Adams Et Al. Ch. 2 Area J 2004-2010 Megiddo V 2013 FULL-Libre

MEGIDDO V THE 2004-2008 SEASONS
VOLUME I
TEL AVIV UNIVERSITY SONIA AND MARCO NADLER INSTITUTE OF ARCHAEOLOGY
MONOGRAPH SERIES NUMBER 31
Executive Editor Editorial Board
Managing Editor
Israel Finkelstein Moshe Fischer Avi Gopher Raphael Greenberg Oded Lipschits Myrna Pollak
MEGIDDO V
THE 2004-2008 SEASONS
VOLUME I
EDITORS: ISRAEL FINKELSTEIN, DAVID USSISHKIN, ERIC H. CLINE MATTHEW J. ADAMS, ERAN ARIE, ERIC H. CLINE, ISRAEL FINKELSTEIN, NORMA FRANKLIN, MARIO A.S. MARTIN, DAVID USSISHKIN
Contributions by
Matthew J. Adams, Yaniv Agmon, Eran Arie, Carolina Aznar, David Ben-Shlomo, Julye Bidmead, Noga Blockman, Elisabetta Boaretto, James M. Bos, Baruch Brandl, Eric H. Cline, Margaret E. Cohen, Adi Eliyahu-Behar, Julie Ellis, Lev Eppelbaum, Israel Finkelstein, Norma Franklin, David Friesem, Yuval Gadot, Mor Gafri, Boaz Gattenio, Ayelet Gilboa, Philippe Guillaume, Christian Herrmann, Sonia Itkis, Othmar Keel, Adi Keinan, Inbar Ktalav, Nili Liphschitz, Shmuel Marco, Mario A.S. Martin, Assaf Nativ, Alexander Pechuro, Rachel Pelta, Laura A. Peri, Daniel Rosenberg, Galit Sameora, Inbal Samet, Benjamin Sass, Aharon Sasson, Ruth Shahack-Gross, Ilan Sharon, David Ussishkin, Lior Weissbrod, Naama Yahalom-Mack, Assaf Yasur-Landau
Project coordinator
Sivan Einhorn
EMERY AND CLAIRE YASS PUBLICATIONS IN ARCHAEOLOGY TEL AVIV UNIVERSITY
WINONA LAKE, INDIANA EISENBRAUNS 2013
Monograph Series under the auspices of the Friends of the Institute of Archaeology of Tel Aviv University
Graphics by Michal Semo-Kovetz, TAU Graphic Design Studio
www.eisenbrauns.com Printed in the U.S.A. Copyright 2013 by the Institute of Archaeology, Tel Aviv University All rights reserved.
Published for the Emery and Claire Yass Publications in Archaeology
(Bequeathed by the Yass Estate, Sydney, Australia)
of the Institute of Archaeology, Tel Aviv University

by Eisenbrauns Winona Lake, Indiana, U.S.A.
Library of Congress Cataloging-in-Publication Data

Adams, Matthew J. (Matthew Joel) Megiddo V : the 20042008 seasons / Matthew J. Adams, Eran Arie, Eric H. Cline, Israel Finkelstein, Norma Franklin, Mario A. S. Martin, David Ussishkin ; editors, Israel Finkelstein, David Ussishkin, Eric H. Cline ; contributions by Matthew J. Adams, Yaniv Agmon, Eran Arie, Carolina Aznar, David Ben-Shlomo, Julye Bidmead, Noga Blockman, Elisabetta Boaretto, James M. Bos, Baruch Brandl, Eric H. Cline, Margaret E. Cohen, Adi Eliyahu-Behar, Julie Ellis, Lev Eppelbaum, Israel Finkelstein, Norma Franklin, David Friesem, Yuval Gadot, Mor Gafri, Boaz Gattenio, Ayelet Gilboa, Philippe Guillaume, Christian Herrmann, Sonia Itkis, Othmar Keel, Adi Keinan, Inbar Ktalav, Nili Liphschitz, Shmuel Marco, Mario A. S. Martin, Assaf Nativ, Alexander Pechuro, Rachel Pelta, Laura A. Peri, Daniel Rosenberg, Galit Sameora, Inbal Samet, Benjamin Sass, Aharon Sasson, Ruth Shahack-Gross, Ilan Sharon, David Ussishkin, Lior Weissbrod, Naama Yahalom-Mack, Assaf Yasur-Landau ; project coordinator, Sivan Einhorn. volume cm. (Monograph series (Tel Aviv University. Sonia and Marco Nadler Institute of Archaeology) ; number 31) Published . . . for the Emery and Claire Yass Publications in ArchaeologyTitle page verso. Includes bibliographical references and index. ISBN 978-1-57506-273-0 (volume 1 : hardback : alkaline paper) ISBN 978-1-57506-274-7 (volume 2 : hardback : alkaline paper) ISBN 978-1-57506-275-4 (volume 3 : hardback : alkaline paper) ISBN 978-1-57506-276-1 (set, 3 volumes : hardback : alkaline paper) 1.Megiddo (Extinct city)2.Excavations (Archaeology)IsraelMegiddo (Extinct city)3.Bronze ageIsraelMegiddo (Extinct city)4.Iron ageIsraelMegiddo (Extinct city)5.Israel Antiquities.I.Finkelstein, Israel.II.Einhorn, Sivan.III.Emery and Claire Yass Publications in Archaeology.IV.Title.V.Title: Megiddo5.VI. Title: Megiddo Five. DS110.M4A33 2013 933.46dc23 2013011695 The paper used in this publication meets the minimum requirements of the American National Standard for Information SciencesPermanence of Paper for Printed Library Materials, ANSI Z39.48-1984.
CHAPTER 2
AREA J PART I: INTRODUCTION

Matthew J. Adams
GOALS AND PROGRESS OF WORK

Excavation in Area J from 2004 to 2010 was supervised by Matthew J. Adams.1 The large excavation area was, in fact, divided into three sub-areas (Fig. 2.1; Table 2.1). The work was carried out with the help of assistant area supervisors Ruslan Shvartsman (2004), Hai Ashkenazi (2006), James M. Bos, Adi Keinan and Julie Ellis (2008, Sub-Area Upper J, Sub-Area Lower J and registration, respectively), and Katia Charbit Nataf and Julie Ellis (2010). In the main part of Area J work continued in the Level J-4 temple sanctuary, that is, in the squares north of Wall 96/J/07 (Fig. 2.24). Upper Area J (20062010) featured two squares located at the northwest edge (and outside) of the University of Chicagos Area BB. In Lower Area J (2008) excavation was initiated in three squares on the slope, east of Temple 4050, in order to answer questions regarding the Picture Pavement (University of Chicago Stratum XIX). This report includes a description of the inds in these areas and a stratigraphic reassessment of some previous seasons work in all of Area J in light of the newly excavated material. The presentation of the results is composed of three sub-chapters devoted to the sub-areas of Area J. The 2010 season proved to be the inal season for Area J. Therefore those results have been included herein. The Early Bronze Age remains have been added as Appendix to Part III: The Main Sector of Area J and Part IV: Sub-Area Upper J has been completely revised to include the 2010 results there. Several goals dictated the progress of our work in the 20042008 seasons. 1. The exposure of the Level J-4 temple. In the 1998 and 2000 seasons the interior of the sanctuary was found to be on the north side of Area J (Finkelstein and Ussishkin 2000a). Because of the boundaries of the University of Chicagos deep cut (Area BB), the area available for excavation of this massive temple-hall was limited to twelve squares (or part thereof). With a hypothesis regarding the size and shape of the Level J-4 temple in hand (Finkelstein and Ussishkin 2003), excavation began in 2004 in six squares (G/7, G/8, H/7, H/8, J/7, J/8) in order to expose the plan of the temple west of the buildings altar, which was exposed in the 1998 season. In 2006, work continued in these units and also expanded to include Square F/7 as well as the following baulks: H/7-H/8, H/8-H/9, G/7-G/8, G/7-F/7, G/8-F/8 and F/7-F/8. In 2008, some of these units continued to expose the plan of the temple, but primarily a number of key-hole incisions were made to answer speciic architectural questions related to the ediice.
1 The author would like to express his gratitude to the W.F. Albright Institute of Archaeological Research, the Educational and Cultural Affairs division of the U.S. Department of State and the Dr. M. Aylwin Cotton Foundation for support during the preparation of this report.
21
M ATTHEW J. A DAMS
Fig. 2.1: Plan of areas excavated, squares backilled, locations of dumps, and locations where dump was used to stabilize existing monuments and facilitate drainage (20042008).
2. The exposure of more information on Levels J-5 and J-6. The seasons reported here have more than doubled the exposure of these levels, which date to the EB III, providing a much clearer picture of their stratiication and layout. 3. The chronological and stratigraphic positions of the three megara-style temples of Level J-7 (the University of Chicago Stratum XV) have been under some debate. Fortunately, there remained three opportunities to shed light on this problem:
22
AREA J, PART I: INTODUCTION
CHAPTER 2: A REA J, PART I: I NTODUCTION
Fig. 2.2: Aerial view of Area J, end of the 2008 season, looking east.
a. exposure of Level J-4 required the removal of the western porch pillar-base of Temple 4040. This base had extensive stone foundations and we were interested in the ceramic material sealed beneath this stone. b. the threshold of Temple 4040 had never been excavated. In the 2008 season we removed the temple loor (still preserved, though exposed by Chicago) and excavated beneath it. c. the University of Chicago team left the northwestern corner of the westernmost temple (5269) unexcavated. Here in 2006, two new squares were opened: Squares S/20 and T/20.2 4. We were interested in tying together the stratigraphy of Area J with that of other areas on the site, thus providing the renewed excavations irst continuous stratigraphic picture of the tell from the Early Bronze Age through the Late Bronze Age. The revised and augmented stratiication of Area J presented below now covers that sequence (Table 2.2). 5. Adi Keinan (2007) raised some stratigraphic and contextual questions regarding the Stratum XIX Picture Pavement. In the 2008 season we opened several units in Lower Area J to re-expose the pavement and
2 The designations for these squares do not match the Area J local grid. Rather, their designations were chosen at the extreme end of the grid lettering so that they will not be confused with units in Area J proper. Their orientation, while similar to that of the regular Area J grid, was chosen based on local topography (see below).
23
M ATTHEW J. A DAMS
excavate a portion of the University of Chicagos northern baulk of the deep trench in Area BB in order to shed light on some of these problems. To these ends, excavation in 2004 was concentrated in six squares (G/7J/8) exposing primarily Level J-7 and Level J-5/J-6 architecture (Table 2.1). Work continued in these squares in 2006 and also resumed the 2000 excavation in Squares F/7 and F/8. The 2006 season also saw the initiation of work in Sub-Area Upper J, Squares S/20 and T/20. In 2008, a number of small incisions were made in order to answer key questions about the plan of the Level J-4 temple. In Square J/7 a 1.5 m incision was made into the western baulk, extending all the way to tomb Structure 5239 in order to locate the sanctuarys western closing wall. A 1.5 3 m incision removed a portion of the F/7-E/7 baulk and Square E/7 in order to expose the entrance through the northern sanctuary wall. A 1.5 2 m incision was made in the south baulk of Square E/9 down through the threshold of Temple 4040 to test for the southern wall of the sanctuary. The E/8-E/9 baulk was removed to better expose one of the basalt tables in the corner of Square E/9. The F/8-F/9 baulk was removed with Temple 4040s western porch pillar base to expose the central axis of the Level J-4 temple. Work also continued in Sub-Area Upper J and new excavation was initiated in Sub-Area Lower J (Squares 1 and 2 and Incision A) to re-expose the University of Chicagos Picture Pavement (Level J-2). TABLE 2.1: SUB-AREAS AND SQUARES EXCAVATED, 2004-2008
Season 2004 2006 2008 1; 2; Incision A Lower J Main J G/7; G/8; H/7; H/8; J/7; J/8 F/7; G/7; G/8; H/7; H/8; J/7; J/8 E/7; E/9; F/7; F/8; F/9; G/7; H/9; J/7 S/20; T/20 S/20; T/20; S/21 Upper J
STRATIGRAPHIC SUMMARY
To date, we have identiied 19 levels in Area J, from Level J-1 (earliest) through Level J-19 (latest), covering the Early EB I through the Iron Age. Table 2.2 summarizes these levels with a brief description and the corresponding Oriental Institute strata. Note that unlike the other areas under investigation by the renewed excavations, Area Js stratigraphic numbering system begins from the earliest occupation (see Finkelstein and Ussishkin 2000a: 34).
24
TABLE 2.2: REVISED AND AUGMENTED STRATIGRAPHY OF AREA J*

Level/ Phase J-1 J-2 J-3 J-4 J-4a J-5 J-6b J-6a J-7 J-8 J-9 U of Chicago Sub-area** stratum -XX, XX XIX XIX XVIII Not detected XVII XVII XVI XV XIVB XIVA LJ LJ LJ MJ MJ MJ MJ MJ MJ UJ UJ MJ J-10 J-11 XIIIA/B XII UJ MJ UJ MJ J-12 J-13 J-14 J-15 J-16 J-17 J-18 J-19 XI X IX (B) IX (B) IX (A?) IX/VIII (?) VIII-VIB (?) V? UJ UJ UJ UJ UJ MJ UJ UJ Description Carved bedrock and associated structures Temple beneath 4050 and Picture Pavement Temple 4050 with curved Wall 08/J/21 Great Temple, terrace Walls 4045 and 4114, Gate 4113 Sporadic activity within the temple. No (or reduced) settlement Palace Compound and Palace 3177 Original Palace Compound and Palace 3177 Original Palace Compound and Palace 3177 Rebuilt Temples 4040, 5192 and 5269 House with burials House rebuilt with chalk-paste coated basins and burials Stone-lined Silo 06/J/140, Temple 4040 reuse Fill 08/J/157, Tombs 06/J/107 and 04/J/056 Walls 08/J/13 and 08/J/05 Stone-lined Pit 04/J/48 Walls 08/J/08 and 08/J/19 Walls 06/J/16 and 06/J/18 Wall 08/J/14 Pebble Floor 08/J/30 and lime Pavement 08/J/54 Upper lime pavement Tomb structure 5239 Deep Wall 08/J/01 Wall 08/J/02 MB II MB III-LBI (?) LB I LB I LB I (?) LB II (?) LB IA (?) IA II MB I-II MB II Period EB I EB IB EB IB EB IB EB IBII EB III EB III EB III EB III-IBA MB I MB I
* The stratigraphic attributions herein supersede those offered in Megiddo III and Megiddo IV and other earlier reports. ** Key to Abbreviations: MJ = Main J; UJ = Upper J; LJ = Lower J.
MEGIDDO III AND MEGIDDO IV CORRIGENDA

Megiddo III: Fig. 24.4 This photograph is of the southeastern corner, not western. Megiddo IV: 38 The earlier loor of the Level J-4 temple was erroneously reported as being thinly plastered (Finkelstein, Ussishkin and Peersmann 2006: 38). The report of a plastered loor was a conlation of the
25
M ATTHEW J. A DAMS
thin compact ash from Square F/9 interpreted as a loor in 1998 and the discovery of this surface in 2000. Note also that our interpretation of what was a loor here and what was not has changed. Megiddo IV: 165 "and four jars (Fig. 10.1: 25)" should read "and four jars (Fig. 10.9: 25)". A few plans in Megiddo III showed a mislettered grid: Megiddo III: Fig. 14.1 Grid letters should shift one square to the west; grid number should shift one square to the south. Megiddo III: Fig. 14.2 Grid letters should shift one square to the west; grid number should shift one square to the south. Megiddo III: Fig. 14.5 Grid letters should shift one square to the west; grid number should shift one square to the south.
NOTES ON THE NORTH ARROWS IN MEGIDDO III AND MEGIDDO IV

1. The Megiddo Expedition utilizes a combination of a site-wide grid and local grids depending on the strategy of excavation at a particular location. The site-wide grid is a subdivision of the University of Chicagos 25 m by 25 m grid (see Megiddo III: Fig. 1.2), which is aligned to the north as determined by the Chicago expedition. This grid has been used in the Area J excavation of Temple 4050 (Stratum XIX; Levels J-2 and J-3), Area M, Area F, the original grid of Area H and Area G. 2. The north arrow shown on the plans for Levels J-2 and J-3 in Megiddo III (Figs. 3.103.11) is in line with the orientation of the grid used for the excavation of the Levels J-2 and J-3 temples, i.e., the main site-wide grid. The north arrow here, therefore, is correct vis--vis the University of Chicago plans. 3. The local grid of the main sector of Area J is aligned with the orientation of Temple 4040. The temples orientation is of 20 east of University of Chicago north. Thus, the north arrow on plans of the main Area J grid should point 20 differently than the main Area J grid. 4. The north arrow on most plans in Megiddo III and Megiddo IV is, in fact, pointing 30. Any attempt to correlate these plans with the University of Chicago plans will result in misaligned architecture. 5. These north arrows should be rotated 10 clockwise to match both the University of Chicago grid and the primary Megiddo Expedition site-wide grid. 6. The following igures are affected by this discrepancy: Megiddo III: Figs. 3.26, 3.43, 3.46; Megiddo IV: Figs. 3.4, 3.7, 3.10, 3.14, 3.16a, 3.20, 3.22, 3.243.25, 3.28. 7. This problem has been corrected on plans.
CONSERVATION, DUMPS AND BACKFILLING

As the preservation of the existing monuments is one of the primary tenets of the Megiddo Expedition (Finkelstein, Ussishkin and Halpern 2000: 10), a number of squares in and around Temples 4040 and 5192 were backilled to the temples loor level. Additionally, it was found that rain water was beginning to undercut the foundations of these temples, as well as Altar 4017. These walls were shored up with excavation dump where needed. An effort was made to direct runoff water away from these structures. Figure 2.1 shows the location of all backilling and dumping operations since 2004. In the 2008 season, it was decided that the western porch Pillar Base 08/J/214 of Temple 4040 (Loud 1948: Figs. 179, 186) needed to be removed in order to properly investigate the Level J-4 altar and its
26
spatial relationship to the rest of the sanctuary. Our concern in this venture was the preservation of the pillar base for future conservation, reconstruction and display. Our plan was to remove the stone and place it temporarily on top of the eastern base until such a time as a restoration scheme could be established. To do this, we enlisted the help of Yehoshua Dray, a restoration expert who specializes in moving and restoring heavy stone (Fig. 2.3).3 The operation was successful. The University of Chicago Expedition excavated well beneath the loor level of Stratum XV Temple 5269 (though no loor was found), but left in place the temples pillar base foundations with the bases precariously balanced on top (Loud 1948: Fig. 185). Over the last several decades, these bases have fallen from their perch and the foundations themselves accumulated a number of foreign stones. At the end of the 2008 season we cleaned away the brush and stones that were not part of the original foundations and replaced the two pillar bases in their original positions.
Fig. 2.3: Removal of Temple 4040s western portico pillar base (Phase J-7b; Stratum XV), facing northeast.
http://www.yeshuat.com
27
AREA J, PART II: SUB-AREA LOWER J PART II: SUB-AREA LOWER J

