With the participation of Dan Browning, John Camp, M. Caplan, SeJin Koh, Robert Mullins, Nava Panitz-Cohen, Yoav Shor, Amir Sumakai-Fink, Dalit Weinblatt and Adi Ziv-Esudri
I. Introduction
Background
Beth-Shean Valley is a part of the Jordan Valley, which is itself part of the great Syro-African Rift. Previous surveys, mainly data collected over the course of 30 years by the late N. Zori, have shown that this area was one of the most heavily settled in antiquity in the Land of Israel (1). Settlement patterns in the valley during the Bronze Age, with emphasis on the Middle Bronze Age, were studied by A. Maeir (2).
The goal of the Beth-Shean Valley Archaeological Project is to explore the settlement history and archaeology of the region in the Bronze and Iron Ages. The project is directed by A. Mazar, on behalf of the Institute of Archaeology of the Hebrew University of Jerusalem. The first part of this project included nine seasons of excavations at Tel Beth-Shean between 1989 and 1996 (3), and the results are currently being prepared for publication by the project's staff. The excavations at Beth-Shean showed that during the second and first millennia B.C.E. (Middle Bronze Age to Iron Age II), the mound was only partially occupied, and its settled area probably did not exceed 3.5 acres (14 dunams). Despite its having been an administrative and military centre of the Egyptian New Kingdom, Beth-Shean never developed into a vital urban centre. Surveys in the region clearly indicated that the main Canaanite and Israelite town in the valley was situated at Tel Rehov (Tell es-Sarem), 5 km. south of Beth-Shean.
When the Tel Beth-Shean excavations ended in 1996, it was only natural to begin excavating the main site in the region: Tel Rehov, an excavation made possible with a generous grant from Mr. John Camp. This paper contains a preliminary report on the first two excavation seasons at Tel Rehov, in 1997 and 1998 (4).
Rehov in Historical Sources
Rehov (the Hebrew word for ‘street', often rendered into English as Rehob) was the name of several cities mentioned in the Bible and in various other ancient Eastern sources. Two cities of that name are referred to in the town-list of Asher (Josh. 19:28-30). An Aramaean town and state in Syria is mentioned, mainly in relation to David's conquests (2 Sam. 10:6, 8). Egyptian sources mention a town called Rehov several times, although some of these most probably refer to Rehov in Western Galilee (5). The following sources most likely refer to our Rehov:
In several other Egyptian references mentioning Rehov, it is hard to judge to which town the inscription refers. A bronze vessel from a place called Rehov is mentioned in a papyrus kept in Torino, Italy, which includes accounts dated to the 20th Dynasty (11); and a notation concerning the production of chariot parts at Rehov appears in Papyrus Anastasi IV:17:3 (12). This is 'a fictitious administrative document, composed for the purpose of training scribes in administrative matters' (13). Rehov in the Execration Texts can either be our Rehov or the one in Western Galilee, while Rehov in Tuthmosis III's topographic list (No. 87) is probably the latter (14).
Identification
The identification of Tel Rehov (Arabic: Tell es-Sarem) with the Rehov of the Egyptian texts was first suggested by F.M. Abel, and later accepted by many scholars (15). It is based on the preservation of this name at the nearby Islamic holy tomb esh-Sheikh er-Rihab (c. 1 km. south of the mound) and on the existence of a Jewish town with the same name (Rohob) near the mound during the Byzantine period. This settlement is mentioned by Eusebius as located on the fourth mile from Beth-Shean. The Byzantine site (Horvat Parva = Khirbet Farwana) is located 1 km. north-west of the mound (16).
The Site
Tel Rehov (Israel Map ref. 197.207; UTM Grid 873.594; Figs. 1, 2, one of the largest mounds in Israel, is located in the centre of Beth-Shean Valley, in an alluvial plain. It is about 6 km. west of the Jordan River, 3 km. east of the Gilboa ridge and 5 km. south of Tel Beth-Shean, dominating the north-south road along the Jordan Valley Its total size is 10.2 hectares (102 dunams, including the slopes), and it is divided into an upper and lower mound. The upper mound comprises c. 5 hectares (50 dunams) and rises c. 20 m. above the plain; the lower mound comprises c. 5 hectares and rises c. 8 m. above the surrounding plain (17). Both parts have steep slopes, but the eastern side of the mound suffered severe erosion. A ravine separates the eastern side of the lower mound from the upper mound, and another deep ravine cuts into the upper mound on its eastern side, opening to the fields east of the mound. The lower mound is elongated: c. 315 m. long and c. 135 m. wide. Its outer slopes on the western, northern and part of the eastern sides are very steep and even. The upper part of the lower mound is a plateau descending to the east: there is a 16 m. drop in elevation between Areas C and E, located 240 m. apart. The ravine cutting into the upper mound is probably a result of a geological subsidiary fault, which is perpendicular to a north-south fault running just east of the mound (18). Gates to the lower city may have been located in the opening of the eastern ravine and in the western side of the mound, near the connection between the lower and upper mounds, where there is a sunken depression used today for the dirt road that ascends the mound.
Figure 1. Beth-Shean Valley:
location map
1. Tel Kitan (Tell Musa)
2. Tell el-Sahina
3. Tell al-Mudawar
4. Tell el-Qeseibeh
5. Tel Yishmael
6. Tell Issachar
7. Tel Kuneizir
8. Hamadiya
9. Tel Huga
10. Tel Hasida
11. Hamadiya South
12. Tel Istava
13. Tel Naharon
14. Tel Zahara
15. Tel Zaharon
16. Tel ʿAmal (Tell el-ʿAsi)
17. Tel Shoqeq
18. Tel Tel Socho
19. Tel Arbaʿin
20. Kh. e-Sheikh Muhammad
21. Del Sufim
22. Del Dove
23. Tel Pahah
24. Tel Nimrod
25. Tel Ashtori
26. Tel Nisa
27. Tel Semed
28. Tel Tahash
29. Tel Hasavim
30. Tel Karpas
31. Tel ʿArtal
32. Tel Qataf
33. Tel Masad
34. Tel Teomim A
35. Tel Teomim B
36. Tel Roʿeh
37. Tel Saharon
38. Tel Maluah
39. Tel Nezem
40. Tel Sofarim
41. Tel Menora (Tell Abu Faraj)
42. Tell esh-Shuneh
43. Tell Abu en-Niaj
44. Tell el-Hayyat
45. Tell Abu Haraz
46. Tel Gema (Jemaʿin)
47. Tel Sheqafim
48. Tel Shalem
49. Tel Muleikat
50. Tel Malquah
51. Tel e-Deir
52. Tell el-Beidha
53. Tell el-Hama
54. Tell Abu e-Sus
55. Tell Abu Habil
To the north of the mound there is a brook leading east. A spring in the brook, close to the north-eastern corner of the mound, was probably the main water source of the ancient city. Additional springs are to be found nowadays at short distances from the mound.
Previous Research
Tel Rehov was described and surveyed in the past by several scholars, including W.F. Albright, A. Bergman (Biran) and particularly N. Zori, who published stray finds from the mound. Noteworthy is an Old Babylonian cylinder seal (19). The surveys indicate occupation during all phases of the Bronze and Iron Ages. It is noteworthy that in two visits to Tel Rehov in the early 1920s, Albright defined the main periods represented at the site as the thirteenth-tenth centuries B.C.E., a conclusion confirmed by our first two seasons in the site. A surface survey carried out by our team revealed that the lower mound produced mainly Iron Age I and IIA pottery, while on the upper mound the main periods represented were the Late Bronze Age, Iron Age I-II and Islamic periods.
South of the mound, beyond a narrow stretch of the alluvial plain, there is a rocky ridge which served as a cemetery during the Middle Bronze Age (20). An Iron Age II cemetery was probably located in the fields of Kibbutz Shluhot, west of the mound (21).
Strategy of the Field Work
Using a computerised contour map of the mound based on vertical aerial photographs, we prepared a north-south grid which divided the mound into 20 main 100×100 m. squares (Fig. 3). Each main square was further divided into 400 squares of 5×5 m., each referred to by a letter (A-Z on the east-west line) and a number (1-20 on the north-south line). A government triangulation point at the uppermost part of the mound is at an elevation of 116 m. below sea level. In order to avoid using negative numbers to refer to levels in our documentation, we defined this point as being 100 m. above a theoretical 0 level. Therefore, converting the field elevations to absolute heights requires the subtraction of 216 m. (for example, our excavation height of 95.6 m. is in fact -120.4 m. below sea level).
