358. Bonify, M. and Piéri, D. 1995. amphores du ve au viie s. à Marseille: nouvelles données sur la typologie et la contenu. Journal of Roman Archaeology 8:94-120. Brenk, B. 2004. Monasteries as rural settlements: Patron-dependence or self-sufficiency? in: Bowden, W., lavan, l. and Machado, C., eds. late Antique Archaeology, Vol. 2: Recent Research on the late Antique Countryside. leiden. pp. 447-475. Brosh, N. 1986. Pottery of the 8th-13th century C.e. (Strata 1-3). in: levine, l.i. and Netzer, e. Excavations at Caesarea-Maritima 1975, 1976, 1979, Final Report. (Qedem 21) Jerusalem. pp. 66-89. Broshi, M. 1980. the population of western Palestine in the Roman-Byzantine period. Bulletin of the American Schools of Oriental Research 236:1-10. Buchennino, a. 2007. Building remains and industrial installations from the early islamic period at Khirbet Deiran, Reúovot. >Atiqot 56:119-144. (Hebrew) Bull, G. and Payne, S. 1982. ageing and sexing animal bones from archaeological sites. in: Wilson, B., Grigson, C. and Payne, S., eds. Animals and Archaeology. (British archaeological Reports international Series 163). Oxford. pp. 55-71. Butler, H.C. 1969. Early Churches in Syria, Fourth to Seventh Centuries. amsterdam. Cahill, J.M. 1992. Chalk vessel assemblages of the Persian/Hellenistic and early Roman periods. in: De Groot, a. and ariel, D.t. Excavations at the City of david 1978-1985 directed by yigal Shiloh, Vol. III. (Qedem 33) Jerusalem. pp. 190-274. Calderon, R. 1999. the pottery. in: Hirschfeld, Y. The Early Byzantine Monastery at Khirbet ed-deir in the Judaean desert: The Excavations in 1981-1987. (Qedem 38) Jerusalem. pp. 135-147. Calderon, R. 2000. Roman and Byzantine pottery. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 91-165. Clamer, Ch. 1997. Fouilles archéologiques de >Aïn ez-Zâra/Callirrhoé, villégiature hérodienne. Beirut. Coates-Stephens, R. 2003. attiudes to spolia in some late antique texts. in: lavan, l. and Bowden, W., eds. late Antique Archaeology, Vol. 1: Theory and Practice in late Antique Archaeology. leiden. pp. 341-358. Coen Uzzielli, t. 1997. Marble decorations, wall mosaics and small finds. in: Hirschfeld, Y. The Roman Baths at Hammath Gader. Jerusalem. pp. 442-455. Cohen Finkelstein, J. 1997. the islamic pottery from Khirbet abu Suwwana. >Atiqot 32:19*-34*. Colt, H.D. 1962. Excavations at Nessana, Vol. 1. london. 233 Conder, C.R. and Kitchener, H.H. 1882. Survey of Western Palestine, Vol. II: Samaria. london. Conder, C.R. and Kitchener, H.H. 1883. Survey of Western Palestine, Vol. III: Judaea. london. Constantelos, D.J. 1972. the Muslem conquests of the Near east as revealed in the Greek sources of the seventh and the eighth centuries. Byzantion 42:325-357. Cope, C.R. 1999. Faunal remains and butchery practises from Byzantine and islamic contexts (1993-94 Seasons). in: Holum, K.G., Raban, a. and Patrich, J., eds. Caesarea Papers, Vol. 2: Herod’s Temple, the Provincial Governor’s Praetorium and Granaries, the later Harbor, a Gold Coin Hoard and Other Studies. (JRa Supplementary Series 35) Portsmouth. pp. 405-417. Corbo, v. 1955. Gli scavi di Kh. Siyar el-Ghanam (Campo dei Pastori) e i monasteri dei dintorni. Jerusalem. Crawford, J.S. 1990. The Byzantine Shops at Sardis. Harvard. Creson, B.C. 1999. the monastery. in: Beit arieh, i., ed. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology, tel aviv University No. 15) tel aviv. pp. 88-96. Crowfoot, J.W., Crowfoot, G.M. and Kenyon, K.M. 1957. The Objects of Samaria. london. Crowfoot, J.W. and Fitzgerald, G.M. 1929. Excavations in the Tyropoeon Valley, Jerusalem 1927. (annual of the Palestine exploration Fund 5) Jerusalem. Dagan, Y. 1991. Bet Shemesh map, survey. Excavations and Surveys in Israel 10:141-142. Dagan, Y. 2006. Archaeological Survey of Israel: Map of Ama½ya (109), Vol. 1: The Northern Sector. Jerusalem. Dagan, Y. Forthcoming. Ramat Bet Shemesh I. landscape of Settlement from Paleolithic Times to the Ottoman Period. (iaa Reports) Jerusalem. Dahari, U. 2000. Monastic Settlements in South Sinai in the Byzantine Period. The Archaeological Remains. (iaa Reports 9) Jerusalem. Dahari, U. 2003. the excavations at îorvat îani. Qadmoniot 126:102-106. (Hebrew) Dahari, U. and >ad, U. 2000. Shoham Bypass Road. Excavations and Surveys in Israel 20:56*-59*. Damati, e. 2002. the irrigation systems in the gardens of the monastery of St. Martyrius (Ma>ale adummim). in: amit, D., Patrich, J. and Hirschfeld, Y., eds. The Aqueducts of Israel. (Journal of Roman archaeology Supplementary Series 46) Portsmouth. pp. 438-444. Dan, Y. 1976. Social life in Eretz-Israel in the Byzantine Period in the Sixth and Seventh Centuries. (Ph.D. dissertation, the Hebrew University of Jerusalem) Jerusalem. (Hebrew) Dan, Y. 1984. the soils of israel and the connection between them and the natural vegetation. in: Waisel, Y., ed. The Plants and Animals of the land of Israel, an Illustrated Encyclopedia, Vol. 8: Vegetation of Israel. tel aviv. pp. 28-39. (Hebrew) Dar, S. 1982. The Settlement Pattern of Western Samaria in the Periods of the Second Temple, Mishnah, the Talmud and the Byzantine Period. tel aviv (Hebrew) Dar, S. 1986. landscape and Pattern. An Archaeological Survey of Samaria, 800 B.C.E. – 636 C.E. (British archaeological Reports international Series 308) Oxford. Dar, S. 1999. Sumaqa. A Roman and Byzantine Jewish Village on Mount Carmel, Israel. (British archaeological Reports international Series 815) Oxford. Dauphin, C. 1991. the excavations of a Byzantine site at Khirbet Jannaba et-taúta. >Atiqot 20:111-117. Dauphin, C. and Kingsley, S.a. 2003. Ceramic evidence for the rise and fall of a late ecclesiastical estate at Shelomi in Phoenicia Maritima. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 61-74. Dayan, t. 1999. Faunal remains: areas a-G. in: in: Beit arieh, i., ed. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology, tel aviv University No. 15) tel aviv. pp. 480-487. Decker, M. 2005. the wine trade of Cilicia in late antiquity. Aram 17:51-59. 234 referenCeS Decker, M. 2007. Water into wine: trade and technology in late antiquity. in: lavan, l., Zanini, e. and Sarantis, a., eds. late Antique Archaeology, Vol. 4: Technology in Transition A.d. 300-650. leiden and Boston. pp. 65-91. Delougaz, P. and Haines, R.C., eds. 1960. A Byzantine Church at Khirbat al-Karak. (the University of Chicago Oriental institute Publications 65) Chicago. Dembińska, M. 1985. Diet: A comparison of food consumption between some eastern and western monasteries in the 4th-12th centuries. Byzantion 55:431-462. de vaux, R. and Steve, a.M. 1950. Fouilles à Qaryet el-‘Enab, Abū Gôsh, Palestine. Paris. de vincenz, a. 2003. Ceramics from the 5th-8th seasons (1994-1998). Campaigns 1994 and 1996. in: Strus, a., ed. Khirbet Fattir – Bet Gemal. Two Ancient Jewish and Christian Sites in Israel. Roma. pp. 249-381. de vincenz, a. 2005. the pottery from Khirbet el-Jiljil (first season). Bulletin of the Anglo-Israel Archaeological Society 23:111-138. de vincenz, a. 2007. the pottery. in: Hirschfeld, Y. En-Gedi Excavations II. Final Report (1996-2002). Jerusalem. pp. 234-427. Di Segni, l. 1999. epigraphic documentation on building in the provinces of Palaestina and arabia, 4th-7th c. in: Humphrey, J.H., ed. The Roman and Byzantine Near East: Some Recent Archaeological Research, Vol. 2. (Journal of Roman archaeology Supplementary Series 31) ann arbor. pp. 149-178. Di Segni, l. 2001. Monks and society: the case of Palestine. in: Patrich, J., ed. The Sabaite Heritage in the Orthodox Church from the Fifth Century to the Present. leuven. pp. 31-36. Di Segni, l. 2003. a Greek inscription in the church at îorvat îanot. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 273-276. Di Segni, l. 2005. an inscription from Khirbet el-Jiljil. Bulletin of the Anglo-Israel Archaeological Society 23:101-105. Di Segni, l. 2007. a fragment of an ampulla. in: Mazar, e. The Temple Mount Excavations in Jerusalem 19681978, directed by Benjamin Mazar: The Byzantine Period. (Qedem 46) Jerusalem. pp. 35-36. Di Segni, l. 2008. the Greek inscription from tel ashdod: a revised reading. >Atiqot 58:31*-36*. Di Segni, l. and Gibson, S. 2007. Greek inscriptions from Khirbet el-Jiljil and Beth Jimal and the identification of Caphar Gamala. Bulletin of the Anglo-Israel Archaeological Society 25: 117-145. Dorsey, D.a. 1991. The Roads and Highways of Ancient Israel. Baltimore and london. Drake, F.M. 1919. an early Christian mosaic at Deir Dakleh. The Burlington Magazine for Connoisseurs 34: 145. Dray, Y. 1994. the oil presses of Zur Natan. Reports on TFAHR Excavations at: Zur Natan, Israel; Silistra, Bulgaria and Ulanci, Macedonia. Houston, tX. pp. 14-15. eisenberg, e. 2000. Naúal Yarmut. îadashot Arkheologiyot – Excavations and Surveys in Israel 112:91*-93*. eisenberg, e. and Ovadiah, R. 1998. a Byzantine monastery at Mevo Modi>im. >Atiqot 36:1*-19*. (Hebrew) elgavish, J. 1977. Archaeological Excavations at Shiqmona, Vol. 3: The Pottery of the Roman Period. Haifa. (Hebrew) elgavish, J. 1994. Shiqmona on the Seacoast of Mount Carmel. tel aviv. (Hebrew) ellenblum, R. 1998. Frankish Rural Settlement in the latin Kingdom of Jerusalem. Cambridge. epstein, C. 1993. Hippos (Sussita). in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 2. Jerusalem. pp. 634-636. eshel, H., Magness, J. and Shenhav, e. 2000. Khirbet Yattir, 1995-1999: Preliminary report. Israel Exploration Journal 50:153-168. Fabian, P. and Goldfus, H. 2004. a Byzantine farmhouse, terraces and agricultural installations at the Goral Hills near Be’er Sheva. >Atiqot 47:1*-14*. (Hebrew) Fantalkin, a. 2000. a salvage excavation at a 6th-7th century C.e. Site on Palmach Street, Beersheba. Tel Aviv 27:257-272. 235 Fatal, a. 1958. le Statut légal des non-musulmans en pays d’Islam. Beyrouth. Feig, N. 1985. Pottery, glass and coins from Magen. Bulletin of the American Schools of Oriental Research 258:33-40. Feig, N. 2003. excavations at Beth Safafa: iron age ii and Byzantine agricultural installations south of Jerusalem. >Atiqot 44:191-238. Fiema, Z.t. 2003. the Byzantine monastic/pilgrimage centre of St. aaron near Petra, Jordan. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 343-357. Fiema, Z.t. 2006. City and countryside in Byzantine Palestine. Prosperity in question. in: lewin, a.S. and Pellegrini, P., eds. Settlement and demography in the Near East in late Antiquity. (Proceedings of the Colloquium, Matera 27-29 October 2005. Pisa and Rome) pp. 67-88. Figueras, P. 1995. Monks and monasteries in the Negev desert. liber Annuus 45:401-450. Figueras, P. 2004a. Sculptured building stones. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev (Final Report of the Excavations 1989-1995) (Beer-Sheva 16/Beer-Sheva archaeological Monographs 1) Beer-Sheva. pp. 109-122. Figueras, P. 2004b. Pottery marks. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev. Final Report of the Excavations 1989-1995. (Beer-Sheva 16/Beer-Sheva archaeological Monographs 1). Beer-Sheva. pp. 210-215. Fischer, M. 1989. an early Byzantine Settlement at Kh. Zikrin (israel). Actes du Congrès International d’Archéologie Chrétienne, lyon, Vienne, Grenoble, Genève et Aoste, 21-28 Septembre. Rome. pp. 1787-1807. Fischer, M. and tal, O. 1999a. the Byzantine and early islamic periods. in: Beit arieh, i. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology of tel aviv University No. 15) tel aviv. pp. 300-345. Fischer, M. and tal, O. 1999b. Pottery from Har Bariyaú (area F). in: Beit arieh, i. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology of tel aviv University No. 15) tel aviv. pp. 346-349. Fischer, M. and tal, O. 1999c. Stone artefacts from the Byzantine period. in: Beit arieh, i. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology of tel aviv University No. 15) tel aviv. pp. 421-427. Fischer, M. and tal, O. 2000. Pottery. in: Fischer, M., tal, O. and Gichon, M., eds. >En Boqeq. Excavations in an Oasis on the Dead Sea, Vol. 2: The Oficina, an Early Roman Building on the Dead Sea Shore. Mainz am Rhein. pp. 29-67. Fitzgerald, G.M. 1939. A Sixth Century Monastery at Beth-Shan. Philadelphia. Foerster, G. 1977. Yeúi>am. îadashot Arkheologiyot 63-64:11. (Hebrew) Forsyth, G.H. and Weizmann, K. 1970. The Monastery of Saint Catherine at Mount Sinai. The Church and Fortress of Justinian. ann arbor. Fowler, a. 1990. the pottery of Zur Natan. in: Matthews, e., Neidinger, W. and ayalon, e. Preliminary Report on the Excavations at Zur Natan, 1989 and 1990 Seasons. Houston, tX. pp. 34-51. Frankel, R. 1992. Some oil presses from western Galilee. Bulletin of the American Schools of Oriental Research 286:39-71. Frankel, R. 1994. ancient oil mills and presses in the land of israel. in: Frankel, R., avitsur, S. and ayalon, e. History and Technology of Olive Oil in the Holy land. tel aviv. pp. 19-90. Frankel, R. 1999. Wine and Oil Production in Antiquity in Israel and Other Mediterranean Countries. Sheffield. Frankel, R. 2003a. Mills and querns in talmudic literature – a reappraisal in light of archaeological evidence. Cathedra 110:43-60. (Hebrew) 236 referenCeS Frankel, R. 2003b. the Olynthus mill, its origin, and diffusion: typology and distribution. American Journal of Archaeology 107:1-21. Frankel, R. and ayalon, e. 1988. Vines, Wine presses and Wine in Antiquity. tel aviv. (Hebrew) Frankel, R., Patrich, J. and tsafrir, Y. 1990. the oil press at îorvat Beth loya. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 287-300. Fritz, v. 1983. Die nestorianische Kloster (areal D): Die architektur. in: Fritz, v. and Kempinski, a., eds. Ergebnisse der Ausgrabungen auf der Hirbat el-Mšāš (Tel Māśōś) 1972-1975. Wiesbaden. pp. 138-153. Gadot, Y. and tepper, Y. 2003. a late Byzantine pottery workshop at Khirbet Baraqa. Tel Aviv 30:130-162. Gal, Z. 1986. vineyard cultivation at >emek îarod and its vicinity during the Roman-Byzantine period. Israel – People and land 2-3 :129-138. (Hebrew) Gass, e. and Zissu, B. 2005. the Monastery of Samson up the Rock of etham in the Byzantine period. Zeitschrift des deutschen Palästina-Vereins 121:168-183. Gayraud, R.P. 2003. la transition céramique en egypte, viie-iXe siècles. in: Bakirtzis, Ch., ed. Actes du VIIe Congrès International sur la Céramique Médiévale en Méditerranée, Thessaloniki, 11-16 Octobre 1999. athens. pp. 558-562. Geiger, J. 1982. the spreading of Christianity in eretz israel – from its beginning to the times of Julian. in: Baras, Z., Safrai, S., tsafrir, Y. and Stern. M., eds. Eretz Israel, from the destruction of the Second Temple to the Muslim Conquest, Vol. 1: Political, Social and Cultural History. Jerusalem. pp. 218-235. (Hebrew) Gershuny, l. 2006. excavations at Khirbet Marmita. >Atiqot 53:139-178. Geva, H. and Hershkovitz, M. 2006. local pottery of the Hellenistic and early Roman periods. in: Geva, H., ed. Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 19691982. Vol. 3: Area E and Other Studies, Final Report. Jerusalem. pp. 94-143. Geva, H. and Rosenthal-Heginbottom, R. 2003. local pottery from area a. in: Geva, H., ed. Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 1969-1982. Vol. 2: The Finds from Areas A, W and X-2, Final Report. Jerusalem. pp. 176-191. Gibson, S. 1985-6. Ras et-tawil: a Byzantine monastery north of Jerusalem. Bulletin of the Anglo-Israel Archaeological Society 6:69-73. Gichon, M. 1993. >En Boqeq, Ausgrabung in einer Oase am Toten Meer, Band 1: Geographie und Geschichte der Oase. das spätrömanisch-byzantinische Kastell. Mainz am Rhein. Gil, M. 1992. A History of Palestine, 634-1099. Cambridge. Glick, D. 2006. a salvage excavation at >ein ez-Zeituna in Naúal >iron. >Atiqot 51:31-70. Goldfus, H. 1998. Tombs and Burials in Churches and Monasteries of Byzantine Palestine (324-628 A.d.). ann arbor. Goldsmith, D., Ben-Dov, R. and Kertesz, t. 1999. Miscellaneous finds. in: Beit arieh, i. Tel >Ira: A Stronghold in the Biblical Negev. (Monograph Series of the institute of archaeology of tel aviv University No. 15). tel aviv. pp. 445-475. Goodwin, t. 2002. arab-Byzantine coinage. in: album, S. and Goodwin, t., eds. Sylloge of Islamic Coins in the Ashmolian, Vol. 1: The Pre-Reform Coinage of the Early Islamic Period. Oxford. pp. 74-109. Gophna, R. and Feig, N. 1993. a Byzantine monastery at Kh. Jemameh. >Atiqot 22:97-108. Gophna, R. and taxel, i. 2007. Pottery. in: Gophna, R., taxel, i. and Feldstein, a. Kafr >ana: a Rural Settlement in the lod valley. Salvage Excavation Reports 4:33-65. Gophna, R. and Zvilichovsky, v. 1959. Roglit. Israel Exploration Journal 9:143. Goren, Y. 2005. appendix: the pottery technology. in: arubas, B. and Goldfus, H., eds. Excavations on the Site of the Jerusalem International Convention Centre (Binyanei Ha’uma): A Settlement of the late First to Second Temple Period, the Tenth legion’s Kilnworks, and a Byzantine Monastic Complex. The Pottery and Small Finds. (Journal of Roman archaeology Supplementary Series 60) Portsmouth. pp. 192-194. 237 Gorin-Rosen, Y. 1999. Glass vessels from Ras abu Ma>aruf (Pisgat Ze<ev east a). >Atiqot 38:205-214. Gorin-Rosen, Y. 2000. the glass vessels from Khirbet tabaliya (Giv>at Hamatos). >Atiqot 40:81*-95*. (Hebrew) Govrin, Y. 1991. Archaeological Survey of Israel: Map of Naúal yattir (139). Jerusalem. Greenhut, Z. 1998. îorvat îermeshit (1988-1990). >Atiqot 34:121-172. (Hebrew) Greenhut, Z. 2001. Maúseya. îadashot Arkheologiyot – Excavations and Surveys in Israel 113:124*. Greenhut, Z. 2004. early islamic remains at îorvat >illin (Upper). >Atiqot 47:15*-32*. (Hebrew) Greenhut, Z., Weiss, D. and Solimany, G. 2000. Maúseya. îadashot Arkheologiyot – Excavations and Surveys in Israel 111:105*. Grierson, P. 1968. Catalogue of the Byzantine Coins in the dumbarton Oaks Collection and in the Whittemore Collection. Washington. Gudovitch, S. 1996. a Byzantine building at the foot of îorbat Sokho. >Atiqot 28:17*-23*. (Hebrew) Gudovitch, S. 1999. Khirbet >asfura. îadashot Arkheologiyot – Excavations and Surveys in Israel 109:79*-81*. Guidoboni, e. 1994. Catalogue of Ancient Earthquakes in the Mediterranean Area up to the 10th Century. Rome. Habas, l. 1999. the marble furniture. in: Hirschfeld, Y. The Early Byzantine Monastery at Khirbet ed-deir in the Judaean desert: The Excavations in 1981-1987. (Qedem 38) Jerusalem. pp. 119-134. Hachlili, R. 2001. The Menorah, the Ancient Seven-Armed Candlebraum: Origin, Form and Signiicance. leiden. Hadad, S. 1998. Glass lamps from the Byzantine through Mamluk periods at Bet Shean, israel. Journal of Glass Studies 40:63-76. Hadad, S. 2002. The Oil lamps from the Hebrew University Excavations at Bet Shean. (Qedem Reports 4) Jerusalem. Hadad, S. 2003. Glass lamps from the “House of the Menorot” in area vi. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 191-195. Hadad, S. 2005. Islamic Glass Vessels from the Hebrew University Excavations at Bet Shean. (Qedem Reports 8) Jerusalem. Haddad, e. 2009. Roman Byzantine amphorae from a terrestrial site and its underwater environs: îorbat Castra and Kfar Samir (southern levant) as a case study. levant 41:79-91. Haiman, M. 1995. an early islamic period farm at Naúal Mitnan in the Negev highlands. >Atiqot 26:1-13. Hamilton, J.a. 1933. Byzantine Architecture and decoration. london. Hamilton, R.W. 1935. Note on a chapel and wine press at >ain el-Jedide. Quarterly of the department of Antiquities in Palestine 4:111-117. Hamilton, R.W. 1944. excavations against the North Wall of Jerusalem, 1937-1938. Quarterly of the department of Antiquities in Palestine 10:1-54. Harden, D.B. 1936. Roman Glass from Karanis Found by the University of Michigan Archaeological Expedition in Egypt, 1924-29. ann arbor. Harding, G.l. 1951. excavations on the Citadel, amman. Annual of the department of Antiquities of Jordan 1:7-16. Harper, R.P. 1995. Upper Zohar, an Early Byzantine Fort in Palaestina Tertia. Final Report of Excavations in 1985-1986. Oxford. Harrell, J.a. and Brown, J.M. 2008. Discovery of a Medieval islamic industry for steatite cooking vessels in egypt’s eastern desert. in: Rowan, Y.M. and ebeling, J.R., eds. New Approaches to Old Stones: Recent Studies on Ground Stone Artefacts. london and Oakville. pp. 41-65. Harrison, t.P. 1994. a sixth-seventh century ceramic assemblage from Madaba, Jordan. Annual of the department of Antiquities of Jordan 38:429-446. Hartal, M. 2005. land of the Ituraeans. Archaeology and History of the Northern Golan in the Hellenistic, Roman and Byzantine Periods. Qazrin. (Hebrew) 238 referenCeS Hayes, J.W. 1972. late Roman Pottery. london. Hayes, J.W. 1985. Sigillata orientali. Encyclopedia dell’arte antica: Atlante delle forme ceramiche, Vol. 2. Rome. Heideman, S. 2003. Der Kleingeldumlauf in der Gazira in früh->abbasidischer Zeit. in: Heideman, S. and Becker, a., eds. Raqqa II – die islamischer Stadt. Mainz. pp. 141-161. Heker, D. 1996. animal bones and molluscs. in: Nahlieli, D., israel, Y. and Ben-Michael, Y. the Naúal la>ana Site: an early islamic farm in the Negev. >Atiqot 30:74*-75*. (Hebrew) Heker, D. 2004. animal remains from the village in Upper Naúal Besor. >Atiqot 48:123*-125*. (Hebrew) Hellwing, S. and Gophna, R. 1984. the animal remains from the early and Middle Bronze ages at tel aphek and tel Dalit: a comparative study. Tel Aviv 11:48-58. Hervé, H. and Zelinger, Y. 2006. a Byzantine monastery at Naúal Kidron. New Studies on Jerusalem 11:289-295. (Hebrew) Hershkovitz, M. 1992. aroer at the end of the Second temple period. Eretz-Israel 23:309-319. (Hebrew) Hirschfeld, Y. 1981. ancient wine presses in the area of the ayalon Park. Eretz-Israel 15:383-390. (Hebrew) Hirschfeld, Y. 1985a. Archaeological Survey of Israel: Map of Herodium (108/2). Jerusalem. Hirschfeld, Y. 1985b. Khirbet el-Quneitra: a Byzantine monastery in the Wilderness of Ziph. Eretz-Israel 18:243-255. (Hebrew) Hirschfeld, Y. 1988-1989. îorvat >amude Qerayot. Excavations and Surveys in Israel 7-8:3-4. Hirschfeld, Y. 1990. list of the Byzantine monasteries in the Judaean desert. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 1-90. Hirschfeld, Y. 1992. The Judaean desert Monasteries in the Byzantine Period. New Haven and london. Hirschfeld, Y. 1993. euthymius and his monastery in the Judaean desert. liber Annuus 43:339-371. Hirschfeld, Y. 1995. The Palestinian dwelling in the Roman-Byzantine Period. Jerusalem. Hirschfeld, Y. 1997. Jewish rural settlement in Judaea in the early Roman period. in: alcock, S.e., ed. The Early Roman Empire in the East. Oxford. pp. 72-88. Hirschfeld, Y. 1999. The Early Byzantine Monastery at Khirbet ed-deir in the Judaean desert: The Excavations in 1981-1987. (Qedem 38) Jerusalem. Hirschfeld, Y. 2000a. îorvat >aqav: architecture and stratigraphy. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 13-87. Hirschfeld, Y. 2000b. îorvat >eleq: architecture and stratigraphy. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 235-370. Hirschfeld, Y. 2000c. General discussion: Ramat Hanadiv in context. in: Hirschfeld, Y. Ramat Hanadiv Excavations. Final Report of the 1984-1988 Seasons. Jerusalem. pp. 679-737. Hirschfeld, Y. 2001-2002. the Monastery of Marda: Masada in the Byzantine period. Bulletin of the AngloIsrael Archaeological Society 19-20:119-156. Hirschfeld, Y. 2002a. Deir Qal>a and the monasteries of western Samaria. in: Humphrey, J.H., ed. The Roman and Byzantine Near East, Vol. 3: late-Antique Petra, Nile Festival Building at Sepphoris, deir Qal>a Monastery, Khirbet Qana Village and Pilgrim Site, >Ain >Arrub Hiding Complex and other Studies. (Journal of Roman archaeology Supplement 49) ann arbor. pp. 155-189. Hirschfeld, Y. 2002b. The desert of the Holy City. The Judaean desert Monasteries in the Byzantine Period. Jerusalem. (Hebrew) Hirschfeld, Y. 2002c. the water supply of the monastery of Chariton. in: amit, D., Patrich, J. and Hirschfeld, Y., eds. The Aqueducts of Israel. (Journal of Roman archaeology Supplementary Series 46) Portsmouth. pp. 428-437. Hirschfeld, Y. 2004a. the monasteries of Gaza: an archaeological review. in: Bitton-ashkeloni, B. and Kofsky, a., eds. Christian Gaza in late Antiquity. leiden. pp. 61-88. 239 Hirschfeld, Y. 2004b. Excavations at Tiberias, 1989-1994. (iaa Reports 22) Jerusalem. Hirschfeld, Y. 2005. the expansion of rural settlement during the fourth-fifth centuries Ce in Palestine. in: lefort, J., Morrison, C. and Sodini, J.