Adi Keinan
The paved courtyard of the cultic complex of Levels J-2 and J-3 was irst excavated by the University of Chicago in the late 1930s (Fig. 2.4). The pavement was preserved in three segments: Locus 4118 in the central part of the complex, Locus 4064 in the southern part, and Locus 4008 in the northeast. Locus 4008 (Pavement 08/J/200 in the renewed excavations) the most elaborate pavement segment included 44 incised stone slabs depicting human and animal igures and signs and symbols. It was thus named the Picture Pavement. Following the renewed excavations in Area J, the pavement and the curving wall, supposedly enclosing the courtyard, were attributed to Level J-2 (Finkelstein and Ussishkin 2000a: 52). Pavement 4008 had two phases (Loud 1948: 61), which were termed upper pavement, associated with the curving stone wall, and lower pavement, attributed to an earlier phase. The upper pavement was removed by Loud, revealing a stepped-like pavement similar to a staircase. The removal of this upper pavement was recorded in Louds report with only one photograph (Fig. 2.5). At the end of the University of Chicago excavation, a high baulk was left along the northern edge of the excavated trench (Fig. 2.6). During the 70 years that passed since Louds excavation, the pavements condition deteriorated. About half of it was covered with earth and stone debris collapsed from the section created along the northern edge of Area BB. Other parts of the exposed pavement, especially those on its edges, were washed away. The impetus for the renewed excavations in Lower Area J was my study of the Picture Pavement (Keinan 2007), in which the relation between the various scenes incised on the slabs, the temple complex and the people who visited the temple were reexamined. Based on stylistic and conceptual considerations I attributed the pavements incisions to the domain of Egyptian art. Questions regarding the function of the pavement within the earliest EB IB cultic complex and the signiicance of the engravings incised on many of its slabs arose from this study. Thus, the goals of the 2008 season in Lower Area J were: 1. To re-expose the Picture Pavement, record its remains to the maximal extent, check if some of the incised slabs are still in situ and look for new or unnoticed incisions; 2. To reassess the relationship between the pavement and the University of Chicagos curving wall (Loud 1948: 61, Fig. 390); 3. To check whether the lower pavement extends farther to the north.
PROGRESS OF WORK
The irst part of the excavation was dedicated to the removal of the collapsed debris from the northern edge of Area BB upon the pavement.1 This was done in two adjoining squares, labelled Squares 1 and 2 (Loci 08/J/010 and 08/J/041 accordingly; Figs. 2.711).2 In both squares the collapse layer sloped steeply from west to east and from north to south.
1 2
The dump zone was set at about 25 m east of the excavated area, at the eastern edge of the tell (see Fig. 2.1). Lower Area J used a local grid, speciically laid for the excavation of the Picture Pavement, referred to here as Square 1, Square 2 and Incision A.
28
CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J
Fig. 2.4: Plan of the Stratum XIX Temple (Loud 1948: Fig. 390).
29
A DI K EINAN
Fig. 2.5: The stepped-like pavement. Note the curving wall at the bottom right and the northern baulk at the bottom, looking south (Loud 1948: Fig. 146).
Fig. 2.6: Picture Pavement 4008; photograph taken during excavation, looking northwest (courtesy of the Oriental Institute, the University of Chicago).
30
Fig. 2.7: Lower Area J, showing Square 1 on the right, Square 2 on the left and Incision A, looking north.
Square 1 was located directly to the north of the exposed pavement segment. The original eastern edge of the pavement was traced in the western part of the square; the eastern part of the square had no inds at the elevation of the pavement. By comparing the slabs that were found with the University of Chicagos original excavation photos, we determined that the original northern baulk of Area BB was located just a few cm to the north of Square 1s northern section. In Square 2 the pavement was revealed in almost the entire area. In its northwestern part some roughly rectangular slabs were found. This part of the pavement, which seemed to have been laid as a staircase, was the segment discovered under the upper pavement removed by Loud. He also removed some of these lower slabs probably an indication that they were incised. This slab removal was not recorded by the University of Chicago team; therefore, if these slabs were incised, the depicted scenes remain unknown. After excavating these two squares, the southern half of the baulk between them was removed, revealing additional pavement slabs. In both squares the incised slabs had been removed during the University of Chicago excavation, hence the negatives of the removed slabs could be traced (Fig. 2.8). Most of those negatives match known slabs that were removed and published (see Appendix to Part II), but in some cases, unexpected negatives of unknown slabs were found. Some slabs were found broken, probably due to the destruction caused by the collapsing stones of the northern baulk. Twelve slabs included some incisions and decoration patterns not known thus far; one of them is of particular interest. The second part of the excavation in Sub-Area Lower J comprised a 3 2 m Incision (A) into the line of the University of Chicagos baulk at the northern edge of Area BB. It was cut at the western half of Square 2s northern baulk.
31
A DI K EINAN
Fig. 2.8: Pavement 4008 after exposure and removal of slabs and before the removal of the upper pavement, looking south (courtesy of the Oriental Institute, the University of Chicago).
While removing the collapsed debris in this incision, we found part of the University of Chicagos curving wall (Wall 08/J/21). This is the southeastern edge of the segment that had been uncovered by Loud (see Fig. 2.5, bottom right). The contour of the original northern baulk left by the Chicago Expedition was detected just to the north of the wall. Continuing the excavation in this incision revealed Pavement 08/J/200 part of the lower pavement unearthed by Loud. The wall was built later than the pavement, as the pavement slabs continued under it, at least a few cm northwards (Fig. 2.9), and as an earth layer of a few cm separated the pavement from the wall. This was also the case with Pavement 4064 and the wall on top of it, which were re-examined in 1994 (Finkelstein and Ussishkin 2000a: 52, where the pavement was assigned to Level J-2 and the later wall to Level J-3). After reaching the pavements level in the 2 2 m test incision, the latter was extended another metre to the north in order to excavate a larger portion of the original University of Chicagos baulk and achieve a better picture of the stratigraphy (Fig. 2.10).
THE STRATIGRAPHY
LEVEL J-2 The earliest remains in the incision included a number of stones that appeared to be laid as a wall (08/J/30, Fig. 2.12). This possible earlier wall was ive rows wide (1.24 m) and had one course of stones. A collapsed layer of medium-sized stones (08/J/176) to its east may have originated from this wall. A layer of dark
32
Fig. 2.9: Stepped-like Pavement 08/J/200 and Wall 08/J/21 (on the left), looking north.
Fig. 2.10: Northern section of Incision A.
brown compact soil (08/J/161) was found above Wall 08/J/30 and its collapse layer. Level J-2 Pavement 08/J/200 did not extend northwards in this area. The revealed edge of Pavement 08/J/200 in this square is exactly the edge that was uncovered by the University of Chicagos excavation. Compact mudbrick material (08/J/215) unearthed under one of the pavements slabs (see Slab D below) is part of the make-up laid prior to the construction of the courtyards pavement. LEVEL J-3 This level included Wall 08/J/21 the University of Chicagos enclosure wall of Stratum XIX and some collapsed debris to its east. The exposed part of Wall 08/J/21 was actually its original eastern edge. It had one course on its northern part and two courses on the southern part. This part of the wall is wider than was previously thought. Loud did not uncover this wall to its entire width in this area, hence the University of Chicagos Stratum XIX plan depicted it as getting narrower as it goes eastwards (see Fig. 2.4). It is now clear that the eastern edge of Wall 08/J/21 has exactly the same width as its western part that was unearthed by the University of Chicago team. Therefore, Wall 08/J/21 can be reconstructed as 1.5-m-wide throughout its length (Fig. 2.13). The relationship between this wall and the upper pavement remains unclear, since the latter had been fully removed by Loud. It is assumed that they were contemporaneous (Finkelstein and Ussishkin 2000a: 52). The remains of Wall 08/J/21 could have served as a stone foundation for a mudbrick superstructure, or more likely, there could have been additional stone courses, as hinted by some large and medium-sized
33
A DI K EINAN
Fig. 2.11: Plan of Lower Area J. The incised slabs are marked by the letters A to I.
stones found to its east, which may have collapsed from it. The latest collapse layer of this level is Locus 08/J/140, which included some plaster patches and a mud surface to the east. The plaster may have originated from Wall 08/J/21. The lighter brown surface may be interpreted as brick collapse. LEVEL J-4(?) The elements described below, all found above Wall 08/J/21, seem to belong to Level J-4, though no physical connection with the monumental construction of this level farther west could be established. This level consisted of a series of light brown/pink clayish surfaces (08/J/134) laid over mudbrick material (08/J/134). Laid in a shallow pit beneath the clay surfaces were the fully articulated remains of a fetus/infant of either a sheep or a goat (08/J/135). Above the clayish surface was a thin layer of phytolith and an occupational layer above it a horizontal ashy accumulation (08/J/116) containing a large quantity of charcoal.
34
Fig. 2.12: Wall 08/J/21 and Wall 08/J/30 just below it on the left, looking north.
Fig. 2.13: Suggested reconstruction of Wall 08/J/21 (in black), after Loud 1948: Fig. 390 (in grey).
35
A DI K EINAN
INCISED SLABS
Some incisions and decoration patterns that have not been observed (or at least not published) thus far were found on 12 of the exposed slabs of Pavement 08/J/200 (see Figs. 2.11, 2.14ab; Table 2.3). TABLE 2.3. INCISED SLABS FOUND DURING THE 2008 SEASON
Slab A B C D E F G H I J K L Square 1 1 and to its south 2 2 2 2 2 Incision A 2 2 2 2 Measurements (cm) 20 19 46 28 55 20 43 25 32 22 40 32 63 21 48 25 42 32 28 16 38 21 46 33 Elevation (m) 150.94 150.78 151.55 151.60 151.91 151.86 151.98 152.00 151.68 151.71 151.59 151.84
Slab A was not fully exposed since it continued into the squares western baulk. It has two parallel lines incised at its bottom right, and a circular depression that may or may not be natural to its left. Slab B has two parallel lines at its left part. A natural crack is located to its right. Slab C has three sets of two parallel short lines: one on the left, one in the centre and one on the right part of the slab. The two lines on the right are only roughly parallel, the bottom line being much shorter than the top line. This slab was also part of the pavement below the upper pavement. Slab D includes the most elaborate and signiicant incisions (Fig. 2.15). It was part of the stepped-like pavement found by Loud below the upper pavement. The main element, incised at its centre, is a human male igure holding a spear. The igure is facing right, holding the spear in his left hand. Three circles in relief represent the eyes and mouth. In addition, the igure has a pointed beard. His breast is emphasized by two circles in relief. He is wearing a short skirt and a belt with a dagger; the belt has two bulges at its centre, and the skirt has three bulges on its right side, all in relief. The right hand was destroyed by what seems to be an intentional strike made at a later stage. Additional blows include a semicircle of strikes on the left half of the slab, a diagonal line of strikes on the right half of the slab and a few strikes on the top edge of the slab. Another incision on the slab seems to depict a bird above a net pattern, on the right half of the slab. A few incisions around it and around the human igure remain unidentiied. Four groups of parallel short lines are incised on the right edge of the slab.
36
Slab E constituted part of the stepped-like pavement. The two hind legs of an animal (note the emphasized calcaneus), maybe depicting a bull, are incised at the bottom left. Parts of the animals belly and tail are also preserved. The top right area of the slab includes incisions that seem to be part of a larger, worn depiction: one is a soft wavy line that forms a pointed hat-like shape. There are additional incisions on its top and bottom, including four short parallel lines. Slab F was also part of the stepped-like pavement segment found under the upper pavement. It has nine incised lines at its southeastern corner. They are arranged in two groups, one of ive parallel lines (top) and the other of four parallel lines (bottom). In the irst group the lines are longer than those in the second group, except for the top-most line, which is very short and may have originally been longer. Slab G was also part of the lower pavement. The large crack at its centre is natural. It has many incised lines on its surface. Four lines are incised above the large crack: two of them are horizontal, and the other two lines, incised closer to the left horizontal line, are diagonal. Two parallel lines with four short parallel lines between them are also incised above the crack, on the right. Additional lines are incised below the large crack: six short parallel lines close to the slabs bottom left edge, and another line to their top right; another line in the centre; two horizontal parallel lines on the right side of the slab, with a net pattern between them. Slab H was also part of the step-like pavement segment found under the upper pavement. It is located one slab away from Wall 08/J/21. This slab has many incised lines, arranged in three main groups. The irst group at the top part of the slab has three long horizontal parallel lines, and another line, not as straight as the others and not exactly parallel to them. The second group is at the bottom of the slab; it has one long horizontal line, and at least three other short lines, parallel to it. Another short line is incised to the right of these lines. The third group of lines is located on the right part of the slab. It is a design of three horizontal parallel lines, and a vertical line that intersects them in the middle. Slab I was already exposed prior to the excavation in this square. It is located at the southern edge of the preserved pavement. It has a circle of at least 19 small depressions, probably made by blows of a stone. Those depressions look like small cup-marks. The four depressions in the eastern part of the slab are deeper and clearer than the other depressions. A few other lines are incised on the surface of this slab; some of them may be natural cracks: three parallel lines are incised at the centre; another line is incised on the western part of the slab; another line is incised on its eastern part. Slab J was part of the step-like pavement segment found under the upper pavement. It includes a few incisions to the right: two parallel lines, ending in one short vertical line, and another line branching twice to additional short lines to their bottom. Slab K has numerous incisions, especially of straight lines and geometrical shapes. In the upper left are four straight lines extending from a single point. A quadrangle and two parallel lines are incised below them. A human-like shape (stick igure) appears at the centre of the slab, with a few incised lines extending from its top, bottom and sides. A rectangle with two lines incised in it can be seen nearby. A triangle with a line dividing it and two parallel lines extending from it are in the top right. A worn portion of a larger incision is located at the bottom: a triangle-like shape is incised next to two curving lines. There are additional incisions inside the shape: a line parallel to its top side, and two short parallel lines on the left. Many other lines are incised on the slabs surface.
37
A DI K EINAN
Fig. 2.14: Incised slabs.
38
39
A DI K EINAN
Fig. 2.15: Photograph and drawing (reconstructed) of Slab D.
40
Fig. 2.16: Human igures incised on the Picture Pavement slabs (Loud 1948: Pl. 273: 56).
Slab L was part of the step-like pavement segment found under the upper pavement. At its centre are two curving lines ending at a single point, with eight short parallel lines between them; one of the short lines continues to curve upwards. The incision as a whole reminds one of a depiction of a scorpion, with its narrowing body and two semicircular claws.
DISCUSSION
The excavation of Incision A showed that the Level J-2 Picture Pavement 08/J/200 did not extend farther to the north. Most of the incisions on the slabs of this pavement seem random and are very dificult to interpret. One motif (the pattern of two parallel lines) seems to repeat itself on at least nine of our 12 slabs. This motif was also incised on some of the slabs discovered by Loud (1948: Pl. 281: 3133). The motif of the net pattern, appearing on two slabs (D and G), is also familiar from the previously found slabs (ibid.: Pls. 271: 1, 272: 4, 280: 2629, 282: 34) and so is the motif of the short parallel lines between two long parallel lines (Slabs G and L; ibid.: Pl. 282: 36). The incised decoration on the edges of Slabs D, F and G (a group of four to six short parallel lines) is similar to a pattern on one of the slabs discovered by Loud ive parallel lines on the edge of a slab (Loud 1948: Pl. 277: 16). Slab H has an incision identical to an Egyptian potters mark (van den Brink 1992: 283, group XLI; example on p. 290, Fig. 11, Group XV No.
41
A DI K EINAN
19-13). Another afinity with contemporary Egyptian motifs may be found in the scorpion incised on Slab L (assuming that the identiication is correct).3 The animal incision on Slab E, represented by its hind legs, is a very typical depiction on the Picture Pavement; characteristic is the emphasized calcaneus on the back of the left leg (e.g., Loud 1948: Pls. 275: 1011, 277: 1416). Slab D deserves special attention. Similar to other human igures incised on slabs from Megiddo, the igure incised on this slab is one of the very few examples of human depictions in EB I in the southern Levant. Its general form and motifs are typical of the Megiddo incisions: the bodys position upper part en fae, head and legs in proile, the beard and the belt, are common to almost all human igures incised on the pavements slabs. The spear is similar to another spear incised on a different slab (see Fig. 2.16: 7, right): a triangle-shaped spearhead placed on a long vertical staff. The igure on Slab D shows clear resemblance to the harpist depicted on another slab (see Fig. 2.16, 5, left); both share the same attributes: the eyes and mouth, the prominent breast, the belt and the dagger. However, whereas the face and breast of the harpist are incised, those of the igure on Slab D are raised in relief. Another interesting incision is a possible bird depiction, facing right, on the right part of the slab. Schematically incised, the bird is depicted only by the outline contour of its body, which is not complete at the tail area. The bird depiction lacks legs; instead there is a net pattern incised just below it. If the bird identiication is correct, it is possible that this bird represents the Egyptian Horus falcon, symbolizing the contemporary pre-dynastic Egyptian king and attesting to his connection to Megiddo. The net pattern, the blows around the slab and on its surface and the groups of short parallel lines incised on the right edge of the slab are all common motifs on the Picture Pavements slabs. Their meaning remains vague they could either convey magical symbolism, or they could be mere decoration on the slab. It is interesting to note that as most of the other incised slabs found at Megiddo by the University of Chicago expedition (Keinan 2007: 3436), the scene incised on this slab is also oriented to the east. It is most probable that the scenes were deliberately oriented towards the entrance to the cultic complex, and were meant to be seen by people ascending towards the temple. Slab I was decorated with a circular pattern of depressions. This slab has at least three equivalents in Area J: 1. According to Ussishkin, this decoration is very similar to the spiral decoration on a slab uncovered in Locus 4064 (Finkelstein and Ussishkin 2000a: 52). We tried to locate this slab, which remained in situ after excavation, but with no success. 2. A very similar slab was found in the 2008 excavations season inside the Level J-4 monumental temple (Fig. 2.17; see Adams in Part III of this chapter). 3. A slab with the same spiral or circular decoration was found in the University of Chicago excavations (Fig. 2.18). This slab depicted four decapitated igures with a net pattern incised on top. The circular decoration, made by blows of a stone, was created on top of these incisions. The net pattern and the circle of stones were interpreted as magical symbols of the capturing or killing of these igures (Kempinski 1989: 173; Schroer and Keel 2005: 240). A similar decoration of incised lines around the edges of a slab was observed on another slab discovered by Loud (1948: Pl. 274: 8). According to Guillaume, these slabs could be used as game boards (Chapter 25).
The pre-dynastic King Scorpion is contemporary with the EB IB in the southern Levant. However, there is no clear synchronism between the early Egyptian kings and the EB IB strata at Megiddo.
42
The Level J-3 Wall 08/J/21 was built above Wall 08/J/30, and enclosed the courtyard at a later stage. It is clear that this wall was later than the Level J-2 08/J/200 Picture Pavement, as it was separated from it by a few cm of earth debris. The eastern edge of Wall 08/J/21 was very clear, and no continuation of this wall was found either by us or by the University of Chicagos team. Therefore, it is possible that this was a north-south entrance to the cultic complex of Level J-3. It is also possible that the face of the wall was plastered, according to the plaster bits that were found to its east. The layers above Wall 08/J/21 may have belonged to Level J-4; however there was no architecture with which to relate them, so their actual afiliation remains unclear. It is possible that the series of clay surfaces (08/J/134) represents multiple resurfacing of an exterior courtyard, maybe serving the same function as the earlier Picture Pavement. It is interesting to note that the earliest phase of the 08/J/134 clay surfaces was laid in a slope, whereas the latest phase of this locus was more or less horizontal as if some sort of levelling was needed. This change could be related to the construction of massive Wall 4045,
Fig. 2.17: Slab with a circle of depressions, found in the Level J-4 temple, looking south.
Fig. 2.18: Incised slab with a circle of depressions (Loud 1948: Pl. 272: 4).
43
A DI K EINAN
unearthed by the University of Chicago team farther to the east. If this wall was used as a terrace wall, it would explain the levelling of the whole area of the cultic complex of this phase. This levelling phase can be observed in the latest clay surface of 08/J/134, as well as the layers above it the phytolith layer with the occupational level on top (08/J/116) both are horizontal. It is also possible that all these clay surfaces represent a single levelling event, carried out in a few stages. If this were the case, the clay surfaces would have been contemporary with the phytolith and ashy layers above them.
CONCLUSIONS
Considering the new information, the stratigraphy of Lower Area J can be reined as follows: 1. The lower pavement (08/J/200) and its foundation (08/J/215) represent the earliest phase of the complex. Following the renewed excavations it was concluded that the Picture Pavement is contemporary with the Level J-2 temple that underlies Level J-3 Temple 4050. 2. The upper pavement and Wall 08/J/21 belong to the next phase, which seems to be contemporary with the Level J-3 Temple 4050. 3. Clay Surfaces 08/J/134, uncovered above Wall 08/J/21, may suggest a later courtyard, contemporary with the monumental temple of Level J-4. 4. The ashy, horizontal layer (08/J/116) above the phytolith layer, representing the occupational debris above the clay surfaces, may also be attributed to Level J-4.
44
APPENDIX: TABLE OF THE UNIVERSITY OF CHICAGOS INCISED SLABS

No. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Plate No. in Megiddo II 271: 1 271: 2 271: 3 272: 4 273: 5 273: 6 273: 7 274: 8 275: 9 275: 10 275: 11 276: 12 276: 13 277: 14 277: 15 277: 16 278: 17 278: 18 278: 19 278: 20 279: 21 279: 22 279: 23 280: 24 280: 25 280: 26 280: 27 280: 28 Field No. Slab location c 307 d 282 c 322 c 334 c 308 c 309 c 311 c 320 d 283 c 312 c 336 c 98 c 314 d 280 c 313 d 284 c 319 d 281 c 315 c 321 d 279 c 344 d 285 (A, B) c 323 d 278 c 332 c 310 c 335 Oriental Institute, Chicago Oriental Institute, Chicago Oriental Institute, Chicago Oriental Institute, Chicago Israel Museum, Jerusalem Oriental Institute, Chicago Unknown Oriental Institute, Chicago Oriental Institute, Chicago Oriental Institute, Chicago Approx. slab meas. (cm) 48 34 57 28 45.5 28 50 30 42 21.5 61 24 48 32 35.5 30.5 58.5 32 Approx. incision meas. (cm) 23 19 87 9.5 8; 11 10 40 22 14 13 32 14 17 10 32 25 13 9 33 17 26 13 36 18 19 12 16 9 22 15 18 11 13 8 14 14 15 11 26 16 20 19 66 12 6; 8 5; 12 5.5; 11.5 6 16 8 9.5 8 16.5 11 74 87 Upper/ Lower Pavement L L L L L L L L L L L L L L L L L L L L L L L L L U L L
Rockefeller Museum, Jerusalem 64 19.5 Oriental Institute, Chicago Oriental Institute, Chicago Oriental Institute, Chicago Oriental Institute, Chicago 60 26 48 30.5 57 35.5 48 33
Rockefeller Museum, Jerusalem 42 29 Unknown Oriental Institute, Chicago Oriental Institute, Chicago 51 45.5 48 27
Rockefeller Museum, Jerusalem 45 20.5 Rockefeller Museum, Jerusalem 45 20.5 Rockefeller Museum, Jerusalem 59 25.5 Oriental Institute, Chicago Unknown Oriental Institute, Chicago Unknown Oriental Institute, Chicago Unknown 61 48 47 25.5 23 12.5 -
45
A DI K EINAN
APPENDIX: TABLE OF THE UNIVERSITY OF CHICAGOS INCISED SLABS (cont.)

No. 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Plate No. in Megiddo II 280: 29 281: 30 281: 31 281: 32 281: 33 282: 34 282: 35 282: 36 Field No. Slab location c 333 c 325 c 331 c 316 c 330 c 318 c 328 c 329 Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Approx. slab meas. (cm) Approx. incision meas. (cm) 18 16 16 11 10 10; 6.5 4.5 15 14 15 14 26 22 15 7; 15 12 23 18 Upper/ Lower Pavement U L U L L L U U L L L L -
Unpublished c 324 Unpublished c 326 Unpublished c 327 Unpublished c 317 Unpublished Unpublished Unpublished Unpublished -
46
AREA J, PART III: THE MAIN SECTOR OF AREA J

Matthew J. Adams
PART III: THE MAIN SECTOR OF AREA J

Matthew J. Adams
This sub-chapter presents new information relating to Levels J-4, J-5, J-6 and J-7 (see Tables 2.2 and 2.4) with an appendix on the 2010 season. The Area J Middle and Late Bronze stratigraphy is discussed in Part IV of this chapter. The Middle Bronze Age ceramic sequence and tombs will be presented in future reports. While the basic stratigraphic results of the 2008 operation in Sub-Area Lower J are discussed by Adi Keinan in Part II of this chapter, the current sub-chapter discusses the larger stratigraphic implications of that work relating to the overall stratigraphic sequence of the Early Bronze Age. Table 2.4 summarizes the current stratigraphic attributions of the Early Bronze Age levels in relation to those previously published. Note that unlike the other areas under investigation by the renewed excavations, Area Js stratigraphic numbering system begins from the earliest occupation (see Finkelstein and Ussishkin 2000a: 34). TABLE 2.4: COMPARISON OF EARLY BRONZE LEVEL ATTRIBUTIONS, 1994PRESENT
Megiddo III Level J-4 Period EB IB Megiddo IV Level J-4 Gap J-4a Gap J-5 J-6 EB III EB III J-5 J-6b J-6a J-7 EB III+ J-7 Period EB IB EB IB EB IB EB II EB III EB III EB III EB III+ J-5 J-6b J-6a J-7 EB III EB III EB III EB III-IBA** J-4a EB IBII* Current (Megiddo V) Level/Phase J-4 Period EB IB
* EB IB according to Finkelstein and Ussishkin. ** EB III according to Finkelstein (Chapter 40), IBA according to Adams (this chapter) and Ussishkin (Chapter 39).
LEVEL J-2
During the 20042008 seasons, remains of Level J-2 were encountered only in the re-exposure of the University of Chicagos Stratum XIX Picture Pavement 4008 in Lower Area J (08/J/200; for new incised slabs see Keinan in Part II of this chapter). This pavement was linked in Megiddo III with the temple found directly beneath the Level J-3 (Stratum XIX) Temple 4050 (Finkelstein and Ussishkin 2000: 38, 55, Fig. 3.10).
47
48
Fig. 2.19: Remains of Level J-2 on the eastern slope.
M ATTHEW J. A DAMS
CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J
Fig. 2.20: Remains of Level J-3 on the eastern slope.
49
M ATTHEW J. A DAMS
The curved northern wall appearing on the Stratum XIX plan, our Wall 08/J/21, was interpreted by the University of Chicagos excavators as contemporary with this picture pavement. This was accepted in Megiddo III (Finkelstein and Ussishkin 2000: 52, Fig. 3.10; see also Level J-3 plan, below, Fig. 2.20). The 2008 season demonstrated that Wall 08/J/21 is later than the picture pavement (Fig. 2.9). This wall has therefore been assigned to Level J-3 (Fig. 2.20). Level J-2, then, is composed of the temple below Temple 4050, its paved ascending courtyard, Picture Pavement 08/J/200 and Wall 08/J/30 (Fig. 2.19).
LEVEL J-3
Level J-3 is deined by Temple 4050, which had been exposed by the University of Chicago and reinvestigated by us in 1994 (Loud 1948: Fig. 390; Finkelstein and Ussishkin 2000: 5052). The 2008 season demonstrated that curved Wall 08/J/21 dates later than Picture Pavement 08/J/200 of Level J-2, resting on some 10 cm of ill (see above, Fig. 2.9; Keinan, Part II of this chapter). This is identical to the situation found in the southeast corner of the Level J-3 courtyard where the right-angled wall was found to be resting on a 10+ cm ill over Pavement 4064 (Finkelstein and Ussishkin 2000: 52), indicating that both walls are part of the same construction and both later than the picture pavement (Fig. 2.20).
Fig. 2.21: Schematic section through Area BB showing the location of Incision A (facing north; after Loud 1948: Fig. 416). Level J-4 is in grey. For clarity, the upper pavement of Level J-3 and Wall 08/J/21 are not indicated.
LEVEL J-4: THE GREAT TEMPLE