In 1997 we opened five excavation areas (A-E), and a sixth area (F) was added in 1998. The areas were selected to ensure that the maximum information on the various parts of the mound would be gained in the first two seasons. The goals of excavation in each area were as follows: Area A) to explore occupation layers at the crest of the upper mound; Area B) to study the fortification systems and occupational history of the upper mound; Areas C and E) to explore the latest occupation levels on the lower mound and the nature of the settlement; Area D) to explore the occupational history and fortifications of the lower mound; and Area F) to further explore this area of the lower mound.
A general feature of the site is the predominance of mudbrick architecture. Stones are virtually missing from the surface of the mound and almost all the walls excavated by us are mudbrick, in most cases lacking stone foundations. Sometimes (so far mainly in the Late Bronze Age and Iron Age I) the brick walls have one or two courses of a stone foundation. In a few cases, large basalt or limestone boulders were used for specific architectural purposes.
The Geo-Physical Survey
A geophysical survey of parts of the lower mound was carried out during a period of two weeks in 1997 by Mr. David Maki (IMA Consulting Inc., Minneapolis, MN), with the purpose of detecting large building and fortification lines in the uppermost 1 m. thick debris layer. Geophysical surveying at a site composed mainly of unbaked mudbrick structures and collapsed mudbricks is exceptionally difficult, yet a magnetometer is able to detect brick walls which were destroyed by fire and thus partly burnt. Since a fire put an end to the uppermost layer of the lower city at Tel Rehov, the magnetometric survey could indeed detect massive walls. The survey was done with a FM 36 fluxgate gradiometer which measured the magnetic field gradient. An attempt was made to use ground penetrating radar (pulse EKKO 1000), but it failed to uncover any mudbrick walls.
The magnetometric survey was carried out in three areas: in Area 1) 6,000 sq.m. along the northern part of the lower mound; in Area 2) 800 sq.m. in the eastern part of the lower city; and in Area 3) 300 sq.m. in the western part of the lower city, near the supposed location of a city gate. In his report, Maki concluded that ‘The destruction episode at Tel Rehov enhanced the magnetic signature of the mudbrick walls through thermoremanent magnetization, or TRM. This remanent magnetization made the city walls and structures visible to the magnetic gradiometer. Only those portions of walls that were heated to above the Curie temperature retained TRM. This is why some portions of walls and buildings are clearly visible in the images and other portions are not visible'.
The survey in Area 1 detected large structures and what appears to be fortifications, as well as a possible street. In Area 2, a large building was detected. In Area 3, a circular feature of unknown function was detected, abutting a large wall. By revealing potential architectural finds the survey enables us to more precisely plan future excavation areas.
II. The Lower Mound
In Areas C, E and F in the lower mound our work focused on the uppermost three or four strata, dating from the tenth-ninth centuries B.C.E. Only in Area D did we excavate earlier remains, dating from the Iron Age I and the Late Bronze Age. Table 1 suggests a tentative correlation between the stratigraphic phases in this part of the mound.
Figure 2. View of Tel Rehov (looking north-west; photograph: R. Cleve)
Figure 3. View of Tel Rehov Tel Rehov: topographic map
Area D (Figs. 4, 5)
This area is a step trench excavated along the steep western slope of the mound. The trench consists of one line of five 5 m. wide squares (L-Q4 in main Square 17), a total area of 125 sq.m. The trench runs along a length of 25 m., with an elevation variance of 7 m. Each square was excavated independently, starting from the highest point. Surprisingly, no fortification lines were detected in this trench; instead, 11 Late Bronze Age and Iron Age I stratigraphic phases, including three major building periods, were defined. In the following report on Area D we deliberately use the term 'Phase' in counting the various stratigraphic features, rather than the term 'Stratum' used in the other areas, because the limited exposure in Area D does not yet permit final definition of strata. In all phases, every east-west wall discovered ended with the slope of the mound; it thus appears that the western enclosing wall of all the structures had been severely eroded away. Since it can hardly be assumed that entire city walls were eroded away, we assume that the LB-Iron I cities identified in this section were not fortified.
Interpretation of the finds in the section was made difficult by various factors, such as the limited exposure of each phase, discrepancies in levels, erosion, and animal holes, which cut many elements. Thus, our conclusions are only partial and preliminary. The area's stratigraphy, the location of each phase, dates and main features are summarised in Table 2.
Figure 4. Area D: general view (looking east).
Figure 5. Area D: schematic plan and section of main architectural remains.
Following are some notes on the main occupational phases in Area D, beginning with the lowest one.
Phase D-11
A layer of dark brown silt with several horizontal lines of black ash was found at the bottom of a narrow probe in Squares L4-S. The little pottery found can be dated to the end of the MB II or the beginning of the LB I, including a few sherds of Chocolate-on-White Ware. In the back-hoe trench excavated a few metres to the north of Area D, this silt layer reached a thickness of approximately one metre. Below it was bedrock composed of yellow travertine. The top of the brown silt deposit was at a height of 77 m., 2 m. below the level of the present field to the west of the mound. This dark brown layer at the foot of the mound is perhaps evidence for a marsh which existed nearby at that time, which would explain the soil's fine dark matrix and the periodically deposited black layers.
Phase D-10
This is a layer of light yellow travertine, over 2 m. thick, found at the foot of the mound. It is void of any pottery or other finds, except for one carinated LB bowl. The same layer was detected in two back-hoe trenches, dug at the foot of the mound north of Area D. In the northern trench, located 20 m. north of Area D, the travertine layer was about 6 m. deep. The top of this layer is c. 1.20 m. above the present field west of the mound. The layer is divided into two sub-layers of similar thickness (c. 1 m.); the upper one is more crumbly, while the lower is more homogeneous and dense. Our own conclusions, together with those of the geologists who examined this layer, is that the travertine must have accumulated here in a body of water: a small lake or pond (22). It should be noted that the travertine layer in the narrow trench c. 20 m. north of Area D was much deeper. Thus, it appears that in Area D we are closer to the edge of this pond. Since this layer is sealed by a LB building and is accumulated above a MB/LB I layer, one must conclude that the entire layer accumulated over a time period of between 200 and 300 years, between c. 1550/1500-1300/1250 B.C.E. During this time, the water reached the foot of the mound and the pond could have been large, covering much of the present-day field to the site's west. The pond's existence raises questions as to possible tectonic changes, which prevented the water from being drained to the north-east where a brook now exists, leading to the east. This and related questions will be the subject of a palaeo-environmental study being initiated by the Rehov Expedition and the Israel Geological Institute.
Phase D-9
A substantial building was founded above the travertine layer sometime in the Late Bronze Age II, probably in the thirteenth century B.C.E. A massive I m. wide east-west mudbrick wall with a foundation of small travertine stones was found in Squares L-M4, with two subsidiary walls continuing to the north and south. Altogether, parts of three or four rooms were detected. The east-west wall disappears with the erosion line at the slope of the mound. Thus, it appears that at least on this side of the mound, a tell in the real sense of the word did not yet exist (unless the building was constructed at the foot of and outside a MB rampart, for which there is currently no evidence). The floor in this first phase was made of beaten earth and was constantly raised, resulting in greyish-white striations. In one of the rooms, a complete clay plaque figurine of a naked female was found. The pottery was typical LB II, including some fragments of Cypriote White Slip Ware. In a second phase (D-9a), a new cobble floor was founded above the striations of the earlier phase in the area south of the building's main wall. No evidence for the nature of the destruction could be detected.
Phase D-8
A thick plaster floor, found at levels 80.66-80.13 m., covered the entire excavated area in Square M4, about 0.30 m. above the floor of the previous phase. It post-dates and covers the remains of the previous building, and thus can be defined as an independent stratigraphic phase, even though no walls were found. A c. 1 m. thick accumulation of fallen mudbricks and debris separated this floor from Phase D-7. The pottery from this layer is LB II, typical of the thirteenth century B.C.E.
Phase D-7
This phase was identified in Squares M-N4. The remains belong to a large room in a building whose southern and eastern walls were found, while the western part was eroded away. In this room, below a thick accumulation of fallen debris, a floor was detected, with several pottery vessels and bronze objects on it. A cooking pit in the floor had a complete cooking pot next to it. Two ‘foundation deposits', each containing two pottery bowls with a lamp between them, were found, one below the stone foundation of one of the walls. Such foundation deposits are a well-known feature in the LB/Iron I transition, and were also discovered at Beth-Shean in the same period (23). Other finds included a gypsum bottle and fragments of a faience vessel decorated with palm leaves. The local pottery from this phase is similar to that found'at Beth-Shean in the strata dated to the Egyptian 20th Dynasty (twelfth century B.C.E.); yet Egyptian forms, so common at Beth-Shean, are very rare here.