-P., eds. les villages dans l’empire byzantine (IVe-XVe siècle). Paris. pp. 523-537. Hirschfeld, Y. 2006. the monasteries of Palestine in the Byzantine period. in: limor, O. and Stroumsa, G.G., eds. Christians and Christianity in the Holy land: From the Origins to the latin Kingdom. turnhout. pp. 401-419. Hizmi, H. 1997. a burial cave in talluza. >Atiqot 32:119-123. (Hebrew) Horwitz, l.K. 1995. Fauna from the Naúal Mitnan farm. >Atiqot 26:15-19. Horwitz, l.K. 1998. animal exploitation during the early islamic period in the Negev: the fauna from elatelot. >Atiqot 36:27-38. Hull, D. 2008. a spatial and morphological analysis of monastic sites in the northern limestone massif, Syria. levant 40:89-113. Hütteroth, D.H. and abdulfatah, K. 1977. Historical Geography of Palestine, Transjordan and Southern Syria in the late Sixteenth Century. erlangen. ilan, Z. 1986. the excavation in the western church at î. >eirav (a>iribin). in: Yedaya, M., ed. The Western Galilee Antiquities. tel aviv. pp. 503-515. (Hebrew) ilan, Z. 1991. Ancient Synagogues in Israel. tel aviv. (Hebrew) ilisch, l. 1993. Sylloge Numorum Arabicorum Tübingen. Palästina IV a Bilād aš-Šām I. tübingen. israel, Y., Nahlieli, D. and Ben Michael, Y. 1995. the Naúal Shaúaq site: an early islamic settlement in the northern arava. >Atiqot 26:1*-14*. (Hebrew) israeli, Y. 2002. Pilgrimage to the Holy land. in: israeli, Y. and Mevorah, D., eds. Cradle of Christianity. Jerusalem. pp. 186-212. Jackson-tal, R.e. 2007. Glass vessels from en-Gedi. in: Hirschfeld, Y., ed. En-Gedi Excavations II. Final Report (1996-2002). Jerusalem. pp. 474-506. Kagan, e.J., agnon, a., Bar-Matthews, M. and ayalon, a. 2005. Dating large infrequent earthquakes by damaged cave deposits. Geology 33:261-264. Kalman, Y. 2001. Olive oil press from the Byzantine period at lower Herodium. Judea and Samaria Research Studies 10:143-146. (Hebrew) Karmon, Y. 1957. topographical influences on the roads in Judaea. Judah and Jerusalem 12:144-150. (Hebrew) Katsnelson, N. 1999a. Glass vessels from the Painted tomb at Migdal ashqelon. >Atiqot 37:67*-82*. Katsnelson, N. 1999b. Glass vessels. in: Kogan-Zehavi, e. late Roman-Byzantine Remains at ashqelon. >Atiqot 38:119*-121*. (Hebrew) Katsnelson, N. 2002. the glass ornaments. excavations in Khirbet el-Shubeika 1991, 1993. in: Gal, Z., ed. Eretz Zafon: Studies in Galilean Archaeology. Jerusalem. pp. 322-331. (Hebrew) Katsnelson, N. 2004. Glass objects. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev (Final Report of the Excavations 1989-1995). (Beer-Sheva 16/Beer-Sheva archaeological Monographs 1). Beer-Sheva. pp. 265-291. Katsnelson, N. and Jackson-tal, R.e. 2004. the glass vessels from ashqelon, Semadar Hotel. >Atiqot 48:99-109. Katz, K., Kahane, P.P. and Broshi, M. 1968. From the Beginning: Archaeology and Art in the Israel Museum, Jerusalem. New York. Kelso, J.l. and Baramki, D.C. 1955. Excavations at New Testament Jericho and Khirbet en-Nitla. (annual of the american Schools of Oriental Research 29-30). New Haven. Kennedy, H. 1991. Nomads and settled people in Bilād al-Shām in the third/ninth and fourth/tenth centuries. in: Bakhit, M.a. and Schick, R., eds. Bilād al-Shām during the Abbasid Period (132 A.H/750 A.D-451 A.H./1059 A.d.). (Proceedings of the 5th International Conference on the History of Bilād al-Shām) 240 referenCeS amman. pp. 105-113. Kennedy, H. 2004. elite incomes in the early islamic state. in: Haldon, J. and Conrad, l.i., eds. The Byzantine and Early Islamic Near East, Vol. 6: Elites Old and New in the Byzantine and Early Islamic Near East. Princeton. pp. 13-28. Kenneth Sems, G. 1982. the weighting implements. in: Bass, G.F. and van Doorninck, F.H., eds. yassi Ada: A Seventh Century Byzantine Shipwreck. College Station, tX. pp. 202-230. Khalidi, W. 1992. All That Remains. The Palestinian Villages Occupied and depopulated by Israel in 1948. Washington. Khalil, l. and Kareem, J. 2002. >abbasid pottery from area e at Khirbat Yajuz, Jordan. levant 34:111-150. Khamis, e. 1996. the metal objects. in: Ben-tor, a., avissar, M. and Portugali, Y. yoqne>am I, the late Periods. (Qedem Reports 3) Jerusalem. pp. 218-235. Khamis, e. 1998. Weights and Balances of the Byzantine and Umayyad Periods from Beth Shean. (M.a. thesis, the Hebrew University of Jerusalem) Jerusalem. (Hebrew) Khroushkova, l.G. 1998. les églises de l’époque Justinienne en Colchide. in: Cambi, N. and Marin, e., eds. Acta XIII Congressus Internationalis Archaeologiae Christianae, Split-Poreč (25.9-1.10 1994). Split. pp. 823-836. Killebrew, a. 1999. the pottery. in: Hachlili, R. and Killebrew, a. Jericho, the Jewish Cemetery of the Second Temple Period. (iaa Reports 7) Jerusalem. pp. 115-133. Kingsley, S.a. 1994-95. Bag-shaped amphorae and Byzantine trade: expanding horizons. Bulletin of the Anglo-Israel Archaeological Society 14:39-56. Kingsley, S.a. 1999. Specialized Production and long-distance Trade in Byzantine Palestine. (Ph.D. dissertation, University of Oxford) Oxford. Kingsley, S.a. 2002. A Sixth-Century Ad Shipwreck off the Carmel Coast, Israel. dor d and the Holy land Wine Trade. (British archaeological Reports international Series 1065) Oxford. Kingsley, S.a. and Decker, M. 2001. New Rome, new theories on inter-regional exchange. an introduction to the east Mediterranean economy in late antiquity. in: Kingsley, S.a. and Decker, M., eds. Economy and Exchange in the Eastern Mediterranean during late Antiquity. Oxford. pp. 1-27. Kingsley, S.a. and Raveh, K. 1996. The Ancient Harbour and Anchorage of dor, Israel. Results of the Underwater Survey 1976-1991. (British archaeological Reports international Series 626) Oxford. Kisch, B. 1965. Scales and Weights. A Historical Outline. New Haven and london. Kletter, R. 2005. early islamic remains at >Opher Park, Ramla. >Atiqot 49:57-99. Kloner, a. 2000. Archaeological Survey of Israel: Survey of Jerusalem, the Southern Sector. Jerusalem. Kloner, a. and Sagiv, N. 2003. Subterranean complexes 44 and 45. in: Kloner, a. Maresha Excavations Final Report I: Subterranean Complexes 21, 44, 70. (iaa Reports 17) Jerusalem. pp. 51-72. Kogan-Zehavi, e. 2008. Bet Shemesh, Khirbat es-Suyyagh. îadashot Arkheologiyot – Excavations and Surveys in Israel 120 (http://www.hadashot-esi.org.il). Kolska Horwitz, l. 2006. the application of ethnographic analogy to the examination of Roman/Byzantine pastoral practises in the Mount Carmel region. in: Maeir, a.M. and de Miroschedji, P., eds. “I Will Speak the Riddles of Ancient Times”. Archaeological and Historical Studies in Honor of Amihai Mazar on the Occasion of His Sixtieth Birthday, Vol. 2. Winona lake. pp. 833-851. Kol-Yaakov, S. 2000. various objects from the Hellenistic, Roman and Byzantine periods. in: Hirschfeld, Y. Ramat Hanadiv Excavations, Final Report of the 1984-1988 Seasons. Jerusalem. pp. 473-503. Kopp, C. and Stève, a.M. 1946. le Désert de Saint Jean, près d’Hébron. Revue Biblique 53:547-575. Kraeling, C.H. 1938. Gerasa, City of the decapolis. New Haven. Krautheimer, R. 1965. Early Christian and Byzantine Architecture. Suffolk. laMotta, v.M. and Schiffer, M.B. 1999. Formation processes of house floor assemblages. in: allison, P.M., ed. The Archaeology of Household Activities. london and New York. pp. 19-29. 241 landgraf, J. 1980. Keisan’s Byzantine pottery. in: Briend, J. and Humbert, J.B., eds. Tell Keisan (1971-1976), une cité phénicienne en Galilée. Paris. pp. 51-99. lane-Poole, S. 1889. Additions to the Oriental Collections 1876-1888, Vols. I-IV. london. lecker, M. 1989. the estates of >amr b. al->Ās in Palestine: Notes on a new Negev Arabic inscription. Bulletin of the School of Oriental and African Studies 52:24-37. lehmann, G. 1996. Khirbet en-Nabi Bulus (B). Excavations and Surveys in Israel 17:89-90. lester, a. 1996. the glass from Yoqne>am: the early islamic, Crusader, and Mamluk periods. in: Ben-tor, a., avissar, M. and Portugali, Y. yoqne>am I, the late Periods. (Qedem Reports 3) Jerusalem. pp. 202-217. lester, a. 2004. the glass. in: Stacey, D. Excavations at Tiberias, 1973-1974. The Early Islamic Periods. (iaa Reports 21) Jerusalem. pp. 166-220. leszczyc, t. 2003. Ceramics from area a-B. in: Strus, a. Khirbet Fattir – Bet Gemal. Two Ancient Jewish and Christian Sites in Israel. Roma. pp. 217-248. levy, S. 1960. the ancient synagogue of Ma>on (Nirim): excavation report. in: avi-Yonah, M., ed. Bulletin III of the louis M. Rabinowitz Fund for the Exploration of Ancient Synagogues. Jerusalem. pp. 6-13. levy-Rubin, M. 2000. New evidence relating to the process of islamization in Palestine in the early Muslim period – the case of Samaria. Journal of the Economic and Social History of the Orient 43:257-276. levy-Rubin, M. 2002. the Samaritans during the early Muslim period according to the continuatio to the chronicle of abu ‘l-Fath. in: Stern, e. and eshel, H., eds. The Samaritans. Jerusalem. pp. 562-586. (Hebrew) levy-Rubin, M. 2003. the reorganization of the Patriarchate of Jerusalem during the early Muslim period. Aram 15:197-226. lightfoot, C.S. and irison, e.a. 2001. the amorium project: the 1998 excavation season. dumbarton Oaks Papers 55:371-399. limor, O. 2006. ‘Holy Journey’: Pilgrimage and Christian sacred landscape. in: limor, O. and Stroumsa, G.G., eds. Christians and Christianity in the Holy land: From the Origins to the latin Kingdom. turnhout. pp. 321-353. liphschitz, N. 2007. Timber in Ancient Israel: dendroarchaeology and dendrochronology. (Monograph Series of the institute of archaeology of tel aviv University No. 26). tel aviv. liphschitz, N., Biger, G. and Mandel, Z. 1990. Did the aleppo Pine – “Oren Yerushalaim” (Pinus halepensis) – cover the mountains of eretz israel in the past? Israel – People and land 5-6:141-150. (Hebrew) loffreda, S. 1990. the Greek inscriptions on the Byzantine lamps from the Holy land. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 475-500. loffreda, S 1996. la Ceramica di Macheronte e dell’Herodion (90 a.C.-135 d.C.). Jerusalem. lombardi, G. 1956-1957. Bolli byzantino-arabi al “Dominus Flevit”. liber Annuus 7:165-190. loverance, 1990. early Byzantine marble church furnishings: Some examples from the episcopal basilica of Kourion in Cyprus. in: Morris, R., ed. Church and People in Byzantium. (Society for the Pomotion of Byzantine Studies twentieth Spring Symposium if Byzantine Studies. Manchester, 1986) Birmingham. pp. 225-243. Macalister, R.a.S. 1889. a Byzantine church at Umm er Rus. Palestine Exploration Fund Quarterly Statement 200-204. Macalister, R.a.S. 1911. The Excavations of Gezer, 1902-1905 and 1907-1909, Vol. 1. london. Macalister, R.a.S. 1912. The Excavations of Gezer, 1902-1905 and 1907-1909, Vol. 3. london. Mackenzie, D. 1911. the excavations at ain Shems, 1911. Annual of the Palestine Exploration Fund 1911:41-94. Mader, a.e. 1918. Altchrisliche Basiliken und lokaltraditionen in Südjudäa: archäologische und topographische Untersuchungen. Paderborn. Magen, Y. 1993. the monastery of St. Martyrius at Ma>ale adummim. in: tsafrir, Y., ed. Ancient Churches Revealed. Jerusalem. pp. 170-196. Magen, Y. 2002a. The Stone Vessel Industry in the Second Temple Period. Excavations at îizma and the Jerusalem Temple Mount. Jerusalem. 242 referenCeS Magen, Y. 2002b. the areas of Samaritan settlement in the Roman-Byzantine period. in: Stern, e. and eshel, H., eds. The Samaritans. Jerusalem. pp. 245-271. (Hebrew) Magen, Y. 2004. Qalandiya – a second temple-period viticulture and wine-manufacturing agricultural settlement. in: Magen, Y., ariel, D.t., Bijovsky, G., tzionit, Y. and Sirkis, O. The land of Banjamin. Jerusalem. pp. 29-144. Magen, Y. and aizik, N. 2008. Deir Qal>a. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 5. Jerusalem. pp. 1693-1695. Magen, Y. and Baruch, Y. 1997. Khirbet abu Rish. >Atiqot 32:135-146. (Hebrew) Magen, Y. and Batz, S. 2008. Umm Deimnah, Khirbet. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 5. Jerusalem. pp. 2058-2059. Magen, Y., Har-even, B. and Sharukh, i. 2008. Qa§r, Khirbet el-. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 5. Jerusalem. pp. 1996-1997. Magen, Y., Peleg, Y. and Sharukh, i. 2008. Rujm Jerida. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 5. Jerusalem. pp. 2023-2024. Magen, Y. and talgam, R. 1990. the monastery of Martyrius at Ma>ale adummim (Khirbet el-Murassas) and its mosaics. in: Bottini, G.C., Di Segni, l. and alliata, e., eds. Christian Archaeology in the Holy land, New discoveries. Essays in Honor of Virgilio C. Corbo, OFM. Jerusalem. pp. 91-152. Magen, Y., tzionit, Y. and Sirkis, O. 2004. Khirbet Badd >isa – Qiryat Sefer. in: Magen, Y., ariel, D.t., Bijovsky, G., tzionit, Y. and Sirkis, O. The land of Banjamin. Jerusalem. pp. 179-242. Magness, J. 1992a. the late Roman and Byzantine pottery from areas 2a and G. in: De Groot, a. and ariel, D.t. Excavations at the City of david 1978-1985 directed by yigal Shiloh, Vol. III. (Qedem 33) Jerusalem. pp. 149-164. Magness, J. 1992b. late Roman and Byzantine pottery: Preliminary report, 1990. in: vann, R.l., ed. Caesarea Papers: Straton’s Tower, Herod’s Harbour, and Roman and Byzantine Caesarea. (JRa Supplementary Series 5) ann arbor. pp. 129-153. Magness, J. 1993. Jerusalem Ceramic Chronology, circa 200-800 CE. Sheffield. Magness, J. 1994. the dating of the black ceramic bowl with a depiction of the torah shrine from Nabratin. levant 26:199-206. Magness, J. 1995. the pottery from area v/4 at Caesarea. Annual of the American Schools of Oriental Research 52:133-145. Magness, J. 1996. Blessings from Jerusalem: evidence for early Christian pilgrimage. Eretz-Israel 25:37*-45*. Magness, J. 1999. Redating the forts of >ein Boqeq, Upper Zohar and other sites in Se Judaea, and the implications for the nature of the limes Palaestinae. in: Humphrey, J.H., ed. The Roman and Byzantine Near East: Some Recent Archaeological Research, Vol. 2. (JRa Supplementary Series 31). Portsmouth. pp. 189-206. Magness, J. 2003a. late Roman and Byzantine pottery. in: Geva, H., ed. Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 1969-1982. Vol. 2: The Finds from Areas A, W and X-2, Final Report. Jerusalem. pp. 423-432. Magness, J. 2003b. The Archaeology of the Early Islamic Settlement in Palestine. Winona lake. Magness, J. 2005. the Roman legionary pottery. in: arubas, B. and Goldfus, H., eds. Excavations on the Site of the Jerusalem International Convention Centre (Binyanei Ha’uma): A Settlement of the late First to Second Temple Period, the Tenth legion’s Kilnworks, and a Byzantine Monastic Complex. The Pottery and Small Finds. (Journal of Roman archaeology Supplementary Series 60) Portsmouth. pp. 68-191. Majcherek, G. 1995. Gazan amphorae: typology reconsidered. in: Meyza, H. and Mlynarczyk, J., eds. Hellenistic and Roman Pottery in the Eastern Mediterranean: Advances in Scientiic Studies. Acts of the Second Workshop at Nieborow. Warsaw. pp. 163-178. 243 Majcherek, G. 2004. alexandria’s long-distance trade in late antiquity – the amphora evidence. in: eiring, J. and lund, J., eds. transport Amphorae and Trade in the Eastern Mediterranean. (acts of the international Colloquium at the Danish institute at athens, September 26-29, 2002) athens. pp. 229-237. Mango, C. 1985. Byzantine Architecture. Milano. Ma>oz, Z.U. 1993. Deir Qeruú. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 1. Jerusalem. pp. 348-349. Marco, S. and agnon, a. 2005. High-resolution stratigraphy reveals repeated earthquake faulting in the Masada Fault Zone, Dead Sea transform. Tectonophysics 408:101-112. Marco, S., Hartal, M., Hazan, N., lev, l. and Stein, M. 2003. archaeology, history, and geology of the a.D. 749 earthquake, Dead Sea transform. Geology 31:665-668. Marco, S., Rockwell, t.K., Heimann, a., Frieslander, U. and agnon, a. 2005. late Holocene slip of the Dead Sea transform revealed in 3D palaeoseismic trenches on the Jordan Gorge Segment. Earth and Planetary Science letters 234:189-205. Margalit, S. 1987. the North Church of Shivta: the Discovery of the First Church. Palestine Exploration Quarterly 119:106-121. Marti, K. 1880. Die alten lauren und Klöster in der Wüste Juda. Zeitschrift des deutschen Palästina-Vereins 3:1-43. Matthews, e., Neidinger, W. and ayalon, e. 1990. Preliminary Report on the Excavations at Zur Natan, 1989 and 1990 Seasons. Houston, tX. pp. 4-28. Mayerson, P. 1992. the Gaza ‘wine’ jar (Gazition) and the ‘lost’ ashkelon jar (ashkalônion). Israel Exploration Journal 42:76-80. Mazar, e. 2003a. areas iii and Xii in the Byzantine period: architecture and stratigraphy. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 203-245. Mazar, e. 2003b. areas Xv and Xvii in the Byzantine period: architecture and stratigraphy. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 3-85. Mazar, e. 2003c. architecture and stratigraphy of the “House of the Menorot”. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 163-186. Mazar, e. 2007. a Byzantine building in area Xvi. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. III: The Byzantine Period. (Qedem 46). Jerusalem. pp. 3-22. Mazar, e. and Gordon, B. 2007. the pottery from the Peristyle and Southern Houses. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. III: The Byzantine Period. (Qedem 46) Jerusalem. pp. 149-176. Mazar, e. and Peleg, O. 2003. the pottery assemblage in the large Byzantine structure in area Xv. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 86-103. McCormick, M. 2003. Rats, communications and plague: toward an ecological history. Journal of Interdisciplinary History 24:1-25. McNicoll, a.W., Smith, R.H. and Hennessy, J.B. 1982. Pella in Jordan 1: First Interim Report of the Joint University of Sydney and the College of Wooster Excavations at Pella, 1979-1981. Canberra. Meimaris, Y.e. 1986. Sacred Names, Saints, Martyrs and Church Oficials in the Greek Inscriptions and Papyri Pertaining to the Christian Church of Palestine. athens. Meyer, C. 1987. Glass from the North theatre Byzantine Church, and Soundings at Jerash, Jordan, 1982-1983. Bulletin of the American Schools of Oriental Research Supplement 25:175-222. 244 referenCeS Meyers, e.M., Strange, J.F. and Meyeres, C.l. 1981. Excavations at Ancient Meiron, Upper Galilee, Israel 1971-72, 1974-75, 1977. Cambridge. Mielsch, H. 1985. Buntmarmore aus Rom im Antikenmuseum Berlin. Berlin. Miles, G.C. 1958. the early islamic coinage of egypt. in: ingholt, H., ed. Centennial Publication of the American Numismatic Society. New York. pp. 471-502. Miller, N.F. and Gleason, K.l. 1994. Fertilizer in the identification and analysis of cultivated soil. in: Miller, N.F. and Gleason, K.l., eds. The Archaeology of Garden and Field. Philadelphia. pp. 25-43. Mlynarczyk, J. 2005. the pottery from Khirbet el-Jiljil (Second Season). Bulletin of the Anglo-Israel Archaeological Society 23:139-166. Monelli, a. 2004. Marble fragments and chancel screens. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev: Final Report of the Excavations 1989-1995. (Beer-Sheva 16/ Beer-Sheva archaeological Monographs 1) Beer-Sheva. pp. 88-108. Nahshoni, P. 1999. a Byzantine site in the Migdal neighborhood, ashqelon. >Atiqot. 38:99*-111*. (Hebrew) Negev, a. 1988. The Architecture of Mampsis, Final Report, Vol. 2: The late Roman and Byzantine Periods. (Qedem 27) Jerusalem. Negev, a. 1997. The Architecture of Oboda, Final Report. (Qedem 36) Jerusalem. Neidinger, W., Matthews, e. and ayalon, e. 1994. excavations at Zur Natan: stratigraphy, architectural and historical report. Reports on TFAHR Excavations at: Zur Natan, Israel; Silistra, Bulgaria; and Ulanci, Macedonia. Houston, tX. pp. 5-14. Netzer, e. 2001. Hasmonean and Herodian Palaces at Jericho. Final Report of the 1973-1987 Excavations, Vol. I: Stratigraphy and Architecture. Jerusalem. Nevo, Y.D. 1991. Pagans and Herders. A Re-Examination of the Negev Runoff Cultivation Systems in the Byzantine and Early Arab Periods. Jerusalem. Nikolsky, v. and Figueras, P. 2004. Descriptive pottery catalogue. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev: Final Report of the Excavations 1989-1995. (BeerSheva 16/Beer-Sheva archaeological Monographs 1). Beer-Sheva. pp. 151-209. Nikolsky, v., Figueras, P., auladell, J. and areal Guerra, R. 2004. Metal objects. in: Figueras, P., ed. îorvat Karkur >Illit. A Byzantine Cemetery Church in the Northern Negev: Final Report of the Excavations 1989-1995. (Beer-Sheva 16/Beer-Sheva archaeological Monographs 1). Beer-Sheva. pp. 237-264. Nir, D. 1975. Géomorphologie d’Israël. Paris. O’Hea, M. 2007. Glass in late antiquity in the Near east. in: lavan, l., Zanini, e. and Sarantis, a., eds. late Antique Archaeology, Vol. 4: Technology in Transition A.d. 300-650. leiden and Boston. pp. 233-248. Oked, S.e. 1993. The Pottery of the late Byzantine and Early Arab Periods at Tel Nessana. (M.a. thesis, Barilan University) Ramat-Gan. (Hebrew) Orren, e. 1976. Operation “danni”, July 1948. tel aviv. (Hebrew) Ovadiah, a. 1970. Corpus of the Byzantine Churches in the Holy land. Bonn. Ovadiah, a., Ovadiah, R. and Gudovitch, S. 1976. Une église byzantine a Matta>. Revue Biblique 83:421-431. Patrich, J. 1988a. the glass vessels. in: tsafrir,Y., Patrich, J., Rosenthal-Heginbottom, R., Hershkovitz, i. and Nevo, Y.D. Excavations at Rehovot-in-the-Negev, Vol. I: The Northern Church. (Qedem 25). Jerusalem. pp. 134-141. Patrich, J. 1988b. architectural sculpture and stone objects. in: tsafrir,Y., Patrich, J., Rosenthal-Heginbottom, R., Hershkovitz, i. and Nevo, Y.D. Excavations at Rehovot-in-the-Negev, Vol. I: The Northern Church. (Qedem 25) Jerusalem. pp. 97-133. Patrich, J. 1993. Monasteries. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 3. Jerusalem. pp. 1063-1067. Patrich, J. 2007. Caesarea in transition from the Byzantine to the Muslim regime: the archaeological evidence from the southwestern zone (areas CC, KK, NN), and the literary sources. Cathedra 122:172-143 (Hebrew) 245 Patrich, J. and tsafrir, Y. 1985. Church and agricultural installations from the Byzantine period in îorvat Beth loya. Qadmoniot 71-72:106-112. (Hebrew) Patrich, J. and tsafrir, Y. 1993. Beth loya, îorvat. in: Stern, e., ed. The New Encyclopedia of Archaeological Excavations in the Holy land, Vol. 1. Jerusalem. pp. 210-213. Payne, S. 1973. Kill-off patterns in sheep and goats: Mandibles from asvan Kale. Anatolian Studies 23:281-303. Peacock, D.P.S. and Williams, D.F. 1986. Amphorae and the Roman Economy, an Introductory Guide. london and New York. Pele, O. 2003. Roof tiles of the Byzantine period from area Xv. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 133-134. Peleg, O. 2003. Decorated chancel screen panels and posts from the temple Mount excavations. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 135-152. Peleg, M. and Reich, R. 1992. excavations of a segment of the Byzantine city wall of Caesarea Maritima. >Atiqot 21:137-170. Perath, i. 1984. Stone Building and Building Stone in Israel: A Historical Review. Jerusalem. Perrone, l. 2006. ‘Rejoice Sion, Mother of all Curches’: Christianity in the Holy land during the Byzantine era. in: limor, O. and Stroumsa, G.G., eds. Christians and Christianity in the Holy land: From the Origins to the latin Kingdom. turnhout. pp. 141-173. Piccirillo, M. 1982. la chiesa della vergina a Madaba. liber Annuus 32:373-408. Piccirillo, M. 1995. la antichità Christiane del villaggio di Mekawer. liber Annuus 45:293-318. Piccirillo, M. 2003. la chiesa della tabula ansata a Umm al-Rasas – Kastrom Mafaa. liber Annuus 53:285-324. Popović, S. 1998. The trapeza in cenobitic monasteries: Architectural and spiritual contexts. dumbarton Oaks Papers 52:281-303. Porath, Y. 2002. Hydraulic plaster in aqueducts as a chronological indicator. in: amit, D., Patrich, J. and Hirschfeld, Y., eds. The Aqueducts of Israel. (Journal of Roman archaeology Supplementary Series 46) Portsmouth. pp. 25-36. Poulter, a. 1995. Nicopolis ad Istrum: A Roman, late Roman, and Early Byzantine City. Excavations 1985-1992. london. Prummel, W. and Frisch, H.J. 1986. a guide for the distinction of species, sex and body side in bone of sheep and goat. Journal of Archaeological Science 13:567-577. Qedar, S. 2001. Weights of eretz israel in the Roman-Byzantine period. in: Rimon, O., ed. Measuring and Weights in Ancient Times. (Reuben and edith Hecht Museum, University of Haifa) Haifa. pp. 23*-25*. Raban, a., Kenneth, G.H. and Bukley, J.a. 1993. The Combined Caesarea Expeditions. Field Reports of the 1992 Season. Haifa. Rahmani, l.Y. 1960. the ancient synagogue of Ma>on (Nirim): the small finds and coins. in: avi-Yonah, M., ed. Bulletin III of the louis M. Rabinowitz Fund for the Exploration of Ancient Synagogues. Jerusalem. pp. 14-18. Rahmani, l.Y. 1964. Mirror-plaques from a fifth-century a.D. tomb. Israel Exploration Journal 14:50-60. Rahmani, l.Y. 1991. two Byzantine wine presses in Jerusalem. >Atiqot 20:95-110. Rappaport, R. a. 1984. Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People. New Haven and london. Rappaport, U. 1983. the Great Revolt and its courses. in: Rappaport, U., ed. Judaea and Rome – The Jewish Revolts. Jerusalem. pp. 22-65. (Hebrew) Rapuano, Y. 1999. the Hellenistic through early islamic pottery from Ras abu Ma>aruf (Pisgat Ze<ev east a). >Atiqot 38:171-203. 246 referenCeS Rapuano, Y. and Yas, J. 1996. Khirbet en-Nabi Bulus (a). Excavations and Surveys in Israel 17:88-89. Rautman, M. 2003. A Cypriote Village of late Antiquity. Kalavasos-Kopetra in the Vasilikos Valley. (Journal of Roman archaeology Supplementary Series 52) Portsmouth. Ravikovitch, S. 1992. The Soils of Israel: Formation, Nature and Properties. tel aviv. (Hebrew) Reich, R. 1990. Miqwa’ot (Jewish Ritual Immersion Baths) in Eretz-Israel in the Second Temple and the Mishnah and Talmud Periods. (Ph.D. dissertation, the Hebrew University of Jerusalem) Jerusalem. (Hebrew) Reitz, e.J. and Wing, e.S. 1999. Archaeozoology. Cambridge. Rielly, K. 1993. the eleventh and twelfth seasons of excavations at Pella (tabaqat Fahl), 1989-1990: the animal bones from tell al-Husn (area XXXiv) and the >abbasid complex (area XXiX). Annual of the department of Antiquities of Jordan 37:218-221. Riley, J.a. 1975. the pottery from the first session of excavation in the Caesarea hippodrome. Bulletin of the American Schools of Oriental Research 218:25-54. Riley, J.a. 1981. the pottery from Cisterns 1977.1, 1977.2 and 1977.3. in: Humphrey, J.H., ed. Excavations at Carthage 1977, Conducted by the University of Michigan, Vol. 6. ann arbor. pp. 85-124. Rodziewitz, M. 1976. Alexandrie I: la céramique romaine tardive d’Alexandrie. Warsaw. Rodziewitz, M. 1984. Alexandrie III: les habitationes romaines tardives d’Alexandrie. Warsaw. Roll, i. 1976. the Roman road network in eretz-israel. Qadmoniot 34-35:38-50 (Hebrew) Roll, i. 1995. Roads and transportation in the Holy land in the early Christian and Byzantine times. in: Dassmann, e. and engemann, J., eds. Akten des XII. Internationalen Kongresses für Cristliche Archäologie, Bonn 22.-28. September 1991. Münster. pp. 1165-1170. Roll, i. and Dagan, Y. 1988. Roman roads around Beth Govrin. in: Stern, e. and Urman, D., eds. Man and Environment in the Southern Shefelah: Studies in Regional Geography and History. Giv‘atayim. pp. 175-179. (Hebrew) Rosen-ayalon, M. and eitan, a. 1969. Ramla Excavations. (israel Museum Catalogue 66) Jerusalem. Rosenthal, R. and Sivan, R. 1978. Ancient lamps in the Schloessinger Collection. (Qedem 8) Jerusalem. Rosenthal-Heginbottom, R. 1988. the pottery. in: tsafrir, Y., Patrich, J., Rosenthal-Heginbottom, R., Hershkovitz, i. and Nevo, Y.D. Excavations at Rehovot-in-the-Negev, Vol. I: The Northern Church. (Qedem 25) Jerusalem. pp. 78-96. Rosenthal-Heginbottom, R. 2003. Hellenistic and early Roman Fine Ware and lamps from area a. in: Geva, H., ed. Jewish Quarter Excavations in the Old City of Jerusalem, Conducted by Nahman Avigad, 19691982. Vol. 2: The Finds from Areas A, W and X-2, Final Report. Jerusalem. pp. 192-223. Roux, G. 1973. Tables chrétiennes en marbre découvertes à Salamine. (Salamine de Chypre 4: anthologie Salamienne) Paris. pp. 133-196. Rubin, Z. 1982. the spreading of Christianity in eretz israel – from the times of Julian to Justinian. in: Baras, Z., Safrai, S., tsafrir, Y. and Stern. M., eds. Eretz Israel, from the destruction of the Second Temple to the Muslim Conquest, Vol. 1: Political, Social and Cultural History. Jerusalem. pp. 236-251. (Hebrew) Rubin, Z. 1999. Jerusalem in the Byzantine period – an historical survey. in: tsafrir, Y. and Safrai, S., eds. The History of Jerusalem. The Roman and Byzantine Periods (70-638 CE). Jerusalem. pp. 199-238. (Hebrew) Russell, J. 1982. Byzantine instrumenta domestica from anemurium: the significance of context. in: Hohlferder, R.l., ed. City, Town and Countryside in the Early Byzantine Era. New York. pp. 133-163. Russell, K.W. 1985. the earthquake chronology of Palestine and northwest arabia from the 2nd through the mid-8th century a.D. Bulletin of the American Schools of Oriental Research 260:37-59. Saar, O. 2003. Superstitions in Israel during the Roman and Early-Byzantine Periods. (M.a. thesis, tel aviv University) tel aviv. (Hebrew) Sade, M. 2006. appendix a: archaeozoological remains. in: taxel, i. and Feldstein, a. Khirbet ibreica: a rural settlement in the southeastern Sharon plain. Salvage Excavation Reports 3:62-63. 247 Sade, M. 2007. Faunal remains. in: Gophna, R., taxel, i. and Feldstein, a. Kafr >ana: a rural settlement in the lod valley. Salvage Excavation Reports 4:103-107. Safrai, Z. 1994. The Economy of Roman Palestine. london and New York. Safrai, Z. 1998. The Missing Century. Palestine in the Fifth Century: Growth and decline. leuven. Safrai, Z. and Dar, S. 1997. îorvat Bireh – an estate in the plain of lod. in: Friedman, Y., Safrai, Z. and Schwartz, J., eds. Hikrei Eretz. Studies in the History of the land of Israel, dedicated to Prof. yehuda Feliks. Ramat-Gan. pp. 57-108. (Hebrew) Safrai, Z. and Sion, O. 2007. Nomad settlement in Palestine during the late Byzantine-early Moslem period. in: edwards, R. and McCollough, C.t., eds. The Archaeology of difference. Gender, Ethnicity, Class and the “Other” in Antiquity: Studies in Honor of Eric M. Meyers. (annual of the american Schools of Oriental Research 60/61) Boston. pp. 397-411. Saller, S.J. 1941. The Memorial of Moses on Mount Nebo, Part 1: The Text. Jerusalem. Saller, S.J. 1946. discoveries at St. John’s, >Ein Karim, 1941-1942. Jerusalem. Saller, S.J. 1957. Excavations at Bethany (1949-1953). Jerusalem. Saller, S.J. and testa, e. 1961. The Archaeological Setting of the Shrine of Bethphage. Jerusalem. Sar-avi, D. 1999. ein el-Sachaniah and the monasteries in the wilderness of Ziph. Judea and Samaria Research Studies 8:185-192 (Hebrew) Schick, R. 1995. The Christian Communities of Palestine from Byzantine to Islamic Rule. A Historical and Archaeological Study. Princeton. Schiffer, M.B. 1995. Behavioral Archaeology: First Principles. Salt lake City. Schmid, e. 1972. Atlas of Animal Bones. amsterdam. Schürer, e. 1973. The History of the Jewish People in the Age of Jesus Christ, Vol. 2. vermes, G. and Millar, F., eds. edinburgh. Schwartz, J.J. 1986. Jewish Settlement in Judaea after the Bar-Kokhba War until the Arab Conquest, 135 CE640 CE. Jerusalem. (Hebrew) Segal, a., Mlynarczyk, J., Burdajewicz, M., Schuler, M. and eisenberg, M. 2005. Hippos-Sussita. Sixth Season of Excavations, July 2005. Haifa. Seligman, J. 1995. Naúal îaggit. Excavations and Surveys in Israel 16:61-63. Seligman, J. 1999. agricultural complexes at Ras abu Ma>aruf (Pisgat Ze<ev east a), north of Jerusalem. >Atiqot 38:137-170. Seligman, J. and abu Raya, R. 2002. a Byzantine ‘monastery’ at Khirbet Umm leisun, Jerusalem. >Atiqot 43:127-140. Seligman, J. and May, N. 1993. Upper îorvat >illin. Excavations and Surveys in Israel 12:76-78. Seligman, J., Zias, J. and Stark, H. 1996. late Hellenistic and Byzantine burial caves at Giv>at Sharet, Bet Shemesh. >Atiqot 29:43-62. Shalem, D. 2002. Nevé Ur – an early islamic period village in the Bet She’an valley. >Atiqot 43:149-176. Shallev, R. 1994. The Emmaus Region during the Roman-Byzantine Period. (M.a. thesis, Bar-ilan University) Ramat-Gan. (Hebrew) Shapira, l. and Peleg, O. 2003a. Pottery lamps of the Byzantine period from area Xv. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43) Jerusalem. pp. 104-108. Shapira, l. and Peleg, O. 2003b. Byzantine and early islamic pottery lamps from the “House of the Menorot” in area vi. in: Mazar, e. The Temple Mount Excavations in Jerusalem 1968-1978, directed by Benjamin Mazar. Final Report, Vol. II: The Byzantine and Early Islamic Periods. (Qedem 43). Jerusalem. pp. 187-190. Shenhav, e. 2003. îorvat îanot. a Byzantine tradition of Goliath’s burial place. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 269-272. 248 referenCeS Shukrun, e. and Sauariego, a. 1993. Jerusalem, Pisgat Ze<ev (villa Quarter). Excavations and Surveys in Israel 12:56-58. Shurkin, O. 2004. Burial grounds and an industrial area in Wadi el-îalaf (Near Khirbet Ras abu Ma>aruf) in Pisgat Ze<ev, Jerusalem. >Atiqot 48:27*-58*. (Hebrew) Sidi, N., amit, D. and >ad, U. 2003. two wine presses from Kefar Sirkin and Mazor. >Atiqot 44:253-266. Silberstein, N. 2000. Hellenistic and Roman pottery. in: Hirschfeld, Y. Ramat Hanadiv Excavations, Final Report of the 1984-1988 Seasons. Jerusalem. pp. 420-472. Silver, i. a. 1969. the ageing of domesticated animals. in: Brothwell, D. R. and Higgs, e., eds. Science in Archaeology. london. pp. 283-302. Simpson, St J. 2000. the clay pipes. in: Harper, R. and Pringle, D., eds. Belmont Castle: The Excavation of a Crusader Stronghold in the Kingdom of Jerusalem. (British academy Monographs in archaeology, No. 10) Oxford. pp. 147-171. Sion, O. 1997a. Khirbet abu Suwwana. >Atiqot 32:183-194. (Hebrew) Sion, O. 1997b. Mishor adummim (Khirbet Handoma). >Atiqot 32:149-158. (Hebrew) Sion, O. 1998. excavations at Khirbet el-Khillia (Building ii). Judea and Samaria Research Studies 7:191-205. (Hebrew) Sion, O. 2004. an early islamic period settlement in Ramla. >Atiqot 46:67-92. (Hebrew) Sodini, J.P. and Kolokotsas, K. 1984. Aliki, II: la Basilique double. Paris. Spaer, M. 1988. the pre-islamic glass bracelets of Palestine. Journal of Glass Studies 30:51-61. Spaer, M. 2001. Ancient Glass in the Israel Museum. Beads and Other Small Objects. Jerusalem. Stacey, D. 1988-89. Umayyad and egyption Red-Slip “a” Ware from tiberias. Bulletin of the Anglo-Israel Archaeological Society 8:21-33. Stacey, D. 2004. Excavations at Tiberias, 1973-1974. The Early Islamic Periods. (iaa Reports 21) Jerusalem. Stern, e.J. 1997. Burial caves at Kisra. >Atiqot 33:103-135. (Hebrew) Strus, a. 2003. Khirbet Fattir – Bet Gemal. Two Ancient Jewish and Christian Sites in Israel. Rome. Strus, a. and Gibson, S. 2005. New excavations at Khirbet el-Jiljil (Bet Gemal) near Beth Shemesh. Bulletin of the Anglo-Israel Archaeological Society 23:29-89. Sussman, v. 1976. a burial cave at Kefar >ara. >Atiqot 11:92-101. Syon, D. 1998. a wine press at akhziv. >Atiqot 34:85-99. (Hebrew) Syon, D. 2003. a church from the early islamic period at Khirbet el-Shubeika. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 75-81. Syon, D. 2004. a late Byzantine oil press at Kefar Barukh. >Atiqot 47:155-168. tacher, a., Nagar, Y. and avshalom-Gorni, D. 2002. the burial caves. in: excavations at Khirbet el-Shubeika 1991, 1993. in: Gal, Z., ed. Eretz Zafon: Studies in Galilean Archaeology. Jerusalem. pp. 263-288. (Hebrew) taha, H. 2002. the sanctuary of Sheikh el-Qatrawani. liber Annuus 52:441-456. taha, H. 2003. a Byzantine tomb at atara. in: Bottini, G.C., Di Segni, l. and Chrupcala, D., eds. One land, Many Cultures. Archaeological Studies in Honor of Stanislao loffreda OFM. Jerusalem. pp. 87-110. tal, N. 1997. Rolling stones – evidence of insecurity during the Byzantine period. Judea and Samaria Research Studies 6:271-290. (Hebrew) tal, O. and taxel, i. 2008. the late Umayyad, >abbasid and Fatimid periods. in: tal, O. and taxel, i. Ramla (South): an early islamic industrial Site and Remains of Previous Periods. Salvage Excavation Reports 5:81-213. talbot, a.M. 2002. Byzantine monastic horticulture: the textual evidence. in: littlewood, a., Maguire, H. and Wolschke-Bulman, J., eds. Byzantine Garden Culture. Washington, D.C. pp. 37-67. talgam, R. 2002. technical aspects of the mosaic craft. Michmanim 16:7-13. (Hebrew) 249 taxel, i. 2005. The Transition between the Byzantine and the Early Islamic Periods (the 7th Century CE) as seen through Rural Settlement – îorvat Zikhrin as a Case Study. (M.a. thesis, tel aviv University) tel aviv. (Hebrew) taxel, i. 2006. Hurvat es-Suyyagh – a Byzantine monastery in the northeastern Judaean Shephelah. Judea and Samaria Research Studies 15:169-183. (Hebrew) taxel, i. 2007a. Glass objects. in: Gophna, R., taxel, i. and Feldstein, a. Kafr >ana: a rural settlement in the lod valley. Salvage Excavation Reports 4:66-76. taxel, i. 2007b. Stone, bone, shell and metal objects. in: Gophna, R., taxel, i. and