Level J-4 is deined by the massive monumental temple, the Great Temple, on the summit of the EB IB acropolis (Figs. 2.222.25; see also Adams, Finkelstein and Ussishkin forthcoming). From 1996 to 2000, excavation revealed three monumental mudbrick-on-stone-socle walls extending some 50 m east to west across Area J and a portion of a broad-room-style sanctuary on their northern side (Finkelstein and Ussishkin 2000a; Finkelstein, Ussishkin and Peersmann 2006). The 20042008 seasons primary goal was to continue to expose the plan of the Level J-4 sanctuary. From 2004 to 2006, work proceeded in eight squares (FJ/78) down to the temple loor, and the 2008 season was dedicated to small key-hole incisions aimed at answering speciic architectural and stratigraphic questions. We have now exposed the maximum area of the sanctuary accessible within the University of Chicagos Area BB (Figs. 2.2, 2.24). With approximately 60% of the structure exposed, it is clear that the temple was
50
51
Fig. 2.22: Aerial photograph of the Great Temple of Level J-4 after the 2008 season, facing south.
52
Fig. 2.23: Remains of Level J-4 in the sanctuary of the Great Temple (basalt slabs marked in black).
M ATTHEW J. A DAMS
53
Fig. 2.24: Remains of Level J-4 (basalt slabs marked in black).
54
Fig. 2.25: Suggested reconstruction of the J-4 temple based on a central axis through the entrance and Altar 08/J/204 (basalt slabs marked in black).
M ATTHEW J. A DAMS
a broad-room-style building in the tradition of the Ein Gedi shrine (Ussishkin 1980; Kempinski 1992). This tradition has a history at Megiddo reaching back to the Level J-3 temple and possibly to the underlying Level J-2 temple (Figs. 2.192.20). The reconstructed sanctuary is a broad hall entered from the centre of the northern long wall (Figs. 2.242.25). Opposite the entrance is a mudbrick and stone altar. The roof was supported by a row of 12 columns established on roughly hewn stone bases (six excavated, six others reconstructed). The column row is lanked by six pairs (12 altogether) of large basalt tables (seven tables excavated, ive others reconstructed). The structure also has a set of three corridors that apparently acted as favissae for the bones of sacriices (Fig. 2.24). THE TEMPLE WALLS The sanctuary is bounded on the south by Wall 96/J/7 and on the north by Wall 00/J/21, forming a consistent interior width of 8.90 m. In Megiddo IV, we reported the thickness of the northern sanctuary wall as 2.6 m (Finkelstein, Ussishkin and Peersmann 2006: 38). Excavation of the northern baulk of Square F/7 has proved this incorrect, demonstrating its thickness as 3.45 m, exactly that of southern sanctuary Wall 96/J/7 and corridor Wall 96/J/1 (Fig. 2.23). The inner face of western sanctuary Wall 08/J/25 was completely robbed (in Level J-5, see below); it was identiiable in the baulk of Squares J/7K/7 by the remains of the eastern chalk-paste coating left in situ by the robbers (Fig. 2.26), as was the case in other portions of the robbed sanctuary walls in Squares H/7 through F/7 (Figs. 2.272.28; see Finkelstein, Ussishkin and Peersmann 2006: 38). The plaster is in perfect alignment with the eastern face of Wall 98/J/33 (Square K/9; Fig. 2.24; see Finkelstein, Ussishkin and Peersmann 2006: 3637). The western face of Wall 08/J/25, then, can be reconstructed in line with the western face of Wall 98/J/33 with a thickness of 3.45 m in agreement with the other walls of the sanctuary. The interior of the sanctuary walls were all coated with a thin mud preparation coat followed by a inishing coat of white chalk paste. The paste was not cooked to create a formal plaster but was simply composed of local chalk mixed with water with a straw binder (Friesem and Shahack-Gross, Part V of this report, RME-2526). In places where we could section this plaster and the loor of the temple (e.g., Fig. 2.28), it was clear that the bottom of this coating was at the approximate level of the loor but did not curve into a loor coating as is common in plastered loor buildings. The loor was left with an earthen inish (see below). In Megiddo IV we reconstructed a continuation of Wall 96/J/1 to meet Wall 00/J/19, forming an entrance to the rear corridors (blocked by Wall 98/J/33) by way of a narrow plastered corridor (00/J/185 and 00/J/201; Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.14). We now consider Walls 08/J/25 and 98/J/33 to comprise the exterior structural wall of the main temple ediice. Plastered Corridor 00/J/201, then, must form an alley that allowed access from the front courtyard of the temple (north of Wall 00/J/21) to the area south of Wall 96/J/23. It is assumed that from this period through the Middle Bronze Age, a series of terrace walls were built successively farther west of this point to support the expanding acropolis (see below, and Adams and Bos in Part IV of this chapter). It is likely that Wall 00/J/19 is the earliest example of this practice, forming the western edge of the acropolis terrace. The temple had no subterranean foundations (Figs. 2.282.29). The structure was constructed upon a levelled surface. This surprising feature was tested by small sondages excavated at the interior face of Wall 00/J/21 in Square H/7 (Fig. 2.28) and within the robber trench of the same wall in Square G/7. The latter revealed the bottom of the wall at elevation 155.95 m. Both test trenches revealed that the wall rested
55
M ATTHEW J. A DAMS
Fig. 2.26: Robber trench of Wall 08/J/25 (arrow 1) with chalk-paste coating in situ (arrow 2), facing northwest. Note the subterranean Level J-17 stone Structure 5239 (left, arrow 3).
Fig. 2.27: Level J-4 northern sanctuary Wall 00/J/21, facing northwest. Note remains of chalk-paste coating at points where the wall was robbed.
Fig. 2.28: Square H/7 east section. Striated remains near the base of the section are Level J-4a squatters debris and temple collapse (see also Fig. 2.38). At left, Probe 06/J/099/PT003 reveals the bottom of Wall 00/J/21. Note phytolith level 08/J/185 in low baulk at right.
56
Fig. 2.29: North-south architectural section through Area J. The Great Temple of Level J4 is at bottom and Temple 4040 of Level J-7 at top (see Fig. 2.3).
Fig. 2.30: Level J-4 temple entrance, facing northeast.
57
M ATTHEW J. A DAMS
on a thin dense compact ill (08/J/202), characterized by high concentration of lint debitage and bladelets (see Chapter 15), which in turn covered a thick construction ill (06/J/91). The surface of this thin layer also acted as the temple loor (see below). The lack of subterranean wall foundations was also recognized in the southern corridors in previous seasons (Finkelstein and Ussishkin 2000: 6063). This observation provides an important datum for the reconstruction of the acropolis. From the later location of round Altar 4017, the general slope of the tell was down to the north, west and east (Franklin and Peersmann, personal communication; see Fig. 2.29). Had the Level J-4 temple been constructed on this slope without preparation, deep foundations would have been required, deeper for each wall north of another. As this is not the case, it is clear that a massive illing and terracing operation had been carried out here in preparation for the construction of the temple (Figs. 2.21, 2.29, 2.42, 2.48). This ill, consisting of dense mudbrick debris of a characteristically greenish colour, 06/J/91 (Fig. 2.37), was detected in Squares H/8 through F/8. Mudbrick lines were apparent but did not form a coherent pattern, leading us to the conclusion that it is construction ill for the temple support terrace. The ill must have been laid behind a large terrace wall extending along the northern slope (but see Appendix to Part III). THE ENTRANCE In 2008, excavation of the baulk between Squares E/7F/7 revealed the western half of the entrance to the sanctuary through Wall 00/J/21 (Figs. 2.302.31). The entrance (directly opposite the altar as in the Level J-3 temple) was paved with basalt slabs (Fig. 2.30). The pavement was cracked, but appears to have been
Fig. 2.31: Level J-4 entrance door socket and orthostat plinth (plinth stone is original; orthostat is reconstructed).
58
composed of at least one large slab at least 1 m long and 0.40 m wide (which continues into the eastern and northern baulks). Several thin broken basalt slabs were found lying haphazardly upon the pavement, the collapse of which seems to be responsible for the broken pavement. These thin slabs are separated by a very thin layer of wind-blown dust and may have fallen onto the pavement early in the process of collapse of the temple. These collapsed slabs may be all that remain of a basalt orthostat dado that may have lined the entrance of the building (Fig. 2.31). The basalt-paved entrance is separated from the western jamb of Wall 00/J/21 by approximately 0.75 m. The remains of well laid stones in this gap in the southern portion of the entrance are arranged around a cavity that may have been a door socket (Fig. 2.31). Immediately northwest of this possible socket, in a narrow channel between the end of (and parallel to) Wall 00/J/21 and the socket reinforcement stones, is a smooth lat-topped piece of basalt that may have formed a plinth for the entrance orthostats. THE COLUMN BASES AND ROOFING The roof of the temple was held up by a row of column bases down the centre of the longitudinal axis of the building, bisecting the sanctuary into north and south (Figs. 2.232.25). Six bases were found west of the altar, ive made from roughly hewn limestone, the sixth from basalt. The ive westernmost bases were equidistant from one another and from the western sanctuary wall (an intercolumnar space of ca. 2.50 m; or ca. 3.203.30 m on-centre). A sixth column base, the one farthest east, is off-centre and much closer to its neighbour than the others. Had this base been placed at the next calculated distance, it would have obscured the view of and interrupted access to the altar from the doorway. It is evident in other examples of the broad-room temple that access from door to altar is preferrably unobstructed (e.g., Fig. 2.20). Simply omitting the next base to leave space was inadvisable from an engineering point of view. Therefore, this off-centre base with a presumed counterpart to the east was intended to help span the extra-long gap with an additional architrave beam and leave the axis of the temple clear (Fig. 2.32).1 We were only able to excavate beneath one of these column bases, that in Square J/8 where it is located in the northern section. The base was placed upon a stone foundation that assisted in distributing the weight of the column. The minimum surface area of these column bases is approximately 0.123 m2, indicating a maximum diameter of a column at around 0.40 m. Local trees growing in forests covering Mount Carmel and the Menashe Hills would have been more than adequate for the needs of this temple, whether a second storey was needed or not. Aleppo pine (Pinus halepensis)2 and Mt. Tabor oak (Quercus ithaburensis) certainly could have provided timber for columns and rooing beams in the 5 m range (or even the full 9+ m of the hall width with above-average specimens of each species).3 Olive (Olea europaea) too, could theoretically provide adequate 5 m straight timber as evidenced, for example, by numerous examples with several-metre-tall trunks seen by the author and Lev-Yadun in the region of Beit Lehem HaGlilit. All three
1 There may have been a less functional desire on the part of the architect to have six column bases on either side of the central axis of the temple. There were six column bases and six basalt tables for a reconstructed total of 12. This number may have played an important symbolic role in the Megiddo cult. See Weinstein-Evron and Lev-Yadun (2000) for the discussion of Pinus halepensis as part of the natural landscape before the Late Bronze Age. Remains are now well attested in the archaeological record for the Early Bronze age, particularly at Megiddo (Lev-Yadun and Weinstein-Evron 2002; Liphschitz 2006; contra Liphschitz, Chapter 30 in this volume). I wish to thank Professor Simcha Lev-Yadun of Haifa University for hours of conversation about the potential of local timber in the construction of this temple and for supplying the arboreal information herein.
59
60
Fig. 2.32: Level J-4 Great Temple sanctuary, facing south. Note alternate alignment of easternmost column base (left) in relation to entrance and altar.
M ATTHEW J. A DAMS
species grew locally in the Early Bronze Age and are well attested in the wood assemblage of Megiddo and other sites in the Jezreel Valley (Liphschitz 2007: 38, Table 2.6; Lev-Yadun and Weinstein-Evron 2002). Speciically for Level J-4, Olea europaea dominates the wood assemblage (Liphschitz 2000; 2006; this volume).4 Thus, there is no need to hypothesize the importation of Cedrus libani (Cedar of Lebanon), which, in any case, is not attested in the wood assemblages of the Jezreel Valley until the Late Bronze Age (Liphschitz 2007: 38, Table 2.6). Given forests in the proximity of the site, the potential to yield timber of the sizes needed (Lev-Yadun and Weinstein-Evron 2002) and the known wood samples from Level J-4 (Liphschitz 2006 and this volume), the most likely candidates for beams and columns within the Level J-4 temple are the local Pinus halepensis (see note 3 above), Quercus ithaburensis and Olea europaea. An extensive lens of grass phytoliths among the collapse within the temple (see below) suggests the possibility of a thatched roof (see Friesem and Shahack-Gross, Part V of this chapter). Whatever the height of the building, the amount of collapsed debris within the temple suggests only a single storey. This, however, does not rule out the possibility of useable space on the roof for storage, or perhaps, astronomical observation. THE BASALT RITUAL TABLES Perhaps the most striking feature of the whole temple is the system of massive basalt ritual tables set in two rows lanking the longitudinal axis of the sanctuary (Figs. 2.232.32). There were at least eight of these massive basalt ritual tables,5 six rectangular and two circular. A convergence of metrological data that will be presented below demands a reconstruction of the temple with the latitudinal axis centred upon the altar and entrance, and forces a reconstruction of 12 total basalt tables. This mirrored reconstruction jibes well with the fact that the altar is lanked by rectangular ritual tables rather than alternating rectangular and circular. The rectangular slabs are approximately 1 m by 1.5 m and the circular are approximately 1.5 m in diameter (see speciic measurements below, Table 2.5). The only slab whose bottom was reached was 08/J/207, where we measured its thickness at 20.5 cm. We tentatively assume that the others are similar, with an average weight near one ton each. The centre of each of these slabs is equidistant from its neighbour in the same row, and the centre line of the ritual table in each row is parallel to the other. In Megiddo IV, it was hypothesized that these basalt slabs were the column bases supporting the temple roof (Finkelstein, Ussishkin and Peersmann 2006: 38 and Fig. 3.29). At the time, however, we had only revealed two of the centre limestone slabs (those in Square F/8), which appeared to have no relationship to the rest of the architecture. With further exposure of the central row of limestone bases, the large basalt slabs must be interpreted differently. The only evidence for their function is the presence of a beautifully carved holemouth bowl in circular Table 08/J/207 (Figs. 2.322.33). The cavity, off-centre in favour of the north, is 12 cm deep with a rim diameter of 23 cm. Presumably this was a receptacle for a liquid that played a role in the cult. The very existence of the cavity, which must have had a functional purpose, eliminates the possibility that the slabs themselves were objects of veneration (as in massebot). The fact that the cavity is off-centre suggests that something else held the centre position, perhaps a cult statue or
4 Most of the wood remains reported by Liphschitz (2000; 2006) do not come from the sanctuary itself. Those that come from the squatters debris (Phase J-4a) include the following species: Olea europaea, Pistacia palaestina, Cupressus sempervirens and Quercus calliprinos (Liphschitz 2006: Table 26.1). Unfortunately, it is impossible to tell what species were reused during the squatters phase from the wooden construction in the original temple. While not exposed, the counterpart to 08/J/211 is virtually assured.
61
M ATTHEW J. A DAMS
other fetish. In this light, the term ritual table seems apt. Receptacles of an architectural nature related to temple sanctuaries and ritual are known from both Mesopotamia and Egypt in the late fourth and third millennia. The ED III Ishtar Temple H at Assur contained several oval hemispherical bowls embedded in the loor near the edge of the internal sanctuary podium (Andrae 1922: 39, Abb. 1112; Ellis 1968: 127). The contemporary Ishtar temple cella at Mari also had numerous similar oval bowls sunk near architectural features such as benches and pedestals (cf. our tables) (Parrot 1956: passim, see especially Fig. 30ad and Pl. XI, 24; Ellis 1968: 127). Something similar may have had a parallel use in the late Pre-Dynastic/ Early Dynastic sacred area at the site of Mendes in Egypt (Adams 2009). I would like to further suggest the possibility that this receptacle served a function similar to the holemouth jar imbedded in the loor near a basalt slab in the Level J-3 temple (Finkelstein and Ussishkin 2000a: Fig 3.24, the bones from within being secondary; Wapnish and Hesse 2000: 444). The characteristically green colour of the terrace Fill 06/J/91 beneath the temple loor (see below) proved advantageous because we could easily trace the outlines of the foundation trenches for the basalt ritual tables (Fig. 2.34). It was clear, most obviously in the case of the circular ritual tables, that foundation trenches only slightly larger than the size and shape of the tables were dug and illed with medium-sized ield stones; the basalt ritual tables were then lowered down into the foundation trench. This allowed the table tops to be only a few cm above the loor of the temple. A sondage (08/J/108) into a portion of the foundation trench for ritual Table 08/J/207 afforded us an opportunity to verify the foundation trench and to ascertain the thickness of one of the tables; ritual Table 08/J/207 proved to be 20.5 cm thick. While we exposed some of the foundation stones beneath the table, we were not able to determine the depth of these foundations. With the thickness of the ritual table, we were able to approximate the weight of this particular table and to estimate the weight of the others (Table 2.5). TABLE 2.5: BASALT RITUAL TABLES DATA6
Ritual Table Shape 08/J/205 08/J/206 08/J/207 08/J/208 08/J/209 08/J/210 08/J/211 Rectangular Rectangular Circular Circular Rectangular Rectangular Rectangular Length 1.65 m* 1.65 m* 1.61m Areac = 1.80 m 1.70 m 1.60 m* Width 0.83 m 0.85 m 1.61 m 1.45 m2 1.02 m 0.89 m 0.80 m Heighta 0.205 m 0.205 m 0.205 m 0.205 m 0.205 m 0.205 m 0.205 m Volume 0.2807 m3* 0.2875 m3* 0.4174 m
3
Weightb 0.7691 tons 0.7878 tons 1.1435 tons 0.8146 tons 1.0313 tons 0.8500 tons 0.7190 tons
0.2973 m3 0.3764 m
3
0.3102 m3 0.2624 m
3*
An asterisk indicates that the measurement has been estimated. a. Since only one of the slabs was excavated to its base, its thickness is assumed for the purpose of this table. b. Based on a standard of 2.74 tons per cubic m of basalt. c. Calculated by chord (0.9 m) and segment height (0.17 m) since most of the table is within the baulk.
62
Fig. 2.33: Cavity within Table 08/J/207, facing north. Collapsed chalk wall coating was found inside the cavity (08/J/101). Note collapsed chalk in section, beneath which was the phytolith layer (08/J/185), both sealing the cavity and table.
Fig. 2.34: Level J-4 Tables 08/J/207 and 08/J/208, in 2006. Note foundation trenches of tables cut into lighter greenish ill (06/J/091). Squares G/7 and G/8, facing northwest.
63
M ATTHEW J. A DAMS
BENCHES On the north side of the sanctuary abutting the southern chalk-paste coated face of Wall 00/J/21 and lanking the western side of the entrance, we found evidence of a mudbrick Bench 08/J/24 coated with chalk paste (Square F/7; Fig. 2.35).7 In Megiddo IV, the presence of two well-plastered steps (here, chalk coating8) lanking the western side of the altar and running parallel to and abutting the southern sanctuary Wall 96/J/7 was indicated (Finkelstein, Ussishkin and Peersmann 2006: 38). In the 2006 and 2008 seasons, we identiied the continuation of the upper step in the baulk between Squares H/8H/9 extending 65 cm from Wall 96/J/7 (Fig. 2.23). Here, there was not a second, lower step. We were thus prompted to re-examine the lower step in Square F/9 (in 2008) and determined that the coating was wall collapse and that there was no sign of a mudbrick substructure. There was indeed only one mudbrick bench along the southern wall. In several places in the Squares H/8H/9 baulk, the putative chalk coating of the bench was found lying over the northern edge of the step, clearly having fallen onto the edge and broken. Also, in places on top of the bench, chalk coating was present lying at a slanted angle away from the wall, and, while it is clearly the same material as the wall coating, its consistency is more puddle-like and sorted, suggesting that it washed down off of the wall onto the bench. The presence of other chalk puddles and water-erosion channels in the post-abandonment phase (Phase J-4a) of the temple attests to the effect of exposure to the elements (see below and Fig. 2.36). Additionally, it appeared in some places that this puddle coating lay over already denuded portions of the bench, suggesting that it accumulated after the abandonment of the temple. Unlike the northern bench, in no place along the southern bench was there any evidence of chalk-paste vertically coating it; all chalk was found on top of it. Thus, in contrast to the bench on the north side of the hall, there is no indication of chalk coating on the southern bench. This bench may have continued all the way to the western sanctuary wall from the altar along Wall 96/J/7 or have stopped short of the western wall (as in the Ein Gedi shrine, Ussishkin 1980: Fig. 3). We found no indication of a comparable bench on the east side of the altar in Square E/9, nor was there evidence of a bench against the western Wall 08/J/25. It should also be noted that the bench clearly was not constructed until the second phase of the altar (see below). THE FLOOR The irst encounters with loors relating to this building outside the sanctuary were in the rear corridors where they were made of clay with a thin layer of ash topped with an accumulation of bones (Finkelstein and Ussishkin 2000: 60). Within the sanctuary, the loor was irst encountered in 1998 in Square F/9. There, west of the altar, the two steps (above) stepped down onto a whitish surface (Finkelstein, Ussishkin and Peersmann 2006: 38), which was originally interpreted as a plastered loor (ibid.). However, re-exposure in 2008 revealed that it was, in fact, a very compacted thin lens of ash that was present 1 to 1.5 m around the altar.
7 8
Because of the mirrored reconstruction mentioned above and elaborated below, we assume that a second bench lanked the eastern side of the doorway. This plaster has been identiied technically as a chalk paste (i.e., it was not cooked; see below and Part V).
64
Fig. 2.35: Level J-4 Bench 08/J/024 against Wall 00/J/21. Square F/7 facing northeast.
Fig. 2.36: Square H/8-H/9 baulk, facing southeast. Note Phase J-4a water erosion channels and chalk Puddle 08/J/177. Note also the brick edge of the Level J-4 bench against Wall 96/J/07.
Work in the 2000 season revealed the inside of the temple in Squares F/78 and indicated that there were two loors. The earliest was a mud-packed loor with a greenish hue (06/J/91).9 According to the interpretation offered in Megiddo IV, a brief period of abandonment as the result of an earthquake followed this loor and ca. 15 cm of mudbrick debris accumulated before renewed habitation at the site saw the construction of a new loor (00/J/166) (Finkelstein and Ussishkin 2003; Finkelstein, Ussishkin and Peersmann 2006).
This was not found thinly plastered as reported in Megiddo IV (Finkelstein, Ussishkin and Peersmann 2006: 38). This report was a conlation of the thin compact ash from Square F/9 interpreted as a loor in 1998 and the discovery of this surface in 2000.
65
M ATTHEW J. A DAMS
In 2006, in Squares H/8 and G/8 we excavated down through the upper loor and eventually reached the greenish mud surface seen in 2000. The upper loor turned out to be a thin phytolith lens found in most parts of the temple (08/J/185; see Friesem and Shahack-Gross, Part V: Table 2.7, RME-30) and overlying a few of the basalt ritual tables and pillar bases (i.e., the pillars had been removed before the phytolith deposition); it thus post-dates the abandonment of the temple and is related to later collapse and sporadic activity within the temple (see Phase J-4a, below). In Square G/8 it was clear that the greenish mud was not a loor but rather mudbrick debris (06/J/91) that predated the construction of the temple. Lines of bricks were readily identiiable. Additionally, it was clear that the circular basalt ritual tables were cut down into this greenish debris (see above and Fig. 2.34). In 2006 and 2008 we found that this greenish debris is overlaid by a 23-cm-thick dense compact mud (08/J/202; Fig. 2.37). This mud was laid in order to level irregularities in the construction ill before building and was found in several squares beneath collapse and Phase J-4a debris and also in every sondage beneath the northern sanctuary wall. In summary, the greenish mudbrick debris encountered in several squares is representative of part of the massive terracing ill upon which the temple was constructed (but see Appendix to Part III). This rough surface was coated with the thin dense mud that smoothed out this platform to prepare for the temple construction. The surface of this smoothing coat served as the loor of the temple (08/J/202). Contrary to what we reported in Megiddo IV, there exists only one loor of the Level J-4 temple the surface of the surface-smoothing construction layer, 08/J/202. THE ALTAR Altar 08/J/204 is a rectangular installation placed against the northern face of Wall 96/J/7. It is constructed of mudbricks and stone (Fig. 2.38), similar to the second altar in the Level J-3 temple (Loud 1948: Figs. 139, 142). The upper portion of the altar is not preserved and most sides of the installation are denuded. There is evidence that at least the northern face was coated with a thin chalk paste. It is unclear how much of the upper portion of the altar is missing. The stone elements appear to have been topped by mudbrick and thus should be considered part of the internal structure. As reported in Megiddo IV, one component of the altar appears to have been a plastered basin (having undergone multiple re-plastering) in the south central portion (Finkelstein, Ussishkin and Peersmann 2006: 39). The podium in the cella of the ED III Ishtar temple at Mari contained several large oblong ceramic basins embedded in the feature that may be parallel to the plastered element here (Parrot 1956: passim, especially Figs. 12, 15, 30, 43 and Pl. XIV). A thin ash layer (08/J/213) was traced on the loor from around the northwest corner of the altar to the rectangular basalt ritual Table 08/J/210 where it hit approximately 12 cm below the surface of the table. It petered out farther to the west, but probably represents ash-yielding activity within the vicinity of the altar. In tracing the ash lens around the altar, it became clear that it went beneath the front of the altar and beneath the lower step, now understood as collapse (see above). Removing a small section of the northwestern corner of the altar, an earlier phase of the installation was exposed. The ash went beneath the addition and met this earlier altar. There are, thus, at least two phases of this altar (Fig. 2.38). The bench along the southern sanctuary wall abutted the later altar, but clearly did not touch the earlier, suggesting that the bench and the later altar were constructed together. This situation is similar to that in Level J-3 Sanctuary 4050 (Loud 1948: Fig. 143). In the temples of both levels, there was no evidence of a site-wide rebuild that would warrant a separation into two sub-levels. The altar was the main focal point of cultic activity within the sanctuary. As great ritual was required in the
66
Fig. 2.37: Squares HG/78, eastern section.
67
M ATTHEW J. A DAMS
Fig. 2.38: Level J-4 Altar 08/J/204 and Table 08/J/210, facing southeast. Note earlier incarnation of altar and ash lens beneath metre.
Fig. 2.39: Level J-4 Table 08/J/210 and foundation deposit, facing south. Small section relects thickness of loor makeup 08/J/190 at this location.
construction of sacred buildings (foundation ceremonies, deposits, etc.), it is possible that the altar within an altar that we ind in Levels J-3 and J-4 is a relection of construction ritual rather than an actual rebuild and the ash lens a product of that ritual. In both Levels J-3 and J-4, the underlying structure is in excellent shape and completely intact, with no sign of dismantlement whatsoever. Evidence for ritual practices in the construction of temple altars is known from Mesopotamia (Ellis 1968: 3233). By late into the 2008 season we understood the basic coniguration of the altar, but one very important question remained outstanding: how did the altar relate to the southern wall of the temple? To better understand this an incision was laid out extending 2 m from Square E/9 and positioned 1.5 m wide, centred within the Level J-7 Temple 4040 threshold. The southeastern corner of the altar was found, but the continuation of the rear wall of the sanctuary was not. As of now, the idea presented in Megiddo IV that there was a niche behind the altar remains a possible scenario. If so, that niche is at least as deep as
68
Fig. 2.40: Level J-4 foundation deposit.
the southern extent of our Square E/9 incision. The possibility of robbing cannot be ruled out, however, and the corner of a stone in the southern section of Square E/9 may be all that remains of the rear, southern wall at this point. This incision also demonstrated that the mudbrick superstructure of the altar was preserved as high as elevation 157.50 m in the southeastern corner, some 50 cm higher than was detected in the excavation of the rest of the altar. A STONE SLAB WITH INCISED MARKS Late in the 2008 season, we decided to remove a portion of the dense loor makeup in Square F/8 between basalt ritual Table 08/J/210 and the nearest limestone column base in order to tidy a section for inal photography. The removal of this loor makeup, 08/J/202, down to the greenish mudbrick Material 06/J/91 revealed a lone piece of limestone pavement with drill-marks around the outer edges (Figs. 2.392.40). This stone was placed into the surface of the greenish brick-material terrace ill and covered over by the dense mud of the inal temple loor. A small patch of grey ash with small bits of charcoal was detected about 10 cm away from the stone, also sandwiched between the construction ill and smoothing coat. Similar slabs are known from Lower Area J Picture Pavement 08/J/200. The irst was discovered by the University of Chicago (Loud 1948: Pl. 272: 4; Keinan 2007: 41); here the marks around the edge appear chipped out. The second was observed in Lower Area J by Ussishkin during the excavation of the courtyard of the Level J-3 temple (Finkelstein and Ussishkin 2000: 52). And the third was recovered by us in Lower Area J (Slab I; see Keinan in Part II of this chapter); the marks on this slab, too, are chipped rather than drilled as in the Level J-4 example. To this category of slabs with perimeter markings I would also add Louds Slab 8 (Loud 1948: Pl. 274), which is marked around the perimeter with incised lines. As a corpus, they are coherent in their perimeter markings, but their original meanings are not readily apparent (for suggestions, see Adams forthcoming a). While the signiicance of the pattern is not clear (see Keinan 2007 and Keinan in Part II of this chapter for hypotheses and references),10 its presence within the loor makeup of the temple must be signiicant.
10 Guillaumes suggestion of a game board (Chapter 24) strikes me as highly doubtful.
69
M ATTHEW J. A DAMS
The planning, construction and dedication of sacred buildings are processes bounteous with sacred meaning. Textual material and archaeological evidence from Egypt and the Near East detail rituals and traditions associated with the erection of important buildings from palaces to temples and tombs. The attention given to these often elaborate rituals and the expense waged on precious deposits speaks to the vital nature of such acts to the ultimate success of the structure, its occupants and the community. With this dedication to the sacred space and building, it is not surprising that similar importance was placed on desacralizing such structures once they had served their purpose or ceased to function. These acts can be called Termination Rituals, a phrase recently introduced to Mesoamerican archaeology by S. Zuckerman (2007). Such was also the case when a sacred space was to be built anew. The old structure would be desacralized and the new structure would go through the foundation rituals. Both actions were required in order to transfer the sanctity to the new building. Theoretically, there should be archaeological signs of such acts in the transitions between the temples of Megiddo from Level J-2 through J-7, and perhaps beyond. These issues in relation to the inds at Megiddo are explored in more detail elsewhere (Adams forthcoming a; Adams, Finkelstein and Ussishkin forthcoming). Given the similarity between the foundation slab beneath the loor of the Great Temple and those from Levels J-2 and J-3, it is likely that it did in fact originate in one or both of the earlier temples. Two needs would have confronted the architect of the Great Temple: 1) to legitimately desacralize the earlier Level J-3 temple and 2) to transfer the sanctity to the new construction. While a number of activities may be involved in satisfying these needs, one appears common in both Mesopotamia and Egypt. The Mesopotamian kal ritual prescribes the removal of one brick of the old temple to be placed in a special location where special foundation offerings for the new temple would take place (Ellis 1968: 13, 184). At the site of Mendes, Egypt, the construction of a new temple pylon by Ramesses II required the dismantling of an earlier structure by Thutmose III. Over one hundred bricks stamped with Thutmoses cartouche were laid carefully within the sand foundation of the new pylon. In the case of the Great Temple at Megiddo, the slab from the earlier temple was placed beneath the dense packed layer 08/J/202 that served as both loor of the temple and the surface upon which the temple walls were built. That is, it was probably laid during the physical laying out of the building. To the Egyptians, this ritual was known as the p-s (stretching of the cord)-ritual where the king himself was required to ix the four corners of the new building (Weinstein 1973: 9). This foundation ritual is known from as early as the 1st Dynasty reign of Den (Wilkinson 1999). It therefore seems prudent to interpret the slab within the Great Temple as a foundation deposit laid during the early ceremonies for the construction of the temple. Since the slab appears to come from the earlier Level J-2 or J-3 temple, its function may have been to transfer the sanctity of the earlier building to the new temple. The ash found near the slab may be the physical remains of the associated ritual. It should be noted that the loor makeup was not removed everywhere within the sanctuary and other such slabs or deposits may yet be found. ACTIVITY WITHIN THE SANCTUARY Virtually no non-architectural evidence for activity within the temple was found. The loor was completely clean and emptied of any contents. A thin ashy layer was detected around Altar 08/J/213 (Fig. 2.38), where it had been compacted into a thin laminate, perhaps as a result of being walked on. Since this thin laminate goes under the renovated portion of the altar, this ash presumably belongs to activity within the temple (as opposed to Phase J-4a activity). No evidence of in situ burning was found anywhere nearby. It is probable
70
that this ash, since it goes beneath the second phase of the altar, belongs to the founding ritual of the altar itself (see above). In the northwestern corner of the temple (Square J/7) a thicker deposit of ash was found (06/J/62, 06/J/82, 08/J/145). It was not possible to establish whether it belongs to the occupation of the temple or to the later Phase J-4a activity (I prefer the latter possibility). The ashes from the burnt sacriices (the burning of which would have taken place outside the sanctuary) were undoubtedly charged with sacred power and perhaps utilized and/or stored within the sanctuary.11 The remains of these ash deposits perhaps allude to such activity. THE CORRIDORS It was argued above that western Alley 00/J/201 runs between the sanctuary ediice and Terrace Wall 00/J/19 rather than leads into the rear corridors. This frees us from the interpretation of these rear corridors as walkways. Better, they can be seen as compartments for the storage of animal bones from the sacriice. That is, they were used as locations for favissae for the charred remains of the sacriiced animals. This conforms well to the faunal assemblage discovered within the compartments. Wapnish and Hesse (2000) determined that different locations within the favissae were reserved for different debris from different stages of the carcass processing. The structured deposition and segregation of this material according to locations within the favissae lends support to the sanctity of the process as well as suggests ritual in the discard process (idem. 449). Note also that tabular scrapers sometimes associated with ritual butchering were found within the favissae (Blockman and Groman-Yeroslavski 2006: 321; see McConaughy 2003: 511). EXCURSUS: WALL 4114 AND BUILDING 4113 Two additional architectural elements should be afiliated with Level J-4. They are University of Chicago Wall 4114 (Loud 1948: Figs. 392394) and University of Chicago Building 4113 (Loud 1948: Fig. 391). Effectively, Wall 4114 is a southern extension of the eastern wall of the Level J-4 temple, and acts as a terrace wall revetting the upper terrace (Fig. 2.41). This wall connects to Building 4113, which serves as the gateway from the lower terrace to the upper terrace. This spatial relationship is preserved in all of the later Early Bronze Age strata, XVIIXVI (Level J-6) and XV (Level J-7) where the location of Building 4113 served as a contact point from east to west, allowing access to the area behind the temple complexes. In Strata XVIIXVI, this access was by staircase controlled by Building 3177 and by the plastered path along the south side of the same building (Fig. 2.55; Loud 1948: Fig. 392). In Stratum XV a monumental staircase replaced Building 3177 (Loud 1948: Fig. 394). Wall 4114 abutted Building 4113, which appears to have been a monumental gate to the sacriicial courtyard behind the sanctuary, the same courtyard that had access to the bone favissae corridors of the Great Temple.12 It should also be noted that Building 4113 was constructed on a slope of ca. 4 down to the east (Loud 1948: Fig. 149), a fact consistent with the Level J-4 activity on the lower terrace (see below).
11 Compare Numbers 19:110. David Ussishkin also reminds me of the later but parallel example of Schumachers Masseboth Temple (University of Chicago Building 338). There the ashes from burnt offerings were stored in a large heap in the corner (Ussishkin 1989: 156). 12 This is contra Golani (1999: 126127), who interprets Building 4113 as a house with rounded corners and Ben-Tor (1973: 9496) and Herzog (1997: 67), both of whom interpreted this structure as the Stratum XVIII incarnation of the Stratum XIX temple, well before our Great Temple of Level J-4 was known.
71
M ATTHEW J. A DAMS
The placement of Wall 4114 on the Level J-4 plan (above) completes the length of the temple whereby the altar and entrance are the axis of symmetry (Figs. 2.25, 2.41). In this reconstruction, the length of the temple from the east face of Wall 4114 to the west face of Wall 08/J/25 is ca. 49 m,13 where the central latitudinal axis is exactly through the centre of the altar and the entrance to the temple. Therefore, we reconstruct this axis of symmetry through the entrance of the building allowing for 12 pillar bases (including two out of alignment that support the axis-spanning beam) and 12 basalt ritual tables (eight rectangular and four circular) in six pairs (Figs. 2.25, 2.41).This is additionally supported by the fact that in such a restoration the fragment of a corner found on the far eastern end of Wall 96/J/1 comes into line perfectly with the inner face of the hypothesized east wall of the sanctuary (Figs. 2.242.25). THE LOWER TERRACE The massive temple was only one aspect of an ambitious construction project undertaken in Level J-4. As is well known, the large 34-m-thick stone Wall 4045 was constructed some 40 m down the slope (Loud 1948: 391). This wall has been the subject of much debate, variously interpreted as a city wall and a terrace wall and dated from the EB I to the EB III (Kenyon 1958; Brandfon 1977; Dunayevsky and Kempinski 1973; Esse 1991; Finkelstein and Ussishkin 2000b: 580583). There is no need to rehash the arguments here as new data from our excavations renders them obsolete.14 The Great Temple of Level J-4 is a new discovery; its presence was only hinted at by the small portion of Wall 96/J/1 on the western edge of the University of Chicagos Stratum XVIII. Its sheer size and monumentality at the top of the acropolis comprise an unprecedented engineering venture that certainly would have needed signiicant terracing of the original mound to support it (especially considering the fact that none of the temple walls had subterranean foundations). Evidence for the attribution Wall 4045 to Level J-4 and its interpretation as a terrace wall comes from our 2008 excavations in Sub-Area Lower J (see Keinan, Part II in this Chapter). Revealed in Incision A was an ashy accumulation laying on a mud and phytolith lens (08/J/116; 08/J/124) that accumulated on the surface of a thick clay Floor 08/J/134. These loci yielded only EB IB pottery. From the section, it was clear that this thick clay surface was actually a series of laminated clay surfaces probably belonging to a pavement. The clay laminations may represent resurfacing over time, but their clean nature suggests that they actually represent stages in the construction process of a thick levelling clay ill (Figs. 2.92.10, 2.12, 2.21). The clay was laid over a regular earthen and bricky ill (08/J/137), which overlaid the Level J-3 remains.
13 Being more speciic with measurements is dificult considering the resolution of accuracy on the University of Chicagos published plans. 14 Esses assessment of this wall should be addressed here, because his conclusions have become widely accepted. He concluded that the ill behind and at the base of the wall contained EB III pottery (Esse 1991: 81; University of Chicago photographic archive negative B-4347). This led him to conclude that the wall dated to the EB III. The only information given by the University of Chicago is that this pottery comes from west of Wall 4045 (W=4045) and therefore could have come from any EB III disturbance at any depth below the loor surfaces of Building 3177. It should be borne in mind that the construction of Building 3177 was a major undertaking that certainly brought in ills and cut into earlier strata. In this case, as in many others, the University of Chicagos loci are simply not useful in this type of situation. Until the renewed excavation, for example, Chicago pottery loci have been used to wrongly date Stratum XVIII to the EB II and/or III (e.g., Kenyon 1958; Dunayevsky and Kempinski 1973; Esse 1991).
72
Fig. 2.41: Level J-4 Great Temple and environs reconstructed (walls marked in grey were unearthed by the University of Chicago team).
73
M ATTHEW J. A DAMS
This clay ill marks an important transition on this portion of the site. The bottom of this ill is at a different angle than the top, suggesting that the clay laminate was intended to change the angle of the slope. The bottom of the laminate is 12 while the inished angle is only 3, effectively levelling the eastern slope.15 As this level lies overtop of the Level J-3 Wall 08/J/W21 and beneath the ill for EB III Building 3177 and as it yields only EB IB pottery, it must be attributed to Level J-4. When the north section of Incision A is placed on Chicagos northern section drawing, it becomes apparent that this clay levelling belongs to the construction of Wall 4045 (Figs. 2.21, 2.42). It should also be noted that the new angle of the lower terrace is the same as the slope of Structure 4113, interpreted above as a gateway. The importance to the builders of this new levelling construction and the new clay pavement is evidenced by the burial of a fully articulated fetus/infant sheep/goat (08/J/135) immediately beneath the clay surface, evidentially laid as a foundation offering.
PHASE J-4A: A PERIOD OF CRISIS