Phase D-6
This phase includes fragmentary remains in Squares N4 and P4. This was an extremely disturbed area and the relationship between the various scanty remains was difficult to establish; nevertheless, they are all post-Phase D-7 and pre-Phase D-5. They include fragments of flimsy walls and floor surfaces in an open area (the lower at 82.32 m.), which were rapidly accumulated, resulting in striations, which at one point included 65 horizontal superimposed thin layers. Remains of two tabuns, a built basin and another installation were related to this open area. The pottery is very similar to that of the previous phase and should be dated to the second half of the twelfth century, either at or immediately following the end of the Egyptian control of Canaan.
Phase D-5
Two mudbrick walls and a floor in Square N4 are attributed to this phase, as well as a thick layer of brick debris.
Phase D-4
Found in Squares N-Q4, this phase is characterised by major building activity. The remains include a system of several walls with related layers of floor surfaces. Here, two sub-phases were observed. The architectural remains appear on two or three terraces built down the slope. The major east-west wall had a two-course foundation, on which seven courses of mudbrick sat. The stone foundation was built directly on top of a wall of the previous phase, D-S. In one of the rooms, a massive cobblestone floor was found with a build-up comprising 0.4 m. of white striations. The latest floor was covered by more than one metre of fallen bricks and debris. East of this structure there was a narrow alley, in which striations resulting from floor raisings accumulated to a total height of almost LS m. East of the alley, remains of an additional structure were uncovered on a somewhat higher level (the uppermost floor was found at 85.43 m.). By this time, the original entrance to the eastern building had been blocked, since it became too low to accommodate the new floor surfaces in the alley.
Phase D-3
This phase, representing extremely intensive activity, is characterised by a series of at least 33 pits found in Squares N-Q4 (diameter: 0.3-1 m.; depth: very shallow to almost I m.). Some of these pits were lined with white plaster and probably used as storage for grain or other products. Other unplastered pits may have been used for refuse. In some cases, three pits were found cutting one another. The pits contained small quantities of grain and olive pits. It thus appears that that this area's function changed considerably after the destruction of Phase D-4, becoming an open area used for storing food and discarding refuse.
The pottery in Phases D-5 to D-3 is identical: typical Iron Age I pottery with a large amount of painted decoration, mostly dull red on a buff surface. The main designs are horizontal stripes and irregular wavy lines. Red slip and burnish are absent, as are ‘collared-rim' jars. Among the finds are five small sherds of unmistakable Philistine Bichrome pottery, noteworthy in light of the near absence of such pottery at Beth-Shean. About three small sherds of imported Myc. HIC were also found in the debris layers of these phases. These three phases should be dated to the period between the end of Egyptian domination (c. 1140?) and the end of the eleventh century B.C.E.
Phase D-2
This flimsy phase, which was difficult to detect, includes two poorly preserved mudbrick walls in Square P4 and few floor surfaces. A small sherd of a typical Philistine vessel was found in this layer, as well as a tiny cone-shaped seal incised with a figure of a crab.
Phase D-1
The uppermost phase in the section was defined only in Square Q4 at the top of the trench. The layers related to this phase can be divided into two; the lower is a deep accumulation of earth containing mainly Iron Age I pottery, while the upper one contains two thin, poorly preserved mudbrick walls and a surface with two ovens. The pottery related to the latter phase is mixed Iron Age I and tenth century, including red slipped and burnished sherds. It appears that this phase is contemporary with Stratum C-2 in Area C (see below).
Other features in the section remain enigmatic. The main question – to be explored in future seasons – is whether the various phases exposed on the slope represent horizontal occupation layers inside the mound, or whether artificial or natural features (such as a MB rampart or natural hill) forced the inhabitants to build on terraces.
The cultural history of Rehov, as seen in the Area D trench, can be divided into five groups: A) the sixteenth-century ‘marsh' (Phase 11), followed by a pond at the bottom of the mound (Phase 10); B) the thirteenth century (Phases 9-8); Q the twelfth century (Egyptian 20th dynasty; Phases 7-6); D) the late twelfth and eleventh centuries (Phases 5-2); and E) the tenth century (Phase 1).
Groups B-D represent a continuous development of Canaanite culture, in which distinctive definitions of the pottery can be observed. In each of these three groups we witness one major building phase (Phases 9, 7 and 5) followed by a destruction and periods of less extensive construction. However, this is only a preliminary impression, based on limited exposure, and it requires further examination.
Area C (Figs. 6, 7, 8, 9)
This area is located on the uppermost part of the lower mound, near its north western corner. Its western limit is 5-8 m. from the edge of the mound, except for Square R4, which includes the edge of the mound and connects Area C with the step trench in Area D. The goal of excavation in this area was to give wide exposure to the uppermost strata in the lower city. A total area of 275 sq.m. was excavated.
Three strata were detected: C-1, C-2 and C-3. Stratum C-1 was excavated in the entire area, C-2 in part of the area, and C-3 in only a few probes. The uppermost stratum (C-1) utilises some of the structures of C-2, yet to a large extent is rebuilt with a new plan. This final stratum was destroyed in a heavy fire. In Stratum C-1, we were able to detect two sub-phases, indicating its long duration.
Stratum C-3
Only one room of this stratum and a small part of another room were uncovered in Square S2. The walls are made of typical grey bricks, unknown in later strata. Three of the walls are right below Stratum C-2 walls, indicating continuity between the two, while the other two walls were not rebuilt in the following stratum. Finds are scarce. This stratum may be contemporary with Phase D-2 in Area D.
Stratum C-2 (Fig. 7)
This stratum was exposed in parts of Area C and was very well preserved in certain places. The building remains seem to belong to two or three architectural complexes, which appear to be extensive and well planned, although they cannot yet be classified typologically, due to incomplete exposure. The walls are 0.6-0.8 m. wide and are made of typical hard yellow-beige mudbricks, measuring c. 0.60ª0.35ª0.17 m. (Fig. 8).
In the southern part of the area (Squares T-Y1-2) a long north-south wall (2468) was found, just below a wall of Stratum C-1. This was perhaps the western wall of a building that continues to the east. An entrance leads through the wall to the east, and a partition wall (2465), extending to the east of this wall, was detected below C-1 layers in Square Y1. West of this building there were accumulations of occupation debris in Squares S-T1, creating striations which contained abundant pottery. Noteworthy is a clay female figurine of distorted proportions and unusual features, with legs crossed (?) below the body.
In the central part of the area (Squares S-T2-3), there is a large hall (2473). Three of its walls were founded on the grey brick walls of Stratum C-3, yet the smaller rooms of the previous stratum were replaced by the new large hall. Its inner dimensions are 3.5ª6.3 m. (22 sq.m.), and its walls are preserved up to a height of 18 courses. Two 0.9 m. wide entrances to this hall were preserved to their entire original height (1.5 m.) in the eastern and northern walls; a depression in the upper part of the eastern entranceway indicates the existence of a wooden lintel, of which only some carbonised charcoal pieces survived.
South-west of the northern entrance there was a large limestone boulder, found turned on its side. The use of stone in this stratum is extremely rare, and this one could have been a pillar base or a working table of some sort. We were not able to determine the floor of the hall with certainty. It appears to have been made of layers of beaten earth, striations and ash, ranging from 85.47 to 84.54 m. The finds included primarily fragmentary pottery vessels, consisting mainly of bowls, jugs and juglets, but no storage vessels. Red slip and hand burnish were abundant, and several calcite vessels were found. A thick mudbrick collapse, with many whole bricks, was found above the floor, although there was no evidence of fire. This severe collapse may be evidence of an earthquake. Another hint of seismic activity is a split or seam in the northern wall of the hall, suggesting that the wall was torn into two.
This hall appears to belong to a much larger architectural complex, the extent of which is still unknown. A unit north of the hall (5.2ª6.3 m.) is divided into a main chamber (inner dimensions 2.5ª3.55 m.) and two small chambers to the south. Occupation surfaces in Square T3 and remains of an oven and a brick platform in Square T4 may indicate the existence of an open courtyard in this area.