Feldstein, a. Kafr >ana: a rural settlement in the lod valley. Salvage Excavation Reports 4:88-98. taxel, i. 2008. Rural monasticism at the foothills of southern Samaria and Judaea in the Byzantine period: asceticism, agriculture and pilgrimage. Bulletin of the Anglo-Israel Archaeological Society 26:57-73. taxel, i. Forthcoming. From prosperity to survival: Rural monasteries in Palestine in the transition from Byzantine to Muslim rule (7th century Ce). in: Challenging Frontiers: Mobility, Transition, and Changes. (Proceedings of the third G.a.O. international Conference, University of Oxford, 4th-5th april 2008) Oxford. tepper, Y. 2003. Survey of the legio Area near Megiddo – Historical-Geographical Research. (M.a. thesis, tel aviv University) tel aviv. (Hebrew) testini, P. 1962. the ‘Kathisma’ church and monastery. in: aharoni, Y. Excavations at Ramat Raúel, Seasons 1959 and 1960. Rome. pp. 73-91. testini, P. 1964. the ‘Kathisma’ church and monastery. in: aharoni, Y. Excavations at Ramat Raḥel, Seasons 1961 and 1962. Rome. pp. 101-106. thomsen, P. 1917. Die römischen Meilsteine der Provinzen Syria, arabia und Palaestina. Zeitschrift des deutschen Palästina-Vereins 40:1-141. tiesenhausen, W.G. 1873. Monnaies des Khalifes Orientaux. St. Peterseburg. tomber, R.S. 1988. Pottery from the 1982-83 excavations. in: Humphrey, J.H., ed. The Circus and Byzantine Cemetery at Carthage. ann arbor. pp. 437-528. tomber, R.S. 1999. Pottery from the sediments of the inner harbour (area i14). in: Holum, K.G., Raban, a. and Patrich, J., eds. Caesarea Papers, Vol. 2: Herod’s Temple, the Provincial Governor’s Praetorium and Granaries, the later Harbor, a Gold Coin Hoard and Other Studies. (JRa Supplementary Series 35) Portsmouth. pp. 295-322. tsafrir, Y. 1984. Eretz Israel from the destruction of the Second Temple to the Muslim Conquest, Vol 2: Archaeology and Art. Jerusalem. (Hebrew) tsafrir, Y. 1986. the maps used by theodosius: On the pilgrim maps of the Holy land and Jerusalem in the Sixth Century C.e. dumbarton Oaks Papers 40:129-145. tsafrir, Y. 1996. Some notes on the settlements and demography of Palestine in the Byzantine Period: the archaeological evidence. in: Seger, J.D., ed. Retrieving the Past: Essays on Archaeological Research and Methodology in Honor of Gus W. Van Beek. Winona lake. pp. 269-183. tsafrir, Y. and Foerster, G. 1997. Urbanism at Scythopolis-Bet Shean in the fourth to seventh centuries. dumbarton Oaks Papers 51:85-146. tsafrir, Y. and Hirschfeld, Y. 1979. the church and mosaics at îorvat Berachot, israel. dumbarton Oaks Papers 33:293-326. tsori, N. 1962. archaeological survey in the Beth Shean valley. in: The Beth Shean Valley: The 17th Archaeological Convention. Jerusalem. pp. 135-198. (Hebrew) tsuk, Z. 1994. the water supply of Hurbat Zikrin. Israel – People and land 7-8:133-148. (Hebrew) tsuf, O. 2003. Red Slip Bowls in the late Roman and Byzantine Period – Social, Economical and Technological Aspects. (Ph.D. dissertation, University of Haifa). Haifa. (Hebrew) 250 referenCeS tubb, J.N. 1986. the pottery from a Byzantine well near tell Fara. Palestine Exploration Quarterly 118:51-65. tushingham, a.D. 1972. The Excavations at Dibon (Dhībân) in Moab, the Third Campaign, 1952-53. (annual of the american Schools of Oriental Research 40) Cambridge. tushingham, a.D. 1985. Excavations in Jerusalem 1961-1967, Vol. I. toronto. tzaferis, v. 1975. the archaeological excavation at Shepherds’ Field. liber Annuus 55:5-52. tzaferis, v. 1983. The excavations of Kursi – Gergesa. (>atiqot 16). Jerusalem. tzaferis, v. 1997. a Greek inscription from Khirbet abu Rish. >Atiqot 32:147-148 (Hebrew) Ustinova, Y. and Nahshoni, P. 1994. Salvage excavations in Ramot Nof, Be’er Sheva. >Atiqot 25:157-177. vailhé, P.S. 1899. Répertoire alphabétique des monastères de Palestine. Revue de l’Orient Chrétien 4:512-542. vikan, G. 1982. Byzantine Pilgrimage Art. Washington, D.C. vincent, l.H. and abel, F.M. 1926. Jérusalem. Recherches de topographie, d’archéologie et d’histoire, Tome 2: Jérusalem nouvelle. Paris. vincent, l.H. and abel, F.M. 1932. Emmaüs – sa basilique et son histoire. Paris. von den Dreisch, a. and Boessneck, J. 1995. Final report of the zooarchaeological investigation of animal bone finds from tell Hesban, Jordan. in: labianca, Ø.S. and von den Dreisch, a., eds. Hesban, Vol. 13: Faunal Remains: Taphonomical and Zooarchaeological Studies of the Animal Remains from Tell Hesban and Vicinity. Berrien Springs, Mich. pp. 65-108. vööbus, a. 1960. History of Ascetism in the Syrian Orient. A Contribution to the History of Culture in the Near East, Vol. 2: Early Monasticism in Mesopotamia and Syria. louvain. Waisel, Y. 1984. the vegetation of the Judaean Shephelah. in: Waisel, Y., ed. The Plants and Animals of the land of Israel, an Illustrated Encyclopedia, Vol. 8: Vegetation of Israel. tel aviv. pp. 187-193. (Hebrew) Waldbaum, J.C. 1983. Metalwork from Sardis: The Finds through 1974. Harvard. Waliszewski, t. 1994. la mosaïque de Deir el->asfur retrouvée: le motif des “rinceaux habitas” en Judée et dans la Shéphéla. Revue Biblique 101: 562-579. Waliszewski, T. and Ortali-Tarazi, R. 2002. Village romain et byzantin à Chhîm-Marjiyat. Rapport préliminaire (1996-2002). Bulletin d’Archéologie et d’Architecture libanaises 6:5-105. Walker, J. 1956. A Catalogue of the Arab-Byzantine and Post-Reform Umaiyad Coins. london. Walmsley, a. 1992. the social and economic regime at Fihl (Pella) between the 7th and 9th centuries. in: Canivet, P. and Rey-Coquais, J.-P., eds. la Syrie de Byzance à l’islam VIIe-VIIIe siècles: Actes du colloque international, lyon-Maison de l’Orient Méditerranéen, Paris – Institute du Monde Arabe, 11-15 septembre 1990. Damas. pp. 249-261. Walmsley, a. 1995. tradition, innovation, and imitation in the material culture of islamic Jordan: the first four centuries. Studies in the History and Archaeology of Jordan 5:657-668. Walmsley, a. 2000. Production, exchange and regional trade in the islamic east Mediterranean: Old structures, new systems? in: Hensen, i.K. and Wickham, C., eds. The long Eighth Century. leiden. pp. 265-343. Walmsley, a. 2007. Early Islamic Syria: An Archaeological Assessment. london. Wapnish, P. and Hesse, B. 2000. Mammal remains from the early Bronze sacred compound. in: Finkelstein, i., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 1992-1996 Seasons. (Monograph Series of the institute of archaeology of tel aviv University No. 18). tel aviv. pp. 429-462. Warnock, P. 2007. Identiication of Ancient Olive Oil Processing Methods Based on Olive Remains. (British archaeological Reports international Series 1635). Oxford. Watson, P.M. 1992. Change in foreign and regional economic links with Pella in the seventh century a.D.: the ceramic evidence. in: Canivet, P. and Rey-Coquais, J.-P., eds. la Syrie de Byzance à l’Islam VIIe-VIIIe siècles: Actes du colloque international, lyon-Maison de l’Orient Méditerranéen, Paris – Institute du Monde Arabe, 11-15 septembre 1990. Damas. pp. 233-248. 251 Watson, P.M. 1995. Ceramic evidence for egyptian links with northern Jordan in the 6th-8th centuries aD. in: Bourke, S. and Descoeudres, J-P., eds. Trade, Contact, and the Movements of Peoples in the Eastern Mediterranean. Studies in Honor of J. Basil Hennessy. Sydney. pp. 303-324. Watson, P.M. 2004. Cultural identity and wine production in northern Jordan. Studies in the History and Archaeology of Jordan 8:485-502. Weiss, D. 1994. Naúal Yo>el. Excavations and Surveys in Israel 14:145. Weiss, D. 1995. Khirbet >ein Shams. Excavations and Surveys in Israel 16:112. Weiss, D., Zissu, B. and Solimany, G. 2004. Archaeological Survey of Israel: Map of Nes Harim (104). Jerusalem. Weksler-Bdolah, S. 1996. îorbat >illin. Excavations and Surveys in Israel 18:95-96. Weksler-Bdolah, S. 2006-2007. the fortifications of Jerusalem in the Byzantine period. Aram 18-19:85-112. Whitcomb, D. 1989. evidence of the Umayyad period from the aqaba excavations. in: Bakhit, M.a. and Schick, R., eds. The Fourth International Conference on the History of Bilād al-Shām during the Umayyad Period. Proceedings of the 3rd Symposium. amman. pp. 164-184. Whitcomb, D. 2009. From pastoral peasantry to tribal urbanites: arab tribes and the foundation of the islamic state in Syria. in Szuchman, J. ed. Nomads, Tribes, and the State in the Ancient Near East: Cross-disciplinary Perspectives. Chicago. pp. 241-259. White, K.D. 1967. Agricultural Implements of the Roman World. Cambridge. White, K.D. 1975. Farm Equipment of the Roman World. Cambridge. Wightman, G.J. 1989. The damascus Gate, Jerusalem. Excavations by C.M. Bennett and J.B. Hennessy at the damascus Gate, Jerusalem, 1964-66. (British archaeological Reports international Series 519) Oxford. Wilson, J. and Sa>d, M. 1984. the domestic material culture of Busra from the Nabataean to the Umayyad periods. Berytus 32:35-147. Winfield, D. and Wainwright, J. 1962. Some Byzantine churches from the Pontus. Anatolian Studies 12:131-161. Womer Katzev, S. 1982. Miscellaneous finds. in: Bass, G.F. and van Doorninck, F.H., eds. yassi Ada: A Seventh Century Byzantine Shipwreck. College Station, tX. pp. 266-281. Wroth, W. 1908. Imperial Byzantine Coins in the British Museum, Vol. 1. london. Yeivin, Z. and Finkielsztejn, G. 1999. îorbat Castra – 1993-1997. îadashot Arkheologiyot – Excavations and Surveys in Israel 109:23*-27*. Yekutieli, Y., Ben-Yishai, Y., talis, S. and Harpak, t. 2001. The Salvage Excavations at Site 77 in the TransIsrael Highway. (Unpublished internal Report, institute of archaeology, Ben-Gurion University). BeerSheva. (Hebrew) Yitach, M. 2001. Khirbet Deir >arab. îadashot Arkheologiyot – Excavations and Surveys in Israel 113:64*-65*. Yusuf, M.D. 1985. Economic Survey of Syria during the Tenth and Eleventh Centuries. Berlin. Zeisel, W.N. 1975. An Economic Survey of the Early Byzantine Church. ann arbor. Zemer, a. 1999. Castra at the Foot of Mount Carmel – the City and its Secrets. Haifa. Zinger, a. 1985. The Olive Growth. Jerusalem. (Hebrew) Zissu, B. 1999. îorbat îushsham. îadashot Arkheologiyot – Excavations and Surveys in Israel 109:81*. Zissu, B. 2000. Matta> (West). îadashot Arkheologiyot – Excavations and Surveys in Israel 112:90*-91*. Zissu, B. 2001. Rural Settlement in the Judaean Hills and Foothills from the late Second Temple Period to the Bar Kokhba Revolt. (Ph.D. dissertation, the Hebrew University of Jerusalem) Jerusalem (Hebrew) 252 lISt of loCI liSt OF lOCi1 locus 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 Square C-D/9 10 C8 C7 C9 C9 C8 C8 C8 C9 C9 C9 B/9-10 C8 C8 C8 C9 C/8-9 C8 C/8-9 C9 C-D/8 B9 C8 C-D/8 C8 C8 C8 C8 C9 C8 C8 C9 C9 C10 C10 B-C/8 C10 C10 B8 C8 C10 C10 C10 C10 B-C/10 B-C/7 B8 C10 C7 C-e/10 C8 Opening height Closing height description/Stratigraphic relation 259.20 257.22 topsoil 258.75 258.27 258.50 258.50 257.97 257.91 258.35 258.20 258. 02 259.10 258.16 258.17 258. 09 258.23 257.80 257.70 258. 02 258.56 257.96 258.67 257.47 257.88 258.52 257.29 258.01 257.55 258.22 257.81 258.46 258.22 257.82 258.82 257.16 257.33 257.72 256.65 256.65 257.43 257.43 256.56 256.56 256.65 256.65 256.20 257.35 257.35 256.20 257.43 256.20 257.22 258.27 257.70 258.20 258.20 257.87 257.54 258.22 258.20 257.62 257.80 257.48 257.95 257.58 257.38 257.25 257.29 257.53 257.48 257.45 258.22 256.91 257.55 258.01 256.91 257.80 257.19 257.57 257.22 258.46 257.57 257.19 257.99 256.20 256.65 257.43 256.56 256.56 257.30 257.23 256.05 256.02 255.56 255.61 255.74 257.23 257.26 255.58 257.23 254.96 256.90 topsoil Fill in Courtyard 28; Phase ii/iii Collapse inside tower 22; phase III Collapse inside tower 22 Fill Fill Collapse inside Storeroom 29; phase II Cleaning of W1 and W8 Collapse topsoil topsoil Collapse Collapse Cleaning between W4 and W7 Fill Fill south of tower 22; phase III Floor of Storeroom 34 topsoil topsoil Collapse inside Storeroom 29; phase II topsoil Collapse between W5 and W36; phase II/III topsoil Bedrock south of tower 22; phase III Sterile bedrock south of Storeroom 34 Bedrock below collapse l122 Collapse inside Storeroom 29; phase II Bedrock inside southwestern room of tower 22 Cleaning of W8 Collapse inside Storeroom 29; Phase II Bedrock north of W9 inside tower 22 Collapse inside tower 22; Phase III Collapse inside basement of tower 22; Phase II Fill between W13 and W14; Phase I/II topsoil Fill between W13 and W14; Phase I/II Fill between W14 and W20; Phase I/II Bedrock west of W12 Bedrock east of W12 Collapse between W13 and W14; Phase I/II Fill between W14 and W20; Phase I/II Fill between W13 and W14; Phase I/II Fill between W14 and W20; Phase I/II Fill over the destroyed part of W11 in tower 22 topsoil Bedrock west of W12 Northern part of basement in tower 22; Phase II Bedrock east of W12 Fill north of W16 Collapse east of W12 1. Not including loci excavated by the iaa. these, when mentioned in the text, are marked with an asterisk [*]; see also Kogan-Zehavi 2008). 