A few years ago Finkelstein and Ussishkin (2003, 2006) offered a reinterpretation of the stratigraphy of the Level J-4 temple and the transition to the EB III based on the then-current understanding of two loors within the temple. As described above, however, what was previously identiied as the lower loor is actually the construction ill for the temple platform. The later loor turned out to be a phytolith lens representing collapse and sporadic Phase J-4a activity within the temple. There is evidence for only one phase of primary use of the temple Level J-4. This phase was followed by a crisis period during which the temple was allowed to deteriorate. Within the corridors, Square D/10 showed evidence of water-washed mudbrick and other sorted debris over the Level J-4 bone accumulation in the corridors (96/J/096; Finkelstein and Ussishkin 2000b: 585). Within the sanctuary, there is evidence for numerous ephemeral hearths attesting to the temples continued sporadic use during this period. OWL PELLETS A total of ive owl pellets were discovered in Area J. The irst two come from the corridors behind the temple sanctuary, which were used as bone favissae (94/J/81 and 96/J/21). Wapnish and Hesse indicated that owl pellets are typically found in areas where owls were nesting and that these roosts are typically found in spaces little used or deserted by humans (Wapnish and Hesse 2000: 444445; see also Chapter 30). Three additional owl pellets were identiied within the temple sanctuary itself (in Loci 98/J/122, 08/J/142) in the 1998 and 2008 seasons Locus 98/J/122 yielded two pellets directly on the loor immediately west of the altar. Both were identiied in situ. The ifth pellet came from the surface of the monumental basalt threshold (08/J/142). The entrance pellet was identiied by the numerous small fauna bones recovered during sifting of earth within 5 cm of the basalt threshold 08/J/144 (identiied in the ield by Aharon Sasson). While the owl pellets found within the corridors were not conclusive evidence of the abandonment of the temple, those within the sanctuary demonstrate unequivocally that owls roosted within the building where they regurgitated the bones of their prey, and that no one returned to clean up after them. Their location directly on the loor demonstrates that the building remained rarely visited as it began to collapse.
15 The range of angles of the slope in Levels J-2 and J-3 is 12 to 15 (see sections in Loud 1948: Figs. 149, 416).
74
Fig. 2.42: Schematic section west-east through the Great Temple and lower terrace.
THE CRISIS AND SPORADIC OCCUPATION: ASSESSING THE SEQUENCE In an effort to better understand the sequence of activity during the crisis phase, a portion of the 2008 season was dedicated to a detailed study of the 1020 cm accumulation (Figs. 2.28, 2.37) in the Squares H/89 baulk, the Squares GH/7+GH/8 baulk and the Square J/7 western incision. To complement this approach, a sampling strategy was coordinated with Ruth Shahack-Gross. (The results of the micromorphological analysis of these samples appear in Part V of this chapter; RME numbers refer to these samples, see Table 2.8.) The simplest of the sequences was excavated in the Square J/7 western incision. Here, the distinct Level J-5 construction Fill 08/J/212 overlaid the robbed Wall 08/J/25, the preserved plaster face of that wall and a 23-cm-thick grey-white ash accumulation that also exhibited traces of ired mudbrick fragments (08/J/145). The ash layer overlaid a mud-packed surface with clear signs of heating (low temperature iring of mud surface). Within the ash and clearly upon the loor were a scattering of bones, including an articulation of a young bos primogenius PH2 and 3 (preliminary ield identiication; notably the species and carcass part itself is statistically rare in the corridor assemblage, Wapnish and Hesse 2000). These bones were not singed; it would appear that the ash, then, was swept away from the primary ire location. The presence of a distinct and contiguous phytolith layer (08/J/185, RME-30) provided a link between the sequences excavated across the sanctuary and supplied important stratigraphic data regarding this period (Phase J-4a; Figs. 2.432.44; see also Fig. 2.28): 1. The phytolith layer clearly overlays the circular basalt tables. It is evidenced in the Square G/8 southern baulk and was detected on top of Tables 08/J/207, 08/J/208 and 08/J/205. This phytolith layer was interpreted in Square F/8 in previous seasons as the upper loor and also overlaid portions of the rectangular basalt Table 08/J/210. The cavity within Table 08/J/207 was illed with collapsed brick and plaster (Fig. 2.33), and was perfectly sealed by the phytolith layer, demonstrating that the phytolith was deposited after the temple had begun to crumble. 2. The phytolith layer completely overlaid the basalt pillar base near the circular tables (Figs. 2.37, 2.44). Certainly the wooden pillars had to have been removed before the deposition of the phytolith layer. 3. The phytolith is clearly overlaid by the super compacted and dense clayish-earth mixed with chunky white chalk fragments excavated in Square G/7 and the Squares G/78 baulk (08/J/18=08/J/193/LB01; RME-36), where it was conined. These chalk fragments are whiter than the wall plaster of the temple and may not belong to the Level J-4 architecture as we understand it. 4. The phytolith layer is clearly overlaid by ca. 15 cm of mudbrick debris and lots of bones. This is the bone layer found in 2000 in Square F/8 (the upper loor; Finkelstein, Ussishkin and Peersmann 2006: 38). An archaeozoological comparison of this assemblage to that in the bone favissae (corridors) has
75
M ATTHEW J. A DAMS
not yet been made. In what may be a contrast with the corridor favissae,16 the inal deposition of the bones in the sanctuary occurred after the Level J-4 temple began to collapse, i.e., in Phase J-4a. 5. The phytolith layer is overlaid and underlaid by ephemeral hearths scattered around the sanctuary (Figs. 2.432.44). Several of these hearths, particularly in the Squares GH/7 baulk, are superimposed as a result of multiple visits. Fragments of collapsed chalk paste from the temple walls were found under, over and between all of these hearths. 6. Most noticeably on the northern end of the Squares GH/7 baulk (Fig. 2.28; but evidenced throughout), the phytolith layer clearly overlies and underlies collapsed plaster debris identical with the plaster preserved on the walls of the Level J-4 temple. This demonstrates that the phytolith layer is but one episode in the slow collapse of the Level J-4 temple after its abandonment. The stratigraphic sequence that resulted from the detailed study is presented from bottom to top in Table 2.6. TABLE 2.6: SCHEMATIC STRATIGRAPHIC SEQUENCE OF THE LEVEL J-4 FLOOR AND THE PHASE J-4A ACTIVITY
Main Locus 06/J/91 08/J/190=08/J/202 08/J/190 08/J/101 08/J/185=06/J/65 08/J/193=08/J/18 08/J/188 08/J/184 08/J/187 08/J/177 08/J/177=178 08/J/129=212 Description Green brick material Level J-4 loor makeup Sandy sorted washed-in debris
a
Sample none taken 08/J/190/LB02 08/J/190/LB06 none taken 08/J/185/LB01b 08/J/193/LB01 08/J/188/LB04 08/J/184/LB04 08/J/187/LB01 08/J/177/LB02 08/J/177/LB01e 08/J/129/LB01
Minor chalk-paste collapse within 08/J/207 cavity Phytolith layer and hearths Dense chalky deposit against north Wall 00/J/21 Ephemeral earthen surface onto which 08/J/184 washed Natural washed-in debrisc Ashy hearths Earth beneath plaster puddle (bottom of locus) Chalk puddle (top of locus) and late hearths Level J-5 ill over Level J-4 templef
d
a. Found primarily on the north side of the H/7-G-7 baulk. b. 08/J/185/LB001 is from the phytolith, not the hearth. c. Located in a 12 m2 patch on the north side of the G/7-H-8 baulk. d. The plaster puddle was found in the southern portions of G/8 and the H/8-H-9 baulk. e. 08/J/177/LB001 is from the plaster puddle, not the hearth. f. This level covers the entire temple and other strata (=08/J/212).
16 It should be noted that because of our new understanding that the deposition of the bones within the sanctuary belong to Phase J-4a, the interpretation that the bones in the corridors belong to Level J-4 is called into question. This issue cannot be resolved at this time.
76
Fig. 2.43: Square H/8, facing north, showing Level J-5 phytolith layer 08/J/185 (foreground) and post-phytolith hearth (background).
INTERPRETATION OF THE SEQUENCE The Phase J-4a crisis period can be said to have begun with the irst deterioration of the temple. This is most plain in the chalky wall and ceiling collapse within the cavity in basalt ritual Table 08/J/207. Since we assume that this cavity had a function related to the primary use of the temple, its illing with collapse marks the end of the temple use as it was previously known and the physical incarnation of the beginning of Phase J-4a. The earliest disintegration occurred in the cavity and doorway, either naturally or intentionally, as the basalt orthostats collapsed onto the pavement. Over time, owls roosted in the sanctuary and the corridors. The temple began to deteriorate. As rain water began to seep into cracks in the roof, sorted sandy, silty and clay puddles formed (08/J/190; see also Fig. 2.36). Water began to percolate between the mudbrick and stone walls and their plaster. The weakened plaster collapsed slowly in places and is evident in small collapses around the sanctuary throughout the abandonment phase. Hearths among the early accumulation with the temple suggest sporadic usage throughout this period. At some point, the wooden columns were pilfered, causing the roof to collapse into the sanctuary. It overlaid several of the basalt ritual tables and column bases and eventually rotted away leaving behind phytolith Layer 08/J/185. After the collapse of the roof, the deterioration intensiied. On the north side of the sanctuary, isolated to Square G/7, a dense and thick deposition of clay and chalky chunks accumulated (08/J/18=08/J/193/ LB01; RME-36). On the south side of the temple, the collapsed roof left the walls more susceptible to rain damage. Rain water dissolved plaster from the walls and collected in pools at the base of Wall 96/J/7 (most evident in the Squares H/89 baulk; Fig. 2.36) where the chalk precipitated in puddles (08/J/177). Bone stuck within the matrix of the puddle supports this hypothesis. Among this post-phytolith collapse, someone continued to make use of the structure as evident from several hearths (e.g., 08/J/185, 08/J/187).
77
M ATTHEW J. A DAMS
Fig. 2.44: Square GH/78 baulk, facing north, showing Phase J-4a phytolith Layer 08/J/185 (foreground) and Hearth 08/J/177 (background). Note the foundation stones of Table 08/J/207, right.
The doorway was relatively free of debris from this period. Because of the fact that occupational debris was found inside the temple, it stands to reason that the entrance remained accessible through most of this period. When the site was reinhabited on a large scale in Level J-5, the new inhabitants would have encountered a structure with its roof caved in, its plastered walls worn from the weather, and 20 to 50 cm of occupational debris, water- and windborne soil and structural collapse within the sanctuary. The main structure was still standing, however, and this fact is critical to how they reused the site (see below). THE QUESTION OF THE SANCTUARY BONES There are two basic deposits of bones in the temple complex. The irst to be identiied came from the corridors behind the temple (Wapnish and Hesse 2000) and was a favissa of spent animal bones from sacriice and/or sacred feasting at the temple. The second deposit came from inside the temple sanctuary. Here, bones were encountered within the ca. 30 cm of Phase J-4a occupation, both below and above the phytolith layer. Few, if any, bones can be associated with the main usage of the temple in Level J-4. The bones belonged either to the Phase J-4a accumulation as primary deposition, or they were secondarily deposited through the action of the Phase J-4a users of the temple during the process of collapse and weathering. The presence of articulations, however, argues for the former, and it appears most likely that
78
the bone deposits within the sanctuary are to be connected to the continuation of cultic activity during Phase J-4a (albeit in a signiicantly altered manner).
AN EB IB EARTHQUAKE: EVIDENCE FOR AND AGAINST