An analysis of the plan of Stratum C-2 at this stage is perhaps premature. Yet it appears that the large hall (2473) and the structural remains in the southern part of the area (Walls 2468 and 246S) could belong to one large unit, with an open courtyard in Squares S-T1-2, leading through Wall 2468 to rooms in the east, from which there may have been access to the large hall's eastern entrance. The fact that Walls 1448 and 1442 are built 'back to back' alludes to the possibility that this is the centre of an insula and that Unit 2491 was a separate building, although its full plan is as yet unknown.
The pottery of Stratum C-2 is characterised by red slip and hand burnish, which appear on many of the small vessels, while painted decoration becomes rare.
Stratum C-1 (Fig. 9)
After the destruction of Stratum C-2, the buildings in Area C were rebuilt on a somewhat different plan, with evidence of centralised integral planning, as indicated'by the very long and straight backbone wall crossing the area from north to south. Almost all the walls of Stratum C-1 were new mudbrick walls. The buildings can be divided into four units.
In the southern part of the area, an architectural unit was partially exposed. Two constructional phases (C-1a and C-1b) were defined, separated by debris, c. 0.5 m. deep. The inner partition and two of the outer walls of this unit were rebuilt in the transition between the two sub-phases, while a third wall continued to be used without change. In both phases the occupation debris, the nature of the finds and the installations indicate domestic activity. The last phase, foundjust below topsoil, ended in a severe destruction; many restorable vessels were found on the floors.
The central part of the area now became an open space. Remnants of a stone installation or pavement were found in the centre of this area, and a white hard-packed floor covered most of the space. The Stratum 2 structure in Squares S3-4 was rebuilt, its inner division was cancelled, and it was turned into a large rectangular courtyard or hall (1474; inner dimensions 4.3×5.15 m.). The new eastern wall of the structure now blocked the northern entrance to Hall 2473 in the south. In Square T4, part of an oven, a plastered bin (?), and a large basalt basin with a central sunken area and eight 'cupmarks' on its rim were found in Phase C-1a; an earlier oven was attributed to Phase C-1b.
On the eastern side of the area there stood a massive building, some of its walls 0.9 m. wide. One of these walls was built right on top of an earlier Stratum C-2 wall. One part of this building consisted of two small chambers, without entrances and surrounded by massive walls on all four sides. These appear to be basement compartments, perhaps entered from above by ladders. One of the chambers contained charred wheat; in the other, a group of pottery vessels was found. South of this unit there were three other spaces, two adjoining and a third which might have been related to the same building. Their widths are 2.55, 3.4 and 2.55 m. respectively; their length is unknown since they continue to the east, beyond the limit of the excavation. A common feature of these rooms is the use of wooden logs or beams as a foundation for the walls and floors. The wood was sometimes found to have been laid in several superimposed and interspersed layers. Such wood was also found in the southern compartment room described above. All the beams were carbonised, and in the southern part of the area they were found tilted at a sharp angle down to the east (Fig. 10). The trees used for this construction were identified by Dr. U. Baruch (Israel Antiquities Authority) as belonging to various species, such as elm, olive, acacia, Judas tree, Syrian ash and mulberry. This type of wooden construction serving as a foundation for both mudbrick walls and clay and plaster floors is unusual, and unparalleled elsewhere in the Levant. It may have been intended to protect the building against earthquakes, which present a hazard in the Jordan Valley, or it could be the roof of a basement, still unexcavated.
In the corner of one of the rooms in this area, a group of pottery vessels was found in situ under a 1.2m. thick layer of collapsed bricks and ashed (Fig. 11). The assemblage included two jars of the ‘hippo' type, a krater, two jugs and several other vessels, including a red-slip amphoriskos painted with geometric designs and a degenerated palm-tree motif. A painted globular jug with a delicate ring base and painted concentric circles in red and black is imported Cypriote Bichrome Ware, inspired by Phoenician prototypes.
The other rooms also contained rich pottery assemblages. In the north-eastern corner of the area, a broad room (2489) was excavated. It was entered from the north and probably belonged to an independent building which continues beyond the excavated area. It stands ‘back to back' in relation to the eastern building described above. A goose (?) egg and the claws of a river crab were found close to the floor, which was covered with an organic burnt matrix and contained a number of vessels, loom weights and beads.
In Square R4, which connects Areas C and D, a Stratum C-1 room was exposed just below topsoil, west of the long backbone wall which runs along the entire length of Area C. A group of over ten restorable storage jars with pottery lids, multi-handled kraters and other vessels was found here in a destruction level. Since this room reaches the western edge of the mound, it leaves no space for a fortification wall; it thus appears that the city of Stratum C-1 was not fortified.
Stratum C-1 came to an end in a violent destruction, testified to by some heavily burnt mudbrick walls and by the many broken vessels found in situ within the thick destruction debris on the floors. There is no other activity in this area after that, except one wall which may suggest a post-destruction phase.
Area E
This area is located on the highest point of the lower mound's eastern side. Ten squares were opened here (E-F13-16 and Dl5-16), covering a total area of 250 sq.m. In the south, parts of a well-preserved building were exposed, while in the north-west there was an open space: a courtyard or piazza, with a cultic corner and a building to its west. One major stratum was excavated here, with two sub-phases: E-1a and E-1b, probably corresponding to Phases C-1a and C-lb in Area C.
A building exposed in the southern part of Area E consists of two rooms, the western of which was badly preserved, while the eastern (in Square F13) was a well-preserved square room (inner dimensions: 2.4×2.4 m.), constructed of mudbrick walls covered with white plaster. Inside the room there were two clay compartments, probably for storing food commodities. The room's eastern and northern walls abutted walls of what appears to be a neighbouring building (Square F14). At least one room of this structure was in use only in Phase E-1b; in Phase E-1a one of its walls was covered by a white lime floor, on which there was a destruction layer close to topsoil, with a rich group of restorable pottery vessels (Fig. 14), resembling the pottery deposits found in Area C.
The open courtyard in the northern and western parts of Area E covered an extensive tract, that continues beyond the limits of the excavation on the north and west. The excavated open area covers c. 90 sq.m. The initial floor surface was made of compact earth and small bits of gravel. Five ovens (tabuns) and six circular clay installations resembling ovens were found in this area; the latter, showing no signs of fire, apparently served a different function relating to food storage and preparation. Some of these ovens and installations replaced earlier ones. Successive surfaces in this area, observed in the northern squares (E14-16, F15-16), accumulated to a total height of c. 0.6-0.9 m.
In Squares D-E15 a square mudbrick platform (3ª3.55 m.) was found, c. 0.4 m. above the original E-1b surface around it. On top of this platform there was a smaller square installation (1.0×1.0 m.), made of fieldstones and large river pebbles, rising to a height of 0.33 m. On its southern side (oriented north-east to south-west) there were four larger stones, three of which were elongated and stood on their narrow side (c. 0.3×0.5 m.). In front of the brick platform (in Square E15) there was a large flat limestone slab (0.7×0.5 m.), supported by five smaller stones, possibly an offering table.
In a tumble of debris next to the platform (also in Square E15), there were fragments of a large square pottery cult stand, with a flat top into which a hollow cut was made after firing (Fig. 15). The protruding rim above this flat part was deliberately removed. The stand's fagade has two rows of triangular windows, standing alternately on their base and apex. Above the windows is a decoration made of reed impressions; only part of this design was preserved. This cult stand recalls two similar stands from Pella, dated to the tenth century B.C.E. (24) Two clay female figurines (Fig. 16) and an animal figurine were also found in the vicinity.
Figure 14. Area E, Stratum E-1: group of potttery vessels
Figure 6. Area C: view at end of 1998 season (looking west)
Figure 8. Area C, Stratum C-2: detail of northern mudbrick wall of Hall 2473 (looking north).
Figure 10. Area C, Stratum C-1, Squares Y1-2: detail of charred timber (looking west).
Figure 11. Area C, Stratum C-1: group of pottery vessels.
Figure 13. Area E: general view at end of 1997 season
Figure 15. Area E: pottery cult stand.
Figure 16. Area E: clay figurine.
It appears that this complex was a high place (bamah) serving the vicinity (Fig. 17). The standing stones can be interpreted as massebot standing on a ritual platform at the edge of a spacious courtyard, where a number of ovens and other installations were used for preparing sacred meals. The flat stone in front of the platform may be an offering table, with the pottery cult stand being used similarly to the small stone altars known from Megiddo and elsewhere. The flat top of these stands could be used for burning offerings, such as pigeons (25). The cult place was probably built in Phase E-1b, and remained in use until the end of Phase E-1a. In Phase E-1a, the surfaces around it were raised almost to the level of the top of the brick podium. This may explain the additional stone square built on top of the square brick podium, which might have been an addition in Phase E-1a.