253 locus Square Opening height Closing height 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179a 179b 179C 180 181 182 183 184 185 186 C10 C-e/10 D10 B10 B10 D10 e9 D10 e9 e9 e10 e9 e9 e9 e8 D10 e8 D10 e8 e9 e8 e8 e8 D10 e9 e10 e10 e10 e10 e10 e8 C8 d8 C8 d/9-10 e9 e9 255.58 256.20 257.20 256.83 256.12 257.20 257.20 257.63 256.99 256.99 256.62 256.81 256.62 257.03 257.27 256.67 257.61 256.92 257.61 256.83 256.55 256.55 256.55 255.40 256.59 256.44 255.20 255.82 253.30 253.10 256.04 258.27 258.10 257.98 256.62 257.54 256.78 255.50 254.96 255.33 256.12 255.03 256.18 256.99 256.92 256.34 256.62 255.93 256.12 256.20 256.73 256.69 256.36 257.50 255.40 257.50 256.55 255.82 254.91 255.18 252.46 255.79 255.34 253.50 253.30 251.90 251.90 255.37 257.98 257.50 257.73 256.30 256.89 256.54 187 e9 256.78 256.45 188 189 190 191 193 194 195 196 197 200 201 202 203 204 205 206 207 e9 e8 e/7-8 e7 e8 e9 e8 C9 C8 d9 d9 d8 d7 d8 d9 d7 d8 256.85 257.45 256.90 256.79 257.45 256.81 256.59 256.06 257.53 259.40 258.17 259.07 258.27 258.80 257.92 258.05 258.41 256.54 257.01 256.70 256.37 255.87 255.60 255.45 255.52 257.45 258.48 257.92 258.80 258.03 258.28 257.85 257.89 258.30 254 description/Stratigraphic relation Cleaning of W11 and W16 Fill north of W16 Fill inside Room 23 topsoil Fill northwest of W11 Fill inside Room 23; Phase II/III Fill south of floor l184 in Oil press 26; Phase II/III topsoil Fill inside plastered installation l503; Phase I Fill inside plastered installation l503; Phase I topsoil Fill inside upper collecting vat l194 in Oil press 26; Phase III Fill inside plastered installation l503; Phase I Base of crushing basin in Oil press; Phase II topsoil Collapse inside Room 23; Phase II Collapse west of W22 in Courtyard 8 Collapse above Cistern 24 Collapse west of W22 in Courtyard 8 Fill above upper collecting vat in Oil press 26 Collapse above upper collecting vat in Oil press 26; Phase III Collapse at the south of Oil press 26; Phase III Fill south of lower collecting vat l195 in Oil press 26; Phase III Collapse inside Cistern 24; Phase III Fill inside lower collecting vat l195 in Oil press 26; Phase III Fill east of W33; Phase II/III Collapse north of Cistern 25 Collapse inside Cistern 25; Phase III Fill inside Cistern 25; Phase III Upper plaster layer of Cistern 25; Phase III Fill below crushed limestone floor l140*; Phase II Fieldstones paving between Storerooms 29 and 34; Phase II Fill below floor l289 in Courtyard 8; Phase II Fill below fieldstones paving l181; Phase II ashlars floor in Oil press 26; Phase II Bedrock east of W22 in Courtyard 8 Fill above staircase of Oil press 26; Phase III Fill below sloping stone pavement l263 around upper collecting vat in Oil press 26; Phase II/III Dismantling of W24 Floor in Room 11; Phase III Fill below floor of built chamber l131* in Room 12; Phase III Dismantling of floor l298 in Room 14; Phase II Dismantling of W30 Upper collecting vat in Oil press 26; Phase II lower collecting vat in Oil press 26; Phase II Southern part of basement in tower 22; Phase II Stone pavement west of Storeroom 29; Phase II topsoil topsoil topsoil topsoil Fill Fill in demolished part of Courtyard 8 Collapse east of W100 topsoil lISt of loCI locus Square Opening height Closing height 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 d7 d9 d9 d8 d9 d8 d9 d8 d8 d/8-9 d8 d7 d7 d8 d9 d8 d8 e8 d-e/6-7 d9 d7 d9 e8 e8 d-e/7 d9 d9 e6 d6 d6 d8 e8 d6 e7 e7 e7 e7 e7 e7 e7 e7 e7 e7 f/5-8 e7 e6 d/6-7 e7 e9 e7 257.90 258.40 258.33 258.30 258.30 258.28 257.85 258.24 258.24 258.48 258.14 257.65 257.62 258.21 257.91 257.61 258.12 258.16 258.09 257.92 257.97 257.79 257.61 257.30 258.80 257.62 257.73 256.75 256.82 257.57 258.70 256.92 257.13 257.81 257.25 256.69 256.69 256.97 256.81 256.84 256.72 256.92 257.75 256.02 256.12 256.43 257.80 256.38 257.12 256.90 257.44 258.33 258.01 257.90 257.91 257.91 257.52 258.12 258.08 258.12 257.80 257.49 257.38 258.08 257.79 257.42 257.86 257.75 256.95 257.73 257.78 257.42 257.29 256.92 258.53 257.47 257.48 256.52 256.15 257.42 258.18 256.74 256.01 257.23 256.92 255.73 255.55 255.69 256.00 256.38 256.20 256.77 257.04 255.21 255.21 255.97 257.49 255.50 256.01 256.79 258 e7 256.52 255.97 259 260 261 262 e6 d6 d/6/-7 C6 256.46 256.80 257.83 257.78 256.19 256.52 257.49 257.31 description/Stratigraphic relation Collapse inside Room 19; Phase III Fill above floor l289 in Courtyard 8; Phase III Fill above floor l289 in Courtyard 8; Phase III Fill above floor l289 in Courtyard 10 Fill in demolished part of Courtyard 8 Fill above floor l289 in Courtyard 8 Fill in demolished part of Courtyard 8 Fill above floor l289 in Courtyard 8 Fill above floor l289 in Courtyard 8 topsoil Fill in demolished part of Courtyard 8 Fill inside Room 19; Phase III Fill inside Room 19; Phase III Fill above floor l289 in Courtyard 8 Fill in demolished part of Courtyard 8 Fill above floor l289 in Courtyard 8 Fill in demolished part of Courtyard 8 topsoil topsoil Fill in demolished part of Courtyard 8 Fill above bedrock in Courtyard 28 Fill in demolished part of Courtyard 8 topsoil Fill inside Room 12; Phase II/III topsoil Fill below floor l289 in Courtyard 8; Phase II Fill in demolished part of Courtyard 8 Fill Fill inside Room 5 Fill inside entrance Corridor 7 topsoil Fill inside Room 12; Phase II/III Fill inside entrance Corridor 7; Phase II/III Fill inside Room 21; Phase II/III Floor in western part of Room 20; Phase II/III Fill in eastern part of Room 20; Phase II/III Fill in southern part of Room 14; Phase II/III Fill in northern part of Room 14; Phase II/III Built chamber in Room 13; Phase III Bedrock in northeastern part of Room 20 Fill inside Room 13; Phase III Fill Fill above floor l189 in Room 11; Phase II/III Fill east of W33; Phase II/III Fill inside Room 13; Phase III topsoil topsoil Fill inside hewn pit l292 in Room 20; Phase III topsoil Fill above floor l298 in Room 14; Phase II/III Fill in foundation trenches of W110 and W124 in Room 20; Phase II Floor of Room 6; Phase II Floor of Room 5; Phase II topsoil topsoil 255 locus Square Opening height Closing height 263 e9 256.90 256.81 264 265 266 267 268 269 270 e6 e6 d6 d9 d7 d9 d6 257.05 256.55 257.07 257.42 257.51 257.87 256.86 256.96 256.41 256.86 256.97 257.50 257.63 256.68 271 e6 256.38 255.37 272 273 274 275 276 277 278 279 280 281 282 283 284 285 e5 e6 e6 d6 d6 e6 C6 e5 C6 d6 C6 C/5-6 e5 e6 255.43 256.12 256.17 256.68 256.82 256.37 257.31 256.47 257.78 257.51 257.54 257.50 256.47 256.17 255.22 255.95 254.27 256.12 256.07 255.95 256.33 256.16 257.54 256.90 256.48 257.37 256.11 254.27 286 e5 255.76 255.37 287 289 290 291 292 293 294 295 296 298 300 301 302 303 304 305 306 307 308 309 310 311 312 313 315 316 317 318 319 d8 d8 d7 d8 e7 d8 d8 d8 d8 e7 C5 C5 C5 C5 C4 C5 C5 C5 C4 C4 C5 d4 C5 C5 d5 C4 C5 C4 d5 258.10 258.18 257.52 258.35 256.38 258.08 258.05 258.35 257.87 256.79 257.29 257.30 257.23 257.36 257.43 257.30 257.15 257.76 257.00 257.36 256.93 257.25 257.13 256.97 257.24 256.96 257.10 256.98 257.04 257.87 257.77 257.50 258.15 255.50 257.55 257.62 258.15 257.63 256.65 257.27 257.27 257.06 257.32 256.77 257.15 257.00 256.48 256.97 256.77 256.21 256.76 256.60 256.65 256.98 256.71 256.40 256.55 256.51 256 description/Stratigraphic relation Sloping stone pavement around upper collecting vat in Oil press 26; Phase II/III Cleaning of W107 and W117 Fill inside Room 4; Phase III Fill inside Room 4; Phase III Fill in demolished part of Courtyard 8 Floor of Room 19 (northern section); Phase II Drainage channel in Courtyard 8; Phase II Fill inside Room 4; Phase III Fill above stone-paved platform between W131 and W147 (room 16); Phase III topsoil Round stone-built installation in Room 15; Phase II/III Fill inside lime kiln l285 in Room 4; Phase III Fill inside Room 4; Phase III Fill inside Room 4; Phase III Fill inside Room 4; Phase III Fill inside Hall 31; Phase II Fill above floor l400; Phase II/III topsoil Fill inside entrance Corridor 7; Phase II/III Fill inside Hall 31; Phase II topsoil topsoil lime kiln in Room 4; Phase III Fill above stone-paved platform between W131 and W147 (Room 16); Phase III Fill below floor l289 in Courtyard 8 Floor of Courtyard 8-10; Phase II Floor of Room 19 (southern section); Phase II late floor in Courtyard 8; Phase III Hewn pit in Room 20; Phase III Fill below floor l289 in Courtyard 8; Phase II Fill below floor l289 in Courtyard 8; Phase II Dismantling of floor l291 in Courtyard 8; Phase II Fill inside drainage channel l269 in Courtyard 8; Phase II Floor of Room 14; Phase II topsoil Fill above northeastern corner of church and Room 33 Fill inside Hall 31 Fill inside church; Phase II/III Fill inside church; Phase II/III topsoil/fill inside church Fill inside church; Phase II/III Collapse inside Hall 31; Phase III Fill inside church; Phase II/III Fill inside church; Phase II/III Round stone-built base in Hall 31; Phase II topsoil Fill inside Hall 31; Phase III Fill inside church; Phase II/III topsoil Fill inside church; Phase II/III Collapse inside Hall 31; Phase III Fill inside church; Phase II/III topsoil lISt of loCI locus Square Opening height Closing height 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 340 341 342 343 345 346 347 348 349 350 351 352 353 C4 C5 d5 C5 C4 d5 d5 d4 C4 d5 d5 C5 C5 C-d/4 d5 d4 d5 d5 b5 d4 d4 d5 d5 d5 d5 d-e/5 d-e/4 d4 d4 C-d/4 b5 d5 256.77 256.48 257.08 257.24 256.55 256.99 256.90 257.19 256.53 256.77 256.44 256.60 256.56 256.51 256.47 257.19 256.99 256.50 257.47 256.56 256.56 256.21 256.00 256.27 256.24 256.35 256.28 256.33 256.70 256.33 257.53 256.27 256.53 256.32 256.29 256.54 255.82 256.44 256.56 256.41 256.50 256.15 255.65 256.47 256.43 255.71 256.21 256.56 256.27 256.00 256.88 256.33 255.79 256.20 255.60 256.24 255.80 256.00 255.91 254.88 256.45 256.33 257.28 255.80 354 d5 256.12 255.77 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 e5 d5 b3 d5 d4 d5 d5 d4 d4 b4 b5 b5 e5 d5 b4 d5 e5 e4 e4 e5 C-d/4 b5 256.93 256.12 256.92 256.15 255.31 257.03 256.29 256.18 256.45 257.15 257.28 257.26 256.06 256.16 257.06 256.15 256.47 255.75 255.75 256.14 256.24 257.98 256.00 255.00 255.82 255.37 255.29 256.24 255.91 255.50 256.20 256.68 256.85 256.86 256.03 255.69 256.39 255.40 256.01 255.10 255.41 255.84 255.60 257.10 description/Stratigraphic relation Fill inside church; Phase II/III Fill inside Hall 31; Phase II/III Fill inside Hall 31; Phase III topsoil trench inside western end of church’s southern isle; Phase II Fill inside entrance Corridor 7; Phase II/III Fill above the western fringes of Room 33 Fill inside church; Phase II/III Fill inside church; Phase II/III Fill above the western fringes of Room 33 Fill inside entrance Corridor 7; Phase II/III Fill inside Hall 31; Phase II/III Bedrock in northwestern corner of church Fill inside church; Phase II/III topsoil topsoil topsoil topsoil topsoil Fill inside church; Phase II/III Fill south of main gate; Phase II/III Fill inside entrance Corridor 7; Phase II/III Fill inside Room 3; Phase III Fill inside Room 1; Phase III Fill inside Room 1; Phase III topsoil/fill above Courtyard 2; Phase III topsoil/fill above Room 3; Phase III trench inside church’s apse; Phase II topsoil/fill Bedrock inside church Fill inside church; Phase II/III Cleaning of W217 in Room 1 Fill above dismantled part of pebble pavement in Room 3; Phase II/III topsoil/fill above Courtyard 2 Section through pebble pavement l385 in Room 33; Phase II topsoil/fill Fill south of main gate; Phase II/III Floor of Room 3; Phase III topsoil Cleaning of W214; Phase III Fill south of main gate; Phase II/III topsoil Bedrock inside church Bedrock inside church Fill inside