The process described above of the Phase J-4a crisis of the Great Temple is opposed to the earthquake hypothesis proffered in Megiddo IV for the destruction and abandonment of the site at the end of the EB IB (Finkelstein and Ussishkin 2003; 2006; Marco et al. 2006). Let us briely review the evidence previously presented: 1. Monumental walls of the Level J-4 temple are fractured along their strike. It is noted (erroneously, see below) that the overlying walls of Temple 4040 are not fractured, suggesting that the shockwave that caused the fractures occurred between the construction of the Level J-4 temple and the Level J-7 temples (Marco et al. 2006: Table 31.1, No. 1; Finkelstein, Ussishkin and Peersmann 2006: 49). If the upper temple were fractured, it would suggest either: a) that if there had been an earthquake it happened later, affecting both strata, or b) the damage is related to something else (see suggestion below). 2. The sequence of two loors within the temple sanctuary separated by ca. 15 cm of collapsed mudbrick debris was interpreted as abandonment of the temple, collapse from the walls and reoccupation of the temple (Finkelstein, Ussishkin and Peersmann 2006: 38). 3. The north face of Wall 96/J/1, in Square D/10, has a vertical crack, and the wall east of the crack has sunk slightly toward the slope (Finkelstein, Ussishkin and Peersmann 2006: 49). Likewise, the north face of Wall 96/J/7 in Square H/9 displays a pulling-apart of the wall and a sinking of the wall west of the separated interstices (Fig. 2.45; Marco and Finkelstein, observations in the ield, 2008). 4. When the inhabitants returned to the temple after the earthquake (Phase J-4a) they attempted to shore-up some of the crumbling walls within the corridors (Finkelstein, Ussishkin and Peersmann 2006: 5051). This included the construction of two limsy mudbrick Walls 96/J/11 and 98/J/12, perpendicularly across the corridors, presumably to keep the towering 3.5-m-thick walls from collapsing (ibid.: Fig. 3.20). 5. Basalt ritual Table 08/J/205 has clearly sunk. The western section of Square H/7 that runs through the table shows that this sinking caused a vertical fracture in the strata above (Fig. 2.46; Marco and Finkelstein, observations in the ield, 2008). 6. Wall 96/J/1 (Squares BC/11) was repaired by laying lat stone slabs atop the stone foundation to receive new mudbricks (ibid. 2006: 51). The 20042006 excavations in Area J provided new material that forces us to reconsider the evidence above. I offer the following data to refute the interpretation of the evidence above (numbers below correspond to the numbers above): 1. Level J-7 Temple 4040 above the Level J-4 temple does display numerous examples of fractures along wall strikes. I observed and photographed no fewer than eight separate examples of fracture lines across numerous stones along their strike in Walls 08/J/37, 08/J/38, 08/J/39 and 08/J/40 (Fig. 2.56). For the sake of brevity, one example is provided in Fig. 2.47.17 The strike fractures in the Level J-4 temple, therefore, provide no evidence for an earthquake at the end of the EB IB.
17 I further suggest that strike fractures are a common feature of hewn-limestone walls. The builders choice of where to strike the stone to create the desired shape directly relates to how the stone is then placed within the wall. If a relatively smooth face is desired for a wall, the builder will naturally place the stone in the wall such that the lines of fracture are
79
M ATTHEW J. A DAMS
2. The evidence for two loors within the sanctuary has been corrected above. There was one loor, followed by slow collapse and irregular activity. 3. The vertical cracking and sinking of wall segments should be expected at these locations. These points are closer to the east and west edges of the upper terrace. The Level J-4 construction would have required enormous ills to support the temple, which would have been weakest along the deeper edges, thus more prone to natural sinking.18 4. The shoring up of the corridor walls is a very important component of Phase J-4a as presented in Megiddo IV. As we shall see in more detail below, these two walls are artefacts of excavation, and are, in fact, part of a laid-mudbrick ill placed between and along the entire length of the corridors that was excavated through. This ill was laid in Level J-5 as part of the dismantling of the Level J-4 temple and the preparation of the site for construction at that time. 5. The fracturing of the accumulation near basalt ritual Table 08/J/205 is clear. There certainly was settling. As in number 3 of my refutations, above, this settling is expected. However, this settling occurred after the deposition of the Phase J-4a activity, which is represented in the striations in the section (Fig. 2.46). While much later settling is my preferred explanation, if one must hypothesize an earthquake, it will have to have taken place after the accumulation of Phase J-4a debris and the collapse of the Level J-4 roof, and therefore cannot be responsible for the abandonment of the site in the EB IB. 6. The technique of laying lat stones at the top of the socle was almost certainly part of the construction technique for the Level J-4 temple. It is an attested EB architectural feature at Qiryat Ata, for example (Golani 2003: 75). Additionally, the lat stones in Wall 96/J/1 can be none other than those robbed from the Levels J-2 and J-3 Picture Pavement. These stones are distinct in their size, shape and thickness, and a good portion of stones at the top of the wall are identical.19 As Level J-3 was almost certainly covered in the course of construction of Wall 4045 and the terrace ill, it is most likely that these stones were robbed and used in the Level J-4 construction project. The proximity of this end of the wall to the location of the earlier pavement explains why it is only here that the builders used these particular stones. Note also the foundation deposit stone within the Level J-4 sanctuary, which almost certainly comes from the Picture Pavement as well (see above). In conclusion, all of the evidence for an earthquake in the EB IB at Megiddo has been refuted by new data. Socio-economic and political factors must underlie the Phase J-4a crisis of the Level J-4 Great Temple, just as they underlie the striking change in settlement patterns throughout the Jezreel Valley at this time (Finkelstein et al. 2006: 763; Finkelstein and Ussishkin 2000b: 585588; Adams, Finkelstein and Ussishkin forthcoming). While activity continued at the site irregularly, the site was nearly vacant during the succeeding EB II period (Finkelstein and Ussishkin 2000b: 585586; Greenberg 2003). When the site was reoccupied in the EB III (Level J-5/J-6), the roof had caved in, but the mudbrick walls, although weathered by years of exposure, were probably standing to at least half if not most of their original height. The Level J-5/J-6 inhabitants of the site dismantled portions of the temple, robbed stone from the walls and illed in the sanctuary to create space for their construction activity.
parallel to the face of the wall, because he will have just hewn along that line to create a smooth faade. These lines of fracture dictate that fractures along strikes are statistically more probable in any roughly hewn limestone wall. 18 Incidentally, it is not clear when this sinking occurred during the life of the temple, in the period of abandonment, or much later, after the building up of the mound. 19 Some of these lat stones also appear within the wall itself, not just on top.
80
Fig. 2.45: North face of Level J-4 Wall 96/J/07 in Square H/9, facing south. Note chalk puddle overlying Level J-4 bench below.
Fig. 2.46: Square H/7 western section showing sunken basalt Table 08/J/205. Note Phase J-4a squatters remains (striations in section) broken as a result of sinking (i.e., the sinking occurred after the deposition of the squatters remains).
81
M ATTHEW J. A DAMS
Fig. 2.47: Strike fractures in Temple 4040 Wall 08/J/38.
LEVEL J-5/J-6
UPDATED STRATIGRAPHIC TERMINOLOGY AND INTRODUCTION Up through the 2000 season our work concentrated in locations where the Level J-7 temples and later burials and pits had obliterated most of the EB III remains below, down to Level J-4. Our picture, therefore, of Level J-5/J-6 was patchy at best. Originally, two levels were identiied between Levels J-4 and J-7. Level J-5 was assigned to squatter activity on the ruins of the Level J-4 temple, representing the reoccupation of the site in the EB III (Finkelstein and Ussishkin 2000a: 6567). Level J-5/J-6 represented the permanent habitation on the site following this squatting activity. Partially on this basis of the presence of EB IB material in Level J-5 loci, in a 2003 article, Finkelstein and Ussishkin reinterpreted some of the Level J-5 loci, speciically the locus containing the Egyptianized Pottery, into a Level J-4 late phase that was a (nonsquatter) reoccupation of the temple in the late EB IB after its destruction by an earthquake (Joffe 2000: 170174; Finkelstein and Ussishkin 2003; Finkelstein, Ussishkin and Peersmann 2006a). The rest of the Level J-5 loci remained in the EB III (Finkelstein and Ussishkin 2003: 28, 37). The remains of Level J-5
82
were scanty enough before the relocation of some loci to Level J-4, so this new interpretation left Level J-5 virtually non-existent. Since the 2000 season, it has been a primary goal to further understand the nature of the Levels J-5 and J-5/J-6 material. The seeming abandonment of the temple and its later sporadic use has been described above. The reoccupation of the site is clearly marked by the laying of a thick ill of well-levigated mudbrick material to prepare for a paved street and building complex. The building endured three main phases of renovation before the construction of Level J-7 Temple 4040. Since the buildings represent continuous usage, they are considered to have belonged to a single level (J-5/J-6). As described above, Phase J-4a has been assigned to the period of sporadic use after the primary use of the Level J-4 temple. Level J-5/J-6 marks the return to the site in the EB III after a period of abandonment (Phase J-4a). Excavation since 2004 in Main Area J has almost doubled the exposure of Level J-5/J-6 and done so in a sizable area, providing a much clearer picture of the levels sandwiched between the Level J-4 Great Temple and the triple temple complex of Level J-7. The Level J-5/J-6 architecture on this northern portion of Area J belongs to a large complex that underwent three major renovations, labelled here Level J-5 (early), Phase J-6b (middle) and Phase J-6a (late). Though these three layers should be considered a single level with subphases, we retain the numbering scheme presented in Megiddo IV. The architecture is deined by an east-west running paved street with three buildings appended to its northern side (Figs. 2.512.53). On the basis of both layout and stratigraphy, Level J-5/J-6 can be equated with Strata XVII and XVI (Loud 1948: Figs. 392393). All of the exposed architecture from this level was rebuilt two to three times, attesting to the longevity of this layer. See also the appendix on the 2010 season at the end of Part III. LEVEL J-5 TERRAFORMING PROJECT: NEW VISIONS OF THE ACROPOLIS While the exact length of the period of the Phase J-4a sporadic activity is unknown, it seems to have occurred for most, if not all of the EB II period (Joffe 2000; Halpern 2000; Finkelstein and Ussishkin 2000b: 585591; Greenberg 2003, 2006; Finkelstein, Ussishkin and Halpern 2006; and see already Kenyon 1958; see also Adams, Chapter 8 in this volume). When the newcomers arrived at the site, they found a town in ruins but with the Level J-4 artiicial topography still intact. The lower terrace revetted by Wall 4045 and the upper terrace revetted by Wall 4114 were still distinct spaces and the Level J-5/J-6 occupants maintained them. The lower terrace was levelled and Building 3177 was constructed (Stratum XVII; Loud 1948: 7078). On the upper terrace, the old temples walls, though ruined, were still standing. The Level J-5 inhabitants opted for transforming the site. The lower terrace was essentially used as found with some minor illing and cutting to prepare for Building 3177. Preparing the upper terrace for reoccupation involved much more work, including the dismantling of the remaining superstructure of the Level J-4 temple and the illing in of spaces around the temple. Three different levels rising from north to south were conceptualized for the upper terrace, consisting of a lower northward sloping level bounded on the south by a street (north of Wall 96/J/7), a mudbrick podium slightly higher than the lower architecture also bounded on the south by a street and a third level, south of this street (Figs. 2.29, 2.48). For the central level, the thick walls of the Level J-4 temple provided a suitable foundation and the standing ruins were renovated accordingly to create a podium. The mudbrick upper portions of Walls 96/J/7 and 96/J/1 were systematically dismantled (not destroyed). The bricks were used to ill the corridors up to the desired elevation. Evidence for this platform emerged during recent excavation seasons, in
83
M ATTHEW J. A DAMS
reinterpreting older strata and in cleaning older sections. The reconsideration of Megiddo III and IV s Level J-5 required the reattribution of those loci to other phases. Loci (and architectural elements) that were clearly EB III have been retained in Level J-5. Those loci that Finkelstein and Ussishkin had moved to Phase J-4a (2003; 2006a), however, had to be looked at afresh. In particular, Walls 98/J/11 and 98/J/12 were problematic (Finkelstein and Ussishkin 2006a: Fig. 3.20). First, the walls raison dtre depended on the interpretation of a post-earthquake repair of the Level J-4 temple. With the earthquake hypothesis controverted, these walls needed to be reassigned stratigraphically. Second, it was highly suspicious that these walls were known only from sections, recognized only after excavation (Figs. 2.492.50; see also Finkelstein and Ussishkin 2000: Fig. 3.433.45; Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.13). Suspicions heightened in 2006 when, after several seasons of weathering, the east and west sections of Square H/10 revealed two new walls (Fig. 2.50) and then after a hard rain in winter 2008, the southern section of the same square revealed contiguous laid mudbrick. Rather than a number of cross-walls through the northern corridor, all coincidentally only found later in sections, we actually have a laid-mudbrick ill within the corridor.20 Additional support is found in the fact that the temple corridor walls (Walls 96/J/1 and 96/J/7) were not robbed,21 mudbrick was found in situ upon these walls in some places and the brick superstructure was clearly excavated through in others (Fig. 2.49; Finkelstein and Ussishkin 2000a: 6061 and Figs. 3.283.39; Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.13). The bottom level of the newly observed mudbrick ill in the corridor of Square H/10 was at elevation 157.14 m. This is the same height as the reported upper loor within the corridor (see Finkelstein and Ussishkin 2000: Fig. 3.26; Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.14). This upper loor should be considered the surface upon which the mudbrick ill was laid, i.e., a surface that represents the end of the natural and/or anthropogenic accumulation of Phase J-4a. Especially when considering the Phase J-4a accumulation within the Level J-4 sanctuary, these observations throw into high relief the problem of dating the bone deposition within the corridors (above, passim). This can only be solved with a reinvestigation of the corridor deposits. Whereas the corridors were illed with an organized mudbrick ill at the outset of Level J-5/J-6, the space north of Wall 96/J/7 received a different treatment. Walls 00/J/21 and 08/J/25 were completely dismantled. The stone socle was almost completely robbed and the stones were evidently used for the construction of the Level J-5 building above as is indicated by the reuse of at least one of these stones in the construction of Wall 06/J/17. Bricks from the walls were crushed and levigated to create a very distinct ill (08/J/212 = 08/J/129), which was laid in the sanctuary and over the robbed walls. This is demonstrated unequivocally by the fact that no robber trenches were found above the walls; the ill lays continuously over them and the sanctuary. Fill 08/J/212 was laid by the Level J-5 contractors directly on top of the Phase J-4a remains within the Level J-4 sanctuary. This debris had accumulated only 10 to 70 cm above the loor of the temple. Only in the immediate vicinity of the altar was there evidence of major wall collapse. This may be due to the fact that the initial robbing of the columns during early Phase J-4a would have brought immediate structural
20 The dificulty in discovering this lies in the fact that these bricks came from the upper portions of the Level J-4 walls themselves, and thus had been quite weathered by the time they were dismantled, making their appearance to the excavators little different than mudbrick collapse. The recognition that these walls are actually a Level J-5 ill has signiicant bearing on the interpretation of the Egyptianizing pottery cache. I deal with this elsewhere. 21 The robbing of Wall 96/J/7 in Square H/9 was done in the Middle Bronze Age. See Adams and Bos in Part IV of this chapter.
84
weakness to the centre where the additional offset columns and support beam had been. Whatever the case, the distinct Level J-5 ill was easily identiiable by its very well crushed mudbrick debris and small fragments of the chalk paste coating (i.e., bricks and chalk coating from the Level J-4 walls themselves). It became so identiiable that it was known as the Snickers ill during excavation. It could be easily distinguished from the Phase J-4a debris below and the collapse on top of the altar. THE STREET AND BUILDINGS The levigated mudbrick Fill 08/J/212 within the sanctuary was laid on a slope that extended from approximately the top of the socle of Wall 96/J/7 down to the north and west (see Figs. 2.37, 2.48). This slope changes 0.83 m over a distance of 16 m (from the southeast corner of Square F/7 to Square J/7). The reason for the slope was probably related to how the architects designed the area farther to the north (beyond Area BBs northern baulk, i.e., beneath Area AA) whether it was terraced down or was allowed to slope all the way to the bottom of the tell. The architecture above the Level J-4 sanctuary consisted of an east-west street paved with ist sized cobbles (00/J/9), approximately 3.5 m north of and parallel to Wall 96/J/7 (again demonstrating a foreknowledge of the earlier temple). North of the street were three buildings labelled A, B and C (from west to east, each separated by a narrow alley; Fig. 2.51). The street was bound on its southern edge by Wall 98/J/2, which stretches at least 20 m from Square E/8 to Square H/7. The space between Wall 98/J/2 and Level J-4 Wall 97/J/7 was all part of the levigated Fill 08/J/212. There is no construction abutting the south side of the street. Wall 96/J/7 was probably preserved higher than the street, forming a terrace to the south and Wall 98/J/2 may have served to revet the ill between it and Wall 98/J/7, effectively extending the central terrace northwards some 4 m. Building A in Square J/7 is the least known. Only a small corner has been excavated in which was found a large storage jar (Fig. 8.4: 1). This storage jar apparently remained in use through Phase J-6b (and Phase J-6a?) at which time it had become partially subterranean with the raising of loors. Building B was the most exposed of the three structures, with its entire east-west dimension revealed. It is 11 m long and over 5.7 m wide. Unfortunately, its central space is poorly preserved, destroyed by later intrusions and rebuildings. The corner of the room preserved in Square J/7 yielded two installations that belong either to this phase or to Phase J-6b: one made from the upper portion of a holemouth jar held in place by stones (04/J/114; Fig. 8.9: 7), and another just north of it, made of thin standing stones formed into a square (04/J/113). One entrance to the building appears to have been on the east side from the alley.
Fig. 2.48: Level J-5 terraformation project showing three levels of the upper acropolis. See also Fig. 2.29.
85
M ATTHEW J. A DAMS
Fig. 2.49: Square F/10 west section showing Level J-5 laid brick ill between Level J-4 walls. Note bricks on top of Wall 06/J/01.
Fig. 2.50: Square H/10 east section showing Level J-5 laid brick ill between Level J-4 walls.
86
Fig. 2.51: Remains of Level J-5.
87
M ATTHEW J. A DAMS
Here, a row of three larger stones (probably robbed from the Level J-4 temple below) may have served as a threshold. Building C consists of four rooms separated by interior Walls 98/J/29 and 08/J/20. In Square F/7, a step descended from the southern room to the northern room in accordance with the general slope of the terrace. Nothing is known about the eastern rooms. The north-south measurements of the Building A room and the Building C southwestern rooms are identical, perhaps suggesting similar plans. PHASE J-6B It is most likely that the three-phase remodelling of the structures in the main sector of Area J is schematic. It is doubtful that all three buildings were dismantled and rebuilt at the same time. Rather, each building was probably repaired or rebuilt in its own time frame. Phase J-6b, then, is an artiicial snapshot sometime in the middle of this long process (Fig. 2.52). Buildings A and C remained basically unchanged in terms of their overall plan. Building As eastern wall was rebuilt as Wall 04/J/6, consciously wider than it was previously, perhaps indicating the addition of a second storey. Within the building, the loor was raised ca. 30 cm and the large store-jar that originally sat on the Level J-5 surface became partially subterranean. Building C underwent an adjustment to its southwestern corner. The southern continuation of Wall 06/J/11 was removed and the southern rooms Floor 98/J/144 was raised and extended out into the western alley. While the later Middle Bronze intrusion (Pit 06/J/140) makes it dificult to conirm, Wall 96/J/1 appears to have been extended to meet the eastern Wall 04/J/29 of Building B. This change appears to have made Buildings B and C into one structure, or at least changed the entry of each so that they had a shared entrance. That new entrance from the former alley came from the north. One arrived at a small pebble Pavement 06/J/97, which acted as a common threshold for both buildings. To turn to the west would be to enter Building B (as it was in Level J-5), but to continue south would be to enter the southwestern room of Building C. Within Building C plastered Bench 00/J/116 was built against Wall 96/J/31. In the northern room the only sign of activity consists of phytolith surfaces and 20-cm-deep ash Pit 00/J/160. While it had a number of superimposed loor surfaces, the general plan of Building B changed only slightly in Phase J-6b. It shifted to a more eastern diagonal (relative to the southern) street. The nature of activity of the westernmost room appears to have been maintained and Level J-5 Installations 04/J/114 and 04/J/113 either continued in use or were built subterranean in this level. Partition Wall 04/J/14 was built to separate these from new Installation 06/J/119 an ashy hearth. From Phase J-6b through Phase J-6a, the architecture in this room remains the same, but a substantial number of loor raisings attests to its longevity and intense usage (Fig. 2.54). A new addition to the complex appears in Phase J-6b. On the eastern end of the street a building was constructed, extending southwards. Two walls were found abutting southern street Wall 96/J/2: Walls 08/J/22 and 08/J/26, the southern portion of which were discovered in Square F/9 in previous seasons. West of Wall 08/J/22, only distinct Level J-5 Fill 08/J/212 material was found, suggesting that the corner formed by the street and this southern room circumvallated a space south of the street. This space was the central locus of the mudbrick podium created in Level J-5 by illing in the Level J-4 walls.
88
Fig. 2.52: Remains of Phase J-6b.
89
90
Fig. 2.53: Remains of Phase J-6a (University of Chicago-exposed walls marked in black).
M ATTHEW J. A DAMS
Fig. 2.54: Square H/7-8 west section indicating the thick loor buildup of Phases J-6b and J-6a. Note Level J-4 basalt table (bottom left), and Phase J-4a accumulation broken by sinking of table (see also Fig. 2.46).
PHASE J-6A Phase J-6a marks the most signiicant change to Building C, which has been so completely altered that it will now be referred to as Building D (Fig. 2.53). Building B also underwent signiicant alterations, but to its eastern rooms only. Building A appears to have continued unchanged save for a loor raising that may or may not have resulted in the cutting off of the upper portion of the storage vessel and paving over it. The western rooms of Building B retain the character that they had attained in Phase J-6b with the addition of two ashy Installations 06/J/108 and 06/J/115. Installation 06/J/108 was a corner of the room cordoned off by large stones. It had a mud and pebble packed surface with some lat-laying pottery with areas of patchy ash. Partition Wall 04/J/14 separated this installation from Installation 06/J/115 (which replaced 06/J/109) cordoned off by a row of mudbricks. Continuing from Phase J-6b, this space witnesses several loor raisings throughout the period (Fig. 2.54). It is from this room that the most striking evidence comes for the nature of these buildings: a cylinder seal (04/J/095/AR1; see Brandl, Chapter 18, this volume) and a pear-shaped macehead (04/J/072/AR1) (Blockman and Sass, Chapter 15, this volume, Fig. 15.3: 4). The eastern exterior wall of Building B (Wall 04/J/26) was moved about 1.5 m to the west and where there were originally two rooms in Level J-5 and Phase J-6b, there was now one large room. Building C was replaced by Building D, little of which is preserved. Wall 98/J/9 was constructed as the western wall of the building. Almost nothing can be said about the interior of the building to the east, as it is heavily disturbed by Middle Bronze pits. The University of Chicago uncovered one wall east of and parallel to Wall 98/J/9 in our Square E/7 (in black in Fig. 2.53; Loud 1948: Fig. 393, Square N/13). The long east-west street and the space between Buildings B and D received more dramatic upgrades. The Level J-5 and Phase J-6b cobbled Street 00/J/9 was repaved with larger lagstones (98/J/136) that raised the level of the street to the elevation of ca. 157.60 m. This new surface height became a benchmark for
91
92
Fig. 2.55: Schematic reconstruction of Level J-6 (University of Chicago-exposed architecture in black).
M ATTHEW J. A DAMS
the space between Buildings B and D, to where the lagstone pavement was partially extended. There, the out-of-use walls of Phase J-6b (part of Walls 06/J/29 and 04/J/27) were removed down to the same height as the street and integrated into the street pavement. The result seems to have been a one-metre northern tongue of lagstone pavement. It is unclear how much of Wall 04/J/29 to the north continued this pavement or was buried beneath a surface. On the west side of this northern extension, there was a gap of about one metre before the western edge of Building B. The gap seems to have had an earthen surface, 04/J/27, and was separated from the street by fence Wall 04/J/5. Within this space was a curious feature, 04/J/62, in which four lat-topped stones were neatly arranged around a smaller lat-topped stone, creating a round lat table. The feature had a foundation of small pebbles approximately 20 cm deep, suggesting that it was designed to support some amount of weight. However, it does not seem logical to interpret it as a column base since it is built up against Wall 04/J/26. The best explanation at this time is that this is some sort of minor cult installation. In this scenario, Installation 04/J/62 is a table supporting some sort of statue, other icon, or fetish (a heavy one?), and fence Wall 04/J/5 delineates the sacred space. Two additional features support this conclusion. First is the discovery of a clay animal igurine on the earthen loor of the delineated space (06/J/050/AR1).22 Second is the elaborately paved red-sherd Floor 98/J/76 adjacent to this installation (Fig. 2.53; Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.27; see also Adams, Chapter 8 in this volume, for further ceramic and stratigraphic discussion). Floor 98/J/76 is bounded on the east by Wall 98/J/9, on the south by Wall 96/J/31, and continues into the northern baulk (Finkelstein, Ussishkin and Peersmann 2006: Fig. 3.27). On the west, the loor comes to an abrupt end even with the western terminus of Wall 96/J/31 (Fig. 2.53, but see Chapter 8). This is also the location of the eastern edge of the northern tongue of the lagstone paved street. It is possible, though not demonstrable, that this street continued to the north but has been completely robbed away. Floor 98/J/76 was originally uncovered in 1998 and the removal of the baulks in 2006 exposed the continuation of the loor to the boundaries described above. The loor occupies two levels of surface separated by a step down from south to north (this step occurs at the precise location of Phase J-6b Wall 98/J/29), from elevation 157.55 m to 157.36 m. The loor itself is composed of a pavement of sherds from more than 18 vessels (most of them red-burnished; Figs. 8.128.14), four of which may have been imported from the Jordan Valley and the Lebanese coast (Greenberg 2006: 165). This contrasts sharply with the rest of the ceramics recovered from the Level J-6 phases (Greenberg 2006; Adams, Chapter 8 in this volume). These elements within the new Phase J-6a space between Buildings B and D should be interpreted as a unit. An entrance on the south leads from the street into a space that is partially segregated by a fence on the west and beautifully paved with extravagant imported pottery on the east. The segregated space is oriented towards a single rounded installation that may have been a plinth for a special object.
22 For animal igurines as EB III assemblages, see Getzov 2006: 9495. He concludes that while these types of igurines may be present in EB II contexts, they do not become widespread until the EB III. Additionally, while cattle igurines are an important part of Early Bronze art and cult, most igurines represent caprines and donkeys. The identiication of the animal in this igurine, however, is dificult to determine.
93
M ATTHEW J. A DAMS
LEVEL J-5/J-6 AND ARCHITECTURAL ELEMENTS UNCOVERED BY THE UNIVERSITY OF CHICAGO Level J-5/J-6 is contemporary with the University of Chicagos Strata XVIIXVI (Fig. 2.55; Loud 1948: Figs. 392393). Several problems occur in any attempt to correlate the Level J-5/J-6 phases of the main sector of Area J with those identiied by the University of Chicago (see also comments in Chapter 8). First, there can be no direct stratigraphic connection between architecture on the upper terrace and the lower terrace. The lower terrace contains Palace 3177 for which the University of Chicago team identiied two major Strata, XVII and XVI, with some minor sub-phases evident in Stratum XVI (e.g., Loud 1948: Fig. 170). One connection between Area J North and the lower terrace is in the style of pavement in Building 3177, Room 1. In Stratum XVII, this room is paved with pebbles (Loud 1948: Fig. 162) while in Stratum XVI, the room is repaved with larger lag-stones (Loud 1948: Fig. 163). This is comparable to the new pavement that the Phase J-6a street received. It is upon this meagre evidence that we generally associate Level J-5 and Phase J-6b with Stratum XVII and Phase J-6a with Stratum XVI. At the base of and against terrace Wall 4045 the University of Chicago uncovered a number of rooms that probably date to the Level J-5 reoccupation of the site (Loud 1948: Fig. 391; Esse 1991: 76). At some point during Level J-6, Wall 4045 was thickened, covering these structures. It seems logical to assign this thickening to the transition between Phases J-6b and J-6a, when Palace 3177 was completely rebuilt (Loud 1948: 7078). There is less dificulty in connecting our Level J-5/J-6 phases with the XVII and XVI strata on the upper terrace, but again, no direct connection exists. Our street is parallel to another paved street located some 30 m to the south (Locus 5215 in Loud 1948: Fig. 393). The University of Chicago street runs from approximately the location of Altar 4017 westwards for more than 50 m. Along the way numerous rooms lank either side of the street. It appears, then, that these streets are contemporary throughout Level J-5/J-6 (Fig. 2.55).23 For the 2010 excavation of this J-5/J-6 architecture, see the appendix to this section below. THE LOCATION OF THE J-5/J-6 TEMPLE The location of the Level J-7 Temple 4040 directly over the Level J-4 Great Temple and the fact that no Level J-5/J-6 (Strata XVIIXVI) temple has yet been identiied raises the question: where is the Level J-5/J-6 temple? Two pieces of evidence shed light on this problem. First, the continuity of cultic tradition on this portion of the mound from the EB I through the Late Bronze Age strongly suggests that a Level J-5/J-6 temple existed in the vicinity. It is reasonable to suggest (though there is no outright proof) that the temple should be sought very close to the location of the Level J-4 sanctuary, especially since that was the location chosen for the Level J-7 Temple 4040. The second piece of evidence lies in the observations that the street and building in the north of Main Area J were constructed with an awareness of the earlier Level J-4 temple walls and that the corridors between these walls were illed to create a platform. The overall plan as we have it suggests that the Level J-5/J-6 temple was located more or less around the central axis of the Level J-4 temple above the mudbrick illed corridors of Squares GJ/1012 (Fig. 2.48). If this is the case, it seems likely that all traces of it have been destroyed in the construction of the Level J-7 temples, leaving round Altar 4017 the only testament to the Level J-5/J-6 cult activity (but see Appendix: The 2010 Season, below).
23 It should also be noted that the University of Chicagos southern street is at a higher elevation. This is why we reconstruct a terrace system creating platforms of occupation. Our street and buildings make one terrace and the mudbrick podium is the next terrace up, with the Chicago street as its southern limit. The rooms south of the Chicago street are at a higher elevation and make up a third higher terrace.
94
LEVEL J-7
The transition from Level J-5/J-6 to Level J-7 marks a signiicant change in the occupation at the site. This transition has been generally underappreciated. All architecture on the site was levelled to make way for a completely new type of construction. Palace 3177 was abandoned and replaced with a monumental staircase (Loud 1948: 7884; Fig. 394). On the upper terrace, the Level J-5/J-6 Street 98/J/136 and Buildings A, B and D in Main Area J were completely dismantled in preparation for the construction of Temple 4040 (Fig. 2.56). Levelling ills were laid around the area, several loci of which contained restorable EB IB pottery, attesting to a major earthmoving operation. The Level J-5/J-6 street in the southern portion of Main Area J was also dismantled and cut by the construction of the dual western temples. The architectural chronology suggests that the transition was abrupt and total (even if the dual-temples were technically later in construction, contra Dunayevsky and Kempinski 1973: Figs. 78; Esse 1991: Fig. 16). The transition is not one of gradual evolution of the town during the EB III. Rather, it is a sign of a complete reorganization of political and religious structure and a major cultural transition. Whereas the nature of the Level J-5/J-6 plan reveals a city engaged in a multiplicity of functions, Level J-7 is monolithic in its attention to the sacred. The University of Chicago excavated the Stratum XV megaron-style temples almost completely. In Temple 4040, they excavated down to the original plastered loor of the temple, leaving it intact. In Temple 5192, the patchy loor was found as was a severely denuded (or uninished?) mudbrick altar. Temple 5269 did not produce a loor or an altar leading to our speculation that it was never completed (see Loud 1948: 7884). Data relating to Level J-7 was obtained in Main Area J over the course of the 2004 through 2008 seasons. A number of levelling ills and foundation trenches were excavated relating to Temple 4040 (Level J-7). In 2010, we exposed the western wall of Temple 5269 in Sub-Area Upper J (see Part IV). RELATIVE CHRONOLOGY OF THE LEVEL J-7 MEGARON-STYLE TEMPLES Loud (1948) published no statement indicating that he and other members of the team thought anything other than that all three temples were constructed simultaneously. The fact that the twin temples were constructed separately (physically, though not necessarily temporally) from Temple 4040 is clear enough from both their orientations and their slightly different construction techniques. Temple 5192 has a continuous stone foundation within the threshold24 whereas Temple 4040 does not (see below). The altar of Temple 4040 was made of ieldstones, whereas that in Temple 5192 was made of mudbrick. The size, proportions and construction grids of the structures are different (Loud 1948: 78; de Miroschedji 2001: 483). Additionally, the pillar bases in Temple 4040 have an average diameter of 53 cm (one 5.25-cm cubit) at the bevel whereas those of the twin temples have an average diameter of 63 cm.25 Kenyon (1958) suggested that the twin temples were built irst and that Temple 4040 cut a portion of the earlier construction. To test this hypothesis, Dunayevsky and Kempinski excavated a few sondages at the site (1966; 1973). They concluded that Temple 4040 was built before the other two on the basis of a wall (A) appended to Temple 4040, which Temple 5192 clearly put out of use (Dunayevsky and Kempinski 1973: 162167; Esse 1991). The
24 My observation in the ield. 25 Incidentally, this would also suggest that these bases were made for the respective temples at the time of construction, rather than having been robbed from earlier structures, such as Building 3177 (contra Ussishkin in Finkelstein and Ussishkin 2000b: 590591).
95
M ATTHEW J. A DAMS
renewed excavations checked afresh Wall A in 1994 and determined that it was not bonded to Temple 4040 and therefore does not by itself provide the stratigraphic evidence that Dunayevsky and Kempinski claimed (Finkelstein and Ussishkin 2000: 7071). To this we can add that with our latest exposure of Level J-6, we now have a good understanding of the general orientation of its architecture, and it would appear that Wall A is better oriented with Level J-5/J-6 than with Temple 4040, and is cut by the construction of Temple 4040 in the same way the rear wall of Temple 5192 cuts into the Level J-5/J-6 architecture to its south (Fig. 2.72). Dunayevsky and Kempinski used their interpretation of Wall A to push back the construction of Temple 4040 to Stratum XVII (Dunayevsky and Kempinski 1973: Fig. 7), contemporary with Palace 3177. This was followed by Nigro (1995: 1623). Based on his metrological analysis, de Miroschedji suggested that the palace and temple were built as a single operation (de Miroschedji 2001: 485 and n. 10).26 This can now be disputed based on the three phases of Level J-6, which match well the full chronology in the rest of the site. Palace 3177 is contemporary with the buildings attached to streets, both of which were put out of commission by the construction of the Level J-7 temples; Level J-5/J-6 ended at the same time everywhere on the site and Level J-7 marks a completely new era at Megiddo. Ultimately, there is no basis for believing that there was any signiicant temporal distinction between Temple 4040 and the other two temples (see also Appendix: The 2010 Season). TEMPLE 4040 New evidence relating to Temple 4040 (Figs. 2.56, 2.59) from the renewed excavations between 2004 and 2008 consists of three separate entities: 1) ills over the Level J-5/J-6 architecture, much of which has been contaminated by Middle Bronze Age pits and modern mixing (encountered in most squares); 2) the excavation of the foundation stones (08/J/73) of western portico Pillar Base 08/J/214; and 3) the excavation of the clearly delineated foundation trench beneath the threshold of Temple 4040 in the southern incision of Square E/9. The removal of the western porch Pillar Base 08/J/214 of Temple 4040 revealed the stone foundations (08/J/73) supporting the massive stone (see Part I of this chapter). The foundation penetrated 0.66 m beneath the bottom of the base and was constructed with a boulder core with additional medium- to large-sized stones for support (Fig. 2.3). The foundation column did not have a diameter larger than the base itself; all of these stones were carefully laid and not simply dumped into the foundation trench, providing a solid foundation for the ca. 2.5 ton column base. The foundation penetrated down through a corner of one room of a Level J-5/J-6 building and into the Level J-5 ill over the Level J-4 temple, but did not reach the latter temple itself. In preparing to remove the Temple 4040 pillar base, the space between the western base and the eastern base was cleaned to re-expose and plan the stone wall that linked the two porch pillar-bases (seen and partially planned by the University of Chicago team; Loud 1948: Figs. 186, 394), to which we assigned the designation Wall 08/J/12 (Fig. 2.56; Loud 1948: Fig. 394). Loud does not discuss these stones, but the plan
26 It should be noted that Building 3177s orientation is 18 east of Chicagos north while Temple 4040s orientation is 21 east of north. The similarity of orientation is not exact enough to have been a product of contemporary planning and must be a coincidence. Rather, the orientations of both structures were determined more or less by the underlying Level J-4 architecture, particularly Wall 4114 (see above).
96
and photograph suggest that they were at least partially covered by the lime loor of Temple 4040 (Loud 1948: Figs. 186, 394). The wall consists of three rows of parallel stones (Fig. 2.56). Visual inspection in the ield and of the plans indicates that the southernmost of these rows lines up perfectly with the northern face of the porch pillar-bases which, despite their differences in size and shape, have their northern faces in alignment (this line is also shared by the northern terminal ends of the western and eastern walls of the temple).27 The second row north is only partially preserved. The southernmost and middle rows are composed of medium-sized ieldstones. The northernmost row is made from much larger lagstones. All of the stones have approximately the same elevation of ca. 158.32 m. One possible explanation is that these stones represent a stone paved street leading to the temple. This paved street perhaps terminated in the north (near the end of the fence Wall 06/J/2) at the top of a staircase that ascended the slope of the tell from the north to Temple 4040 (compare Staircase 5263 in Stratum XII, Loud 1948: Fig. 398; see also Fig 2.69 in this chapter). This would be congruous to the monumental Staircase 3160 on the eastern slope (Loud 1948: Fig. 394). Near the end of the 2008 season, some questions remained about the relationship between the Level J-4 sanctuary altar and the rear wall of the sanctuary (see above). To address this issue, an incision was made through the threshold of Temple 4040 down to the Level J-4 temple (see also above). This probe extended 2 m south from the southern edge of the shortened Square E/9, positioned 1.5 m wide, centred within the Level J-7 Megaron 4040 threshold (Fig. 2.56). This incision cut through the Level J-7 temple Floor 08/J/133 and revealed the foundation trench for the Temple Wall 08/J/40 (Figs. 2.562.58). The foundation trench (08/J/147) is 1.25 m deep (from the plaster loor) and is ca. 0.90 m wider than the thickness of Wall 08/J/40.28 While this certainly represents a continuous foundation trench, the continuation of Wall 08/J/40 was not therein. Rather, the wall foundations appear to have stopped at the jams and this portion of the unused trench was simply back-illed in preparation for the tan/brown plaster loor. The sealed ceramics from the wall foundation and the portico pillar base foundation provide a terminus post quem for the date of construction of Temple 4040 in the terminal EB III or perhaps the onset of the Intermediate Bronze Age. New information relating to Temple 5269 is reported in the Appendix to Part III and in Part IV of this chapter.
27 The possibility that these stones comprise a stylobate for the column bases was checked in the ield. These stones were not connected to the actual foundation of the bases. It is also not clear if a similar row of stones was present in Temple 5192. 28 This is an estimate since the entire width of the foundation was not exposed. The north side of the foundation trench extends 0.45 m beyond the face of the wall.
97
98
Fig. 2.56: Remains of Level J-7 (in black elements uncovered by the University of Chicago team but not drawn by us).
M ATTHEW J. A DAMS
Fig. 2.57: Square E/9 southern incision eastern section. Note lighter coloured plaster loor of Temple 4040 at surface.
Fig. 2.58: Square E/9 southern incision eastern section.
99
100
Fig. 2.59: Aerial photograph of the Level J-7 temples at the end of the 2008 season, facing south. Note Upper Area J Squares TS/20 at right.
M ATTHEW J. A DAMS
APPENDIX TO PART III: THE 2010 SEASON

The 2010 season in Area J was initiated with four major goals. The irst was to assess the levels underlying the Level J-4 Great Temple. To accomplish this, we reopened Squares H/7 and H/8 to provide a section through the loor of the temple. Second, we wished to sample the Level J-4 activity south of the Great Temple to assess the prospects for future work in this sector of Area J; two squares were selected in the southern portion of Area J, behind the twin Level J-7 temples in antis. Third, previous seasons demonstrated that Level J-5/J-6 was a signiicant elite compound (see Part III of this chapter), and we wished to explore it further by cleaning and examining those levels exposed by the University of Chicago in the southern portion of the area. Fourth, we wished to inish our excavations in the upper part of Area J in order to complete our stratiication of the Early and Middle Bronze Ages (see Part IV of this chapter). The results of the irst three endeavors of the 2010 excavations are reported here, while the 2010 excavations in the upper portion of Area J have been incorporated into the report in Part IV of this chapter. Midway through the 2010 season, the directors and I consulted about the future prospects of excavation in Area J. Since many of our major goals for this part of the mound had been satisied (see Finkelstein and Ussishkin 2000a: 3034), a decision was taken that the 2010 season would be our last in Area J.
SONDAGE BENEATH THE LEVEL J-4 TEMPLE

As one of three foci of investigation in the 2010 season, we continued excavation in Squares H/7 and H/8 and the H/8H/9 baulk down beneath the loor of the Level J-4 Great Temple in order to clarify the underlying stratigraphy. We established that the Great Temple was not built on a large ill at this point. Rather, the underlying stratigraphy had been cut lat to prepare for the construction of the temple. Beneath we revealed four phases of construction and occupation dating to the EB I. It is now abundantly clear that the pre-J-4 levels on the mound are not insigniicant, and, while the Great Temple represents a leap forward in monumental construction, it is perhaps not as punctuated as we once thought. No stratigraphic link can be made between the stratigraphy beneath the Great Temple and Levels J-1 to J-3 on the eastern slope. Therefore, we have numbered the stratigraphy here from youngest to oldest relative to J-4: i.e., the phase beneath is -1, then -2, and so on (Table 2.7). While bedrock was not reached in these squares, it appears that pre-J-4 activity was more intense here than on the eastern slope. At the end of the 2010 season, the squares of this sondage were backilled in order to prevent erosion and preserve the integrity of the Level J-4 and Level J-7 temples. We have marked the bottom of the backill with imperishable material.
101
M ATTHEW J. A DAMS
Fig. 2.60: Composite plan of the remains of Level -4 Phases d through a beneath the Great Temple.
TABLE 2.7: STRATIGRAPHIC SEQUENCE BENEATH THE LEVEL J-4 TEMPLE