In Phase E-1a, a room was built to the west of the podium (Square D15), with its eastern wall attached to - and even overlapping by 0.05 m. - the western edge of the podium. The room was found just below topsoil. A roof collapse was observed inside it, in the shape of clay lumps with many reed impressions, covering a destruction layer that contained abundant pottery vessels.
Many animal bones were found in the open area in front and east of the podium. Unlike those found in the other areas at Tel Rehov, which included mainly sheep, goat and cattle, in Area C there were many bones of wild goat, suggesting that this animal was used specifically for ritual here. The pottery found to the south-east of this open area included a number of chalices, a further indication of cultic activity.
In sum, this area provides evidence for a cultic centre which may have served a small community – perhaps an extended family living in this neighbourhood. There is no similar phenomenon found in Iron Age Israel, although the concept of a small cult place serving a local community is well known from approximately the same time at Megiddo, Taanach and Lachish (26). At none of these sites, however, are standing stones and podia reported, although massebot of a similar rough shape and small size are known from the city gate area at Dan (27).
Area F
This area, excavated only during the 1998 season, is located 10 m. south of Area E, on the slope descending towards the ravine separating the lower from the upper city. The purpose of the excavation here was to expose more of the upper occupation layers in the lower city, continuing the wide exposure in Area E. An area of c. 95 sq.m. was excavated here.
The stratigraphy distinctly recalls that of Area E, and the upper layer is definitely a continuation of the upper layer in that area. For example, the northern wall of a rectangular room found in the northern part of Area F continues the southern wall in Area E. Together, the two areas demonstrate the sophisticated urban planning of this period. The fact that this area is located on a slope creates some stratigraphic problems, since buildings of the same period may have been built on different levels in adjacent squares. The stratigraphic analysis in such cases is based on structures which were evidently superimposed.
Four strata were defined. The lowest (F-4), exposed in a very small area, is attributed to the last phase of the Iron Age I, on the basis of a few painted sherds. Stratum F-3 includes a rectangular architectural unit with a row of five pillar bases made of unworked flat stones. Red slip and hand burnish appear among the few pottery sherds from this level. Stratum F-2, defined only in the centre of the area, includes a strange installation made of three large flat stones (the largest 1.5 m. long), two parallel to one another and the third perpendicular to the others. These exceptional large and well-worked stones are surrounded on three sides by a narrow mudbrick wall (in fact, a parapet), which creates a small room continuing to the east beyond the line of the stones. A north-south wall was found to the west of the installation. Finds were sparse, and there is no indication as to the function of this unusual installation. Three large stones that had been removed prior to our excavation were located about 3 m. to the north of our installation, and may belong to the same structure, although their stratigraphic attribution is not secure: they could equally well belong to Stratum F-1. In Stratum F-1, a new building was erected, on top of the installation of F-2. A large rectangular room of this building was exposed (2.3 m. wide by at least 5 m. long). A square area paved with small stones was found to its south.
Most of the walls and floors in this area were found tilted towards the south or south-east. This could be due to seismic and tectonic activity, since the area is located very close to a geological fault.
III. The Upper Mound
(Table 3)
Area A
Excavations in this area began with eight squares arranged in an L-shape, close to the summit of the upper mound (Grid 10, Squares G3-6; G-K2). One week of excavation here yielded evidence of Islamic occupation, but almost nothing in the way of architectural remains. In some of the northern squares we found plaster floors and the stone foundations of a wall, but much was removed by erosion. In the eastern squares, sloping layers of soft brown earth alternating with black ash probably indicate that this was an open area where refuse was thrown and accumulated over a long period of time. A probe in Square K2 showed that the Islamic debris reached a depth of over 2.5 m. The earliest Islamic pottery found on the mound dates from the Umayyad-Abbasid period. The bulk of the sherds, however, are Medieval, dating mainly from the twelfth-thirteenth centuries C.E. (28)
Figure 17. Area E: detail of high place and standing stones (looking east)
Figure 18. Area F: general view (looking east)
Due to the extreme depth of the Islamic debris and our original intention to explore the earlier periods, work in this part of Area A was suspended and the squares back-filled. We then continued excavating in the lower eastern part of the upper mound, where we opened 3.5 squares on an east-west line (Squares R-Y2; Fig 20). Here we defined a total of six occupation phases. In the following, we use the term 'Stratum' for each of these phases, although future research may determine that some of these 'strata' are merely phases of one single stratum.
Stratum A-5
The earliest Iron Age II level was excavated in only a few small areas. In Square T2, we uncovered the top of a mudbrick wall (1061) amongst fallen and burnt mudbrick debris. Beaten earth surfaces in Square S2 also apparently belong to this level. The floors yielded Iron Age II pottery, possibly of the late ninth-early eighth centuries B.C.E.
Strata A-4 and A-4'
Remains of an Iron Age II dwelling were excavated in Squares T-Y2. The building continues to the north and east, and includes several mudbrick walls and the remains of two brick bins. An open courtyard was uncovered in Squares R-S2 to the west. The courtyard surface abuts a mudbrick wall on the east. In about the middle of the courtyard the partial remains of an oven (tabun) were uncovered. A 3.5 m. wide north-south street passes between the courtyard and the dwelling, and may intersect another street or a plaza in Square T2, which abuts the southern side of the dwelling. At a later point in time, the inhabitants of Rehov added a new floor to the dwelling in Square Y2 and laid down a new courtyard over the old one, building a new oven slightly further to the west. Only the northern half of this new oven survived, containing eighth-century B.C.E. pottery. The eastern edge of the new courtyard continued to abut the same wall as the earlier surface.
Stratum A-3
Changes to the Stratum 4 building were made by attaching a new wing to the south and raising the floors in the old wing. The new addition (1158 and 1169) covered the earlier east-west street, although the north-south road continued to exist, with evidence of continued street build-up. A brick wall (1134), bordering the western side of the street, was built directly over its predecessor. A new courtyard was laid down in Square R2, with the addition of two new walls (1124 and 1174). The western side of Wall 1174 nearly abuts a limestone and basalt floor (1173), which is still largely hidden in section. Room 1158 and the north-western corner of the courtyard (1145) were destroyed in a heavy conflagration. Ash in Locus 1145 was as much as US m. thick. In two rooms, human skeletons were found. One of these, in Room 1158, was in a contracted position with the skull missing. Nearby lay two northern-style decanters and several other vessels typical of the eighth century B.C.E., as well as many unfired clay loom weights. In an adjacent room, the skeleton of a young woman was found in a position indicating that she was thrown into the corner of the room and died on the spot, her hands holding the knees of her contracted legs (Fig. 21). This may possibly have been a result of a massacre of people in their homes, and is clear evidence for a violent and dramatic end of the Iron Age II city. We are tentatively attributing this destruction to the Assyrian conquest of 732 B.C.E.
Stratum A-2b
This stratum is architecturally poor, and is represented by only a few north-south walls, plaster courtyard surfaces and an oven at the eastern edge of Square R2. Two walls of the previous stratum continued to be used in this one. We assume that Stratum A-2b represents a settlement following the Assyrian destruction of 732 B.C.E.
Stratum A-2a
A burial found inside a shallow pit in Square S2 post-dates Stratum A-2b. A poorly preserved human skeleton, oriented north-south (the head aimed south) lay in a contracted position; at the edge of the pit lay an intact Assyrian bottle. This appears to be a chance burial during the Assyrian period, although no evidence for an occupation on the mound from this time has been found so far.
Stratum A-1
The top layer contained Early Islamic occupation in the form of a paved stone courtyard and a nearby oven in Square S2. This was covered over by Medieval period debris with no architectural remains. In the westernmost square, this layer reached nearly 1 m. deep, thinning out to the east due to the slope of the mound. Several deep Medieval pits and one burial of unknown date cut into the Iron Age II layers, damaging some of the features below.