church; Phase II/III Floor of Courtyard 2; Phase II Fill inside entrance Corridor 7; Phase II/III topsoil Fill south of main gate; Phase II/III Fill in Courtyard 2 topsoil Fill south of Room 3 Fill inside entrance Corridor 7; Phase II/III Fill inside Room 4; Phase III Fill south of church 257 locus Square Opening height Closing height 378 379 380 381 382 383 384 385 386 387 388 b4 d4 b4 d4 d4-e4 d4-e4 d5 d5 d5 d5 C5 257.17 256.11 257.13 255.31 257.30 256.60 256.91 256.12 256.12 256.91 256.93 257.13 255.40 256.77 255.29 255.68 256.18 256.89 256.07 255.93 256.70 256.21 389 d5 256.60 256.70 390 391 392 393 394 400 401 402 403 404 500 501 502 503 504 505 506 510 511 512 513 514 d5 C5 d5-e5 C5 C5 e5 e7 e8 e8 d7/8 C8-d8 C8-d8 b5 e9 e9 e9/10 d8-e8 b10 b10 b10 b10 b10 257.10 256.91 256.06 255.31 256.85 256.16 257.22 257.20 256.40 257.30 258.46 258.01 256.00 256.62 257.41 257.43 257.90 255.88 255.56 255.37 255.26 255.26 256.15 256.85 255.60 255.10 256.60 256.10 256.75 256.62 256.10 257.20 258.01 257.95 255.90 256.20 256.62 253.02 257.60 255.56 255.35 254.62 254.66 254.76 515 b10 255.56 255.51 516 517 518 519 520 600 b10 b10 e10 e10 C8-d8 d9 254.75 255.51 252.35 251.75 257.70 258.00 254.62 255.45 251.75 251.65 257.30 257.40 258 description/Stratigraphic relation topsoil Fill south of main gate; Phase II/III Fill above southwestern corner of church Cleaning of floor l359 in Room 3; Phase III Fill south of Room 3 Fill south of Room 3 Mosaic floor in northern isle of church; Phase II Pebbles pavement in Room 33; Phase II Fill below pebbles pavement l385 in Room 33; Phase II Fill below mosaic floor l384 in northern isle of church; Phase II Dismantling of round stone-built base l310 Plaster surface below mosaic floor l384 in northern isle of church; Phase II Dismantling eastern section of W210 Fill below mosaic floor l384 in northern isle of church; Phase II Section through floor l367 of Courtyard 2; Phase II Fill below floor l359 in Room 3; Phase II Tabun in Hall 31; Phase III Floor in Room 16; Phase III Built chamber in Room 13; Phase III Fill/collapse below floor l189; Phase III Fill below floor in Room 11; Phase II Fill below floor of Courtyard 10; Phase II Dismantling of W1 Floor of southern wing of tower 22; Phase II Bedrock in northwestern corner of church Roman-period plastered installation in Oil press 26; Phase I Dismantling of W26 Dismantling of W33 Fill below upper screw weight (outside Oil press 26); Phase III Collapse inside collecting vat of Wine press 35; Phase II/III Wall debris south of late Roman-early Byzantine wine press Fill inside late Roman-early Byzantine wine press; Phase I Fill east of W50; Phase II Fill west of W50; Phase II Plaster layer over mosaic floor of collecting vat of Wine press 35; Phase II Mosaic floor of late Roman-early Byzantine wine press; Phase I Mosaic floor of collecting vat of Wine press 35; Phase II Fill below later floor of Cistern 24; Phase I early floor of Cistern 24; Phase I Fill below floor l501 of southern wing of tower 22; Phase II Fill below floor l289 in Courtyard 8; Phase II lISt of wallS liSt OF WallS2 Wall Square Axis Opening height Closing height 1 4 5 6 7 8 C9 C8 C8 C8 C8 D8/9 e-W e-W e-W N-S e-W N-S 258.46 258.10 257.95 257.86 258.42 258.80 258.01 257.39 257.17 257.54 256.92 258.07 1.1-1.5 0.6 0.9 0.8 1.5-2 0.6-0.8 2.6 3.2 3.2 3.2 10 10.5 9 11 12 13 C9 C/8-10 B-C/7-8 C/9-10 e-W N-S N-S N-S 257.81 257.38 257.51 256.82 257.22 255.03 257.26 256.15 1.5 1.5-2 1.3 1.5 5 15.8 13 2.5 14 C/9-10 N-S 257.10 255.49 1.5 3.5 15 16 B-C/8 C-e/10 e-W e-W 257.54 256.08 257.04 252.96 1.5 1.3 3 24 19 C/9-10 e-W 256.06 255.52 0.6 2.5 20 D/9-10 N-S 257.17 256.15 1.5 3 22 23 D/8-9 D-e/9 N-S e-W 257.70 257.27 256.16 256.62 0.5 0.3 6.7 1.7 24 25 D-e/8 D-e/8 e-W e-W 257.31 256.98 256.87 256.17 0.5 0.2-0.3 5.5 1.7 26 e/8-9 N-S 257.41 256.62 0.7 3 27 D-e/9 e-W 257.20 256.62 0.3 1.5 29 e8 e-W 256.79 255.45 0.5 2 30 e8 N-S 257.27 256.75 0.5-0.7 6 31 32 33 e10 D/9-10 e-F/5-10 e-W N-S N-S 255.02 255.73 257.43 252.46 252.46 253.02 0.7 0.6 0.9-2.5 3.8 3.8 50 34 35 36 37 38 39 40 e9 e9 C-D/8 C-D/9 D9 C/8-9 e/8-9 e-W e-W e-W e-W e-W N-S e-W 255.82 255.22 257.95 258.70 257.10 257.80 256.55 251.90 251.90 257.17 258.20 256.90 257.22 256.25 0.3-0.7 0.3-0.5 0.5 1.3 0.9 1 1 5.1 5.1 3 5 4.2 3 2 41 45 C/7-8 B9 N-S N-S 258.05 255.88 257.75 258.56 0.5 0.8 3 3.3 2 Thickness length description/Stratigraphic relation late wall in tower 22; Phase III Northern wall of Storeroom 29; Phase II Southern wall of Storeroom 29; Phase II Western wall of Storeroom 29; Phase II Southern wall of tower 22; Phase II Central section of western peripheral wall of main monastery complex; Phase II internal wall in tower 22; Phase II Western wall of tower 22; Phase II Western wall of Courtyard 28; Phase II internal foundation wall at the north of tower 22; Phase II internal foundation wall at the north of tower 22; Phase II Northern wall of Courtyard 28; Phase II Northern peripheral wall of main monastery complex; Phase II internal wall in basement of tower 22; Phase II Foundation wall at the north of tower 22; Phase II Western wall of Oil press 26; Phase II internal wall at the western part of Oil press 26; Phase II Southern wall of Oil press 26; Phase II internal wall at the western part of Oil press 26; Phase II internal wall at the southwestern part of Oil press 26; Phase II late internal wall at the western part of Oil press 26; Phase III internal wall at the southeastern part of Oil press 26; Phase II Western external wall at the southeastern part of Oil press 26; Phase II Northern wall of Cistern 24; Phase I Western wall of Cistern 24; Phase I eastern peripheral wall of main monastery complex; Phase II Southern wall of Cistern 25; Phase II Northern wall of Cistern 25; Phase II Northern wall of Storeroom 34; Phase II internal wall in tower 22; Phase II Northern wall of Courtyard 8; Phase II internal wall in tower 22; Phase II internal wall at the southeastern part of Oil press 26; Phase II Western wall of Storeroom 34; Phase II Western wall of collecting vat of Wine press 35; Phase II Not including walls excavated by the iaa (these, when mentioned in the text, are marked with an asterisk [*]; see also Kogan-Zehavi 2008). 259 Wall Square Axis Opening height Closing height 46 B9 e-W 256.10 258.56 0.8 2.5 47 B9 N-S 255.93 258.56 0.15 3 48 B9 e-W 255.37 254.62 0.10 8 49 B9/10 N-S 255.26 254.75 0.8 2.2 50 B10 N-S 255.26 254.81 0.3 1.8 100 D6-8 N-S 258.19 257.24 0.7 16 104 105 106 D9 D8 D7 e-W N-S N-S 258.53 258.48 258.01 258.27 258.10 257.72 1 0.4 0.5 2.1 1.3 1.9 107 108 109 110 111 112 e6/7 D7 D7 D7-e7 D7-e7 D8-e8 N-S N-S N-S e-W e-W e-W 257.48 257.59 258.43 258.42 258.26 258.58 257.00 256.91 257.41 256.05 257.22 257.75 0.5 0.7 0.6 0.9 0.9 0.8 4.3 3.5 2.1 6 4 7 113 e7/8 N-S 257.87 257.40 0.7 7 115 e7 N-S 257.22 256.91 0.4 1.7 116 e7 e-W 257.18 256.88 0.4 1.1 117 D6-e6 e-W 257.52 256.52 0.8-1.2 15 118 120 D6-e6 e7 N-S N-S 256.85 257.22 256.48 256.90 1.2-1.5 0.4 3.8 1.6 121 124 126 128 129 D8 e6/7 e7 e7 D5/6 e-W N-S e-W e-W N-S 258.52 257.10 256.93 256.92 257.52 258.14 256.05 256.47 256.69 256.09 0.5 0.6 0.8 0.7 0.7 1.3 4.3 3.5 3.2 10 130 D6-e6 e-W 256.68 256.25 0.7 10 131 132 133 134 135 136 137 e5/6 D6-e6 D5-e5 D6 e6 e5 B6-C6 N-S e-W e-W N-S e-W e-W e-W 256.38 256.73 256.68 256.73 256.12 255.73 257.31 255.37 256.12 256.07 256.12 255.77 255.27 256.33 0.8 0.5 0.5 0.5 0.8 0.8 1.1 14 5 5 2 2.5 3.4 16 138 139 C6 D5-F5 e-W e-W 257.37 256.50 256.80 256.00 0.7 1 5.4 18.5 140 D/6-7 N-S 257.49 257.24 0.7 6 260 Thickness length description/Stratigraphic relation Southern wall of collecting vat of Wine press 35; Phase II eastern wall of collecting vat of Wine press 35; Phase II Southern wall of late Roman-early Byzantine wine press; Phase I eastern wall of late Roman-early Byzantine wine press; Phase I Wall between late Roman-early Byzantine wine press and tower 22; Phase II Central section of western peripheral wall of main monastery complex; Phase II late internal wall in Courtyard 8; Phase III late internal wall in Courtyard 8; Phase III late wall between Room 18 and Room 19; Phase III internal wall in Room 20; Phase II internal wall in Room 20; Phase II Western wall of Room 21; Phase II Wall between Rooms 20 and 21; Phase II Wall between Rooms 17 and 21; Phase II Wall between Rooms 12 and 17 and rooms 11 and 27; Phase II Wall between Rooms 12 and 13 and Rooms 17 and 21; Phase II eastern wall of northern late chamber in Room 13; Phase III Wall between two late chambers in Room 13; Phase III Wall between Rooms 5, 6 and 15 and Rooms 14, 19 and 20; Phase II Wall between Rooms 5 and 6; Phase II eastern wall of southern late chamber in Room 13; Phase III late internal wall in Courtyard 8; Phase III Wall between Rooms 14 and 20; Phase II Wall between Rooms 12 and 13; Phase II Wall between Rooms 13 and 14; Phase II Southern section of western peripheral wall of main monastery complex; Phase II Wall between Room 4 and Rooms 5 and 6; Phase II Western wall of Rooms 15 and 16; Phase II late internal wall in Room 4; Phase III late northern wall of Courtyard 2; Phase III late internal wall in Room 4; Phase III late wall between Rooms 15 and 16; Phase III Wall between Rooms 15 and 16; Phase II Northern wall of Hall 31 and Room 32; Phase II internal wall in Hall 31; Phase II Southern peripheral wall of main monastery complex; Phase II Wall between Rooms 19 and 20; Phase II lISt of wallS Wall Square Axis Opening height Closing height 144 145 146 147 B/5-6 e6 e/5-6 e/5-6 N-S N-S N-S N-S 256.33 256.90 256.70 256.35 256.00 256.19 256.20 255.33 0.8 0.7-1.2 0.7 0.7 3.2 4.1 3.2 13 148 149 150 e7 D-e/5 e7 e-W e-W N-S 257.23 256.47 257.12 256.93 256.07 256.90 0.4 0.7 0.4 2.5 5 1.6 151 e7 e-W 257.12 256.90 0.4 1.6 200 201 C4 B-D/5 N-S e-W 257.12 257.10 256.60 256.06 0.5 0.9 1 16.7 202 203 204 205 206 207 B-D/3-4 B/3-5 B-C/4 C5 B-D/4 D/5-6 e-W N-S e-W N-S e-W N-S 257.08 256.93 256.96 257.08 256.67 256.99 256.10 255.94 256.56 256.58 255.67 256.12 1 1 1 0.7 1 1 13 13.4 9 4.2 9 12 208 209 210 211 212 213 B/5-6 C5 D-e/5 D/4-5 D4 C5 N-S N-S e-W e-W N-S N-S 257.10 256.85 257.10 256.54 256.56 256.71 256.57 256.34 256.62 254.98 255.80 255.30 1 0.3-1 0.9 1 1 1 7.6 5 7.5 4 3 3 214 215 216 217 218 219 220 223 225 227 229 D/4-5 D5 D/4-5 D5 D/4-5 D4 D/4-5 D4 D5 B/3-5 D4 N-S N-S N-S N-S N-S e-W N-S e-W e-W N-S N-S 256.28 256.37 255.92 256.35 256.20 255.75 256.56 256.56 256.24 256.95 255.91 255.83 255.90 255.13 255.75 255.65 255.41 255.50 255.50 255.60 256.64 255.28 0.7 0.7 0.7-1 0.7 0.3 0.7 1 1.2 0.7 1 0.7 5 4 5 2 1.5 4 4.2 3.2 4.2 13.4 2 230 D/3-4 N-S 255.80 255.51 0.5 2 231 232 233 B4 D5 e5 e-W N-S N-S 256.95 256.37 256.40 256.77 256.05 256.00 0.6 0.6 0.6 2 3.5 2 Thickness length description/Stratigraphic relation Western wall of Room 32; Phase II eastern wall of Room 6; Phase II late internal wall in Room 4; Phase III late western wall of Rooms 15 and 16; Phase III late internal wall in Room 13; Phase III Northern wall of Courtyard 2; Phase II eastern wall of late chamber in Room 12; Phase III Northern wall of late chamber in Room 12; Phase III Western wall of bema in church; Phase II Wall between church and Hall 31 and Room 32; Phase II Southern wall of church; Phase II eastern wall of church’s narthex; Phase II Northern stylobate wall of church; Phase II internal wall in Hall 31; Phase II Southern stylobate wall of church; Phase II Wall between entrance Corridor 7 and Hall 31; Phase II Wall between Hall 31 and Room 32; Phase II late internal wall in Hall 31; Phase III late wall north of main gate; Phase III Northern wall of church’s apse; Phase II eastern wall of church’s southern aisle; Phase II eastern wall of church’s northern aisle; Phase II Western wall of Room 3; Phase II Western wall of Courtyard 2; Phase II eastern wall of Room 3; Phase II late wall in Room 1; Phase III eastern wall of Room 33; Phase II Southern wall of of Room 3; Phase II eastern wall of church’s apse; Phase II Southern wall of church’s apse; Phase II Wall between Rooms 1 and 4; Phase II Western wall of church’s narthex; Phase II eastern wall of path south of main gate; Phase II Western wall of path south of main gate; Phase II internal wall in church’s narthex; Phase II late western wall of Courtyard 2; Phase III internal wall in Courtyard 2; Phase II 261