Level/ Phase J-4 -1 -2 -3 -4 a -4 b -4 c -4 d Description Great Temple, Terrace Walls 4045 and 4114 Wall 10/J/56 with French Drain 10/J/168 Greenish Fill 10/J/23 = 06/J/91 with Stone Pillar Base (?) Greenish-clay-coated Circular Platform 10/J/152; Wall 10/J/29; Floor 10/J/58 Rounded revetment Wall 10/J/53; Wall 10/J/55, Ash and Bone Accumulation 10/J/151 Rounded revetment Wall 10/J/53; Hearth 10/J/188; Accumulation 10/J/191, 10/J/183 Rounded revetment Wall 10/J/53; Hearth 10/J/188 Rounded revetment Wall 10/J/53 Period EB IB late* EB IB late EB IB late EB IB late EB IB early EB IB early EB IB early EB IB early
* The ceramics from this sondage will be published in a future report. Sufice it to say at this time that the sequence presented here appears to stretch from the middle of the EB I through the end of the EB I. The use of the terms early and late here is intended only for relative purposes, and not intended to favour any particular EB I ceramic chronology. Further study is required to determine what contribution this assemblage can make to the internal chronology of the EB I.
102
Fig. 2.61: Phase -4 b Surface 10/J/183 facing east. Phase -4 d/c Hearth 10/J/188, at left; Wall 10/J/53 at bottom. Compare eastern section with Fig. 2.37.
LEVEL -4 (FIG. 2.60) The earliest architecture and occupational debris reached in the sondage consisted of a slightly curving revetment wall (10/J/53) that remained a feature of this area through four sub-phases, Phase -4 d through a. The single-row wall was constructed with its eastern face neatly arranged sloping lightly to the west against the debris (10/J/175) that it was intended to support. The back side of the wall was left jagged. It is not clear if this revetment was intended to separate two levels on the northern slope, or formed part of a mound-like installation. The bottom of this wall was not reached, but it stands at least 0.81-m-high. The hearthy deposits at the lowest portions of excavation in Square H/7 (see Fig. 2.37) perhaps represent the earliest occupational material associated with this feature. If this is the case, the revetment feature stood over 1.14 m above its original occupation, Phase -4 d. In Phase -4 c, a stony ill (10/J/196) supported the construction of a large stone hearth (10/J/188), consisting of a stone superstructure enclosing a circular pit 59 cm in diameter. Within the pit there was evidence of in situ ire and a 15 cm accumulation of large charcoal pieces (Fig. 2.61). Charcoal was also evident across a surface south of the hearth (10/J/194; Fig. 2.37). Phase -4 b witnessed the continued accumulation of occupational debris south of Hearth 10/J/188 and east of Revetment 10/J/53. Wall 10/J/54 was added in this phase, perhaps to segregate the area of Hearth 10/J/188 from the area to the south. The surface Accumulation 10/J/191 culminates in thick ashy hearth deposits (10/J/183; Fig. 2.61). Finally, in the last phase of revetment Wall 10/J/53, Phase -4 a, Wall 10/J/55 was added and ash and bone (10/J/151) accumulated directly to its south. This ash and bone accumulation is consistent with cultic activity through all periods of the site where the temple area continually witnessed the accumulation of sacred garbage (Loud 1948: passim; Epstein 1965: 208213; Wapnish and Hesse 2000; Finkelstein and
103
M ATTHEW J. A DAMS
Ussishkin 2000a: 5565; see also Level J-4, below). This accumulation partially overlays Wall 10/J/53, suggesting that the preserved height of 155.62 m ASL represents the approximate original height of the revetment feature. East of the ash and bone deposit, an amorphously preserved group of exceptionally dense mudbricks (10/J/170) suggested itself as some sort of installation. Further support came from patchy evidence of a pinkish clay coating extending westward beneath Ash 10/J/151. A curious vertical hole in the northern side was found, 34 cm in diameter and over 10-cm-deep, illed with a light gray ash (Fig. 2.60). Hearth 10/J/188 appears to have gone out of use (based on levels), but this could not be veriied by stratigraphic connection to the other elements of this phase. LEVEL -3 (FIG. 2.62) This phase witnessed a refreshment of the architectural scheme of the preceding levels. Wall 10/J/53 was rebuilt as Wall 10/J/46 over ash deposit 10/J/151. To the east, a new large curvilinear feature was constructed consisting of curving, stone-based Wall 10/J/29 (one course high) that acted as a step up to stone Platform 10/J/152 (one course high) (Fig. 2.63). Wall 10/J/29 and Platform 10/J/152 were both coated with a thick mud mortar (10/J/42, 10/J/83, respectively). Surface 10/J/58 extended north and west of Wall 10/J/29. The loor, wall and platform were all coated with a greenish coating (see especially Fig. 2.37) to create a circular stepped structure. A group of stones in the southeast corner of Square H/7 may be the remains of a wall that bound the greenish sloping loor 10/J/58 (Figs. 2.37, 2.62). The calcareous greenish coating is similar to mudbricks found on the site in the Early Bronze and Middle Bronze (Levels J-8, J-9; apparently also rampart bricks in Area K). The source for this material appears to be a major calcareous marl deposit of the Mount Scopus Group (Senonian/Paleocene; Sneh et al. 19981) surrounding the spring at Ain el-Qubbi on the northeastern edge of the mound (Adams, David and Homsher forthcoming) and stretching around to the south side of Megiddo.2 The greenish coating over the circular platform was applied wet in one complete coating, not built with bricks. However, there is some evidence that elsewhere in the building, perhaps at the top of the circular structure beyond the southern section, there existed a structure built with bricks composed of the same calcareous material. These bricks were used as a levelling ill in the succeeding phase. LEVEL -2 (FIG. 2.64) This layer is almost completely obliterated by the level-cutting operation of Level J-4 (see below). Evidence of it can be identiied by two features. The irst is a thick greenish ill (10/J/23) that covers most elements of the preceding phase (see Fig. 2.37). This ill is made of bricks composed of the greenish calcareous material discussed above. This ill was also noted in Square G/7 as the greenish Fill 06/J/91 (see discussion of the Level J-4 loor above). In several places, brick lines were clearly identiiable in this ill, suggesting that it was composed of bricks dismantled from the Level -3 architecture associated with the circular stepped structure. The western portion of this ill was mixed with a pile of small- to medium-sized ieldstones (10/J/49), perhaps from the dismantling of Wall 10/J/46 or another Level -3 structure. The second element of this phase is a large lat stone in the northeast corner of Square H/8, which appears to be coexistent
1 2 http://www.gsi.gov.il/Eng/Index.asp?ArticleID=172&CategoryID=119&Page=1 This same deposit was also the source of the Late Bronze Amarna letters from Megiddo (Goren et al. 2004: 246). Note that there it is identiied as the Maastrichtian Ghareb formation. The material closer to the spring is probably part of the Taqiye formation of the same Mount Scopus Group (Y. Goren, personal communication).
104
Fig. 2.62: Remains of Level -3 beneath the Great Temple.
Fig. 2.63: Level -3 Wall 10/J/29 and Podium 10/J/152, facing east. Note Floor 10/J/58 in section at right (compare with Fig. 2.37). Level -4 architecture appears beneath.
105
M ATTHEW J. A DAMS
Fig. 2.64: Remains of Level -2 beneath the Great Temple.
with greenish ill deposit. As most of this building has been cut away by Level J-4 activity, one can only speculate that this new architecture represents a shift from the preceding circular structures/installation to a larger building with a pillar(?), more in keeping with the later Level J-4 architecture. LEVEL -1 (FIG. 2.65) This phase is deined by wide foundation Trench 10/J/168 constructed to receive Wall 10/J/56 (Figs. 2.37, 2.652.66). Chalk paste was preserved on the northern face of the brick-on-stone-socle wall, indicating that the interior of the building to which it belonged was to the north. South of the wall, the oversized foundation trench was lined with a clean silty mud. The side of the wall within the trench was coated with a mud-withstraw-binder coating (no chalk paste). The trench was then backilled with small- to medium-sized stones and a sandy earthen ill (10/J/168). Given these preparations, it appears that the purpose of the foundation trench was to facilitate drainage from up-slope to the south; this is one of the earliest examples of a French drain, a common drainage solution for retaining walls. A similar retaining wall drainage technique may have been in use behind Wall 94/J/2 of the Level J-3 temple on the eastern slope, which also served as a terrace for material up-slope (Fig. 2.20; Finkelstein and Ussishkin 2000a: 4042, Fig. 3.15). This would be a tenuous link, at best, for the contemporaneity of Level J-3 on the eastern slope and Level -1 here. Because of the later Level J-4 construction activity, it is dificult to determine if the foundation Trench 10/J/168 cut through Level -2 green ill, and is therefore later, or if the poorly preserved structure of Level -2 was intended to be the building terraced by Wall 10/J/56, and is thus contemporary. I prefer the former. The last two phases, Levels -2 and -1, before the construction of the Great Temple with their thick linear walls and the possible pillar base, perhaps represent the beginning of broad-room-style temple architecture at the site and may be contemporary with Level J-3 on the eastern slope.
106
Fig. 2.65: Remains of Level -1 beneath the Great Temple (University of Chicago-exposed walls in grey).
Fig. 2.66: Level -1 Wall 10/J/56 and French Drain 10/J/168, facing east. Note Level J-4 temple Wall 00/J/21 at left with bricks of Wall 10/J/56 below and chalk coating of northern face of Wall 10/J/56 at far left. Stone socle of Wall 10/J/56 is evident beneath the string marking the southern face of the wall. French Drain 10/J/168 is evident in the east section (compare Fig. 2.63). Stones at base of the wall are the lower portions of the stone ill of the drain.
107
M ATTHEW J. A DAMS
Fig. 2.67: Remains of pre-Level J-4 curvilinear Wall 10/J/62 in M/15. Note bedrock level indicated beneath wall.
LEVEL J-4 (FIGS. 2.222.41) Very little new material was added regarding Level J-4. The Great Temple was constructed by cutting the previous occupation down to a level construction surface rather than building it up with terraces (Fig. 2.37). Our assessment of the Level J-4 loor (above) was conirmed with the discovery of the Levels -1 to -3 architecture beneath. In that assessment we concluded that the loor of the temple (08/J/202) was simply thin compact mud laid to smooth the irregularities in the underlying cutting. Logically, this also conirmed that the foundation deposit slab (Figs. 2.392.40), which was found above the green ill but beneath the temple loor, was laid during the construction process of the Great Temple, i.e., after the area had been cut lat to prepare for construction of the temple. This cutting procedure eliminated many aspects of the underlying strata, and it is not possible to determine how many strata have been completely eradicated on this portion of the mound. Similarly, it is dificult to determine if the Level J-3 temple on the eastern slope immediately preceded the Level J-4 temple or simply represents the elevation to which Level J-4 cut down.
108
EXCAVATIONS SOUTH OF THE LEVEL J-7 TEMPLES

Work commenced in Squares O/13 through J/16: Squares M/15 and M/14 were tasked with reaching Level J-4 while work in the other squares served to clarify architectural details of Level J-5 and J-6 (Figs. 2.682.72). BEDROCK Bedrock was reached in Square M/15 at 155.23 m ASL (Figs. 2.672.68). This complements our two adjacent bedrock level data: 1) Schumacher reached bedrock at ca. 153.00 m ASL in approximately Square P/15 on our grid; 2) we reached bedrock at 155.24 m ASL in Square J/13 in the 2000 season (Finkelstein, Ussishkin and Peersmann 2000: 36). Overall, these data indicate relatively level underlying bedrock in the vicinity of Level J-7 Temple 5192. At some point between Squares M/15 and P/15, the bedrock begins to descend more steeply. In Square M/15, no evidence of human activity on the bedrock was present. PRE-LEVEL J-4 In Square M/15, Debris 10/J/193 overlaid the bedrock, yielding EB IB pottery (Fig. 2.68). Above this was an occupational level (10/J/181 = 10/J/187) with restorable EB I pottery not associated with any architecture. This occupational debris was built over by curvilinear Wall 10/J/62 with no clearly preserved occupational surfaces (Figs. 2.672.68). How these elements correspond to other levels (Level J-4 and earlier) around the site is unclear. LEVEL/PHASE J-4/J-4a Squares M/15 and M/14 both reached what we assume to be Level J-4 material in the form of extensive bone (with many articulations in each locus) and ash deposits (10/J/160; Fig. 2.68). No architecture could be associated with these bone deposits, but in Square M/15, the bones appeared to be part of a rubbish pile that included ash and stones. The tip lines of this deposit suggested that the material had been thrown from the east/southeast. Without stratigraphic linkage to other sectors of the site, and given the fact that some bone deposits or the disturbance of these deposits may come from Phase J-4a (see above), we cannot conclusively associate this material with Level J-4. However, the fact that a Level J-5 ill overlays the ash and bone deposits directly narrows the attribution to one of the two. Perhaps signiicantly, none of the pre-Level J-4 intense stratiication known from the eastern slope and from beneath the Great Temple is to be found here. This matches the results from previous seasons in, for example, Square J/13 (Finkelstein, Ussishkin and Peersmann 2000: Fig. 3.14). The emerging picture (though far from complete) is that in the EB IB (Levels J-3 and J-4, in particular) the area south of the Level J-4 Great Temple and west of the Level J-3 temple was an open space with a variety of installations (e.g., Wall 00/J/01, plaster Feature 00/J/121, the burial of the complete bos primigenius; Finkelstein, Ussishkin and Peersmann 2000: Fig. 3.14) and signiicant bone deposits (e.g., 98/J/21 and 98/J/28 in Squares CD/1314, Fig. 2.24; 10/J/138, 10/J/160, 10/J/189 in Squares M/1415, Fig. 2.68). In both Levels J-3 and J-4, this area was bounded by a terrace wall on at least one side: Wall 94/J/02 (Fig. 2.20) and Wall 98/J/23 (Fig. 2.24), respectively. Given this evidence and the signiicant number of bones associated with these levels, it is tempting to identify this open area as the EB I communal feasting area.
109
110
Fig. 2.68: Eastern section drawing of Squares M/1415 showing stratiication from Level J-6 to bedrock.
M ATTHEW J. A DAMS
Fig. 2.69: Remains of Level J-5 south of the Level J-7 temples.
LEVEL J-5 FILL (FIG. 2.68) Characteristically, everywhere in Area J where Level J-4 was encountered, it was covered by a thick bricky ill deined by small sherds and occasional chunks of plaster (e.g., 08/J/129 = 08/J/212 in the sanctuary, Figs. 2.37, 2.48; see Level J-5, below). This is also the case in Squares M/1415, where the ash and bone deposits are covered by a dense bricky deposit over 1-m-thick (Fig. 2.68). As elsewhere, this ill comes from the dismantling of Level J-4 and the levelling operations carried out at the beginning of the Level J-5 reinhabitation of the site. Within this ill, Square M/14 produced a thick deposit of brick material and larger pieces of EB IB ceramic material, some restorable, in large quantities (08/J/13310/J/171; Fig. 2.68). As part of the Level J-5 levelling operation, this material represents J-4 occupational debris removed from elsewhere on the mound and redeposited with the bricky ill, and may represent the best Level J-4 ceramic assemblage to date. LEVEL J-5/J-6 (FIGS. 2.692.72) The nature of Level J-5/J-6 to the south of the Level J-7 temples is similar to that excavated to the north (see Level J-5/J-6, above). Here, too, at least three major phases were identiied Level J-5, Phase J-6b and Phase J-6a which appear to correspond roughly to the three phases identiied elsewhere (Figs. 2.692.72).
111
M ATTHEW J. A DAMS
Fig. 2.70: Remains of Phase J-6b (basalt slabs in black) south of the Level J-7 temples (in light grey).
Similar to the area north of Temple 5192, this phase is deined by an east-west street that also acted as a terracing feature: the buildings to the north of the street are lower than those to the south (see Figs. 2.29, 2.48). The street was originally unpaved, and through the course of Level J-5 and Phase J-6b, there was signiicant refuse build-up within it (Fig. 2.68). At some point in Phase J-6b, this street (like the one to the north) was paved with lagstones (10/J/24). The street slopes down to the west (the slope occasionally demanding a step) where we traced it as far as the edge of the University of Chicagos Area BB. The EB III city, therefore, extends at least some distance in that direction, perhaps as far as the boundaries of the mound. To the east, the street passes south of Altar 4017 (see Fig. 2.48), leading to the entrance of the altar enclosure and the buildings on the eastern edge of the upper terrace. Primarily, the Level J-5/J-6 architecture exposed by the University of Chicago was cleaned to prepare a new plan. Only in Squares M/14, M/15 and L/15 were the phases dissected. For the most part, the results of this seasons operation can be deduced from the plans presented here (Figs. 2.682.72). Otherwise, the following observations can be made: Level J-5: The area slopes down to the north and west. Many of the living spaces were terraced from room to room. In some cases different levels in rooms could be identiied by the presence of terrace architecture, e.g., in Fig. 2.69, Wall 10/J/13: the parts of the wall that were subterranean on one side were constructed with irregular faces. A doorway through Wall 10/J/30 was identiied by its door socket, giving some idea of the internal layout. While we have little preserved architecture from Level J-5, 2030 cm of occupational material remained in several places.
112
Phase J-6b: This level is the best exposed (Figs. 2.70, 2.73). To the north of the street a large, ine building was constructed (which apparently existed into Phase J-6a). Its walls are thicker than most of the other walls of this level, suggesting either a second story or superior construction procedures. Compared to the other Level J-5/J-6 architecture, this building is certainly of higher quality (cf. Figs. 2.512.53). Unfortunately, the northern portions of the building have been cut away with the construction of the Level J-7 temples. This building was already exposed by the University of Chicago and only cleaned by us. Room 5221 appears to be the focal point of the building its walls and loors are well plastered (as was Room 5222), and a pillar base (no longer in situ since its exposure) stands in the east-west centre of the room. Room 5199 in the southwestern corner of the building contains a drain through Wall 10/J/23 into the street. To the south of the paved street is another large building (Fig. 2.70). The layout is fairly clear as excavated by the University of Chicago team (that removed most of the loors), but we revealed new features in Squares L/15 east of Wall 10/J/02 (Figs. 2.702.71). This room featured a series of loor levels. With the earliest loors (10/J/120, 10/J/74) the room featured at least two basalt slabs, whose purpose is unknown. The room was entered through two doorways: from the west through Wall 10/J/02 and from the north through Wall 10/J/22. The walls were coated in a ine plaster. With the raising of the loor to Floor 10/J/51, the basalt slabs were covered. The loor in the adjacent western room was not raised, and a basalt step and door socket were added to the west entrance. The north entrance also received a step (Wall 10/J/24) and a socket. The northern step led to newly constructed stone and plaster Pavement 10/J/82. A basalt grinding stone installation was established on the south side of the room. In all three loor levels, ashy deposits characterized the loor remains. Phase J-6a: This phase is characterized by the continuation of the lagstone pavement and the remodelling of several portions of the Phase J-6b buildings (Fig. 2.72). The far western portion of the street was expanded and Wall 10/J/60 gives evidence for an additional building. A yellowish clay-like plaster was found to line both the street pavement and Wall 10/J/60, but it is unclear if this was a feature of the entire street or just this location. The basalt-slab room in Square L/15 was apparently illed with a thick bricky ill, which raised the loor level signiicantly. The room was eventually paved with ist-sized river pebbles (10/J/04), a feature known also from Palace 3177 on the eastern terrace. The addition of blocking Wall 10/J/28 over former Pavement 10/J/82 suggests a redirection of the low of trafic though the building. Much of the rest of the building south of the street was destroyed by later construction (including a deep stone drain that cut through pavement 10/J/04 Stratum XII?). In Square H/13, a brief probe revealed a holemouth jar installation (10/J/69) containing a lamp type characteristic of Level J-5/J-6 (see Chapter 8, Fig. 8.6: 111) and a juglet. Elevation suggests that it belongs to Phase J-6a.
113
114
Fig. 2.71: Square L/15 northern and eastern section drawings.
M ATTHEW J. A DAMS
Fig. 2.72: Remains of Phase J-6a (basalt slabs in black) south of the Level J-7 temples (in light grey).
115
116
Fig. 2.73: Plan of the EB III city in Phase J-6b/a with suggested reconstruction of Level J-6 Temple 5221. Dark grey indicates walls seen by the University of Chicago team as Stratum XVI, but not re-excavated by us. Lightest grey indicates underlying Stratum XVII or our Phase J-6b walls which complete the overall picture. Note also in light grey Walls 4114 and 4045 which continued in use as terrace walls from Level J-4.
M ATTHEW J. A DAMS
LEVEL J-5/J-6 SUMMARY The ceramic assemblage from this portion of Level J-5/J-6 was similar to that in other sectors of the area. Speciically, pattern burnished platters are common and Fine Pinkish Fabric (see Chapter 8) bowls/lamps abound (Figs. 8.48.16). The Level J-5/J-6 assemblage presented in Chapter 8 should be considered representative of the material recovered in the 2010 season. Overall, Level J-5/J-6 appears to represent a large, well-planned city. The eastern terrace supported a palatial building (3177) while the western portion was constructed on an orthogonal grid with streets. The buildings here were probably a mixture of public ediices (e.g., Altar 4017; Temple 5221, see below) and elite residences. In addition to the original planning in Level J-5, the universal paving of the streets with lagstones indicates centralized organization of public works projects. THE LEVEL J-6 TEMPLE (FIG. 2.73) A long-standing question regarding Level J-5/J-6 (Strata XVIIXVI) has been the location of the contemporary temple(s) (see above; Finkelstein and Ussishkin 2000b: 588). Given the earlier and later use of space in this sector of the mound and the location of Altar 4017, it is likely that a Level J-5/J-6 cultic ediice is to be found within the current limits of Area J (though a location south of the altar beyond the southern section cannot be ruled out). Building 5221 (Loud 1948: Fig. 393, northwest of Square N/12) presents us with the best candidate for such a building. Figure 2.73 provides a reconstruction of the EB III city at its most prosperous moment, with Palace 3177 on the eastern slope and the reconstructed Temple 5221 in Squares LM/1214. The overall design of this building is typologically similar to the later Level J-7 temples in antis, not following the tradition of the Level J-2 through Level J-4 broad-room temples with ritual tables (for this typological distinction, see Adams, Finkelstein and Ussishkin forthcoming). Therefore, our reconstruction of the width of the building is based on the general metrology of the later Level J-7 temples, where the intercolumnar space is twice the length from the exterior wall to the column. The main sanctuary, then, has internal dimensions of 7.5 m by 7 m with two columns. The loor and walls are plastered and a stubby partition wall (10/J/18) juts into the room. The sanctuary is fronted by an anteroom. The overall dimensions of the reconstruction are 11.3 m by 8.5 m. The contemporary Khirbet ezZeraqon temple provides parallels for the basic structure and for the general spatial relationship to a round altar (Ibrahim and Mittmann 1994). We have, therefore, borrowed the concept of pillars on a continuous foundation lanking the anteroom entrance for this reconstruction. Between reconstructed Temple 5221 and Altar 4017 there is much evidence for cultic activity. Locus 98/J/152 in Square J/13 (Finkelstein, Ussishkin and Peersmann 2000: 42, Fig. 3.22) consisted of a highly burnt ash and bone accumulation on a loor just before the entrance of the temple. The accumulation included such pottery as a cult-stand (Greenburg 2006: Fig. 10.2: 12, erroneously attributed to Level J-4). In adjacent Square G/13, a large amorphous plaster installation with grooves was also surrounded by ash and bone deposits (Finkelstein, Ussishkin and Peersmann 2000: 42, Fig. 3.22). The overall impression of this material is of an open cultic activity space. LEVEL J-7 (FIG. 2.56) Portions of the thick wall (10/J/11) behind the twin temples in antis were planned and removed as part of our exposure of Level J-5/J-6. We conirmed what appeared to be the case in the University of Chicagos
117
M ATTHEW J. A DAMS
plan: that this wall is thicker behind Temple 5192 and thinner behind Temple 5269 (Fig. 2.56, Square M/14); the reason for this is not clear, but it may be related to the fact that the rear wall of Temple 5269 was thicker than the rear wall of Temple 5192. We also recognized that Temple 5269s back wall is supported by a rough wall (10/J/65) leaning against the building (previously assigned by the University of Chicago team to Stratum XVI). The same feature occurs on the temples west side (10/J/61). This is the only one of the three temples with such a feature. It might be related to the fact that the temples were built on a slope down to the west and that this temple was constructed on a steeper sector of the slope and required sturdier foundations (note also that the rear wall of Temple 5269 is thicker than any of the other walls of the three temples). The slope is also evident from the bottom levels of the temple walls proper, where Wall 10/J/44 (down slope) of Temple 5269 is founded over 1 metre deeper than the western wall of Temple 5192. While Temple 4040 has fairly consistent foundations, each of the four north-south walls of the twin temples had to be progressively stepped down westward. It is possible that the support walls along the southern and western walls of Temple 5269 were a last-ditch attempt by the builders to support an unstable construction other evidence suggests that this temple was never completed (see above; Loud 1948: 78). The stratigraphy of these three temples has been called into question by other investigators (Dunayevsky and Kempinski 1973; Kempinski 1989; de Miroschedji 2001; see also Level J-7, above). Essentially, these attempts at re-stratiication have tried to place Stratum XV Temple 4040 back into Strata XVIIXVI, contemporary with Building 3177. In our excavations all three temples cut into and overlay Level J-5/J-6 (Stratum XVII-XVI) architecture everywhere investigated (above; Finkelstein, Ussishkin and Halpern 2006: 846847). De Miroschedjis argument (2001: 483485) that Building 3177 and Temple 4040 were laid out according to the same orientation and metrological module, was rechecked. We found that their orientations actually differ by 3 to 5 degrees depending on which walls in Building 3177 were chosen. The orientation of Building 3177 is dictated by the preexisting terrace Wall 4045. As for his argument of the metrological module, while the 52.5-cm cubit does appear to have been in use during the layout of these buildings, we could not replicate any satisfactory results with his 11-cubit module on Stratum XV Temple 4040 or on Building 3177. Thus, de Miroschedjis arguments do not stand and there is no connection between Building 3177 and Temple 4040. Ultimately, the renewed excavations have demonstrated that the three Level J-7 temples in antis comprise a single stratum constructed after the total clearing of Level J-5/J-6, as already indicated by the original excavators.
118
AREA J, PART IV: SUB-AREA UPPER J
PART IV: SUB-AREA UPPER J