Area B (Figs. 22, 23)
A total of 12 squares (300 sq.m.) was excavated on the northern slope of the upper mound, at a location overlooking Areas C and D. The following six strata were defined:
Strata B-6 and B-5
The partial remains of at least two Iron Age IIA strata (tenth-ninth centuries B.C.E.) were found in small probes excavated in the lower squares. The remains include a fragmentary floor surface and oven in Square G3, and parts of two mudbrick walls: a lower wall (Stratum B-6) in Square F2 and a higher wall (Stratum B-5) in Square G2. A body sherd of a storage jar or jug with a fragmentary inscription incised on it was found in the accumulation above a B-6 floor in Square G3. The inscription contains three letters: lnh, perhaps 'belonging to Nah[um]'. However, the letter h is unclear, since it includes various other incised lines (29). The small amount of pottery related to these strata is identical to that found in the two uppermost strata in the lower city (Strata C-1-2 and E-1-2), and dates from the tenth-ninth centuries B.C.E.
Stratum B-4
An Iron Age II occupation layer in Square G2, below the foundations of the Stratum B-3 city wall, contained a broken clay figurine depicting a woman playing the tambourine, and fragments of Phoenician red-slipped ‘Achziv' Ware jugs. To the south of this, the partial remains of a dwelling were found in Squares G19-20. The northern end of this structure, which appears to have been some sort of a brick platform, was cut when builders laid the foundations for the mudbrick city wall in the subsequent phase.
Stratum B-3
A wide and massive mudbrick fortification wall (1200) was found in most of the excavated area. It is at least 9 m. wide and was exposed across the entire width of the area (17 m.). It is oriented north-east to south-west, parallel to the edge of the upper mound. Wall 1200 was built of poor-quality square and rectangular mudbricks (average sizes: 0.35×0.35×0.10 m.; 0.50×0.35×0.12 m.). It also lacked a stone foundation, which is unusual for a wall of such magnitude. The northern face of the wall stood to a height of one or two courses, while the southern face was preserved to a height of over 1.60 m. with more than 14 courses of mudbrick. Numerous Islamic graves disturbed the wall in many places; nevertheless, the bottom two courses of mudbricks were well preserved in most places.
The north-western part of the wall within Area B was badly damaged by erosion, impeding our understanding of the wall's course at this point. Nevertheless, there are two points where it seems to turn or to make a recess: one in Square El and the other in Square G2. The inner face is preserved much better, and we were able to trace it along a distance of 9 m. At the western end, in Square F19, the wall makes an inset of 2 m., corresponding to what appear to be offsets on the northern face. It may be possible, then, to reconstruct it as an offset-inset wall, a fortification with inner and outer buttresses, or perhaps part of a bastion incorporated into a narrower wall.
In 1997, the eroded and damaged north-eastern part of the wall in Squares F-G1 was measured and drawn. When this section was removed in 1998, we discovered a thin layer of decayed mudbrick debris below (c. 0.10-0.15 m. thick), covering a double-wall system that looks more or less like part of a casemate wall: an outer wall (1.35 m. wide), a parallel inner wall (1.2 m. wide) and a division wall (1 m. wide), creating two, or possibly three, rooms. These walls were made of mudbrick, like those of the fortification above. They were preserved to a height of two courses and had no stone foundations. The only evidence for a floor was found in Locus 2280 (Square Fl). The interpretation of this structure is as yet uncertain, but three possibilities, to be explored in detail in future seasons, present themselves: 1) that it was a structural feature designed to stabilise the foundations of the massive mudbrick wall above; 2) that it represents an independent architectural phase, probably a casemate fortification; 3) that it was part of an unfinished or badly damaged building.
In Squares G19-20, the building of Stratum B-4 appeared to continue in use in Stratum B-3, though its orientation is entirely different from that of Wall 1200, which cuts its northern end. The remains in the dwelling include a brick platform and a 3.6 m. wide courtyard with an oven and plastered basin. In a destruction layer found in the courtyard, there were several restorable vessels typical of the eighth century B.C.E. Thus, it seems that this dwelling was destroyed together with the city wall. This supports our general impression that the fortification wall was hastily constructed over earlier occupation debris, which may be as late as the eighth century B.C.E. The wall may have been built as an emergency measure, protecting the town against an anticipated Assyrian siege. We also assume that the huge width of the wall was intended to withstand the threat of the Assyrian battering rams.
Stratum B-2
A few scattered remains of Iron Age II activity post-date the destruction of the city wall and the dwelling mentioned above. In Squares G19-20, these include three east-west mudbrick walls above the ruined eighth-century building. In Square G20, above the brick platform of the earlier dwelling, a living surface (1226) was found abutting the inner face of the ruined city wall. This, together with an open area in Square F19, indicate that structures continued to exist inside the city wall even after it had ceased to function as a fortification. Moreover, on the ruined remains of the city wall itself there were a few isolated fragmentary 'surfaces' with restorable pottery vessels and concentrations of loom weights. This may suggest a short period of reoccupation following the Assyrian conquest of 732 B.C.E. This level corresponds to the ‘post-destruction' occupation defined as Stratum A-2 in Area A.
Stratum B-1
Islamic graves were found just below topsoil, mostly cut as rectangular or oval pits into the mudbrick superstructure of the Iron Age II fortification and dwellings. In each grave there was one skeleton of an adult or a child lying on its side with its face towards the south. There were no gifts and little ceramic evidence for the date of these graves. A couple of graves yielded fragments of Early Islamic pottery. In a few cases, complete bricks pulled up from the ruined Iron Age fortification were used to line the pit and construct gabled coverings for the individual graves (30). This indicates that the top of the eighth-century city wall was exposed in the Islamic period. Finally, the discovery of Islamic period loom weights and basalt grinding stones shows that activity took place in the vicinity of the old Iron Age city wall.
IV. Potttery and Chronology
The first two seasons at Tel Rehov revealed a rich variety of pottery and other artefacts from the thirteenth to the eighth centuries B.C.E. These ceramic assemblages are still under study and the following are only a few preliminary remarks.
As mentioned above, the various phases in the step trench in Area D exhibit a continuous development of pottery from the twelfth until the late eleventh centuries B.C.E. In this sequence, three main assemblages can be defined:
The Canaanite tradition in the pottery industry changes dramatically in Strata C-2 and E-2. The painted tradition gives way to a new industry, characterised by red slip and hand burnish on bowls, kraters, jugs and juglets. Painting is limited to a few geometric designs, often on red slip. Black paint on red slip appears in small quantities, and Phoenician Black-on-Red starts to appear. The inner development between the two uppermost strata in the lower city still needs to be studied in detail. A preliminary observation shows a great deal of continuity between these two strata. Rich assemblages with dozens of complete vessels are known from the destruction of Strata C-1 and E-1. These include many jars of the 'Hippo' type (31), a large number of cooking pots with elongated rims, chalices and many other vessels typical of the tenth century and the early part of the ninth century B.C.E. (Fig. 24). Parallels are mainly with Megiddo Stratum IVB-VA, Taanach Stratum IIA-IIB (32), Jezreel (final phase) (33) Hazor Strata X-IX and to some extent Stratum VIII (34), and Horvat Rosh Zayit (35).
Although this assemblage is traditionally dated to the tenth century B.C.E., the discovery of very similar vessels in the destruction level of Jezreel, which must be dated to after the end of the Omride dynasty (36), led Finkelstein to suggest that all the similar assemblages mentioned above should also be dated to the ninth century B.C.E., while assemblages that were traditionally dated to the eleventh century B.C.E. (like Megiddo VIA) should be lowered to the tenth century (37). In accordance with the evidence from Jezreel, we accept the lowering of the latest possible date of this assemblage to between 840 and 830 B.C.E., and this may well be the date for the end of our Stratum 1 in the lower mound. Such a date is also supported by the find of a Cypriote Bichrome jug, which is dated in Cyprus to no earlier than 830 B.C.E. (although this date is based on mere typological seriation of pottery from tomb groups), as well as by C14 dates, although the latter also justify an earlier date (see below). However, unlike Finkelstein, it is my claim that this assemblage had a long duration, and that it was well known already during the tenth century B.C.E. (38) This conclusion is supported by the ceramic assemblage from the lower stratum at Tel Rehov (Stratum 2 in the lower city), which is very similar to that of Stratum 1 and is characterised by an abundance of red slip and hand burnish. In both Strata 1 and 2 in the lower city we found stratigraphic sub-phases, indicating a rather long time span for each one of them. Stratum 2 can be dated to the tenth century B.C.E., or at least to the second half of that century.