Matthew J. Adams and James M. Bos
Since the University of Chicago completely excavated most portions of the Level J-7 temples, the northwestern corner of Temple 5269 was the only area in which their stratigraphy could be rechecked (Finkelstein and Ussishkin 2000: 71). Another advantage of excavating this sector was the possibility of rechecking the University of Chicagos post Early Bronze Age stratigraphy in Area BB. In 2006 we undertook this project in two squares directly over this corner of the temple, in Upper Area J, Squares S/20 and T/20 (Figs. 2.12.2).1 This work was completed in 2010. As a result, the stratigraphic sequence of Area J now extends from the Early Bronze Age through the Late Bronze Age (see Table 2.1). This section reports on the Middle and Late Bronze Age architecture and stratigraphy beginning with the Level J-7 Temples. Sub-Area Upper J is situated in a stratigraphically and topographically precarious position straddling the University of Chicagos Areas BB and DD (Fig. 2.1). Area DD was intended to connect the stratigraphy of Area AA with that of Area BB (Fig. 2.1; Loud 1948: 113114). This was attempted with broad exposure of Square K/11 (of the University of Chicago expedition grid) and its surroundings. Later, a southern trench was extended from Area DDs southeastern corner to connect to Area BB, more or less along the line of Schumachers north-south trench. This was never completed and Loud (1948: 116) reports that excavation reached no deeper than Stratum VIA. The southernmost limit of the Area DD as indicated on the Stratum VIA plan is an east-west wall in Square M/12 (of the University of Chicago expedition grid see Loud 1948: Fig. 413). The horizontal position matches perfectly our Walls 08/J/1 and 08/J/2 (Wall 08/J/2 being built directly on top of Wall 08/J/1; Fig. 2.85) which form the northern baulk of Squares S/20 and T/20 (Fig. 2.80). The upper wall (08/J/1) could be seen on the surface prior to our excavations and extends eastwards towards the location of Tomb 5239 (see below). The architecture within ca. 5 m north of this wall seen on the Stratum VIA plan lies beneath the modern tourist path. The stratiication of Walls 08/J/1 and 08/J/2 relative to University of Chicagos numbering is not perfectly clear, but some suggestions will be presented below. More certain is the so-called Deep Wall on the Stratum VIA plan. It is still preserved north of the tourist path and comes into Sub-Area Upper J as Wall 08/J/18. As we shall see below, this wall is earlier than Stratum VIA.2 The University of Chicago plan suggests that there was no excavation south of our Wall 08/J/2 (compare Loud 1948: Figs. 413, 394399; Harrison 2004: Fig. 6). It is clear, however, that they had indeed excavated the full length of this trench into Area BB, fully exposing Stratum VIA as well as earlier walls (see, for example, the photographs in Harrison 2004: Fig. 121; Loud 1948: Fig. 183). This was clear during our excavation as most of Squares S/20 and T/20 were illed with modern debris down to Level J-12. The architecture of Levels J-12 through J-19 remained intact as the University of Chicago team often exposed several levels of architecture at once, leaving later ones pedestalled for some time (see, for example, Loud
1 2 These squares appear on a newly oriented Upper J local grid that is distinct from that used in Main Area J. See Part I of this chapter. The architecture in the Area DD southern extension was so cursorily studied by the University of Chicago that its stratigraphic attribution of much of the architecture therein is questionable.
119
M ATTHEW J. A DAMS AND JAMES M. BOS
1948: Fig. 148; Harrison 2004: Fig. 6), removing only loors and occupational debris. Unfortunately, the Level J-12 and later architecture and the occupational debris within were not presented by Loud (1948), presumably because it was only excavated very near to the cessation of work in 1939. In any case, the preserved remains appear to be only the deepest foundations beneath any loor levels. Schumacher, too, before the University of Chicago had excavated some of this material as his long north-south trench, went partially through this location (see Harrison 2004: Fig. 3; Schumacher 1908: Taf. II; plan accessible in Kempinski 1989: Plan 1). Sub-Area Upper J features a noticeable change in elevation from the higher east to the lower west. Excavation in Area J and the bedrock topography study conducted by N. Franklin and J. Peersmann (unpublished) has shown that throughout the history of the formation of the site, this was the general point at which the higher acropolis met the lower portion of the western town. From very early on, perhaps as early as Level J-4 (see Part III of this chapter), terrace walls revetted the western edge of the acropolis. Through the Early and Middle Bronze Ages, new terraces were constructed farther and farther west, to extend the acropolis surface (more below, Fig. 2.82). For example, the strata are so drastically terraced and sloping that Area BBs Stratum XII shared similar levels as Area DDs Stratum VIA farther north (see Loud 1948: Fig. 416). Note that Upper Area J continues the Level-numbering sequence employed in the rest of Area J; thus the lower numbers represent earlier occupation than the higher numbers (Tables 2.22.3; see Finkelstein and Ussishkin 2000a: 34).
LEVEL J-8
Stratum XIV is one of the most problematic phases in the Megiddo sequence. It was completely removed in Area BB by the University of Chicago wherever they encountered it. As presented by Loud, it is a very poorly preserved stratum with two superimposed building phases, marked by the reuse of Temple 4040 (Figs. 2.56, 2.742.75; Loud 1948: 84, Fig. 395). The poor preservation of the stratum can be attributed to the Stratum XIII and later cutting of burial, storage and refuse pits deeply into the underlying strata (see Hallote 2001).3 Loud already admitted that two stages were apparent in this stratum but that the poor preservation of the walls did not allow them to be separated, and thus, he included them together on the same plan. There are also major problems with the inds. The pottery presented from this stratum is an amalgamation of pottery that we now recognize as Intermediate Bronze, MB I and MB II (Loud 1948: Pls. 1015). Secondary scholarship has not reached an agreement on the date of Stratum XIV with ranges from the IBA to the MB II (e.g., Kenyon 1958; 1969; Gerstenblith 1983; Aharoni 1993; Hallote 2001). The northern continuation of the Stratum XIV structure in Squares M/11M/12 (University of Chicago expedition grid; Fig. 2.61) was uncovered in Sub-Area Upper J in 2008. It is a well constructed building that dates to the MB I (on the basis of occupational and tomb ceramics). Beneath this building we found the earlier remains of a similar structure constructed directly on top of the ruins of the Level J-7 Temple 5269 (Fig. 2.74). Together these two architectural levels have been assigned to Levels J-8 and J-9 (equivalent
This phenomenon has not only affected Stratum XIV, but also our Levels J-7, J-5/J-6, and, to a limited extent, Level J-4; take the extreme example of the massive burial pit in Square J/8 that completely cut through Stratum XIV down to Level J-4, penetrating beneath the Level J-4 temple loor.
120
CHAPTER 2: A REA J, PART IV: SUB -A REA U PPER J
Fig. 2.74: Remains of Level J-8 in Sub-Area Upper J. Note underlining Level J-7 walls in grey.
to the two phases of Stratum XIV; Loud 1948: 84), and represent the earliest Middle Bronze occupation at the site after the Level J-7 Temples.
LEVEL J-9
In Sub-Area Upper J, Level J-9 is characterized by the northern continuation of Stratum XIV architecture in the University of Chicagos Squares M/1112 (Figs. 2.742.75). The building is at least partially a rebuild of Level J-8 (cf. Figs. 2.742.75). Wall 08/J/23 enters Sub-Area Upper J obliquely from the south. This wall was originally revealed by the University of Chicago straddling Squares M/11 and M/12 (Figs. 2.75, 2.77). It crosses the southeastern corner of Square S/20 and continues into the east baulk, where it probably runs for at least 23 m, enough to accommodate at least one more room to the north. Wall 08/J/23 is intersected perpendicularly by Wall 08/J/28, an interior stone-socle wall upon which some mudbrick was still preserved. This wall meets Wall 08/J/29 some 3 m to the west. The southern portion of Wall 08/J/29 was also excavated by the University of Chicago (Fig. 2.75).
121
In the northeastern corner of this room was a square plastered bin. The bin was constructed using the pre-existing walls as two sides and constructing abutting thin mudbrick (or just mud?) walls for the other two sides (Walls 08/J/31 and 08/J/32). The inside of the bin was completely coated with a chalk paste walls and loor (Fig. 2.77). As concluded by Ruth Shahack-Gross, the chalky paste is composed of local chalk bedrock crushed and mixed with water and straw, and is not cooked like a proper plaster. This point is particularly important because uncooked paste would have been unsuitable for containing liquid. Thus, whatever was kept in this bin and its counterpart to the north of Wall 08/J/28 (see below) was certainly dry. Nothing was found within the bin aside from the later ill (the northern bin has not yet been completely excavated). The outside of the bin was also coated, as was Wall 08/J/28 (seamlessly with the exterior of the bin). A rough calculation of the minimum volume based on the preserved height is ca. 0.63 m3. Signiicantly, at elevation 157.62 m, the coating on the western face of this installation curved outwards, probably suggesting an otherwise undetected loor level. Wall 08/J/28 continued down, so this loor probably represented a later addition to the room. The inside bottom of the installation was found at 157.73 m, approximately the same height at this outward curving coating outside. This suggests that the installation was inserted into the building in a second phase. The earlier occupational accumulation 08/J/201, below the plaster curve, contained a whole vessel among other occupational debris (ash, etc.) suggesting an MB I date. West of Wall 08/J/29, in what must have been an exterior space, fence Wall 08/J/35 perpendicularly abutted Wall 08/J/29 and Tabun 08/J/203, nestled into their corner. The base of the tabun, though not completely excavated, was at approximately 157.50 m. Just to the west and north of this we observed several different ash and phytolith layers that correspond to occupational debris (08/J/198) contemporary with the tabun. Another bin, slightly larger, was found on the north side of Wall 08/J/28. It does not seem likely that these rooms were connected by a doorway, but the later cutting of a pit has destroyed the location where one might be expected. This bin was also constructed in the corner of a room, in this case in the southeast corner at the meeting of Walls 08/J/28 and 08/J/23. It was slightly larger and its two thin walls were constructed of stone rather than mudbrick (Walls 08/J/33 and 08/J/34). It too was coated with chalk paste. Six burials were uncovered in Upper J belonging to the Level J-8 and Level J-9 buildings. Burials 10/J/126 and 10/J/190, both infants, almost certainly belong to Level J-8 (Fig. 2.74). Burials 10/J/100 and 10/J/113, both infants, almost certainly belong to Level J-9. The two shaft and crypt burials 10/J/136 and 10/J/171, may belong to either level, but may have been in use in both (compare large shaft and crypt burial in Squares JH/89, below; Fig. 2.92). These tombs will be fully published in a future report. In the main sector of Area J, Pit 06/J/140 belongs to Level J-9 (Fig. 2.94). It is located exactly beneath and within the Stratum XIV building removed by the University of Chicago team from Square M/13 (Fig. 2.92; Loud 1948: Fig. 395). It is a stone-lined pit more than 2-m-deep. It appears to have been located in a room within the building. It was probably used for subterranean storage, perhaps of grain. Its location near the cultic heart of the acropolis suggests that it may have been part of an administrative unit related to the cult-place. Compare the similar stone-lined Pit 06/J/055 from Level J-11, which we suggest had a similar function vis--vis the cult-centre. Pit 06/J/140 was illed with refuse at the end of its life, probably in the transition to Level J-10 (Stratum XIII), when the building in which it was located was rebuilt. Therefore, it was in use in Level J-9, though
122
Fig. 2.75: Remains of Level J-9 in Sub Area Upper J reconstructed with the University of Chicagos Stratum XIV architecture in black (after Loud 1948: Fig. 395). Note Level J-8 in grey.
123
Fig. 2.76: Level J-9 facing east. Note the relative level of the Level J-7 temples.
124
Fig. 2.77: Level J-9 chalk-paste coated bins, facing east. Note the termination of the paste on the southern bin; this is the projected location of the upper loor.
it may have been constructed earlier. The contents from this ill are comparable to the Level J-10 Fill 08/J/157 discussed below. There we suggest that the ill is debris from a dismantled Level J-9 building. Two other pits (08/J/086 and 98/J/012) clearly pre-date Level J-10, but they are too isolated to connect to Level J-8 or Level J-9 (Fig. 2.94).
LEVEL J-10
Level J-10 corresponds to Stratum XIII. Though the University of Chicago team determined two sub-strata, XIIIA and XIIIB, no such distinction could be made in our encounter with this level in Sub-Area Upper J. Here, the Level J-10 architecture was truncated by a later Level J-11 building and is represented only by the thick Fill 08/J/157 (Fig. 2.78). As noted above, a slope down to the north and west existed in Level J-9. Fill 08/J/157 appears to have been part of an attempt to raise this portion of the site, which would have required a terrace wall somewhere to the west beyond our excavation area. This created a platform upon which the Stratum XIII architecture was constructed. The earlier structures of Level J-9 were vacated, the terrace wall was constructed and the ill was placed immediately thereafter. The chalk coating of both Level J-9 bins remained in excellent condition, suggesting that a signiicant amount of time had not passed between the end of Level J-9 and the laying of the massive construction ill over top in Level J-10.
125
Fig. 2.78: Level J-10 brick ill over Level J-9, facing north. Note Wall 08/J/23 at right.
On top of this new terrace, Level J-10 saw the construction of a large building covering well over 500 m (Loud 1948: Figs. 396397). The location of this building near the cultic centre attests to its importance, as does the fact that in the succeeding Level J-11 it is replaced by a palace (see below). The plaster-paved rooms and columned rooms unearthed by the University of Chicago team are all features associated with palatial architecture at Megiddo (Loud 1948: Fig. 397). The owner of this structure apparently had the power to commandeer structures such as that in Level J-9 and transform a portion of the tell to accommodate his new construction. To a late phase of Level J-10 must go Tombs 04/J/056, 06/J/107 and 06/J/143, which are built over by Stratum XII architecture (Fig. 2.94; the tomb assemblages will be dealt with in a future report).
2
FILL 08/J/157 Fill 08/J/157 was composed of unlevigated building materials. Typical ills in other strata around the site consist of crushed and mixed mudbrick debris. In this ill, greenish yellow bricks were tossed over the Level J-9 architecture. These mudbricks were evident in the north baulk of Square S/21 (Fig. 2.78). Large quantities of animal bones and pottery (primarily pithoi) were also present. We were able to determine two basic discrete portions of the ill: the mudbricks (plaster still on many of them) were in one deposition and the pottery and bones were in another. Though only excavated in the limited area of Sub-Area Upper J, it appeared as though most of the mudbricks were thrown in irst and then the pottery and bones with additional mudbrick thrown in on top. Tip lines were evident down towards the west, suggesting that the dumping occurred from the east (Fig. 2.78). Mixed in with the bones was a bronze censer (08/J/195/ AR1; see Blockman, Chapter 15, Fig. 15.2). It is clear that a building was dismantled during the terrace construction process, and its constituent architectural parts, as well as its contents, were thrown here behind the terracing retaining wall.
126
Fig. 2.79: Remains of Level J-11 in Sub-Area Upper J.
This piece of information offers some important data to our discussion of the Level J-10 transformation of the site. First, the pottery is all MB I. Since this dismantled building material undoubtedly comes from Level J-9, it gives us an additional datum for that level. Additionally, the contents themselves have much to say. While thorough analysis of the bones is yet to be conducted, the parts are consistent with animal sacriice. The bronze censer is an indication of an assemblage not consonant with everyday refuse. It is tempting to connect this with the dismantling of a Level J-9 cultic building. Two options for the location of this building stand out. It may have stood where the large house of Stratum XIII was built. It would thus be represented both by this ill and by, perhaps, some of the foundations preserved on the University of Chicago plan west of Temple 4040. Alternatively (and preferred by the authors of this chapter), these remains may have come from Temple 4040 itself. Thus, if the remodelled and reused interior of Temple 4040 can be attributed to the Levels J-8 and J-9 occupation at the site, the Level J-10 ill may represent the dismantling of that cult-space to make way for a new cult space consonant with the Level J-10 remodelling of the city. While it is dificult to accept Epsteins suggestion to assign the Migdol Temple 2048 to this stratum (Epstein 1965; Adams forthcoming b), we must conclude that some aspect of cultic worship continued in the area in Stratum XIII. At the opposite end of the spectrum, Hallotes analysis concluding that no central cultic space was needed for cultural reasons fails to account for the fact that a central cult space was, indeed, needed in Level J-9 when Temple 4040 was clearly reused. Ultimately, Hallote does not argue convincingly that no temple existed in the MB (Hallote 2001). The going out of use of Temple 4040 while maintaining
127
the boundaries of the space architecturally and temporally strongly supports the reconstruction of a cultic structure above the Temple 4040. We suggest the hypothesis that Fill 08/J/157 represents the dismantling of Temple 4040 to make way for the new cultic structure (the issue of an MB temple is further explored in Adams forthcoming b). Contemporary to this illing in Sub-Area Upper J is the illing of the Level J-9 stone-lined Pit 06/J/140 in main Area J (see above). This ill appears to have a similar composition to that in Sub-Area Upper J.
LEVEL J-11
Level J-11 (Stratum XII) inaugurates a period of more monumental palatial buildings on this portion of the mound. Near the end of Stratum XIII a city wall was built, and shortly thereafter, the Level J-10 large structure was replaced by a new, well-planned palace (Loud 1948: 8492, Fig. 398). In Sub-Area Upper J, the new palace is represented by the construction of Walls 08/J/5 and 08/J/13 (Fig. 2.79). These walls form a northwestern interior corner of the building seen on the Stratum XII plan (Fig. 2.80; Loud 1948: Fig. 398) which in turn is further exposed in Schumachers Nordburg (a conglomeration of MB and LB strata; Schumacher 1908). No loor was found associated with these walls as only the deep foundations survived. We also encountered Level J-11 in the main sector of Area J by way of an Middle Bronze intrusion found in the northwestern corner of Square H/7 (straddling the baulk with Square J/7; Figs. 2.54, 2.81). This stone-lined pit (06/J/5) was seen by the University of Chicago and appears on their Area BB Stratum XII plan (Loud 1948: 398); it can also be seen in the distance in one of the photos (Loud 1948: Fig. 183). Based on their elevations and the silty wash that we found inside, it appears that they excavated the contents all the way down to the bottom (the pit cuts into the Level J-4 temple) over 3.5-m-deep. We did discover that the foundation trench for this pit was quite wide, up to 0.5 m wider than the stone lining (at the radius). Our excavations revealed no additional clues as to the purpose of this stone-lined pit, but the plan of the building in which it was found seems to suggest that the building itself was located at the top of a staircase leading to the acropolis. The structure itself appears to be a gateway with a drain beneath the threshold. The volume of the pit can be calculated at 4.84 m3. If we use wheat as an example (785 kg per m3; Kemp 2006: 178), this amounts to 3,800 kg of cereal. According to New Kingdom Egyptian documents, this is the equivalent of more than 6,000 one-day individual rations (Kemp 2006: 177179).4 The presence of such a large storage pit near the entrance of the sacred precinct may have been connected to the distribution of grain controlled by the cult.
LEVEL J-12
In Sub-Area Upper J, we witness the transition from the University of Chicagos Stratum XII palace to the Stratum XI palace with the construction of Walls 08/J/8 and 08/J/19, which form a northwestern corner (Figs. 2.822.83). Some additional smaller walls were part of this level. Walls 08/J/4 and 08/J/3 create a square room with the corner formed by Walls 08/J/8 and 08/J/19 (Figs. 2.822.83). These two walls are
4
Note that the daily ration in these documents is a unit of exchange, not necessarily an indication of the daily caloric intake of a single individual. One must assume that these grain rations were supplemented with other foods as part of the regular diet.
128
Fig. 2.80: Level J-11 reconstructed, with University of Chicago Stratum XII architecture in black (after Loud 1948: Figs. 398 and 415).
129
Fig. 2.81: Remains of Level J-11 in the main sector of Area J (06/J/55) with Stratum XII architecture in black (after Loud 1948: Fig. 398).
Fig. 2.82: Remains of Level J-12 in Sub-Area Upper J. Note the cutting of the northwest portion of these walls by Level J-18 Walls 08/08 and 10/37 (Fig. 2.91).
130
Fig. 2.83: Level J-12 reconstructed with University of Chicago Stratum XI architecture in black (after Loud 1948: Figs. 399 and 415).
thinner and must be interior walls. The same is true of Walls 08/J/7 and 08/J/6. A couple of large lat stones in Wall 08/J/3 may indicate a threshold to the room. A glance at the Stratum XI plan will demonstrate that our Wall 08/J/8 is in perfect alignment with a wall in the western portion of Square M/12, and Wall 08/J/19 is in perfect alignment with a wall straddling the Square N/11-12 line (Fig. 2.83). Thus, this forms part of the Stratum XI palatial complex. It is tempting to reconstruct the eastern face of Wall 08/J/8 as the western jamb of the entrance to the building, or at least the jamb of the entrance to the large lime-paved courtyard. The small room, then, could possibly be interpreted as a gate house.
131
One important piece of evidence for the dating of this level is a scarab that was found sealed by Wall 08/J/19 and clearly above Level J-11s Wall 08/J/13 (08/J/97/AR1; see Keel, Chapter 17, Fig. 17.2). The patterns and hieroglyphs on the scarab conform well with the Middle Kingdom scarabs of Strata XIIXI as treated by Tufnell (1973).
LEVEL J-13
Level J-13 heralds the construction of yet another palace, that of Stratum X. In Sub-Area Upper J this transition sees the levelling down of the Level J-12 palace and the construction of new walls directly over the old: Wall 06/J/16 and its perpendicular Crosswalls 08/J/9 and 06/J/18 (Figs. 2.842.86). There was also some evidence of a robbed wall parallel to Wall 06/J/18. Between Wall 06/J/18 and the robbed wall is a stone-lined pit that penetrated down to Level J-7. It has parallels elsewhere in Stratum X, in the westernmost portion of the palace. These bins seem isolated from the western chambers of the palace and perhaps represent the storage area, while the nice plastered rooms to the east are living and reception rooms. Overall, Level J-13 relects the continuation from Levels J-12 and J-11. The most signiicant architectural change will come in Level J-14 where this portion of the mound is no longer used for the palace grounds.
132
Fig. 2.85: Level J-13 superimposed over Level J-12 in Sub-Area Upper J, facing east. Note Walls 08/J/1 and 08/J/2 of Levels J-18 and J-19 in left section.
LEVEL J-14
Level J-14 is only represented by a single wall running north-south along the eastern edge of Sub-Area Upper J (Fig. 2.87). Its foundations were dug down where the builders discovered a portion of Wall 06/J/18. Instead of removing the stones in the way, they simply used them in the construction of their wall. No surfaces were associated with this wall and almost nothing can be said about it. The wall itself is at a completely different alignment and scale from that which came before, indicating a major change in the architectural use of this portion of the mound. After Level J-13, the space west of the sacred area is devoid of architecture. This is certainly due to later construction on the site,5 but it appears that this sector of the mound is no longer used for the construction of palaces. We are inclined to follow Kempinski in recognizing the relocation of the palace to Area AA (1989). The architecture here of Levels J-14 and J-15 is simply different and less ambitious than that which came before it. While it is regrettable that we could not add new ceramic data from these levels, there is agreement that the change from Level J-13 (Stratum X) to Level J-14 (Stratum IX) marks the ceramic transition between the MB III and the LB I, with changes already beginning in Level J-13 (Stratum X) (Hallote 2001; Gadot,
5 It seems reasonable to assign this denudation to Stratum VA-IVB and/or the IVA stables, but of this we cannot be certain. Area BB seems to have been cut down (especially in its western portion) in such a way as to leave all levels between IX and VB poorly represented in this area. It is dificult to estimate the damage done by Schumacher and the University of Chicago, either of which may be responsible for the dearth of material at this point.
133
Fig. 2.86: Level J-13 remains, reconstructed with University of Chicago Stratum X architecture in black (after Loud 1948: Fig. 400). Note the circular subterranean stone-lined bins.
134
135
Fig. 2.89: Level J-15 in Sub-Area Upper J, facing east.
Yasur-Landau and Ilan 2006). This transition is considered to be dificult to isolate because of the high degree of cultural continuity between the MB and LB. However, the transition from the palatial space of Level J-13 to the decidedly unpalatial space of Level J-14 represents a major transformation in the spatial organization of the acropolis.6
LEVEL J-15
At the extreme eastern edge of the square almost directly above Level J-14s Wall 08/J/14 (but at a slightly different orientation7) sit Walls 08/J/10 and 08/J/11, which form the corner of a room that goes into the eastern baulk. The room is paved with small cobbles and an entrance at the walls junction is evident (Figs. 2.882.89). Around the north side of this structure was a plastered surface seen only in the section (see the plaster in the distant section in Fig. 2.85). The east section shows that the cobbled surface extended south only so far before it was replaced by a plastered surface. The extensive use of plaster is consonant with the Stratum IX plan. Wall 08/J/10 does seem to align perfectly with a wall from the University of Chicagos plan. If these can be drawn together, something more coherent comes out of the Stratum IX plan (Fig. 2.90).
This has been evident since the University of Chicagos publication. It is therefore dificult to understand why Hallote does not acknowledge any major changes until Stratum VIII (Hallote 2001: 203), other than to support the problematic thesis that there are no temples in the Middle Bronze Age (Adams forthcoming b). During the excavation, Wall 08/J/14 was originally thought to have been the foundation for Wall 08/J/10, but it became clear in the dismantling of the Level J-15 architecture that they are oriented differently. Still, they may represent two phases of the same building, but this cannot be determined with the information at hand.
136
Fig. 2.90: Level J-15 reconstructed with University of Chicago Stratum IX architecture in black (after Loud 1948: Fig. 401).
137
Fig. 2.91: Remains of Level J-18 in Upper J (University of Chicago-exposed walls in black; after Loud 1949: Fig. 413).
138
139
Fig. 2.92: Sub-Area Upper J, northern section drawing after the 2010 excavation season.
Fig. 2.93: Sub-Area Upper J, north and east sections.
LEVEL J-16
In the early phase of our excavation in Sub-Area Upper J we noticed a number of features that could be seen in the north and east sections (Fig. 2.93). Above the Level J-15 architecture in the east section is a thick plaster loor. It extends north and south through the entire section. In the north section it continues from the east section westward for about one metre. Its elevation is generally around 160.15 m, which would match well with the plaster of Stratum IX. Until additional excavation can add data, we tentatively assign this plaster to a late phase of Stratum IX. Note also that plaster is a notable feature of Level J-15.
LEVEL J-17
To Level J-17 we assign the University of Chicagos Structure 5239, which appears on their Stratum IX plan (Loud 1948: Fig. 401). Following a hypothesis by D. Ilan (2001), we concur that it likely belongs to Stratum IX or VIII. A sondage against the eastern face of the substructure and the foundation trench was excavated in the Square J/7-K/7 incision down to the level of the Level J-4 temple (Fig. 2.26). Structure 5239 penetrated the already robbed (in Level J-5) western sanctuary wall. The foundation trench was not larger than the structure itself at this point. Unfortunately, no ceramics were found that could add any new data to the dating. The outer face of the structure exposed here was much rougher than the well-cut inner face, demonstrating (with the foundation trench) that at least the surviving part of the structure was, indeed, subterranean.
LEVEL J-18
Level J-18 is characterized by a large subterranean stone structure (Figs. 2.912.92). Wall 08/J/18 is preserved as high as 2.5 metres. It cuts through Level J-17 down through the Level J-9 tabun-courtyard
140
Fig. 2.94: Middle Bronze Age tombs and pits in the main sector of Sub-Area Upper J (vessels marked in black).
into Level J-8. This wall is apparently the same as the Deep Wall on the University of Chicagos Stratum VIA plan (Loud 1948: Fig. 413), which continues northwards from Sub-Area Upper J for about 13 m before it makes a 90 turn towards the west (see Loud 1948: Fig. 413; Harrison 2004: Figs. 120121). The structure was constructed by excavating a large rectangular foundation and then facing the exposed sections with large ieldstones (a few reused large ashlars were also used), which acted as a revetment. The interior of the structure was paved with ist-sized cobbles. Generally, the construction is reminiscent of the construction of Structure 5239, some 10 metres to the east. It is dificult to date the original construction, but it certainly post-dates the Middle Bronze Age. At some point, the large structure was backilled with over 2 m of ine ash (see Fig. 2.92). A couple of fragments of IA IIB pottery suggest that this backilling event happened late in the Iron Age, perhaps with the construction of the Stratum IVA stables complex. East of Wall 08/J/18 are two walls, which are probably contemporary with the subterranean building. East-west Wall 08/J/1 forms the boundary of the northern baulk of Sub-Area Upper J. Wall 08/J/15 is a perpendicular wall a few centimetres east of Wall 08/J/2. Both can only be seen in the northern section (Figs. 2.912.93). The location of Wall 08/J/1 is exactly where we expected to encounter the wall that appears on the University of Chicagos Area DD plan of Stratum VIA in the northwest corner of Square M/12 (Loud 1948: Fig. 413). The attached architecture on the immediate north side of this wall on the plan by Chicago is no longer visible and lies beneath the tourist path leading up to Palace 6000 (Area L) and the northern observation platform. However, it should be noted that our levels for this wall do not match those of the
141
University of Chicago team our bottom levels being ca. 50 cm lower. It is not clear if this is a coincidence, an earlier version of the Stratum VIA wall, or a mistake in levels on Chicagos plan. What is clear is that it is stratigraphically later than the Level J-16 plaster surface, and therefore dates later than Stratum IX.
LEVEL J-198
Wall 08/J/2 is the only element that may be attributed to this phase (Figs. 2.85, 2.922.93). It was built over Wall 08/J/1, but continues the entire length of Sub-Area Upper Js northern section and can be seen on the surface east of Sub-Area Upper J as far as the location of Tomb 5239. The wall may date to the University of Chicagos Stratum V.
UNSTRATIFIABLE MB PITS AND BURIALS

The subterranean burial complex of Squares HJ/89 cannot yet be stratiied (Fig. 2.94). This complex appears to be a uniied crypt accessed by a shaft located at approximately the location of the Squares HJ/89 grid marker. This shaft penetrated down to the Level J-4 Wall 96/J/7. Some of the stones from this wall were removed at the base of this shaft (see Figs. 2.222.23, and especially 2.32 and 2.45). From this shaft, horizontal shafts and burial niches were hollowed out to receive internments (Fig. 2.94). Once the bodies were placed, the niches were blocked with stone walls (see Burials 04/J/75 and 06/J/75). Several individuals were found in Tomb 06/J/37. Stratigraphically and ceramically, we tentatively date the tomb to the MB III. The osteological data is presented by G. Samaora in this volume. Detailed stratigraphic and ceramic discussion of this interesting tomb complex will be presented in a future report.
SUMMARY
Our excavations in Sub-Area Upper J have established a new sequence through the Middle Bronze Age parallel to that of the University of Chicago. We have conirmed an MB I date for Stratum XIV (Levels J-8 and J-9). This suggests that much of the architecture appearing in Chicagos Stratum XIV (and the reuse of Temple 4040) belong to the original MB I reoccupation of the site. An Intermediate Bronze Age level has not been identiied. The true nature of the site during this irst Middle Bronze reoccupation remains obscure, but it is clear that Level J-10 (Stratum XIII) shows signiicant signs of town planning and major construction efforts. We have also suggested that a new ritual space was established over the buried remains of Temple 4040 in this phase. Levels J-11 through J-13 yielded portions of the Strata XIIX MB palaces, conirming the basic stratigraphy on this portion of the mound as established by the University of Chicago. Unfortunately, little new data has been added regarding the post-palatial character of this area. As others have recognized, with Stratum IX, the palace is relocated to Area AA. Because of later reshaping of the mound at this point, the Late Bronze occupation here is poorly represented, and while we have some uncertain evidence for Stratum VI architecture nearby (southern extension of Area DD), it is not until Stratum VA-IVB and the construction of the northern stable complex in Stratum IVA that any coherent plan can be established for this portion of the site.
8 Level J-19 is the next logical level designation, but since the Stratum IVA northern stable construction removed large portions of Strata V and VI (see Loud 1948; Harrison 2004), several University of Chicago strata are not accounted for in the Area J numbering scheme.
142
AREA J, PART V:
ANALYSES OF SEDIMENTS FROM THE LEVEL
PART V: ANALYSES OF SEDIMENTS FROM THE LEVEL J-4 TEMPLE FLOOR