Carbon 14 dates from lower city Stratum 1 support this chronology, although they are ambiguous to some extent. Nine samples from a large quantity of charred grain found in a single sealed context in the destruction layer of Stratum C-1 were dated by the University of Arizona laboratory to 2750 ± 16 years BP, with great precision (39) The calibrated date is 906-843 B.C.E. (Sigma 1, probability 65%) or 916-832 B.C.E. (Sigma 2, probability 98%). The lower end of this time scale fits a destruction of the lower city at the same time as Jezreel, while the upper end is just a few years after the raid of Shishak, traditionally dated to 925 B.C.E. Since Rehov is mentioned as one of the cities captured by Shishak, we may look for a destruction level caused by his conquest (although Na'aman recently claimed that Shishak did not actually destroy cities (40). Most of the pottery and the upper limits of the C14 dates from the destruction of Stratum 1 in the lower mound may allow us to attribute this destruction to Shishak, yet it is also possible to lower the date of this destruction to the ninth century B.C.E., and to attribute it to the same event that caused the destruction of Jezreel. Candidates may be the invasion by Hazael (some time after 841 B.C.E.), Jehu's revolt, or the western. campaign of Shalmaneser III. The earlier wars with Ben Hadad I can also be taken into account in the case of Rehov. Some evidence for the construction date of Stratum 1 in Area C can be obtained from C14 dates of charred timber found in the foundations of the structures. Two of these logs were dated at the Weizmann Institute. An olive-tree beam was dated, as can be expected, to a much earlier period (1400-1260 B.C.E.), while a beam of an elm tree, which has much shorter life span, was dated to 1120-990 B.C.E. Both beams could somewhat pre-date their use for construction, yet the elm tree date hints that the transition between Strata 2 and 1 in the lower city occurred sometime in the tenth century B.C.E. Even though we do not have C14 dates from Stratum 2, it appears, on stratigraphic and ceramic grounds, that it was both constructed and destroyed sometime in the tenth century B.C.E.
At this point in our research, two chronological schemes are possible. The first is that Stratum 1 in the lower city was destroyed by Shishak, and that Stratum 2 belongs to a still earlier part of the tenth century B.C.E. The second possibility is that the lower city Stratum 1 was destroyed sometime in the mid-ninth century B.C.E., and that Stratum 2, with its sub-phases, represents the tenth century in general, perhaps having been destroyed by Shishak. In either case, we cannot accept the lowering of the Iron Age I assemblages, like those of Megiddo VIA, to the second half of the tenth century B.C.E., as suggested by Finkelstein. This would leave insufficient time for the two strata at Tel Rehov, with their sub-phases (41) Thus, the comparative stratigraphy, pottery assemblages and C14 dates, confirm, in my opinion, the following sequence: 1) Megiddo VIA assemblage and related sites: the eleventh until the early tenth century B.C.E.; 2) northern Iron IIA assemblage: from the first half of the tenth century B.C.E. until c. 840 or 830 B.C.E. The long duration of this assemblage does not necessarily suggest that all strata with the same pottery were destroyed at the same time: a time span of 60-70 years between destructions of sites with the same pottery should not be surprising.
V. Conclusions
The survey and the first two seasons at Tel Rehov clearly indicate that this was one of the largest Late Bronze Age and Iron Age I-IIA sites in Israel. The extent of the mound in earlier periods has yet to be studied. The stratigraphic section in Area D and the results of the work in Areas C, E and F show that during the thirteenthninth centuries B.C.E., the entire mound was heavily settled and there was a continuity of tirban life on the mound, despite the rather large number of destructions and rebuildings. In spite of its thriving urban character, the city came to an end in a conflagration which can be dated either to Shishak's raid, or, more plausibly, to one of the events in the mid-ninth century B.C.E.
Following this destruction, the lower mound was abandoned and the city was restricted to the upper mound. This reduced Iron Age II city survived, with various occupation phases, until the Assyrian conquest of 734-732 B.C.E. The huge defence wall found in Area B was perhaps constructed shortly before the Assyrian attack, with the Assyrian conquest dramatically documented by the destruction and evidence of slaughter in the Israelite houses.
Slight stratigraphic evidence in Areas A and B indicates a short period of occupation after the Assyrian conquest, yet the duration and scope of this settlement requires further clarification, Following this last Iron Age settlement there was a long occupation gap. During the Early Islamic period, the summit of the upper mound was resettled, and this settlement probably flourished during the Medieval period, continuing until the Mamluk or early Ottoman period.
The promising results from the first two seasons of excavation at Tel Rehov strongly suggest that this will become a key site for Iron Age archaeology in Israel in the coming years.
Footnotes
1. N. Zori: An Archaeological Survey of the Beth-Shean Valley, in J. Aviram (ed.): The Beth-Shean Valley. The 17th Archaeological Convention, Jerusalem, 1962, pp. 13S-198 (Hebrew). (back)
2. A. Maeir: The Material Culture of the Central Jordan Valley during the Middle Bronze II Period: Pottery and Settlement Pattern (unpublished Ph.D. diss., The Hebrew University), Jerusalem, 1997. (back)
3. A. Mazar: The Excavations at Tel Beth-Shean in 1989-1990, in A. Biran and J. Aviram (eds.): Biblical Archaeology Today 1990Z. Proceedings of the Second International Congress on Biblical Archaeology, Jerusalem, 1993, pp. 606-619; idem, Beth-Shean in the Iron Age: Preliminary Report and Conclusions of the 1990-1991 Excavations, IEJ 43 (1993), pp. 201-229; idem, Tel Beth-Shean, in NEAEHL 1, pp. 214-223; idem, Four Thousand Years of History at Tel Beth-Shean: An Account of the Renewed Excavations, BA 60 (1997), pp. 62-76; idem, The Excavations at Beth-Shean during the Years 1989-94, in N.A. Silberman and D. Small (eds.):The Archaeology of Israel (Journal for the Study of the Old Testament Supplement Series 237), Sheffield, 1997, pp. 144-164; idem, Beth-Shean, in E.M. Meyers (ed.): The Oxford Encyclopedia of Archaeology in the Near East I, New York, 1997, pp. 305-309. (back)
4. The first season took place between 22 June-1 August, 1997 and the second between 28 June-7 August, 1998. Area supervisors: Area A: R. Mullins; Area B: D. Weinblatt and D. Browning (1997) and R. Mullins and D. Browning (1998); Area C: N. Panitz-Cohen; Area D: Y. Shur and A. Sumakai-Fink; Area E: S.J. Koh; and Area F: A. Ziv-Esudri. Registration: D. Davis. Field photography: J. Camp (1997-1998) and N. Ben-David (1998). Studio photography: G. Laron and Y. Porath. Surveying and plans: M. Caplan (1997-1998), B. Arubas (1998) and S. Laemmel (1998). Restoration: 0. Mazar. Conservation: M. Lavi. Geophysical survey: D. Maki (1997). Palaeobotany: U. Baruch. Osteology: M. Craig. Computer programming: N. Appelbaum. Physical anthropology: Dr. Mark Spiegelam. Ground stones study: L. Petit.
Assistant area supervisors: 1997 and 1998: P. Cowie, E. Hasinoff, T. Koren, L. Shapira, L. Silvermann, W, Wilson, 0. Khalsa, M.Y. Im, L. Petit, T. Sogimoto; in 1997 alone: M. Burns, T. Hofecker, J. Redford, D. Zboboder; in 1998 alone: T. McCoullough, L. Cole, D. Stoutenberg, N. Yahalom. More than 100 volunteers participated in the excavation each season, including groups organised by the Jerusalem University College, The Hebrew University School of Overseas Students and William Carey College. The expedition headquarters was at Kibbutz Nir David (1997) and Kibbutz Mesillot (1998).