David Friesem and Ruth Shahack-Gross
Sediment samples were collected during the 2008 excavation season from within the Level J-4 temple (EB IB) in Area J for the purpose of material-identiication. Speciically, we aimed at identifying the composition of the following features: 1. The white coating of the interior faces of the temples walls 2. The thin white layers found above the temples loor (Phase J-4a) 3. The dark grey sediment patches found above the temples loor (Phase J-4a) 4. The brown sediments identiied as the temples loor
METHODS
Sediments were collected in the ield as bulk samples, based on differences in colour and texture. Analyses were conducted at the Kimmel Center for Archaeological Science, Weizmann Institute of Science, Israel. Analyses included Fourier Transform Infrared spectroscopy (FTIR; Nicolet 380, Thermo Electron Corp.). Minerals and organic materials were identiied using a reference library as well as additional available literature. Samples were prepared using the KBr method and spectra collected between 4000 and 400 cm-1 at 4 cm-1 resolution. Samples were also prepared as grain mounts on microscope slides for observation using a petrographic microscope (Nikon 50POL). The analyses performed are complementary and allow for accurate identiications of phytoliths, wood ash, animal dung and plaster. For samples that contained the mineral calcite in their FTIR spectrum, the heights of the 2 vibration of calcite at 875 cm-1 and the 4 vibration at 713 cm-1 were measured relative to a baseline, and the ratio between them (2/4) , termed ratio, was calculated. According to Chu et al. (2008) this ratio changes due to heating of calcite in the process of transforming limestone into lime plaster. They showed that the ratio is ca. 3.0 for geological unheated limestone, and close to 7.0 when lime plaster has been settled out of heated and slaked limestone. Additionally, they showed that the ratio for wood ash (i.e., calcite that forms after heating biogenic calcium-oxalate in trees) is around 4.0. By conducting this measurement we intended to determine whether calcite in the studied sediments was geogenic or a result of anthropogenic burning activities. For samples that contained clay minerals, we used criteria set forward by Berna et al. (2007) to infer whether clay minerals have been exposed to high temperatures. Berna et al.s study showed that although various clay minerals respond differently to heating, a general pattern of thermal changes can be detected using infrared spectroscopy. The major changes occur at ca. 500C when structural water is lost and the water absorptions at 3650 and 3695cm-1 disappear from the infrared spectrum, and at 800C when the silicate component undergoes structural changes evident in the infrared spectrum by a shift of the main absorption from ca. 1035cm-1 to ca. 1080cm-1 and a decrease in the height of the absorption at 530cm-1. By conducting these observations on the infrared spectra of clay minerals we intended to determine whether heated clays are present in the studied sediments, for example, as a byproduct of preparation of lime plaster.
143
DAVID F RIESEM AND RUTH SHAHACK-GROSS
RESULTS
The results are presented in Table 2.7. Most studied sediments are composed of clay and calcite. The clay minerals were not exposed to temperatures higher than 500C based on the presence of structural water in the infrared spectra (Berna et al. 2007). There is no evidence for intensive burning within the temple, supporting the interpretation that the temple was abandoned rather than destroyed. None of the identiied calcite had been heated to temperatures at which calcination occurs (i.e., minimum of 700C; Chu et al. 2008) indicating that lime plaster and/or ash are either absent or present in minor amounts. For the latter, microscopic observations enabled us to identify wood ash, but these crystals were quite infrequent, except for sample RME-31, which represents the remains of a hearth (see below). Quartz occurs in all samples as grains in the size range of ine sand (50200 m). Due to its low concentrations in the studied sediments, quartz was identiied microscopically but not using the FTIR. Its source is probably from blown dust or soil material (if for example mud bricks had decayed in the studied area). Dahllite, a phosphate-bearing mineral, is present in many sediments. Based on the study of Gafri and Shahack-Gross (Chapter 34, also Shahack-Gross and Mor Gafri), as well as on preliminary analysis of sediments from Area H (Einhorn 2011), dahllite in Megiddo sediments is abundant, originating from microscopic bone fragments and nodules that formed from phosphate-rich solutions after organic matter degradation. Dahllite nodules are usually found associated with phytoliths, thus it is reasonable to assume that in many cases the source of phosphate was from decomposing vegetal matter. Speciic feature identiications largely support M.J. Adamss ield interpretations of materials above the temple loor (Phase J-4a): 1) the thin white layers are indeed composed of opaline phytoliths; 2) the ashy features include wood ash crystals (i.e., pseudomorphs of calcite after calcium-oxalate biogenic crystals) and abundant microscopic charcoal fragments. The phytolith patches do not seem to originate from livestock dung based on the absence of dung spherulites and dahllite (cf. Shahack-Gross et al. 2003; 2005). The phytoliths in these patches can be generally assigned to the grass family, but in the absence of quantitative analyses their diversity was not studied and they cannot be assigned a genus and/or species. The samples of white coating on the inner surface of the temples walls provide no evidence that it was prepared through calcination at high temperatures. The evidence for lack of heating includes the low ratio as well as the presence of foraminifera microfossils typical of chalk. Had the chalk been heated to produce lime plaster, these microfossils would have disintegrated. In a few samples (especially from Wall 96/J/07), macroscopic elongated voids occur in random orientations in the white mass, and some voids include grass phytoliths. This observation indicates that the coating on the temples walls was prepared from crushed or powdered chalk (abundant in the vicinity of Megiddo) tempered with grass stems (i.e., straw). Brown sediments studied here are composed of a mixture of all mineral types, i.e., clay, calcite, quartz, opal and dahllite, and include phytoliths and sometimes wood ash and microscopic fragments of bones and charcoal. The large variety of materials included in these sediments, with no speciic domination of one speciic material, might relect a yet unstudied aspect of the origin of ill sediments. Crystal masses of low interference colours were identiied in several samples during the microscopic examination. Because the sediments were prepared as grain mounts and not at ixed 30 m thickness, the identiication of these crystal masses as gypsum is still uncertain. Future micromorphological analyses might aid in resolving this observation.
144
CHAPTER 2: A REA J, PART V: A NALYSES OF SEDIMENTS FROM THE LEVEL J-4 T EMPLE F LOOR
DISCUSSION AND CONCLUSIONS

The study performed here was done on a small sample of sediments and did not use all available quantitative methods. It should thus be regarded as a pilot study. The initial results support M.J. Adamss ield interpretations regarding the white and grey patches above the temples loor originating from phytoliths and wood ash respectively. If the identiication of the loor level (08/J/190) is correct, then these remains accumulated after the abandonment of the temple (Phase J-4a). The phytolith patches do not seem to originate from livestock dung, thus other possible sources for phytoliths should be considered for example a collapsed thatch roof, sporadic storage of straw, or remains of grass matting. The localized patches of wood ash may be representative of ephemeral hearths. The white plaster on the inner surfaces of the temples walls was prepared from crushed/powdered chalk tempered with grass stalks. In a recent study at the site of Neot Kedumim, where traditional agricultural practices are demonstrated to the general public, the practice of preparation of crushed/powdered chalk surfaces was recorded (Gafri and Shahack-Gross, this volume). According to local Arab villagers the surface of a threshing loor at Neot Kedumim was prepared from crushed chalk tempered with straw by pounding the mixture using stone rollers, a practice they have used for generations. It is possible that this practice was also conducted at Megiddo in the Bronze Age. TABLE 2.8: RESULTS OF FTIR AND GRAIN MOUNT ANALYSES
Sample registration No.* RME-23 Field description Square H/8, east sectionthin white layer (=08/J/185) Square F/9, surface thin white layer (=08/J/185) Identiied minerals 2/4 (FTIR analysis)** ratio*** Opal, Calcite n.d. Microscopic identiications Grass phytoliths, microsparitic chalk fragments/powder, foraminifera microfossils, quartz (as ine sand grains) Grass phytoliths, microsparitic chalk fragments/powder, foraminifera microfossils, quartz (as ine sand grains), microscopic charcoal fragments Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), few phytoliths Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains) Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), gypsum crystals?
RME-24
Opal, Calcite, n.d. Dahllite, Clay (n/a)
RME-25
Square F/9 white coating on face of 96/J/W07 Square G/7 white coating on face of 00/J/W21 G/7-G/8 baulk white material on surface of round stone table
Calcite, Clay, Dahllite?
3.3
RME-26
Calcite, Clay, Dahllite
3.4
RME-27
Calcite, Clay, Dahllite
3.5
145
DAVID F RIESEM AND RUTH SHAHACK-GROSS
TABLE 2.8: RESULTS OF FTIR AND GRAIN MOUNT ANALYSES (cont.)

Sample registration No.* Field description Identiied minerals 2/4 (FTIR analysis)** ratio*** Microscopic identiications
RME-30 (08/J/185/LB01) Phytolith layer and hearths a: white sediment
Opal, Calcite
n.d.
b: yellow sediment
Clay (n/a), Calcite 3.6
c: grey sediment
Clay (n/a), Calcite, 3.2 Dahllite
Many phytoliths, few microsparitic chalk fragments/ powder Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains). Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), tree and grass phytoliths, microscopic charcoal fragments, wood ash, gypsum crystals? Wood ash, abundant microscopic charcoal fragments, microscopic bone fragments, microsparitic chalk fragments/powder, foraminifera microfossils, quartz (as ine sand grains), few phytoliths Quartz (as ine sand grains), sparitic limestone fragments, foraminifera microfossils, few phytoliths
RME-31 (08/J/187/LB01) Ashy Hearths dark grey sediment Calcite, Clay (n/a?), Dahllite? 3.1
RME-32 (08/J/129/LB01) Level J-5 ill over Level J-4 temple brown sediment
Clay, Calcite, Dahllite? Quartz?
2.4
RME-33 (08/J/177/LB01) Plaster puddle and late hearths a: white sediment
Calcite, Clay (n/a) 2.9
b: dark grey sediment
Calcite, Clay (n/a), 2.6 Dahllite?
Microsparitic chalk fragments/ powder, foraminifera microfossils Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), few phytoliths
RME-34 (08/J/184/LB04) Natural washed-in debris brown sediment
Clay (n/a), Calcite, 2.6 Dahllite?
Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), few phytoliths, microscopic charcoal fragments, wood ash, gypsum crystals?
146
TABLE 2.8: RESULTS OF FTIR AND GRAIN MOUNT ANALYSES (cont.)

Sample registration No.* Field description Identiied minerals 2/4 (FTIR analysis)** ratio*** Microscopic identiications
RME-35 (08/J/188/LB04) Ephemeral earthen surface brown sediment
Calcite, Clay (n/a), 2.6 Dahllite?
Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), phytoliths, microscopic charcoal fragments, wood ash Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), gypsum crystals? Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), phytoliths, microscopic charcoal fragments, gypsum crystals?
RME-36 (08/J/193/LB01) Dense chalk deposit a: white sediment Calcite, Clay (n/a), 2.9 Quartz
b: brown-grey sediment
Clay (n/a), Calcite, 2.4 Dahllite?
RME-37 (08/J/190/LB02) Floor makeup under sandy sorted debris and phytoliths a: white sediment
Calcite, Clay (n/a), 2.9 Opal, Dahllite?
b: brown-grey sediment
Clay (n/a), Calcite, 2.4 Opal? Dahllite?
Phytoliths, microsparitic chalk, ine sand quartz, foraminifera fossils, microscopic bone fragments, dung spherulites? gypsum crystals? Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), wood ash, phytoliths, microscopic bone fragments? gypsum crystals?
RME-38 (08/J/190/LB06) Sandy sorted washedin debris brown sediment
Clay (n/a), Calcite, 2.2 Dahllite? Opal?
Microsparitic chalk fragments/ powder, foraminifera microfossils, quartz (as ine sand grains), few phytoliths, microscopic charcoal fragments
RME designations refer to laboratory analyses. Parenthetical designations refer to the Megiddo Expedition registration system (see Table 2.5). ** (n/a): clay not altered due to heating (i.e., not exposed to temperatures higher than 500C). *** The 2/4 ratio was calculated based on Chu et al. (2008). n.d.: not determined (because calcite was present in very small amounts which affects the ability to accurately determine the baseline for calculation of the ratio).
147
M ATTHEW J. A DAMS ET AL.
REFERENCES
Adams, M.J. 2009. An Interim Report on the Naqada III First Intermediate Period Stratiication at Mendes, 19992005. Delta Reports 1. Adams, M.J. Forthcoming a. Notes on the Early Temples at Megiddo. Levels J-2 and J-3. Adams, M.J. Forthcoming b. The Reurbanization of Megiddo in the Middle Bronze Age: New Data and New Reconstructions. Adams, M.J., David, J.K. and Homsher, R. Forthcoming. The Jezreel Valley Regional Project 1: Report on the 2010 Test Season of Excavation at Tel Megiddo East and Ain el-Qubbi. Adams, M.J., Finkelstein, I. and Ussishkin, D. Forthcoming. The Great Temple of Early Bronze Age Megiddo. Aharoni, Y. 1993. Megiddo. The New Encyclopedia of Archaeological Excavations in the Holy Land 3: 1003 1012. Andrae, W. 1922. Die Archaischen Ischtar-Temple in Assur. Leipzig. Ben-Tor, A. 1973. Plans of Dwellings and Temples in Early Bronze Age Palestine. Eretz Israel 11: 9298 (Hebrew). Ben-Tor, A. 2008. Hazor. The New Encyclopedia of Archaeological Excavations in the Holy Land 5: 17691776. Ben Tor, A., Bonil, R. and Zuckerman, S. 2003. Tel Qashish, a Village in the Jezreel Valley: Final Report of the Archaeological Excavations (19781987) (Qedem 5). Jerusalem. Berna, F., Behar, A., Shahack-Gross, R., Berg, J., Boaretto, E., Gilboa, A., Sharon, I., Shalev, S., Shilstein, S., Yahalom-Mack, N., Zorn, J.R. and Weiner, S. 2007. Sediments Exposed to High Temperatures: Reconstructing Pyrotechnological Processes in Late Bronze and Iron Age Strata at Tel Dor (Israel). Journal of Archaeological Science 34: 358373. Blockman, N. and Groman-Yeroslavski, I. 2006. The Early Bronze Age Flint Assemblage. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982000 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 315342. Brandfon, F.R. 1977. The Earliest City Wall at Megiddo. Tel Aviv 4: 7984. Van den Brink, E.C.M. 1992. Corpus and Numerical Evaluation of the Thinite Potmarks. In: Friedman, R. and Adams, B., eds. The Followers of Horus. Studies Dedicated to Michael Allen Hoffman. Oxford: 265296. Chu, V., Regev, L., Weiner, S. and Boaretto, E. 2008. Differentiating between Anthropogenic Calcite in Plaster, Ash and Natural Calcite using Infrared Spectroscopy: Implications in Archaeology. Journal of Archaeological Science 35: 905911. Cohen, S. 2002. Canaanites, Chronologies, and Connections: The Relationship of Middle Bronze IIA Canaan to Middle Kingdom Egypt (Studies in the Archaeology and History of the Levant 3). Winona Lake. de Miroschedji, P. 1988. Yarmouth I. Rapport sur les trios premires campagnes de fouilles Tel Yarmouth (Isral) (19801982). Paris. de Miroschedji, P. 2001. Notes on Early Bronze Age Metrology and the Birth of Architecture in Ancient Palestine. In: Wolff, S.R., ed. Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse (Studies in Ancient Oriental Civilization 59). Chicago: 465492. de Miroschedji, P., ed. 1989. Lurbanisation de la Palestine lge du Bronze ancient (British Archaeological Reports International 527). Oxford. Dunayevsky, I. and Kempinski, A. 1966. Notes and News: Megiddo. Israel Exploration Journal 16: 142.
148
Dunayevsky, I. and Kempinski, A. 1973. The Megiddo Temples. Zeitschrift des Deutschen Palstina-Vereins 89: 161187. Einhorn, S. 2011. Microarchaeology of Destruction and Abandonment Layers: A Case Study at Tel Megiddo (Unpublished M.A. Thesis, Tel Aviv University). Tel Aviv (Hebrew). Ellis, R. 1968. Foundation Deposits in Ancient Mesopotamia. New Haven. Eppelbaum, L. and Itkis, S. 2000. Magnetic Investigations in the Proto-historic Site to the East of Tel Megiddo. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 504514. Epstein, C. 1965. An Interpretation of the Megiddo Sacred Area During Middle Bronze II. Israel Exploration Journal 15: 204221. Epstein, C. 1973. The Sacred Area at Megiddo in Stratum XIX. Eretz Israel 11: 5457 (Hebrew). Esse, D. 1989. Secondary State Formation and Collapse in Early Bronze Age Palestine. In: de Miroschedji, P., ed. Lurbanisation de la Palestine lge du Bronze ancient (British Archaeological Reports International 527). Oxford: 8196. Esse, D. 1991. Subsistence, Trade and Social Change in the Early Bronze Age Palestine (Studies in Ancient Oriental Civilization 50). Chicago. Finkelstein, I. 1995. Two Notes on Early Bronze Urbanization and Urbanism. Tel Aviv 22: 4769. Finkelstein, I., Halpern, B., Lehmann, G. and Niemann, H.M. 2006. The Megiddo Hinterland Project. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 705776. Finkelstein, I. and Ussishkin, D. 2000a. Area J. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 2574. Finkelstein, I. and Ussishkin, D. 2000b. Archaeological and Historical Conclusions. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 576605. Finkelstein, I. and Ussishkin, D. 2003. The Cache of Egyptianized Vessels from Megiddo: A Stratigraphical Update. Tel Aviv 30: 2741. Finkelstein, I., Ussishkin, D., and Halpern, B. 2000. Introduction: The Megiddo Expedition. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 113. Finkelstein, I., Ussishkin, D. and Halpern, B. 2006. Archaeological and Historical Conclusions. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 843857. Finkelstein, I., Ussishkin, D. and Peersmann, J. 2006. Area J (The 19982000 Seasons). In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 2953. Gadot, Y., Yasur-Landau, A. and Ilan, D. 2006. The Middle Bronze III and Late Bronze I Pottery from Areas F and N. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 171190. Garinkel, Y. and Cohen, S., eds. 2007. The Middle Bronze Age IIA Cemetery at Gesher: Final Report. (Annual of the American Schools of Oriental Research 62). Boston.
149
Gerstenblith, P. 1983. The Levant at the Beginning of the Middle Bronze Age (American Schools of Oriental Research Dissertation Series 5). Philadelphia. Getzov, N. 2006. The Tel Bet Yera Excavations 19941995 (Israel Antiquities Authority Reports 28). Jerusalem. Golani, A. 1999. New Perspectives on Domestic Architecture and the Initial Stages of Urbanization in Canaan. Levant 31: 123133. Golani, A. 2003. Salvage Excavations at the Early Bronze Age Site of Qiryat Ata (Israel Antiquities Authority Reports 18). Jerusalem. Goren, Y. 2000. Technology, Provenience and Interpretation of the Early Bronze Age Egyptian Ceramics. In: In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 496501. Goren, Y., Finkelstein, I. and Naaman, N. 2004. Inscribed in Clay: Provenance Study of the Amarna Tablets and Other Ancient Near Eastern Texts (Monograph Series of the Institute of Archaeology of Tel Aviv University 23). Tel Aviv. Greenberg, R. 2000. Changes in Ceramic Production between Early Bronze Age II and III in Northern Israel Based on the Pottery of Tel Hazor and Tel Dan. In: Philip, G. and Baird, D., eds. Ceramics and Change in the Early Bronze Age of the Southern Levant. Shefield: 183200. Greenberg, R. 2002. Early Urbanizations in the Levant. A Regional Narrative. London. Greenberg, R. 2003a. What Happened to Megiddo in the Early Bronze Age II? Eretz Israel 27: 6672 (Hebrew). Greenberg, R. 2003b. Early Bronze Age Megiddo and Bet Shean: Discontinuous Settlement in Sociopolitical Context. Journal of Mediterranean Archaeology 16: 1732. Greenberg, R. 2006. Notes on the Early Bronze Age Pottery (The 19982000 Seasons). In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 149165. Greenberg, R., Eisenberg, E., Paz, S. and Paz, Y. 2006. Bet Yerah: The Early Bronze Age Mound. Volume I: Excavation Reports, 19331986 (Israel Antiquities Authority Reports 30). Jerusalem. Guy, P.L.O. 1938. Megiddo Tombs. Chicago. Hallote, R.S. 2001. Tombs, Cult and Chronology: A Reexamination of the Middle Bronze Age Strata of Megiddo. In: Wolff, S.R., ed. Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse (Studies in Ancient Oriental Civilizations 59). Chicago: 199214. Halpern, B. 2000. Centre and Sentry: Megiddos Role in Transit, Administration and Trade. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 535575. Harrison, T.P. 2004. Megiddo 3: Final Report on the Stratum VI Excavations. Chicago. Herzog, Z. 1997. Archaeology of the City: Urban Planning in Ancient Israel and Its Social Implications (Monograph Series of the Institute of Archaeology of Tel Aviv University 13). Tel Aviv. Ibrahim, M. and Mittmann, S. 1994. Excavations at Khirbet ez-Zeraqon, 1993. Newsletter of the Institute of Archaeology and Anthropology, Yarmouk University 16: 1115. Ilan, D. 2001. The Riddle of Structure 5239 at Megiddo, Stratum IX. In: Wolff, S.R., ed. Studies in the Archaeology of Israel and Neighboring Lands in Memory of Douglas L. Esse (Studies in Ancient Oriental Civilizations 59). Chicago: 307316. Ilan, O. and Goren, Y. 2003. The Egyptianized Pottery Vessels of Early Bronze Age Megiddo. Tel Aviv 30: 4253.
150
Joffe, A. 1993. Settlement and Society in the Early Bronze Age I and II, Southern Levant (Monographs in Mediterranean Archaeology 1). Shefield. Joffe, A. 2000. The Early Bronze Age Pottery from Area J. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 161185. Keinan, A. 2007. The Megiddo Picture Pavement: Evidence for Egyptian Presence in Northern Israel during Early Bronze Age I. (M.A. thesis, Tel Aviv University). Tel Aviv (Hebrew). Kemp, B. 2006. Ancient Egypt: Anatomy of a Civilization. London. Kempinski, A. 1989. Megiddo, A City-State and Royal Centre in North Israel. Munich. Kempinski, A. 1992. Chalcolithic and Early Bronze Age Temples. In: Kempinski, A., ed. The Architecture of Ancient Israel from the Prehistoric to the Persian Periods. Jerusalem: 5359. Kenyon, K. 1958. Some Notes on the Early and Middle Bronze Age Strata of Megiddo. Eretz Israel 5: 51*60*. Kenyon, K. 1969. The Middle and Late Bronze Age Strata at Megiddo. Levant 1: 2560. Lamon, R. and Shipton, G. 1939. Megiddo I: Seasons of 192534. Strata IV. Chicago. Lev-Yadun, S. and Weinstein-Evron, M. 2002. The Role of Pinus Halepensis (Aleppo Pine) in the Landscape of Early Bronze Age Megiddo. Tel Aviv 29: 332343. Liphschitz, N. 2000. The Archaeobotanical Finds. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo III: The 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 487495. Liphschitz, N. 2006. Wood Remains. In: Finkelstein, I., Ussishkin, D. and Halpern, B., eds. Megiddo IV: The 19982002 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 24). Tel Aviv: 505518. Liphschitz, N. 2007. Timber in Ancient Israel: Dendroarchaeology and Dendrochronology (Monograph Series of the Institute of Archaeology of Tel Aviv University 26). Tel Aviv. Loud, G. 1948. Megiddo II: Seasons of 193539. Chicago. Marquet-Krause, J. 1949. Les fouilles de Ai (et-Tel) 19331935. Paris. Master, D.M., Monson, J.M., Lass, E.H.E. and Pierce, G.A., eds. 2005. Dothan I. Remains from the Tell (1953 1964). Winona Lake. Mazar, A. 2006. Tel Beth-Shean and the Fate of Mounds in the Intermediate Bronze Age. In: Gitin, S., Wright, J.E. and Dessel, J.P., eds. Confronting the Past: Archaeological and Historical Essays on Ancient Israel in Honor of William G. Dever. Winona Lake: 105118. McConaughy, M.A. 2003. Chipped Stone Tools at Bb edh-Dhr. In: Rast, W.E. and Schaub, R.T., eds., Bb edh-Dhr: Excavations at the Town Site (19751981). Winona Lake: 473512. Nigro, L. 1994. The Nordburg of Megiddo: A New Reconstruction on the Basis of Schumachers plan. Bulletin of the American Schools of Oriental Research 293: 1529. Nigro, L. 1995. Ricerche sularchitettura palaziale della Palestina nelle et del Bronze e del Ferro. Contesto archeologico e sviluppo storico (Contributi e Materiali de Archeologia Orientale V). Rome. Parrot, A. 1956. Le Temple dIshtar (Mission Archologique de Mari I). Paris. Philip, G. and Baird. D. 2000. Early Bronze Age Ceramics in the Southern Levant: An Overview. In: Philip, G. and Baird, D., eds. Ceramics and Change in the Early Bronze Age of the Southern Levant. Shefield: 330. Portugali, J. and Gophna, R. 1993. Crisis, Progress and Urbanization: The Transition from the Early Bronze I to Early Bronze II in Palestine. Tel Aviv 20: 164186.
151
Prent, M. 2003. Glories of the Past in the Past: Ritual Activities at Palatial Ruins in Early Iron Age Crete. In: van Dyke, R.M. and Alcock, S.E., eds. Archaeologies of Memory. Oxford: 81103. Schroer, S. and Keel, O. 2005. Die Ikonographie Palstinas/Israels und der Alte Orient: Eine Religionsgeschichte in Bildern, Band 1: Vom ausgehenden Mesolithikum bis zur Frhbronzezeit. Freiburg. Schumacher, G. 1908. Tell el-Mutesellim. Leipzig. Shahack-Gross, R., Marshall, F. and Weiner, S. 2003. Geo-ethnoarchaeology of Pastoral Sites: The Identiication of Livestock Enclosures in Abandoned Massai Settlements. Journal of Archaeological Science 30: 439459. Shahack-Gross, R., Albert, R.M., Gilboa, A., Nagar-Hilman, O., Sharon, I. and Weiner, S. 2005. Geoarchaeology in an Urban Context: The Uses of Space in a Phoenician Monumental Building at Tel Dor (Israel). Journal of Archaeological Science 32: 14171431. Shahack-Gross, R., Gafri, M. and Finkelstein, I. 2009. Identifying Threshing Floors in the Archaeological Record: A Test Case at Iron Age Tel Megiddo, Israel. Journal of Field Archaeology 34: 171184. Shipton, G.M. 1939. Notes on the Megiddo Pottery of Strata VIXX (Studies in Ancient Oriental Civilization 17). Chicago. Sneh, A., Bartov, Y., Weissbrod, T. and Rosensaft, M. 1998. Geological Map of Israel, 1:1200,000. Israel Geological Survey. (4 sheets). http://www.gsi.gov.il/Eng/Index.asp?ArticleID=172&CategoryID=119&Pa ge=1 Thompson, T.L. 1970. The Dating of the Megiddo Temples in Strata XVXIV. Zeitschrift des Deutschen Palstina-Vereins 86: 3849. Tufnell, O. 1973. The Middle Bronze Age Scarab-seals from Burials on the Mound at Megiddo. Levant 5: 6982. Ussishkin, D. 1980. The Ghassulian Shrine at En-Gedi. Tel Aviv 7: 144. Ussishkin, D. 1989. Schumachers Shrine in Building 338 at Megiddo. Israel Exploration Journal 39: 149172. Van der Steen, E.J. 2001. Megiddo in the Early Bronze Age. Bibliotheca Orientalis 58: 202211. Van der Steen, E.J. 2005. The Sanctuaries of Early Bronze IB Megiddo: Evidence of a Tribal Polity? American Journal of Archaeology 109: 120. 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 19921996 Seasons (Monograph Series of the Institute of Archaeology of Tel Aviv University 18). Tel Aviv: 429462. Weinstein, J. 1973. Foundation Deposits in Ancient Egypt (Ph.D. dissertation, University of Pennsylvania). Philadelphia. Weinstein-Evron, M. and Lev-Yadun, S. 2000. Palaeoecology of Pinus halepensis in Israel in the Light of Archaeobotanical Data. In: Neeman, G. and Trabaud, L., eds. Ecology, Biogeography and Management of Pinus Halepensis and Pinus Brutia Forest Ecosystems in the Mediterranean Basin. Leiden: 119130. Wilkinson, T.A.H. 1999. Royal Annals of Ancient Egypt, The Palermo Stone and Its Associated Fragments. London. Zuckerman, S. 2007. Anatomy of a Destruction: Crisis Architecture, Termination Rituals and the Fall of Canaanite Hazor. Journal of Mediterranean Archaeology 20: 332.
152

Adams Et Al. Ch. 2 Area J 2004-2010 Megiddo V 2013 FULL-Libre

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Adams Et Al. Ch. 2 Area J 2004-2010 Megiddo V 2013 FULL-Libre

Transféré par

Droits d'auteur :

Formats disponibles

MEGIDDO V THE 2004-2008 SEASONS

TEL AVIV UNIVERSITY SONIA AND MARCO NADLER INSTITUTE OF ARCHAEOLOGY

MONOGRAPH SERIES NUMBER 31

Executive Editor Editorial Board

EMERY AND CLAIRE YASS PUBLICATIONS IN ARCHAEOLOGY TEL AVIV UNIVERSITY

WINONA LAKE, INDIANA EISENBRAUNS 2013

of the Institute of Archaeology, Tel Aviv University

Library of Congress Cataloging-in-Publication Data

AREA J PART I: INTRODUCTION

GOALS AND PROGRESS OF WORK

AREA J, PART I: INTODUCTION

CHAPTER 2: A REA J, PART I: I NTODUCTION

CHAPTER 2: A REA J, PART I: I NTODUCTION

TABLE 2.2: REVISED AND AUGMENTED STRATIGRAPHY OF AREA J*

MEGIDDO III AND MEGIDDO IV CORRIGENDA

NOTES ON THE NORTH ARROWS IN MEGIDDO III AND MEGIDDO IV

CONSERVATION, DUMPS AND BACKFILLING

CHAPTER 2: A REA J, PART I: I NTODUCTION

AREA J, PART II: SUB-AREA LOWER J PART II: SUB-AREA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

Fig. 2.10: Northern section of Incision A.

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

Fig. 2.14: Incised slabs.

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

Fig. 2.15: Photograph and drawing (reconstructed) of Slab D.

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

CHAPTER 2: A REA J, PART II: SUB -A REA LOWER J

APPENDIX: TABLE OF THE UNIVERSITY OF CHICAGOS INCISED SLABS

APPENDIX: TABLE OF THE UNIVERSITY OF CHICAGOS INCISED SLABS (cont.)

AREA J, PART III: THE MAIN SECTOR OF AREA J

PART III: THE MAIN SECTOR OF AREA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.20: Remains of Level J-3 on the eastern slope.

LEVEL J-4: THE GREAT TEMPLE

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.24: Remains of Level J-4 (basalt slabs marked in black).

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.30: Level J-4 temple entrance, facing northeast.

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.37: Squares HG/78, eastern section.

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.40: Level J-4 foundation deposit.

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

PHASE J-4A: A PERIOD OF CRISIS

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

AN EB IB EARTHQUAKE: EVIDENCE FOR AND AGAINST

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

Fig. 2.47: Strike fractures in Temple 4040 Wall 08/J/38.

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J

CHAPTER 2: A REA J, PART III: T HE M AIN SECTOR OF A REA J