For a preliminary survey of the 1997 season, see A. Mazar, in S. Wolff: Archaeology in Israel, AJA 102 (1988) pp. 775-776, which requires some modifications in light of the 1998 season. (back)
5. S. Ahituv: Canaanite Toponyms in Ancient Egyptian Documents, Jerusalem, 1984, pp. 163-16S. Ahituv classified the Egyptian sources as referring to three cities of that name: 1) in Akko Valley; 2) in the northern part of Western Galilee; and 3) in Beth-Shean Valley. (back)
6. F. Hrozny, in E. Sellin: Tel Tacannek, Vienna, 1904, p. 115; W.F. Albright: A Prince of Taanach in the Fifteenth Century B.C., BASOR 94 (1944), p. 23; B. Maisler (Mazar): The Taanach Tablets, in N.H. Tortchiner et al.(eds.): Klausner Volume, Tel Aviv, 1936, p. 55 (Hebrew). (back)
7. ANET, p. 253; K.A. Kitchen: Ramesside Inscriptions I, Oxford, 1968, No. 12:10; Ahituv (above, n. 5), p. 164. (back)
8. ANET, pp. 253-254. (back)
9. Ibid., p. 479. (back)
10. J. Simons: Handbook for the Study of Egyptian Topographical Lists Relating to Western Asia, Leiden, 1937, p. 181; K.A. Kitchen: The Third Intermediate Period in Egypt, Warminster, 1973, pp. 293-300, 432-447; B. Mazar: The Early Biblical Period, Jerusalem, 1986, pp. 145-146; Ahituv (above, n. 5). (back)
11. W. Helck: Matrialien zur Wirtschaftsgeschichte des neuen ReichesI-V, Mainz, 1961-1970, p. 985; Ahituv (above, n. 5), p. 164. (back)
12. A.H. Gardiner: Late Egyptian Miscellanies, Bruxelles, 1937, No. 17:3; Ahituv (above, n. 5). (back)
13. Ahituv (above, n. 5), p. 41. (back)
14. ANET, p. 243; Y. Aharoni: The Land of the Bible. Historical Geography, Philadelphia, 1979, pp. 144-146, 162; ANET, p. 243 (where the identification is not secure; the name could refer to Rehov in Akko Valley); Ahituv (above, n. 5), p. 164, identifies this city with the northern Rehov of Asher. Kempinski suggests identifying Rehov of Asher and of several Egyptian sources (including Rehov in the Execration Texts) with the mound of Kabri. See A. Kempinski: Excavations at Kabri. Preliminary Report of 1986 Season, Tel Aviv, 1987, p. 18. (back)
15. F.M. Abel: Melange, I. Exploration de la Valée du Jordan, RB 10 (1913), pp. 218-220; WF Albright: The Jordan Valley in the Bronze Age, AASOR VI (1926) pp. 38-39; F.M. Abel: Geographie de la Palestine, II, Paris, 1938, pp. 433-434. (back)
16. M. Avi-Yonah: Gazetteer of Roman Palestine (Qedem 5), Jerusalem, 1976, p. 91; for inscription on mosaic floor from Rehov, see NEAEHL 4, pp. 1272-1274 (with bibliography); J. Naveh: On Stone and Mosaic, Jerusalem, 1978, pp. 79-85 (Hebrew). (back)
17. These numbers correct those mentioned in Wolff (above, n. 4), p. 775 (who cites 12 hectares for the entire mound and 4 hectares for the upper mound, based on preliminary measurements). (back)
18. This is a subsidiary fault of the great Syro-African fault. Geological consultation was given to us by E. Zilberman and R. Amit (Israel Geological Institute). (back)
19. Albright (above, n. 15); Zori (above, n. 1), pp. 176-178. (back)
20. O. Yogev: A Middle Bronze Age Cemetery South of Tel Rehov, cAtiqot (English Series) XVII (1985) pp. 90-113. (back)
21. Based on finds published by Zori (above, n. 1). (back)
22. The layer was examined by E. Zilberman and R. Amit (Israel Geological Institute). Our thanks are extended to A. Segev from the Institute for his help. (back)
23. S. Bunimovitz and O. Zimhoni: 'Lamp and Bowl' Foundation deposits in Canaan, IEJ 43 (1993), pp. 99-125. (back)
24. A.W. McNicoll et al.: Pella in Jordan 2 (Mediterranean Archaeology supplement 2), Sydney, 1992, pp. 97-100, Pls. 70-71. (back)
25. In Are B we found a fragment of the top corner of such a stand, which had a horn similar to that of the horned stone altars. The stand from Area E probably also had horns in its corners. (back)
26. For discussion and references, see J.S. Holladay: Religion in Israel and Judah under the Monarchy: An Explicitly Archaeological Approach, in P.D. Miller et al. (eds.): Ancient Israelite Religion. Essays in Honor of F.M. Cross, Philadelphia, 1987, pp. 249-299; W.G. Dever: Material Remains and the Cult in Ancient Israel: An Essay in Archaeological Systematics, in C.L. Meyers and M. O'Connor (eds.): The Word of the Lord Shall Go Forth. Essays in Honor of D.N. Freedman, Winona Lake, 1983, pp. 571-585. (back)
27. A. Biran: Biblical Dan, Jerusalem, 1994, pp. 244-245. (back)
28. The pottery was identified by Dr. Adrian J. Boas (The Hebrew University of Jerusalem). (back)
29. We first read this letter as b. After a lecture at the Annual Meeting of the Society of Biblical Literature in Orlando, FL (November 1998) Prof. Kyle McCarter suggested the reading h, a reading confirmed by Profs. F.M. Cross and A. Lemaire. These scholars were of the opinion that the letters correspond to the late tenth-early ninth century B.C.E., confirming the dating of the floor. (back)
30. The human remains from these graves were removed and reburied at the foot of the mound. (back)
31. Y. Alexandre: The 'Hippo' Jar and Other Storage Jars at Hurvat Rosh Zayit, Tel Aviv 22 (1995), pp. 77-88. (back)
32. W.E. Rast: Tacanach I. Studies in the Iron Age Pottery, Cambridge, MA, 1978, pp. 17-38, Figs. 18-69. (back)
33. O. Zimhoni: Clues from the Enclosure-Fills: Pre Omride Settlement at Tel Jezreel, Tel Aviv 24 (1997), pp. 93-108. The pottery from Loci 154, 214 and 484 from Jezreel is extremely important for the chronology of this period, as these loci were related by the excavators to the destruction of the royal enclosure there, probably around 841 B.C.E. (back)
34. For recent treatment, see A. Ben-Tor and D. Ben-Ami: Hazor and the Archaeology of the Tenth Century B.C.E., IEJ 48 (1998), pp. 1-38; also A. Zarzeki-Peleg: Hazor, Jokneam and Megiddo in the Tenth Century B.C.E.,Tel Aviv 24 (1997), pp. 258-288. (back)
35. Z. Gal: Rosh Zayit, Horvat, in NEAEHL 4, pp. 1289-1291. A final report by Z. Gal and Y. Alexandre (Israel Antiquities Authority Reports), currently in press, presents a very rich assemblage of pottery which is similar to ours in many respects. (back)
37. I. Finkelstein: The Archaeology of the United Monarchy: An Alternative View, Levant 28 (1996), pp. 177-187. (back)
38. A. Mazar: Iron Age Chronology: A Reply to I. Finkelstein, Levant 29 (1997), pp. 157-167. For Finkelstein's reply to this response, see I. Finkelstein: Bible Archaeology or Archaeology of Palestine in the Iron Age? A Rejoinder, Levant 30 (1998), pp. 167-173. In this rejoinder, Finkelstein preaches against anchoring archaeological phenomena in Biblical data, yet he does exactly the same by utilising the destruction of Jezreel as a datum point for his chronology! (back)
39. Dates were supplied by Prof. D. Donahue (Department of Physics and Atmospheric Sciences, The University of Arizona). In a letter from 8 October, 1998, Prof. Donahue writes: ‘The error quoted is the standard deviation of the average of the nine measurements. In this instance, the error resulting from the scatter of the nine measurements was equal to the uncertainty from statistics. This agreement indicates that the final result is a very good one. In fact, it is the best that we have done in our laboratory'. (back)
40. N. Na'aman: Shishak's Raid to the Land of Israel in Light of the Egyptian Inscriptions, the Bible and the Archaeological Data, Zion 63 (1998), pp. 247-276 (Hebrew). (back)
41. Finkelstein (above, n. 38,1998), p. 170, brings three C14 dates from Megiddo Stratum VIA which point to a date in the eleventh century B.C.E. for this stratum (although one of them slides into the tenth century as well). Since they come from charcoal which perhaps originated from timber, they can be considered as providing a terminus post quem, as, indeed, Finkelstein claims. Yet the fact that all three almost agree in date strengthens, in my view, the dating of Stratum VIA to the eleventh century B.C.E. (although its end could be as low as c. 980 B.C.E., see below). See, for example, A. Mazar: The Archaeology of the Land of the Bible (ca. 10000-586 BCE), New York, 1990, p. 301. (back)
Figure 21. Area A: detail of human skeleton found in ruined dwelling.
Figure 22. Area B: general view of fortification at end of 1998 season (looking east).